An intelligent power dispatching method for a wind power energy storage system

Abstract

The application discloses an intelligent power dispatching method of a wind power generation energy storage system, and relates to the technical field of wind power generation. The method extracts electrolytic cell module actual state process characteristic data, converts the data into four state data such as heat state maintenance and residual gas occupation, and determines to generate absorption, maintenance, release or isolation seats. The method reads wind power output, local load and energy storage state, reorganizes seat groups and generates a power acceptance seat chain. The method allocates power along the power acceptance seat chain, preferentially guarantees local necessary load, distributes residual power to the absorption seat group, limits the energy storage unit to provide heat preservation current to the maintenance seat group in combination with the energy storage supportable state, and intercepts power overstepping to flow into the release seat group. According to the execution result, the seat migration operation rule is executed to generate a next period seat to trigger a closed loop. The application converts bottom process constraints into dispatching boundaries, and avoids the misplacement of power supply and equipment acceptance capacity.

Description

Technical Field

[0001] This invention relates to the field of wind power generation technology, and more specifically to an intelligent power dispatching method for a wind power generation energy storage system. Background Technology

[0002] With the increasing demand for large-scale integration and consumption of new energy sources, coupled systems of wind power generation, energy storage, and water electrolysis for hydrogen production under off-grid and weak grid conditions have become a key development direction. In complex microgrid environments, wind power output exhibits significant intermittency and volatility, requiring electrolyzer clusters to function as hubs for smoothing energy fluctuations and acting as large-capacity adjustable loads. Wind-storage-hydrogen production systems constitute a dynamic control network with deep coupling between source, grid, load, and storage. This requires intelligent dispatching systems to coordinate wind power output destinations, control energy storage unit throughput, and rationally allocate hydrogen production loads from the underlying electrolyzer clusters within extremely short timeframes, thereby maintaining real-time balance of the system's bus active power.

[0003] Existing smart grid dispatching technologies, when dealing with wind-storage-hydrogen production systems, generally treat chemical equipment as an ideal adjustable load and use simple follow-up logic for power distribution. When wind power is abundant, the system directly instructs the electrolyzer to absorb power at its rated upper limit; when wind power is insufficient, the load is reduced and energy storage is used for compensation. This mechanism severely ignores the complex gas-liquid multiphase flow dynamics and electrochemical process constraints inside the electrolyzer. There is a significant separation between the external hot-state operability of the electrolyzer and the actual internal state of sustainable acceptance of new power supply. When the electrolyzer is restarted after experiencing low-load operation, hot-state maintenance, or power withdrawal, its external indicators such as tank temperature still meet the hot-state start-up criteria, but the internal separation chamber often retains a large amount of residual gas, the gas-liquid separation space is occupied, and the circulating medium has not yet re-established a stable flow field. Existing dispatching systems cannot perceive the actual process boundary of the unreleased internal acceptance space and still issue new power supply commands according to the nominal rated power of the equipment, resulting in serious failure of dispatching commands at the execution level. Misjudged electrolyzers, due to their limited internal space, cannot handle sudden power surges, easily leading to converter disconnection or abnormal operation of exhaust valves. This can cause a large amount of surplus wind power to flow back into the microgrid as surge currents or force energy storage devices to withstand over-limit surges. A critical technical challenge under off-grid and weak grid conditions is how to transform the complex residual gas occupancy and process reset states within the electrolyzer into a directly analyzable basis for the dispatch system's acceptance order. This would prevent equipment in the occupancy release phase from being misjudged as immediately available loads, thereby eliminating control failures and grid oscillations caused by the misalignment between system power dispatch and the actual load capacity of equipment. Summary of the Invention

[0004] To address the shortcomings of existing technologies, this invention proposes an intelligent power dispatching method for wind power energy storage systems. This method solves the problem that existing dispatching systems cannot detect the occupancy of residual gas in the separation chamber inside the electrolyzer and the process reset status. They are prone to misjudging equipment in the occupancy release stage as immediately available loads, leading to a misalignment between power dispatching and the actual load-bearing boundary of the underlying equipment, resulting in control failure and grid oscillation.

[0005] To achieve the above objectives, the present invention provides the following technical solution: Extract real-state process characteristic data of the electrolytic cell module and convert it into hot state holding state data, residual gas occupancy state data, cycle reset state data, and current load acceptance state data. Based on the four state data, execute logic to determine and generate the current seat category, including absorption seat, maintenance seat, release seat, and isolation seat. Read the current wind power output, local necessary load, and energy storage support status. Based on the current seat category, reorganize the absorption seat group, maintenance seat group, release seat group, and isolation seat group to generate a power absorption seat chain. The matching order is absorption seat group, maintenance seat group, release seat group, and isolation seat group. Distribute power along the power acceptance seat chain, prioritize local necessary loads with the current wind power output, allocate the remaining allocable power to the absorption seat group, combine the energy storage unit to provide insulation current to the maintenance seat group and transfer the surplus wind power into the energy storage unit, intercept the excess power flowing into the release seat group, and output the actual execution result that represents the response of the underlying equipment. Based on the current seat category and the actual execution result, the seat migration operation rules are executed to generate the seat category for the next cycle. The input register is written back as the basic input for reconstructing the power admission seat chain in the next cycle to trigger a new scheduling loop.

[0006] Furthermore, the actual process characteristic data of the electrolytic cell module are extracted, including: Read continuous physical analog quantities and compare them with preset safe operation physical thresholds to obtain threshold comparison results; The underlying controller performs anti-jitter filtering based on the local scan cycle. Only when the threshold comparison result remains consistent within the set continuous time step, the underlying controller confirms and refreshes the status parameter value in the status register, which is then latched and read by the upper-level intelligent scheduling system within the scheduling clock cycle.

[0007] Furthermore, this data is converted into thermal hold-up status data, residual gas occupancy status data, cyclic reset status data, and current load acceptance status data, including: When the underlying controller determines that the current tank temperature is higher than the set insulation threshold, it generates thermal state maintenance data with thermal basis; otherwise, it generates thermal state maintenance data without thermal basis. When the underlying controller determines that the pressure inside the separation chamber drops to the ambient pressure threshold or the exhaust valve has been continuously open for a preset time, it generates residual gas occupancy status data with no residual gas occupancy; otherwise, it generates residual gas occupancy status data with occupancy. When the underlying controller determines that the circulating pump has reached the lower limit of the rated speed, it generates circulating reset state data indicating that the circulating pump is in a reset state; otherwise, it generates circulating reset state data indicating that the pump is not in a reset state. When the underlying controller determines that the current active power is greater than the standby insulation power threshold, it generates current load acceptance status data indicating that the system is in actual operation and acceptance. Otherwise, it generates current load acceptance status data indicating that the system is not in actual operation and acceptance.

[0008] Furthermore, the specific rules for determining the current seat category based on the four status data are as follows: When the data for maintaining the hot state has a hot state basis, the data for the residual gas occupancy state has no residual gas occupancy, the data for the cyclic reset state is in the reset state, and the data for the current load acceptance state is in the actual operation acceptance state, an absorption seat is generated. A maintenance position is generated when the hot-state maintenance data has a hot-state basis, the residual gas occupancy status data has no residual gas occupancy, the cyclic reset status data is in the reset state, and the current load acceptance status data is not in the actual operation acceptance state. A release seat is generated when the hot-state holding data has a hot-state basis and the residual gas occupancy state data is in an occupied state or the cyclic reset state data is not in a reset state. If the conditions for generating, maintaining, or releasing seats are not triggered, an isolated seat will be generated.

[0009] Furthermore, power is distributed along the power acceptance seat chain, including: The intelligent scheduling system treats multiple electrolytic cell modules within the same seat group as an aggregated adjustable point and issues overall instructions accordingly. The fine current sharing and micro power allocation of each module in the group are executed based on the original current sharing control strategy of the underlying grid-connected converter. Based on the matching order determined in the power acceptance seat chain, mandatory routing isolation is solidified within the controller to prevent grid-side power dispatch commands from being sent to the electrolytic cell module which is in the internal process recovery period.

[0010] Furthermore, the output represents the actual execution result that characterizes the response of the underlying device, including: The actual execution results recorded and output by the intelligent scheduling system include new acceptance result components, hot state maintenance result components, release continuous result components, and isolation result components; The actual execution result is generated by hard-wired mapping of the current status fed back from the underlying converter and the action result of the exhaust valve by the intelligent scheduling system. The data synchronously output by the intelligent dispatch system includes the execution type of the energy storage unit in this cycle and the destination record of wind power generation in this cycle.

[0011] Furthermore, in conjunction with the energy storage support status, surplus wind power is transferred to energy storage units, including: The intelligent dispatch system analyzes the charge / discharge interlock permission flag in the energy storage's supportable state; When the intelligent dispatch system determines that the allow charging flag in the charge-discharge interlock permission flag is valid and the remaining space meets the requirements, the intelligent dispatch system issues a surplus absorption instruction to the energy storage unit, transferring the surplus wind power that cannot be accepted by the absorption seat group into the energy storage unit. When the intelligent dispatch system determines that the charging permission flag is invalid or the remaining space is insufficient, the intelligent dispatch system issues a power limiting load reduction or pitch control command to the wind power generation unit through the control loop to forcibly reduce the wind power output.

[0012] Furthermore, the energy storage unit, in conjunction with the energy storage support condition limited energy storage unit, provides insulation current to the maintenance seat group, including: When wind power generation decreases and the already activated absorption or maintenance positions still require power supply, the intelligent dispatch system determines that the allow discharge flag in the charge-discharge interlock permission flag is valid, and the intelligent dispatch system issues a continuous support instruction to the energy storage unit. The intelligent scheduling system prioritizes continuous support commands by prioritizing absorption seats that have already taken on new loads over maintenance seats that need to be kept in a hot state. Furthermore, the intelligent scheduling system prohibits releasing seats to receive new support input commands.

[0013] Furthermore, the path for generating the seat category for the next cycle by executing the seat migration operation rules includes: When the current seat category is a release seat, the actual execution result indicates that the release of the seat in this cycle has been completed, and the underlying sensor indicates that the current tank temperature is still higher than the set insulation threshold, the intelligent scheduling system will assign the seat category to a maintenance seat for the next cycle. When the current seat category is a maintenance seat, and the actual execution result feedback indicates that the hot state was successfully maintained in this cycle, there is no residual gas occupying the seat at the end of the cycle, and the cycle is in a reset state, the intelligent scheduling system will assign the seat category of the next cycle to an absorption seat. When the current seat category is an absorption seat, and the actual execution result shows that the new power supply was taken over in this cycle but residual gas occupancy reappears at the end of the cycle or the cycle is not reset, the intelligent scheduling system will revert the seat category to a release seat for the next cycle.

[0014] Furthermore, the path for generating the seat category for the next cycle by executing the seat migration operation rules also includes: When the underlying controller of the electrolytic cell module reports hardware fault codes such as converter disconnection and water pump failure, or when the electrolytic cell module is in the release position and the venting is not completed for a continuous period exceeding the set maximum waiting period threshold, the intelligent scheduling system will assign the position category of the next period to the isolation position. When the electrolytic cell module does not trigger the state change condition, the intelligent scheduling system executes the state self-maintaining path, and the seat category of the next cycle inherits the current seat category; For the electrolytic cell module in the release position, the intelligent scheduling system simultaneously performs an accumulation operation on the state stagnation counter when executing the state self-holding path, providing a definite time-domain calculation basis for triggering the set maximum waiting period threshold.

[0015] Compared with existing technologies, it has the following advantages: This solution proposes an intelligent power dispatching method for wind power energy storage systems. It overcomes the inherent limitation of traditional microgrid control that treats water electrolysis hydrogen production equipment as a completely ideal adjustable load, solving the control failure problem caused by a severe misalignment between power supply commands and the actual load-bearing limits of equipment in off-grid and weak-grid scenarios. The core innovation of this solution lies in accurately abstracting and reshaping the underlying process conditions—such as the hidden residual gas occupancy, separation chamber margin, and fluid circulation reset—into directly analyzable power receiving slot categories on the grid side. By constructing a multi-dimensional acceptance order encompassing absorption, maintenance, release, and isolation, this method ensures that the power allocation logic of the top-level dispatch is strictly controlled by the actual thermodynamics and gas-liquid multiphase flow tolerance limits of the underlying equipment. The intelligent dispatching system no longer blindly issues new power supply commands based on the nominal rated limits of the equipment or a single temperature phenomenon, thus completely eliminating the potential risks of converter disconnection and inrush current backflow caused by equipment in the process recovery period being forced to bear sudden load changes.

[0016] Based on the precise mapping of the underlying state and dynamic seat chain reorganization, this invention achieves deep coordination and refined management of power generation, grid, load, and storage at the system-level power dispatching dimension. In complex and fluctuating environments with surges or drops in wind power generation, the dispatching system can accurately direct surplus power to generating units with substantial capacity based on a defined topology routing mechanism, and limit energy storage units to providing stable support only to specific units that need to maintain a hot state. This mechanism not only effectively curbs losses in the energy storage system during ineffective charge-discharge cycles but also ensures continuous power supply to equipment that has established stable hydrogen production conditions. By transforming discrete power allocation into a closed-loop iterative state machine model on the time axis, this solution self-corrects the dispatching path for the next cycle based on the actual power supply response results at the underlying level, ensuring high continuity and stability of the complex microgrid system at the execution level, and effectively improving the operational safety and energy conversion efficiency of the entire wind-storage-hydrogen production chain. Attached Figure Description

[0017] Figure 1 This is a schematic diagram of the method flow of the present invention. Detailed Implementation

[0018] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0019] Please see Figure 1 This application provides an intelligent power dispatching method for a wind power generation energy storage system; The method specifically includes the following steps: Step 1: Perform the operation of converting the electrolytic cell group into power supply acceptance seats. The intelligent scheduling system extracts the underlying state and classifies and reshapes the real-time acceptance eligibility of each electrolytic cell module in the system.

[0020] At the start of the current scheduling cycle, the intelligent scheduling system latches and reads the actual state process characteristic data of the i-th controlled electrolytic cell module within the same scheduling clock cycle. The station's underlying equipment-level controller reads the cell temperature sensor data, the gas-liquid separator exhaust valve feedback command, and the circulating pump inverter operating frequency, comparing the continuous physical analog quantities with preset safe operating physical thresholds. The comparison results are converted into four binary state parameters in the form of communication messages and uploaded to the intelligent scheduling system. These four state parameters represent the thermal hold-up state data. Residual gas occupancy status data Cyclic reset status data and current load acceptance status data .

[0021] Specifically, the i-th electrolyzer module refers to the specific physical unit in the wind power generation system that participates in the controlled response to wide power fluctuations during water electrolysis to produce hydrogen. (Hot state data) This refers to the status parameter where the underlying controller outputs 1 when the current tank temperature is higher than the set insulation threshold, and 0 otherwise. It characterizes whether the module has the foundation for hot-state operation. It should be noted that this set insulation threshold is a system preset threshold directly determined by technicians based on the thermodynamic characteristics and rated factory parameters of a specific model of alkaline or proton exchange membrane electrolyzer; its specific value varies depending on the equipment model. Residual gas occupancy status data. This refers to the state parameter where the underlying controller outputs 1 when it determines that the separation chamber has not completed the preset venting procedure, and 0 otherwise. The physical criterion for completing the preset venting procedure is that the internal pressure of the separation chamber drops to the ambient pressure threshold or the vent valve remains open for a preset duration. The residual gas occupancy status data characterizes whether there is residual gas occupying the gas-liquid separation space inside the module that could affect power ramp-up. The aforementioned ambient pressure threshold and preset duration are conventional parameters that can be directly determined through a limited number of routine experiments, combining basic fluid mechanics principles with the aerodynamic characteristics of the specific equipment's vent valve. (Cycle reset state data) This refers to the status parameter where the underlying controller outputs 1 when it determines that the circulating pump has reached the lower limit of its rated speed, and 0 otherwise. It characterizes whether the module's gas-liquid separation and circulating medium path have returned to a stable acceptable state. Current load acceptance status data. This refers to the status parameter of the underlying converter, which outputs 1 when the current active power is greater than the set standby insulation power threshold and 0 otherwise, indicating whether the module is currently undertaking actual hydrogen production load.

[0022] To meet the data parsing requirements between the lower-level controller and the upper-level scheduling system, the above four binary state parameters correspond to clear text state interpretations: a value of 1 for the thermal hold state indicates the presence of a thermal foundation, while a value of 0 indicates the absence of a thermal foundation; a value of 0 for the residual gas occupancy state indicates the absence of residual gas occupancy, while a value of 1 indicates the presence of occupancy; a value of 1 for the cyclic reset state indicates that the cycle is in a reset state, while a value of 0 indicates that the cycle is not in a reset state; and a value of 1 for the current load acceptance state indicates that the load is in a substantive operating acceptance state, while a value of 0 indicates that the load is not in a substantive operating acceptance state.

[0023] The standby power threshold is the minimum rated power boundary required to maintain the thermal balance of the corresponding electrolytic cell, which can be obtained directly from the equipment technical manual. The core of this implementation lies in the logical reorganization of the above-mentioned state parameters and the complete reconstruction of the power supply acceptance order.

[0024] Electromagnetic interference or sensor mechanical vibration in industrial settings can easily cause high-frequency jumps in the acquired signals. The lower-level controller performs anti-jitter filtering based on a local scan cycle. Only when the threshold comparison result remains consistent within a set continuous time step does the lower-level controller confirm and refresh the status parameter value in the status register, allowing the upper-level intelligent scheduling system to latch and read it within the scheduling clock cycle. The local filtering and clock latching mechanism under the hierarchical control architecture can avoid logic decision distortion caused by communication delays or signal glitches, ensuring that process fluid dynamics constraints are stably acquired by the grid-side scheduling system.

[0025] The intelligent scheduling system is based on the collected parameters , , and Perform deterministic logic gating to assign an exclusive current seat category to the i-th electrolytic cell module. The decision-making logic is based on the following four sets of mutually exclusive conditional branches.

[0026] When the system determines , , and When both conditions are met, the system outputs... To accommodate new wind power storage loads, the following criteria are established: the module has a thermal foundation and its internal gas occupancy has been cleared; the circulation is stable and the module is currently in a state of substantive operation and can accept additional power fluctuations.

[0027] When the system determines , , and When both conditions are met, the system outputs... To maintain the seat, the system filters out idle modules whose internal occupancy has been released and whose loop has been reset, and which are in a hot state but not currently connected to any actual workload. The constraint system prioritizes sending hot-state hold-up low-power commands to such modules to prevent them from cooling down.

[0028] When the system determines and satisfy or In any state, the system outputs To release seats. The judgment criteria are for physical control of transitional equipment that appears to be in a hot state but has not vented residual gas or established a circulating flow field. The system prohibits the supply of new power to it.

[0029] If any of the above conditions are not met, the system outputs... These are for isolated seating.

[0030] Specifically, the current seat category This refers to the power constraint status flag assigned to the electrolyzer module within this scheduling cycle. The "Absorb" position corresponds to a scheduling status that allows the converter to receive additional set power and convert it into continuous hydrogen production load. The "Maintain" position corresponds to a scheduling status that only allows the converter to output the standby power required to maintain its hot state and does not accept ramp-up loads. The "Release" position corresponds to a protective transitional status that controls the converter to maintain zero power output while waiting for the venting of internal residual gas. The "Isolate" position corresponds to a disconnected status that disconnects the main hydrogen production power circuit electrical connection in this cycle and does not participate in scheduling, while retaining the module's eligibility to receive power for its auxiliary heating circuit.

[0031] After completing the traversal calculation of all controlled electrolytic cell modules within the control domain, the intelligent scheduling system packages the identity flags of all modules and outputs them as the current seat category result set. The current seat category result set is written into the storage unit as the basic input parameter for generating the subsequent power acceptance seat chain, triggering the system power supply diversion path reconfiguration action.

[0032] Specifically, traversal operation refers to the iterative operation by which the intelligent scheduling system executes logical comparison instructions one by one for each electrolytic cell module within the control domain. The current seat category result set is composed of the acceptance identity flags of all modules, forming a system-level state matrix. The underlying discrete process physical state is thus transformed into a power grid scheduling data structure. In subsequent steps, the scheduling system implements power distribution solely based on this state matrix, ensuring the continuity and stability of power allocation at the physical execution level.

[0033] Step Two: Execute the operation of constructing a power acceptance seat chain based on seat category. After renaming the identities of each electrolyzer module, the intelligent scheduling system reads the current wind power output Pw, local necessary load Pn, and energy storage support status Eb via the industrial communication bus. Simultaneously, the system's central processing unit retrieves the current seat category of each electrolyzer module generated in the previous steps. Based on the current seat category The intelligent scheduling system reorganizes the entire electrolytic cell group into four mutually exclusive physical device sets: the absorption seat group (GA), the maintenance seat group (GM), the release seat group (GR), and the isolation seat group (GX). Based on this, the intelligent scheduling system generates a power admission seat chain (SC) in memory. Within the power admission seat chain (SC), the traversal hierarchy and rule matching order of the power supply control strategy are set from highest to lowest as follows: absorption seat group (GA), maintenance seat group (GM), release seat group (GR), and isolation seat group (GX).

[0034] Specifically, the current wind power output Pw refers to the real-time active power output from the wind power generation unit to the system bus, with data sourced from wind power grid connection measurement points or wind farm control systems. Local necessary load Pn refers to the total basic power supply load that must be prioritized and guaranteed in off-grid or weak-grid systems. Energy storage support status Eb is a state vector reported by the energy storage management system. The state vector data structure contains continuous numerical fields representing available remaining energy, and Boolean flags representing the charge / discharge interlock permission status. The power acceptance seat chain SC is a dynamically generated ordered routing list in the memory of the intelligent dispatch system. The absorption seat group GA, maintenance seat group GM, release seat group GR, and isolation seat group GX respectively map to the electrolytic cell physical arrays with corresponding acceptance identity flags within the current dispatch cycle.

[0035] In the Power Admission Seat Chain (SC) operation mechanism, multiple modules within the same seat group are treated as an aggregated adjustable point by the intelligent scheduling system for overall command issuance. Fine-grained current sharing and micro-power allocation within the group rely on the existing current sharing control strategy of the underlying grid-connected converter, avoiding the waste of computing power caused by secondary micro-optimization at the global macro-scheduling layer. Simultaneously, in this step, the intelligent scheduling system only considers the current seat category. Perform topology logic grouping and channel construction for physical devices. The current wind power output Pw, local necessary load Pn, and energy storage support status Eb, which are captured synchronously, are independent of the linked list sorting process in this step. They are directly encapsulated together with the constructed power acceptance seat chain SC to form the data environment for the next scheduling stage, providing the necessary numerical calculation basis for subsequent physical power diversion.

[0036] The underlying control logic for constructing the power admission seat chain (SC) involves transforming the previously chaotic and disordered underlying process physical states into a unified global network-side access control list. This access control list defines the rule matching hierarchy of the global topology and binds specific access constraints to each level. Specifically, the scheduling system sequentially traverses the access control list, opening the admission channel for newly allocated power to the top-ranked absorption seat group (GA); issuing only channel commands to the second-ranked maintenance seat group (GM) to acquire the maintenance power required to maintain a hot state; and executing hard blocking of power admission and hard isolation of electrical loops to the third and fourth-ranked release seat group (GR) and isolation seat group (GX), respectively. By including nodes lacking admission conditions at the end of the chain and assigning them a forced blocking label, a robust access interception mechanism is formed. The defined rule matching order solidifies mandatory routing isolation within the controller, preventing grid-side scheduling power commands from being bypassed and sent to the electrolyzer module during its internal process recovery phase. This ensures that the subsequently generated actual physical diversion paths accurately match the real-time gas-liquid thermodynamic capacity of the underlying electrolyzer.

[0037] Step 3: Execute power distribution along the power acceptance seat chain, including wind power generation, energy storage, and electrolyzer clusters. After encapsulating the data environment, the intelligent dispatch system reads the current wind power output Pw, local necessary load Pn, energy storage support status Eb, and the power acceptance seat chain SC. The system forcibly distributes power step-by-step along the set topology order, specifically executing the following five power supply actions: The first power supply action prioritizes ensuring necessary local power supply. The intelligent dispatch system compares Pw and Pn. When At that time, wind power generation directly supplies the local load. At this time, the intelligent dispatch system issues a power supply support command to the energy storage unit, which then compensates for the power shortage. In this scenario, the energy storage unit's output prioritizes only the necessary local load. In off-grid or weak grid systems, the supply of necessary local power takes precedence over the absorption of hydrogen production load.

[0038] The second and third power supply actions respectively direct power input to the absorption seat group GA and the maintenance seat group GM. With the necessary local load already guaranteed, the intelligent dispatch system prioritizes allocating remaining allocable power to the absorption seat group GA in the power absorption seat chain SC. The system issues a new power supply input command to the electrolyzer module corresponding to the absorption seat group GA, allowing the corresponding module to enter the new acceptance state and take on new hydrogen production load. When the absorption seat group GA has completed the new acceptance allocation, and the system needs to keep some electrolyzer modules in a state of rapid recovery, the intelligent dispatch system issues a hot-state maintenance power supply command to the maintenance seat group GM, limiting the supply of a stable heat preservation current from the energy storage unit through the DC bus. The hot-state maintenance power supply command controls the module to maintain a hot state, limiting its ability to take on new hydrogen production load, creating conditions for seat migration in the next cycle. The execution process of the hot-state maintenance command avoids prematurely pulling modules that are close to availability but have not yet officially entered acceptance into the new load, thus re-triggering the internal occupancy problem.

[0039] The fourth and fifth power supply actions execute energy storage interaction responses based on safety constraints. When the absorption seat group GA has no additional capacity but wind power generation still has surplus capacity, the intelligent dispatch system analyzes the charge / discharge interlock permission flag in the energy storage support state Eb. If the allowed charging flag is determined to be valid and the remaining space meets the requirements, the system issues a surplus absorption command to the energy storage unit, transferring the surplus wind power that cannot be absorbed by the absorption seat group into the energy storage. If the allowed charging flag is determined to be invalid or the remaining space is insufficient, the system issues a power limiting load reduction or pitch adjustment command to the wind power generation unit through the control loop, forcibly reducing wind power output to maintain the real-time balance of active power on the system bus. The system intercepts the action of surplus power flowing into the release seat group GR through the control logic and physical loop, avoiding direct impact of wind power on the electrolyzer module that is still in the occupied release state. In subsequent dispatch cycles, if wind power generation decreases but the already put into absorption or maintenance seats still need to supply power, the system issues a continuous support command to the energy storage unit when it determines that the allowed discharge flag in Eb is valid. The priority of continuous support commands is defined as follows: the absorption seat group (GA) that has already taken on new load takes precedence over the maintenance seat group (GM) that needs to be kept in a hot state. The release seat group (GR) is prohibited from receiving new support command to ensure that the energy storage unit does not mistakenly pull up modules that are in the pre-emptive release phase.

[0040] Specifically, the above power supply diversion path DP can be defined by the control expression to specify the flow constraints: wind power generation is first used for necessary local loads; surplus power enters the absorption seat group GA; surplus power that cannot immediately enter the absorption seat group is absorbed by the energy storage unit; the maintenance seat group GM is supported by the energy storage unit only when it is necessary to maintain thermal continuity.

[0041] After the aforementioned physical distribution, the intelligent scheduling system records and outputs the actual execution result of the i-th electrolytic cell module in this cycle. Actual execution results This includes a new acceptance result component indicating whether new power supply was actually received in this cycle, a hot-state maintenance result component indicating whether only hot-state holding was performed, a release continuity result component indicating whether the device was still in a preemptive release state after the end of this cycle, and an isolation result component indicating whether it did not enter the power supply acceptance chain. The assignment of each component of the actual execution result vector is not based on theoretical deduction from upper-level instructions, but is generated by hard-wired mapping by reading the actual current state feedback from the lower-level converter and the exhaust valve action results, ensuring the physical authenticity of the state record. The synchronously output data also includes the execution type of the energy storage unit in this cycle and the destination record of wind power generation in this cycle.

[0042] The power supply organization between wind power generation, energy storage, and hydrogen production flexible loads is completed along the seat chain formed by the characteristics of hot residual gas occupancy, ensuring that power diversion is strictly controlled by the underlying thermodynamics and gas-liquid bearing boundaries. Initial seat category Compared with actual execution results It will be directly extracted and written to the shared storage space as the calculation input for determining which type of seat each electrolytic cell module will migrate to at the end of this cycle.

[0043] Step 4: Perform seat migration based on the execution results and form the entry point for the next scheduling cycle. After the power supply diversion action in the current cycle is completed, the intelligent scheduling system reads the initial seat category of the i-th electrolytic cell module at the beginning of the current cycle from the shared storage space. and the actual execution results at the end of the current cycle. The central processing unit (CPU) performs state transition calculations for each electrolytic cell module according to the seat migration operation rule TR, and generates its seat category for the next cycle. .

[0044] Specifically, the seat migration operation rule TR contains deterministic conditional branch instructions. When the current seat category of the i-th electrolytic cell module... To release seats, the actual execution result If the current cycle's seat release has been completed, and the underlying sensors indicate that the current tank temperature is still higher than the set insulation threshold, the system will change the seat category for the next cycle. The current module is assigned a maintenance seat; if the end-of-period review finds that the tank temperature has fallen below the set insulation threshold, the system will downgrade the corresponding module to an isolation seat. (The current seat category is...) To maintain the seat, and the actual results of the implementation When the system successfully maintains the hot state during the current cycle, detects no residual gas occupying space at the end of the cycle, and the cycle is in a reset state, the system will classify the seats for the next cycle. Assign a value to the absorbed seats. This applies to the current seat category. To attract seats, and the actual results of the implementation If the system receives additional power supply during the current cycle but residual gas occupancy reappears at the end of the cycle or the cycle fails to reset, the system will adjust the seat category for the next cycle. The rollback assignment is set to the release position. This occurs when any module's underlying controller reports hardware fault codes such as converter disconnection or pump failure, or when the module remains in the release position and venting is incomplete for an extended period exceeding the set maximum waiting period threshold. At that time, the system will classify the seats for the next cycle. The value is assigned to the isolation seat.

[0045] Among them, the maximum waiting period threshold The time boundary engineering tuning parameters, set according to the station's process safety specifications and equipment shutdown protection strategies, can be flexibly configured by technicians based on actual on-site conditions. When the module status does not trigger any of the aforementioned status change conditions, the system executes the status self-maintaining path, i.e., the next cycle's seat category. Inherit current seat category The system simultaneously increments the state retention counter of modules in the release position while executing the state self-holding path, setting it to the maximum waiting period threshold. The triggering provides a definite basis for time-domain calculations.

[0046] Specifically, initial seat categories The representation module's power acceptance identity in the current cycle, and the actual execution result. Characterizes the underlying device response, specifically the physical boundary changes triggered by the acceptance identity after a physical power supply action. Next cycle seat category. The fundamental variables for the new round of scheduling are written back to the system's input registers. The seat migration operation rule TR is a rule-based code logic that can be directly executed by the programmable logic controller or the central processing unit, and does not use machine learning models or empirical fuzzy functions that include implicit weight allocation. For modules in isolated seats that have not produced substantial operational results, the wake-up action of the module returning to the admission chain relies on the re-polling determination of the underlying physical sensors in the initial stage of the next cycle, and is not within the scope of the migration driven by the execution result in this step.

[0047] The underlying control logic driving seat migration based on actual physical execution results involves transforming discrete single power allocations into a continuous state machine model over time. The system solidifies the actual occupancy and thermodynamic changes resulting from the current cycle's wind power supply, energy storage support, and electrolyzer hydrogen production into an acceptance identity that can be directly read in the next cycle. The newly generated seat category for the next cycle... This triggers the system to enter a new round of power acceptance seat chain reconfiguration and power supply diversion operations, thereby constructing a self-driven intelligent scheduling closed loop based on underlying real-world process characteristic data.

[0048] The above embodiments are only used to illustrate the technical methods of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical methods of the present invention without departing from the spirit and scope of the technical methods of the present invention.

Claims

1. A smart power dispatching method for a wind power generation and energy storage system, characterized in that, include: Extract real-state process characteristic data of the electrolytic cell module and convert it into hot state holding state data, residual gas occupancy state data, cycle reset state data, and current load acceptance state data. Based on the four state data, execute logic to determine and generate the current seat category, including absorption seat, maintenance seat, release seat, and isolation seat. Read the current wind power output, local necessary load, and energy storage support status. Based on the current seat category, reorganize the absorption seat group, maintenance seat group, release seat group, and isolation seat group to generate a power absorption seat chain. The matching order is absorption seat group, maintenance seat group, release seat group, and isolation seat group. Distribute power along the power acceptance seat chain, prioritize local necessary loads with the current wind power output, allocate the remaining allocable power to the absorption seat group, combine the energy storage unit to provide insulation current to the maintenance seat group and transfer the surplus wind power into the energy storage unit, intercept the excess power flowing into the release seat group, and output the actual execution result that represents the response of the underlying equipment. Based on the current seat category and the actual execution result, the seat migration operation rules are executed to generate the seat category for the next cycle. The input register is written back as the basic input for reconstructing the power admission seat chain in the next cycle to trigger a new scheduling loop.

2. The intelligent power dispatching method for a wind power energy storage system according to claim 1, characterized in that, Extract the actual process characteristic data of the electrolytic cell module, including: Read continuous physical analog quantities and compare them with preset safe operation physical thresholds to obtain threshold comparison results; The underlying controller performs anti-jitter filtering based on the local scan cycle. Only when the threshold comparison result remains consistent within the set continuous time step, the underlying controller confirms and refreshes the status parameter value in the status register, which is then latched and read by the upper-level intelligent scheduling system within the scheduling clock cycle.

3. The intelligent power dispatching method for a wind power energy storage system according to claim 1, characterized in that, Converted into hot-state holding status data, residual gas occupancy status data, cyclic reset status data, and current load acceptance status data, including: When the underlying controller determines that the current tank temperature is higher than the set insulation threshold, it generates thermal state maintenance data with thermal basis; otherwise, it generates thermal state maintenance data without thermal basis. When the underlying controller determines that the pressure inside the separation chamber drops to the ambient pressure threshold or the exhaust valve has been continuously open for a preset time, it generates residual gas occupancy status data with no residual gas occupancy; otherwise, it generates residual gas occupancy status data with occupancy. When the underlying controller determines that the circulating pump has reached the lower limit of the rated speed, it generates circulating reset state data indicating that the circulating pump is in a reset state; otherwise, it generates circulating reset state data indicating that the pump is not in a reset state. When the underlying controller determines that the current active power is greater than the standby insulation power threshold, it generates current load acceptance status data indicating that the system is in actual operation and acceptance. Otherwise, it generates current load acceptance status data indicating that the system is not in actual operation and acceptance.

4. The intelligent power dispatching method for a wind power energy storage system according to claim 3, characterized in that, The specific rules for determining the current seat category based on the four status data are as follows: When the data for maintaining the hot state has a hot state basis, the data for the residual gas occupancy state has no residual gas occupancy, the data for the cyclic reset state is in the reset state, and the data for the current load acceptance state is in the actual operation acceptance state, an absorption seat is generated. A maintenance position is generated when the hot-state maintenance data has a hot-state basis, the residual gas occupancy status data has no residual gas occupancy, the cyclic reset status data is in the reset state, and the current load acceptance status data is not in the actual operation acceptance state. A release seat is generated when the hot-state holding data has a hot-state basis and the residual gas occupancy state data is in an occupied state or the cyclic reset state data is not in a reset state. If the conditions for generating, maintaining, or releasing seats are not triggered, an isolated seat will be generated.

5. The intelligent power dispatching method for a wind power energy storage system according to claim 1, characterized in that, Distributing power along the power admission seat chain includes: The intelligent scheduling system treats multiple electrolytic cell modules within the same seat group as an aggregated adjustable point and issues overall instructions accordingly. The fine current sharing and micro power allocation of each module in the group are executed based on the original current sharing control strategy of the underlying grid-connected converter. Based on the matching order determined in the power acceptance seat chain, mandatory routing isolation is solidified within the controller to prevent grid-side power dispatch commands from being sent to the electrolytic cell module which is in the internal process recovery period.

6. The intelligent power dispatching method for a wind power energy storage system according to claim 1, characterized in that, The output represents the actual execution result that characterizes the response of the underlying device, including: The actual execution results recorded and output by the intelligent scheduling system include new acceptance result components, hot state maintenance result components, release continuous result components, and isolation result components; The actual execution result is generated by hard-wired mapping of the current status fed back from the underlying converter and the action result of the exhaust valve by the intelligent scheduling system. The data synchronously output by the intelligent dispatch system includes the execution type of the energy storage unit in this cycle and the destination record of wind power generation in this cycle.

7. The intelligent power dispatching method for a wind power energy storage system according to claim 1, characterized in that, In conjunction with the energy storage availability status, surplus wind power will be transferred to energy storage units, including: The intelligent dispatch system analyzes the charge / discharge interlock permission flag in the energy storage's supportable state; When the intelligent dispatch system determines that the allow charging flag in the charge-discharge interlock permission flag is valid and the remaining space meets the requirements, the intelligent dispatch system issues a surplus absorption instruction to the energy storage unit, transferring the surplus wind power that cannot be accepted by the absorption seat group into the energy storage unit. When the intelligent dispatch system determines that the charging permission flag is invalid or the remaining space is insufficient, the intelligent dispatch system issues a power limiting load reduction or pitch control command to the wind power generation unit through the control loop to forcibly reduce the wind power output.

8. The intelligent power dispatching method for a wind power energy storage system according to claim 7, characterized in that, The energy storage unit, in conjunction with the energy storage support condition limited energy storage unit, provides insulation current to the maintenance seat group, including: When wind power generation decreases and the already activated absorption or maintenance positions still need to supply power, the intelligent dispatch system determines that the allow discharge flag in the charge-discharge interlock permission flag is valid, and the intelligent dispatch system issues a continuous support instruction to the energy storage unit. The intelligent scheduling system prioritizes continuous support commands by prioritizing absorption seats that have already taken on new loads over maintenance seats that need to be kept in a hot state. Furthermore, the intelligent scheduling system prohibits releasing seats to receive new support input commands.

9. The intelligent power dispatching method for a wind power energy storage system according to claim 1, characterized in that, The path for generating the seat category for the next cycle by executing the seat migration operation rules includes: When the current seat category is a release seat, the actual execution result indicates that the release of the seat in this cycle has been completed, and the underlying sensor indicates that the current tank temperature is still higher than the set insulation threshold, the intelligent scheduling system will assign the seat category to a maintenance seat for the next cycle. When the current seat category is a maintenance seat, and the actual execution result feedback indicates that the hot state was successfully maintained in this cycle, there is no residual gas occupying the seat at the end of the cycle, and the cycle is in a reset state, the intelligent scheduling system will assign the seat category of the next cycle to an absorption seat. When the current seat category is an absorption seat, and the actual execution result shows that the new power supply was taken over in this cycle but residual gas occupancy reappears at the end of the cycle or the cycle is not reset, the intelligent scheduling system will revert the seat category to a release seat for the next cycle.

10. The intelligent power dispatching method for a wind power energy storage system according to claim 9, characterized in that, The path for generating the seat category for the next cycle by executing the seat migration operation rules also includes: When the underlying controller of the electrolytic cell module reports hardware fault codes such as converter disconnection and water pump failure, or when the electrolytic cell module is in the release position and the venting is not completed for a continuous period exceeding the set maximum waiting period threshold, the intelligent scheduling system will assign the position category of the next period to the isolation position. When the electrolytic cell module does not trigger the state change condition, the intelligent scheduling system executes the state self-maintaining path, and the seat category of the next cycle inherits the current seat category; For the electrolytic cell module in the release position, the intelligent scheduling system simultaneously performs an accumulation operation on the state retention counter when executing the state self-holding path, providing a definite time-domain calculation basis for triggering the set maximum waiting period threshold.

Images