Rolling optimization method for initial state of charge of multi-type hybrid energy storage batteries in energy system

Abstract

The application provides a method for rolling optimization of initial state of charge (SOC) of mixed energy storage batteries in an energy system, which comprises the following steps: obtaining historical operation data of a distributed renewable energy system, and clustering the historical operation data based on the difference between power generated by a power generation system and power consumed by a load; predicting the power generated by the power generation system and the power consumed by the load for each day to be optimized, and classifying the predicted values according to the clustering results; traversing all initial SOC and terminal SOC in the SOC data subset corresponding to each day to be optimized, obtaining a target function value sequence corresponding to each day to be optimized, and performing weighted addition on the target function value sequences of multiple days in the future to generate a final target function; and solving the final target function to obtain a target terminal SOC of a target day as the initial SOC of the next day of the target day. The method is used for optimizing the initial SOC of the mixed energy storage batteries in a long time scale, and is beneficial to continuously optimizing the operation scheduling of the energy system.

Description

Technical Field

[0001] This application relates to the field of power system energy storage technology, and in particular to a method for daily initial SOC rolling optimization of multi-type hybrid energy storage batteries in an energy system. Background Technology

[0002] With the development of renewable energy technologies, wind power and photovoltaic power generation are increasingly accounting for a larger share of the power supply system. Energy storage systems, such as battery banks, can store the electricity generated by these renewable energy systems and discharge it at appropriate times. As a result, the adoption rate of distributed energy systems incorporating various types of battery storage is gradually increasing. In practical applications, it is necessary to optimize the operation of these distributed energy systems.

[0003] In related technologies, when optimizing, the existing operation optimization scheduling method for energy storage devices in distributed renewable energy systems usually sets the initial SOC value of the energy storage system on the current day to be equal to the final SOC value of the energy storage system on the previous day, thereby ensuring the safe operation of the energy storage system on that day.

[0004] However, the above method lacks rationality in practical application. If renewable energy generation is sufficient on day N-1, but insufficient on day N, using this traditional SOC setting method will result in a low end-of-day SOC on day N-1, leading to low power utilization of the wind and solar renewable energy system on the next day, ultimately causing wind and solar curtailment.

[0005] Therefore, how to reasonably set the initial state of charge (SOC) and improve the utilization rate of renewable energy system output has become an urgent problem to be solved. Summary of the Invention

[0006] This application aims to at least partially address one of the technical problems in the related art.

[0007] Therefore, the first objective of this application is to propose a rolling optimization method for the initial SOC of multiple types of hybrid energy storage batteries in an energy system. This method can optimize the initial SOC of multiple types of hybrid energy storage batteries over a long time scale, and reasonably set the initial SOC value, which is conducive to ensuring the stable and efficient operation of the energy system.

[0008] The second objective of this application is to propose a rolling optimization system for initial SOC of multi-type hybrid energy storage batteries within an energy system;

[0009] The third objective of this application is to provide a non-transitory computer-readable storage medium.

[0010] To achieve the above objectives, a first aspect of this application provides a method for daily initial SOC rolling optimization of multi-type hybrid energy storage batteries within an energy system, the method comprising the following steps:

[0011] Historical operating data of distributed renewable energy systems are obtained, and the historical operating data is clustered based on the difference between the power generation and the power load of the renewable energy power generation system.

[0012] Assuming the initial SOC of the target day to be optimized, the power generation and the power consumption of each day to be optimized, including the target day, are predicted, and the predicted values ​​are classified according to the clustering results to generate a subset of SOC data for each day to be optimized.

[0013] In each SOC data subset, all the initial SOC and final SOC of the day are traversed to obtain the objective function value sequence corresponding to each day to be optimized, and the objective function value sequences of multiple future days to be optimized are weighted and summed to generate the final objective function;

[0014] Solve the final objective function to obtain the target day's end SOC, and use the target day's end SOC as the day's beginning SOC of the day following the target day.

[0015] Optionally, in one embodiment of this application, after using the target day-end SOC as the day-start SOC of the day following the target day, the method further includes: based on the day-start SOC of the day following the target day, performing day-start SOC optimization repeatedly for each day within a preset number of days after the target day.

[0016] Optionally, in one embodiment of this application, clustering the historical operating data includes: clustering the historical operating data using the K-means clustering algorithm to determine the optimal number of clusters; for each cluster, processing the centroid data to obtain the renewable energy power generation and electricity load power corresponding to the centroid data; and selecting the corresponding initial daily SOC and final daily SOC from the historical operating data based on the renewable energy power generation and electricity load power corresponding to the centroid data.

[0017] Optionally, in one embodiment of this application, when clustering the historical running data using the K-means clustering algorithm, the sum of squares of the distances from each sample point in each cluster to its corresponding centroid is represented by the following formula:

[0018]

[0019] in, Let i be the i-th feature of the j-th sample in the l-th cluster. Let k be the i-th feature of the centroid of the l-th cluster; k is the number of clusters, n is the number of features, and m is the number of features. l It is the number of samples in the l-th cluster.

[0020] Optionally, in one embodiment of this application, when the day to be optimized includes: the target day, the day after the target day, and the two days after the target day, the corresponding final objective function is represented by the following formula:

[0021] min f = Obj N +k×Obj N+1 +k 2 ×Obj N+2

[0022] Among them, Obj N It is the sequence of objective function values ​​for the target date, Obj N+1 It is the sequence of objective function values ​​for the day following the target date, Obj N+2 It is a sequence of objective function values ​​for the two days following the target date, where k is the weighting coefficient, 0 <k<1。

[0023] Optionally, in one embodiment of this application, the method further includes: determining an optimization model for rolling optimization of the initial SOC of multi-type hybrid energy storage batteries; and obtaining a SOC dataset based on the evolution of the optimal value of the optimization model as the initial SOC and the final SOC change.

[0024] To achieve the above objectives, a second aspect of this application also proposes a rolling optimization system for initial state of charge (SOC) of multiple types of hybrid energy storage batteries within an energy system, comprising the following modules:

[0025] The clustering module is used to acquire historical operating data of the distributed renewable energy system and cluster the historical operating data based on the difference between the power generation and the power load of the renewable energy power generation system.

[0026] The first generation module is used to assume the initial SOC of the target day to be optimized, predict the power generation and the power consumption of each day to be optimized, including the target day, and classify the predicted values ​​according to the clustering results to generate a subset of SOC data for each day to be optimized.

[0027] The second generation module is used to traverse all the initial SOC and final SOC in each SOC data subset, obtain the objective function value sequence corresponding to each day to be optimized, and perform a weighted sum of the objective function value sequences of multiple future days to be optimized to generate the final objective function.

[0028] The solution module is used to solve the final objective function, obtain the target day-end SOC of the target day, and use the target day-end SOC as the day-end SOC of the day following the target day.

[0029] Optionally, in one embodiment of this application, the solving module is further configured to: perform initial SOC optimization on each day within a preset number of days after the target day based on the initial SOC of the following day.

[0030] Optionally, in one embodiment of this application, the clustering module is specifically used to: cluster the historical operating data using the K-means clustering algorithm to determine the optimal number of clusters; for each cluster, process the centroid data to obtain the renewable energy power generation and electricity load power corresponding to the centroid data; and select the corresponding initial daily SOC and final daily SOC from the historical operating data based on the renewable energy power generation and electricity load power corresponding to the centroid data.

[0031] To implement the above embodiments, a third aspect of this application also proposes a non-transitory computer-readable storage medium storing a computer program thereon. When the computer program is executed by a processor, it implements the daily initial SOC rolling optimization method for multi-type hybrid energy storage batteries in the energy system described in the above embodiments.

[0032] The technical solution provided by the embodiments of this application brings at least the following beneficial effects: This application optimizes the daily end-of-day SOC of multi-type hybrid energy storage batteries in a distributed renewable energy system by selectively clustering local historical operating data, and then determines the initial-day SOC for the next day based on the optimized daily end-of-day SOC, thus achieving rolling optimization of the initial-day SOC of multi-type hybrid energy storage batteries over a long time scale. This allows for the reasonable setting of the initial-day SOC value, enabling the energy storage system to operate accordingly based on the renewable energy output of the next day, improving the utilization rate of the renewable energy system and ensuring the stable and efficient operation of the energy system. Furthermore, clustering historical operating data improves the efficiency of optimization solutions and reduces computational load, effectively achieving optimization of the operation and scheduling of distributed energy systems over a longer time scale.

[0033] Additional aspects and advantages of the invention will be set forth in part in the description which follows, and in part will be obvious from the description, or may be learned by practice of the invention. Attached Figure Description

[0034] The above and / or additional aspects and advantages of this application will become apparent and readily understood from the following description of the embodiments taken in conjunction with the accompanying drawings, wherein:

[0035] Figure 1This is a flowchart of a method for daily initial SOC rolling optimization of multi-type hybrid energy storage batteries in an energy system, as proposed in an embodiment of this application.

[0036] Figure 2 A flowchart illustrating a specific method for daily initial SOC rolling optimization of multi-type hybrid energy storage batteries in an energy system, as proposed in an embodiment of this application;

[0037] Figure 3 This is a schematic diagram of a specific daily initial SOC rolling optimization process proposed in an embodiment of this application;

[0038] Figure 4 This is a schematic diagram of the structure of a rolling optimization system for the initial state of charge (SOC) of multiple types of hybrid energy storage batteries in an energy system, as proposed in an embodiment of this application. Detailed Implementation

[0039] Embodiments of the present invention are described in detail below, examples of which are illustrated in the accompanying drawings, wherein the same or similar reference numerals denote the same or similar elements or elements having the same or similar functions throughout. The embodiments described below with reference to the accompanying drawings are exemplary and intended to explain the present invention, and should not be construed as limiting the present invention.

[0040] The following description, with reference to the accompanying drawings, illustrates a method and system for daily initial SOC rolling optimization of multi-type hybrid energy storage batteries within an energy system, as proposed in an embodiment of the present invention.

[0041] Figure 1 This is a flowchart of a method for daily initial SOC rolling optimization of multi-type hybrid energy storage batteries in an energy system, as proposed in an embodiment of this application. Figure 1 As shown, the method includes the following steps:

[0042] Step S101: Obtain historical operating data of the distributed renewable energy system, and cluster the historical operating data based on the difference between the power generation and the power load of the renewable energy power generation system.

[0043] Distributed renewable energy systems refer to distributed energy systems that include multiple types of renewable energy power generation systems (such as wind power generation systems and photovoltaic power generation systems) and multiple energy storage systems. Each energy storage battery can store the electricity output by a certain renewable energy power generation system, and the types of energy storage batteries can also be different.

[0044] The historical operating data of the distributed renewable energy system includes pre-recorded operating data of the entire distributed renewable energy system and its subsystems and components at different times. It can be understood that the historical operating data includes the initial state of charge (SOC) of the hybrid energy storage battery in the past day and the end state of charge of the day.

[0045] Specifically, this application can obtain historical operating data of distributed renewable energy systems through various methods, such as reading pre-stored historical operating data from the energy system's database. To effectively reduce computational load, this application clusters the historical operating data, dividing it into several types, which facilitates the subsequent selection of different SOC data from the historical operating data for each type of characteristic data.

[0046] It should be noted that, in order for the clustering features to be sufficiently obvious and direct, data with the highest correlation to energy storage operation needs to be generated. Since the operating status of energy storage is closely related to the difference between the power generation and load of wind and solar power generation, the power difference can reflect the operating status of energy storage to a certain extent. Therefore, when selecting feature data for clustering, this application first obtains the daily power difference curves of historical data, and then uses the power difference to determine the magnitude of the electricity difference. This allows us to obtain the corresponding electricity generated by the energy storage operation and the electricity exchanged between the external power consumer and the grid.

[0047] In one embodiment of this application, clustering can be performed using the K-means clustering algorithm. K-means is an unsupervised Euclidean distance clustering algorithm that achieves good clustering results even with low dimensionality. This algorithm divides a feature matrix X of N samples into K non-overlapping clusters, where a set of data within each cluster can be considered as one class. The mean of all data in each cluster is... The centroid of a cluster is called the centroid. The clustering process involves continuously updating the centroids to find the optimal centroid that minimizes the overall Euclidean distance. In this example, the sum of the squares of the distances from each sample point in each cluster to its corresponding centroid is expressed by the following formula:

[0048]

[0049] in, Let i be the i-th feature of the j-th sample in the l-th cluster. Let k be the i-th feature of the centroid of the l-th cluster; k is the number of clusters, n is the number of features, and m is the number of features. l It is the number of samples in the l-th cluster.

[0050] In this embodiment, when performing clustering, one possible implementation is to first cluster historical data, select different k values, and determine the optimal number of clusters through comparison. As an example of a method for determining the k value, to determine the optimal number of clusters, this application first uses the K-means clustering algorithm to initially divide the historical data, determining an initial number of clusters K. That is, K cluster points are obtained through the initial division, and corresponding clusters are generated for each. Then, the Silhouette criterion is used to evaluate the similarity between all the initially determined cluster points, i.e., these K cluster points, and other sample data in their corresponding clusters.

[0051] Then, for each cluster, the centroid data is processed to obtain the corresponding renewable energy generation power and electricity load power. Since the characteristics of each data cluster are quite different, different SOC data need to be selected for different types of data. Then, based on the renewable energy generation power and electricity load power corresponding to the centroid data, the corresponding initial daily SOC and final daily SOC are selected from the historical operating data.

[0052] Step S102: Assuming the initial SOC of the target day to be optimized, predict the power generation and power consumption of each day to be optimized, including the target day, and classify the predicted values ​​according to the clustering results to generate a subset of SOC data for each day to be optimized.

[0053] Specifically, this application selects a specific day to be optimized (i.e., the target day) and several days within a preset number after that day as the days to be optimized. Then, it predicts the power generation and load of the renewable energy power generation system (wind power and photovoltaic power) for each day to be optimized. The predicted values ​​are classified according to the clustering results, that is, the data types obtained by processing the centroid data after clustering the power prediction data in step S101 are classified. That is, according to the clustering method in the previous step, the type of predicted data for each day to be optimized is determined, and then, based on the determined data type, the corresponding initial SOC and final SOC are selected from the historical operation dataset to generate a data subset corresponding to each day to be optimized.

[0054] For example, when the days to be optimized include: target day N, the day after the target day N+1, and the two days after the target day N+2, the power of wind, solar, and load on days N, N+1, and N+2 is predicted. The predicted values ​​for these three days are then classified according to the clustering results, and the corresponding data subset D is selected. N D N+1 D N+2 Further optimizations will be carried out.

[0055] In one embodiment of this application, when performing power prediction, the initial SOC of the target day to be optimized can be used for prediction.

[0056] Step S103: Traverse all the initial SOC and final SOC of each day in each SOC data subset to obtain the objective function value sequence corresponding to each day to be optimized, and perform a weighted sum of the objective function value sequences of multiple future days to be optimized to generate the final objective function.

[0057] In this embodiment, all the initial SOC and final SOC of each day are traversed on a subset of data for each day, and a sequence of objective function values ​​is generated for each day. Each sequence of objective function values ​​includes multiple candidate final SOCs. Then, the sequence of objective function values ​​for the next few days selected in step S102 that need to be optimized is weighted and summed to generate the final objective function.

[0058] Continuing with the example in step S102, when the days to be optimized include: the target day, the day after the target day, and the two days after the target day, the known initial SOC for day N is... We need to determine the final initial SOC on day N, which is: In D N Traversing based on a subset of data Obj, the sequence of objective function values ​​obtained on day N. N Then in D N+1 Traversing based on a subset of data Obj is the sequence of objective function values ​​obtained on day N+1. N+1 Similarly, Obj is obtained. N+2 Furthermore, the corresponding final objective function can be expressed by the following formula:

[0059] min f = Obj N +k×Obj N+1 +k 2 ×Obj N+2

[0060] Among them, Obj N It is the sequence of objective function values ​​for the target date, Obj N+1 It is the sequence of objective function values ​​for the day following the target date, Obj N+2 It is a sequence of objective function values ​​for the two days following the target date, where k is the weighting coefficient, 0 <k<1。

[0061] Step S104: Solve the final objective function to obtain the target day's end SOC, and use the target day's end SOC as the day's beginning SOC of the day following the target day.

[0062] Specifically, the final objective function is solved to determine the initial SOC of the target day from the sequence of objective function values ​​that minimizes the final objective function. This initial SOC is the optimized final SOC setpoint for the energy storage system on the target day. Then, this final SOC is set as the initial SOC for the day following the target day.

[0063] In one embodiment of this application, after setting the target day-end SOC as the day-beginning SOC of the day following the target day, the method further includes: based on the day-beginning SOC of the day following the target day, repeatedly performing day-beginning SOC optimization for each day within a preset number of days after the target day. For example, the day-end SOC setting value of the energy storage system on day N is also the initial SOC setting value of the energy storage system on day N+1. The day-end SOC setting value of the energy storage system on day N+1 is also determined using the above optimization method. Here, the day-beginning SOC of day N+1 is a calculated known value, and no assumptions are needed. Power prediction and subsequent optimization steps can be performed again based on this day-beginning SOC.

[0064] In one embodiment of this application, before and after obtaining historical operating data of the distributed renewable energy system, an optimization model for rolling optimization of the initial SOC of multiple types of hybrid energy storage batteries is further included. When selecting SOC data in the future, the SOC dataset can be obtained based on the evolution of the optimal value of the optimization model when the initial SOC and the final SOC change.

[0065] Therefore, this application uses the daily end-of-day SOC calculated by optimization on the current day as the daily beginning-of-day SOC for the next day, and then repeats the optimization steps based on the daily beginning-of-day SOC for the next day, thus realizing rolling optimization of the daily end-of-day SOC for the next day based on the currently calculated daily end-of-day SOC.

[0066] To more clearly illustrate the specific implementation process of the daily initial SOC rolling optimization method for multi-type hybrid energy storage batteries in the energy system of this application, the following description uses a specific optimization method proposed in one embodiment of this application as an example. Figure 2 This is a flowchart illustrating a specific method for daily initial SOC rolling optimization of multi-type hybrid energy storage batteries within an energy system, as proposed in an embodiment of this application. Figure 2 As shown, the method includes the following steps:

[0067] Step S201: Determine the optimization model.

[0068] Specifically, the optimization objective of the optimization model can be set according to the needs, such as the commonly used objective of optimizing the system economy, the objective of renewable energy consumption, or a multi-objective optimization model.

[0069] Step S202: Process the historical data.

[0070] Specifically, after determining the optimization model, the historical operating data needs to be processed. To effectively reduce the computational load, this application clusters the historical operating data, dividing it into K types. Step S203 is then performed on the feature data of each type to obtain K data subsets, which together form a dataset.

[0071] In this embodiment, the main steps of the clustering algorithm are as follows:

[0072] Step S2021: First, cluster the data and select different k values ​​to find the optimal number of clusters.

[0073] Step S2022: Process the centroid data to obtain the power data of photovoltaic, wind power, and load corresponding to the centroid data, and perform the processing steps in step S203 on the obtained data. Select a granularity change of 0.1, and a daily SOC change range of 0.1 to 0.8 at the beginning and end of the day to meet the safe operation of energy storage.

[0074] Step S203: Obtain the SOC dataset.

[0075] Specifically, after determining the optimization model, it is necessary to obtain the evolution of the model's optimal value as the SOC changes at the beginning and end of the day. By traversing the beginning and end SOCs at an appropriate granularity, a three-dimensional data subset of the optimal value with respect to the beginning and end SOCs is obtained.

[0076] Step S204: Select the SOC dataset for each optimization day.

[0077] Specifically, in the case of day-ahead optimization, assuming the initial SOC of day N is known, and the final SOC of day N needs to be determined, firstly, the power of wind, solar, and load on days N, N+1, and N+2 is predicted. The predicted values ​​for these three days are then classified according to the data type described in step S202. That is, following the processing method described in step S202, the SOC data is selected in step S203 to generate the corresponding data subset D. N D N+1 D N+2 Proceed to the next optimization step.

[0078] Step S205: Generate the final objective function.

[0079] Specifically, the known initial SOC on day N is... We need to determine the final initial SOC on day N, which is: In D N Traversing based on a subset of data Obj, the sequence of objective function values ​​obtained on day N. N Then in D N+1 Traversing based on a subset of data Obj is the sequence of objective function values ​​obtained on day N+1. N+1 Similarly, Obj is obtained. N+2 The objective function values ​​for the three days are weighted:

[0080] min f = Obj N +k×Obj N+1 +k 2 ×Obj N+2

[0081] Among them, 0 <k<1。

[0082] Step S206: Determine the optimized end-of-day SOC setting value for the day.

[0083] Specifically, find the value that makes the function f reach its minimum. in, This refers to the end-of-day SOC setting value of the energy storage system that needs to be determined on day N.

[0084] Step S207: Optimize the end-of-day SOC for the subsequent days to be optimized.

[0085] Specifically, the end-of-day SOC setting value of the energy storage system on day N is the initial SOC setting value of the energy storage system on day N+1. The end-of-day SOC setting value of the energy storage system on day N+1 is determined by repeatedly executing steps S205 to S207.

[0086] Based on the above embodiments, in order to more clearly illustrate the specific operation flow of the daily initial SOC rolling optimization method for multiple types of hybrid energy storage batteries in the energy system of this application, the following is an exemplary description of a specific operation flow proposed in one embodiment of this application. Figure 3 This is a schematic diagram of a specific daily initial SOC rolling optimization process proposed in an embodiment of this application.

[0087] like Figure 3 As shown, this embodiment clusters historical operating data to determine the cluster type. Then, assuming an initial SOC on day N, power predictions are made for days N, N+1, and N+2. The predicted data is categorized according to the cluster type. Data is selected from the dataset of electricity exchanged with the grid corresponding to the initial and final SOC changes, and the electricity exchanged with the grid corresponding to the initial and final SOC changes is predicted for days N, N+1, and N+2, respectively. Then, following step S205 in the above embodiment, the SOC is traversed to generate a sequence of objective function values ​​for the three days. Finally, the optimal value is calculated by weighting the three days' data to determine the final optimized daily SOC for day N, which is then used as the initial daily SOC for day N+1.

[0088] In summary, the daily initial SOC rolling optimization method for multi-type hybrid energy storage batteries in the energy system of this application optimizes the daily final SOC of multi-type hybrid energy storage batteries in the distributed renewable energy system by selectively clustering local historical operating data. Then, based on the optimized daily final SOC, the daily initial SOC for the next day is determined, achieving rolling optimization of the daily initial SOC of multi-type hybrid energy storage batteries over a long time scale. Therefore, this method can reasonably set the daily initial SOC value, allowing the energy storage system to adapt its operation according to the renewable energy output of the next day, improving the utilization rate of the renewable energy system and contributing to the stable and efficient operation of the energy system. Furthermore, clustering historical operating data improves the optimization solution efficiency and reduces the computational load, effectively achieving optimization of the operation and scheduling of distributed energy systems over a longer time scale.

[0089] To achieve the above embodiments, this application also proposes a rolling optimization system for initial state of charge (SOC) of multiple types of hybrid energy storage batteries within an energy system. Figure 4 This is a schematic diagram of the structure of a rolling optimization system for initial state of charge (SOC) of multiple types of hybrid energy storage batteries in an energy system, as proposed in an embodiment of this application. Figure 4 As shown, the system includes: a clustering module 100, a first generation module 200, a second generation module 300, and a solution module 400.

[0090] Among them, the clustering module 100 is used to obtain historical operating data of distributed renewable energy systems and cluster the historical operating data based on the difference between the power generation capacity and the power load capacity of the renewable energy power generation system.

[0091] The first generation module 200 is used to assume the initial SOC of the target day to be optimized, predict the power generation and power consumption of each day to be optimized, including the target day, and classify the predicted values ​​according to the clustering results to generate a subset of SOC data corresponding to each day to be optimized.

[0092] The second generation module 300 is used to traverse all the initial SOC and final SOC of each SOC data subset, obtain the objective function value sequence corresponding to each day to be optimized, and perform a weighted sum of the objective function value sequences of multiple future days to be optimized to generate the final objective function.

[0093] Solver module 400 is used to solve the final objective function, obtain the target day's end SOC, and use the target day's end SOC as the day's beginning SOC of the next day.

[0094] Optionally, in one embodiment of this application, the solving module 400 is further configured to: perform daily initial SOC optimization repeatedly for each day within a preset number of days after the target day, based on the daily initial SOC of the next day.

[0095] Optionally, in one embodiment of this application, the clustering module 100 is specifically used to: cluster historical operating data using the K-means clustering algorithm to determine the optimal number of clusters; for each cluster, process the centroid data to obtain the renewable energy power generation and power consumption corresponding to the centroid data; and select the corresponding initial daily SOC and final daily SOC from the historical operating data based on the renewable energy power generation and power consumption corresponding to the centroid data.

[0096] Optionally, in one embodiment of this application, the clustering module 100 is specifically used to represent the sum of squares of the distances from the sample points in each cluster to their corresponding centroids using the following formula:

[0097]

[0098] in, Let i be the i-th feature of the j-th sample in the l-th cluster. Let k be the i-th feature of the centroid of the l-th cluster; k is the number of clusters, n is the number of features, and m is the number of features. l It is the number of samples in the l-th cluster.

[0099] Optionally, in one embodiment of this application, when the days to be optimized include: the target day, the day after the target day, and the two days after the target day, the second generation module 300 specifically generates the corresponding final objective function using the following formula:

[0100] min f = Obj N +k×Obj N+1 +k 2 ×Obj N+2

[0101] Among them, Obj N It is the sequence of objective function values ​​for the target date, Obj N+1 It is the sequence of objective function values ​​for the day following the target date, Obj N+2 It is a sequence of objective function values ​​for the two days following the target date, where k is the weighting coefficient, 0 <k<1。

[0102] Optionally, in one embodiment of this application, the system further includes a determination module 500, used to determine an optimization model for rolling optimization of the initial SOC of multi-type hybrid energy storage batteries; and to obtain a SOC dataset based on the evolution of the optimal value of the optimization model as the initial SOC and the final SOC change.

[0103] It should be noted that the explanation of the aforementioned embodiment of the daily initial SOC rolling optimization method for multi-type hybrid energy storage batteries in the energy system also applies to the system in this embodiment, and will not be repeated here.

[0104] In summary, the daily initial SOC rolling optimization system for multi-type hybrid energy storage batteries in the energy system of this application optimizes the daily final SOC of multi-type hybrid energy storage batteries in the distributed renewable energy system by selectively clustering local historical operating data. Then, based on the optimized daily final SOC, the daily initial SOC for the next day is determined, achieving rolling optimization of the daily initial SOC of multi-type hybrid energy storage batteries over a long time scale. Therefore, this system can reasonably set the daily initial SOC value, allowing the energy storage system to adapt its operation according to the renewable energy output of the next day, improving the utilization rate of the renewable energy system and contributing to the stable and efficient operation of the energy system. Furthermore, clustering historical operating data improves the optimization solution efficiency and reduces the computational load, effectively achieving optimization of the operation and scheduling of the distributed energy system over a longer time scale.

[0105] To implement the above embodiments, this application also proposes a non-transitory computer-readable storage medium storing a computer program thereon, which, when executed by a processor, implements the daily initial SOC rolling optimization method for multi-type hybrid energy storage batteries in an energy system as described in any of the above embodiments.

[0106] In the description of this specification, the references to terms such as "one embodiment," "some embodiments," "example," "specific example," or "some examples," etc., refer to specific features, structures, materials, or characteristics described in connection with that embodiment or example, which are included in at least one embodiment or example of this application. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples. Moreover, without contradiction, those skilled in the art can combine and integrate the different embodiments or examples described in this specification, as well as the features of different embodiments or examples.

[0107] Furthermore, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of technical features indicated. Thus, a feature defined as "first" or "second" may explicitly or implicitly include at least one of that feature. In the description of this application, "multiple" means at least two, such as two, three, etc., unless otherwise explicitly specified.

[0108] Any process or method description in the flowchart or otherwise herein can be understood as representing a module, segment, or portion of code comprising one or more executable instructions for implementing custom logic functions or processes, and the scope of the preferred embodiments of this application includes additional implementations in which functions may be performed not in the order shown or discussed, including substantially simultaneously or in reverse order depending on the functions involved, as should be understood by those skilled in the art to which embodiments of this application pertain.

[0109] The logic and / or steps represented in the flowchart or otherwise described herein, for example, can be considered as a sequenced list of executable instructions for implementing logical functions, and can be embodied in any computer-readable medium for use by, or in conjunction with, an instruction execution system, apparatus, or device (such as a computer-based system, a processor-included system, or other system that can fetch and execute instructions from, an instruction execution system, apparatus, or device). For the purposes of this specification, "computer-readable medium" can be any means that can contain, store, communicate, propagate, or transmit programs for use by, or in conjunction with, an instruction execution system, apparatus, or device. More specific examples (a non-exhaustive list) of computer-readable media include: an electrical connection having one or more wires (electronic device), a portable computer disk drive (magnetic device), random access memory (RAM), read-only memory (ROM), erasable and editable read-only memory (EPROM or flash memory), fiber optic devices, and portable optical disc read-only memory (CDROM). Alternatively, the computer-readable medium may be paper or other suitable media on which the program can be printed, since the program can be obtained electronically, for example, by optically scanning the paper or other medium, followed by editing, interpreting, or otherwise processing as necessary, and then stored in a computer memory.

[0110] It should be understood that various parts of this application can be implemented using hardware, software, firmware, or a combination thereof. In the above embodiments, multiple steps or methods can be implemented using software or firmware stored in memory and executed by a suitable instruction execution system. For example, if implemented in hardware as in another embodiment, it can be implemented using any one or a combination of the following techniques known in the art: discrete logic circuits having logic gates for implementing logical functions on data signals, application-specific integrated circuits (ASICs) having suitable combinational logic gates, programmable gate arrays (PGAs), field-programmable gate arrays (FPGAs), etc.

[0111] Those skilled in the art will understand that all or part of the steps of the methods in the above embodiments can be implemented by a program instructing related hardware. The program can be stored in a computer-readable storage medium, and when executed, the program includes one or a combination of the steps of the method embodiments.

[0112] Furthermore, the functional units in the various embodiments of this application can be integrated into a processing module, or each unit can exist physically separately, or two or more units can be integrated into a module. The integrated module can be implemented in hardware or as a software functional module. If the integrated module is implemented as a software functional module and sold or used as an independent product, it can also be stored in a computer-readable storage medium.

[0113] The storage medium mentioned above can be a read-only memory, a disk, or an optical disk, etc. Although embodiments of this application have been shown and described above, it is understood that the above embodiments are exemplary and should not be construed as limiting this application. Those skilled in the art can make changes, modifications, substitutions, and variations to the above embodiments within the scope of this application.

Claims

1. A method for daily initial SOC rolling optimization of multi-type hybrid energy storage batteries in an energy system, characterized in that, Includes the following steps: Historical operating data of distributed renewable energy systems are obtained, and the historical operating data is clustered based on the difference between the power generation and the power load of the renewable energy power generation system. Assuming the initial SOC of the target day to be optimized, the power generation and the power consumption of each day to be optimized, including the target day, are predicted, and the predicted values ​​are classified according to the clustering results to generate a subset of SOC data for each day to be optimized. In each SOC data subset, all the initial SOC and final SOC of the day are traversed to obtain the objective function value sequence corresponding to each day to be optimized, and the objective function value sequences of multiple future days to be optimized are weighted and summed to generate the final objective function; Solve the final objective function to obtain the target day's end SOC, and use the target day's end SOC as the day's beginning SOC of the day following the target day.

2. The method according to claim 1, characterized in that, After setting the target day's end SOC as the beginning SOC of the day following the target day, the method further includes: Based on the initial SOC of the following day, the initial SOC optimization is repeated for each day within a preset number of days after the target day.

3. The method according to claim 1, characterized in that, Clustering the historical operational data includes: clustering the historical operational data using the K-means clustering algorithm to determine the optimal number of clusters; For each cluster, the centroid data is processed to obtain the renewable energy power generation and electricity load corresponding to the centroid data; Based on the renewable energy power generation and power consumption corresponding to the centroid data, the corresponding initial daily SOC and final daily SOC are selected from the historical operating data.

4. The method according to claim 3, characterized in that, When clustering the historical data using the K-means clustering algorithm, the sum of squared distances from each sample point in each cluster to its corresponding centroid is represented by the following formula: in, For the first l The th cluster j The first sample i One characteristic, For the first l The first centroid of a cluster i One feature; k For the number of clusters, n It is the number of features. m l It is the first l The number of samples in each cluster.

5. The method according to claim 1, characterized in that, When the days to be optimized include the target day, the day after the target day, and the two days after the target day, the corresponding final objective function is expressed by the following formula: in, It is a sequence of objective function values ​​for the target date. It is a sequence of objective function values ​​for the day following the target date. It is a sequence of objective function values ​​for the two days following the target date. k These are weighting coefficients, 0 < k< 1.

6. The method according to claim 1, characterized in that, Also includes: An optimization model for rolling optimization of the initial state of charge (SOC) of multiple types of hybrid energy storage batteries was determined. Based on the evolution of the optimal value of the optimization model as the SOC changes at the beginning and end of the day, the SOC dataset is obtained.

7. A rolling optimization system for initial state of charge (SOC) of multiple types of hybrid energy storage batteries within an energy system, characterized in that, Includes the following modules: The clustering module is used to acquire historical operating data of the distributed renewable energy system and cluster the historical operating data based on the difference between the power generation and the power load of the renewable energy power generation system. The first generation module is used to assume the initial SOC of the target day to be optimized, predict the power generation and the power consumption of each day to be optimized, including the target day, and classify the predicted values ​​according to the clustering results to generate a subset of SOC data for each day to be optimized. The second generation module is used to traverse all the initial SOC and final SOC in each SOC data subset, obtain the objective function value sequence corresponding to each day to be optimized, and perform a weighted sum of the objective function value sequences of multiple future days to be optimized to generate the final objective function. The solution module is used to solve the final objective function, obtain the target day-end SOC of the target day, and use the target day-end SOC as the day-end SOC of the day following the target day.

8. The system according to claim 7, characterized in that, The solution module is also used for: Based on the initial SOC of the following day, the initial SOC optimization is repeated for each day within a preset number of days after the target day.

9. The system according to claim 7, characterized in that, The clustering module is specifically used for: The historical data was clustered using the K-means clustering algorithm to determine the optimal number of clusters. For each cluster, the centroid data is processed to obtain the renewable energy power generation and electricity load corresponding to the centroid data; Based on the renewable energy power generation and power consumption corresponding to the centroid data, the corresponding initial daily SOC and final daily SOC are selected from the historical operating data.

10. A non-transitory computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, it implements the daily initial SOC rolling optimization method for multi-type hybrid energy storage batteries in an energy system as described in any one of claims 1-6.

Images