Energy control method and device, energy scheduling method and device, energy storage system

Abstract

The embodiment of the application provides an energy control method and device, an energy scheduling method and device and an energy storage system, relates to the technical field of energy storage equipment, and the energy control method comprises the following steps: receiving a charging and discharging user instruction; in response to the charging and discharging user instruction, switching from an automatic scheduling state to a cooperative transition state, in the cooperative transition state, controlling the energy storage equipment to perform charging and discharging according to a fusion result of the automatic scheduling instruction and the charging and discharging user instruction, and gradually increasing the control weight of the charging and discharging user instruction in the fusion result with time and gradually reducing the control weight of the automatic scheduling instruction with time; in the case that the control weight of the charging and discharging user instruction reaches a first preset threshold, switching from the cooperative transition state to a user leading state, and controlling the energy storage equipment to perform charging and discharging according to the charging and discharging user instruction. Through the scheme of the application, the operation stability of the energy storage equipment is ensured while the instant demand of the user is considered.

Description

Technical Field

[0001] This invention relates to the field of energy storage equipment technology, and more specifically, to an energy control method and device, an energy dispatching method and device, and an energy storage system. Background Technology

[0002] Residential energy storage devices typically achieve energy management through an edge-cloud collaborative architecture. The cloud-based energy management platform generates automatic dispatch plans based on power generation forecasts, load forecasts, and electricity price data. Users can also issue temporary operation commands through a client or device panel to meet personalized energy needs. In actual operation, the automatic dispatch plan generated in the cloud and the user's temporary operation may act on the same energy storage device at the same time. Once the user manually intervenes, the residential energy storage device will operate entirely according to the user's command, overriding the original automatic dispatch plan, which cannot meet the requirements for stable operation of the device, thus affecting the stability of the device's operation. Summary of the Invention

[0003] The purpose of this invention is to provide an energy control method and device, an energy dispatching method and device, and an energy storage system that can solve the problem that when users manually intervene, the household energy storage equipment operates entirely according to user instructions, overriding the original automatic dispatching plan, which fails to meet the requirements for stable equipment operation and thus affects the stability of equipment operation.

[0004] In view of this, an embodiment of the first aspect of the present invention provides an energy control method.

[0005] An embodiment of the second aspect of the present invention provides an energy scheduling method.

[0006] An embodiment of the third aspect of the present invention provides an energy control device.

[0007] An embodiment of the fourth aspect of the present invention provides an energy dispatching device.

[0008] An embodiment of the fifth aspect of the present invention provides an energy storage system.

[0009] To achieve the above objectives, an embodiment of the first aspect of the present invention provides an energy control method applied to the energy management unit of an energy storage device. The energy control method includes: receiving a charging / discharging user instruction from a user terminal when the energy storage device is controlled to charge and discharge according to an automatic scheduling instruction issued by a cloud platform; responding to the charging / discharging user instruction, switching from an automatic scheduling state to a cooperative transition state; in the cooperative transition state, controlling the energy storage device to perform charging and discharging according to the fusion result of the automatic scheduling instruction and the charging / discharging user instruction, wherein the control weight of the charging / discharging user instruction gradually increases over time, and the control weight of the automatic scheduling instruction gradually decreases over time; when the control weight of the charging / discharging user instruction reaches a first preset threshold, switching from the cooperative transition state to a user-dominated state; in the user-dominated state, controlling the energy storage device to perform charging and discharging according to the charging / discharging user instruction.

[0010] This invention discloses an energy control method applied to the energy management unit of an energy storage device. During the process of the energy management unit controlling the charging and discharging of the energy storage device according to automatic scheduling instructions issued by the cloud platform, it receives charging and discharging user instructions from the user terminal. In response to the charging and discharging user instructions, the system switches from an automatic scheduling state to a collaborative transition state. In this state, the two types of instructions are weighted and fused to generate a fusion result. In the fusion result, the control weight of the charging and discharging user instructions gradually increases over time, while the control weight of the automatic scheduling instructions gradually decreases over time, to form a smooth transition. When the control weight of the charging and discharging user instructions reaches a first preset threshold, the system switches to a user-dominated state, controlling the energy storage device to perform charging and discharging entirely according to the charging and discharging user instructions.

[0011] Understandably, this invention, through a collaborative transition state and weighted gradual instruction fusion mechanism, does not immediately and completely overwrite the automatic scheduling instructions when the user issues instructions. Instead, it allows the control weights of the two types of instructions to transition smoothly over time, avoiding step jumps in operating state and power mutations caused by sudden takeover by user instructions. This eliminates the impact of manual intervention on the stable operation of the equipment, ensuring the operational stability of the energy storage device while taking into account the user's immediate needs.

[0012] In some technical solutions, the energy control method may optionally include: during the user-dominated state, if a preset return condition is detected, switching from the user-dominated state to a collaborative transition state; in the collaborative transition state, controlling the energy storage device to perform charging and discharging according to the fusion result of the automatic scheduling command and the charging and discharging user command, wherein the control weight of the charging and discharging user command gradually decreases over time and the control weight of the automatic scheduling command gradually increases over time; when the control weight of the automatic scheduling command reaches a second preset threshold, switching back from the collaborative transition state to the automatic scheduling state; in the automatic scheduling state, controlling the energy storage device to perform charging and discharging according to the automatic scheduling command issued by the cloud platform.

[0013] After the user's temporary needs are met, a reverse collaborative transition is triggered by preset return conditions, so that the system smoothly returns control from user-led operation to the cloud platform. This avoids the operational status from changing again due to a direct switch back to automatic scheduling, ensuring the flexibility of user operation and ensuring that the energy storage equipment can smoothly return to the optimal energy management track after the user intervention ends.

[0014] In some technical solutions, optionally, the preset return conditions include at least one of the following: the takeover time reaches a preset takeover duration, a user's end-of-takeover instruction is received, and the target operating parameters corresponding to the charging / discharging user instruction reach target values. During the user-dominated state, if the preset return conditions are detected, the system switches from the user-dominated state to the collaborative transition state. This includes: obtaining the takeover start time corresponding to the user-dominated state, and detecting the takeover duration and operating parameters of the energy storage device corresponding to the user-dominated state; determining whether the user-dominated state meets the preset return conditions based on the takeover duration, the user's end-of-takeover instruction, and the operating parameters; and switching from the user-dominated state to the collaborative transition state if the preset return conditions are met.

[0015] Once user needs are met, the process of returning control is automatically initiated. This avoids the user-dominated state being maintained indefinitely and thus blocking cloud scheduling. It ensures the flexibility of user operation and maintains the recoverability of long-term energy management goals in the cloud and the stability of equipment operation.

[0016] In some technical solutions, the energy control method optionally includes: during automatic scheduling, coordinated transition, or user-dominated states, if the charging and discharging control parameters of the energy storage device meet preset safety trigger conditions, switching from the current state to a safety fallback state; in the safety fallback state, controlling the energy storage device to perform charging and discharging according to the corresponding safety operation constraints of the energy storage device, so that the operating state of the energy storage device is within the safe execution range; wherein, the priority of the preset safety trigger conditions is higher than that of the charging and discharging user instructions and automatic scheduling instructions; the preset safety trigger conditions include at least the energy storage device's state of charge reaching the upper limit of the safe state of charge, the energy storage device's state of charge reaching the lower limit of the safe state of charge, the charging power reaching the maximum charging power, the discharging power reaching the maximum discharging power, the grid-connected power reaching the upper limit of the grid-connected power, and the state of charge being lower than the minimum backup charge state.

[0017] To address the issue of potential overcharging, over-discharging, and power exceeding limits due to conflicts between user commands and cloud platform commands, or under extreme operating conditions, this invention sets a safety fallback state as the highest priority. Once a preset safety trigger condition is triggered, user and cloud control commands are forcibly blocked, and charging and discharging are directly executed according to hardware safety constraints. This fundamentally avoids the risk of hardware damage caused by command conflicts, ensuring that energy storage devices maintain physical safety even under extreme conditions.

[0018] In some technical solutions, optionally, after switching from the current state to the safety fallback state, the energy control method further includes: if it is detected that the currently effective charge and discharge control parameters no longer meet the preset safety trigger conditions, obtaining the control state before entering the safety fallback state; determining the target control state based on the control state before entering the safety fallback state, so that the energy storage device returns to the automatic scheduling state, the cooperative transition state, or the user-led state.

[0019] After the safety hazard is eliminated, the system automatically restores to the control state before the trigger, rather than remaining in a safety fallback state or defaulting back to a single state. This ensures that temporary user operations can be seamlessly resumed when conditions permit, avoids unnecessary scheduling interruptions, and balances hardware security protection with system operation continuity.

[0020] In some technical solutions, optionally, the energy storage device is controlled to perform charging and discharging according to the fusion result of the automatic dispatch command and the charging and discharging user command, including: determining the automatic dispatch power according to the automatic dispatch command and determining the user's desired power according to the charging and discharging user command; weighting the automatic dispatch power and the user's desired power according to the control weight of the automatic dispatch command and the control weight of the charging and discharging user command to obtain the fused power; and controlling the energy storage device to perform charging and discharging according to the fused power.

[0021] By weighting and fusing the power automatically scheduled in the cloud with the power expected by the user, the power setting value is obtained, avoiding the step change caused by the user command suddenly completely overriding the automatic scheduling command.

[0022] In some technical solutions, optionally, the automatic scheduling power and the user's desired power are weighted according to the control weight of the automatic scheduling instruction and the control weight of the charging / discharging user instruction. This includes: obtaining the transition start time, transition target weight, and transition time parameter corresponding to the coordinated transition state; determining the control weight corresponding to the current time according to the time interval between the current time and the transition start time and the transition time parameter, and according to a preset weight change rule; and weighting the automatic scheduling power and the user's desired power according to the control weight corresponding to the current time, so that the fused power changes smoothly with the change of control weight.

[0023] By dynamically calculating the control weights based on the transition start time, transition time parameters, and preset weight change rules, the fused power changes continuously and smoothly with the weights, rather than abruptly. This fundamentally eliminates the impact of sudden power changes on the battery management module and controllable loads, prevents the triggering of overcurrent protection or operational abnormalities, and ensures the continuous and stable operation of the energy storage device throughout the entire process of user intervention or return of control.

[0024] In some technical solutions, the preset weight change rule optionally includes at least one of the exponential change rule, the linear change rule, and the piecewise change rule. The preset weight change rule is used to make the control weight gradually change from the current weight to the transition target weight according to the corresponding change trajectory.

[0025] By providing multiple weight change rules, including exponential, linear, and piecewise, the system can select the most suitable change trajectory based on actual operating conditions such as battery state of charge, ensuring that the control weights can smoothly transition to the target value under various operating conditions, and further guaranteeing the continuous and stable operation of energy storage equipment under different operating conditions.

[0026] In some technical solutions, the energy control method may optionally include: acquiring the safe operating parameters corresponding to the energy storage device; determining the safe execution range corresponding to the energy storage device based on the safe operating parameters; detecting whether the operating status parameters of the energy storage device exceed the safe execution range during the process of controlling the energy storage device to perform charging and discharging according to the fusion result; if any operating status parameter is detected to exceed the safe execution range, adjusting the fusion result to make the adjusted charging and discharging control parameters within the safe execution range; and controlling the energy storage device to perform charging and discharging according to the adjusted charging and discharging control parameters.

[0027] When the fusion result exceeds the safety boundary due to instruction conflict or calculation abnormality, the energy management unit automatically limits and trims the charging and discharging control parameters to bring them back to the safe execution range. This prevents hardware safety risks such as power over-limit and battery overcharging and over-discharging caused by temporary user operations, and ensures the continuous and stable operation of the energy storage device under all operating conditions.

[0028] In some technical solutions, optionally, when any operating state parameter is detected to exceed the safe execution range, the fusion result is restricted and adjusted so that the adjusted charging and discharging control parameters are within the safe execution range. This includes: when the user's expected power corresponding to the charging and discharging user command conflicts with the safe execution range, restricting the user's expected power according to the safe execution range to determine the restricted power; weighting the automatic scheduling power and the restricted power according to the control weight of the automatic scheduling command and the control weight of the charging and discharging user command to determine the restricted fusion power; and controlling the energy storage device to perform charging and discharging according to the restricted fusion power so that the energy storage device executes the charging and discharging user command while meeting the safe execution range.

[0029] Before fusion computing, the user's expected power is individually limited and then included in the weighted processing. This not only prevents the user's over-limit command from exceeding the limit and the cloud command from exceeding the safety boundary, but also preserves the effective execution of the user's operation intention within the limited range. At the same time, it maintains the participation of the automatic scheduling command in the fusion process, and achieves the optimal balance between security protection, user demand response and cloud scheduling continuity.

[0030] In some technical solutions, optionally, the automatic dispatching instructions are generated by the cloud platform and sent to the energy management unit. The energy control method also includes: acquiring the operating data of the energy storage device and sending the operating data to the cloud platform so that the cloud platform can determine the automatic dispatching instructions based on the operating data; the energy management unit receives the automatic dispatching instructions sent by the cloud platform and controls the energy storage device to perform charging and discharging according to the automatic dispatching instructions.

[0031] By placing computationally intensive prediction and scheduling calculations on the cloud platform, the energy management unit only needs to be responsible for data collection and local execution, which leverages the advantages of the cloud platform's global data perspective and strong computing power while reducing local hardware costs.

[0032] In some technical solutions, the energy control method may optionally include: the energy management unit reporting device identifier, scene identifier, and device capability information to the cloud platform; the energy management unit receiving an adapted automatic scheduling instruction issued by the cloud platform based on the device capability information; wherein the device capability information includes at least: whether it supports AC charging power setting, whether it supports inverter discharge power setting, whether it supports photovoltaic charging power limitation, whether it supports grid-connected power upper limit setting, whether it supports target state of charge setting, and whether it supports local policy caching.

[0033] To address the issue of mismatch between cloud-based scheduling commands and the hardware capabilities of different user terminals, this invention enables the cloud platform to perceive the actual hardware configuration of the energy management unit by reporting device capability information. This allows for the dynamic trimming of parameter items in automatic scheduling commands, issuing only control commands supported by the device. This ensures correct implementation on residential energy storage devices with different hardware configurations, thereby improving the compatibility and reliability of the end-cloud collaborative scheduling system.

[0034] In some technical solutions, the energy control method may optionally include: receiving constraint preference parameters from a user terminal, the constraint preference parameters including setting a target state of charge, pre-trip backup power, and power maintenance range; and uploading the constraint preference parameters to a cloud platform so that the cloud platform generates an automatic scheduling instruction including the constraint preference parameters.

[0035] Users can upload their energy consumption preferences and travel plans as constraint parameters to the cloud platform through their user terminals. When generating automatic dispatch instructions, the cloud platform incorporates these subjective needs into its calculations, avoiding conflicts between dispatch instructions and user schedules, and improving user acceptance and satisfaction with the energy storage system.

[0036] In some technical solutions, the energy control method may optionally include: when the energy management unit communicates normally with the cloud platform, storing at least one of the automatic scheduling instructions, preset weight change rules, and preset return conditions in the energy management unit; when the communication between the energy management unit and the cloud platform is interrupted, controlling the energy storage device to perform charging and discharging according to at least one of the automatic scheduling instructions, preset weight change rules, and preset return conditions stored in the energy management unit.

[0037] When communication is interrupted and the user sends a local command, the energy management unit calls the cached weight change rules and return condition parameters to perform weighted fusion and smooth state switching between the charging and discharging user commands and the cached automatic scheduling commands. This avoids power jumps caused by user intervention in network outage scenarios and ensures the continuous stability of energy storage equipment operation during communication failures.

[0038] In some technical solutions, optionally, in the event of a communication interruption between the energy management unit and the cloud platform, the energy control method further includes: receiving charging and discharging user instructions sent by the user terminal, and adjusting the control weights corresponding to the automatic scheduling instructions and the charging and discharging user instructions according to preset weight change rules stored in the energy management unit; determining the fusion result based on the adjusted control weights, and controlling the energy storage device to perform charging and discharging according to the fusion result; determining whether to switch from the user-dominated state to the collaborative transition state according to preset return conditions stored in the energy management unit, and switching to the automatic scheduling state when the control weight corresponding to the automatic scheduling instruction reaches a second preset threshold.

[0039] In network outage scenarios, the system still uses pre-defined weight change rules and pre-defined return conditions in the local cache to perform weighted fusion of instructions and smooth state switching, rather than directly replacing cached automatic scheduling instructions with user commands. This avoids power step jumps caused by temporary user operations during network outages, ensuring that the transition of control in the offline state is consistent with that in the online state, and guaranteeing the continuous stability of energy storage equipment operation during communication failures.

[0040] A second aspect of the present invention provides an energy dispatching method applied to a cloud platform. The cloud platform is communicatively connected to the energy management unit of an energy storage device. The energy dispatching method includes: receiving operating data and device capacity information of the energy storage device reported by the energy management unit; acquiring dispatching data corresponding to the energy storage device, including photovoltaic power generation forecast data, household load forecast data, and electricity price data; determining device operating constraints corresponding to the energy storage device based on the operating data, and determining automatic dispatching control parameters based on the dispatching data and device operating constraints; generating automatic dispatching instructions based on the automatic dispatching control parameters, and sending the automatic dispatching instructions to the energy management unit so that the energy management unit controls the energy storage device to perform charging and discharging according to the automatic dispatching instructions.

[0041] This invention discloses an energy dispatching method applied to a cloud platform. The cloud platform receives energy storage device operation data and device capacity information reported by the energy management unit, obtains photovoltaic power generation forecast data, household load forecast data and electricity price data, determines device operation constraints based on the operation data, calculates automatic dispatching control parameters by combining dispatching data and constraints, generates automatic dispatching instructions and sends them to the energy management unit for charging and discharging.

[0042] Understandably, by placing computationally intensive prediction and optimization calculations on a cloud platform, and leveraging a global data perspective to integrate power generation forecasts, load forecasts, electricity prices, and real-time equipment status and capabilities, optimal scheduling instructions that balance electricity costs and hardware security are generated. The introduction of equipment capability information allows instructions to automatically adapt to different hardware configurations, preventing execution failures caused by issuing parameters unsupported by the equipment. This cloud-based scheduling architecture fully utilizes the powerful computing capabilities of the cloud platform, enabling precise, secure, and personalized energy management of distributed residential energy storage systems.

[0043] In some technical solutions, the energy scheduling method may optionally further include: receiving constraint preference parameters and / or user takeover records uploaded by the energy management unit. The constraint preference parameters are sent by the user terminal to the energy management unit and uploaded to the cloud platform. The user takeover records include the user takeover time, user takeover type, and user takeover duration. If the constraint preference parameters exist, the optimization constraint conditions are determined based on the constraint preference parameters, and a new automatic scheduling instruction is generated and sent to the energy management unit. If the user takeover records exist, the user takeover behavior is determined based on the user takeover records within a preset time period to determine whether the user takeover behavior meets the preset preference conditions. If the user takeover behavior meets the preset preference conditions, user preference parameters are generated based on the user takeover behavior, and the automatic scheduling control parameters are adjusted based on the user preference parameters. A new automatic scheduling instruction is generated based on the adjusted automatic scheduling control parameters and sent to the energy management unit.

[0044] By incorporating user-defined energy demands into scheduling constraints through constraint preference parameters, new automatic scheduling instructions are generated. Simultaneously, by analyzing user takeover records, user preference parameters are automatically learned and adjusted when user behavior meets preset preference conditions, reducing the frequency of manual user intervention and balancing grid safety with user experience.

[0045] A third aspect of the present invention provides an energy control device applied to the energy management unit of an energy storage device. The energy control device includes: an instruction receiving module, used to receive charging and discharging user instructions from a user terminal when controlling the energy storage device to charge and discharge according to automatic scheduling instructions issued by a cloud platform; a cooperative transition module, used to switch from an automatic scheduling state to a cooperative transition state in response to the charging and discharging user instructions, and in the cooperative transition state, controlling the energy storage device to perform charging and discharging according to the fusion result of the automatic scheduling instructions and the charging and discharging user instructions, wherein the control weight of the charging and discharging user instructions gradually increases over time, and the control weight of the automatic scheduling instructions gradually decreases over time; and a user-dominated module, used to switch from the cooperative transition state to the user-dominated state when the control weight of the charging and discharging user instructions reaches a first preset threshold, and in the user-dominated state, controlling the energy storage device to perform charging and discharging according to the charging and discharging user instructions.

[0046] An embodiment of the fourth aspect of the present invention provides an energy dispatching device applied to a cloud platform. The cloud platform is communicatively connected to the energy management unit of an energy storage device. The energy dispatching device includes: a data receiving module for receiving operating data and device capacity information of the energy storage device reported by the energy management unit; a data dispatching module for acquiring dispatching data corresponding to the energy storage device, including photovoltaic power generation forecast data, household load forecast data, and electricity price data; a constraint determination module for determining the device operating constraints corresponding to the energy storage device based on the operating data, and determining automatic dispatching control parameters based on the dispatching data and the device operating constraints; and an instruction issuing module for generating automatic dispatching instructions based on the automatic dispatching control parameters, and issuing the automatic dispatching instructions to the energy management unit so that the energy management unit controls the energy storage device to perform charging and discharging according to the automatic dispatching instructions.

[0047] An embodiment of the fifth aspect of the present invention provides an energy storage system, which includes an energy storage device and a cloud platform. The energy storage device includes an energy management unit, and the cloud platform is communicatively connected to the energy management unit. The energy management unit includes the energy control device of the third aspect, and the cloud platform includes the energy dispatching device of the fourth aspect.

[0048] Additional aspects and advantages of the technical solutions of the present invention will become apparent in the following description or may be learned by practice of the invention. Attached Figure Description

[0049] Figure 1 One of the schematic flowcharts of the energy control method of the present invention is shown;

[0050] Figure 2 A second schematic flowchart of the energy control method of the present invention is shown;

[0051] Figure 3 The third schematic flowchart of the energy control method of the present invention is shown;

[0052] Figure 4 One of the flowcharts of the energy dispatching method of the present invention is shown;

[0053] Figure 5 A second schematic flowchart of the energy dispatching method of the present invention is shown;

[0054] Figure 6 A schematic block diagram of the energy control device of the present invention is shown;

[0055] Figure 7 A schematic block diagram of the energy dispatching device of the present invention is shown;

[0056] Figure 8 A schematic block diagram of the energy storage system of the present invention is shown;

[0057] Figure 9 The diagram shows the framework of the human-machine co-driving software for the energy storage system.

[0058] Figure 10 This shows one of the overall flowcharts for human-machine co-driving scheduling;

[0059] Figure 11 The second part of the overall flowchart for human-machine co-driving scheduling is shown.

[0060] Among them, 800: Energy control device; 802: Command receiving module; 804: Cooperative transition module; 806: User-led module; 900: Energy scheduling device; 902: Data receiving module; 904: Data scheduling module; 906: Constraint determination module; 908: Command issuing module;

[0061] 1000: Energy storage system; 1100: Energy storage device; 1200: Cloud platform; 1102: Energy management unit. Detailed Implementation

[0062] To better understand the above-described objectives, features, and advantages of the embodiments of the present invention, the embodiments of the present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments. It should be noted that, unless otherwise specified, the embodiments of the present invention and the features thereof can be combined with each other.

[0063] Many specific details are set forth in the following description in order to provide a full understanding of the invention. However, embodiments of the invention may be implemented in other ways different from those described herein. Therefore, the scope of protection of the invention is not limited to the specific embodiments disclosed below.

[0064] With the increasing popularity of balcony photovoltaic (PV) systems, residential rooftop PV systems, and household energy storage devices, integrated energy storage units (including battery packs, battery management modules, charging / discharging and inverter modules, energy management modules, and communication modules) are increasingly being connected to cloud-based energy management platforms, forming an energy system of "cloud-based scheduling and device execution." Based on PV power generation forecasts, household load forecasts, time-of-use or spot electricity prices, and battery state of charge, the cloud-based energy management platform automatically generates charging, discharging, or hold-up scheduling plans for the integrated energy storage unit and distributes these plans to the unit's energy management module for execution via the Internet of Things (IoT) channel.

[0065] In this type of system, in addition to automatic scheduling, users can also temporarily intervene through mobile application clients or device panels, such as immediately charging for travel, temporarily discharging to power critical loads, or maintaining low power during holidays. In other words, the energy storage unit must simultaneously accept automatic scheduling from cloud-based artificial intelligence algorithms (hereinafter referred to as "machine side") and respond to temporary operational intentions from users (hereinafter referred to as "human side" or "user side"). The two jointly determine the charging and discharging behavior of the same energy storage unit. This operating mode, in which humans and machines jointly and collaboratively control the same execution unit, can be compared to human-machine co-driving in the field of intelligent driving: the machine assumes the main control most of the time, and the user can intervene and take over when needed. After the intervention, the control should smoothly and controllably return to the machine side.

[0066] Almost all related human-machine co-driving technologies are concentrated in the field of intelligent driving, and their control object is vehicle movement (steering, acceleration and deceleration). The goal of negotiation and takeover is driving safety. On the other hand, existing energy storage scheduling technologies are almost all purely machine-side optimization solutions, and user interaction is limited to the level of "mode selection" or "manual hard overlay". The two have not yet been combined.

[0067] Once a user manually intervenes, the machine-side scheduling is suspended entirely. There is no proportional integration and negotiation mechanism between user operations and algorithm optimization, which means that temporary and localized user intentions (such as "charging the battery before traveling") often completely overturn the originally reasonable economic scheduling.

[0068] After user intervention, most systems require the user to manually switch back to automatic mode, or wait for a fixed period of time unrelated to the scenario before resuming automatic scheduling; if the user forgets to switch back, the energy storage unit may remain in an uneconomical or even detrimental operating state for a long time.

[0069] When switching from user control back to machine control (or vice versa), the charging and discharging power often undergoes a step jump, which is not conducive to the stable operation of the battery and inverter, and may also cause instantaneous impact on the power supply to household loads or grid feeders.

[0070] Furthermore, when user manual commands conflict with battery safety boundaries, grid connection and feeder limitations, or minimum backup power requirements, existing solutions lack unified and interpretable arbitration rules, which may lead to a dilemma of either blindly obeying users and breaking safety boundaries, or simply rejecting users and resulting in a poor user experience.

[0071] The energy control method and device, energy dispatching method and device, and energy storage system provided by the embodiments of the present invention will be described in detail below through specific examples and application scenarios.

[0072] This embodiment provides an energy control method applied to the energy management unit of an energy storage device, such as... Figure 1 As shown, the energy control methods include:

[0073] Step S100: While controlling the energy storage device to charge and discharge according to the automatic scheduling instructions issued by the cloud platform, receive the charging and discharging user instructions from the user terminal;

[0074] Step S102: In response to the charging and discharging user command, switch from the automatic scheduling state to the cooperative transition state. In the cooperative transition state, control the energy storage device to perform charging and discharging according to the fusion result of the automatic scheduling command and the charging and discharging user command. In the fusion result, the control weight of the charging and discharging user command gradually increases over time, while the control weight of the automatic scheduling command gradually decreases over time.

[0075] Step S104: When the control weight of the charging and discharging user command reaches the first preset threshold, switch from the collaborative transition state to the user-dominated state. In the user-dominated state, control the energy storage device to perform charging and discharging according to the charging and discharging user command.

[0076] This invention discloses an energy control method applied to the energy management unit of an energy storage device. During the process of the energy management unit controlling the charging and discharging of the energy storage device according to automatic scheduling instructions issued by the cloud platform, it receives charging and discharging user instructions from the user terminal. In response to the charging and discharging user instructions, the system switches from an automatic scheduling state to a collaborative transition state. In this state, the two types of instructions are weighted and fused to generate a fusion result. In the fusion result, the control weight of the charging and discharging user instructions gradually increases over time, while the control weight of the automatic scheduling instructions gradually decreases over time, to form a smooth transition. When the control weight of the charging and discharging user instructions reaches a first preset threshold, the system switches to a user-dominated state, controlling the energy storage device to perform charging and discharging entirely according to the charging and discharging user instructions.

[0077] Understandably, this invention, through a collaborative transition state and weighted gradual instruction fusion mechanism, does not immediately and completely overwrite the automatic scheduling instructions when the user issues instructions. Instead, it allows the control weights of the two types of instructions to transition smoothly over time, avoiding step jumps in operating state and power mutations caused by sudden takeover by user instructions. This eliminates the impact of manual intervention on the stable operation of the equipment, ensuring the operational stability of the energy storage device while taking into account the user's immediate needs.

[0078] In the actual operation of energy storage systems (residential photovoltaic-storage systems), the energy storage devices establish a communication connection with the cloud platform through an energy management unit. The cloud platform generates automatic dispatch instructions based on photovoltaic power generation forecast data, household load forecast data, and electricity price data, and sends these instructions to the energy management unit. Upon receiving the automatic dispatch instructions, the energy management unit controls the energy storage devices to perform charging or discharging operations according to the instructions; at this time, the energy management unit is in automatic dispatch mode. In automatic dispatch mode, the charging and discharging behavior of the energy storage devices is entirely determined by the automatic dispatch instructions issued by the cloud platform, aiming to achieve a comprehensive optimization between electricity costs, battery life protection, and photovoltaic power consumption.

[0079] User terminals include mobile application clients or device panels. During actual use, users may experience temporary energy needs at any time, such as needing to charge immediately for backup power during travel or needing to discharge immediately to handle temporary loads. In these cases, users send charge / discharge user commands to the energy management unit through their user terminals. The energy management unit, while in automatic scheduling mode and controlling the energy storage devices to perform charging and discharging according to automatic scheduling instructions, receives the charge / discharge user commands from the user terminals.

[0080] In response to received charge / discharge user commands, the energy management unit switches from automatic dispatch state to a cooperative transition state. The cooperative transition state is an intermediate state between the automatic dispatch state and the user-dominated state. In the cooperative transition state, the energy management unit no longer controls the energy storage device solely based on the automatic dispatch commands or solely based on the charge / discharge user commands. Instead, it controls the energy storage device to perform charge / discharge operations according to the fusion result of the automatic dispatch commands and the charge / discharge user commands. The fusion result is obtained by weighting the automatic dispatch power determined by the automatic dispatch commands and the user-expected power determined by the charge / discharge user commands. The fusion result simultaneously contains the control components of both the automatic dispatch commands and the charge / discharge user commands.

[0081] During the coordinated transition, the control weight of charging and discharging user commands gradually increases over time, while the control weight of automatic scheduling commands gradually decreases. The control weight represents the proportion of influence of the corresponding command in the fusion result. The control weight of charging and discharging user commands gradually increases from its initial value over time, while the control weight of automatic scheduling commands gradually decreases from its initial value over time. This weighting mechanism ensures that the charging and discharging power of the energy storage device maintains a continuous and smooth change during the transition from automatic scheduling to user-dominated mode, avoiding abrupt power jumps. The specific trajectory of the weighting changes is determined by preset weighting rules.

[0082] When the control weight of the charging / discharging user command reaches the first preset threshold, the energy management unit switches from a collaborative transition state to a user-dominated state. The first preset threshold is a pre-set weight judgment value used to determine the entry condition for the user-dominated state. In the user-dominated state, the energy management unit no longer refers to the automatic dispatch command and controls the energy storage device to perform charging and discharging entirely according to the charging / discharging user command. At this time, the charging and discharging behavior of the energy storage device fully responds to the user's temporary operational intentions, and the automatic dispatch command issued by the cloud platform is suspended during the user-dominated state.

[0083] The energy management unit (EMU), as the local control core on the energy storage device side, is responsible for switching decisions and charging / discharging control between automatic scheduling, collaborative transition, and user-led states. The user terminal, as the user interaction interface, sends charging / discharging user commands to the EMU. The cloud platform, as the remote optimization decision-making end, is responsible for generating and issuing automatic scheduling commands. The EMU, user terminal, and cloud platform interact via a communication network, collectively forming an edge-cloud collaborative energy storage system architecture.

[0084] In some embodiments, optionally, in the fusion result of charging / discharging user commands and automatic scheduling commands, the weight change curves of the two types of commands are pre-configured continuous nonlinear functions, and / or the shape parameters of the continuous nonlinear functions are obtained through offline training based on user satisfaction scores and equipment operation stability indicators in the historical switching records of energy storage devices, to ensure that the fusion transition process can maintain optimal dynamic response characteristics under different load characteristics and battery states.

[0085] In some embodiments, the first preset threshold is optionally set according to the trade-off requirements of the energy storage device for control switching sensitivity and system stability in different application scenarios, and / or the first preset threshold is mapped to at least one of the current battery charging and discharging power limit parameters, the rated power of the controllable load, and the upper limit of the grid interaction power, so that the switching triggering condition can be adaptively adjusted according to the hardware capability boundary, and avoid system oscillation caused by switching too early or too late due to a fixed threshold.

[0086] In some embodiments, the user terminal's permission levels are pre-defined, and / or the charging and discharging user commands corresponding to different permission levels have different initial weight values ​​and different weight increment rates in the collaborative transition state, so that the emergency operations of high-privilege users can gain control more quickly, while the routine adjustments of low-privilege users are still mainly based on a smooth transition, taking into account both system security and user operation flexibility, and is used to adapt to the use of different members in the family.

[0087] In some embodiments, optionally, such as Figure 2 As shown, the energy control method also includes:

[0088] Step S110: During the user-dominated state, if the preset return conditions are met, switch from the user-dominated state to the collaborative transition state. In the collaborative transition state, control the energy storage device to perform charging and discharging according to the fusion result of the automatic scheduling command and the charging and discharging user command. In the fusion result, the control weight of the charging and discharging user command gradually decreases over time, while the control weight of the automatic scheduling command gradually increases over time.

[0089] Step S112: When the control weight of the automatic scheduling command reaches the second preset threshold, switch back from the collaborative transition state to the automatic scheduling state. In the automatic scheduling state, control the energy storage device to perform charging and discharging according to the automatic scheduling command issued by the cloud platform.

[0090] In this embodiment, when a preset return condition is met during user-led operation, the system switches to a collaborative transition state, merging automatic scheduling commands with user charging / discharging commands. The control weight of the automatic scheduling commands gradually increases over time, while the control weight of the user charging / discharging commands gradually decreases. When the control weight of the automatic scheduling commands reaches a second preset threshold, the system switches back to automatic scheduling, controlling the energy storage device to charge and discharge entirely according to the automatic scheduling commands issued by the cloud platform.

[0091] Understandably, after the user's temporary needs are met, a reverse collaborative transition is triggered by preset return conditions, so that the system smoothly returns control from user-led operation to the cloud platform. This avoids the operational status from changing again due to a direct switch back to automatic scheduling, ensuring the flexibility of user operation and ensuring that the energy storage device can smoothly return to the optimal energy management track after the user intervention ends.

[0092] During user-controlled operation, the energy management unit controls the energy storage device to perform charging and discharging operations entirely according to the charging and discharging user commands from the user terminal.

[0093] User-dominated mode cannot continue indefinitely because the automatic dispatch commands issued by the cloud platform carry long-term optimized dispatch plans generated based on photovoltaic power generation forecast data, household load forecast data, and electricity price data. Prolonged suspension of automatic dispatch commands can lead to increased risks of battery overcharging and over-discharging, reduced peak-valley arbitrage profits, or a decrease in photovoltaic grid integration rates. Therefore, it is necessary to return control from user-dominated mode to automatic dispatch mode at an appropriate time.

[0094] The preset return conditions are used to determine when to initiate the transition from user-led to collaborative mode. These preset return conditions include at least one of the following: the takeover time reaches a preset takeover duration, a user's command to end takeover is received, or the target operating parameter corresponding to a charge / discharge user command reaches a target value.

[0095] The preset takeover duration is obtained and stored by the energy management unit from the cloud platform when communication is normal, or it is pre-configured locally by the energy management unit.

[0096] The user's command to end takeover is sent from the user terminal to the energy management unit. Target operating parameters include the energy storage device's state of charge (SOC), charging power, or discharging power. Target values ​​are determined by the settings carried in the user's charging / discharging command. During user-dominated operation, the energy management unit continuously monitors whether the above return conditions are met; once any one of them is met, a state switch is triggered.

[0097] After switching from the user-dominated state to the collaborative transition state, the energy management unit re-enters the collaborative transition state. In the collaborative transition state of the return path, the energy management unit again controls the energy storage device to perform charging and discharging according to the fusion result of the automatic dispatch instructions and the charging / discharging user instructions. Unlike the gradual increase in the weight of user instructions when transitioning from the automatic dispatch state to the collaborative transition state, in the collaborative transition state of the return path, the control weight of charging / discharging user instructions gradually decreases over time, while the control weight of automatic dispatch instructions gradually increases over time.

[0098] This reverse weighting change allows the charging and discharging power of energy storage devices to gradually transition from being user-intention-driven back to being cloud-driven, maintaining a continuous and smooth power change process and avoiding the impact of abrupt changes on the battery pack and inverter.

[0099] The control weight of the automatic scheduling command gradually increases from its initial value when entering the return coordination transition state, while the control weight of the charging / discharging user command gradually decreases from its initial value when entering the return coordination transition state. The specific trajectory of the weight change is also determined by the preset weight change rules, which include exponential change rules, linear change rules, or piecewise change rules.

[0100] The rate of weight change and transition time parameters are determined by the energy management unit based on preset return conditions stored locally.

[0101] When the control weight of the automatic dispatch command reaches the second preset threshold, the energy management unit switches from the cooperative transition state back to the automatic dispatch state. The second preset threshold is a pre-set weight judgment value used to determine the conditions for restoring the automatic dispatch state.

[0102] The second preset threshold can be the same as the first preset threshold, or it can be configured to a different value according to the specific needs of the return scenario.

[0103] In automatic scheduling mode, the energy management unit resumes controlling the energy storage device to perform charging and discharging according to the automatic scheduling instructions issued by the cloud platform. The automatic scheduling instructions generated by the cloud platform once again become the sole control basis for the charging and discharging of the energy storage device.

[0104] As the decision-making and execution entity for state switching, the energy management unit continuously monitors the return conditions during the user-dominated state and autonomously triggers state switching when the conditions are met, without the need for real-time intervention from the cloud platform.

[0105] As one of the triggering sources for the return process, the user terminal can proactively terminate the user-dominated state by sending a user termination takeover command. When communication is normal, the cloud platform is responsible for providing parameters such as the preset takeover duration, preset weight change rules, and a second preset threshold. The energy management unit stores these parameters locally and calls them during the return process. The energy management unit, user terminal, and cloud platform collaborate through the communication network to complete the entire process of control handover from user takeover to automatic scheduling recovery.

[0106] In some embodiments, optionally, during the transition from a user-led state to a collaborative transition state, the energy management unit adopts an asymmetric configuration for the control weight decay curve of the charging and discharging user command and the weight rise curve of the automatic scheduling command, and / or the parameters of the asymmetric configuration are selected from a predefined set of curve shapes to adapt to the power transition requirements under two different return triggering scenarios: user-initiated termination of takeover and natural achievement of operating parameters. This allows the weight change trajectory of the return transition to cover multiple asymmetric shapes, improving the matching degree of the state switching process to the actual operating conditions.

[0107] In some embodiments, optionally, during the cooperative transition state, the energy management unit also detects in real time whether the charging and discharging user instructions contain an explicit continuous dominance flag, and / or when a continuous dominance flag is detected, the weight increment rate of the automatic scheduling instructions is forcibly reduced or the increment is paused, so that user operations with high priority continuous control requirements can obtain additional weight maintenance capabilities during the return transition process, and avoid the control being prematurely taken back in emergency situations.

[0108] In some embodiments, optionally, the preset return conditions include at least one of the following: the takeover time reaches a preset takeover duration, a user's end-of-takeover instruction is received, and the target operating parameters corresponding to the charging / discharging user instruction reach a target value. During the user-dominated state, if the preset return conditions are detected, the user-dominated state is switched to a collaborative transition state, including: obtaining the takeover start time corresponding to the user-dominated state, and detecting the takeover duration corresponding to the user-dominated state and the operating parameters of the energy storage device; determining whether the user-dominated state meets the preset return conditions based on the takeover duration, the user's end-of-takeover instruction, and the operating parameters; and switching from the user-dominated state to the collaborative transition state if the preset return conditions are met.

[0109] In this embodiment, under user-dominated mode, the takeover start time corresponding to the user-dominated mode is obtained, and the takeover duration and operating parameters of the energy storage device are detected. Preset return conditions include at least one of the following: the takeover time reaches a preset takeover duration, a user's end-of-take command is received, and the target operating parameters corresponding to the charge / discharge user command reach target values. When the preset return conditions are met based on the takeover duration, the user's end-of-take command, and the operating parameters, the system switches from user-dominated mode to a collaborative transition mode, gradually returning control to the cloud platform.

[0110] Understandably, the automatic initiation of the control return process after the user's needs are met avoids the user's dominant state being maintained indefinitely and thus blocking cloud scheduling. This ensures both the flexibility of user operations and the recoverability of the long-term energy management goals of the cloud and the stability of equipment operation.

[0111] In the actual operation of a residential photovoltaic and energy storage system, after the user sends a charging and discharging user command to the energy management unit through the user terminal, the energy management unit switches from the automatic scheduling state to the user-led state through the cooperative transition state.

[0112] In user-controlled mode, the charging and discharging behavior of energy storage devices is executed entirely according to the user's charging and discharging instructions, and the automatic scheduling instructions issued by the cloud platform are temporarily suspended.

[0113] User-dominated mode cannot continue indefinitely, as prolonged suspension of automatic dispatch commands can increase the risk of battery overcharging and over-discharging, reduce peak-valley arbitrage profits, or decrease photovoltaic grid integration rates. Therefore, the energy management unit needs to automatically or, based on user-initiated actions, transfer control from user-dominated mode to a collaborative transition mode at appropriate times, ultimately restoring to automatic dispatch mode.

[0114] The preset return conditions are the basis for the energy management unit to determine whether to initiate the transition from user-led to collaborative state.

[0115] The preset return conditions include at least one of the following: the takeover time reaches the preset takeover duration, a user's command to end takeover is received, and the target operating parameter corresponding to the user's charge / discharge command reaches the target value. These three conditions cover different actual use scenarios and user operating habits. The energy management unit continuously monitors whether any one of these three conditions is met during user-controlled operation.

[0116] A takeover time exceeding the preset takeover duration means that the duration of the user-dominated state exceeds the pre-set maximum allowable takeover time. The preset takeover duration needs to be determined by balancing the duration of the user's temporary energy demand with the urgency of the automatic scheduling plan's recovery. If the preset takeover duration is too short, the user may be forced to reclaim control before completing the temporary energy demand operation, affecting the user experience. If the preset takeover duration is too long, the automatic scheduling command will be delayed for too long, potentially leading to missed peak-valley arbitrage windows or battery overcharging and over-discharging.

[0117] Therefore, the specific value of the preset takeover duration is determined by the cloud platform based on historical electricity consumption data, electricity price distribution during different time periods, and user behavior patterns.

[0118] Receiving a user's command to end takeover means that the user has proactively sent a command to the energy management unit through the user terminal to end the current user-dominated state.

[0119] The user terminal provides an interactive entry point for ending takeover. After completing the temporary power consumption operation, the user can trigger the generation and sending of the end-of-takeover command by clicking the end-of-takeover button or closing the temporary operation interface.

[0120] The energy management unit receives the user's command to end takeover via the communication network and immediately responds as if one of the preset return conditions has been met.

[0121] The target operating parameters corresponding to the charge / discharge user command reaching the target value means that the energy storage device's actual operating state reaches the target set in the charge / discharge user command during the execution of the command. The target operating parameters include at least one of the energy storage device's state of charge, charging power, or discharging power.

[0122] The target value is determined by the set value carried in the charge / discharge user command. For example, the charge / discharge user command requires the energy storage device to be charged to a specific state of charge value. When the current state of charge value read by the energy management unit from the battery management unit reaches the specific state of charge value, it is determined that the target operating parameter has reached the target value.

[0123] For example, if a user command requests a specific power for charging or discharging, and the actual discharge power decreases to zero due to the battery being depleted or cannot maintain the specific power, it is also considered that the target operating parameter has reached the target value.

[0124] During the user-dominated state, the energy management unit acquires the takeover start time corresponding to the user-dominated state. The takeover start time is the system time recorded by the energy management unit when switching from the cooperative transition state to the user-dominated state. The takeover start time is provided by the real-time clock circuit inside the energy management unit.

[0125] The real-time clock circuit synchronizes with the cloud platform or via a network time protocol after the system is powered on to ensure the accuracy of the timestamps.

[0126] The energy management unit continuously monitors the takeover duration during user-dominated states. The takeover duration is obtained by subtracting the takeover start time from the current time. The current time is read by the real-time clock circuit at the beginning of each control cycle. The energy management unit compares the takeover duration with a preset takeover duration. When the takeover duration is greater than or equal to the preset takeover duration, it determines that the takeover time has reached the preset takeover duration.

[0127] During user-dominated operation, the energy management unit continuously monitors the communication network interface to detect whether a user's termination command has been received. When a user termination command arrives, the energy management unit immediately marks the condition of receiving the command as true.

[0128] During user-controlled operation, the energy management unit continuously reads the state of charge (SOC) values ​​of the energy storage device from the battery management unit (BMU) and the real-time charging or discharging power values ​​from the inverter's power sampling circuit. The energy management unit compares the read operating parameters with the target values ​​carried in the charging / discharging user commands. When the operating parameters meet the conditions defined for achieving the target value, the energy management unit determines that the target operating parameter corresponding to the charging / discharging user command has reached the target value.

[0129] The energy management unit determines whether the user-dominated state meets the preset return conditions based on the takeover duration, the user's command to end takeover, and operating parameters. The determination method is as follows: when the takeover duration reaches the preset takeover duration, the preset return conditions are met. When a user's command to end takeover is received, the preset return conditions are met. When operating parameters reach the target value, the preset return conditions are met. The determination result is triggered upon the fulfillment of any one of the above three conditions.

[0130] When preset return conditions are met, the energy management unit switches from a user-dominated state to a collaborative transition state. This switching is accomplished by the energy management unit's microprocessor updating state variables. The state variables change from values ​​representing the user-dominated state to values ​​representing the collaborative transition state.

[0131] After the switchover is completed, the energy management unit re-enters the control logic of the collaborative transition state and begins to control the energy storage device to perform charging and discharging according to the fusion result of the automatic scheduling command and the charging and discharging user command. The control weight of the charging and discharging user command gradually decreases over time from the initial value when entering the return collaborative transition state, while the control weight of the automatic scheduling command gradually increases over time from the initial value when entering the return collaborative transition state.

[0132] In some embodiments, the preset takeover duration is optionally determined dynamically based on the user permission level bound to the user terminal, and / or the preset takeover duration for high-privilege users is longer than that for low-privilege users, so that users with different permission levels can enjoy the longest takeover duration that matches their operation permissions after obtaining user-led control, and avoid premature interruption of the reasonable long-term operation of high-privilege users due to uniform duration restrictions.

[0133] In some embodiments, optionally, after receiving a user's termination of takeover instruction, the energy management unit also verifies whether the identity authentication signature carried in the user's termination of takeover instruction matches the legitimate key of the currently logged-in user terminal, and / or when the identity authentication signature verification fails, ignores the user's termination of takeover instruction and continues to maintain the user-dominated state, so that the system can resist forged control release signals sent by unauthorized terminals and prevent the control from being maliciously taken back prematurely due to communication hijacking or misoperation.

[0134] In some embodiments, optionally, such as Figure 3 As shown, the energy control method also includes:

[0135] Step S120: When the charging and discharging control parameters of the energy storage device meet the preset safety trigger conditions during the automatic scheduling state, collaborative transition state, or user-dominated state, switch from the current state to the safety fallback state.

[0136] Step S122: Under the safety fallback state, control the energy storage device to perform charging and discharging according to the safety operation constraints corresponding to the energy storage device, so that the operating state of the energy storage device is within the safe execution range;

[0137] Among them, the priority of the preset safety trigger conditions is higher than that of the charging / discharging user instructions and the automatic scheduling instructions;

[0138] The preset safety trigger conditions include at least the following: the energy storage device's state of charge reaches the upper limit of the safe state of charge; the energy storage device's state of charge reaches the lower limit of the safe state of charge; the charging power reaches the maximum charging power; the discharging power reaches the maximum discharging power; the grid-connected power reaches the upper limit of the grid-connected power; and the state of charge is lower than the minimum backup charge state.

[0139] In this embodiment, the charging and discharging control parameters of the energy storage device are monitored in real time during automatic scheduling, collaborative transition, or user-led states. When the parameters meet the preset safety trigger conditions, the device immediately switches from the current state to the safety fallback state, and controls the energy storage device according to safety operation constraints to ensure that the operating state is within the safe execution range.

[0140] The preset safety trigger conditions have higher priority than charging / discharging user commands and automatic scheduling commands.

[0141] Understandably, to address the issue of potential battery overcharging, over-discharging, or power exceeding limits due to conflicts between user commands and cloud platform commands, or under extreme operating conditions, this invention sets a safety fallback state as the highest priority. Once a preset safety trigger condition is triggered, user and cloud control commands are forcibly blocked, and charging and discharging are directly executed according to hardware safety constraints. This fundamentally avoids the risk of hardware damage caused by command conflicts, ensuring that energy storage devices can maintain physical safety even under extreme conditions.

[0142] In the actual operation of residential photovoltaic-storage systems, the operating state of energy storage devices during charging and discharging is affected by multiple factors, including battery physical characteristics, inverter hardware capabilities, grid connection regulations, and real-time user operations. In automatic dispatch mode, the energy management unit controls the energy storage devices according to automatic dispatch instructions issued by the cloud platform; in collaborative transition mode, it controls the energy storage devices according to the fusion result of automatic dispatch instructions and user charging / discharging instructions; and in user-dominated mode, it controls the energy storage devices according to user charging / discharging instructions.

[0143] In all three states described above, the charging and discharging control parameters of the energy storage device are directly determined by the corresponding commands or fusion results. However, regardless of the state, the operating status of the energy storage device must not exceed the bottom-line constraints of ensuring device safety and compliant system operation.

[0144] The charging and discharging control parameters include at least one of the following: the state of charge of the energy storage device, the real-time charging power, the real-time discharging power, and the grid-connected power.

[0145] During automatic scheduling, collaborative transition, or user-led operation, the energy management unit continuously collects the operating data of the energy storage device, extracts the charging and discharging control parameters, and compares the charging and discharging control parameters with the preset safety trigger conditions in real time.

[0146] The preset safety trigger conditions include at least one of the following: the energy storage device's state of charge reaches the upper limit of the safe state of charge; the energy storage device's state of charge reaches the lower limit of the safe state of charge; the charging power reaches the maximum charging power; the discharging power reaches the maximum discharging power; the grid-connected power reaches the upper limit of the grid-connected power; and the state of charge is lower than the minimum standby state of charge.

[0147] The safe state of charge upper limit is provided by the battery management unit and represents the highest state of charge value that the battery pack is allowed to reach by charging.

[0148] The safe state of charge limit is provided by the battery management unit and represents the lowest state of charge value that the battery pack is allowed to reach during discharge.

[0149] The maximum charging power is determined by the inverter hardware specifications and represents the maximum allowable charging power value of the energy storage device.

[0150] The maximum discharge power is determined by the inverter hardware specifications and represents the maximum allowable discharge power value of the energy storage device.

[0151] The upper limit of grid-connected power is determined by local grid connection regulations and inverter configuration, representing the maximum allowable power that energy storage devices can feed to the grid.

[0152] The minimum standby state of charge is preset by the user or the system and represents the minimum state of charge value that must be maintained to cope with sudden load or photovoltaic fluctuations.

[0153] When the energy management unit detects that the charging and discharging control parameters meet any of the aforementioned preset safety trigger conditions, it determines that there is a safety or compliance risk. At this time, the energy management unit immediately switches from the current automatic scheduling state, collaborative transition state, or user-led state to the safety fallback state.

[0154] The safety fallback state is a global highest priority state. Its triggering is not limited by the current control state, nor is it affected by the control intent carried in the charging / discharging user command or automatic scheduling command.

[0155] The preset safety trigger conditions have a higher priority than charging / discharging user commands and automatic dispatch commands. This priority relationship is determined by the basic safety principle of energy storage systems: commands from any source must not exceed the safety boundaries provided by the battery management unit, the power limits defined by the inverter hardware specifications, and the upper limit of grid-feed power required by grid connection regulations.

[0156] User charging and discharging instructions represent users' temporary energy consumption intentions, while automatic dispatch instructions represent the dispatch intentions generated by the cloud platform based on the optimal combination of economic efficiency and photovoltaic consumption. However, neither of them can override safety constraints.

[0157] In the safety fallback state, the energy management unit no longer executes automatic scheduling commands, responds to user commands, or maintains the fusion weight change process during the collaborative transition state. Instead, it controls the energy storage devices to perform charging and discharging entirely based on the safety operation constraints corresponding to the energy storage devices. The safety operation constraints are obtained by the energy management unit from the battery management unit, inverter hardware parameters, and grid-connected configuration parameters stored locally.

[0158] The energy management unit directly limits the charging power, discharging power, grid-connected power, and state of charge of the energy storage device according to safe operating constraints, ensuring that the energy storage device operates within a safe operating range. This safe operating range, defined by the safe operating constraints, is the parameter interval within which the energy storage device can operate safely.

[0159] As the local control core on the energy storage device side, the energy management unit is connected to the battery management unit through a communication bus to obtain the upper and lower limits of the safe state of charge, and is connected to the inverter through a control interface to obtain the maximum charging and discharging power value. It also reads the upper limit of grid-connected power and the minimum standby charge state of charge value from the local storage unit.

[0160] While the cloud platform considers user-defined upper and lower limits of state of charge when generating automatic dispatch commands, the safe upper and lower limits of state of charge provided by the battery management unit have the final say. When the control parameters implicit in the automatic dispatch commands issued by the cloud platform or the charging / discharging user commands sent by the user terminal conflict with the safety constraints of the safety fallback state, the energy management unit directly executes the control according to the safety constraints in the safety fallback state, thereby ensuring that the operating state of the energy storage device is always within the safe execution range.

[0161] In some embodiments, optionally, at least one threshold of the preset safety trigger condition determination parameters is associated with the current operating stage of the energy storage device, and / or the threshold of the determination parameter is dynamically calibrated based on at least one of the battery pack health status, cell consistency parameters and ambient temperature, so that the sensitivity of the safety fallback trigger can be adaptively adjusted according to the actual aging degree of the energy storage device and the operating environment, avoiding frequent false triggering due to fixed threshold after battery performance degradation or missed triggering under high temperature and high humidity conditions.

[0162] In some embodiments, optionally, the triggering priority of the preset safety triggering condition that the state of charge is lower than the minimum backup state of charge is higher than other safety triggering conditions, and / or when the state of charge is detected to be lower than the minimum backup state of charge, the energy management unit sends a low power alarm signal to the user terminal and cloud platform through the communication module while performing the safety fallback state, so that the system can prioritize ensuring the minimum backup power requirement when the battery reserve power is insufficient to support the critical load, and prompt the user or cloud to take timely measures to replenish power or reduce load through alarm notification.

[0163] In some embodiments, optionally, during the determination process of the preset safety trigger condition that the grid-connected power reaches the upper limit of the grid-connected power, the energy management unit also makes a comprehensive judgment based on the real-time grid voltage and frequency parameters reported by the metering device, and / or when the grid voltage or frequency exceeds the allowable range, even if the grid-connected power has not yet reached the upper limit of the grid-connected power, the safety backup state is triggered, so that the safety backup mechanism can cover the passive protection needs under abnormal grid operating conditions and expand the protection boundary for the safety of energy storage devices and grid interaction.

[0164] In some embodiments, optionally, after switching from the current state to the safety fallback state, the energy control method further includes: if it is detected that the currently effective charge and discharge control parameters no longer meet the preset safety triggering conditions, obtaining the control state before entering the safety fallback state; determining the target control state based on the control state before entering the safety fallback state, so as to make the energy storage device return to the automatic scheduling state, the cooperative transition state, or the user-led state.

[0165] In this embodiment, after the energy management unit enters the safety fallback state, it continuously monitors whether the currently effective charge and discharge control parameters no longer meet the preset safety trigger conditions. When the parameters are detected to have returned to normal, the control state before entering the safety fallback state is obtained and determined as the target control state, enabling the energy storage device to return to the automatic scheduling state, the collaborative transition state, or the user-led state, and restore normal energy control logic.

[0166] Understandably, after the safety hazard is eliminated, the system automatically restores to the control state before the trigger, rather than remaining in a safety fallback state or defaulting back to a single state. This ensures that temporary user operations can be seamlessly resumed when conditions permit, avoids unnecessary scheduling interruptions, and balances hardware security protection with system operation continuity.

[0167] During the period when the energy storage device is in a safety fallback state, the energy management unit directly controls the charging and discharging process of the energy storage device entirely based on the safety operation constraints. At this time, automatic scheduling commands, charging and discharging user commands, and the fusion weight change process in the collaborative transition state are all suspended.

[0168] The safety fallback state is a temporary protective state designed to remove energy storage devices from operating conditions that could jeopardize batteries, inverters, or grid connection compliance. Once the abnormal operating condition that triggered the safety fallback state is eliminated, the energy management unit needs to restore the energy storage device to its original control state before the safety fallback state was triggered, in order to continue executing the original charge and discharge scheduling logic.

[0169] The currently effective charge and discharge control parameters include at least one of the following: the current state of charge of the energy storage device, the real-time charging power, the real-time discharging power, and the grid-connected power.

[0170] The energy management unit continuously collects the aforementioned charge and discharge control parameters under the safety fallback state and compares the collected values ​​with the preset safety trigger conditions in real time. When all currently effective charge and discharge control parameters no longer meet the preset safety trigger conditions, the energy management unit determines that the operating state of the energy storage device has re-entered the safe execution range, triggering the exit condition of the safety fallback state.

[0171] The control state prior to entering the safety fallback state refers to the charging and discharging control mode that the energy management unit is executing before receiving the safety trigger signal that triggers the safety fallback state. The control state can be any one of the following: automatic scheduling state, cooperative transition state, or user-controlled state.

[0172] When the energy management unit enters the safety fallback state, it records the control state it was in before entering the safety fallback state in the local storage unit.

[0173] The recorded information includes status indicators and the currently active instruction parameters or weight values.

[0174] When the energy management unit detects that the currently effective charge / discharge control parameters no longer meet the preset safety trigger conditions, it reads the control state it was in before entering the safety fallback state from the local storage unit. Based on the read control state before entering the safety fallback state, the energy management unit determines the target control state. The target control state is the control state that the energy storage device should return to after exiting the safety fallback state.

[0175] If the control state before entering the safety fallback state is the automatic scheduling state, the target control state is determined to be the automatic scheduling state.

[0176] After the energy management unit exits the safety fallback state, it resumes controlling the energy storage device to perform charging and discharging according to the automatic scheduling instructions issued by the cloud platform. If the control state before entering the safety fallback state was a coordinated transition state, the target control state is determined to be the coordinated transition state.

[0177] After the energy management unit exits the safety fallback state, it resumes controlling the energy storage device to perform charging and discharging according to the fusion result of automatic dispatch instructions and user charging and discharging instructions, and resumes the control change process corresponding to the cooperative transition state. When the control state before entering the safety fallback state was the user-dominated state, the target control state is determined to be the cooperative transition state rather than directly returning to the user-dominated state.

[0178] In user-controlled mode, the energy storage device executes the charging and discharging user commands completely. If it returns directly from the safety fallback state to the user-controlled state, the control weight of the user commands will immediately be restored to the first preset threshold or a higher level, which may cause the energy storage device to approach the safety limit again when the safety boundary has just been restored.

[0179] Therefore, when the energy management unit is in a user-dominated control state before returning from and entering the safety fallback state, it determines the target control state as a collaborative transition state, so that the control right first goes through a transition process of gradual weight changes before deciding whether to finally enter the user-dominated state.

[0180] The Energy Management Unit (EMU) connects to the Battery Management Unit (BMU) via a communication bus to obtain real-time updates on the upper and lower limits of the safe state of charge (SOC). The EMU also connects to the inverter via a control interface to obtain real-time maximum charge / discharge power. Furthermore, the EMU reads the upper limit of grid-connected power and the minimum standby charge SOC from its local storage unit. These hardware devices, along with the local storage unit, collectively provide the EMU with the criteria for determining whether to enter or exit a safe backup state.

[0181] After determining the target control state, the energy management unit sends a power control signal corresponding to the target control state to the inverter, so that the energy storage device can resume charging and discharging operations according to the control logic corresponding to the target control state.

[0182] In some embodiments, optionally, during the process of controlling the energy storage device to return to the automatic scheduling state, the cooperative transition state, or the user-led state according to the target control state, the energy management unit gradually restores the charging and discharging power from the limit value in the safety fallback state to the set value corresponding to the target control state at a preset ramp rate, and / or the ramp rate is determined according to the cell temperature equalization state reported by the battery management module, so that the energy storage device can smoothly resume bearing the load after exiting the safety protection, and avoids the power sudden change triggering the protection again.

[0183] In some embodiments, optionally, during the safety fallback state, the energy management unit also detects whether it receives a new automatic scheduling instruction issued by the cloud platform or an instruction cancellation signal sent by the user terminal, and / or when an instruction cancellation signal is detected, updates the control state before entering the safety fallback state to the automatic scheduling state or the user-led state, so that the system can determine the target control state based on the latest instruction environment after the security threat is eliminated, and avoids the control logic from being out of touch with actual needs because the user has actively cancelled the operation but still returns to the user-led state.

[0184] In some embodiments, optionally, controlling the energy storage device to perform charging and discharging according to the fusion result of the automatic scheduling instruction and the charging and discharging user instruction includes: determining the automatic scheduling power according to the automatic scheduling instruction and determining the user's desired power according to the charging and discharging user instruction; weighting the automatic scheduling power and the user's desired power according to the control weight of the automatic scheduling instruction and the control weight of the charging and discharging user instruction to obtain the fused power; and controlling the energy storage device to perform charging and discharging according to the fused power.

[0185] In this embodiment, during the collaborative transition state, the automatic scheduling power is determined according to the automatic scheduling instructions issued by the cloud platform, the user's desired power is determined according to the charging and discharging user instructions sent by the user terminal, and the control weights of the automatic scheduling instructions and the charging and discharging user instructions are obtained. The automatic scheduling power and the user's desired power are weighted to obtain the fused power, and finally the energy storage device is controlled to perform charging and discharging according to the fused power.

[0186] Understandably, by weighting and fusing the power automatically scheduled in the cloud with the power expected by the user to obtain the fused power, the step change in the power setting value caused by the user command suddenly completely overriding the automatic scheduling command is avoided.

[0187] Even if the two types of commands conflict in power direction or value, the fused power can provide an intermediate execution value for a smooth transition between the two, eliminating the impact of manual intervention on the stability of equipment operation and preventing power surges from triggering battery overcurrent protection or abnormal operation of controllable loads.

[0188] During the collaborative transition state, the energy management unit needs to respond simultaneously to the automatic scheduling instructions issued by the cloud platform and the charging and discharging user instructions sent by the user terminal.

[0189] Automatic scheduling commands and charging / discharging user commands are generated independently, and they may differ in command format, numerical meaning, and time granularity. The energy management unit cannot directly superimpose and execute the two commands, nor can it simply choose one of the two commands to execute. Instead, it needs to convert the two commands into quantifiable power values, perform weighted fusion, and then convert the fusion result into actual power control signals for the inverter and battery management unit.

[0190] The automatic dispatch power is obtained by the energy management unit from the automatic dispatch instructions issued by the cloud platform. When generating automatic dispatch instructions, the cloud platform takes electricity purchase cost, battery degradation, photovoltaic power consumption, and backup power guarantee as comprehensive optimization objectives, and uses the target power value in the automatic dispatch instructions as the automatic dispatch power. A positive value of the automatic dispatch power indicates the charging direction, and a negative value indicates the discharging direction.

[0191] The magnitude of the automatic power dispatch reflects the cloud platform's economically optimal recommendation for the charging and discharging behavior of energy storage devices at the current moment.

[0192] The user's desired power is obtained by the energy management unit from the charge / discharge user commands sent by the user terminal. These commands contain the user-defined target power value or target operating state. The energy management unit uses the target power value carried in the charge / discharge user commands as the user's desired power.

[0193] A positive value for the user's desired power indicates that the user requires the energy storage device to charge, while a negative value indicates that the user requires the energy storage device to discharge.

[0194] The magnitude of the user's expected power reflects the user's temporary operational intention regarding the charging and discharging behavior of the energy storage device at the current moment.

[0195] The control weights of automatic dispatch commands and charging / discharging user commands are determined by the energy management unit based on the transition start time, transition target weight, and transition time parameters corresponding to the coordinated transition state. When transitioning from the automatic dispatch state to the coordinated transition state, the initial control weights of the automatic dispatch commands are higher, while the initial control weights of the charging / discharging user commands are lower. Conversely, when returning from the user-dominated state to the coordinated transition state, the initial control weights of the charging / discharging user commands are higher, while the initial control weights of the automatic dispatch commands are lower.

[0196] The control weights range from zero to one, and the sum of the two control weights remains one at any given time.

[0197] The energy management unit performs weighted processing on the automatic dispatch power and the user's desired power according to the control weight of the automatic dispatch command and the control weight of the charging and discharging user command.

[0198] The specific calculation method for weighted processing is as follows: multiply the automatic scheduling power and the control weight of the automatic scheduling command to obtain the automatic scheduling weighted component; multiply the user's expected power and the control weight of the charging and discharging user command to obtain the user's expected weighted component; add the two weighted components to obtain the fused power.

[0199] For example, the formula for calculating the fused power is: the fused power equals the product of the automatic scheduling power and the control weight of the automatic scheduling command, plus the product of the user's expected power and the control weight of the charging and discharging user command.

[0200] Fusion power is a scalar value; a positive value indicates that the energy storage device should charge the battery pack, while a negative value indicates that the energy storage device should discharge to the load or the grid. The value of fusion power combines the cloud's economically optimal recommendations with the user's temporary operational intentions, and fusion power changes continuously and smoothly with changes in control weights.

[0201] The energy management unit sends the combined power to the inverter via a control interface. The inverter adjusts the duty cycle of the DC-side or AC-side power conversion circuit based on the combined power value, thereby controlling the actual charging or discharging power output of the energy storage device. Simultaneously, the energy management unit sends the current limit information corresponding to the combined power to the battery management unit, which monitors and protects the battery pack's charging and discharging process in real time based on the current limit information.

[0202] The energy management unit reads the transition start time, transition target weight, and transition time parameters from the local storage unit, calculates the current control weight based on the current time, and then calls the weighted calculation program to obtain the merged power. The entire calculation process is repeated within each control cycle, with typical control cycles on the order of seconds or hundreds of milliseconds, to ensure that the merged power can be updated in real time with changes in the control weight.

[0203] In some embodiments, optionally, after obtaining the fused power, the energy management unit will also compare the fused power with the maximum allowable charge and discharge power reported in real time by the battery management module, and / or when the fused power exceeds the maximum allowable charge and discharge power, the fused power will be limited and clipped with the maximum allowable charge and discharge power as the upper limit, so that the fusion result passes through the last hardware security interception before being sent to the charge, discharge and inverter module for execution, ensuring that even if the weighted fusion algorithm is abnormal, it will not send an over-limit command to the power device.

[0204] In some embodiments, optionally, the automatic scheduling power and the user-expected power are weighted according to the control weight of the automatic scheduling instruction and the control weight of the charging / discharging user instruction, including: obtaining the transition start time, transition target weight, and transition time parameter corresponding to the cooperative transition state; determining the control weight corresponding to the current time according to the time interval between the current time and the transition start time and the transition time parameter, and according to a preset weight change rule; and weighting the automatic scheduling power and the user-expected power according to the control weight corresponding to the current time, so that the fused power changes smoothly with the change of control weight.

[0205] In this embodiment, during the coordinated transition state, the transition start time, transition target weight, and transition time parameter corresponding to the coordinated transition state are obtained. Based on the time interval between the current time and the transition start time and the transition time parameter, the control weight corresponding to the current time is determined according to the preset weight change rule. The automatic scheduling power and the user's expected power are weighted based on the control weight, so that the fused power changes smoothly with the change of control weight, thereby realizing the continuous transition of the charging and discharging power of the energy storage device on the time axis.

[0206] Understandably, by dynamically calculating the control weight through the transition start time, transition time parameters, and preset weight change rules, the fused power changes continuously and smoothly with the weight rather than abruptly, preventing the triggering of overcurrent protection or abnormal operation, and ensuring the continuous and stable operation of the energy storage device throughout the entire process of user intervention or return of control.

[0207] During the collaborative transition state, the energy management unit needs to perform weighted fusion of automatically dispatched power and user-expected power based on real-time changing control weights.

[0208] The control weights are not fixed values, but variables that evolve continuously over time. If the control weights change abruptly during state transitions, the combined power will jump accordingly, causing a sudden change in the inverter's output power and resulting in a current surge to the battery pack and power devices.

[0209] Therefore, the energy management unit must use a predictable, continuously changing approach to determine the control weight values ​​for each control cycle.

[0210] The transition start time is the point at which the cooperative transition state officially takes effect. When the energy management unit switches from automatic scheduling state to cooperative transition state, the transition start time is recorded as the system time of this switch.

[0211] When the energy management unit switches from a user-controlled state to a collaborative transition state, the transition start time is recorded as the system time at which the switch occurs. The transition start time is provided by the real-time clock circuit inside the energy management unit. After the system is powered on, the real-time clock circuit synchronizes with the cloud platform or via a network time protocol to ensure the accuracy of the timestamp.

[0212] The transition target weight is the value that the control weight ultimately needs to reach during the coordinated transition process. When transitioning from the automatic scheduling state to the coordinated transition state, the target value of the control weight for the charging and discharging user command is set to a first preset threshold, and the target value of the control weight for the automatic scheduling command is set to one minus the first preset threshold.

[0213] When returning from the user-dominated state to the collaborative transition state, the target value of the control weight of the automatic scheduling command is set to the second preset threshold, and the target value of the control weight of the charging and discharging user command is set to one minus the second preset threshold.

[0214] The transition target weight is read by the energy management unit from the local storage unit. The value stored in the local storage unit is sent by the cloud platform and written by the energy management unit when communication is normal.

[0215] The transition time parameter is the length of time required for the control weight to change from its initial value to the target weight. The value of the transition time parameter determines how quickly the weight changes. A transition time parameter that is too short will cause the merged power to change too quickly, while a transition time parameter that is too long will cause a perceived delay in the control response for the user. The energy management unit stores the transition time parameter in its local storage unit. The value of the transition time parameter is determined by the cloud platform based on the scale of the photovoltaic system connected to the energy storage device, the battery capacity, and the user's operating habits, and then distributed to the energy management unit.

[0216] The current time is read by the real-time clock circuit of the energy management unit at the beginning of each control cycle. The energy management unit calculates the time interval between the current time and the transition start time, divides the time interval by the transition time parameter, and obtains a normalized time progress ratio. When the time progress ratio is less than one, the current time is in the transition process. When the time progress ratio reaches or exceeds one, the control weight has reached the transition target weight, and the control weight will no longer change in subsequent control cycles.

[0217] The preset weight change rules define the mathematical mapping relationship between the control weight and the time progress ratio. These rules include at least one of exponential, linear, and piecewise change rules. Under the linear change rule, the control weight is linearly proportional to the time progress ratio. Under the exponential change rule, the control weight approaches the transition target weight in an exponential function manner, changing slowly in the initial stage and accelerating near the end of the transition, or vice versa.

[0218] Under the segmented change rule, the transition process is divided into multiple time intervals, each employing a different rate of change or a different mathematical function. The energy management unit selects the corresponding calculation program based on the rule identifier stored in the local storage unit, inputs the time progress ratio into the calculation program, and outputs the control weight value corresponding to the current moment.

[0219] The control weights of the automatic dispatch command and the charge / discharge user command are calculated and stored independently in each control cycle. The energy management unit calls the weighted processing program to multiply the automatic dispatch power by the control weight of the automatic dispatch command to obtain the automatic dispatch weighted component, and multiply the user's desired power by the control weight of the charge / discharge user command to obtain the user's desired weighted component. The two weighted components are then added together to obtain the merged power.

[0220] Since the control weights change only slightly in each control cycle, the value of the fused power also changes only slightly, thus ensuring that the actual charging and discharging power of the energy storage device remains continuously and smoothly changed throughout the entire coordinated transition process, avoiding the impact on the inverter power switching devices and the internal chemical system of the battery pack.

[0221] In each control cycle, the microprocessor of the energy management unit sequentially executes the following steps: reading the current time, calculating the time interval, looking up a table or performing a calculation to obtain the control weight, reading the automatically dispatched power and the user's desired power, performing weighted multiplication and addition operations, and writing the merged power into the inverter's control register. These steps are repeated cyclically throughout each control cycle of the coordinated transition state until the control weight reaches the transition target weight or the state changes again.

[0222] In some embodiments, optionally, during the collaborative transition state, the energy management unit also records the control weight, fusion power and actual output charging and discharging power corresponding to the current moment as a transition process log, and / or the transition process log is reported to the cloud platform to evaluate the actual smoothing effect of the weight change rule, so that the cloud platform can remotely optimize and update the shape parameters and transition time parameters of the preset weight change rule based on the actual transition data, forming a continuous self-improvement closed loop of end-cloud collaboration.

[0223] In some embodiments, the preset weight change rule may optionally include at least one of the exponential change rule, the linear change rule, and the piecewise change rule. The preset weight change rule is used to make the control weight gradually change from the current weight to the transition target weight according to the corresponding change trajectory.

[0224] In this embodiment, during the cooperative transition state, the energy management unit selects at least one of three rules based on the current operating conditions, so that the control weight of the automatic scheduling command and the control weight of the charging and discharging user command change gradually and continuously from the current weight value to the transition target weight according to the corresponding change trajectory, driving the fused power to transition along a smooth trajectory rather than abruptly changing.

[0225] Understandably, by providing multiple weight change rules, including exponential, linear, and piecewise, the system can select the most suitable change trajectory based on actual operating conditions such as battery state of charge, ensuring that the control weights can smoothly transition to the target value under various operating conditions, and further guaranteeing the continuous and stable operation of energy storage equipment under different operating conditions.

[0226] During the coordinated transition state, the energy management unit needs to ensure that the control weights of the automatic dispatch commands and the control weights of the charging and discharging user commands evolve continuously over time. The way the control weights change directly determines the trajectory of the merged power over time. If the trajectory is not chosen properly, the merged power may experience a sudden change in the rate of change at the beginning or end of the transition, which in turn causes a jump in the derivative of the inverter output current, adversely affecting the chemical reaction rate of the cells inside the battery pack and the stress on the inverter's power switching devices.

[0227] Therefore, the energy management unit needs to preset a variety of selectable weight change rules to adapt to the hardware response characteristics of different energy storage systems and the needs of different user operation scenarios.

[0228] The preset weight change rules are stored in the local storage unit of the energy management unit. Each time the energy management unit enters a coordinated transition state, it reads the identifier information of the preset weight change rule from the local storage unit and calls the corresponding calculation program based on the identifier information to determine the control weight value for each control cycle. The preset weight change rules include at least one of exponential change rules, linear change rules, and piecewise change rules.

[0229] The exponential change rule refers to the mathematical relationship where the control weights change over time according to an exponential function. Under the exponential change rule, there is an exponential mapping between the control weights and the normalized time progress ratio. The normalized time progress ratio is obtained by dividing the time interval between the current moment and the transition start moment by the transition time parameter.

[0230] The exponential change rule causes the control weights to change slowly in the initial stage of the transition process. As the normalized time progress ratio increases, the rate of change gradually accelerates, and the control weights rapidly approach the transition target weight near the end of the transition. This trajectory is suitable for scenarios where users want the energy storage device to maintain its original operating state for a longer period before actual power changes are perceived, and then quickly switch over at the end of the transition. Conversely, an exponential form can be used, where the control weights change rapidly in the initial stage and gradually slow down as they approach the transition target weight. This is suitable for scenarios where the energy storage device needs to respond quickly to user commands but requires slow convergence before final stabilization. The specific time constant of the exponential change rule is determined by the cloud platform based on the battery chemistry type of the energy storage device and the dynamic response capability of the inverter, and then distributed to the energy management unit for storage.

[0231] The linear change rule refers to a linear relationship between the control weight and the normalized time progress ratio. Under the linear change rule, the control weight equals the initial weight plus the product of the difference between the transition target weight and the initial weight and the normalized time progress ratio. The linear change rule ensures that the control weight changes at a constant rate throughout the transition. The change in merged power over time also exhibits a linear shape with a constant slope. The linear change rule has the lowest computational complexity; the energy management unit's microprocessor only needs to perform one multiplication and one addition in each control cycle to update the control weight, minimizing computational resource consumption. The linear change rule is suitable for conventional scenarios where energy storage equipment hardware response bandwidth is relatively wide and users have no particular preference for the power change rate.

[0232] The segmented change rule refers to dividing the entire transition time into multiple consecutive time intervals, with different change functions or different change rates used in each time interval.

[0233] The specific division methods for the segmented transition rules include: using a linear segment with a slower rate of change in the early stage of the transition, a linear segment with a faster rate of change in the middle stage of the transition, and a linear segment with a slower rate of change again in the later stage of the transition; or using an exponential form in the early stage of the transition and switching to a linear form in the later stage. The division nodes of the segmented transition rules are determined by at least one preset time ratio threshold, and the value of the time ratio threshold is stored in the local storage unit. The segmented transition rules are suitable for scenarios where energy storage devices have intermediate state constraints during the transition process, such as waiting for the battery management unit to complete a balancing operation halfway through the transition, or prioritizing photovoltaic absorption before gradually responding to user intentions at a certain time point. The segmented transition rules enable the energy management unit to flexibly adjust the pace of control handover at different stages of the transition process.

[0234] Regardless of whether exponential, linear, or piecewise change rules are used, the control weights gradually change from the current weights to the transition target weights according to the corresponding change trajectory.

[0235] The current weight is the initial value of the control weight when entering the cooperative transition state. When transitioning from the automatic scheduling state to the cooperative transition state, the initial value of the control weight of the automatic scheduling command is higher, while the initial value of the control weight of the charging and discharging user command is lower.

[0236] When transitioning from a user-dominated state to a collaborative transition state, the initial control weights for charging and discharging user commands are higher, while the initial control weights for automatic scheduling commands are lower. This gradual change means that the control weights undergo only minor incremental adjustments within each control cycle, preventing jumps from one extreme value to another within the same cycle. This gradual change characteristic ensures that the variation in fused power between any two adjacent control cycles is limited to an allowable range, thus maintaining a continuous and smooth trajectory for the actual charging and discharging power of the energy storage device throughout the entire collaborative transition process.

[0237] The microprocessor in the energy management unit contains the calculation program code corresponding to each preset weight change rule. The local storage unit stores the rule selection identifier, transition start time, transition target weight, transition time parameter, and time ratio threshold under the segmented change rule. In each control cycle, the microprocessor reads the rule selection identifier, jumps to the corresponding calculation program entry point, inputs the normalized time progress ratio, and outputs the control weight value corresponding to the current moment.

[0238] The output results are simultaneously written to the control weight storage variables of the automatic scheduling instruction and the charge / discharge user instruction, for the weighted processing program to call in subsequent steps.

[0239] In some embodiments, optionally, the slope of the weight change corresponding to each time interval in the segmented change rule is dynamically updated according to the real-time operating status of the energy storage device within the time interval, and / or the real-time operating status includes the temperature uniformity of the battery pack cells, the power margin of the grid connection and feed, and the instantaneous power demand of the controllable load, so that the segmented change rule can adaptively adjust the transition rate of each stage according to the actual operating conditions of the energy storage device, and avoid the transition being too fast or too slow due to the fixed slope being unable to match the changes in operating conditions.

[0240] In some embodiments, the energy control method may optionally further include: acquiring safe operating parameters corresponding to the energy storage device; determining the safe execution range corresponding to the energy storage device based on the safe operating parameters; detecting whether the operating state parameters of the energy storage device exceed the safe execution range during the process of controlling the energy storage device to perform charging and discharging according to the fusion result; if any operating state parameter is detected to exceed the safe execution range, adjusting the fusion result to make the adjusted charging and discharging control parameters within the safe execution range; and controlling the energy storage device to perform charging and discharging according to the adjusted charging and discharging control parameters.

[0241] In this embodiment, during the process of controlling the energy storage device to perform charging and discharging according to the fusion result of the automatic scheduling command and the charging and discharging user command, the operating status parameters of the energy storage device are detected in real time to see if they exceed the safe execution range. When any operating status parameter is detected to exceed the safe execution range, the fusion result is adjusted to restrict the charging and discharging control parameters so that the adjusted charging and discharging control parameters are constrained within the safe execution range, and finally the energy storage device is controlled to perform charging and discharging according to the adjusted charging and discharging control parameters.

[0242] Understandably, when the fusion result exceeds the safety boundary due to instruction conflict or calculation abnormality, the energy management unit automatically limits and trims the charging and discharging control parameters to bring them back to the safe execution range. This prevents hardware safety risks such as power over-limit and battery overcharging and over-discharging caused by temporary user operations, and ensures the continuous and stable operation of the energy storage device under all operating conditions.

[0243] In the coordinated transition state, the energy management unit controls the energy storage device to perform charging and discharging according to the fusion result of automatic scheduling instructions and charging and discharging user instructions.

[0244] The fusion result is obtained by weighting the automatically scheduled power and the user's expected power, and its value reflects the combination of the cloud's economically optimal suggestion and the user's temporary operational intention.

[0245] However, the charging and discharging power values ​​represented by the fusion result are not inherently guaranteed to be within the hardware capabilities and safe operating boundaries of the energy storage device. If the absolute value of the fusion power exceeds the maximum power allowed by the inverter, or if the charging direction corresponding to the fusion power causes the battery pack's state of charge to approach the safe upper limit, or if the discharging direction corresponding to the fusion power causes the battery pack's state of charge to approach the safe lower limit, or if the discharging power exceeds the upper limit of the grid-connected power supply, the energy storage device will face the risk of hardware damage, battery thermal runaway, or violation of grid connection regulations.

[0246] Therefore, the energy management unit must perform safety boundary verification and necessary limitation adjustments on the fusion results before executing the fusion results.

[0247] Safe operating parameters are fundamental data used to define the safe operating boundaries of energy storage devices. Safe operating parameters include at least one of the following: upper limit of safe state of charge, lower limit of safe state of charge, maximum charging power, maximum discharging power, upper limit of grid-connected power, and minimum standby state of charge.

[0248] The upper and lower limits of the safe state of charge (SPC) are provided to the energy management unit (EMU) in real time by the battery management unit (BMU) via the communication bus. The BMU calculates the current allowable SPC limits based on cell chemical characteristics, battery pack temperature distribution collected by temperature sensors, and historical cycle counts, and sends these limits to the EMU via periodic data frames. The maximum charging and discharging power are determined by the inverter hardware specifications. During system initialization, the EMU reads these power limits from the inverter's device description information and stores them in its local storage unit. The grid-connected power limit is jointly determined by local grid-connection regulations and inverter configuration. The EMU reads the pre-configured grid-connected limit value from its local storage unit. The minimum standby charge SPC is preset by the user or the system and stored in the local storage unit to ensure that the energy storage device retains sufficient charge to maintain power supply to critical loads during sudden load increases or a sudden drop in photovoltaic power generation.

[0249] The energy management unit determines the safe operating range based on safe operating parameters. The safe operating range is a multi-dimensional parameter range that covers the allowable range of state of charge, the allowable range of charging power, the allowable range of discharging power, and the allowable range of grid-connected power.

[0250] The lower bound of the allowable state of charge (SOC) range is the upper limit of the safe SOC range, and the upper bound is the upper limit of the safe SOC range. It must also ensure that there is still a margin after subtracting the amount of charge corresponding to the minimum reserve charge state from the current SOC range. The upper bound of the allowable charging power range is the maximum charging power, and the lower bound is zero. The upper bound of the allowable discharging power range is the maximum discharging power, and the lower bound is zero. The upper bound of the allowable grid-connected power range is the upper limit of the grid-connected power, and the lower bound is zero. The energy management unit combines the upper and lower bound values ​​of the above dimensions to form the safe execution range and stores it in the local storage unit for subsequent detection steps.

[0251] During the process of controlling the energy storage device to perform charging and discharging according to the fusion results, the energy management unit (EMU) performs operational status parameter detection in each control cycle. These operational status parameters include the current state of charge (SOC) value of the energy storage device, real-time charging power value, real-time discharging power value, and grid-connected power supply value. The SOC value is periodically read by the EMU from the battery management unit (BMU) via the communication bus. The real-time charging and discharging power values ​​are obtained by the EMU from the inverter's power sampling circuit. The grid-connected power supply value is read by the EMU from the grid metering module. The EMU compares the acquired operational status parameters one by one with the corresponding dimensions within the safe execution range.

[0252] The detection process distinguishes the direction of the fused power. When the fused power is positive, indicating the charging direction, the energy management unit detects whether the current state of charge value has reached or exceeded the safe state of charge limit, and whether the fused power value has exceeded the maximum charging power.

[0253] When the fusion power is negative, indicating the direction of discharge, the energy management unit detects whether the current state of charge value has reached or fallen below the safe state of charge limit, whether the absolute value of the fusion power has exceeded the maximum discharge power, whether the absolute value of the fusion power has exceeded the grid-connected power limit, and whether the state of charge after discharge will fall below the minimum standby state of charge.

[0254] If any operating status parameter is detected to be outside the safe execution range, the energy management unit will adjust the limits on the fusion results.

[0255] The specific adjustment methods are as follows: When the merged power is positive and exceeds the maximum charging power, the merged power is limited to the maximum charging power. When the current state of charge (SOC) value reaches the upper limit of the safe SOC and the merged power is positive, the merged power is limited to zero. When the merged power is negative and its absolute value exceeds the maximum discharge power, the merged power is limited to the negative maximum discharge power. When the merged power is negative and its absolute value exceeds the upper limit of the grid-connected power supply, the merged power is limited to the negative upper limit of the grid-connected power supply. When discharging would cause the SOC to fall below the minimum standby charge SOC, the maximum allowable discharge capacity is calculated based on the difference between the minimum standby charge SOC and the current SOC, and the merged power is limited to the corresponding discharge capacity or limited to zero. The charge and discharge control parameters obtained after the adjustment are within the safe execution range.

[0256] The energy management unit sends the target power value, obtained after adjusting and limiting the fusion results, to the inverter via the control interface. The inverter adjusts the duty cycle of its internal power conversion circuit based on the target power value, controlling the energy storage device to perform charging or discharging operations according to the adjusted charging and discharging control parameters.

[0257] Simultaneously, the energy management unit sends the current limit information corresponding to the adjusted target power value to the battery management unit, which then monitors the charging and discharging process of the battery pack in real time based on the current limit information. The entire limit adjustment process is executed independently in each control cycle, ensuring that the actual operating parameters of the energy storage device at any time do not exceed the boundaries of the safe operating range.

[0258] In some embodiments, optionally, after controlling the energy storage device to perform charging and discharging according to the adjusted charging and discharging control parameters, the energy management unit also continuously monitors whether all operating status parameters have fallen back to the safe execution range, and / or after all operating status parameters have remained within the safe execution range for a preset stable period of time, gradually relaxes the adjustment range of the fusion result, so that the system can gradually reuse the original power setting value in the fusion result after the safety conditions are restored, rather than being in a limited state for a long time, affecting the smoothness of the transition.

[0259] In some embodiments, optionally, when any operating state parameter is detected to exceed the safe execution range, the fusion result is restricted and adjusted so that the adjusted charge and discharge control parameters are within the safe execution range. This includes: when the user's expected power corresponding to the charge and discharge user command conflicts with the safe execution range, restricting the user's expected power according to the safe execution range to determine the restricted power; weighting the automatic scheduling power and the restricted power according to the control weight of the automatic scheduling command and the control weight of the charge and discharge user command to determine the restricted fusion power; and controlling the energy storage device to perform charge and discharge according to the restricted fusion power so that the energy storage device executes the charge and discharge user command while meeting the safe execution range.

[0260] In this embodiment, when the user's desired power corresponding to the charging / discharging command is detected to exceed the safe execution range, the user's desired power is first limited according to the safe execution range. The automatically scheduled power and the limited power are then weighted and fused according to their respective control weights to obtain the limited fused power. The energy storage device is controlled to perform charging and discharging according to the limited fused power, so that the energy storage device responds to the user command as much as possible while meeting the safe execution range.

[0261] Understandably, the user's expected power is individually limited before being included in the weighted processing. This prevents the user's over-limit command from exceeding the limit and the cloud command from exceeding the limit and breaking the safety boundary. It also preserves the effective execution of the user's operation intention within the limited range, and maintains the participation of the automatic scheduling command in the fusion process. This achieves the optimal balance between security protection, user demand response and cloud scheduling continuity.

[0262] In the coordinated transition state, the energy management unit controls the energy storage device to perform charging and discharging according to the fusion result of automatic scheduling instructions and charging / discharging user instructions. When the operating status parameters of the energy storage device are detected to exceed the safe execution range, the fusion result needs to be adjusted to limit it.

[0263] The goal of the limit adjustment is to bring the adjusted charge and discharge control parameters back into the safe execution range without switching the current control state.

[0264] This adjustment process follows a unified arbitration priority that prioritizes security constraints, followed by explicit user intent, and then machine-side economic optimization.

[0265] Specifically, when a conflict is detected, priority is given to ensuring that the safe execution range is not breached. Without breaching the safe execution range, the temporary operational intent of the user represented by the charging and discharging user command is preserved as much as possible. The charging and discharging user command is only trimmed as necessary within the safe boundary, rather than simply rejecting or discarding the charging and discharging user command.

[0266] The user's desired power corresponding to the charge / discharge user command is obtained by the energy management unit from the charge / discharge user command sent by the user terminal. The user's desired power reflects the charging or discharging target that the user requires the energy storage device to perform at the current moment. The safe execution range is determined by the energy management unit based on safe operating parameters, covering the allowable range of state of charge, the allowable range of charging power, the allowable range of discharging power, and the allowable range of grid-connected power.

[0267] A conflict between user-expected power and safe execution range refers to a situation where the value of user-expected power exceeds the allowable boundary of the corresponding dimension within the safe execution range.

[0268] Specific scenarios of conflict include: the user's expected power is positive and exceeds the maximum charging power; the user's expected power is positive and the current state of charge has reached the upper limit of the safe state of charge; the user's expected power is negative and the absolute value exceeds the maximum discharge power; the user's expected power is negative and the absolute value exceeds the upper limit of the grid-connected power supply; and the user's expected power is negative and the state of charge after discharge will be lower than the minimum standby state of charge.

[0269] In the event that the user's desired power corresponding to the charging / discharging command conflicts with the safe execution range, the energy management unit limits the user's desired power according to the safe execution range and determines the limited power.

[0270] The specific method of limitation is as follows: when the user's expected power is positive and exceeds the maximum charging power, the user's expected power is limited to the maximum charging power, and the limited power is equal to the maximum charging power.

[0271] When the user's desired power is positive and the current state of charge has reached the upper limit of the safe state of charge, the user's desired power is limited to zero, and the limited power is equal to zero.

[0272] When the user's expected power is negative and its absolute value exceeds the maximum discharge power, the user's expected power will be limited to the negative maximum discharge power, and the limited power will be equal to the negative maximum discharge power.

[0273] When the user's expected power is negative and its absolute value exceeds the grid-connected power limit, the user's expected power will be limited to the negative grid-connected power limit, and the limited power will be equal to the negative grid-connected power limit.

[0274] When the user's desired power is negative and the state of charge after discharge will be lower than the minimum reserve charge state of charge, the maximum allowable discharge capacity is calculated based on the difference between the minimum reserve charge state of charge and the current state of charge. The user's desired power is then limited to the negative power of the corresponding discharge capacity, and the limited power is equal to the negative power.

[0275] The power limit determined by the above limiting steps is the projected value of the user's expected power within the safe operating range.

[0276] After obtaining the power limit, the energy management unit performs weighted processing on the automatic dispatch power and the power limit according to the control weight of the automatic dispatch command and the control weight of the charging and discharging user command to determine the power limit fusion.

[0277] The control weights of the automatic scheduling command and the charging / discharging user command are real-time values ​​calculated using preset weight change rules during the current collaborative transition state.

[0278] For example, the automatic scheduling power and the control weight of the automatic scheduling command are multiplied to obtain the automatic scheduling weighted component, the limit power and the control weight of the charging and discharging user command are multiplied to obtain the user expectation weighted component, and the two weighted components are added to obtain the limit fused power.

[0279] Since the power limit is the value of the user's expected power after being clipped to a safety boundary, the power limit for fusion is naturally within the safe execution range.

[0280] The energy management unit sends the limit on the combined power to the inverter via the control interface. The inverter adjusts the duty cycle of its internal power conversion circuit based on the limit on the combined power, controlling the energy storage device to perform charging or discharging operations according to the limit.

[0281] The energy management unit sends the current limit information corresponding to the power limit to the battery management unit, and the battery management unit monitors and protects the charging and discharging process of the battery pack in real time based on the current limit information.

[0282] Through the above steps, the energy storage device executes user charging and discharging commands within safe operating limits. Temporary operational intentions issued by the user through the user terminal receive maximum response within safety boundaries, rather than being directly rejected by the system. The energy management unit can also send limitation reason information to the user terminal via the communication network, allowing the user to understand why the charging and discharging user commands have been partially adjusted.

[0283] In some embodiments, optionally, before limiting the user's expected power according to the safe execution range, the energy management unit also detects whether the charging and discharging user command carries an emergency flag, and / or when an emergency flag is detected, the upper limit threshold of the safe execution range is temporarily raised or the lower limit threshold is temporarily lowered to expand the allowable range of the limited power, so that user operations with emergency attributes can obtain a larger executable power space under the condition of reduced safety margin, and avoid completely blocking the user's energy demand in emergency conditions due to the fixed safe execution range.

[0284] In some embodiments, optionally, after controlling the energy storage device to perform charging and discharging according to the limited fusion power, the energy management unit also detects whether the difference between the limited fusion power and the user's expected power is continuously zero, and / or when the difference between the limited fusion power and the user's expected power is continuously zero for a preset duration, the energy management unit determines that the user's expected power has been completely absorbed by the safety limit, triggers the reporting of an event log to the cloud platform that the user's command has been completely intercepted, so that the cloud can evaluate the user terminal's command input habits based on the interception frequency and remotely push safe operation guidance information.

[0285] In some embodiments, optionally, the automatic scheduling instruction is generated by the cloud platform and sent to the energy management unit. The energy control method further includes: acquiring the operating data of the energy storage device and sending the operating data to the cloud platform so that the cloud platform determines the automatic scheduling instruction based on the operating data; the energy management unit receives the automatic scheduling instruction sent by the cloud platform and controls the energy storage device to perform charging and discharging according to the automatic scheduling instruction.

[0286] In this embodiment, the energy management unit acquires the operating data of the energy storage device and sends it to the cloud platform, so that the cloud platform can dynamically determine automatic scheduling instructions based on the latest operating data. The energy management unit receives and executes the instructions to control the charging and discharging of the energy storage device.

[0287] Understandably, by placing computationally intensive prediction and scheduling calculations on the cloud platform, the energy management unit only needs to be responsible for data collection and local execution, which leverages the advantages of the cloud platform's global data perspective and strong computing power while reducing local hardware costs.

[0288] In the edge-cloud collaborative architecture of residential photovoltaic-storage systems, energy storage devices establish a two-way communication connection with the cloud platform through an energy management unit. The cloud platform has stronger computing power and a global data perspective, enabling it to perform multi-objective optimization calculations based on photovoltaic power generation forecast data, household load forecast data, and electricity price data, and generate automatic scheduling instructions for future periods.

[0289] The energy management unit (EMU), as the local control core of the energy storage device, is responsible for receiving and executing automatic dispatch commands issued by the cloud platform, and also has the ability to collect local operating data of the energy storage device. The generation and issuance of automatic dispatch commands rely on a closed-loop data interaction between the EMU and the cloud platform.

[0290] The operational data of the energy storage device is acquired by the energy management unit through an internal communication interface and sampling circuit. This operational data includes the current state of charge (SOC) of the energy storage device, battery pack temperature distribution data, maximum allowable charging power, maximum allowable discharging power, real-time inverter output power, grid-connected power, real-time photovoltaic array power generation, and SOC upper and lower limits set by the user through the user terminal. The energy management unit packages the acquired operational data according to a predefined communication protocol and sends it to the cloud platform via the communication network.

[0291] Communication networks include wireless network connections or wired network connections.

[0292] When communication is normal, the energy management unit periodically uploads operating data to the cloud platform. The upload cycle is configured by the cloud platform and distributed to the energy management unit for storage in the local storage unit.

[0293] After receiving the operational data from the energy storage devices, the cloud platform determines the automatic scheduling control parameters by combining them with locally stored scheduling data. This scheduling data includes photovoltaic power generation forecasts, household load forecasts, and electricity price data.

[0294] The photovoltaic power generation forecast data is obtained by the cloud platform based on meteorological data sources and historical power generation curves of the photovoltaic array.

[0295] Household load forecast data is obtained by the cloud platform based on users' historical electricity consumption records and seasonal electricity consumption patterns.

[0296] Electricity price data is obtained by the cloud platform from the electricity market data interface. Negative electricity price data is allowed to reflect negative electricity price scenarios in the electricity market.

[0297] The cloud platform takes electricity purchase cost, battery degradation, photovoltaic consumption and backup power guarantee as comprehensive optimization goals. It obtains the final charge constraint by taking the intersection of the upper and lower limits of the state of charge set by the user in the operation data and the upper and lower limits of the safe state of charge provided by the battery management unit, and incorporates the final charge constraint into the equipment operation constraint.

[0298] Based on this, the cloud platform determines the automatic scheduling and control parameters for each time step through optimized calculations. The automatic scheduling and control parameters include the target charging and discharging power value and time allocation information.

[0299] The cloud platform generates automatic scheduling instructions based on automatic scheduling control parameters. These instructions contain the target power value, power direction identifier, and execution time window for one or more future time steps.

[0300] The cloud platform generates automatic dispatch instructions and sends them to the energy management unit via the communication network. The energy management unit receives the automatic dispatch instructions from the cloud platform through the communication network interface and stores them in its local storage unit.

[0301] In automatic scheduling mode, the energy management unit controls the energy storage equipment to perform charging and discharging according to the automatic scheduling instructions issued by the cloud platform.

[0302] The energy management unit reads the target power value from the automatic scheduling command from the local storage unit, sends the target power value to the inverter through the control interface, and controls the inverter to adjust the duty cycle of the internal power conversion circuit so that the energy storage device outputs the corresponding charging power or discharging power.

[0303] At the same time, the energy management unit sends the current limit information corresponding to the target power value to the battery management unit, which then monitors and protects the charging and discharging process of the battery pack in real time based on the current limit information.

[0304] The energy management unit, acting as both the data acquisition and command execution end, is responsible for feeding back the real-time operating status of the energy storage equipment to the cloud platform and faithfully executing the automatic scheduling commands issued by the cloud platform.

[0305] The cloud platform, acting as a remote optimization decision-making hub, is responsible for generating automatic scheduling commands using global data and optimization algorithms. The energy management unit (EMU) and the cloud platform form a closed-loop link for data uplink and command downlink via a communication network. The battery management unit (BMU) in the energy storage device provides the EMU with safe upper and lower limits of state of charge (SOC) and real-time SOC values ​​via a communication bus. The inverter receives the target power value from the EMU through its control interface and executes the corresponding power conversion operation. The user terminal interacts with the EMU via the communication network, allowing users to view the execution status of automatic scheduling commands or send temporary operation commands.

[0306] In some embodiments, the energy control method may optionally further include: the energy management unit reporting device identifier, scene identifier and device capability information to the cloud platform; the energy management unit receiving an adapted automatic scheduling instruction issued by the cloud platform based on the device capability information; wherein the device capability information includes at least: whether it supports AC charging power setting, whether it supports inverter discharge power setting, whether it supports photovoltaic charging power limitation, whether it supports grid-connected power upper limit setting, whether it supports target state of charge setting and whether it supports local policy caching.

[0307] In this embodiment, the energy management unit reports device identifiers, scene identifiers, and device capability information to the cloud platform. The cloud platform adapts and modifies the automatic scheduling instructions based on the reported device capability information, and the energy management unit receives and executes the adapted automatic scheduling instructions.

[0308] Understandably, by reporting equipment capability information, the cloud platform can perceive the actual hardware configuration of the energy management unit, dynamically trim parameters in the automatic scheduling instructions, and only issue control instructions supported by the equipment, thereby improving the compatibility and reliability of the end-cloud collaborative scheduling system.

[0309] In the edge-cloud collaborative architecture of residential photovoltaic-storage systems, there are significant differences in hardware configuration, installation scenarios, and functional support among different users' energy storage devices.

[0310] Some energy storage devices support fine-tuning of AC-side charging power, while others only support start-stop control. Some energy storage devices are connected to photovoltaic arrays and support photovoltaic charging power limiting, while others are not equipped with photovoltaic modules. Grid connection regulations in some regions allow a certain limit on grid feeding power, while grid connection regulations in other regions completely prohibit grid feeding.

[0311] When generating automatic dispatch instructions, the cloud platform must ensure that the content of the instructions is compatible with the actual hardware capabilities of the energy storage equipment and local compliance requirements. Otherwise, the target power value in the automatic dispatch instructions may be unavailable due to equipment incompatibility or may be rejected by the energy management unit due to violation of local regulations.

[0312] Therefore, the energy management unit needs to report the device identifier, scene identifier, and device capability information to the cloud platform during the system initialization phase or when the configuration is changed, so that the cloud platform can generate adapted automatic scheduling instructions accordingly.

[0313] The device identifier is a unique identification number for energy storage devices on the cloud platform. The device identifier is assigned by the cloud platform when the energy storage device is first connected to the system, or it is written to the local storage unit by the energy management unit at the factory and reported to the cloud platform during the first communication. The device identifier enables the cloud platform to accurately associate received operational data, device capability information, and issued automatic scheduling commands with specific energy storage devices, avoiding data confusion between multiple energy storage devices.

[0314] Scene identifiers represent the installation and usage scenarios of energy storage devices. Scene identifiers include different types such as residential single-phase systems, residential three-phase systems, small-scale industrial and commercial systems, or off-grid systems.

[0315] The load characteristics, grid connection requirements, and scheduling strategies differ fundamentally in different scenarios.

[0316] For example, the resolution of inverter discharge power settings differs between single-phase and three-phase systems, and there is no concept of grid-connected power limit in off-grid systems. The energy management unit reports the scenario identifier along with the device identifier to the cloud platform, and the cloud platform selects the corresponding scheduling model and constraint template based on the scenario identifier.

[0317] Equipment capability information is a set of data describing the extent to which the hardware functions of energy storage devices are supported. Equipment capability information includes at least whether it supports AC charging power settings, whether it supports inverter discharge power settings, whether it supports photovoltaic charging power limits, whether it supports grid-connected feeder power limit settings, whether it supports target state of charge settings, and whether it supports local policy caching.

[0318] Each capability information is represented as a Boolean value or an enumerated value indicating whether it is supported or not. The energy management unit determines the specific value of each of the above capability information by reading the inverter device description information, the battery management unit communication protocol version, and the function configuration flags in the local storage unit.

[0319] Whether AC charging power setting is supported indicates whether the energy management unit can send a target value for AC charging power to the inverter. If supported, the cloud platform can include the precise AC charging power value in the automatic scheduling command. If not supported, the cloud platform can only issue charging start / stop commands in the automatic scheduling command, and the inverter determines the charging power according to its own logic.

[0320] Whether inverter discharge power setting is supported indicates whether the energy management unit can send a target discharge power value to the inverter. If supported, the cloud platform can include the precise discharge power value in the automatic scheduling command. If not supported, the cloud platform can only issue discharge start / stop commands.

[0321] Whether photovoltaic (PV) charging power limiting is supported indicates whether the energy management unit (EMU) can control the upper limit of the output power of the PV maximum power point tracking (MPPT) circuit. If supported, the cloud platform can include the PV charging power limit value in the automatic scheduling instructions to limit the PV charging rate when there is excess power. If not supported, the cloud platform will not generate instructions related to PV charging power limiting.

[0322] Whether the grid-connected feeder power limit setting is supported indicates whether the energy management unit can set the maximum feeder power for the inverter or grid-connected management module. If supported, the cloud platform includes the grid-connected feeder power limit value in the automatic dispatch command. If not supported, the cloud platform does not consider the feeder power limit dimension when generating automatic dispatch commands.

[0323] Whether target state of charge (SOC) setting is supported indicates whether the energy management unit allows users or the cloud platform to set the target SOC value for the battery pack. If supported, the cloud platform can include the target SOC value in the automatic scheduling command, and the energy management unit can then control the energy storage device to stop charging after reaching the target value. If not supported, the cloud platform can only issue charge / discharge start / stop commands and cannot directly specify the target SOC.

[0324] Does the energy management unit support local policy caching? Does its local storage unit have sufficient storage space and corresponding software logic to store automatic scheduling instructions, preset weight change rules, and preset return conditions issued by the cloud platform locally when communication is normal, and to autonomously control the energy storage device by calling the locally stored parameters when communication is interrupted? If supported, the cloud platform can issue complete scheduling policies and transition parameters to the energy management unit for caching. If not supported, the cloud platform only issues automatic scheduling instructions that are executed immediately, and the energy management unit cannot maintain collaborative transition or automatic scheduling functions after communication is interrupted.

[0325] The energy management unit packages the aforementioned device identifier, scene identifier, and device capability information according to a predefined communication protocol and sends it to the cloud platform via the communication network. Upon receiving the device capability information, the cloud platform stores it in the device file corresponding to the device identifier. When generating subsequent automatic scheduling instructions, the cloud platform tailors and adapts the automatic scheduling control parameters based on the device capability information. For functional dimensions not supported by the device, the cloud platform does not generate corresponding instruction fields, or replaces the dimension's control objective with equivalent start / stop control logic. The automatic scheduling instruction after adaptation is called the adapted automatic scheduling instruction.

[0326] The energy management unit receives adapted automatic dispatch instructions from the cloud platform based on device capability information and stores these instructions in its local storage unit. In automatic dispatch mode, the energy management unit controls the energy storage device to perform charging and discharging according to the content of the adapted automatic dispatch instructions. Because the adapted automatic dispatch instructions have been tailored by the cloud platform based on device capability information, each control field matches the actual hardware capabilities of the energy storage device, ensuring that the instructions can be correctly parsed and executed by the inverter and battery management unit.

[0327] The energy management unit sends device identifiers, scene identifiers, and device capability information to the cloud platform via a communication network interface. Upon receiving the reported information, the cloud platform updates the device profile and generates adapted automatic scheduling instructions, which are then sent to the energy management unit via the same communication network interface.

[0328] In some embodiments, optionally, after detecting that the firmware upgrade is complete or the hardware module status has changed, the energy management unit re-executes the device capability information detection process and reports the updated device capability information to the cloud platform, and / or the cloud platform regenerates the adapted automatic scheduling instructions based on the updated device capability information, so that the system can automatically trigger the re-adaptation of the cloud scheduling instructions after the device function is expanded or the hardware is replaced, ensuring that the adapted automatic scheduling instructions are always consistent with the latest hardware capabilities of the energy management unit.

[0329] In some embodiments, optionally, after receiving the device identifier, the cloud platform may also retrieve the historical power constraint record bound to the device identifier from a preset database, and / or combine the historical power constraint record with the device capability information to generate an adapted automatic scheduling instruction, so that the cloud platform can further refine the instruction parameters based on the power execution performance of the same device in different historical periods, and improve the matching degree between the automatic scheduling instruction and the actual aging state and operating habits of the device.

[0330] In some embodiments, the energy control method may optionally further include: receiving constraint preference parameters from a user terminal, the constraint preference parameters including setting a target state of charge, pre-trip backup power, and power maintenance range; and uploading the constraint preference parameters to a cloud platform so that the cloud platform generates an automatic scheduling instruction including the constraint preference parameters.

[0331] In this embodiment, the energy management unit receives constraint preference parameters sent by the user terminal. These parameters include the target state of charge, the amount of backup power before travel, and the power maintenance range. The energy management unit uploads these constraint preference parameters to the cloud platform. When generating automatic scheduling instructions, the cloud platform incorporates these constraint preference parameters as optimization constraints into the scheduling model, generates automatic scheduling instructions containing the constraint preference parameters, and sends them to the energy management unit for execution.

[0332] Understandably, users can upload their energy consumption preferences and travel plans to the cloud platform as constraint preference parameters through their user terminals. When generating automatic dispatch instructions, the cloud platform incorporates these subjective needs into its calculations, avoiding conflicts between dispatch instructions and user schedules, and improving user acceptance and satisfaction with the energy storage system.

[0333] The user terminal is a smart device held by the user, on which an application bound to the energy management unit runs.

[0334] In the daily use of residential photovoltaic-storage systems, different users have significantly different energy demands on energy storage devices at different times. Some users are concerned about backup power when traveling, hoping that the energy storage device's state of charge will be at a high level before leaving home the next morning. Some users are concerned about power supply during holidays, hoping that the energy storage device's state of charge will remain within a stable range during extended periods away from home to cope with sudden grid outages. Some users are concerned about optimizing electricity costs, hoping to maintain a lower state of charge during off-peak electricity pricing periods to reduce capacity charges.

[0335] These needs cannot be fully covered by the automatic dispatch instructions automatically generated by the cloud platform based on historical and predictive data, because the cloud platform mainly pursues the comprehensive optimization of electricity purchase cost, battery degradation, photovoltaic consumption and backup power guarantee when generating automatic dispatch instructions, and cannot predict users' specific travel plans or vacation arrangements on specific dates.

[0336] The energy management unit needs to provide a mechanism that allows users to upload personalized constraint preference parameters to the cloud platform through their user terminals, so that the cloud platform can incorporate these constraint preference parameters as optimization constraints when generating automatic scheduling instructions.

[0337] Constraint preference parameters are a set of data set that users set through their terminals to limit the operational boundaries of energy storage devices within a specific time period. These parameters include setting the target state of charge, the amount of backup power required before travel, and the power maintenance range.

[0338] The target state of charge (SOC) is the SOC value that the user expects the energy storage device to reach before a specific point in time. Pre-trip backup power is the minimum amount of usable electricity the user expects the energy storage device to retain before the start of the trip, used to support vehicle charging or power consumption during the journey. The power maintenance range is a range consisting of two SOC values; the user expects the energy storage device's SOC to remain within this range for a specified period, neither falling below the lower limit nor exceeding the upper limit.

[0339] The user terminal provides an input interface for constraint preference parameters. Users select the type of constraint preference parameter, input the corresponding value or range, and specify the effective time window for the constraint preference parameter. The user terminal packages the user-input constraint preference parameters according to a predefined communication protocol and sends them to the energy management unit via the communication network. The energy management unit receives the constraint preference parameters from the user terminal through the communication network interface.

[0340] Upon receiving the constraint preference parameters, the energy management unit stores them in its local storage. Simultaneously, it uploads these parameters to the cloud platform via the communication network. The uploaded data packet includes the device identifier, the type identifier of the constraint preference parameter, its numerical value, and the effective time window. Upon receiving the constraint preference parameters, the cloud platform stores them in the user preference profile corresponding to the device identifier.

[0341] When generating automatic scheduling instructions, the cloud platform reads constraint preference parameters from the user's preference profile. The cloud platform incorporates these constraint preference parameters as optimization constraints into the process of determining the automatic scheduling control parameters.

[0342] Specifically, when solving multi-objective optimization problems, the cloud platform transforms constraint preference parameters into mathematical constraint inequalities and adds them to the original set of device operation constraints. For example, when the constraint preference parameter is a set target state of charge (SOC), the cloud platform constrains the target power value carried in the automatic scheduling command at the corresponding time point, ensuring that the energy storage device completes charging to the set target SOC before the time point. When the constraint preference parameter is the pre-trip backup power level, the cloud platform constrains the discharge power in the automatic scheduling command within the pre-trip time window, ensuring that the SOC is not lower than the level corresponding to the pre-trip backup power level. When the constraint preference parameter is the power maintenance range, the cloud platform constrains the charging and discharging operations in the automatic scheduling command throughout the entire effective time window, ensuring that the energy storage device's SOC always remains within the power maintenance range.

[0343] After the above constraint processing, the automatic dispatch instructions generated by the cloud platform include the execution logic of the constraint preference parameters. The cloud platform sends the automatic dispatch instructions containing the constraint preference parameters to the energy management unit. The energy management unit receives the automatic dispatch instructions containing the constraint preference parameters from the cloud platform and, in automatic dispatch mode, controls the energy storage device to perform charging and discharging according to the automatic dispatch instructions containing the constraint preference parameters. Since the automatic dispatch instructions have been processed by the cloud platform with the constraint preference parameters as optimization constraints, the energy storage device naturally meets the personalized energy consumption needs set by the user through the user terminal during the execution of the automatic dispatch instructions.

[0344] The energy management unit, acting as a node for receiving and forwarding constraint preference parameters, establishes connections with the user terminal and the cloud platform respectively through the communication network interface.

[0345] The user terminal, acting as the input point for constraint preference parameters, sends these parameters to the energy management unit (EMU) via the communication network. The cloud platform, as the consumer of these parameters, updates the optimization model and generates automatic scheduling instructions containing them upon receipt. The battery management unit (BMU) provides real-time state of charge (SOC) values ​​to the EMU via the communication bus. When executing automatic scheduling instructions containing constraint preference parameters, the EMU compares the real-time SOC values ​​with the target values ​​in the constraint preference parameters to determine if the constraints have been met. The inverter receives the target power values ​​from the EMU through its control interface and performs the corresponding power conversion operations.

[0346] In some embodiments, the energy control method may optionally further include: when the energy management unit communicates normally with the cloud platform, storing at least one of the automatic scheduling instructions, preset weight change rules, and preset return conditions in the energy management unit; when the communication between the energy management unit and the cloud platform is interrupted, controlling the energy storage device to perform charging and discharging according to at least one of the automatic scheduling instructions, preset weight change rules, and preset return conditions stored in the energy management unit.

[0347] In this embodiment, when the energy management unit communicates normally with the cloud platform, the automatic scheduling instructions, preset weight change rules and preset return conditions issued by the cloud platform are cached in the local non-volatile memory.

[0348] When communication is interrupted, the energy management unit reads the automatic scheduling instructions stored locally to control the charging and discharging of the energy storage device. When it receives a local instruction from the user, it calls the cached preset weight change rules and preset return condition parameters to execute instruction weight fusion and state switching control, so that a smooth transition can still be maintained during the network outage.

[0349] Understandably, when communication is interrupted and the user sends a local command, the energy management unit calls the cached weight change rules and return condition parameters to perform weighted fusion and smooth state switching between the charging and discharging user commands and the cached automatic scheduling commands. This avoids power jumps caused by user intervention in network outage scenarios and ensures the continuous stability of energy storage equipment operation during communication failures.

[0350] In the actual deployment environment of residential photovoltaic storage systems, the communication network between the energy management unit and the cloud platform relies on a home broadband network or wireless network connection.

[0351] Communication networks may be interrupted due to reasons such as operator failures, power outages to home routers, changes in network configuration, or severe weather. When communication is interrupted, the energy management unit cannot receive automatic dispatch instructions from the cloud platform in real time, nor can it upload operational data to the cloud platform.

[0352] If the energy management unit completely stops charging and discharging control during a communication interruption, the energy storage device will be out of control, which may lead to over-discharging of the battery to the point of damage or loss of backup power.

[0353] If the energy management unit continues to blindly execute the last automatic dispatch command received before the communication interruption during the communication interruption, and the photovoltaic power generation forecast data or electricity price data on which the command is based is outdated, it may lead to a serious deviation between the charging and discharging decisions and the actual operating conditions.

[0354] Therefore, when communication is normal, the energy management unit needs to store at least one of the automatic scheduling instructions, preset weight change rules, and preset return conditions issued by the cloud platform locally, so that it can continue to control the energy storage device to perform charging and discharging when communication is interrupted.

[0355] Normal communication between the energy management unit and the cloud platform means that the energy management unit can establish two-way data interaction with the cloud platform through the communication network interface, and the data transmission latency and packet loss rate are within acceptable limits. The energy management unit determines whether the communication is normal by periodically sending heartbeat packets to the cloud platform and receiving responses from the cloud platform. The heartbeat packet sending cycle is stored in the local storage unit and is configured by the cloud platform when communication is normal.

[0356] When communication between the energy management unit and the cloud platform is normal, the energy management unit stores automatic scheduling instructions in its local storage unit. These instructions are generated by the cloud platform and sent to the energy management unit. Upon receiving the instructions, the energy management unit writes them to a designated storage area in its local storage unit. The local storage unit includes non-volatile memory, ensuring that the automatic scheduling instructions stored therein can still be read even after a power outage and restart.

[0357] When the energy management unit communicates normally with the cloud platform, the energy management unit stores the preset weight change rules in the local storage unit.

[0358] The preset weight change rules include at least one of the following: exponential change rules, linear change rules, and piecewise change rules. When communication is normal, the cloud platform sends the identifier information of the preset weight change rules and the corresponding parameter values ​​to the energy management unit, which then writes them into the corresponding storage area in its local storage unit.

[0359] When the energy management unit communicates normally with the cloud platform, the energy management unit will store the preset return conditions in the local storage unit.

[0360] The preset return conditions include the preset takeover duration, the identification code for the user's end-of-take command, and the target values ​​corresponding to the target operating parameters. When communication is normal, the cloud platform sends these parameters to the energy management unit, which then stores them in its local storage unit.

[0361] When the energy management unit detects a communication interruption with the cloud platform, it switches to local autonomous control mode. The communication interruption is detected by failing to receive a response from the cloud platform for several consecutive heartbeat cycles.

[0362] After entering the local autonomous control mode, the energy management unit no longer waits for real-time instructions from the cloud platform. Instead, it reads at least one of the automatic scheduling instructions, preset weight change rules, and preset return conditions stored in the local storage unit, and uses these locally stored parameters to control the energy storage device to perform charging and discharging.

[0363] In local autonomous control mode, the energy management unit controls the energy storage device to perform charging and discharging according to the automatic dispatch instructions stored in the local storage unit. The energy management unit reads the target power value from the automatic dispatch instructions from the local storage unit, sends the target power value to the inverter through the control interface, and controls the inverter to adjust the duty cycle of the internal power conversion circuit.

[0364] At the same time, the energy management unit sends the current limit information corresponding to the target power value to the battery management unit, which then monitors and protects the charging and discharging process of the battery pack in real time based on the current limit information.

[0365] In local autonomous control mode, if a user sends a charge / discharge command to the energy management unit via a locally connected user terminal during a communication interruption, the energy management unit reads the preset weight change rules stored in the local storage unit and determines the control weights of the automatic dispatch command and the charge / discharge command according to the preset weight change rules. The energy management unit then weights the automatic dispatch power corresponding to the automatic dispatch command and the user's desired power corresponding to the charge / discharge command based on the determined control weights to obtain a merged power, and controls the energy storage device to perform charging and discharging according to the merged power. The execution logic of this process is consistent with the collaborative transition state when communication is normal; the only difference is that the automatic dispatch command and the preset weight change rules are both issued from the local storage unit rather than in real time from the cloud platform.

[0366] In local autonomous control mode, the energy management unit reads the preset return conditions stored in the local storage unit and uses these parameters to determine whether the conditions for switching from the user-controlled state to the collaborative transition state are met. When the preset return conditions are detected, the energy management unit executes the corresponding state switch and adjusts the control weights according to the preset weight change rules stored locally.

[0367] In some embodiments, optionally, when communication between the energy management unit and the cloud platform is interrupted, the energy control method further includes: receiving a charging / discharging user instruction sent by a user terminal, and adjusting the control weight corresponding to the automatic scheduling instruction and the control weight corresponding to the charging / discharging user instruction according to a preset weight change rule stored in the energy management unit; determining the fusion result according to the adjusted control weight, and controlling the energy storage device to perform charging / discharging according to the fusion result; determining whether to switch from the user-dominated state to the collaborative transition state according to a preset return condition stored in the energy management unit, and switching to the automatic scheduling state when the control weight corresponding to the automatic scheduling instruction reaches a second preset threshold.

[0368] In this embodiment, when communication between the energy management unit and the cloud platform is interrupted, the unit receives charging and discharging user commands sent by the user terminal through local communication. It then calls preset weight change rules stored locally to adjust the control weights of the automatic scheduling command and the charging and discharging user commands. Based on the adjusted control weights, it determines the fusion result and controls the energy storage device to perform charging and discharging. Simultaneously, based on preset return conditions stored locally, it determines whether to switch from a user-dominated state to a collaborative transition state. When the control weight of the automatic scheduling command reaches a second preset threshold, it switches to the automatic scheduling state, ensuring that charging and discharging control can still be maintained according to the cached automatic scheduling commands during network outages.

[0369] Understandably, in network outage scenarios, the execution of commands is still based on the preset weight change rules and preset return conditions of the local cache, resulting in weighted fusion and smooth state switching, rather than directly replacing the cached automatic scheduling commands with user commands. This avoids power step jumps caused by temporary user operations during network outages, ensuring that the transition of control in the offline state is consistent with that in the online state, and guaranteeing the continuous stability of energy storage equipment operation during communication failures.

[0370] During the communication interruption, the cloud platform was unable to issue new automatic scheduling instructions to the energy management unit, and the energy management unit was also unable to upload operational data to the cloud platform.

[0371] When communication is interrupted, the user terminal can directly establish communication with the energy management unit via local area network or Bluetooth, and send charging and discharging user commands to the energy management unit. At this time, the energy management unit needs to rely on locally stored parameters to complete the entire process control from user takeover to return, and cannot rely on the real-time participation of the cloud platform.

[0372] When communication is normal, the energy management unit stores the automatic scheduling instructions, preset weight change rules, and preset return conditions issued by the cloud platform in the local storage unit.

[0373] When a communication interruption occurs, the energy management unit reads the automatic dispatch instructions stored in the local storage unit as the basis for current control. The automatic dispatch instructions contain the target power value and time window information last issued by the cloud platform before the communication interruption. The energy management unit parses the target power value carried in the automatic dispatch instructions into the automatic dispatch power. The automatic dispatch power remains unchanged during the communication interruption until the cloud platform issues a new automatic dispatch instruction after communication is restored.

[0374] In the event of a communication interruption between the energy management unit (EMU) and the cloud platform, the EMU receives charge / discharge user commands from the user terminal. The user terminal transmits these commands to the EMU via a local communication link. The EMU then parses the user's desired power output from these commands. This desired power output reflects the user's temporary operational intention to charge or discharge the energy storage device during the communication interruption.

[0375] The energy management unit reads the preset weight change rules stored in the local storage unit. The preset weight change rules include at least one of the following: exponential change rules, linear change rules, and piecewise change rules.

[0376] The preset weight change rules include several parameters such as the transition start time, the transition target weight, and the transition time parameter.

[0377] The energy management unit determines the control weights corresponding to automatic scheduling commands and charging / discharging user commands based on preset weight change rules.

[0378] When transitioning from automatic scheduling to a cooperative transition state, the initial control weights for automatic scheduling commands are higher, while those for charging / discharging user commands are lower. Conversely, when returning from a user-dominated state to a cooperative transition state, the initial control weights for charging / discharging user commands are higher, while those for automatic scheduling commands are lower.

[0379] The process of changing the control weights is exactly the same as when communication is normal. The only difference is that the transition start time is set to the system time when the energy management unit detects the communication interruption and receives the charging / discharging user command.

[0380] The energy management unit determines the fusion result based on the adjusted control weights. The fusion result is obtained by weighting the automatically dispatched power and the user's desired power.

[0381] For example, the automatic dispatch power and the control weight of the automatic dispatch command are multiplied to obtain the automatic dispatch weighted component, and the user-expected power and the control weight of the charge / discharge user command are multiplied to obtain the user-expected weighted component. The two weighted components are added together to obtain the merged power. The energy management unit sends the merged power to the inverter through the control interface, controlling the inverter to adjust the duty cycle of the internal power conversion circuit, so that the energy storage device performs charging or discharging operations according to the merged result. At the same time, the energy management unit sends the current limit information corresponding to the merged power to the battery management unit, which monitors and protects the charging and discharging process of the battery pack in real time based on the current limit information.

[0382] During a communication interruption, if a user-dominated state is triggered, the energy management unit reads the preset takeover conditions stored in the local storage unit. Several parameters in the preset takeover conditions include a preset takeover duration, a user's end-of-takeover command identifier, and target values ​​corresponding to the target operating parameters. The energy management unit determines whether the takeover duration has been reached based on the preset takeover duration.

[0383] The energy management unit continuously monitors the local communication link to detect whether it has received a user termination command from the user terminal. The energy management unit continuously reads the state of charge (SOC) values ​​of the energy storage device from the battery management unit and the real-time charging or discharging power values ​​from the inverter's power sampling circuit, comparing the read operating parameters with the target values ​​carried in the charging / discharging user commands.

[0384] When any one of the following conditions is met—that the takeover duration reaches the preset takeover duration, a user's command to end the takeover is received, or the operating parameters reach the target value—the energy management unit determines that the preset return conditions are met and switches from the user-led state to the collaborative transition state.

[0385] During the collaborative transition phase of the return path, the energy management unit continues to adjust the control weights according to the preset weight change rules stored in the local storage unit. The control weights corresponding to charging and discharging user commands gradually decrease over time, while the control weights corresponding to automatic scheduling commands gradually increase over time. When the control weight corresponding to an automatic scheduling command reaches a second preset threshold, the energy management unit switches from the collaborative transition phase to the automatic scheduling phase. The value of the second preset threshold is stored in the local storage unit and is pre-delivered by the cloud platform when communication is normal. After switching to the automatic scheduling phase, the energy management unit resumes controlling the energy storage device to perform charging and discharging according to the automatic scheduling commands stored in the local storage unit, and no longer refers to the charging and discharging user commands.

[0386] In some embodiments, optionally, in the event of an interruption in communication between the energy management unit and the cloud platform, the energy management unit receives charging and discharging user instructions sent by the user terminal through a local wireless communication link, and / or the energy management unit authenticates the user terminal based on a pre-stored device binding relationship, so that user intervention in the offline state must undergo local authentication, preventing unauthorized devices from illegally controlling the energy storage device through the local network.

[0387] This invention provides an energy dispatching method applied to a cloud platform, wherein the cloud platform and the energy management unit of an energy storage device are communicatively connected, such as... Figure 4 As shown, the energy scheduling methods include:

[0388] Step S200: Receive the operating data and equipment capacity information of the energy storage device reported by the energy management unit;

[0389] Step S202: Obtain the scheduling data corresponding to the energy storage device. The scheduling data includes photovoltaic power generation forecast data, household load forecast data, and electricity price data.

[0390] Step S204: Determine the equipment operation constraints corresponding to the energy storage device based on the operation data, and determine the automatic dispatch control parameters based on the dispatch data and equipment operation constraints;

[0391] Step S206: Generate automatic dispatch instructions based on automatic dispatch control parameters, and send the automatic dispatch instructions to the energy management unit so that the energy management unit controls the energy storage device to perform charging and discharging according to the automatic dispatch instructions.

[0392] This invention discloses an energy dispatching method applied to a cloud platform. The cloud platform receives energy storage device operation data and device capacity information reported by the energy management unit, obtains photovoltaic power generation forecast data, household load forecast data and electricity price data, determines device operation constraints based on the operation data, calculates automatic dispatching control parameters by combining dispatching data and constraints, generates automatic dispatching instructions and sends them to the energy management unit for charging and discharging.

[0393] Understandably, by placing computationally intensive prediction and optimization calculations on a cloud platform, and leveraging a global data perspective to integrate power generation forecasts, load forecasts, electricity prices, and real-time equipment status and capabilities, optimal scheduling instructions that balance electricity costs and hardware security are generated. The introduction of equipment capability information allows instructions to automatically adapt to different hardware configurations, avoiding execution failures caused by issuing parameters unsupported by the equipment. The cloud-based scheduling architecture fully utilizes the powerful computing capabilities of the cloud platform, enabling precise, secure, and personalized energy management of distributed residential energy storage systems.

[0394] In the end-cloud collaborative architecture of residential photovoltaic storage systems, the cloud platform, as the remote optimization decision-making end, establishes a two-way connection with the energy management unit, which is the local control core, through a communication network.

[0395] The cloud platform does not have the ability to directly collect local sensor data from energy storage devices, nor can it directly control the hardware operations of the inverter and battery management unit.

[0396] The core function of the cloud platform is to generate automatic scheduling instructions based on global data and optimization algorithms, and then distribute these instructions to the energy management unit for execution. The energy management unit is responsible for data acquisition, instruction parsing, and hardware driver implementation locally. This division of labor between the cloud platform and the energy management unit allows computationally intensive multi-objective optimization tasks to be completed in the cloud, while real-time power control tasks are completed locally.

[0397] The cloud platform receives operational data and equipment capacity information from the energy management unit. The energy management unit periodically packages the operational data and sends it to the cloud platform via the communication network according to a predefined communication protocol.

[0398] The operational data includes the current state of charge (SOC) of the energy storage device, battery pack temperature distribution data, maximum allowable charging power, maximum allowable discharging power, real-time inverter output power, grid-connected power, real-time photovoltaic array power generation, and SOC upper and lower limits set by the user through the user terminal.

[0399] The equipment capability information includes equipment identification, scenario identification, and whether it supports AC charging power setting, inverter discharge power setting, photovoltaic charging power limitation, grid-connected feeder power upper limit setting, target state of charge setting, and local policy caching.

[0400] After receiving operational data and device capability information, the cloud platform stores them in a device profile database corresponding to the device identifier for subsequent optimization calculations.

[0401] The cloud platform acquires scheduling data corresponding to the energy storage devices. The scheduling data includes photovoltaic power generation forecast data, household load forecast data, and electricity price data.

[0402] Photovoltaic power generation forecast data is generated by the cloud platform based on irradiance forecasts, temperature forecasts, and historical power generation curves of the photovoltaic array provided by meteorological data sources. The forecast is calculated using a time-series prediction model to obtain the expected power generation of the photovoltaic array at each future time step. Household load forecast data is generated by the cloud platform based on users' historical electricity consumption records, seasonal electricity consumption patterns, and holiday factors. The forecast is calculated using a load prediction model to obtain the expected household electricity consumption at each future time step.

[0403] Electricity price data is obtained by the cloud platform from the electricity market data interface, covering the purchase and sale prices of electricity at every future time step. Negative electricity price values ​​are allowed to reflect scenarios of negative electricity prices caused by overcapacity in renewable energy generation in the electricity market.

[0404] The scheduling data covers the next day in the time domain, with time steps in hours. These three types of scheduling data together constitute the input boundary conditions for the cloud platform to perform multi-objective optimization.

[0405] The cloud platform determines the corresponding equipment operation constraints for energy storage devices based on operational data. Equipment operation constraints are the hardware and compliance boundaries that energy storage devices must adhere to when performing charging and discharging operations.

[0406] The cloud platform reads the user-defined upper and lower bounds of the state of charge (SOC) from the operational data, and simultaneously obtains the safe SOC upper and lower bounds provided by the battery management unit (BMU) from the data reported by the energy management unit (EMU). The cloud platform then takes the intersection of the user-defined SOC upper and lower bounds and the safe SOC upper and lower bounds to obtain the final charge constraint. The logic for taking the intersection is as follows: the upper bound of the final charge constraint is the smaller of the user-defined safe SOC upper bound and the safe SOC upper bound provided by the BMU; the lower bound of the final charge constraint is the larger of the user-defined safe SOC lower bound and the safe SOC lower bound provided by the BMU.

[0407] Furthermore, the cloud platform determines the upper and lower bounds of power constraints based on the maximum charging power, maximum discharging power, and grid-connected power values ​​carried in the operational data, combined with the identifiers in the equipment capability information indicating whether the corresponding functions are supported. The cloud platform incorporates the final charge constraint and power constraints into the equipment operation constraint set.

[0408] The cloud platform determines automatic scheduling control parameters based on scheduling data and equipment operating constraints. These parameters represent the cloud platform's suggested charging / discharging power values ​​and execution time windows for each future time step. With electricity purchase cost, battery degradation, photovoltaic power consumption, and backup power guarantee as comprehensive optimization objectives, the cloud platform solves for the optimal automatic scheduling control parameters within the constraints of equipment operating conditions.

[0409] The electricity purchase cost target is obtained by multiplying the purchased power at each time step by the electricity price for the corresponding time period and summing them up. The battery degradation target is to reduce battery cycle loss by limiting the depth and rate of charge / discharge.

[0410] The photovoltaic (PV) power consumption target is achieved by prioritizing the use of PV power generation to meet the load and storing excess PV power in batteries as much as possible.

[0411] The backup power guarantee objective is achieved by ensuring that the state of charge is not lower than the user-defined backup power lower limit within a preset time period.

[0412] The cloud platform uses mathematical optimization algorithms to search for the optimal solution within the feasible region of the equipment operation constraints, and obtains the automatic scheduling and control parameters for each time step.

[0413] The cloud platform generates automatic scheduling instructions based on automatic scheduling control parameters.

[0414] The automatic scheduling command carries the target power value, power direction identifier, and execution time window for one or more future time steps.

[0415] The cloud platform encodes and encapsulates automatic dispatch instructions according to a predefined communication protocol. The cloud platform then sends these instructions to the energy management unit via the communication network. Upon receiving the automatic dispatch instructions, the energy management unit stores them in its local storage unit and, in automatic dispatch mode, controls the energy storage device to perform charging and discharging according to the target power value specified in the automatic dispatch instructions.

[0416] In some embodiments, optionally, such as Figure 5 As shown, the energy scheduling method also includes:

[0417] Step S210: Receive constraint preference parameters and / or user takeover records uploaded by the energy management unit. The constraint preference parameters are sent by the user terminal to the energy management unit and uploaded to the cloud platform. The user takeover records include the user takeover time, user takeover type and user takeover duration.

[0418] Step S212: If the constraint preference parameters exist, determine the optimization constraints based on the constraint preference parameters, and generate new automatic scheduling instructions to be sent to the energy management unit;

[0419] Step S214: If user takeover records exist, determine whether the user takeover behavior meets the preset preference conditions based on the user takeover records within the preset time period;

[0420] Step S216: If the user takeover behavior meets the preset preference conditions, generate user preference parameters based on the user takeover behavior, adjust the automatic scheduling control parameters based on the user preference parameters, generate a new automatic scheduling instruction based on the adjusted automatic scheduling control parameters, and send it to the energy management unit.

[0421] In this embodiment, the cloud platform receives constraint preference parameters and user takeover records uploaded by the energy management unit. The constraint preference parameters are sent by the user terminal, and the user takeover records include the takeover time, type, and duration. When constraint preference parameters exist, the cloud platform determines the optimization constraints based on them and generates new automatic scheduling instructions for issuance.

[0422] When a user takeover record exists, the cloud platform determines whether the preset preference conditions are met based on the takeover records within a preset period. If they are met, user preference parameters are generated based on the user takeover behavior, the automatic scheduling control parameters are adjusted, and new automatic scheduling instructions are generated and issued.

[0423] Understandably, by constraining preference parameters, user-defined energy demands are directly incorporated into dispatch constraints, generating new automatic dispatch instructions. Simultaneously, by analyzing user takeover records, user preference parameters are automatically learned and adjusted when user behavior meets preset preference conditions, reducing the frequency of manual user intervention and balancing grid safety with user experience.

[0424] The automatic scheduling instructions generated by the cloud platform take electricity purchase cost, battery degradation, photovoltaic power consumption and backup power guarantee as comprehensive optimization objectives, and determine the charging and discharging control parameters for each time step in the future within the constraints of equipment operation.

[0425] If the cloud platform relies solely on historical and forecast data to generate automatic dispatch instructions, it may lead to a continuous deviation between the dispatch plan and the user's actual lifestyle.

[0426] Therefore, the cloud platform needs to dynamically adjust the automatic scheduling instructions by receiving the constraint preference parameters and user takeover records uploaded by the energy management unit, so that the scheduling plan can adapt to the user's personalized preferences and behavior patterns.

[0427] The cloud platform receives constraint preference parameters and / or user takeover records uploaded by the energy management unit. Constraint preference parameters are set and sent by the user to the energy management unit via the user terminal, and the energy management unit uploads these parameters to the cloud platform via the communication network. Constraint preference parameters include the target state of charge, pre-trip backup power, and power maintenance range. User takeover records are automatically generated and stored locally by the energy management unit each time a user sends a charge / discharge command and executes a control switch, and are uploaded to the cloud platform when communication is normal.

[0428] The user takeover log includes the user takeover time, user takeover type, and user takeover duration. The user takeover time records the timestamp when the energy management unit switches from automatic scheduling or cooperative transition state to user-dominated state. The user takeover type identifies whether the user action is a charging or discharging request, and whether it is initiated through a mobile application client or the device panel. The user takeover duration records the total duration of the user-dominated state from entry to exit.

[0429] When constraint preference parameters exist, the cloud platform parses the type identifier, numerical content, and effective time window of the constraint preference parameters from the received uploaded data. The cloud platform then transforms the constraint preference parameters into optimization constraints.

[0430] Specifically, when the constraint preference parameter is a set target state of charge (SOC), the cloud platform adds an equality or inequality constraint to the device operation constraints at the corresponding time point, requiring that the cumulative charging amount generated by the automatic scheduling command before the time point is sufficient to enable the energy storage device to reach the set target SOC value. When the constraint preference parameter is the pre-trip backup power level, the cloud platform adds a lower bound constraint within the pre-trip time window, ensuring that the SOC is not lower than the corresponding value of the pre-trip backup power level. When the constraint preference parameter is a power maintenance interval, the cloud platform adds an interval constraint at each time step within the effective time window, ensuring that the SOC is between the upper and lower bounds of the power maintenance interval.

[0431] The cloud platform overlays the transformed optimized constraints onto the original set of equipment operation constraints, resolves the multi-objective optimization problem, determines the adjusted automatic scheduling control parameters, and generates new automatic scheduling instructions based on these parameters. The cloud platform then sends these new automatic scheduling instructions to the energy management unit via the communication network. The energy management unit controls the energy storage equipment to perform charging and discharging according to the new instructions.

[0432] If user takeover records exist, the cloud platform extracts these records from the received uploaded data. The cloud platform then performs behavioral pattern analysis on the accumulated user takeover records over a preset time period.

[0433] The preset time period is configured by the cloud platform, with typical values ​​being the most recent seven days, the most recent thirty days, or the most recent ninety days. The cloud platform organizes all user takeover records within the preset time period into a sequence of actions in chronological order.

[0434] The cloud platform determines whether a user takeover behavior meets preset preference conditions based on the behavior sequence. These preset preference conditions include: the user takeover time exhibits a regular clustering over multiple consecutive preset time periods, such as charging requests occurring between 8:00 and 9:00 AM on multiple consecutive weekdays; the user takeover type remains consistent over multiple consecutive preset time periods, such as all being discharge requests with target operating parameters reaching specific state of charge values; and the user takeover duration remains stable over multiple consecutive preset time periods, such as each takeover lasting between 30 and 60 minutes.

[0435] When any of the above conditions are met, the cloud platform determines that the user's takeover behavior meets the preset preference conditions, that is, the user's temporary operation behavior actually reflects a stable energy consumption preference.

[0436] When a user's takeover behavior meets preset preference conditions, the cloud platform generates user preference parameters based on the user's takeover behavior.

[0437] The process of generating user preference parameters is as follows: The cloud platform extracts the most frequently occurring user takeover time from user takeover records that meet preset preference conditions as the preference effective time period, extracts the target operating parameter corresponding to the most frequently occurring user takeover type as the preference target value, and extracts the most frequently occurring user takeover duration as the preference duration. The cloud platform combines the extracted preference effective time period, preference target value, and preference duration into user preference parameters.

[0438] User preference parameters are essentially long-term constraint preferences derived from users' historical behavior. Their format is similar to constraint preference parameters, but their values ​​are derived from learning user takeover records rather than from users' active input.

[0439] The cloud platform adjusts the automatic scheduling control parameters based on user preference parameters. The adjustment method is as follows: the cloud platform maps the user preference parameters into optimization constraints according to the same transformation rules as the constraint preference parameters, adds them to the set of equipment operation constraints, and re-executes multi-objective optimization calculations to obtain the adjusted automatic scheduling control parameters.

[0440] Since the user preference parameters are derived from statistical analysis of user takeover behavior, the adjusted automatic scheduling control parameters can automatically adapt future automatic scheduling commands to the energy consumption patterns already exhibited by the user without relying on manual input from the user each time.

[0441] The cloud platform generates new automatic dispatch instructions based on the adjusted automatic dispatch control parameters and sends these instructions to the energy management unit via the communication network. The energy management unit receives the new automatic dispatch instructions and stores them in its local storage unit. In automatic dispatch mode, it controls the energy storage device to perform charging and discharging according to the new instructions.

[0442] In one specific embodiment, the present invention optionally proposes a human-machine co-driving scheduling and control method for an energy storage system. The solution is designed for an end-to-cloud collaborative energy system consisting of photovoltaic modules, an integrated energy storage unit (including a battery pack, battery management module, charging / discharging and inverter module, energy management module, and communication module), a metering device, a gateway, a mobile application client, and a cloud-based energy management platform.

[0443] Its core idea is to regard the charging and discharging control of the integrated energy storage unit as a shared resource that can be proportionally integrated, arbitrated according to priority, and automatically returned between the "machine side (automatic scheduling in the cloud)" and the "human side (temporary operation by the user)," thereby achieving true human-machine co-driving on the same execution unit.

[0444] For ease of description, the software modules inside the energy storage unit that are responsible for receiving instructions, executing arbitration and fusion, and driving charging and discharging actions are collectively referred to as the energy management unit of the energy storage unit; the software in the cloud that is responsible for prediction, optimization and generation of scheduling plans is collectively referred to as the cloud energy management platform, i.e., the cloud platform.

[0445] The software framework diagram for human-machine co-driving in the energy storage system is as follows: Figure 9 As shown: The user intent management module in the cloud-based energy management platform receives user patterns and intent inputs from the mobile application client, performs structured parsing of the intents, and stores them in a preference library for subsequent use by the optimization scheduling module. The prediction module generates photovoltaic power generation forecasts and household load forecasts based on meteorological and historical electricity consumption data, providing boundary inputs for future time periods for the optimization scheduling module. The optimization scheduling module uses the prediction results and user intents as inputs, combines safety constraints, and performs multi-objective optimization calculations to generate a machine-side automatic scheduling plan. The distribution and traceability module distributes the automatic scheduling instructions output by the optimization scheduling module to the energy management module of the integrated energy storage unit, and simultaneously uploads and archives the instruction versions and execution records for traceability.

[0446] On the one hand, the mobile application client sends the user's set mode and intent input to the user intent management module in the cloud energy management platform. On the other hand, it sends the real-time charging and discharging operation instructions directly to the energy management module of the integrated energy storage unit. The interface presents the current control status and execution progress to the user, i.e., the plan and takeover status display.

[0447] The energy management module of the integrated energy storage unit, serving as the core of human-machine collaborative operation, receives automatic scheduling commands from the issuing and tracing module and real-time operation commands from the mobile application client, synchronously writing the command information to the edge cache unit for use in case of network outages. The control state machine determines its control state based on the current command source and operating status, and outputs the status signal to the fusion and arbitration unit. The battery management module collects cell safety data in real time, providing the fusion and arbitration unit with constraint parameters such as upper and lower limits of safe state of charge. When communication is normal, the edge cache unit locally stores the policy parameters issued by the cloud platform and user takeover records; when communication is interrupted, it provides offline control basis for the fusion and arbitration unit.

[0448] The fusion and arbitration unit obtains the current control state from the control state machine, safety constraints from the battery management module, and cached scheduling parameters and weighting rules from the edge cache unit. It then performs weighted fusion and priority arbitration on the automatic scheduling commands and user charging / discharging commands, outputting a fused power signal to the smooth transition and automatic return unit. The smooth transition and automatic return unit executes the smooth transition and expiration return judgment of control rights between automatic scheduling, collaborative transition, and user-led operation according to preset weighting change rules, outputting the final target power to the charging / discharging and inverter module. The charging / discharging and inverter module drives the inverter power circuit to perform the corresponding charging or discharging operation based on the target power.

[0449] The overall process of human-machine co-driving scheduling is as follows: Figure 10 and Figure 11 As shown:

[0450] Step S300: Binding registration and reporting of capability information;

[0451] Step S302: The integrated energy storage unit collects operational data;

[0452] Step S304: Obtain photovoltaic and load forecasts and electricity prices from the cloud;

[0453] Step S306: Cloud-based multi-objective optimization generates machine-side plan;

[0454] Step S308: The user inputs their intent via the mobile application client;

[0455] Step S310: Intent classification: Immediate takeover or constraint preference;

[0456] Step S312: The energy management module determines the control status;

[0457] Step S314: Is the user in the takeover window?

[0458] If the judgment result of step S314 is negative, then proceed to step S316: Machine master: execute according to the machine-side plan;

[0459] If the judgment result of step S314 is yes, then step S318 is executed: Fusion and Arbitration: Executed according to the weighted fusion mechanism and user power;

[0460] Step S320: Has the safety boundary been reached?

[0461] If the judgment result of step S320 is yes, then execute step S322: safety fallback: project to the safe feasible domain for execution;

[0462] If the judgment result of step S320 is negative, then step S324 is executed: issue the actual charging and discharging power and working mode;

[0463] Step S326: Determine whether the takeover window has expired, the goal has been achieved, or the user has ended the process;

[0464] If the judgment result of step S326 is that any condition is met, then step S328 is executed: negotiate co-driving: smoothly transition and automatically return to the machine side;

[0465] If the judgment result of step S326 is that none of the conditions are met, then proceed to step S330: continue execution in the current state;

[0466] Step S332: Record the takeover behavior and consolidate it into preference priorities;

[0467] Step S334: Edge cache the latest plans and co-driving rules (supports offline).

[0468] 1. Overall process of edge-cloud collaboration:

[0469] The method includes the following steps:

[0470] Step 1 (Establish Binding and Capability Registration): The cloud-based energy management platform receives the device identifier, scenario identifier, and device capability information reported by the integrated energy storage unit, and establishes a binding relationship between the user's energy system and the integrated energy storage unit. Device capability information includes at least: whether it supports AC charging power setting, whether it supports inverter discharge power setting, whether it supports photovoltaic charging power limiting, whether it supports grid-connected power limiting, whether it supports target state of charge setting, and whether it supports local policy caching.

[0471] Step Two (Data Acquisition): The energy management module of the integrated energy storage unit collects its own operational data and reports it to the cloud-based energy management platform. The operational data includes at least: battery capacity, maximum charging power, maximum discharging power, charging / discharging efficiency, current state of charge, safe upper and lower charge limits provided by the battery management module, user-set upper and lower charge limits, grid-connected power limit, and device online status.

[0472] Step 3 (Obtaining Forecasts and Electricity Prices): The cloud-based energy management platform obtains photovoltaic power generation forecasts, household load forecasts, and electricity price data for the future scheduling time domain. Preferably, the scheduling time domain is the next day, with hourly time steps; negative electricity price data is allowed.

[0473] Step Four (Receiving User Intent): The mobile application client receives the user's selected scheduling mode, temporary charging / discharging operation, travel time, backup power, and vacation maintenance intent, and sends them to the cloud energy management platform and the energy management module of the integrated energy storage unit. This invention categorizes user intents into two types: one is "instant takeover" intents, such as charging immediately or discharging immediately; the other is "constraint preference" intents, such as reaching a certain power level before travel or maintaining a certain power level range during vacation.

[0474] Step 5 (Machine-side scheduling plan generation): Based on forecast data, electricity price data, equipment configuration, and user "constraint preference" intent, the cloud-based energy management platform performs multi-objective optimization to generate a machine-side suggested time-sharing charging and discharging plan, which serves as the machine-side input in human-machine co-driving.

[0475] Step Six (Control Status Determination): The energy management module of the energy storage unit determines the current control status (one of four: machine master, user takeover, negotiated co-operation, or safety backup) based on whether there is immediate user takeover, whether it is currently within the takeover time window, and whether the safety boundary has been reached.

[0476] Step 7 (Control Integration and Arbitration): The energy management module of the energy storage unit integrates the recommended power on the machine side and the user's expected power proportionally according to the current control status, and arbitrates the power in conjunction with the safety boundary to obtain the actual charging and discharging power and working mode executed in this time step.

[0477] Step 8 (Smooth Transition and Automatic Return): When user takeover begins or ends, the energy management module of the energy storage unit adjusts the integration weight according to the smooth transition rules to ensure a smooth transfer of control between the human and the machine. When the takeover time window expires or the return conditions are triggered, the energy management module automatically returns control to the machine side and resumes the execution of the machine side scheduling plan.

[0478] Step Nine (Intent Sedimentation and Preference Learning): The cloud-based energy management platform records the time period, type, and frequency of user takeovers. High-frequency and stable takeover behaviors are sedimented into "constraint preference" intents and proactively incorporated into subsequent machine-side scheduling to reduce the number of times users are forced to take over.

[0479] Step 10 (Edge Caching and Offline Co-driving): The energy management module of the energy storage unit caches the latest machine-side plans and human-machine co-driving rules locally; when the cloud is unreachable, the energy management module continues to maintain the human-machine co-driving semantics of "users can temporarily take over and automatically return to the machine side when the time comes" locally, and continuously checks the safety boundaries.

[0480] 2. Human-machine co-driving control state machine:

[0481] This invention organizes the control of the integrated energy storage unit into four states, and the energy management module of the integrated energy storage unit transitions between these states based on user operation, takeover time window, and safety boundary:

[0482] (1) Machine master: No user takes over immediately. The energy management module of the energy storage machine performs charging and discharging according to the machine-side plan of the cloud energy management platform.

[0483] (2) User takeover: Users initiate immediate takeover through mobile application clients or device panels. During the takeover time window, the user's intention is dominant, but is still subject to security boundaries.

[0484] (3) Co-operation: During the transition phase at the beginning or end of the takeover, the energy management module of the integrated energy storage unit takes into account both the machine-side plan and the user's intention according to the integration weight, so as to achieve a smooth transfer of control.

[0485] (4) Safety backup: When the machine side plan or user intention touches the battery safety boundary, grid connection and feeder limit or minimum backup power requirement, the energy management module of the energy storage unit takes over the execution with the safety constraint as the highest priority, and returns to the corresponding state after the risk is eliminated.

[0486] 3. Control Integration Methods:

[0487] At each time step, the energy management module of the integrated energy storage unit does not simply choose one of two options, but rather weights the recommended power from the machine side and the expected power from the user according to the control weighting to obtain the actual charging and discharging power.

[0488] 4. Methods for a smooth transition of control:

[0489] To avoid a step jump in charging and discharging power during the control switch, this invention allows the fusion weight λ to change continuously with the transition process in the "negotiated co-driving" state, rather than switching instantaneously between 0 and 1.

[0490] 5. Takeover time window and automatic return mechanism:

[0491] When a user takes over immediately, the energy management module of the integrated energy storage unit records the start time of the takeover and sets the takeover time window duration; within the takeover time window, the user's intention is dominant. When any of the following return conditions are met, the energy management module automatically and smoothly returns control to the machine side without requiring manual operation from the user again:

[0492] (1) The takeover time window expires, that is, the interval between the current time and the start time of the takeover reaches the preset window length;

[0493] (2) The user actively terminates the takeover process in the mobile application client;

[0494] (3) The user’s goal of immediate takeover has been achieved, for example, the user requests “immediate charging to a certain level” and the current state of charge has reached that level;

[0495] (4) Continuing to maintain user takeover will touch the safety boundary (such as the battery is about to be overcharged or over-discharged, exceeding the grid connection and feeder limit, or falling below the minimum backup power requirement).

[0496] Automatic return does not involve directly cutting off user control and instantly switching back to the machine side. Instead, it enters a "negotiated co-driving" state, gradually shifting the fusion weight back from user-biased to machine-biased using a smooth transition method. This balances the continuity of temporary user needs with daily automatic scheduling.

[0497] 6. Conflict Arbitration and Security Priorities:

[0498] When conflicts arise between machine-side recommendations, user intent, and safety constraints, the energy management module of the integrated energy storage unit arbitrates according to a unified and interpretable priority, from highest to lowest:

[0499] (1) Safety constraints are the highest: The safety and compliance constraints given by the battery management module, such as the upper and lower limits of safe charge, the maximum charging and discharging power, the upper limit of grid-connected power, and the minimum backup power, must not be exceeded by any instruction from any source;

[0500] (2) User explicit intent is secondary: Under the premise of not breaking the security constraints, the user's immediate intent within the takeover time window takes priority over the machine side's economic optimality.

[0501] (3) Machine-side economic optimization again: Under the premise of not breaking the safety constraints and not violating the current effective user intentions, the machine-side plan pursues the comprehensive optimization of economy, battery life and photovoltaic consumption.

[0502] When user commands conflict with safety constraints, the energy management module of the integrated energy storage unit does not simply reject them. Instead, it projects the user commands into the safe feasible domain for execution, that is, it tries to meet the user's intentions within the safety boundary (for example, if the user requests to charge to full capacity but is already close to the safety limit, it will charge to the safety limit). The reason is explained to the user through the mobile application client.

[0503] 7. User intent retention and preference learning:

[0504] The cloud-based energy management platform's user intent management module records the time, type, and frequency of user takeovers. When a certain type of takeover behavior occurs consistently and frequently over several days (for example, a user manually recharges the battery almost every morning for commuting), the intent management module reclassifies it from "immediate takeover" to "constraint preference" intents and pre-writes it into the constraints or preferences for subsequent machine-side optimization. This allows the machine-side plan to meet the demand in advance, thus proactively reducing the number of times the user is forced to take over. Conversely, if a user does not trigger a pre-defined preference for a long period, the intent management module can gradually weaken the preference's weight. Therefore, human-machine co-driving is not only about "current collaboration" but also about continuous learning and evolving towards "machines proactively adapting to humans" as usage progresses.

[0505] 8. Edge caching execution and human-machine co-driving during network outages:

[0506] The energy management module of the integrated energy storage unit locally caches the latest machine-side plans, control fusion rules, smooth transition rules, and takeover time window settings. When the cloud-based energy management platform is unreachable or communication is interrupted, the energy management module continues to maintain human-machine co-driving semantics locally: users can still temporarily take over via the device panel or a locally connected mobile application client. The energy management module still processes according to the locally cached fusion, transition, and return rules, and checks the battery safety boundary at each time step. When communication is restored, the energy management module reports the takeover record during the network outage to the cloud for intent storage and traceability. This ensures that the human-machine co-driving experience and safety are not lost in network outage scenarios.

[0507] For example, consider the human-machine co-driving process during pre-trip charging: A user sets up a trip for the following morning on their mobile app, hoping for a fully charged battery before departure. The cloud-based energy management platform's user intent management module identifies the intent as a "constraint preference" and pre-writes it into the machine-side optimization, causing the machine to schedule charging during the nighttime low-price period. If the user still clicks "Charge Now" that evening, the energy management module of the integrated energy storage unit enters a negotiated co-driving phase: Since the machine-side plan already includes charging, the user's intent aligns with the machine's direction, and the weighted calculations moderately increase the actual charging power within a safe range; once the trip's charging target is reached, the automatic return condition is triggered, and the energy management module smoothly reclaims control from the machine.

[0508] For example, during periods of high prices, a user temporarily discharges to power a load: During a period of high prices, a user temporarily requests the integrated energy storage unit to discharge for a critical household load. The energy management module of the integrated energy storage unit enters user takeover and discharges according to the user's intention within the takeover time window. At the same time, the energy management module continuously checks the battery's safe lower limit and minimum backup power requirements. When the state of charge approaches the safe lower limit, it enters a safety fallback mode, projecting the discharge power into the safe feasible domain (discharging until the safe lower limit is reached), and notifies the user via a mobile application client. After the takeover time window expires, the energy management module, through negotiation and co-operation, smoothly returns control to the machine.

[0509] For example, a smooth transition of control avoids sudden power spikes: When a user ends an immediate charging takeover, if the system directly switches back to the machine-side plan, the charging power will drop sharply from a higher value. With this invention, the energy management module of the integrated energy storage unit enters negotiated co-driving mode. Using a smooth transition method, the fusion weight gradually shifts from biased towards the user to biased towards the machine side within the time period determined by the transition time constant, allowing the actual charging and discharging power to change continuously, and the battery and inverter to smoothly transition to the machine-side plan.

[0510] For example, consider the arbitration of conflicts between user commands and safety boundaries: A user requests that the energy storage unit be fully charged, but the battery management module's safety limit is lower than full charge. The energy management module arbitrates according to safety priority, projecting the user's "fully charged" intention into the safe feasible domain, stopping charging at the safety limit, and notifying the user via a mobile application that "charging has been performed according to the battery's safe limit." This approach neither violates the safety boundary nor compromises the user's intention to the greatest extent possible within feasible limits.

[0511] For example, preference learning proactively reduces the need for manual intervention: The system observed for several consecutive days that users manually recharged their batteries every morning for their commute. The user intent management module of the cloud-based energy management platform categorized this behavior as "constraint preference" intents, pre-emptively scheduling them on the machine side. This enabled the machine side to automatically complete the recharging before dawn. Afterward, users no longer needed to manually intervene daily, and human-machine collaboration shifted from a passive response to an active adaptation.

[0512] For example, in a scenario of human-machine co-operation during network outages: During a cloud communication interruption, the energy management module of the energy storage unit operates according to the locally cached machine-side plan and human-machine co-operation rules. The user temporarily takes over discharge via the device panel, and the energy management module processes this locally according to fusion, transition, and return rules, while also checking safety boundaries. After the takeover time window expires, the local system automatically returns the machine-side plan. Once communication is restored, the energy management module reports the takeover records from the network outage period to the cloud-based energy management platform for preference learning and traceability.

[0513] like Figure 6 and Figure 8 As shown, this embodiment of the invention also provides an energy control device 800, applied to the energy management unit 1102 of an energy storage device 1100. The energy control device 800 includes: an instruction receiving module 802, used to receive charging and discharging user instructions from a user terminal when controlling the energy storage device to charge and discharge according to the automatic scheduling instructions issued by the cloud platform; a cooperative transition module 804, used to switch from an automatic scheduling state to a cooperative transition state in response to the charging and discharging user instructions, and in the cooperative transition state, control the energy storage device to perform charging and discharging according to the fusion result of the automatic scheduling instructions and the charging and discharging user instructions, wherein the control weight of the charging and discharging user instructions gradually increases over time, and the control weight of the automatic scheduling instructions gradually decreases over time; and a user-dominated module 806, used to switch from the cooperative transition state to the user-dominated state when the control weight of the charging and discharging user instructions reaches a first preset threshold, and in the user-dominated state, control the energy storage device to perform charging and discharging according to the charging and discharging user instructions.

[0514] like Figure 7 and Figure 8As shown, this embodiment of the invention also provides an energy dispatching device 900, applied to a cloud platform 1200. The cloud platform 1200 is communicatively connected to the energy management unit 1102 of the energy storage device 1100. The energy dispatching device 900 includes: a data receiving module 902, used to receive the operating data and device capacity information of the energy storage device reported by the energy management unit; a data dispatching module 904, used to obtain dispatching data corresponding to the energy storage device, including photovoltaic power generation forecast data, household load forecast data, and electricity price data; a constraint determination module 906, used to determine the device operating constraints corresponding to the energy storage device based on the operating data, and determine automatic dispatching control parameters based on the dispatching data and device operating constraints; and an instruction issuing module 908, used to generate automatic dispatching instructions based on the automatic dispatching control parameters, and issue the automatic dispatching instructions to the energy management unit so that the energy management unit controls the energy storage device to perform charging and discharging according to the automatic dispatching instructions.

[0515] like Figure 8 As shown, this embodiment of the invention also provides an energy storage system 1000, which includes an energy storage device 1100 and a cloud platform 1200. The energy storage device 1100 includes an energy management unit 1102, and the cloud platform 1200 is communicatively connected to the energy management unit 1102. The energy management unit 1102 includes an energy control device 800, and the cloud platform 1200 includes an energy dispatching device 900.

[0516] In this invention, the terms "first," "second," and "third" are used for descriptive purposes only and should not be construed as indicating or implying relative importance; the term "multiple" refers to two or more unless otherwise explicitly defined. The terms "install," "connect," "link," and "fix" should be interpreted broadly. For example, "connect" can be a fixed connection, a detachable connection, or an integral connection; "link" can be a direct connection or an indirect connection through an intermediate medium. Those skilled in the art can understand the specific meaning of the above terms in this invention according to the specific circumstances.

[0517] In the description of this invention, it should be understood that the terms "upper," "lower," "left," "right," "front," "rear," etc., indicate the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings. They are only for the convenience of describing this invention and simplifying the description, and do not indicate or imply that the device or unit referred to must have a specific orientation or be constructed and operated in a specific orientation. Therefore, they should not be construed as limitations on this invention.

[0518] In the description of this specification, the terms "one embodiment," "some embodiments," "specific embodiment," etc., refer to specific features, structures, materials, or characteristics described in connection with an embodiment or example that are included in at least one embodiment or example of the present invention. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples.

[0519] The above are merely preferred embodiments of the present invention and are not intended to limit the invention. Various modifications and variations can be made to the present invention by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. An energy control method, characterized in that, An energy management unit applied to an energy storage device, wherein the energy control method includes: While controlling the energy storage device to charge and discharge according to the automatic scheduling instructions issued by the cloud platform, it receives charging and discharging user instructions from the user terminal. In response to the charging and discharging user command, the system switches from an automatic scheduling state to a cooperative transition state. In the cooperative transition state, the energy storage device is controlled to perform charging and discharging according to the fusion result of the automatic scheduling command and the charging and discharging user command. In the fusion result, the control weight of the charging and discharging user command gradually increases over time, while the control weight of the automatic scheduling command gradually decreases over time. When the control weight of the charging and discharging user command reaches a first preset threshold, the system switches from the cooperative transition state to the user-dominated state. In the user-dominated state, the system controls the energy storage device to perform charging and discharging according to the charging and discharging user command. During the user-dominated state, if a preset return condition is detected, the user-dominated state is switched to the collaborative transition state. In the collaborative transition state, the energy storage device is controlled to perform charging and discharging according to the fusion result of the automatic scheduling instruction and the charging and discharging user instruction. In the fusion result, the control weight of the charging and discharging user instruction gradually decreases over time, while the control weight of the automatic scheduling instruction gradually increases over time. When the control weight of the automatic scheduling command reaches the second preset threshold, the system switches back from the collaborative transition state to the automatic scheduling state. In the automatic scheduling state, the system controls the energy storage device to perform charging and discharging according to the automatic scheduling command issued by the cloud platform.

2. The energy control method according to claim 1, characterized in that, The preset return conditions include at least one of the following: the takeover time reaches a preset takeover duration, a user's command to end takeover is received, and the target operating parameter corresponding to the charging / discharging user command reaches a target value. The step of switching from the user-dominated state to the cooperative transition state when the preset return conditions are met during the user-dominated state includes: Obtain the takeover start time corresponding to the user-dominated state, and detect the takeover duration corresponding to the user-dominated state and the operating parameters of the energy storage device; Determine whether the user-dominated state meets the preset return conditions based on the takeover duration, the user's takeover termination instruction, and the operating parameters. Under the condition that the preset return conditions are met, the system switches from the user-led state to the collaborative transition state.

3. The energy control method according to claim 1, characterized in that, The energy control method further includes: If, during the automatic scheduling state, the collaborative transition state, or the user-dominated state, the charging and discharging control parameters of the energy storage device meet the preset safety triggering conditions, the current state is switched to the safety fallback state. Under the aforementioned safety fallback state, the energy storage device is controlled to perform charging and discharging according to the safety operation constraints corresponding to the energy storage device, so that the operating state of the energy storage device is within the safe execution range; The priority of the preset safety trigger condition is higher than that of the charging / discharging user instruction and the automatic scheduling instruction. The preset safety triggering conditions include at least the following: the energy storage device's state of charge reaches the upper limit of the safe state of charge; the energy storage device's state of charge reaches the lower limit of the safe state of charge; the charging power reaches the maximum charging power; the discharging power reaches the maximum discharging power; the grid-connected power reaches the upper limit of the grid-connected power; and the state of charge is lower than the minimum standby charge state.

4. The energy control method according to claim 3, characterized in that, After switching from the current state to the safety fallback state, the energy control method further includes: If the currently effective charge / discharge control parameters no longer meet the preset safety triggering conditions, the control state before entering the safety fallback state is obtained. The target control state is determined based on the control state prior to entering the safety fallback state, so that the energy storage device returns to the automatic scheduling state, the cooperative transition state, or the user-led state.

5. The energy control method according to claim 1, characterized in that, The step of controlling the energy storage device to perform charging and discharging according to the fusion result of the automatic scheduling command and the charging and discharging user command includes: The automatic scheduling power is determined according to the automatic scheduling instruction, and the user's desired power is determined according to the charging and discharging user instruction; Based on the control weights of the automatic scheduling instructions and the control weights of the charging and discharging user instructions, the automatic scheduling power and the user-expected power are weighted to obtain the fused power. The energy storage device is controlled to perform charging and discharging based on the fused power.

6. The energy control method according to claim 5, characterized in that, The step of weighting the automatic scheduling power and the user's desired power according to the control weight of the automatic scheduling instruction and the control weight of the charging / discharging user instruction includes: Obtain the transition start time, transition target weight, and transition time parameters corresponding to the cooperative transition state; Based on the time interval between the current time and the transition start time, and the transition time parameter, the control weight corresponding to the current time is determined according to the preset weight change rule; The automatic scheduling power and the user-expected power are weighted according to the control weight corresponding to the current moment, so that the fused power changes smoothly with the change of the control weight.

7. The energy control method according to claim 6, characterized in that, The preset weight change rule includes at least one of the exponential change rule, the linear change rule, and the piecewise change rule. The preset weight change rule is used to make the control weight gradually change from the current weight to the transition target weight according to the corresponding change trajectory.

8. The energy control method according to claim 1, characterized in that, The energy control method further includes: Obtain the safety operating parameters corresponding to the energy storage device; The safe operating range of the energy storage device is determined based on the safe operating parameters. During the process of controlling the energy storage device to perform charging and discharging according to the fusion result, it is detected whether the operating status parameters of the energy storage device exceed the safe execution range; If any of the aforementioned operating state parameters is detected to exceed the safe execution range, the fusion result is restricted and adjusted so that the adjusted charge and discharge control parameters are within the safe execution range; The energy storage device is controlled to perform charging and discharging according to the adjusted charging and discharging control parameters.

9. The energy control method according to claim 8, characterized in that, The step of adjusting the fusion result to ensure that the adjusted charge / discharge control parameters are within the safe execution range when any of the operating state parameters are detected to exceed the safe execution range includes: If the user's desired power corresponding to the charging / discharging user instruction conflicts with the safe execution range, the user's desired power is limited according to the safe execution range to determine the limited power; The automatic scheduling power and the limited power are weighted according to the control weight of the automatic scheduling instruction and the control weight of the charging and discharging user instruction to determine the limited fusion power; The energy storage device is controlled to perform charging and discharging according to the limited fusion power, so that the energy storage device executes the charging and discharging user command while meeting the safe execution range.

10. The energy control method according to claim 1, characterized in that, The automatic scheduling command is generated by the cloud platform and sent to the energy management unit. The energy control method further includes: The system acquires the operating data of the energy storage device and sends the operating data to the cloud platform, so that the cloud platform can determine the automatic scheduling command based on the operating data. The energy management unit receives the automatic scheduling command issued by the cloud platform and controls the energy storage device to perform charging and discharging according to the automatic scheduling command.

11. The energy control method according to claim 1, characterized in that, The energy control method further includes: The energy management unit reports the device identifier, scene identifier, and device capability information to the cloud platform. The energy management unit receives an adapted automatic scheduling instruction issued by the cloud platform based on the device capability information; The device capability information includes at least: whether it supports AC charging power setting, whether it supports inverter discharge power setting, whether it supports photovoltaic charging power limitation, whether it supports grid-connected power upper limit setting, whether it supports target state of charge setting, and whether it supports local policy caching.

12. The energy control method according to claim 1, characterized in that, The energy control method further includes: Receive constraint preference parameters from the user terminal, the constraint preference parameters including setting a target state of charge, pre-trip reserve power, and power maintenance range; The constraint preference parameters are uploaded to the cloud platform so that the cloud platform can generate automatic scheduling instructions that include the constraint preference parameters.

13. The energy control method according to claim 1, characterized in that, The energy control method further includes: When the energy management unit communicates normally with the cloud platform, at least one of the automatic scheduling instructions, preset weight change rules, and preset return conditions is stored in the energy management unit. In the event of a communication interruption between the energy management unit and the cloud platform, the energy management unit controls the energy storage device to perform charging and discharging based on at least one of the automatic scheduling instructions, the preset weight change rules, and the preset return conditions stored in the energy management unit.

14. The energy control method according to claim 13, characterized in that, In the event of a communication interruption between the energy management unit and the cloud platform, the energy control method further includes: The system receives the charging and discharging user instruction sent by the user terminal, and adjusts the control weight corresponding to the automatic scheduling instruction and the control weight corresponding to the charging and discharging user instruction according to the preset weight change rules stored in the energy management unit. The fusion result is determined based on the adjusted control weights, and the energy storage device is controlled to perform charging and discharging based on the fusion result; Based on the preset return conditions stored in the energy management unit, it is determined whether to switch from the user-led state to the collaborative transition state, and if the control weight corresponding to the automatic scheduling instruction reaches a second preset threshold, it switches to the automatic scheduling state.

15. An energy scheduling method, characterized in that, The method is applied to a cloud platform, which is communicatively connected to the energy management unit of the energy storage device. The energy scheduling method includes: Receive the operating data and equipment capacity information of the energy storage device reported by the energy management unit; Obtain the scheduling data corresponding to the energy storage device, the scheduling data including photovoltaic power generation forecast data, household load forecast data and electricity price data; Based on the operating data, determine the equipment operating constraints corresponding to the energy storage device, and based on the scheduling data and the equipment operating constraints, determine the automatic scheduling control parameters; Automatic scheduling instructions are generated based on the automatic scheduling control parameters, and the automatic scheduling instructions are sent to the energy management unit so that the energy management unit controls the energy storage device to perform charging and discharging according to the automatic scheduling instructions.

16. The energy dispatching method according to claim 15, characterized in that, The energy scheduling method further includes: The system receives constraint preference parameters and / or user takeover records uploaded by the energy management unit. The constraint preference parameters are sent by the user terminal to the energy management unit and uploaded to the cloud platform. The user takeover records include user takeover time, user takeover type, and user takeover duration. If the constraint preference parameters exist, the optimization constraints are determined based on the constraint preference parameters, and a new automatic scheduling instruction is generated and sent to the energy management unit. If the user takeover record exists, determine whether the user takeover behavior meets the preset preference conditions based on the user takeover record within the preset time period; When the user takeover behavior meets the preset preference conditions, user preference parameters are generated based on the user takeover behavior, and the automatic scheduling control parameters are adjusted based on the user preference parameters. A new automatic scheduling instruction is generated based on the adjusted automatic scheduling control parameters and sent to the energy management unit.

17. An energy control device, characterized in that, An energy management unit applied to an energy storage device, wherein the energy control device includes: The instruction receiving module is used to receive charging and discharging user instructions from the user terminal when controlling the energy storage device to charge and discharge according to the automatic scheduling instructions issued by the cloud platform. The collaborative transition module is used to switch from the automatic scheduling state to the collaborative transition state in response to the charging and discharging user command. In the collaborative transition state, the energy storage device is controlled to perform charging and discharging according to the fusion result of the automatic scheduling command and the charging and discharging user command. In the fusion result, the control weight of the charging and discharging user command gradually increases over time, while the control weight of the automatic scheduling command gradually decreases over time. The user-dominated module is used to switch from the cooperative transition state to the user-dominated state when the control weight of the charging and discharging user command reaches a first preset threshold. In the user-dominated state, the module controls the energy storage device to perform charging and discharging according to the charging and discharging user command. During the user-dominated state, if a preset return condition is detected, the user-dominated state is switched to the collaborative transition state. In the collaborative transition state, the energy storage device is controlled to perform charging and discharging according to the fusion result of the automatic scheduling instruction and the charging and discharging user instruction. In the fusion result, the control weight of the charging and discharging user instruction gradually decreases over time, while the control weight of the automatic scheduling instruction gradually increases over time. When the control weight of the automatic scheduling command reaches the second preset threshold, the system switches back from the collaborative transition state to the automatic scheduling state. In the automatic scheduling state, the system controls the energy storage device to perform charging and discharging according to the automatic scheduling command issued by the cloud platform.

18. An energy dispatching device, characterized in that, The energy dispatching device is applied to a cloud platform, which is communicatively connected to the energy management unit of the energy storage device. The data receiving module is used to receive the operating data and equipment capacity information of the energy storage device reported by the energy management unit; The data scheduling module is used to acquire scheduling data corresponding to the energy storage device. The scheduling data includes photovoltaic power generation forecast data, household load forecast data, and electricity price data. The constraint determination module is used to determine the equipment operation constraints corresponding to the energy storage device based on the operation data, and to determine the automatic scheduling control parameters based on the scheduling data and the equipment operation constraints. The instruction issuing module is used to generate an automatic scheduling instruction based on the automatic scheduling control parameters, and issue the automatic scheduling instruction to the energy management unit so that the energy management unit controls the energy storage device to perform charging and discharging according to the automatic scheduling instruction.

19. An energy storage system, characterized in that, The energy storage system includes an energy storage device and a cloud platform. The energy storage device includes an energy management unit, and the cloud platform is communicatively connected to the energy management unit. The energy management unit includes the energy control device as described in claim 17; The cloud platform includes the energy dispatching device as described in claim 18.

Images