A control method, a central controller and a storage medium for an energy storage node

Abstract

The present invention discloses a control method, a central controller and a storage medium for an energy storage node. The method includes: determining a target energy storage node, where the target energy storage node is an energy storage node registered to the central controller and capable of participating in power dispatching; obtaining the current behavior data of the user and the current operation data of the target energy storage node, and inputting the current behavior data and the current operation data into a pre-trained prediction model to obtain current demand data, where the prediction model is a model trained with a machine learning algorithm aiming at predicting the total power demand; generating a dispatching plan according to the current demand data, where the dispatching plan includes a plurality of dispatching schemes; optimizing the dispatching plan with the objectives of minimizing the fluctuation of the total power demand and the total cost, and controlling the target energy storage node to work according to the corresponding dispatching scheme in the optimized dispatching plan. This solution can achieve precise and efficient management of the energy storage node, thereby improving the energy utilization rate.

Description

Technical Field

[0001] The present invention relates to the field of power technology, and particularly to a control method for energy storage nodes, a central controller, and a storage medium. Background Art

[0002] A distributed energy storage system refers to a system that connects multiple small, decentralized energy storage nodes to the power grid or power system to achieve energy storage and management.

[0003] Existing distributed energy storage systems mainly use distributed energy storage facilities (such as battery energy storage systems) to adjust the load demand of the power system to balance the grid load, optimize energy use, and manage the power supply and demand balance. However, the distributed energy storage facilities lack effective real-time monitoring means and data transmission mechanisms for energy storage nodes, making it difficult to achieve comprehensive control and precise control of energy storage nodes, resulting in too low overall efficiency of the system. Summary of the Invention

[0004] The present invention provides a control method for energy storage nodes, a central controller, and a storage medium, which can achieve precise and efficient management of energy storage nodes, thereby improving energy utilization efficiency.

[0005] According to one aspect of the present invention, there is provided a control method for energy storage nodes, which is applied to a central controller. The method includes:

[0006] Determine a target energy storage node, where the target energy storage node is an energy storage node registered to the central controller and capable of participating in power dispatching;

[0007] Obtain the current behavior data of the user and the current operation data of the target energy storage node, and input the current behavior data and the current operation data into a pre-trained prediction model to obtain current demand data, where the prediction model is a model trained with a machine learning algorithm aiming at predicting the total power demand;

[0008] Generate a dispatching plan according to the current demand data, where the dispatching plan includes multiple dispatching schemes, and one dispatching scheme corresponds to one target energy storage node;

[0009] Optimize the dispatching plan with the optimization goals of minimizing the fluctuation of the total power demand and the total cost, and control the target energy storage node to work according to the corresponding dispatching scheme in the optimized dispatching plan.

[0010] According to another aspect of the present invention, there is provided a central controller, which includes:

[0011] At least one processor; and a memory communicatively connected to the at least one processor; wherein,

[0012] The memory stores a computer program executable by at least one processor. The computer program is executed by the at least one processor to enable the at least one processor to execute the control method of the energy storage node according to any embodiment of the present invention.

[0013] According to another aspect of the present invention, there is provided a computer-readable storage medium storing computer instructions for causing a processor to implement the control method of the energy storage node according to any embodiment of the present invention when executed.

[0014] The technical solution of the embodiment of the present invention first determines target energy storage nodes registered with the central controller and capable of participating in power dispatching; then obtains the current behavior data of the user and the current operation data of the target energy storage nodes, and inputs them into a pre-trained prediction model to obtain current demand data; further generates a dispatching plan according to the current demand data; finally, optimizes the dispatching plan with the objectives of minimizing the fluctuation of the total power demand and the total cost, and controls the target energy storage nodes to work according to the corresponding dispatching scheme in the optimized dispatching plan. For this control method of the energy storage node, on the one hand, selecting suitable target energy storage nodes to participate in subsequent power dispatching can achieve comprehensive control and precise control of all energy storage nodes. On the other hand, obtaining the current demand data through the prediction model can achieve precise prediction of the total power demand, providing a data basis for generating subsequent dispatching plans. On the third hand, since the dispatching plan includes multiple dispatching schemes, and one dispatching scheme corresponds to one target energy storage node, the purpose of decomposing the overall dispatching problem into multiple dispatching sub-problems is realized, thereby improving the calculation speed and dispatching efficiency. On the fourth hand, further optimizing the dispatching plan with the objectives of minimizing the fluctuation of the total power demand and the total cost can ensure the coordinated operation of each target energy storage node, achieve precise and efficient management of the target energy storage nodes, thereby improving the energy utilization rate and ensuring the maximization of the overall benefit.

[0015] It should be understood that the content described in this part is not intended to identify the key or important features of the embodiments of the present invention, nor is it used to limit the scope of the present invention. Other features of the present invention will become easily understood through the following description. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following will briefly introduce the drawings required for the description of the embodiments. Obviously, the following drawings are only some embodiments of the present invention. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative efforts.

[0017] Figure 1 It is a schematic structural diagram of a distributed energy storage system provided in Embodiment 1 of the present invention;

[0018] Figure 2 It is a schematic flowchart of a control method for an energy storage node provided in the first embodiment of the present invention;

[0019] Figure 3 It is a schematic flowchart of a control method for an energy storage node provided in the second embodiment of the present invention;

[0020] Figure 4 It is a schematic structural diagram of a control device for an energy storage node provided in the third embodiment of the present invention;

[0021] Figure 5 It is a schematic structural diagram of another control device for an energy storage node provided in the third embodiment of the present invention;

[0022] Figure 6 It is a schematic structural diagram of a central controller provided in the fourth embodiment of the present invention. Detailed implementation manners

[0023] In order to enable those skilled in the art to better understand the solution of the present invention, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present invention.

[0024] It should be noted that the terms "first", "second", "original", "candidate", "target", etc. in the specification and claims of the present invention and the above drawings are used to distinguish similar objects, and do not necessarily need to describe a specific order or sequence. It should be understood that such data can be interchanged under appropriate circumstances so that the embodiments of the present invention described here can be implemented in an order other than those illustrated or described here. In addition, the terms "include" and "have" and any variations thereof are intended to cover non-exclusive inclusion. For example, a process, method, system, product, or device that includes a series of steps or units does not necessarily have to be limited to those steps or units clearly listed, but may include other steps or units not clearly listed or inherent to these processes, methods, products, or devices.

[0025] Embodiment 1

[0026] Figure 1 It is a schematic structural diagram of a distributed energy storage system provided in the first embodiment of the present invention. As Figure 1 shown, the distributed energy storage system includes a central controller 101, a plurality of energy storage nodes 102, and a distributed energy trading platform 103.

[0027] The central controller 101 is a device capable of scheduling and management, and is the core of the distributed energy storage system. It can execute the control method of the energy storage node provided by the present invention. Exemplarily, the central controller 101 can be a centralized computer system or a cloud platform. The central controller 101 communicates with each energy storage node 102 through the Internet of Things (IoT). A database can be configured in the central controller 101, or the database can exist independently of the central controller 101. The database is used to store information related to each energy storage node 102.

[0028] The number of energy storage nodes 102 is M. For ease of understanding, Figure 1 it is drawn with M = 3 as an example. The three energy storage nodes 102 are respectively labeled as energy storage node A, energy storage node B, and energy storage node C. The energy storage node 102 can be registered to the central controller 101 so that an IoT connection is established between the central controller 101 and the energy storage node 102, thereby realizing the monitoring, scheduling, control, etc. of the central controller 101 over the energy storage node 102. In the present invention, the energy storage nodes 102 registered to the central controller 101 (such as Figure 1 energy storage node B and energy storage node C in Figure 1 ) are called candidate energy storage nodes, and the energy storage nodes 102 not registered to the central controller 101 (such as

[0029] The distributed energy trading platform 103 provides an energy trading function for users. The distributed energy trading platform 103 can be built using blockchain technology to ensure the security and transparency of energy trading. The distributed energy trading platform 103 can display the operating status and revenue of the distributed energy storage system to users through a smartphone application or a web interface. Optionally, a user incentive mechanism can be designed to encourage users to adjust their electricity consumption behavior and participate in the demand regulation process through dynamic electricity prices and reward policies.

[0030] Figure 2 is a schematic flowchart of a control method for an energy storage node provided in Embodiment 1 of the present invention. This embodiment is applicable to Figure 1 the situation where the central controller shown in Figure 1 controls the energy storage node. This method can be executed by a control device of the energy storage node. The control device of the energy storage node can be implemented in the form of hardware and / or software, and the control device of the energy storage node can be configured in Figure 2 the central controller shown in

[0031] S210. Determine the target energy storage node, where the target energy storage node is an energy storage node registered to the central controller and capable of participating in power dispatching.

[0032] As can be seen from the above description, since a distributed energy storage system includes multiple energy storage nodes, but not all of these energy storage nodes are necessarily registered with the central controller, and not all the energy storage nodes registered with the central controller need to participate in this power dispatch. Therefore, the central controller first needs to determine the target energy storage nodes, where the target energy storage nodes are the energy storage nodes that are registered with the central controller and can participate in power dispatch.

[0033] Specifically, the target energy storage nodes can be preset energy storage nodes; they can also be energy storage nodes selected by the user; or the following method can be used to determine them: first determine the candidate energy storage nodes registered with the central controller, and then screen out the energy storage nodes that can participate in power dispatch from the candidate energy storage nodes as the target energy storage nodes.

[0034] In an optional implementation manner, the central controller can consider that all candidate energy storage nodes can participate in power dispatch, that is, regard all candidate energy storage nodes as the target energy storage nodes.

[0035] In another optional implementation manner, the central controller can determine the candidate energy storage nodes registered with the central controller, and then judge whether each candidate energy storage node can participate in power dispatch, so as to determine the target energy storage nodes according to the judgment results.

[0036] Exemplarily, after multiple candidate energy storage nodes are registered with the central controller, the central controller obtains the node parameters and monitoring data of each candidate energy storage node; according to the node parameters and monitoring data, determines the current state of each candidate energy storage node, where the current state includes an available state and an unavailable state; regards the candidate energy storage nodes in the available state as the target energy storage nodes.

[0037] The node parameters are used to reflect the self-attributes of the candidate energy storage nodes, and the node parameters can include but are not limited to: geographical location, energy storage capacity, charge and discharge rate, and health status. The monitoring data is used to reflect the operation conditions of the candidate energy storage nodes, and can usually be obtained by the acquisition module based on IoT technology. The monitoring data can include but are not limited to: current power, load condition, grid electricity price, and ambient conditions.

[0038] The current state of the candidate energy storage nodes includes an available state and an unavailable state. The available state means that the candidate energy storage node is in a state where it can participate in power dispatch; the unavailable state means that the candidate energy storage node is in a state where it cannot participate in power dispatch, such as situations of excessive load or poor health status.

[0039] In this way, selecting suitable target energy storage nodes to participate in subsequent power dispatch can achieve comprehensive control and precise control of all energy storage nodes.

[0040] Optionally, when multiple candidate energy storage nodes are registered with the central controller, the central control node can assign a unique identifier to each candidate energy storage node to achieve unified and rapid management of the candidate energy storage nodes.

[0041] S220. Obtain the current behavior data of the user and the current operation data of the target energy storage node, and input the current behavior data and the current operation data into a pre-trained prediction model to obtain the current demand data, where the prediction model is a model trained using machine learning algorithms with the goal of predicting the total power demand.

[0042] The current behavior data of the user refers to the behavior data of the user in the current scheduling period, and the behavior data is used to reflect the user's electricity consumption behavior (such as electricity consumption patterns, seasonal data, weather characteristic data, etc.). The current operation data of the target energy storage node refers to the operation data of the target energy storage node in the current scheduling period, such as the current power, load condition, health status, etc.

[0043] Since the prediction model is a model trained using machine learning algorithms with the goal of predicting the total power demand, inputting the current behavior data and the current operation data into the pre-trained prediction model can obtain the current demand data. Machine learning algorithms can be a series of mathematical and statistical methods used in the field of machine learning for data analysis, pattern recognition, and prediction.

[0044] In one embodiment, the method for training the prediction model may include the following steps a1 - a4:

[0045] Step a1: Obtain a training data set, where the training data set includes a number of training samples, and one training sample includes historical behavior data, historical operation data, and historical demand data.

[0046] Step a2: Input the historical behavior data and the historical operation data of the current training sample into the original model to obtain the predicted demand data corresponding to the current training sample, where the current training sample is any one training sample in the training data set.

[0047] For example, the original model can be a Long Short - Term Memory (LSTM) model.

[0048] Optionally, before inputting the historical behavior data and the historical operation data of the current training sample into the original model, the current training sample can be pre - processed (such as cleaning and normalizing data, denoising, filling missing values, normalizing processing, etc.), and then time - series data is formed based on the historical behavior data and the historical operation data to input the time - series data into the original model. Time - series data can reflect the relationship between data and time characteristics, thereby improving the accuracy of model training.

[0049] Step a3: If the difference between the predicted demand data and the historical demand data of the current training sample is less than or equal to the preset threshold, then use the original model as the preset model.

[0050] Step a4: If the difference between the predicted demand data and the historical demand data of the current training sample is greater than the preset threshold, then adjust the parameters of the original model, use the next training sample in the training dataset as the current training sample, and return to execute Step a2.

[0051] Obtaining the current demand data through the prediction model can achieve accurate prediction of the total power demand, providing a data basis for generating subsequent scheduling plans.

[0052] S230. Generate a scheduling plan according to the current demand data, where the scheduling plan includes multiple scheduling schemes, and one scheduling scheme corresponds to one target energy storage node.

[0053] In a possible implementation manner, the central controller can generate a scheduling scheme for each target energy storage node respectively according to the node parameters and monitoring data of each target energy storage node.

[0054] In another possible implementation manner, the central controller can construct a total scheduling function according to the current demand data; decompose the total scheduling function to obtain multiple sub-scheduling functions, where one sub-scheduling function corresponds to one target energy storage node; perform parallel solution on the sub-scheduling functions based on the first optimization algorithm, and use the solution result as the scheduling plan.

[0055] Since the scheduling plan includes multiple scheduling schemes and one scheduling scheme corresponds to one target energy storage node, the purpose of decomposing the overall scheduling problem into multiple scheduling sub-problems is achieved, thereby improving the calculation speed and scheduling efficiency.

[0056] S240. Optimize the scheduling plan with the objectives of minimizing the fluctuation of the total power demand and the total cost, and control the target energy storage nodes to work according to the corresponding scheduling schemes in the optimized scheduling plan.

[0057] After generating the scheduling plan, further optimizing the scheduling plan with the objectives of minimizing the fluctuation of the total power demand and the total cost can ensure the coordinated work of each target energy storage node, achieve accurate and efficient management of the target energy storage nodes, thereby improving the energy utilization rate and ensuring the maximization of the overall benefits.

[0058] Specifically, the central controller can construct an optimization objective function according to the scheduling plan and the current demand data, and perform iterative optimization calculation on the optimization objective function based on the second optimization algorithm to determine the optimized scheduling plan.

[0059] In one embodiment, the first optimization algorithm and the second optimization algorithm can be the same algorithm or different algorithms. For example, both the first optimization algorithm and the second optimization algorithm can be distributed optimization algorithms.

[0060] Optionally, based on the above embodiment, the central controller can obtain the execution data and health data of the scheduling plan of the target energy storage node; according to the execution data and health data of the scheduling plan, dynamically adjust the algorithm parameters of the first optimization algorithm and / or the second optimization algorithm. Thus, the continuous and efficient operation of the distributed energy storage system can be achieved.

[0061] Embodiment 2

[0062] Figure 3 FIG. is a schematic flowchart of a control method for an energy storage node provided in Embodiment 2 of the present invention. Based on the above Embodiment 1, this embodiment provides a specific implementation manner for generating and optimizing a scheduling plan. As Figure 3 shown, the method includes:

[0063] S310. Determine the target energy storage node, where the target energy storage node is an energy storage node registered with the central controller and capable of participating in power scheduling.

[0064] Since the distributed energy storage system includes multiple energy storage nodes, but not all of these energy storage nodes are necessarily registered with the central controller, and not all the energy storage nodes registered with the central controller need to participate in this power scheduling. Therefore, the central controller first needs to determine the target energy storage node, and the target energy storage node is an energy storage node registered with the central controller and capable of participating in power scheduling.

[0065] Exemplarily, after multiple candidate energy storage nodes are registered with the central controller, the central controller obtains the node parameters and monitoring data of each candidate energy storage node; according to the node parameters and monitoring data, determines the current state of each candidate energy storage node, where the current state includes an available state and an unavailable state; and uses the candidate energy storage nodes in the available state as the target energy storage nodes.

[0066] The node parameters are used to reflect the self-attributes of the candidate energy storage nodes, and the node parameters can include but are not limited to: geographical location, energy storage capacity, charge and discharge rate, and health status. The monitoring data is used to reflect the operation conditions of the candidate energy storage nodes and can usually be obtained by a collection module based on IoT technology. The monitoring data can include but is not limited to: current power, load conditions, grid electricity price, and environmental conditions.

[0067] The current state of the candidate energy storage node includes an available state and an unavailable state. The available state means that the candidate energy storage node is in a state capable of participating in power scheduling; the unavailable state means that the candidate energy storage node is in a state unable to participate in power scheduling, such as situations of excessive load or poor health status.

[0068] In this way, selecting suitable target energy storage nodes to participate in subsequent power dispatching can achieve comprehensive control and precise control of all energy storage nodes.

[0069] S320. Obtain the current behavior data of the user and the current operation data of the target energy storage node, and input the current behavior data and the current operation data into a pre-trained prediction model to obtain the current demand data.

[0070] Among them, the prediction model is a model trained using machine learning algorithms with the goal of predicting the total power demand. The method for training the prediction model may include: obtaining a training data set, where the training data set includes a number of training samples, and a training sample includes historical behavior data, historical operation data, and historical demand data; inputting the historical behavior data and historical operation data of the current training sample into the original model to obtain the predicted demand data corresponding to the current training sample, where the current training sample is any one of the training samples in the training data set; if the difference between the predicted demand data and the historical demand data of the current training sample is less than or equal to a preset threshold, then use the original model as the preset model; if the difference between the predicted demand data and the historical demand data of the current training sample is greater than the preset threshold, then adjust the parameters of the original model, and use the next training sample in the training data set as the current training sample, and return to execute the step of inputting the historical behavior data and historical operation data of the current training sample into the original model to obtain the predicted demand data corresponding to the current training sample.

[0071] Optionally, before inputting the historical behavior data and historical operation data of the current training sample into the original model, the current training sample can be preprocessed (such as cleaning and standardizing data, denoising, filling missing values, normalizing, etc.), and then time series data is formed based on the historical behavior data and historical operation data to input the time series data into the original model. The time series data can reflect the relationship between the data and the time characteristics, thereby improving the accuracy of model training.

[0072] S330. Construct a total scheduling function according to the current demand data.

[0073] When generating a scheduling plan, the central controller first constructs a total scheduling function according to the current demand data. For example, maximizing the overall system efficiency or balancing power demand and supply. The scheduling function can cover the optimization objectives of all target energy storage nodes.

[0074] Specifically, the total scheduling function is where N is the total number of target energy storage nodes, f i (,) is the benefit function of the i-th target energy storage node, Ei is the electric quantity of the i-th target energy storage node, P iis the charging and discharging power of the i-th target energy storage node.

[0075] S340. Decompose the total scheduling function to obtain multiple sub-scheduling functions, where one sub-scheduling function corresponds to one target energy storage node.

[0076] Based on the characteristics of the target energy storage node, decompose the total scheduling function to obtain multiple sub-scheduling functions. Each sub-scheduling function is carried out in a local scope (i.e., for the corresponding target energy storage node), and the optimization objective is usually the charging and discharging plan. The constraint conditions include the energy storage capacity, maximum charging and discharging rate, health status, etc. of the target energy storage node.

[0077] Specifically, the sub-scheduling function is subject to E i (t), P i (t) constraints; where E i (t) is the power of the i-th target energy storage node at time t, and P i (t) is the charging and discharging power of the i-th target energy storage node at time t.

[0078] S350. Based on the first optimization algorithm, perform parallel solution on the sub-scheduling function, and use the solution result as the scheduling plan.

[0079] In one embodiment, the first optimization algorithm can be a distributed optimization algorithm, such as a distributed convex optimization algorithm.

[0080] Perform parallel solution on the sub-scheduling function based on the first optimization algorithm to obtain the charging and discharging plan of the target energy storage node. In this way, both the calculation speed and the scheduling efficiency can be improved, and the central controller can coordinate to ensure the global consistency among the target energy storage nodes, ensuring that there will be no conflicts or uneven resource allocation among the target energy storage nodes.

[0081] Optionally, the coordination mechanism can be implemented through an iterative algorithm.

[0082] Alternatively, each sub-scheduling function can be set with constraint conditions to limit the behavior of the target energy storage node. Necessary information such as electricity price and load can also be shared among the target energy storage nodes.

[0083] S360. With the optimization objectives of minimizing the fluctuation of the total power demand and the total cost, construct an optimization objective function according to the scheduling plan and the current demand data, where the optimization objective function has constraint conditions.

[0084] Specifically, the optimization objective function can be

[0085]

[0086] Among them, J represents the total power demand fluctuation and total cost within a given time range T, N is the total number of target energy storage nodes, P i (t) is the charge and discharge power of the i-th target energy storage node at time t, λ i is the electricity price coefficient of the i-th target energy storage node, α i is the charging efficiency coefficient of the i-th target energy storage node, γ is the balance coefficient used to weigh the relationship between the total power demand fluctuation and the total cost, and D(t) is the total power demand at time t.

[0087] The constraint conditions are

[0088] Among them, represents the power limit of the target energy storage node, E i (t) is the power of the i-th target energy storage node at time t, is the maximum power of the i-th target energy storage node; represents the charge and discharge power limit of the target energy storage node, is the minimum charge and discharge power of the i-th target energy storage node, is the maximum charge and discharge power of the i-th target energy storage node; represents the dynamic change limit of the power, β i is the discharge efficiency coefficient of the i-th target energy storage node; represents the limit on the overall system demand balance.

[0089] In one embodiment, the update equation of E i (t) is generally E i (t + 1) = E i (t) + η i P i (t)Δt.

[0090] Among them, η i is the charge and discharge efficiency coefficient, which is used to describe the energy loss or efficiency during the charging and discharging processes of the target energy storage node and can reflect the actual energy utilization efficiency of the system during charging or discharging; Δt is the time interval.

[0091] E i (t) determines the charge and discharge capacity of the target energy storage node. In the charging plan, if the battery is close to full load (such as exceeding 90% capacity), the system will reduce the charging power of this node; conversely, if the battery power is too low, the system will preferentially allocate more power to it to restore the energy storage state; these adjustments will affect subsequent power scheduling.

[0092] S370. Perform iterative optimization calculation on the optimization objective function based on the second optimization algorithm to determine the optimized scheduling plan.

[0093] Specifically, the optimized scheduling plan can be

[0094]

[0095] Among them, is the optimized scheduling scheme of the i-th target energy storage node at time t, represents the weighted sum of the squares of the powers of all target energy storage nodes. During the optimization process, this part represents the contribution of the power distribution of all target energy storage nodes to the optimization objective function; represents the square of the difference between the total power demand and the powers of all target energy storage nodes, ensuring the balance between the system demand and the total power output of all target energy storage nodes; represents the influence of all other target energy storage nodes except the target energy storage node i on the target energy storage node i. It is adjusted through the time-related function h(t), and the cumulative effect of its change rate on the current node is calculated, i≠j. This part ensures that when optimizing the scheduling scheme of the target energy storage node i, the state changes and power distribution situations of all other target energy storage nodes are considered; h(t) is a time-related function used to adjust the smoothness of power changes, representing a weight or influencing factor depending on a certain time t; f(E j (t)) is the energy storage state function of the target energy storage node j, representing the influence of the energy storage state of the target energy storage node j on the target energy storage node i; g(P j (t)) is the power function of the target energy storage node j, representing the influence of the power of the target energy storage node j on the target energy storage node i.

[0096] S380. Control the target energy storage nodes to work according to the corresponding scheduling schemes in the optimized scheduling plan, and obtain the scheduling scheme execution data and health data of the target energy storage nodes.

[0097] S390. Dynamically adjust the algorithm parameters of the first optimization algorithm and / or the second optimization algorithm according to the scheduling scheme execution data and health data.

[0098] Dynamically adjusting the algorithm parameters of the first optimization algorithm and / or the second optimization algorithm can ensure the accuracy of generating the scheduling scheme in subsequent scheduling cycles. Further optionally, the prediction model can also be updated periodically.

[0099] Exemplarily, assume that the distributed energy storage system includes three target energy storage nodes (Node 1, Node 2, Node 3), and the basic information and optimization parameters of each target energy storage node are as follows:

[0100] Node 1: Located in City A, energy storage capacity Charge and discharge power range to α1 = 0.9, β1 = 0.95, λ1 = 0.1;

[0101] Node 2: Located in City B, energy storage capacity Charge and discharge power range to α2 = 0.85, β2 = 0.9, λ2 = 0.12;

[0102] Node 3: Located in City C, energy storage capacity Charge and discharge power range to α3 = 0.92, β3 = 0.93, λ3 = 0.15.

[0103] The optimized scheduling plan is as follows:

[0104] Node 1: It is planned to be 20 kW in the 1st hour, 30 kW in the 2nd hour, and 40 kW in the 3rd hour;

[0105] Node 2: It is planned to be 25 kW in the 1st hour, 35 kW in the 2nd hour, and 45 kW in the 3rd hour;

[0106] Node 3: It is planned to be 30 kW in the 1st hour, 40 kW in the 2nd hour, and 50 kW in the 3rd hour.

[0107] The central controller monitors the execution data and health data of the scheduling schemes of the 3 nodes in real time through the Internet of Things, compares the deviation between the actual operation data and the optimized scheduling scheme. If the actual deviation is too large (for example, the actual power is 15 kW while the plan is 20 kW), the scheduling scheme is adjusted, the optimization problem is re-solved, and a new scheduling scheme is generated.

[0108] Table 1 is a comparison table of the predicted load, actual load, charge and discharge conditions, and grid price of the three target energy storage nodes.

[0109] Table 1 Comparison of Predicted Load, Actual Load, Charge and Discharge Conditions, and Grid Price

[0110]

[0111] It can be seen from Table 1 that the control method of the energy storage node provided by the embodiment of the present invention can keep the state of the node within a reasonable range. For example, the battery charge state of Node 1 gradually drops from 80% to 68% within 6 hours, the battery charge state of Node 2 drops from 70% to 52% within 6 hours, and the battery charge state of Node 3 drops from 90% to 66% within 6 hours, all remaining within the safe range. Thus, the situation of excessive discharge or charge of a single node is avoided, further improving the stability of the system and reducing equipment damage.

[0112] An embodiment of the present invention provides a control method for an energy storage node, including: determining a target energy storage node, where the target energy storage node is an energy storage node registered with a central controller and capable of participating in power dispatching; obtaining the current behavior data of a user and the current operation data of the target energy storage node, and inputting the current behavior data and the current operation data into a pre-trained prediction model to obtain current demand data, where the prediction model is a model trained with a machine learning algorithm aiming at predicting the total power demand; generating a dispatching plan according to the current demand data, where the dispatching plan includes multiple dispatching schemes, and one dispatching scheme corresponds to one target energy storage node; optimizing the dispatching plan with the objectives of minimizing the fluctuation of the total power demand and the total cost, and controlling the target energy storage node to work according to the corresponding dispatching scheme in the optimized dispatching plan. The technical solution of the embodiment of the present invention first determines the target energy storage node registered with the central controller and capable of participating in power dispatching; then obtains the current behavior data of the user and the current operation data of the target energy storage node, and inputs them into the pre-trained prediction model to obtain current demand data; and then generates a dispatching plan according to the current demand data; finally, optimizes the dispatching plan with the objectives of minimizing the fluctuation of the total power demand and the total cost, and controls the target energy storage node to work according to the corresponding dispatching scheme in the optimized dispatching plan. For this control method of the energy storage node, on the one hand, selecting a suitable target energy storage node to participate in subsequent power dispatching can achieve comprehensive control and precise control of all energy storage nodes. On the other hand, obtaining the current demand data through the prediction model can achieve precise prediction of the total power demand, providing a data basis for generating subsequent dispatching plans. On the third hand, since the dispatching plan includes multiple dispatching schemes and one dispatching scheme corresponds to one target energy storage node, the purpose of decomposing the overall dispatching problem into multiple dispatching sub-problems is realized, thereby improving the calculation speed and dispatching efficiency. On the fourth hand, further optimizing the dispatching plan with the objectives of minimizing the fluctuation of the total power demand and the total cost can ensure the coordinated work of each target energy storage node, achieve precise and efficient management of the target energy storage node, thereby improving the energy utilization rate and ensuring the maximization of the overall benefit.

[0113] Embodiment III

[0114] Figure 4 is a schematic structural diagram of a control device for an energy storage node provided by Embodiment III of the present invention. As Figure 4 shown, the device includes: a node determination module 401, a demand prediction module 402, a plan generation module 403, a plan optimization module 404, and a plan execution module 405.

[0115] A node determination module 401, configured to determine a target energy storage node, where the target energy storage node is an energy storage node registered with a central controller and capable of participating in power dispatching;

[0116] A demand prediction module 402, configured to obtain current behavior data of a user and current operation data of the target energy storage node, and input the current behavior data and the current operation data into a pre-trained prediction model to obtain current demand data, where the prediction model is a model trained with a machine learning algorithm for the target of predicting the total power demand;

[0117] A plan generation module 403, configured to generate a dispatching plan according to the current demand data, where the dispatching plan includes multiple dispatching schemes, and one dispatching scheme corresponds to one target energy storage node;

[0118] A plan optimization module 404, configured to optimize the dispatching plan with the minimum fluctuations in the total power demand and the total cost as the optimization objectives;

[0119] A plan execution module 405, configured to control the target energy storage node to work according to the corresponding dispatching scheme in the optimized dispatching plan.

[0120] Optionally, the node determination module 401 is specifically configured to, after multiple candidate energy storage nodes are registered with the central controller, obtain node parameters and monitoring data of each candidate energy storage node, where the node parameters are used to reflect the self-attributes of the candidate energy storage node, and the monitoring data is used to reflect the operation conditions of the candidate energy storage node; after multiple candidate energy storage nodes are registered with the central controller, obtain node parameters and monitoring data of each candidate energy storage node, where the node parameters are used to reflect the self-attributes of the candidate energy storage node, and the monitoring data is used to reflect the operation conditions of the candidate energy storage node; and use the candidate energy storage nodes in the available state as the target energy storage nodes.

[0121] Optionally, combined with Figure 4 , Figure 5 is a schematic structural diagram of another control device for an energy storage node provided in Embodiment III of the present invention. As Figure 5 shown, the device further includes: a model training module 406.

[0122] The model training module 406 is configured to obtain a training data set, where the training data set includes a plurality of training samples, and a training sample includes historical behavior data, historical operation data, and historical demand data; input the historical behavior data and historical operation data of the current training sample into the original model to obtain predicted demand data corresponding to the current training sample, where the current training sample is any one of the training samples in the training data set; if the difference between the predicted demand data and the historical demand data of the current training sample is less than or equal to a preset threshold, then use the original model as the preset model; if the difference between the predicted demand data and the historical demand data of the current training sample is greater than the preset threshold, then adjust the parameters of the original model, and use the next training sample in the training data set as the current training sample, and return to execute the step of inputting the historical behavior data and historical operation data of the current training sample into the original model to obtain the predicted demand data corresponding to the current training sample.

[0123] Optionally, the plan generation module 403 is configured to construct a total scheduling function according to the current demand data; decompose the total scheduling function to obtain a plurality of sub-scheduling functions, where one sub-scheduling function corresponds to one target energy storage node; perform parallel solution on the sub-scheduling functions based on the first optimization algorithm, and use the solution result as the scheduling plan.

[0124] Optionally, the total scheduling function is

[0125] The sub-scheduling function is Subject to E i (t), P i (t) constraints;

[0126] where N is the total number of target energy storage nodes, and f i (,) is the benefit function of the i-th target energy storage node, E i is the power of the i-th target energy storage node, and P i is the charge and discharge power of the i-th target energy storage node; E i (t) is the power of the i-th target energy storage node at time t, and P i (t) is the charge and discharge power of the i-th target energy storage node at time t.

[0127] Optionally, the plan optimization module 404 is specifically configured to take the fluctuation of the total power demand and the minimum total cost as the optimization objectives, construct an optimization objective function according to the scheduling plan and the current demand data, where the optimization objective function has constraint conditions; perform iterative optimization calculation on the optimization objective function based on the second optimization algorithm to determine the optimized scheduling plan.

[0128] Optionally, the optimization objective function is

[0129]

[0130] Among them, J represents the total power demand fluctuation and total cost within a given time range T, N is the total number of target energy storage nodes, and P i (t) is the charge and discharge power of the i-th target energy storage node at time t, and λ i is the electricity price coefficient of the i-th target energy storage node, and α i is the charging efficiency coefficient of the i-th target energy storage node. γ is a balance coefficient used to weigh the relationship between the total power demand fluctuation and the total cost, and D(t) is the total power demand at time t;

[0131] The constraint conditions are

[0132] Among them, E i (t) is the electricity quantity of the i-th target energy storage node at time t, is the maximum electricity quantity of the i-th target energy storage node, is the minimum charge and discharge power of the i-th target energy storage node, is the maximum charge and discharge power of the i-th target energy storage node, and β i is the discharge efficiency coefficient of the i-th target energy storage node;

[0133] The optimized scheduling plan is

[0134]

[0135] Among them, is the optimized scheduling scheme of the i-th target energy storage node at time t, represents the weighted sum of the squares of the powers of all target energy storage nodes, represents the square of the difference between the total power demand and the powers of all target energy storage nodes, represents the influence of all other target energy storage nodes except the i-th target energy storage node on the i-th target energy storage node. h(t) is a time-related function used to adjust the smoothness of power changes, f(E j (t)) is the energy storage state function of the target energy storage node j, and g(P j (t)) is the power function of the target energy storage node j.

[0136] Optionally, the plan generation module 403 is further configured to obtain the execution data and health data of the scheduling scheme of the target energy storage node; dynamically adjust the algorithm parameters of the first optimization algorithm according to the execution data and health data of the scheduling scheme;

[0137] The plan optimization module 404 is further configured to obtain the execution data and health data of the scheduling scheme of the target energy storage node; dynamically adjust the algorithm parameters of the second optimization algorithm according to the execution data and health data of the scheduling scheme.

[0138] The control device of the energy storage node provided by the embodiment of the present invention can execute the control method of the energy storage node provided by any embodiment of the present invention, and has the corresponding functional modules and beneficial effects for executing the method.

[0139] Embodiment 4

[0140] Figure 6 FIG. is a schematic structural diagram of a central controller provided by Embodiment 4 of the present invention. The central controller is intended to represent various forms of digital computers, such as, for example, laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The central controller may also represent various forms of mobile devices, such as, for example, personal digital processors, cellular telephones, smart phones, wearable devices (such as helmets, glasses, watches, etc.) and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely illustrative and are not intended to limit the implementation of the present invention described and / or claimed herein.

[0141] As Figure 6 shown, the central controller 10 includes at least one processor 11, and a memory communicatively connected to the at least one processor 11, such as a read-only memory (ROM) 12, a random access memory (RAM) 13, etc. Among them, the memory stores a computer program executable by the at least one processor. The processor 11 can perform various appropriate actions and processes according to the computer program stored in the read-only memory (ROM) 12 or the computer program loaded from the storage unit 18 into the random access memory (RAM) 13. In the RAM 13, various programs and data required for the operation of the central controller 10 can also be stored. The processor 11, the ROM 12, and the RAM 13 are connected to each other through a bus 14. The input / output (I / O) interface 15 is also connected to the bus 14.

[0142] A plurality of components in the central controller 10 are connected to the I / O interface 15, including: an input unit 16, such as a keyboard, a mouse, etc.; an output unit 17, such as various types of displays, speakers, etc.; a storage unit 18, such as a magnetic disk, an optical disk, etc.; and a communication unit 19, such as a network card, a modem, a wireless communication transceiver, etc. The communication unit 19 allows the central controller 10 to exchange information / data with other devices through a computer network such as the Internet and / or various telecommunication networks.

[0143] The processor 11 may be various general-purpose and / or special-purpose processing components with processing and computing capabilities. Some examples of the processor 11 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various dedicated artificial intelligence (AI) computing chips, various processors running machine learning model algorithms, a digital signal processor (DSP), and any suitable processor, controller, microcontroller, etc. The processor 11 executes the various methods and processes described above, such as the control method of the energy storage node.

[0144] In some embodiments, the control method of the energy storage node may be implemented as a computer program tangibly embodied in a computer-readable storage medium, such as the storage unit 18. In some embodiments, part or all of the computer program may be loaded and / or installed onto the central controller 10 via the ROM 12 and / or the communication unit 19. When the computer program is loaded into the RAM 13 and executed by the processor 11, one or more steps of the control method of the energy storage node described above may be executed. Alternatively, in other embodiments, the processor 11 may be configured to execute the control method of the energy storage node in any other suitable manner (e.g., by means of firmware).

[0145] Various embodiments of the systems and techniques described above herein can be implemented in digital electronic circuit systems, integrated circuit systems, field-programmable gate arrays (FPGA), application-specific integrated circuits (ASIC), application-specific standard products (ASSP), systems-on-a-chip (SOC), complex programmable logic devices (CPLD), computer hardware, firmware, software, and / or combinations thereof. These various embodiments can include: implemented in one or more computer programs that can be executed and / or interpreted on a programmable system including at least one programmable processor, which can be a dedicated or general-purpose programmable processor, and can receive data and instructions from a storage system, at least one input device, and at least one output device, and transmit the data and instructions to the storage system, the at least one input device, and the at least one output device.

[0146] The computer program for implementing the method of the present invention can be written in any combination of one or more programming languages. These computer programs can be provided to the processor of a general-purpose computer, a special-purpose computer, or other programmable data processing device, such that when the computer program is executed by the processor, the functions / operations specified in the flowchart and / or block diagram are implemented. The computer program can be executed entirely on the machine, partially on the machine, as a stand-alone software package partially on the machine and partially on a remote machine, or entirely on a remote machine or server.

[0147] In the context of the present invention, a computer-readable storage medium can be a tangible medium that can contain or store a computer program for use by or in connection with an instruction execution system, apparatus, or device. The computer-readable storage medium can include, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, apparatuses, or devices, or any suitable combination of the foregoing. Alternatively, the computer-readable storage medium can be a machine-readable signal medium. More specific examples of the machine-readable storage medium would include an electrical connection based on one or more wires, a portable computer diskette, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or Flash memory), an optical fiber, a portable compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.

[0148] To provide for interaction with a user, the systems and techniques described herein can be implemented on a central controller having: a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to the user; and a keyboard and a pointing device (e.g., a mouse or a trackball) by which the user can provide input to the central controller. Other kinds of devices can also be used to provide for interaction with the user; for example, feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user can be received in any form (including acoustic input, voice input, or tactile input).

[0149] The systems and techniques described herein can be implemented in a computing system that includes back-end components (e.g., as a data server), or a computing system that includes middleware components (e.g., an application server), or a computing system that includes front-end components (e.g., a user computer having a graphical user interface or a web browser through which the user can interact with an implementation of the systems and techniques described herein), or a computing system that includes any combination of such back-end components, middleware components, or front-end components. The components of the system can be interconnected by any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include: a local area network (LAN), a wide area network (WAN), a blockchain network, and the Internet.

[0150] A computing system may include a client and a server. The client and the server are generally far from each other and usually interact via a communication network. The client-server relationship is generated by computer programs running on respective computers and having a client-server relationship with each other. The server may be a cloud server, also known as a cloud computing server or a cloud host, which is a host product in the cloud computing service system, solving the defects of difficult management and weak business scalability existing in traditional physical hosts and VPS services.

[0151] It should be understood that various forms of processes shown above can be used, with steps reordered, added or deleted. For example, the steps described in the present invention can be executed in parallel, sequentially or in different orders, as long as the desired results of the technical solution of the present invention can be achieved, and no limitation is made herein.

[0152] The above specific embodiments do not constitute a limitation on the protection scope of the present invention. Those skilled in the art should understand that various modifications, combinations, sub-combinations and substitutions can be made according to design requirements and other factors. Any modifications, equivalent substitutions and improvements made within the spirit and principle of the present invention shall be included within the protection scope of the present invention.

Claims

1. A control method for an energy storage node, characterized in that Applied to a central controller, the method includes: Determining a target energy storage node, wherein the target energy storage node is an energy storage node registered with the central controller and capable of participating in power dispatching; Obtaining the user's current behavior data and the current operating data of the target energy storage node, and inputting the current behavior data and the current operating data into a pre-trained prediction model to obtain current demand data, wherein the prediction model is a model trained using a machine learning algorithm with the goal of predicting total power demand; Generate a scheduling plan based on the current demand data, wherein the scheduling plan includes multiple scheduling schemes, and each scheduling scheme corresponds to one target energy storage node; Taking the fluctuations in the total electricity demand and the minimum total cost as the optimization objectives, an optimization objective function is constructed according to the scheduling plan and the current demand data, where the optimization objective function is ; represents the fluctuations in the total electricity demand and the total cost within a given time range T, N is the total number of target energy storage nodes, is the charging and discharging power of the i-th target energy storage node at time t, is the electricity price coefficient of the i-th target energy storage node, is the charging efficiency coefficient of the i-th target energy storage node, is the balance coefficient used to weigh the relationship between the fluctuations in the total electricity demand and the total cost, is the total electricity demand at time t; the optimization objective function has constraint conditions, and the constraint conditions are ; is the electricity quantity of the i-th target energy storage node at time t, is the maximum electricity quantity of the i-th target energy storage node, is the minimum charging and discharging power of the i-th target energy storage node, is the maximum charging and discharging power of the i-th target energy storage node, is the discharging efficiency coefficient of the i-th target energy storage node; Performing iterative optimization calculation on the optimization objective function based on the second optimization algorithm to determine the optimized scheduling plan, where the optimized scheduling plan is ; is the optimized scheduling scheme of the i-th target energy storage node at time t, represents the weighted sum of the squares of the powers of all target energy storage nodes, represents the square of the difference between the total power demand and the powers of all target energy storage nodes, represents the influence of all other target energy storage nodes except the target energy storage node i on the target energy storage node i, is a time-related function for adjusting the smoothness of power changes, is the energy storage state function of the target energy storage node j, is the power function of the target energy storage node j; The target energy storage node is controlled to operate according to the corresponding scheduling scheme in the optimized scheduling plan.

2. The control method of the energy storage node according to claim 1, wherein Determining the target energy storage node includes: After multiple candidate energy storage nodes are registered with the central controller, node parameters and monitoring data of each candidate energy storage node are obtained, wherein the node parameters are used to reflect the own attributes of the candidate energy storage node, and the monitoring data are used to reflect the operating status of the candidate energy storage node; Determine a current state of each candidate energy storage node according to the node parameters and the monitoring data, wherein the current state includes an available state and an unavailable state; The candidate energy storage node in an available state is used as the target energy storage node.

3. The control method of the energy storage node according to claim 1, characterized in that, The method for training the prediction model comprises: Acquire a training data set, wherein the training data set includes a plurality of training samples, and one training sample includes historical behavior data, historical operation data, and historical demand data; Inputting historical behavior data and historical operation data of a current training sample into the original model to obtain predicted demand data corresponding to the current training sample, wherein the current training sample is any training sample in the training data set; If the difference between the predicted demand data and the historical demand data of the current training sample is less than or equal to a preset threshold, the original model is used as the prediction model; If the difference between the predicted demand data and the historical demand data of the current training sample is greater than a preset threshold, the parameters of the original model are adjusted, and the next training sample in the training data set is used as the current training sample, and the process returns to the step of inputting the historical behavior data and historical operation data of the current training sample into the original model to obtain the predicted demand data corresponding to the current training sample.

4. The control method of the energy storage node according to claim 1, wherein Generating a scheduling plan according to the current demand data includes: Constructing a total scheduling function based on the current demand data; Decomposing the overall dispatch function to obtain a plurality of sub-dispatching functions, wherein one sub-dispatching function corresponds to one target energy storage node; The sub-scheduling functions are solved in parallel based on the first optimization algorithm, and the solution results are used as the scheduling plan.

5. The control method of the energy storage node according to claim 4, characterized in that: The total scheduling function is ; The sub-scheduling function is ; where N is the total number of target energy storage nodes, is the benefit function of the i-th target energy storage node, is the power of the i-th target energy storage node, is the charge and discharge power of the i-th target energy storage node; is the power of the i-th target energy storage node at time t, is the charge and discharge power of the i-th target energy storage node at time t.

6. The control method of the energy storage node according to claim 1, characterized in that: Also includes: Obtaining scheduling plan execution data and health data of the target energy storage node; Execute the data and the health data according to the scheduling scheme, and dynamically adjust the algorithm parameters of the first optimization algorithm and / or the second optimization algorithm.

7. A central controller, characterized in that, The central controller includes: At least one processor; and a memory communicatively connected to the at least one processor; wherein, The memory stores a computer program executable by the at least one processor, and the computer program is executed by the at least one processor so that the at least one processor can execute the control method of the energy storage node according to any one of claims 1-6.

8. A computer-readable storage medium, characterized in that The computer-readable storage medium stores computer instructions for implementing the control method of the energy storage node according to any one of claims 1-6 when executed by a processor.

Images