A method for optimizing energy storage configuration based on multi-source data fusion

Abstract

The invention discloses an energy storage configuration optimization method based on multi-source data fusion, which relates to the technical field of multi-source data fusion. The method collects comprehensive resource data of each energy storage device in an electric power system; processes the comprehensive resource data to obtain electric power operation parameters and a load characteristic curve diagram, analyzes the load characteristic curve diagram to obtain a peak segment and a low value segment; sets decision indicators, target threads and restriction conditions according to the obtained electric power operation parameters, peak segments and low value segments; generates dynamic parameters according to the decision indicators and combines them to obtain a random initial solution, obtains candidate individuals according to the random initial solution, the target thread and the restriction conditions, cross-replaces the candidate individuals to obtain cross-individuals, and cyclically compares the cross-individuals to obtain excellent individuals; effectively enhances the configuration efficiency of the electric power system, improves the operation efficiency and stability of the electric power system, and improves the data utilization efficiency and the response speed of the energy storage device.

Description

Technical Field

[0001] The present invention relates to the technical field of multi-source data fusion, and in particular to an energy storage configuration optimization method based on multi-source data fusion. Background Art

[0002] Multi-source data fusion refers to the integration, processing and analysis of data from different sources, types and forms to obtain more accurate, comprehensive and valuable information.

[0003] Energy storage configuration refers to the configuration relationship between energy storage devices (such as batteries, compressed air, supercapacitors, etc.) and power grids and loads (i.e. users) in power systems. However, the existing energy storage configuration methods still have some shortcomings: insufficient energy storage scale leads to power shortages in power systems during peak hours, affecting power supply stability and reliability; current energy storage technology is not yet fully mature, and low-cost, large-capacity, long-term, and cross-seasonal energy storage technology still needs to be broken through, which makes the energy storage system limited in terms of performance, cost, and sustainability, and it is difficult to meet the high requirements of the power system for energy storage; however, the existing energy storage system still has shortcomings in terms of rapid response and precise control, which may lead to increased risks in power grid operation; therefore, optimizing the configuration of traditional high-energy consumption energy storage equipment has important theoretical and practical significance.

[0004] How to use multi-source data fusion technology to collect and process the comprehensive resource data of energy storage equipment, set decision indicators, target threads and constraints according to the processed comprehensive resource data, obtain the energy loss value of the energy storage equipment, screen the energy loss value, obtain candidate individuals, cross-replace and cyclically compare the candidate individuals, and obtain excellent individuals are the problems we need to solve. To this end, a method for energy storage configuration optimization based on multi-source data fusion is provided. Summary of the invention

[0005] The purpose of the present invention can be achieved through the following technical solutions:

[0006] Step S1: Collecting comprehensive resource data of each energy storage device in the power system;

[0007] Step S2: Processing the obtained comprehensive resource data to obtain power operation parameters and load characteristic curves, and analyzing the load characteristic curves to obtain peak segments and low value segments;

[0008] Step S3: setting decision indicators, target threads and constraints according to the obtained power operation parameters, peak segments and low value segments, generating dynamic parameters according to the obtained decision indicators and target threads, combining the dynamic parameters and obtaining a random initial solution;

[0009] Step S4: Obtaining energy loss values ​​of energy storage devices according to random initial solutions, sorting the energy loss values, obtaining candidate individuals, performing crossover replacement on the random initial solutions of the candidate individuals, obtaining crossover individuals, setting intervention factors, obtaining fitness values ​​of crossover individuals according to the intervention factors, and comparing the fitness values ​​until a stop condition is met;

[0010] Step S5: Repeat step S4 until both the restriction condition and the stop condition are met and an excellent individual is obtained.

[0011] The process of collecting comprehensive resource data includes:

[0012] Setting a capture end according to the energy storage device in the power system, wherein the capture end includes a collection node and a monitoring node;

[0013] Associating the obtained energy storage device with the corresponding collection node and monitoring node;

[0014] Acquire power data through a collection node at the capture end, wherein the power data includes generator data, transformer data, and transmission line data;

[0015] The load data is acquired through the monitoring node at the capture end, and the load data includes the power load amount and the load demand amount.

[0016] The process of processing comprehensive resource data includes:

[0017] Extracting the obtained generator data, transformer data and transmission line data to obtain power operation parameters, and associating the obtained power operation parameters with corresponding energy storage devices;

[0018] According to the obtained power load, a two-dimensional rectangular coordinate system with respect to the power load in time is established, and a power load variation curve is generated according to the obtained power load. The obtained power load variation curve is mapped into the two-dimensional rectangular coordinate system to obtain a load characteristic curve diagram.

[0019] The process of analyzing the load characteristic curve diagram includes:

[0020] Set the statistical period according to the obtained load characteristic curve and mark the duration of the statistical period;

[0021] Obtain the average power consumption within the statistical period according to the obtained load characteristic curve diagram;

[0022] Setting a first power usage threshold and a second power usage threshold according to the obtained power usage average value of the statistical period and the duration of the statistical period;

[0023] The obtained first power consumption threshold and second power consumption threshold are marked on the load characteristic curve diagram, the time between two adjacent intersection points of the power load change curve and the first power consumption threshold is marked as a low value segment, and the time between two adjacent intersection points of the power load change curve and the second power consumption threshold is marked as a peak segment.

[0024] The process of obtaining decision indicators, target threads, and constraints includes:

[0025] Setting decision indicators according to the obtained peak value segment, low value segment and power operation parameters, wherein the decision indicators include energy demand, charging and discharging efficiency and power balance;

[0026] Obtaining a target thread according to the obtained energy demand, charge and discharge efficiency, and power balance;

[0027] Set the limit condition based on the target thread obtained.

[0028] The process of generating dynamic parameters and combining them to obtain a random initial solution includes:

[0029] Setting a threshold range for the target thread according to the obtained constraint condition;

[0030] Generate a dynamic parameter according to the obtained threshold range, wherein the dynamic parameter includes a dynamic integer and a dynamic floating point number;

[0031] The obtained dynamic integers and dynamic floating-point numbers are combined to obtain a random initial solution.

[0032] The process of obtaining the fitness value of the energy storage device according to the random initial solution includes:

[0033] Obtaining an energy loss value of the energy storage device according to the obtained random initial solution;

[0034] Obtaining an individual selection rate according to the energy loss value of the obtained energy storage device;

[0035] Determine candidate individuals according to the individual selection rate of each energy storage device;

[0036] Obtaining the collection nodes of the energy storage device, selecting one of the candidate individuals as the reference individual, performing crossover replacement on the random initial solutions at the collection nodes of the reference individual and the collection nodes corresponding to the remaining candidate individuals to obtain the crossover individuals;

[0037] An intervention factor is set for the crossover individual according to the obtained load demand, and a fitness value of the crossover individual is obtained according to the obtained intervention factor, the crossover individual and the energy loss value of the random initial solution;

[0038] The fitness values ​​of the obtained crossover individuals are screened to obtain a number of excellent individuals, and the fitness mean is obtained according to the fitness values ​​of the obtained excellent individuals;

[0039] The fitness threshold is set according to the obtained fitness value and the fitness mean, and the obtained fitness value of the crossover individual is compared with the fitness threshold. If the fitness value is not greater than the fitness threshold, it means that the stopping condition is met.

[0040] The process of obtaining outstanding individuals includes:

[0041] Obtain the random initial solution of the acquisition node corresponding to the crossover individual when the stop condition is met, and record the time when the stop condition is met as a crossover cycle, mark the obtained random initial solution as the pre-cycle initial solution, use the pre-cycle initial solution as the random initial solution of the next crossover cycle, and repeat the process of obtaining the fitness value of the energy storage device until both the stop condition and the restriction condition are met;

[0042] Obtain the fitness values ​​of all crossover individuals in several crossover cycles, and mark the crossover individual with the smallest fitness value as an excellent individual.

[0043] Compared with the prior art, the beneficial effects of the present invention are: processing the collected comprehensive resource data of the energy storage device to obtain the power operation parameters and the load characteristic curve diagram, analyzing the load characteristic curve diagram to obtain the peak value segment and the low value segment, and setting the decision index, target thread and restriction condition according to the obtained power operation parameters, peak value segment and low value segment;

[0044] According to the obtained restriction conditions, a threshold range is set for the target thread, dynamic parameters are generated within the threshold range, the dynamic parameters are randomly combined to obtain a random initial solution, the energy loss value of the energy storage device is obtained according to the obtained random initial solution, the energy loss value is screened to obtain candidate individuals, the candidate individuals are cross-replaced to obtain crossover individuals, and intervention factors are introduced to obtain the fitness value of the crossover individual, and the fitness value is compared with the fitness mean until the stopping condition is met, and the random initial solution of the crossover individual that meets the stopping condition is recorded as the pre-cycle initial solution as the random initial solution of the next crossover cycle until the stopping condition and the restriction conditions are met at the same time to obtain an excellent individual;

[0045] Effectively enhance the configuration efficiency of the power system, improve the operating efficiency and stability of the power system, improve data utilization efficiency and the response speed of energy storage equipment, and ensure that the optimization results have high reliability while meeting operational requirements. BRIEF DESCRIPTION OF THE DRAWINGS

[0046] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0047] Figure 1 It is a schematic diagram of the present invention. DETAILED DESCRIPTION

[0048] The technical solution of the present invention will be clearly and completely described below in conjunction with the embodiments. Obviously, the described embodiments are only 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 ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0049] like Figure 1 As shown, a method for optimizing energy storage configuration based on multi-source data fusion includes the following steps:

[0050] Step S1: Collecting comprehensive resource data of each energy storage device in the power system;

[0051] Step S2: Processing the obtained comprehensive resource data to obtain power operation parameters and load characteristic curves, and analyzing the load characteristic curves to obtain peak segments and low value segments;

[0052] Step S3: setting decision indicators, target threads and constraints according to the obtained power operation parameters, peak segments and low value segments, generating dynamic parameters according to the obtained decision indicators and target threads, combining the dynamic parameters and obtaining a random initial solution;

[0053] Step S4: Obtaining energy loss values ​​of energy storage devices according to random initial solutions, sorting the energy loss values, obtaining candidate individuals, performing crossover replacement on the random initial solutions of the candidate individuals, obtaining crossover individuals, setting intervention factors, obtaining fitness values ​​of crossover individuals according to the intervention factors, and comparing the fitness values ​​until a stop condition is met;

[0054] Step S5: Repeat step S4 until both the restriction condition and the stop condition are met and an excellent individual is obtained.

[0055] It should be further explained that, in the specific implementation process, the process of collecting comprehensive resource data includes:

[0056] A capture end is set according to the energy storage device, wherein the capture end includes a collection node and a monitoring node, and the obtained collection node is marked as i, wherein i=1, 2, 3, ..., n1, n1 is a positive integer, and the obtained monitoring node is marked as j, wherein j=1, 2, 3, ..., n2, n2 is a positive integer;

[0057] Associating the obtained energy storage device with the corresponding collection node and monitoring node, and marking the energy storage device as z, where z=1, 2, 3, ..., n3, n3 is a positive integer;

[0058] It should be further explained that, in the specific implementation process, the energy storage equipment includes pumped storage, compressed air storage, lithium-ion battery storage, supercapacitor storage, sodium-sulfur battery storage, flywheel storage, electromagnetic storage, etc.;

[0059] Collecting comprehensive resource data of the energy storage device through the capture end, wherein the comprehensive resource data includes power data and load data;

[0060] Acquire power data through a collection node, wherein the power data includes generator data, transformer data, and transmission line data;

[0061] Acquiring load data through monitoring nodes, wherein the load data includes power load and load demand;

[0062] The obtained power load is marked as W j , and mark the obtained load demand as WX k , where k is the user, k = 1, 2, 3, ..., n4, n4 is a positive integer.

[0063] Extracting the obtained generator data, transformer data and transmission line data to obtain power operation parameters, and associating the obtained power operation parameters with corresponding energy storage devices;

[0064] The power operation parameters include generator power, transformer input voltage, transformer output voltage, transformer current, transmission line current load and transmission line voltage loss. The obtained generator power, transformer input voltage, transformer output voltage, transformer current, transmission line current load and transmission line voltage loss are marked as FG i , BR i , B.C. i BL i DF i and DS i ;

[0065] According to the obtained power load, a two-dimensional rectangular coordinate system with respect to the power load is established, a power load variation curve is generated according to the obtained power load, and the obtained power load variation curve is mapped into the two-dimensional rectangular coordinate system to obtain a load characteristic curve diagram;

[0066] According to the obtained load characteristic curve, a statistical period is set, which is denoted as m, where m=1, 2, 3, ..., n5, n5 is a positive integer, and the duration of the statistical period is marked as T;

[0067] The average power consumption of the statistical period is obtained according to the obtained load characteristic curve, and the obtained average power consumption is marked as in f1, f2, ..., f j is the impact factor;

[0068] The first power usage threshold and the second power usage threshold are set according to the obtained power usage average value, and the obtained first power usage threshold is marked as YZ1, where: α1 is the threshold coefficient; the obtained second power consumption threshold is marked as YZ2, where,

[0069] The first power consumption threshold and the second power consumption threshold obtained are marked on the load characteristic curve diagram, and the time between two adjacent intersections of the power load change curve and the first power consumption threshold is marked as a low value segment, denoted as W min ;

[0070] The time between two adjacent intersections of the power load change curve and the second power consumption threshold is marked as the peak segment, denoted as W max .

[0071] Setting decision indicators according to the obtained peak value segment, low value segment and power operation parameters, wherein the decision indicators include energy demand, charge and discharge efficiency and power balance, and marking the obtained energy demand, charge and discharge efficiency and power balance as E, C and P respectively;

[0072] It should be further explained that, in the specific implementation process, the energy demand includes the load demand and the transmission line current load, the charging and discharging efficiency includes the generator power, the transformer input voltage and the transformer output voltage, and the power balance includes the peak segment, the low value segment, the transformer current and the transmission line voltage loss;

[0073] The target thread is obtained according to the obtained energy demand, charge and discharge efficiency, and power balance, and the obtained target thread is marked as f(E, C, P), where f(E, C, P) = [E] 2 +[C] 2 +[P] 2 ,and And the values ​​of e and k are the same; Δd1 and Δd2 are transformer influencing variables,

[0074] Setting restriction conditions according to the obtained target thread, wherein the restriction conditions include a first restriction range, a second restriction range, a third restriction range, and a loop restriction range;

[0075] The first limit range is denoted as E min ≤E≤E max , E max Indicates the maximum energy demand, E min Indicates the minimum energy requirement;

[0076] The second limit range is denoted as C min ≤C≤C max , C max Indicates the maximum charge and discharge efficiency, C min Indicates the minimum value of charge and discharge efficiency;

[0077] The third limit range is denoted as P min ≤P≤P max ;P max Indicates the maximum power balance, P min Indicates the minimum value of power balance;

[0078] Set the number of loops, mark the obtained number of loops as N, and satisfy N≤N max , where N max Indicates the maximum value of the loop limit range;

[0079] An energy threshold range, an efficiency threshold range, and a power threshold range are set for energy demand, charge and discharge efficiency, and power balance according to the obtained constraint conditions, and the energy threshold range, the efficiency threshold range, and the power threshold range are respectively provided with an upper threshold limit and a lower threshold limit;

[0080] Generate a dynamic parameter according to the obtained energy threshold range, efficiency threshold range and power threshold range, wherein the dynamic parameter includes a dynamic integer and a dynamic floating point number, wherein the value of the dynamic integer includes an upper threshold limit and a lower threshold limit; the value of the dynamic floating point number includes the lower threshold limit but does not include the upper threshold limit. It should be further explained that, in a specific implementation process, the dynamic floating point number represents a randomly generated decimal within the threshold range, such as 3.7, 6.2;

[0081] The obtained dynamic integers and dynamic floating-point numbers are randomly combined to obtain a random initial solution, and the obtained random initial solution is marked as y, where y=1, 2, 3, ..., n6, and n6 is a positive integer;

[0082] Constructing a dynamic parameter set according to the obtained random initial solution, uploading the obtained random initial solution to the dynamic parameter set, marking the size of the obtained dynamic parameter set as g, in particular, the size of the dynamic parameter set represents the number of random initial solutions in the dynamic parameter set;

[0083] Associating the obtained dynamic parameter set with the corresponding energy storage device, and repeating the above process of generating dynamic parameters to obtain the dynamic parameter set corresponding to each energy storage device;

[0084] Constructing an energy storage group, and uploading the obtained dynamic parameter set of each energy storage device to the energy storage group;

[0085] The energy loss value of the energy storage device is obtained according to the random initial solution in the obtained energy storage set group, and the obtained energy loss value of the energy storage device is marked as F z-y ,in a1 represents the loss change. When a large energy loss value is required, a1>2, and when a small energy loss value is required, 0<a1<2;

[0086] The individual selection rate is obtained according to the energy loss value of the random initial solution of the energy storage device, and the obtained individual selection rate is marked as XL z-y ,in

[0087] The obtained individual selection rates are sorted from large to small, and the energy storage devices ranked in the top h1 positions are marked as pre-selected individuals;

[0088] The energy loss values ​​of the pre-selected individuals are sorted from small to large, and the pre-selected individuals in the first h2 positions are marked as candidate individuals;

[0089] Obtain the collection node of the energy storage device, select one of the candidate individuals as the reference individual, and cross-replace the collection node of the reference individual with the random initial solution of the collection node corresponding to the remaining candidate individuals in turn. Each time the cross-replacement is performed, the reference individual after the cross-replacement is marked once until the reference individual and all the candidate individuals are replaced. The reference individual after the cross-replacement is marked as the cross-individual, and the power operation parameters at the collection node of the cross-individual are obtained and marked as FG respectively. i ', BR i ', BC i ', BL i ', DF i ' and DS i ';

[0090] Similarly, the remaining candidate individuals are marked as reference individuals in turn, and the above cross-replacement process is repeated until all candidate individuals have completed cross-replacement;

[0091] It should be further explained that, in the specific implementation process, the crossover replacement in which the random initial solution at the collection node of the reference individual is sequentially replaced with the collection nodes corresponding to the remaining candidate individuals is a random replacement, and the number of crossover replacements is large enough to obtain enough crossover individuals;

[0092] According to the obtained load demand, an intervention factor is set for the crossover individual. According to the obtained intervention factor, the crossover individual and the energy loss value of the random initial solution, the fitness value of the crossover individual is obtained. The obtained fitness value is marked as J XL ,in, α1 is the charge fluctuation index;

[0093] The fitness values ​​of the obtained crossover individuals are screened and sorted in order from small to large. The crossover individuals ranked in the top q% are marked as excellent individuals. The fitness mean is obtained according to the fitness values ​​of the obtained excellent individuals, and the obtained fitness mean is marked as in Q is the total number of crossover individuals;

[0094] The fitness threshold is set according to the obtained fitness value and the fitness mean.

[0095] The response threshold is marked as ΔJ, where α2 is the threshold influencing factor, and α2 satisfies 0<α2<q%;

[0096] Obtain the fitness value of the crossover individual, compare the obtained fitness value of the crossover individual with the fitness threshold, and if ΔJ≥J is satisfied XL , it means that the stop condition is met.

[0097] Obtain the random initial solution of the acquisition node corresponding to the crossover individual when the stop condition is met, and record the time when the stop condition is met as a crossover cycle, mark the obtained random initial solution as the pre-cycle initial solution, and use the pre-cycle initial solution as the random initial solution of the next crossover cycle, and repeat step S4 until ΔJ ≥ J is met. XL , and the number of cycles N≤N max , then stop the cycle;

[0098] Obtain the fitness values ​​of all crossover individuals in N cycles, mark the crossover individual with the smallest fitness value in N cycles as an excellent individual, obtain the random initial solution of the crossover replacement of the collection nodes of the excellent individual, and replace the original energy storage device accordingly according to the random initial solution of the crossover replacement of the collection nodes of the excellent individual.

[0099] The preferred embodiments of the present invention disclosed above are only used to help explain the present invention. The preferred embodiments do not describe all the details in detail, nor do they limit the invention to only specific implementation methods. Obviously, many modifications and changes can be made according to the content of this specification. This specification selects and specifically describes these embodiments in order to better explain the principles and practical applications of the present invention, so that those skilled in the art can understand and use the present invention well. The present invention is limited only by the claims and their full scope and equivalents.

Claims

1. A method for optimizing energy storage configuration based on multi-source data fusion, characterized in that: The following steps are involved: Step S1: Collecting comprehensive resource data of each energy storage device in the power system; Step S2: Processing the obtained comprehensive resource data to obtain power operation parameters and load characteristic curves, wherein the power operation parameters include generator power, transformer input voltage, transformer output voltage, transformer current, transmission line current load, and transmission line voltage loss, and analyzing the load characteristic curve to obtain peak segments and low value segments; Step S3: setting decision indicators, target threads and constraints according to the obtained power operation parameters, peak segments and low value segments, generating dynamic parameters according to the obtained decision indicators and target threads, combining the dynamic parameters and obtaining a random initial solution; Setting decision indicators according to the obtained peak value segment, low value segment and power operation parameters, wherein the decision indicators include energy demand, charging and discharging efficiency and power balance; Obtaining a target thread according to the obtained energy demand, charge and discharge efficiency, and power balance; The obtained target thread is marked as f(E, C, P), where f(E, C, P) = [E] 2 +[C] 2 +[P] 2 , where E represents energy demand, C represents charge and discharge efficiency, and P represents power balance; Set restriction conditions based on the target thread obtained; Setting a threshold range for the target thread according to the obtained constraint condition; Generate a dynamic parameter according to the obtained threshold range, wherein the dynamic parameter includes a dynamic integer and a dynamic floating point number; The obtained dynamic integers and dynamic floating-point numbers are combined to obtain a random initial solution; Step S4: Obtaining energy loss values ​​of energy storage devices according to random initial solutions, sorting the energy loss values, obtaining candidate individuals, performing crossover replacement on the random initial solutions of the candidate individuals, obtaining crossover individuals, setting intervention factors, obtaining fitness values ​​of crossover individuals according to the intervention factors, and comparing the fitness values ​​until a stop condition is met; Obtaining an energy loss value of the energy storage device according to the obtained random initial solution; Obtaining an individual selection rate according to the energy loss value of the obtained energy storage device; Determine candidate individuals according to the individual selection rate of each energy storage device; Obtaining the collection nodes of the energy storage device, selecting one of the candidate individuals as the reference individual, performing crossover replacement on the random initial solutions at the collection nodes of the reference individual and the collection nodes corresponding to the remaining candidate individuals to obtain the crossover individuals; An intervention factor is set for the crossover individual according to the obtained load demand, and a fitness value of the crossover individual is obtained according to the obtained intervention factor, the crossover individual and the energy loss value of the random initial solution; The obtained fitness value is marked as J XL ,in, α1 is the charge fluctuation index, WX K Indicates load demand, FG i ' represents the generator power at the cross-individual collection node, BR i ' represents the transformer input voltage at the cross-individual acquisition node, BC i ' represents the transformer output voltage at the cross-individual acquisition node, BL i ' represents the transformer current at the cross-individual acquisition node, DF i ' represents the transmission line current load at the cross-individual collection node, DS i ' represents the voltage loss of the transmission line at the cross-individual collection node; The fitness values ​​of the obtained crossover individuals are screened to obtain a number of excellent individuals, and the fitness mean is obtained according to the fitness values ​​of the obtained excellent individuals; The fitness threshold is set according to the obtained fitness value and fitness mean, and the obtained fitness value of the crossover individual is compared with the fitness threshold. If the fitness value is not greater than the fitness threshold, it means that the stop condition is met; Step S5: Repeat step S4 until both the restriction condition and the stop condition are met and an excellent individual is obtained.

2. The energy storage configuration optimization method based on multi-source data fusion according to claim 1 is characterized in that: The process of collecting comprehensive resource data includes: Setting a capture end according to the energy storage device in the power system, wherein the capture end includes a collection node and a monitoring node; Associating the obtained energy storage device with the corresponding collection node and monitoring node; Acquire power data through a collection node at the capture end, wherein the power data includes generator data, transformer data, and transmission line data; The load data is acquired through the monitoring node at the capture end, and the load data includes the power load amount and the load demand amount.

3. The energy storage configuration optimization method based on multi-source data fusion according to claim 2 is characterized in that: The process of processing comprehensive resource data includes: Extracting the obtained generator data, transformer data and transmission line data to obtain power operation parameters, and associating the obtained power operation parameters with corresponding energy storage devices; According to the obtained power load, a two-dimensional rectangular coordinate system with respect to the power load in time is established, and a power load variation curve is generated according to the obtained power load. The obtained power load variation curve is mapped into the two-dimensional rectangular coordinate system to obtain a load characteristic curve diagram.

4. The energy storage configuration optimization method based on multi-source data fusion according to claim 3 is characterized in that: The process of analyzing the load characteristic curve diagram includes: Set the statistical period according to the obtained load characteristic curve and mark the duration of the statistical period; Obtain the average power consumption within the statistical period according to the obtained load characteristic curve diagram; Setting a first power usage threshold and a second power usage threshold according to the obtained power usage average value of the statistical period and the duration of the statistical period; The obtained first power consumption threshold and second power consumption threshold are marked on the load characteristic curve diagram, the time between two adjacent intersection points of the power load change curve and the first power consumption threshold is marked as a low value segment, and the time between two adjacent intersection points of the power load change curve and the second power consumption threshold is marked as a peak segment.

5. The energy storage configuration optimization method based on multi-source data fusion according to claim 4 is characterized in that: The process of obtaining outstanding individuals includes: Obtain the random initial solution of the acquisition node corresponding to the crossover individual when the stop condition is met, and record the time when the stop condition is met as a crossover cycle, mark the obtained random initial solution as the pre-cycle initial solution, use the pre-cycle initial solution as the random initial solution of the next crossover cycle, and repeat the process of obtaining the fitness value of the energy storage device until both the stop condition and the restriction condition are met; Obtain the fitness values ​​of all crossover individuals in several crossover cycles, and mark the crossover individual with the smallest fitness value as an excellent individual.

Images