A coordinated control method for energy storage based on big data

Abstract

The present invention relates to the field of data processing technology, and more specifically, the present invention relates to a method for coordinated energy storage control based on big data, including: obtaining the historical power consumption of the energy storage system and the storage capacity, power supply and attenuation rate of each battery; inputting the historical power consumption into the prediction model to obtain the power consumption in the next time period; calculating the power supply priority of each battery; iteratively obtaining the objective function of the power supply sequence of the energy storage system through a genetic algorithm; and realizing coordinated energy storage control based on big data according to the solution of the objective function of each round of iteration. The present invention obtains future power demand through a prediction model, calculates the priority of battery power supply according to the attenuation rate and storage capacity of each battery, and prompts the energy storage system to use batteries with low attenuation rates for power supply; by determining the heat dissipation coefficient, the heat dissipation demand of the battery during operation is taken into account; through the solution of the objective function, an efficient and reliable power supply plan is realized to ensure the life and efficiency of the energy storage system.

Description

Technical Field

[0001] The present invention relates to the field of data processing technology, and more specifically, to a method for coordinated control of energy storage based on big data. Background Art

[0002] Energy storage technology has been widely used in traditional power generation, power transmission and distribution, power demand, new energy, etc. As a key link in the smart grid, the energy storage system is suitable for various power consumption places and supports multiple power consumption modes. It can not only improve the operating efficiency of power equipment and reduce power supply costs, but also improve the system operation stability and achieve uninterrupted power supply.

[0003] The patent document with announcement number CN112103979B discloses a coordinated control method for an integrated energy storage system, including the following steps: Step 1, establishing an equivalent cycle life model for energy storage; Step 2, modeling and analyzing the energy storage cost; Step 3, dividing the energy storage system into multiple groups of energy storage with different charging and discharging characteristics, and performing coordinated control of the integrated energy storage system. The coordinated control method for an integrated energy storage system proposed in the patent document can improve the utilization efficiency of the integrated energy storage system, reduce the number of charging and discharging switching times of the system, and reduce the overall operating cost of the system.

[0004] However, the above patent document does not determine the optimal power supply sequence when powering power in energy storage coordination. In the hybrid energy system, the energy storage system is the main power source. When the energy storage energy or power is insufficient, the diesel generator set charges the energy storage system or provides power compensation for the energy storage system. However, during the continuous charging and discharging process of the energy storage system, the capacity of the battery in the energy storage system will decrease, resulting in accelerated battery aging. Deep discharge will also damage the battery, thereby affecting the life and efficiency of the energy storage system and increasing costs. Summary of the invention

[0005] In order to solve the technical problem that during the continuous charging and discharging process of the energy storage system, the capacity of the battery in the energy storage system will decrease, resulting in accelerated battery aging, and deep discharge will also cause damage to the battery, thereby affecting the life and efficiency of the energy storage system, the present invention proposes a big data-based energy storage coordination control method, the method comprising the following steps: obtaining the historical power consumption of the energy storage system and the storage capacity, power supply and attenuation rate of each battery; inputting the historical power consumption into the prediction model to obtain the power consumption in the next time period; calculating the power supply priority of each battery, and the power supply priority meets the requirements of the prediction model. , For the The power supply priority of each battery, and Respectively The decay rate and normalized storage capacity of a battery, is a natural exponential function; the objective function of the power supply sequence of the energy storage system is obtained through the iteration of the genetic algorithm. In each round of iteration, the following steps are performed: for multiple batteries pre-selected by the genetic algorithm for power supply, the pre-selected batteries are divided into two categories according to the storage capacity; the pre-selected batteries in each category are clustered according to the positions of the pre-selected batteries to obtain several initial clustering clusters for each category; the clustering clusters that do not belong to the same category are merged according to the distances between the clustering centers of the initial clustering clusters to obtain several final clustering clusters; the distributed heat dissipation coefficient of the pre-selected batteries is calculated. , For the The density of the final clusters, For the The shortest distance between the final cluster and the cluster centers of the remaining final clusters is calculated; the objective function of this round of iteration is calculated based on the power supply priority and distributed heat dissipation coefficient of the pre-selected batteries and the power consumption in the next time period; according to the solution of the objective function of each round of iteration, the coordinated control of energy storage based on big data is realized.

[0006] The beneficial effects are: by reasonably scheduling according to the power supply priority and attenuation rate of the battery, damage to the battery caused by deep discharge is avoided, thereby effectively extending the battery life; controlling the charge and discharge depth of the battery to keep it within the optimal working range and slowing down the rate of battery capacity decay; using historical power consumption and prediction models to accurately predict the power consumption in the next time period, improving the response speed and power supply stability of the energy storage system, and reducing possible energy losses during the power supply process; intelligently adjusting the power supply priority according to the dynamic changes of the battery attenuation rate and storage capacity to ensure that the energy storage system operates efficiently under different load conditions; determining the power supply priority of the battery through cluster analysis, classifying the battery according to location and performance, optimizing the battery combination, and improving the flexibility and efficiency of power supply; reducing unnecessary energy losses and achieving more efficient energy utilization by optimizing scheduling and power supply strategies.

[0007] Furthermore, the prediction model adopts an autoregressive moving average model.

[0008] Furthermore, the pre-selected batteries are divided into two categories according to the amount of storage capacity, including: the pre-selected batteries with a storage capacity not less than a charging threshold are divided into a discharging battery category; the pre-selected batteries with a storage capacity less than the charging threshold are divided into a charging and discharging battery category.

[0009] The beneficial effect is that classifying the pre-selected batteries according to their storage capacity not only helps to improve the efficiency and safety of battery use, but also provides strong support for the optimization of the overall battery management system.

[0010] Furthermore, the clustering adopts a mean shift algorithm.

[0011] Furthermore, the method for obtaining the final clustering cluster is as follows: the clustering clusters whose distances between cluster centers of the initial clustering clusters are less than the fusion threshold and whose distances are closest are merged; the clustering clusters whose distances between cluster centers of the fused clustering clusters and the unfused clustering clusters are less than the fusion threshold and whose distances are closest are continued to be merged until the distances between cluster centers of the fused clustering clusters and the unfused clustering clusters are both greater than the fusion threshold, the fusion is terminated, and several final clustering clusters are obtained, and the final clustering clusters include the fused clustering clusters and the unfused clustering clusters.

[0012] The beneficial effects are: by fusing clusters with close distances, the redundancy between clusters can be effectively reduced and the accuracy of clustering results can be improved; cluster fusion helps to simplify data representation, making the final cluster results more representative, and merging similar clusters can reduce information loss.

[0013] Furthermore, the density of the final cluster is obtained by: determining the minimum convex polygon containing all pre-selected batteries in the cluster by a convex hull algorithm, and taking the ratio of the number of all pre-selected batteries in the cluster to the area of ​​the minimum convex polygon as the density of the cluster; and taking the maximum value of the density of the fused cluster and the initial cluster at the end of the fusion contained in the final cluster as the density of the final cluster.

[0014] The beneficial effects are: the convex hull algorithm can effectively process clusters of different densities, avoid the confusion of clustering results in the case of uneven density, and ensure the separation and clarity between clusters; by combining the density information of the initial cluster and the fused cluster, comprehensively considering data from different sources, the effectiveness of clustering and the fusion effect are enhanced, and more comprehensive information is provided for subsequent calculations.

[0015] Furthermore, the objective function satisfies the following relationship:

[0016] ; In the formula, is the objective function of the power supply sequence of the energy storage system in this iteration, For the The power supply of the pre-selected battery, is the power consumption in the next time period, For the The priority of the block pre-selected battery power supply, is the distributed heat dissipation coefficient of the preselected battery.

[0017] The beneficial effects are: by introducing priority, the objective function allows the energy storage system to more flexibly select power supply strategies, ensuring that high-priority batteries provide more power during high-demand periods, thereby improving the reliability and response speed of the system; by adding a heat dissipation coefficient, it helps prevent performance degradation or safety hazards caused by overheating, thereby improving the operating efficiency and service life of the battery.

[0018] Furthermore, the implementation of energy storage coordinated control based on big data includes: taking the maximum value in the solution of the objective function of the power supply sequence obtained in each round of iteration as the optimal solution, and the power supply sequence corresponding to the optimal solution is the final power supply sequence for the next time period, thereby completing the energy storage coordinated control based on big data.

[0019] The present invention has the following beneficial effects:

[0020] The prediction model is used to obtain future electricity demand, and the priority of battery power supply is calculated according to the attenuation rate and storage capacity of each battery, so as to encourage the energy storage system to use batteries with low attenuation rates for power supply, thereby improving power supply efficiency. In the iterative process, the batteries are clustered and analyzed, and the heat dissipation coefficient is determined. Taking into account the heat dissipation requirements of the batteries during operation, the problem of concentrated distribution of power supply batteries affecting battery heat dissipation and causing potential safety hazards is solved. The objective function of the power supply sequence is determined, and the optimal power supply strategy is obtained using a genetic algorithm to achieve an efficient and reliable power supply solution, thereby ensuring the life and efficiency of the energy storage system, as well as its stability and economy during long-term operation. In summary, the present invention can effectively extend battery life, reduce deep discharge, improve system efficiency and reliability, and reduce maintenance costs. BRIEF DESCRIPTION OF THE DRAWINGS

[0021] The above and other objects, features and advantages of the exemplary embodiments of the present invention will become readily understood by reading the following detailed description with reference to the accompanying drawings. In the accompanying drawings, several embodiments of the present invention are shown in an exemplary and non-limiting manner, and the same or corresponding reference numerals represent the same or corresponding parts, wherein:

[0022] Figure 1 It is a flowchart of the steps of a big data-based energy storage coordinated control method according to an embodiment of the present invention.

[0023] Figure 2 It is a flowchart of the steps of iteratively acquiring an objective function of a power supply sequence of an energy storage system in a big data-based energy storage coordinated control method according to an embodiment of the present invention. DETAILED DESCRIPTION

[0024] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative work are within the scope of protection of the present invention.

[0025] The specific embodiments of the present invention are described in detail below with reference to the accompanying drawings.

[0026] See also Figure 1 , which shows a flow chart of the steps of a big data-based energy storage coordinated control method provided by an embodiment of the present invention, the method comprising steps S1 to S5:

[0027] S1: Obtain the historical power consumption of the energy storage system and the storage capacity, power supply and attenuation rate of each battery.

[0028] The historical power consumption of the energy storage system, that is, the historical load power consumption data, including the historical load power consumption and its corresponding time, wherein the historical load power consumption is the power supply data recorded when the energy storage system supplies power to the load; for abnormal values ​​in the historical load power consumption data caused by equipment failure or other reasons, they are deleted and treated as missing values, and the missing values ​​are supplemented by spline interpolation to ensure the integrity of the data; the battery capacity of each battery in the energy storage system is tested to obtain the attenuation rate of each battery.

[0029] Implementers can set the collection time according to specific implementation circumstances, for example, collecting data once every hour.

[0030] S2: Input the historical electricity consumption into the prediction model to obtain the electricity consumption in the next time period.

[0031] It should be noted that, based on the trend of power consumption over time in historical data, the prediction model is used to predict the next time period. In this embodiment, the power consumption data for the next hour is used to coordinate the power supply of the energy storage system battery for the next hour.

[0032] Specifically, the prediction model adopts an autoregressive moving average model.

[0033] Since historical load power consumption data has strong time series characteristics, the AutoRegressive Integrated Moving Average (ARIMA) model can more effectively capture the dynamic characteristics of time series data and make predictions.

[0034] S3: Calculate the power supply priority of each battery.

[0035] It should be noted that as the battery is used and time goes by, the internal resistance of the battery will usually increase, which will result in the battery not being able to fully use the electricity it discharges when it is supplying power to the load. This part of the electricity is the power loss. As the battery attenuation rate increases, the power loss when the battery is supplying power to the load will also increase, so batteries with lower attenuation rates should be given priority. In addition, deep discharge will accelerate battery loss and reduce battery life, so batteries with more power storage should also be given priority. Therefore, combining the above two factors will determine the priority of each battery in the power supply process of the energy storage system.

[0036] The power supply priority satisfies the following relationship:

[0037] ;

[0038] In the formula, For the The priority of battery power supply, For the The decay rate of a battery, For the Normalized result of battery storage capacity.

[0039] Among them, The decay rate of a battery The lower the power supply, the higher the priority of the load. Battery capacity The more there are, the higher the priority of supplying power to the load.

[0040] S4: Obtain the objective function of the power supply sequence of the energy storage system through genetic algorithm iteration.

[0041] After the maximum number of iterations is reached, the iteration is terminated. The implementer can set the maximum number of iterations according to the specific implementation situation, for example, 100 times.

[0042] In each iteration, refer to Figure 2 , step S4 includes steps S401 to S405, which are specifically as follows:

[0043] S401: For multiple batteries pre-selected by the genetic algorithm for power supply, the pre-selected batteries are divided into two categories according to the amount of power storage.

[0044] It should be noted that when selecting a power supply battery for powering a load, in addition to considering the priority, it is necessary to set a threshold for the battery storage capacity to determine whether the battery needs to be charged.

[0045] Specifically, the pre-selected batteries are divided into two categories according to the amount of power storage, including:

[0046] Classify the pre-selected batteries having a storage capacity not less than a charging threshold into a discharge battery category;

[0047] The pre-selected batteries with a storage capacity less than a charging threshold are classified into charging and discharging battery categories.

[0048] Implementers can set the charging threshold, for example, 50%, based on specific implementation circumstances.

[0049] S402: Clustering the pre-selected batteries in each category according to the positions of the pre-selected batteries to obtain a number of initial clusters for each category.

[0050] It should be noted that since batteries generate heat during the charging and discharging process, if the selected batteries that need to be powered are too concentrated, the heat dissipation effect of the batteries will be affected. If the heat is not effectively dissipated, the battery temperature may be too high, thereby accelerating its attenuation, and further affecting the battery life, performance and safety. The heat dissipation requirements of discharge batteries and charge and discharge batteries are not exactly the same, so it is necessary to hierarchically cluster the pre-selected batteries for subsequent density analysis.

[0051] Specifically, the clustering adopts a mean shift algorithm.

[0052] S403: Clusters that do not belong to the same category are merged according to the distances between the cluster centers of the initial clusters to obtain several final clusters.

[0053] Specifically, the method for obtaining the final cluster is:

[0054] The clusters whose distances between the cluster centers of the initial clusters are less than the fusion threshold and whose distances are closest are merged;

[0055] The clusters whose distances between the cluster centers of the fused clusters and the unfused clusters are less than the fusion threshold and whose distances are closest are continued to be fused until the distances between the cluster centers of the fused clusters and the unfused clusters are both greater than the fusion threshold. The fusion is completed and several final clusters are obtained. The final clusters include the fused clusters and the unfused clusters.

[0056] Implementers can set the fusion threshold according to specific implementation circumstances, for example, 1.

[0057] S404: Calculate the distributed heat dissipation coefficient of the preselected battery.

[0058] It should be noted that due to the concentrated distribution characteristics of batteries in the energy storage system, the smaller the distance between the second clusters, the more concentrated the distribution of the energy storage batteries, the greater the adverse effect on the heat dissipation of the working batteries, and the more likely it is to cause safety hazards. Therefore, the lower limit of the distance should be limited to reduce the impact on battery life. Therefore, the shortest cluster center distance is selected in the subsequent calculation of the heat dissipation coefficient.

[0059] The distributed heat dissipation coefficient satisfies the following relationship:

[0060] ;

[0061] In the formula, To preselect the distributed heat dissipation coefficient of the battery, For the The density of the second cluster, For the The shortest cluster center distance between the second cluster and the remaining second clusters, is the natural exponential function, is the summation symbol.

[0062] in, Indicates The density of the second cluster, Indicates The shortest cluster center distance between the second cluster and the rest of the second clusters. The greater the cluster density, the denser the distribution of batteries contained in the cluster, the less conducive to battery heat dissipation, and the smaller the heat dissipation coefficient; the smaller the distance between clusters, the denser the overall battery distribution, the less conducive to battery heat dissipation, and the smaller the heat dissipation coefficient. Therefore, when selecting a battery to power the load, the smaller the heat dissipation coefficient, the greater the impact on battery life, and the more adjustments should be made to the selection of the power supply battery.

[0063] It should be noted that for the concentrated distribution characteristics of batteries in energy storage systems, the greater the cluster density, the more unfavorable it is for battery heat dissipation and the more likely it is to cause safety hazards. Therefore, the upper limit of the density should be limited to minimize the impact on battery life, so the maximum value of the density should be selected.

[0064] Specifically, the density of the final cluster is obtained as follows:

[0065] Determine the minimum convex polygon containing all pre-selected batteries in the cluster by using the convex hull algorithm, and take the ratio of the number of all pre-selected batteries in the cluster to the area of ​​the minimum convex polygon as the density of the cluster;

[0066] The maximum value of the density of the fused cluster and the initial cluster at the end of the fusion contained in the final cluster is taken as the density of the final cluster.

[0067] S405: Calculate the objective function of this round of iteration based on the power supply priority and distributed heat dissipation coefficient of the pre-selected battery and the power consumption in the next time period.

[0068] It should be noted that the adjustment of the power supply strategy should be based on the total power supply of the pre-selected batteries meeting the power consumption in the next time period, and the battery power supply priority should be calculated according to the battery attenuation rate and storage capacity. Taking into account the heat dissipation characteristics of the battery, the pre-selected batteries should be clustered and then analyzed for concentration, so as to minimize the loss of battery life and determine the objective function of the power supply sequence of the energy storage system.

[0069] Specifically, the objective function satisfies the following relationship:

[0070] ;

[0071] In the formula, is the objective function of the power supply sequence of the energy storage system in this iteration, For the The power supply of the pre-selected battery, is the power consumption in the next time period, For the The priority of the block pre-selected battery power supply, To preselect the distributed heat dissipation coefficient of the battery, is the summation symbol.

[0072] in, Indicates the difference between the total power supply of the pre-selected battery and the power consumption in the next time period. Indicates The priority of the pre-selected battery power supply is determined, and the power supply of the pre-selected battery is weighted by the priority. The higher the priority, the greater the power supply of the battery, thereby reducing battery loss.

[0073] S5: Based on the solution of the objective function of each round of iteration, the coordinated control of energy storage based on big data is realized.

[0074] It should be noted that in the process of solving the objective function, the solution of the objective function in each iteration is different. The heat dissipation coefficient is the result of the concentration analysis of the power supply battery selected in each iteration. The larger the heat dissipation coefficient, the more dispersed the distribution of the selected power supply battery is, and the more conducive to the heat dissipation of the battery. The power supply battery selection strategy is adjusted in this way, and finally the maximum value of all solutions is selected as the optimal solution.

[0075] Specifically, the energy storage coordinated control based on big data includes:

[0076] The power supply sequence corresponding to the maximum value in the solution of the objective function of the power supply sequence obtained in each round of iteration is used as the final power supply sequence for the next time period to complete the energy storage coordinated control based on big data.

[0077] The embodiment of the present invention implements an efficient and reliable power supply solution through the above steps, thereby ensuring the stability and economy of the energy storage system during long-term operation.

[0078] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the principles of the present invention should be included in the protection scope of the present invention.

Claims

1. A method for coordinated control of energy storage based on big data, characterized in that: include: Obtain the historical power consumption of the energy storage system and the storage capacity, power supply and attenuation rate of each battery; input the historical power consumption into the prediction model to obtain the power consumption in the next time period; Calculate the power supply priority of each battery to meet , For the The power supply priority of each battery, and Respectively The decay rate and normalized storage capacity of a battery, is the natural exponential function; The objective function of the power supply sequence of the energy storage system is obtained through the iteration of the genetic algorithm. In each round of iteration, the multiple batteries pre-selected by the genetic algorithm for power supply are divided into two categories according to the storage capacity; the pre-selected batteries in each category are clustered according to the positions of the pre-selected batteries, and several initial clusters of each category are obtained; Clusters that do not belong to the same category are merged according to the distance between the cluster centers of the initial clusters. The method is: the clusters whose distance between the cluster centers of the initial clusters is less than the fusion threshold and the closest distance are merged; The clusters whose distances between the cluster centers of the fused clusters and the unfused clusters are less than the fusion threshold and whose closest distances are close to each other are continued to be fused until the distances between the cluster centers of the fused clusters and the unfused clusters are both greater than the fusion threshold. After the fusion is completed, several final clusters are obtained. Calculate the distributed heat dissipation coefficient of the pre-selected battery , For the The density of the final clusters, For the The shortest distance between the final cluster and the cluster centers of the remaining final clusters; based on the power supply priority and distribution heat dissipation coefficient of the pre-selected battery and the power consumption in the next time period, calculate the objective function of this round of iteration; The density of the final cluster is: determine the minimum convex polygon containing all pre-selected batteries in the cluster by the convex hull algorithm, and take the ratio of the number of all pre-selected batteries in the cluster to the area of ​​the minimum convex polygon as the density of the cluster; take the maximum value of the density of the fused cluster and the initial cluster at the end of the fusion contained in the final cluster as the density of the final cluster; According to the solution of the objective function of each round of iteration, energy storage coordinated control based on big data can be realized.

2. The energy storage coordinated control method based on big data according to claim 1 is characterized in that: The prediction model adopts an autoregressive moving average model.

3. The energy storage coordinated control method based on big data according to claim 1 is characterized in that: The pre-selected batteries are divided into two categories according to the amount of power storage, including: Classify the pre-selected batteries having a storage capacity not less than a charging threshold into a discharge battery category; The pre-selected batteries with a storage capacity less than a charging threshold are classified into charging and discharging battery categories.

4. The energy storage coordinated control method based on big data according to claim 1 is characterized in that: The clustering adopts the mean shift algorithm.

5. The energy storage coordinated control method based on big data according to claim 1 is characterized in that: The objective function satisfies the following relationship: ； In the formula, is the objective function of the power supply sequence of the energy storage system in this iteration, For the The power supply of the pre-selected battery, is the power consumption in the next time period, For the The priority of the block pre-selected battery power supply, is the distributed heat dissipation coefficient of the preselected battery.

6. The energy storage coordinated control method based on big data according to claim 1 is characterized in that: The energy storage coordinated control based on big data includes: The maximum value of the solution of the objective function of the power supply sequence obtained in each round of iteration is taken as the optimal solution. The power supply sequence corresponding to the optimal solution is the final power supply sequence for the next time period, completing the energy storage coordinated control based on big data.

Images