Centralized control chain energy storage grid auxiliary service system and method integrating multiple energy storage technologies

Abstract

The present invention belongs to the field of energy storage grid auxiliary technology, and discloses a centralized control chain energy storage grid auxiliary service system and method integrating multiple energy storage technologies; the method comprises: configuring M different types of energy storage technology units; connecting the communication and data transmission between the M energy storage technology units; collecting the energy storage characteristic data of the M energy storage technology units, and calculating the corresponding energy storage efficiency; predicting the energy demand value of the power grid, and combining the energy storage efficiency of the M energy storage technology units, and collaboratively managing the M energy storage technology units; the M energy storage technology units are interconnected with the power grid to provide power grid auxiliary services; the present invention realizes unified management of multiple energy storage technology units and energy configuration optimization to improve the energy utilization efficiency of the energy storage technology units, thereby avoiding the situation of temporary energy shortage caused by insufficient energy storage of energy storage technology units with low storage efficiency in the power grid auxiliary process, and effectively improving the reliability and operation efficiency of the power grid.

Description

Technical Field

[0001] The present invention relates to the field of energy storage grid auxiliary technology, and more specifically, to a centralized control chain energy storage grid auxiliary service system and method integrating multiple energy storage technologies. Background Art

[0002] With the rapid development of renewable energy and the transformation of power systems, the importance of energy storage technology in power systems has become increasingly prominent. However, different types of energy storage technologies are complementary and have complementary advantages. Their integration and application in grid auxiliary services can further improve the reliability and operating efficiency of power systems. Traditional energy storage grid auxiliary service methods have the following shortcomings:

[0003] 1. Limitations of single energy storage technology: Traditional energy storage technology can only meet specific power system needs and cannot fully meet the diverse needs of grid auxiliary services;

[0004] 2. Lack of collaborative management mechanism between energy storage technologies: Traditional energy storage technologies often work independently and lack collaborative management mechanisms, resulting in low energy utilization efficiency of the system;

[0005] 3. Lack of centralized control and collaborative management: Traditional energy storage grid auxiliary service systems lack centralized control and collaborative management, and cannot achieve unified management and optimized utilization of multiple energy storage technologies;

[0006] The Chinese patent application with publication number CN117833360A discloses a control method and system for a grid-connected converter based on a distributed architecture; the method includes: loading grid demand information, obtaining the output power of the first to the Nth distributed energy nodes, obtaining the output power of the energy storage battery, and then performing scheduling optimization to generate the distributed energy nodes to be scheduled, connecting to the grid-connected converter, monitoring the real-time output power and the fluctuation period, monitoring the real-time output power and the fluctuation period time, and performing grid-connected control when the power balance is met; the problem is that the traditional grid-connected converter has limited grid-connected capability when facing multiple distributed energy sources, and cannot fully consider the differences between different energy sources, so it is difficult to achieve the optimal grid-connected strategy, resulting in energy loss during the grid-connected process;

[0007] In the above technology, the goal is to minimize the power loss ratio of multiple energy nodes and meet the peak power demand of the power grid for grid connection control; however, in reality, different energy storage technologies have different storage efficiencies, that is, different energy storage technologies have different storage difficulties. If the storage efficiency is not considered, the energy storage technology with low energy storage efficiency will require a long time to store energy after the energy is exhausted. During this period, only the energy storage technology with high energy storage efficiency can be used for grid assistance, resulting in energy shortages, thereby affecting the performance and efficiency of grid auxiliary services;

[0008] In view of this, the present invention proposes a centralized control chain energy storage grid auxiliary service system and method that integrates multiple energy storage technologies to solve the above problems. Summary of the invention

[0009] In order to overcome the above-mentioned defects of the prior art and to achieve the above-mentioned purpose, the present invention provides the following technical solution: a centralized control chain energy storage grid auxiliary service method integrating multiple energy storage technologies, comprising:

[0010] Configure M different types of energy storage technology units, where M is an integer greater than 1;

[0011] Connecting M types of energy storage technology units for communication and data transmission;

[0012] Collect energy storage characteristic data of M types of energy storage technology units and calculate the corresponding energy storage efficiency;

[0013] Predict the energy demand value of the power grid, and combine the energy storage efficiency of M energy storage technology units to coordinate the management of M energy storage technology units;

[0014] M types of energy storage technology units are interconnected with the power grid to provide grid auxiliary services; grid auxiliary services include grid capacity expansion, peak shaving and valley filling, demand response and grid frequency regulation.

[0015] Furthermore, the M types of energy storage technology units realize communication and data transmission through a communication protocol stack, and the communication protocol stack includes communication standards, data transmission protocols and data formats.

[0016] The energy storage characteristic data include charging time, charging efficiency, discharge efficiency, self-discharge efficiency, cycle number and total cycle number;

[0017] The calculation methods of energy storage efficiency include:

[0018] ;

[0019] In the formula, is the energy storage efficiency, is the total number of cycles, is the number of cycles, For charging time, For charging efficiency, is the self-discharge efficiency, , , is the preset weight coefficient.

[0020] Furthermore, the method for predicting the power grid energy demand value includes:

[0021] A collection interval is preset, and N power grid energy demand values ​​are continuously collected according to the collection interval, where N is an integer greater than 1, and the continuously collected N power grid energy demand values ​​are input into the energy prediction model to predict the power grid energy demand values ​​at n future moments, where n is an integer greater than 1, and the predicted power grid energy demand values ​​are marked as predicted energy;

[0022] The training method of the energy prediction model includes:

[0023] Pre-collection continuous grid energy demand values ​​and construct a grid energy demand value set, , based on the set of grid energy demand values, training a prediction model for predicting grid energy demand values ​​at future moments;

[0024] Preset a sliding step size L and a sliding window length; use a sliding window method to convert the grid energy demand values ​​in the grid energy demand value set into multiple training samples, use the training samples as the input of the recurrent neural network model, predict the grid energy demand value after the sliding step size L as the output, use the subsequent grid energy demand value of each training sample as the prediction target, use the prediction accuracy as the training target, use the mean absolute percentage error MAPE to evaluate the model accuracy of the prediction result, and when the calculated MAPE is less than the preset MAPE, the energy prediction model training is completed; generate an energy prediction model that predicts the grid energy demand value at a future moment according to the grid energy demand value; wherein the energy prediction model is an RNN neural network model.

[0025] Furthermore, the step of collaboratively managing the M types of energy storage technology units includes:

[0026] Step 1: Randomly combine M energy storage technology units to form a set of R units. ; Each unit set includes h types of energy storage technology units, , and each unit set is different; set digital labels for R unit sets, and the range of digital labels is , mark the numerical labels of the unit set as the set label;

[0027] Step 2: Encode the set labels, obtain chromosomes, and construct the initial population;

[0028] Step 3: Determine the fitness function;

[0029] Step 4: Natural selection of chromosomes in the population;

[0030] Step 5: Perform crossover recombination on chromosomes in the population;

[0031] Step 6: mutate the chromosomes in the population;

[0032] Step 7: Get a new population, preset the population generation number as F, the fitness threshold as S, where F is an integer greater than 0 and S is a real number greater than 0; loop steps 4 to 6 until the generation number corresponding to the new population is F or the fitness corresponding to the chromosome in the new population is greater than or equal to the fitness threshold S, the loop ends, sort the fitness corresponding to each chromosome in the new population, and generate a first sorting table;

[0033] Step 8: Obtain the set label corresponding to the corresponding chromosome according to the positive order of the first sorting table, obtain the corresponding unit set according to the set label, obtain the corresponding energy storage technology unit according to the unit set, and mark it as a selected unit; obtain the remaining energy of the selected unit in real time, and the remaining energy is obtained by the built-in metering system of the energy storage technology unit; add the remaining energy corresponding to the selected unit to obtain the total remaining energy;

[0034] Step 9: According to the prediction order, select a predicted energy that is not marked as selected energy, compare the remaining energy sum with the selected predicted energy, if the remaining energy sum is greater than or equal to the predicted energy, mark the selected predicted energy as selected energy, and loop through steps 2 to 9 until all predicted energies are marked as selected energies and the loop ends; if the remaining energy sum is less than the selected predicted energy, remove the corresponding fitness from the first sorting table, and loop through steps 8 to 9;

[0035] Step 10: Determine whether the selection unit corresponding to the first prediction energy is in a working state, the first prediction energy being the first predicted prediction energy.

[0036] Furthermore, in step 2, the set label is encoded as X, where X is the chromosome and the range of X is ; Randomly generate G chromosomes to form the initial population ;

[0037] In step 3, the expression of the fitness function is: ;

[0038] In the formula, is the fitness corresponding to the i-th chromosome, is the discharge efficiency of the qth energy storage technology unit, is the energy storage efficiency of the qth energy storage technology unit, , ;

[0039] In step 4, natural selection is carried out by combining the elite method and the rotation method; wherein the elite method produces For a population with a capacity of G, the fitness of the G chromosomes is arranged from large to small, and the one at the front is Each chromosome produces one offspring chromosome; the rotation method produces offspring chromosomes, that is, G chromosomes are generated according to the corresponding rotation probability offspring chromosomes; ;

[0040] The expression of rotation probability is: ;

[0041] In the formula, is the rotation probability corresponding to the i-th chromosome.

[0042] Furthermore, in step 5, U chromosomes are randomly selected from the population for crossover recombination to obtain U new chromosomes; the crossover recombination adopts the PMX method; after the chromosome crossover recombination, the fitness of the U new chromosomes is calculated, the fitness of the U new chromosomes and the fitness of the U chromosomes are sorted from large to small and a second sorting table is generated, and the U chromosomes in the second sorting table are replaced with the U chromosomes that are crossover recombined in the population in positive order;

[0043] In step 6, the mutation probability is preset to H, and the G chromosomes in the population are mutated according to the mutation probability. The mutation method is to randomly select the positions of two genes in the chromosome and exchange the values ​​of the two genes.

[0044] Further, in step 10, from the remaining From the selected units corresponding to the prediction energies, select the selected units adjacent to each selected unit corresponding to the first prediction energy; calculate the output energy of each selected unit corresponding to each prediction energy, and take the same selected units as a group of analysis sets; add the output energy of each group of analysis sets to obtain the sum of adjacent energies of each group of analysis sets; then calculate the storage energy corresponding to each group of analysis sets, add the sum of the remaining energy to the storage energy and then subtract the sum of the adjacent energies to obtain the residual energy of each group of analysis sets; if the residual energy of each group of analysis sets is greater than or equal to 0, the selected unit corresponding to the first prediction energy is in a working state; if there is an analysis set with a residual energy less than 0, the selected unit corresponding to the residual energy less than 0 is marked as a deleted unit, and the fitness corresponding to the deleted unit is deleted from the first sorting table, and steps 8 to 10 are repeated, wherein only the corresponding selected unit is re-obtained for the first prediction energy;

[0045] The expression of output energy is: ; In the formula, is the output energy of the kth energy storage technology unit in the jth predicted energy, is the discharge efficiency of the kth energy storage technology unit in the jth predicted energy, is the j-th predicted energy, , , A is the number of energy storage technology units in each predicted energy;

[0046] The expression for stored energy is: ; In the formula, To store energy, To analyze the amount of predicted energy between the corresponding predicted energies of two selected units in the set, is the collection interval.

[0047] Further, the interconnection between the M types of energy storage technology units and the power grid includes power transmission and data transmission;

[0048] The grid capacity expansion is that M types of energy storage technology units transmit the stored electric energy to the grid, expanding the grid load capacity;

[0049] The methods of peak shaving and valley filling include:

[0050] A sampling frequency is preset, and multiple power grid loads are collected according to the sampling frequency, where the power grid load is the power load carried by the power grid; a load curve is drawn according to the collected multiple power grid loads, and the peak and valley values ​​of the power grid load are obtained according to the load curve;

[0051] When the grid load rises to a peak value, M types of energy storage technology units transmit electricity to the grid to reduce the peak value of the grid load; when the grid load drops to a valley value, the grid transmits electricity to M types of energy storage technology units to increase the valley value of the grid load;

[0052] Demand response is that M types of energy storage technology units adjust their working status according to the real-time dispatch instructions received from the power grid;

[0053] The methods of grid frequency regulation include:

[0054] When the frequency of the power grid system decreases, that is, the power supplied by the power grid is insufficient to meet the power grid load demand, the power grid dispatching center issues a discharge instruction to the M types of energy storage technology units; the M types of energy storage technology units transmit the stored power to the power grid according to the discharge instruction, and increase the frequency of the power grid system through the electromagnetic induction effect;

[0055] When the frequency of the power grid system increases, that is, the electric energy supplied by the power grid exceeds the power grid load demand, the power grid dispatching center issues a charging instruction; according to the charging instruction, the power grid transmits electric energy to M types of energy storage technology units to reduce the frequency of the power grid system.

[0056] Further, a suitable energy storage technology unit is selected from the M types of energy storage technology units to replace the deleted unit;

[0057] Mark the selected unit corresponding to the first predicted energy as the first unit, and mark the energy storage technology unit that is not the first unit among the M types of energy storage technology units as the first candidate unit; obtain the residual energy corresponding to each first candidate unit, and mark the candidate unit whose residual energy is greater than the output energy corresponding to the deleted unit as the second candidate unit; use the discharge efficiency and energy storage efficiency of each second candidate unit and the discharge efficiency and energy storage efficiency of the deleted unit as a group of analysis data; input each group of analysis data into the trained probability analysis model respectively, and obtain the similarity probability corresponding to each group of analysis data; preset a probability threshold, sort each similar probability and probability threshold from large to small, mark the second candidate unit corresponding to the similarity probability that is ranked before the probability threshold and ranks first as the positively selected unit, and use the positively selected unit to replace the deleted unit; if each similar probability is ranked after the probability threshold, delete the fitness corresponding to the deleted unit from the first sorting table, and loop steps 8 to 10, where only the corresponding selected unit is re-obtained for the first predicted energy;

[0058] The training process of the probability analysis model includes:

[0059] Collect a group of analysis data and similarity probabilities corresponding to the group of analysis data in advance, where a is an integer greater than 1;

[0060] Convert the analysis data and the corresponding similarity probabilities into a corresponding set of feature vectors;

[0061] Each group of feature vectors is used as the input of the probability analysis model. The probability analysis model takes a group of predicted similarity probabilities corresponding to each group of analysis data as output, and takes the actual similarity probability corresponding to each group of analysis data as the prediction target, where the actual similarity probability is the pre-set similarity probability corresponding to the analysis data; minimizing the sum of prediction errors of all analysis data is used as the training goal; the probability analysis model is trained until the sum of prediction errors reaches convergence and the training is stopped; the probability analysis model is a deep neural network model.

[0062] The centralized control chain energy storage grid auxiliary service system integrating multiple energy storage technologies implements the centralized control chain energy storage grid auxiliary service method integrating multiple energy storage technologies, including:

[0063] An energy storage technology unit configuration module, used to configure M different types of energy storage technology units, where M is an integer greater than 1;

[0064] Centralized control chain communication module, used to connect the communication and data transmission between M types of energy storage technology units;

[0065] An efficiency calculation module is used to collect energy storage characteristic data of M types of energy storage technology units and calculate the corresponding energy storage efficiency;

[0066] Energy collaborative management module, used to predict the energy demand value of the power grid and to collaboratively manage M types of energy storage technology units in combination with their energy storage efficiency;

[0067] The grid auxiliary service module is used to interconnect M types of energy storage technology units with the grid and provide grid auxiliary services; grid auxiliary services include grid capacity expansion, peak shaving and valley filling, demand response and grid frequency regulation.

[0068] The technical effects and advantages of the centralized control chain energy storage grid auxiliary service system and method integrating multiple energy storage technologies of the present invention are as follows:

[0069] 1. A centralized control chain is used to realize communication and data transmission among various energy storage technology units. A neural network model is used to predict the energy demand value of the power grid, and an energy storage efficiency calculation mechanism is established. With the help of genetic algorithms, coordinated scheduling of various energy storage technology units is carried out according to the energy storage efficiency and discharge efficiency of different energy storage technology units, so that after each energy storage technology unit completes a work and stores energy, its remaining energy can still complete the next work; in addition, it also provides a number of power grid auxiliary services such as capacity expansion, peak shaving and valley filling, demand response and frequency modulation; unified management of various energy storage technology units and energy configuration optimization are realized to improve the energy utilization efficiency of energy storage technology units, thereby avoiding the situation of insufficient energy storage of energy storage technology units with low storage efficiency in the power grid auxiliary process, resulting in temporary energy shortage, and effectively improving the reliability and operation efficiency of the power grid.

[0070] 2. Through the deep neural network model, the optimal candidate unit is selected for replacing the deleted unit based on the parameter comparison of each candidate unit. The deep neural network model can also learn complex feature relationships to improve judgment accuracy; and set a reasonable probability threshold to clarify the replacement standard to solve the problems of large computational complexity and poor convergence; improve the collaborative management efficiency of energy storage technology units and the operating efficiency of the power grid, enhance the auxiliary service capabilities of the power grid, and have a positive impact on the flexibility and reliability of the power grid. BRIEF DESCRIPTION OF THE DRAWINGS

[0071] Figure 1 This is a schematic diagram of a centralized control chain energy storage grid auxiliary service system integrating multiple energy storage technologies according to Example 1 of the present invention;

[0072] Figure 2 A schematic diagram of a centralized control chain energy storage grid auxiliary service system integrating multiple energy storage technologies according to Embodiment 2 of the present invention;

[0073] Figure 3 This is a flow chart of a centralized control chain energy storage grid auxiliary service method integrating multiple energy storage technologies according to Embodiment 3 of the present invention;

[0074] Figure 4 This is a schematic diagram of an electronic device according to Embodiment 4 of the present invention;

[0075] Figure 5 This is a schematic diagram of a storage medium according to Embodiment 5 of the present invention. DETAILED DESCRIPTION

[0076] 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 only 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 ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0077] Example 1

[0078] See also Figure 1 As shown, the centralized control chain energy storage grid auxiliary service system integrating multiple energy storage technologies described in this embodiment includes an energy storage technology unit configuration module, a centralized control chain communication module, an efficiency calculation module, an energy collaborative management module and a grid auxiliary service module; each module is connected by wire and / or wireless means to realize data transmission between modules;

[0079] The energy storage technology unit configuration module is used to configure M different types of energy storage technology units, where M is an integer greater than 1; energy storage technology units include battery energy storage, supercapacitor energy storage, compressed air energy storage, etc., each of which can independently store and release electrical energy.

[0080] The centralized control chain communication module is used to connect the communication and data transmission between M types of energy storage technology units to achieve centralized control and collaborative management; among them, the M types of energy storage technology units realize communication and data transmission through a communication protocol stack, which includes communication standards (such as Zigbee, WiFi, LoRa), data transmission protocols (such as TCP, IP) and data formats (such as JSON, XML);

[0081] The efficiency calculation module is used to collect energy storage characteristic data of M types of energy storage technology units and calculate the corresponding energy storage efficiency.

[0082] Energy storage characteristic data include charging time, charging efficiency, discharge efficiency, self-discharge efficiency, number of cycles and total number of cycles;

[0083] The charging time is the time required to fully charge the energy storage technology unit from a fully discharged state. The longer the charging time, the lower the energy storage efficiency of the energy storage technology unit, and vice versa. The charging efficiency is the energy conversion efficiency of the energy storage technology unit during the charging process. The lower the charging efficiency, the lower the energy storage efficiency of the energy storage technology unit, and vice versa. The discharge efficiency is the energy conversion efficiency of the energy storage technology unit during the discharge process. The self-discharge efficiency is the natural energy loss rate of the energy storage technology unit when it is not in use. The lower the self-discharge efficiency, the lower the energy storage efficiency of the energy storage technology unit, and vice versa. The number of cycles is the number of complete charge and discharge cycles experienced by the energy storage technology unit. The more cycles, the greater the degree of performance degradation of the energy storage technology unit, and therefore the lower the energy storage efficiency of the energy storage unit, and vice versa. The total number of cycles is the maximum number of charge and discharge cycles that the energy storage technology unit can experience during its life cycle. Among them, the charging time, charging efficiency, discharge efficiency, self-discharge efficiency and total number of cycles are all obtained from performance data provided by the manufacturer of the energy storage technology unit; the cycle life is obtained by the power analyzer installed in each energy storage technology unit.

[0084] The calculation methods of energy storage efficiency include:

[0085] ;

[0086] In the formula, is the energy storage efficiency, is the total number of cycles, is the number of cycles, For charging time, For charging efficiency, is the self-discharge efficiency, , , is the preset weight coefficient.

[0087] The specific value of the weight coefficient in the formula can be set according to actual conditions. The weight coefficient reflects the degree of influence of each energy storage characteristic data on the energy storage efficiency. Technical personnel in this field can preset the corresponding weight coefficient according to the actual degree of influence of each energy storage characteristic data on the energy storage efficiency, so as to accurately evaluate the energy storage efficiency of the energy storage technology unit; since the energy storage efficiency is only used to represent the energy storage speed of the energy storage technology unit, the calculation process of the energy storage efficiency is a dimensionless calculation.

[0088] The energy collaborative management module is used to predict the energy demand value of the power grid and to collaboratively manage the M types of energy storage technology units in combination with the energy storage efficiency of the M types of energy storage technology units.

[0089] Methods for predicting grid energy demand values ​​include:

[0090] A collection interval is preset, and N power grid energy demand values ​​are continuously collected according to the collection interval, where N is an integer greater than 1. The continuously collected N power grid energy demand values ​​are input into an energy prediction model to predict the power grid energy demand values ​​at n future moments, where n is an integer greater than 1. The predicted power grid energy demand values ​​are marked as predicted energy. The power grid energy demand value is the energy required for the power grid to work, and the power grid energy demand value is acquired by the power grid system. The collection interval of the N power grid energy demand values ​​is set by technical personnel in this field according to the frequency of changes in the power grid energy demand value during the historical operation of the power grid system.

[0091] The training method of the energy prediction model includes:

[0092] Pre-collection continuous grid energy demand values ​​and construct a grid energy demand value set, , based on the set of grid energy demand values, training a prediction model for predicting grid energy demand values ​​at future moments;

[0093] The sliding step size L and the sliding window length are preset according to the actual experience of those skilled in the art; the power grid energy demand value in the power grid energy demand value set is converted into multiple training samples using the sliding window method, the training samples are used as the input of the recurrent neural network model, the power grid energy demand value after the sliding step size L is predicted as the output, the subsequent power grid energy demand value of each training sample is used as the prediction target, the prediction accuracy is used as the training target, and the prediction result is evaluated for model accuracy using the mean absolute percentage error MAPE. When the calculated MAPE is less than the preset MAPE, the energy prediction model training is completed; wherein, the calculation formula of the mean absolute percentage error MAPE is: ,in, is the prediction target corresponding to the d-th predicted grid energy demand value, is the dth predicted grid energy demand value, H is the number of predicted grid energy demand values; an energy prediction model is generated based on the grid energy demand value to predict the grid energy demand value at future moments; wherein the energy prediction model is an RNN neural network model.

[0094] For example, assuming that the grid energy demand value set Contains 10 sets of grid energy demand values, , is the energy demand value of the e-th grid, , use sliding windows to construct multiple training samples, define the length of the sliding window as 3, the sliding step L as 1, each training sample contains 3 consecutive grid energy demand values, and take the next grid energy demand value among the 3 consecutive grid energy demand values ​​as the prediction target; for example:

[0095] As training data, The corresponding prediction target is ;

[0096] As training data, The corresponding prediction target is ; and so on, for energy prediction models.

[0097] The steps for collaborative management of M types of energy storage technology units include:

[0098] Step 1: Randomly combine M energy storage technology units to form a set of R units. ; Each unit set includes h types of energy storage technology units, , and each unit set is different; set digital labels for R unit sets, and the range of digital labels is , mark the numerical labels of the unit set as the set label;

[0099] Step 2: Encode the set labels, obtain chromosomes, and construct the initial population;

[0100] Step 3: Determine the fitness function;

[0101] Step 4: Natural selection of chromosomes in the population;

[0102] Step 5: Perform crossover recombination on chromosomes in the population;

[0103] Step 6: mutate the chromosomes in the population;

[0104] Step 7: Obtain a new population, preset the population generation number to be F, the fitness threshold to be S, where F is an integer greater than 0, and S is a real number greater than 0; loop through steps 4 to 6 until the generation number corresponding to the new population is F or the fitness corresponding to a chromosome in the new population is greater than or equal to the fitness threshold S, the loop ends, sorts the fitness corresponding to each chromosome in the new population, and generates a first sorting table; illustratively, if the preset population generation number is 1, the chromosomes in the initial population are subjected to natural selection, crossover recombination, and mutation to obtain a new population, at which point the generation number corresponding to the new population is 1, and thus the loop ends;

[0105] Step 8: Obtain the set label corresponding to the corresponding chromosome according to the positive order of the first sorting table, obtain the corresponding unit set according to the set label, obtain the corresponding energy storage technology unit according to the unit set, and mark it as a selected unit; obtain the remaining energy of the selected unit in real time, and the remaining energy is obtained by the built-in metering system of the energy storage technology unit; add the remaining energy corresponding to the selected unit to obtain the total remaining energy;

[0106] Step 9: According to the prediction order, select a predicted energy that is not marked as selected energy, compare the remaining energy sum with the selected predicted energy, if the remaining energy sum is greater than or equal to the predicted energy, mark the selected predicted energy as selected energy, and loop through steps 2 to 9 until all predicted energies are marked as selected energies and the loop ends; if the remaining energy sum is less than the selected predicted energy, remove the corresponding fitness from the first sorting table, and loop through steps 8 to 9;

[0107] Step 10: From the rest From the selected units corresponding to the predicted energies, select the selected units adjacent to each selected unit corresponding to the first predicted energy, wherein the first predicted energy is the first predicted energy; exemplarily, the first predicted energy corresponds to battery energy storage and supercapacitor energy storage, the second predicted predicted energy corresponds to battery energy storage and compressed air energy storage, and the third predicted predicted energy corresponds to supercapacitor energy storage and compressed air energy storage, thus selecting the battery energy storage in the second predicted predicted energy and the supercapacitor energy storage in the third predicted predicted energy; calculating the output energy of each selected unit corresponding to each predicted energy, and taking the same selected units as a group of analysis sets; adding the output energy of each group of analysis sets to obtain the sum of adjacent energies of each group of analysis sets; then calculating the storage energy corresponding to each group of analysis sets, adding the remaining energy sum to the storage energy and then subtracting the adjacent The residual energy of each analysis set is obtained by summing up the energy. If the residual energy of each analysis set is greater than or equal to 0, the selected unit corresponding to the first predicted energy is in working state to realize demand-driven intelligent use of energy storage, indicating that the selected unit corresponding to the first predicted energy can complete the next work after the current work is completed, and there will be no energy exhaustion. If the residual energy of the analysis set is less than 0, the selected unit corresponding to the residual energy less than 0 is marked as a deleted unit, and the fitness corresponding to the deleted unit is deleted from the first sorting table, and steps 8 to 10 are looped, wherein only the corresponding selected unit is re-obtained for the first predicted energy, indicating that after the current work is completed, the selected unit corresponding to the first predicted energy can store energy, but the remaining energy cannot complete the next work, and it is necessary to reselect a suitable selected unit to avoid energy shortage.

[0108] It should be noted that the fitness threshold S is preset by technical personnel in this field according to the algorithm accuracy, and the population algebra F is obtained by technical personnel in this field under multiple different prediction energy conditions by using genetic algorithms multiple times to obtain the corresponding set labels. In each process of using the genetic algorithm, when the fitness corresponding to the chromosome in the new population is greater than or equal to the fitness threshold S, the loop ends and the algebra corresponding to the new population is obtained; the largest algebra among the multiple algebras is taken as the population algebra F.

[0109] In step 2 above, the set label is encoded as X, where X is the chromosome and the range of X is ; Randomly generate G chromosomes to form the initial population .

[0110] In the above step 3, the expression of the fitness function is: ;

[0111] In the formula, is the fitness corresponding to the i-th chromosome, is the discharge efficiency of the qth energy storage technology unit, is the energy storage efficiency of the qth energy storage technology unit, , .

[0112] In step 4 above, natural selection is carried out by combining the elite method and the rotation method; the elite method produces For a population with a capacity of G, the fitness of the G chromosomes is arranged from large to small, and the one at the front is Each chromosome produces one offspring chromosome; the rotation method produces offspring chromosomes, that is, G chromosomes are generated according to the corresponding rotation probability offspring chromosomes; , in order to keep the offspring population capacity G unchanged and the population increases algebraically;

[0113] The expression of rotation probability is: ;

[0114] In the formula, is the rotation probability corresponding to the i-th chromosome.

[0115] In the above step 5, U chromosomes are randomly selected from the population for crossover recombination to obtain U new chromosomes; the crossover recombination adopts the PMX method, which is a prior art method and will not be described in detail here; after the chromosome crossover recombination, the fitness of the U new chromosomes is calculated, and the fitness of the U new chromosomes and the fitness of the U chromosomes are sorted from large to small to generate a second sorting table, and the U chromosomes in the second sorting table are replaced in positive order with the U chromosomes in the population for crossover recombination; this embodiment is preferred If the calculated U is not an integer, U is rounded up to ensure that the calculated U is an integer.

[0116] In the above step 6, the mutation probability is preset to H, and the G chromosomes in the population are mutated according to the mutation probability. The mutation method is to randomly select the positions of two genes in the chromosome and exchange the values ​​of the two genes. In this embodiment, H is preferably 0.02; the mutation probability is preset by technical personnel in this field according to the algorithm efficiency and algorithm accuracy.

[0117] In the above step 10, the expression of output energy is: ; In the formula, is the output energy of the kth energy storage technology unit in the jth predicted energy, is the discharge efficiency of the kth energy storage technology unit in the jth predicted energy, is the j-th predicted energy, , , A is the number of energy storage technology units in each predicted energy;

[0118] The expression for stored energy is: ; In the formula, To store energy, To analyze the amount of predicted energy between the corresponding predicted energies of two selected units in the set, is the collection interval; for example, the prediction energies corresponding to two selected units in the analysis set are the first predicted prediction energy and the second predicted prediction energy, respectively, then the number of prediction energies between the corresponding prediction energies of the two selected units in the analysis set is 0, that is, there is no other prediction energy.

[0119] The grid auxiliary service module is used to interconnect M types of energy storage technology units with the grid and provide grid auxiliary services.

[0120] The interconnection between M types of energy storage technology units and the power grid includes power transmission and data transmission; M types of energy storage technology units realize power transmission with the power grid through PCS (power conversion system) inverter equipment; M types of energy storage technology units realize data transmission with the power grid through wireless communication networks (such as 4G / 5G).

[0121] Grid ancillary services include grid expansion, peak load shifting, demand response and grid frequency regulation;

[0122] The grid capacity expansion is to use M types of energy storage technology units to transmit the stored electric energy to the grid, expand the overall load capacity of the grid, and alleviate the peak load of the grid.

[0123] The methods of peak shaving and valley filling include:

[0124] A sampling frequency is preset, and multiple grid loads are collected according to the sampling frequency, where the grid load is the power load carried by the grid; a load curve is drawn according to the collected multiple grid loads, and the peak and valley values ​​of the grid load are obtained according to the load curve; the grid load is obtained by a power sensor installed on the grid side; the sampling frequency is preset by a technician in this field according to actual experience;

[0125] When the grid load rises to a peak value, M types of energy storage technology units transmit electricity to the grid to reduce the peak value of the grid load; when the grid load drops to a valley value, the grid transmits electricity to M types of energy storage technology units to increase the valley value of the grid load; thereby alleviating the rate of change of the grid load curve from the valley value to the peak value, allowing the load curve to transition more smoothly from the valley value to the peak value.

[0126] Demand response is that M types of energy storage technology units adjust their working status according to the real-time dispatch instructions received from the power grid;

[0127] Scheduling instructions such as adjusting the charging rate, delaying the charging time period, stopping charging, etc.;

[0128] Adjust the working status, such as reducing the charging rate, delaying the start time of charging, stopping charging, etc.

[0129] The methods of grid frequency regulation include:

[0130] When the frequency of the power grid system decreases, that is, the power supplied by the grid is insufficient to meet the grid load demand, the grid dispatching center issues a discharge instruction to the M types of energy storage technology units; the M types of energy storage technology units transmit the stored power to the grid according to the discharge instruction, so that the power supplied by the grid can meet the grid load demand and increase the frequency of the grid system through the electromagnetic induction effect;

[0131] When the frequency of the power grid system increases, that is, the power supplied by the power grid exceeds the power grid load demand, the power grid dispatching center issues a charging instruction; according to the charging instruction, the power grid transmits the power to M types of energy storage technology units, so that the power supplied by the power grid is close to the power grid load demand, thereby reducing the frequency of the power grid system;

[0132] M types of energy storage technology units can work together with the power grid to adjust the frequency of the power grid system, thereby ensuring the stable operation of the power grid system frequency.

[0133] This embodiment adopts a centralized control chain mode to realize communication and data transmission among various energy storage technology units, applies a neural network model to predict the energy demand value of the power grid, and establishes an energy storage efficiency calculation mechanism. With the help of a genetic algorithm, the energy storage technology units are coordinated and dispatched according to the energy storage efficiency and discharge efficiency of different energy storage technology units, so that after each energy storage technology unit completes a work and stores energy, its remaining energy can still complete the next work; in addition, it also provides multiple power grid auxiliary services such as capacity expansion, peak shaving and valley filling, demand response and frequency modulation; realizes unified management of various energy storage technology units and energy configuration optimization to improve the energy utilization efficiency of energy storage technology units, thereby avoiding the situation of insufficient energy storage of energy storage technology units with low storage efficiency in the power grid auxiliary process, causing temporary energy shortage, and effectively improving the reliability and operation efficiency of the power grid.

[0134] Example 2

[0135] See also Figure 2 As shown, this embodiment further improves the design on the basis of embodiment 1. In embodiment 1, when there is an analysis set with residual energy less than 0 in the analysis set corresponding to the first predicted energy, re-acquiring the selected unit corresponding to the first predicted energy will increase the amount of calculation, thereby reducing the collaborative management efficiency of the energy storage technology unit; and re-acquiring the selected unit may also result in a situation where the residual energy is less than 0, and the convergence is not strong, thereby reducing the operation efficiency of the power grid; therefore, this embodiment provides a centralized control chain energy storage power grid auxiliary service system integrating multiple energy storage technologies, and also includes a unit replacement module;

[0136] The unit replacement module is used to select suitable energy storage technology units from M types of energy storage technology units to replace the deleted units.

[0137] Mark the selected unit corresponding to the first predicted energy as the first unit, and mark the energy storage technology unit that is not the first unit among the M energy storage technology units as the first candidate unit; obtain the residual energy corresponding to each first candidate unit, and mark the candidate unit whose residual energy is greater than the output energy corresponding to the deleted unit as the second candidate unit; take the discharge efficiency and energy storage efficiency of each second candidate unit and the discharge efficiency and energy storage efficiency of the deleted unit as a group of analysis data; input each group of analysis data into the trained probability analysis model respectively, and obtain the similarity probability corresponding to each group of analysis data; preset a probability threshold, sort each similarity probability and probability threshold from large to small, and sort the similarity probability and probability threshold that are ranked in front of the probability threshold and ranked first The second candidate unit corresponding to the similar probability is marked as the positively selected unit, and the positively selected unit is used to replace the deleted unit; if each similar probability is ranked behind the probability threshold, the fitness corresponding to the deleted unit is deleted from the first sorting table, and steps 8 to 10 are looped, wherein only the corresponding selected unit is re-acquired for the first predicted energy; the probability threshold is determined by technicians in the historical energy storage technology unit collaborative management process, collecting multiple positively selected units that successfully provide grid auxiliary services after replacing the deleted units, and obtaining the corresponding discharge efficiency and energy storage efficiency, analyzing the similarity probabilities between multiple positively selected units and the corresponding deleted units, and taking the average of multiple similar probabilities as the probability threshold. The preferred probability threshold in this embodiment is 96%.

[0138] The specific training process of the probability analysis model includes:

[0139] A group of analysis data and similarity probabilities corresponding to the group of analysis data are collected in advance, where a is an integer greater than 1. The similarity probabilities corresponding to the analysis data are collected by a person skilled in the art during the collaborative management of historical energy storage technology units. A group of different analysis data are collected. The person skilled in the art sequentially determines the similarity probabilities between the second candidate unit and the deleted unit under the conditions of the group of different analysis data based on actual experience, and sequentially sets corresponding similarity probabilities for the group of different analysis data;

[0140] Convert the analysis data and the corresponding similarity probabilities into a corresponding set of feature vectors;

[0141] Each set of feature vectors is used as the input of the probability analysis model. The probability analysis model takes a set of predicted similarity probabilities corresponding to each set of analysis data as output, and takes the actual similarity probability corresponding to each set of analysis data as the prediction target. The actual similarity probability is the pre-set similarity probability corresponding to the analysis data. The training goal is to minimize the sum of the prediction errors of all analysis data. The calculation formula of the prediction error is: ,in is the prediction error, p is the group number of the eigenvector corresponding to the analysis data, is the similarity probability corresponding to the p-th group of analysis data, is the actual similarity probability corresponding to the pth group of analyzed data; the probability analysis model is trained until the sum of the prediction errors reaches convergence and the training is stopped; the probability analysis model is specifically a deep neural network model, which includes an input layer, a hidden layer and an output layer; each hidden layer includes multiple neurons, each neuron is connected to the next layer of neurons, and the connection contains weights, which determine the importance and influence of data transmission in the neural network; an activation function is applied to each neuron between the hidden layer and the output layer, and the activation function is mapped into nonlinearity, allowing the network to learn more complex patterns and features.

[0142] This embodiment uses a deep neural network model to select the best candidate unit for replacing the deleted unit based on parameter comparison of each candidate unit. The deep neural network model can also learn complex feature relationships to improve judgment accuracy; and set a reasonable probability threshold to clarify the replacement standard to solve the problems of large computational complexity and poor convergence; improve the collaborative management efficiency of energy storage technology units and the operating efficiency of the power grid, enhance the auxiliary service capabilities of the power grid, and have a positive impact on the flexibility and reliability of the power grid.

[0143] Example 3

[0144] See also Figure 3 As shown, for the parts not described in detail in this embodiment, please refer to the description of Embodiment 1 and Embodiment 2, and provide a centralized control chain energy storage grid auxiliary service method integrating multiple energy storage technologies, the method includes:

[0145] Configure M different types of energy storage technology units, where M is an integer greater than 1;

[0146] Connecting M types of energy storage technology units for communication and data transmission;

[0147] Collect energy storage characteristic data of M types of energy storage technology units and calculate the corresponding energy storage efficiency;

[0148] Predict the energy demand value of the power grid, and combine the energy storage efficiency of M energy storage technology units to coordinate the management of M energy storage technology units;

[0149] M types of energy storage technology units are interconnected with the power grid to provide grid auxiliary services; grid auxiliary services include grid capacity expansion, peak shaving and valley filling, demand response and grid frequency regulation.

[0150] Furthermore, the M types of energy storage technology units realize communication and data transmission through a communication protocol stack, and the communication protocol stack includes communication standards, data transmission protocols and data formats.

[0151] The energy storage characteristic data include charging time, charging efficiency, discharge efficiency, self-discharge efficiency, cycle number and total cycle number;

[0152] The calculation methods of energy storage efficiency include:

[0153] ;

[0154] In the formula, is the energy storage efficiency, is the total number of cycles, is the number of cycles, For charging time, For charging efficiency, is the self-discharge efficiency, , , is the preset weight coefficient.

[0155] Furthermore, the method for predicting the power grid energy demand value includes:

[0156] A collection interval is preset, and N power grid energy demand values ​​are continuously collected according to the collection interval, where N is an integer greater than 1, and the continuously collected N power grid energy demand values ​​are input into the energy prediction model to predict the power grid energy demand values ​​at n future moments, where n is an integer greater than 1, and the predicted power grid energy demand values ​​are marked as predicted energy;

[0157] The training method of the energy prediction model includes:

[0158] Pre-collection continuous grid energy demand values ​​and construct a grid energy demand value set, , based on the set of grid energy demand values, training a prediction model for predicting grid energy demand values ​​at future moments;

[0159] Preset a sliding step size L and a sliding window length; use a sliding window method to convert the grid energy demand values ​​in the grid energy demand value set into multiple training samples, use the training samples as the input of the recurrent neural network model, predict the grid energy demand value after the sliding step size L as the output, use the subsequent grid energy demand value of each training sample as the prediction target, use the prediction accuracy as the training target, use the mean absolute percentage error MAPE to evaluate the model accuracy of the prediction result, and when the calculated MAPE is less than the preset MAPE, the energy prediction model training is completed; generate an energy prediction model that predicts the grid energy demand value at a future moment according to the grid energy demand value; wherein the energy prediction model is an RNN neural network model.

[0160] Furthermore, the step of collaboratively managing the M types of energy storage technology units includes:

[0161] Step 1: Randomly combine M energy storage technology units to form a set of R units. ; Each unit set includes h types of energy storage technology units, , and each unit set is different; set digital labels for R unit sets, and the range of digital labels is , mark the numerical labels of the unit set as the set label;

[0162] Step 2: Encode the set labels, obtain chromosomes, and construct the initial population;

[0163] Step 3: Determine the fitness function;

[0164] Step 4: Natural selection of chromosomes in the population;

[0165] Step 5: Perform crossover recombination on chromosomes in the population;

[0166] Step 6: mutate the chromosomes in the population;

[0167] Step 7: Get a new population, preset the population generation number as F, the fitness threshold as S, where F is an integer greater than 0 and S is a real number greater than 0; loop steps 4 to 6 until the generation number corresponding to the new population is F or the fitness corresponding to the chromosome in the new population is greater than or equal to the fitness threshold S, the loop ends, sort the fitness corresponding to each chromosome in the new population, and generate a first sorting table;

[0168] Step 8: Obtain the set label corresponding to the corresponding chromosome according to the positive order of the first sorting table, obtain the corresponding unit set according to the set label, obtain the corresponding energy storage technology unit according to the unit set, and mark it as a selected unit; obtain the remaining energy of the selected unit in real time, and the remaining energy is obtained by the built-in metering system of the energy storage technology unit; add the remaining energy corresponding to the selected unit to obtain the total remaining energy;

[0169] Step 9: According to the prediction order, select a predicted energy that is not marked as selected energy, compare the remaining energy sum with the selected predicted energy, if the remaining energy sum is greater than or equal to the predicted energy, mark the selected predicted energy as selected energy, and loop through steps 2 to 9 until all predicted energies are marked as selected energies and the loop ends; if the remaining energy sum is less than the selected predicted energy, remove the corresponding fitness from the first sorting table, and loop through steps 8 to 9;

[0170] Step 10: Determine whether the selection unit corresponding to the first prediction energy is in a working state, the first prediction energy being the first predicted prediction energy.

[0171] Furthermore, in step 2, the set label is encoded as X, where X is the chromosome and the range of X is ; Randomly generate G chromosomes to form the initial population ;

[0172] In step 3, the expression of the fitness function is: ;

[0173] In the formula, is the fitness corresponding to the i-th chromosome, is the discharge efficiency of the qth energy storage technology unit, is the energy storage efficiency of the qth energy storage technology unit, , ;

[0174] In step 4, natural selection is carried out by combining the elite method and the rotation method; wherein the elite method produces For a population with a capacity of G, the fitness of the G chromosomes is arranged from large to small, and the one at the front is Each chromosome produces one offspring chromosome; the rotation method produces offspring chromosomes, that is, G chromosomes are generated according to the corresponding rotation probability offspring chromosomes; ;

[0175] The expression of rotation probability is: ;

[0176] In the formula, is the rotation probability corresponding to the i-th chromosome.

[0177] Furthermore, in step 5, U chromosomes are randomly selected from the population for crossover recombination to obtain U new chromosomes; the crossover recombination adopts the PMX method; after the chromosome crossover recombination, the fitness of the U new chromosomes is calculated, the fitness of the U new chromosomes and the fitness of the U chromosomes are sorted from large to small and a second sorting table is generated, and the U chromosomes in the second sorting table are replaced with the U chromosomes that are crossover recombined in the population in positive order;

[0178] In step 6, the mutation probability is preset to H, and the G chromosomes in the population are mutated according to the mutation probability. The mutation method is to randomly select the positions of two genes in the chromosome and exchange the values ​​of the two genes.

[0179] Further, in step 10, from the remaining From the selected units corresponding to the prediction energies, select the selected units adjacent to each selected unit corresponding to the first prediction energy; calculate the output energy of each selected unit corresponding to each prediction energy, and take the same selected units as a group of analysis sets; add the output energy of each group of analysis sets to obtain the sum of adjacent energies of each group of analysis sets; then calculate the storage energy corresponding to each group of analysis sets, add the sum of the remaining energy to the storage energy and then subtract the sum of the adjacent energies to obtain the residual energy of each group of analysis sets; if the residual energy of each group of analysis sets is greater than or equal to 0, the selected unit corresponding to the first prediction energy is in a working state; if there is an analysis set with a residual energy less than 0, the selected unit corresponding to the residual energy less than 0 is marked as a deleted unit, and the fitness corresponding to the deleted unit is deleted from the first sorting table, and steps 8 to 10 are repeated, wherein only the corresponding selected unit is re-obtained for the first prediction energy;

[0180] The expression of output energy is: ; In the formula, is the output energy of the kth energy storage technology unit in the jth predicted energy, is the discharge efficiency of the kth energy storage technology unit in the jth predicted energy, is the j-th predicted energy, , , A is the number of energy storage technology units in each predicted energy;

[0181] The expression for stored energy is: ; In the formula, To store energy, To analyze the amount of predicted energy between the corresponding predicted energies of two selected units in the set, is the collection interval.

[0182] Further, the interconnection between the M types of energy storage technology units and the power grid includes power transmission and data transmission;

[0183] The grid capacity expansion is that M types of energy storage technology units transmit the stored electric energy to the grid, expanding the grid load capacity;

[0184] The methods of peak shaving and valley filling include:

[0185] A sampling frequency is preset, and multiple power grid loads are collected according to the sampling frequency, where the power grid load is the power load carried by the power grid; a load curve is drawn according to the collected multiple power grid loads, and the peak and valley values ​​of the power grid load are obtained according to the load curve;

[0186] When the grid load rises to a peak value, M types of energy storage technology units transmit electricity to the grid to reduce the peak value of the grid load; when the grid load drops to a valley value, the grid transmits electricity to M types of energy storage technology units to increase the valley value of the grid load;

[0187] Demand response is that M types of energy storage technology units adjust their working status according to the real-time dispatch instructions received from the power grid;

[0188] The methods of grid frequency regulation include:

[0189] When the frequency of the power grid system decreases, that is, the power supplied by the power grid is insufficient to meet the power grid load demand, the power grid dispatching center issues a discharge instruction to the M types of energy storage technology units; the M types of energy storage technology units transmit the stored power to the power grid according to the discharge instruction, and increase the frequency of the power grid system through the electromagnetic induction effect;

[0190] When the frequency of the power grid system increases, that is, the electric energy supplied by the power grid exceeds the power grid load demand, the power grid dispatching center issues a charging instruction; according to the charging instruction, the power grid transmits electric energy to M types of energy storage technology units to reduce the frequency of the power grid system.

[0191] Further, a suitable energy storage technology unit is selected from the M types of energy storage technology units to replace the deleted unit;

[0192] Mark the selected unit corresponding to the first predicted energy as the first unit, and mark the energy storage technology unit that is not the first unit among the M types of energy storage technology units as the first candidate unit; obtain the residual energy corresponding to each first candidate unit, and mark the candidate unit whose residual energy is greater than the output energy corresponding to the deleted unit as the second candidate unit; use the discharge efficiency and energy storage efficiency of each second candidate unit and the discharge efficiency and energy storage efficiency of the deleted unit as a group of analysis data; input each group of analysis data into the trained probability analysis model respectively, and obtain the similarity probability corresponding to each group of analysis data; preset a probability threshold, sort each similar probability and probability threshold from large to small, mark the second candidate unit corresponding to the similarity probability that is ranked before the probability threshold and ranks first as the positively selected unit, and use the positively selected unit to replace the deleted unit; if each similar probability is ranked after the probability threshold, delete the fitness corresponding to the deleted unit from the first sorting table, and loop steps 8 to 10, where only the corresponding selected unit is re-obtained for the first predicted energy;

[0193] The training process of the probability analysis model includes:

[0194] Collect a group of analysis data and similarity probabilities corresponding to the group of analysis data in advance, where a is an integer greater than 1;

[0195] Convert the analysis data and the corresponding similarity probabilities into a corresponding set of feature vectors;

[0196] Each group of feature vectors is used as the input of the probability analysis model. The probability analysis model takes a group of predicted similarity probabilities corresponding to each group of analysis data as output, and takes the actual similarity probability corresponding to each group of analysis data as the prediction target, where the actual similarity probability is the pre-set similarity probability corresponding to the analysis data; minimizing the sum of prediction errors of all analysis data is used as the training goal; the probability analysis model is trained until the sum of prediction errors reaches convergence and the training is stopped; the probability analysis model is a deep neural network model.

[0197] Example 4

[0198] See also Figure 4 As shown, according to another aspect of the present application, an electronic device 500 is also provided. The electronic device 500 may include one or more processors and one or more memories. The memory stores a computer-readable code, and when the computer-readable code is executed by one or more processors, the centralized control chain energy storage grid auxiliary service method integrating multiple energy storage technologies as described above can be executed.

[0199] The method or system according to the embodiment of the present application can also be used by Figure 4 The electronic device architecture shown in FIG. Figure 4As shown, the electronic device 500 may include a bus 501, one or more CPUs 502, ROM 503, RAM 504, a communication port 505 connected to a network, an input / output 506, a hard disk 507, etc. The storage device in the electronic device 500, such as ROM 503 or hard disk 507, may store the centralized control chain energy storage grid auxiliary service method for integrating multiple energy storage technologies provided in this application. Furthermore, the electronic device 500 may also include a user interface 508. Of course, Figure 4 The architecture shown is only exemplary and can be omitted according to actual needs when implementing different devices. Figure 4 One or more components of an electronic device are shown.

[0200] Example 5

[0201] See also Figure 5 As shown, a computer-readable storage medium 600 according to an embodiment of the present application is shown. Computer-readable instructions are stored on the computer-readable storage medium 600. When the computer-readable instructions are executed by the processor, the centralized control chain energy storage grid auxiliary service method for integrating multiple energy storage technologies according to the embodiment of the present application described with reference to the above figures can be executed. The storage medium 600 includes, but is not limited to, for example, volatile memory and / or non-volatile memory. Volatile memory may include, for example, random access memory (RAM) and cache memory (cache). Non-volatile memory may include, for example, read-only memory (ROM), hard disk, flash memory, etc.

[0202] In addition, according to the implementation of the present application, the process described above with reference to the flowchart can be implemented as a computer software program. For example, the present application provides a non-transitory machine-readable storage medium, which stores machine-readable instructions, and the machine-readable instructions can be executed by a processor to execute instructions corresponding to the method steps provided in the present application, for example: a centralized control chain energy storage grid auxiliary service method integrating multiple energy storage technologies. When the computer program is executed by a central processing unit (CPU), the above functions defined in the method of the present application are executed.

[0203] The above is only a specific embodiment of the present invention, but the protection scope of the present invention is not limited thereto. Any person skilled in the art can easily think of changes or substitutions within the technical scope disclosed by the present invention, which should be included in the protection scope of the present invention. Therefore, the protection scope of the present invention should be based on the protection scope of the claims.

[0204] Finally: 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 spirit and principles of the present invention should be included in the protection scope of the present invention.

Claims

1. A centralized control chain energy storage grid auxiliary service method integrating multiple energy storage technologies, characterized in that: include: Configure M different types of energy storage technology units, where M is an integer greater than 1; Connecting M types of energy storage technology units for communication and data transmission; Collecting energy storage characteristic data of M types of energy storage technology units and calculating the corresponding energy storage efficiency; the energy storage characteristic data includes charging time, charging efficiency, discharge efficiency, self-discharge efficiency, number of cycles and total number of cycles; The calculation methods of energy storage efficiency include: ； In the formula, is the energy storage efficiency, is the total number of cycles, is the number of cycles, For charging time, For charging efficiency, is the self-discharge efficiency, , , is the preset weight coefficient; Predicting the energy demand value of the power grid, and combining the energy storage efficiency of the M energy storage technology units, collaboratively managing the M energy storage technology units; the steps of collaboratively managing the M energy storage technology units include: Step 1: Randomly combine M energy storage technology units to form a set of R units. ; Each unit set includes h types of energy storage technology units, , and each unit set is different; set digital labels for R unit sets, and the range of digital labels is , mark the numerical labels of the unit set as the set label; Step 2: Encode the set labels, obtain chromosomes, and construct the initial population; Step 3: Determine the fitness function; the expression of the fitness function is: ; In the formula, is the fitness corresponding to the i-th chromosome, is the discharge efficiency of the qth energy storage technology unit, is the energy storage efficiency of the qth energy storage technology unit, , ; Step 4: Natural selection of chromosomes in the population; Step 5: Perform crossover recombination on chromosomes in the population; Step 6: mutate the chromosomes in the population; Step 7: Get a new population, preset the population generation number as F, the fitness threshold as S, where F is an integer greater than 0 and S is a real number greater than 0; loop steps 4 to 6 until the generation number corresponding to the new population is F or the fitness corresponding to the chromosome in the new population is greater than or equal to the fitness threshold S, the loop ends, sort the fitness corresponding to each chromosome in the new population, and generate a first sorting table; Step 8: Obtain the set label corresponding to the corresponding chromosome according to the positive order of the first sorting table, obtain the corresponding unit set according to the set label, obtain the corresponding energy storage technology unit according to the unit set, and mark it as a selected unit; obtain the residual energy of the selected unit in real time; add the residual energy corresponding to the selected unit to obtain the total residual energy; Step 9: According to the prediction order, select a predicted energy that is not marked as selected energy, compare the remaining energy sum with the selected predicted energy, if the remaining energy sum is greater than or equal to the predicted energy, mark the selected predicted energy as selected energy, and loop through steps 2 to 9 until all predicted energies are marked as selected energies and the loop ends; if the remaining energy sum is less than the selected predicted energy, remove the corresponding fitness from the first sorting table, and loop through steps 8 to 9; Step 10: determining whether the selection unit corresponding to the first prediction energy is in a working state, the first prediction energy being the first predicted prediction energy; In step 10, from the remaining From the selected units corresponding to the prediction energies, select the selected units adjacent to each selected unit corresponding to the first prediction energy; calculate the output energy of each selected unit corresponding to each prediction energy, and take the same selected units as a group of analysis sets; add the output energy of each group of analysis sets to obtain the sum of adjacent energies of each group of analysis sets; then calculate the storage energy corresponding to each group of analysis sets, add the sum of the remaining energy to the storage energy and then subtract the sum of the adjacent energies to obtain the residual energy of each group of analysis sets; if the residual energy of each group of analysis sets is greater than or equal to 0, the selected unit corresponding to the first prediction energy is in working state; if there is an analysis set with residual energy less than 0, the selected unit corresponding to the residual energy less than 0 is marked as a deleted unit, and the fitness corresponding to the deleted unit is deleted from the first sorting table, and steps 8 to 10 are repeated, wherein only the corresponding selected unit is re-obtained for the first prediction energy; the expression of the stored energy is: ; In the formula, To store energy, To analyze the amount of predicted energy between the corresponding predicted energies of two selected units in the set, is the collection interval; M types of energy storage technology units are interconnected with the power grid to provide grid auxiliary services; grid auxiliary services include grid expansion, peak load shifting, demand response and grid frequency regulation; The peak shaving and valley filling method comprises: A sampling frequency is preset, and multiple power grid loads are collected according to the sampling frequency, where the power grid load is the power load carried by the power grid; a load curve is drawn according to the collected multiple power grid loads, and the peak and valley values ​​of the power grid load are obtained according to the load curve; When the grid load rises to a peak value, M types of energy storage technology units transmit electricity to the grid to reduce the peak value of the grid load; when the grid load drops to a valley value, the grid transmits electricity to M types of energy storage technology units to increase the valley value of the grid load; Demand response is that M types of energy storage technology units adjust their working status according to the real-time dispatch instructions received from the power grid; The methods of grid frequency regulation include: When the frequency of the power grid system decreases, that is, the power supplied by the power grid is insufficient to meet the power grid load demand, the power grid dispatching center issues a discharge instruction to the M types of energy storage technology units; the M types of energy storage technology units transmit the stored power to the power grid according to the discharge instruction, and increase the frequency of the power grid system through the electromagnetic induction effect; When the frequency of the power grid system increases, that is, the power supplied by the power grid exceeds the power grid load demand, the power grid dispatching center issues a charging instruction; according to the charging instruction, the power grid transmits the power to M types of energy storage technology units to reduce the frequency of the power grid system; It also includes selecting suitable energy storage technology units from the M types of energy storage technology units to replace the deleted units; Mark the selected unit corresponding to the first predicted energy as the first unit, and mark the energy storage technology unit that is not the first unit among the M types of energy storage technology units as the first candidate unit; obtain the residual energy corresponding to each first candidate unit, and mark the candidate unit whose residual energy is greater than the output energy corresponding to the deleted unit as the second candidate unit; use the discharge efficiency and energy storage efficiency of each second candidate unit and the discharge efficiency and energy storage efficiency of the deleted unit as a group of analysis data; input each group of analysis data into the trained probability analysis model respectively, and obtain the similarity probability corresponding to each group of analysis data; preset a probability threshold, sort each similar probability and probability threshold from large to small, mark the second candidate unit corresponding to the similarity probability that is ranked before the probability threshold and ranks first as the positively selected unit, and use the positively selected unit to replace the deleted unit; if each similar probability is ranked after the probability threshold, delete the fitness corresponding to the deleted unit from the first sorting table, and loop steps 8 to 10, where only the corresponding selected unit is re-obtained for the first predicted energy; The training process of the probability analysis model includes: Collect a group of analysis data and similarity probabilities corresponding to the group of analysis data in advance, where a is an integer greater than 1; Convert the analysis data and the corresponding similarity probabilities into a corresponding set of feature vectors; Each group of feature vectors is used as the input of the probability analysis model. The probability analysis model takes a group of predicted similarity probabilities corresponding to each group of analysis data as output, and takes the actual similarity probability corresponding to each group of analysis data as the prediction target, where the actual similarity probability is the pre-set similarity probability corresponding to the analysis data; minimizing the sum of prediction errors of all analysis data is used as the training goal; the probability analysis model is trained until the sum of prediction errors reaches convergence and the training is stopped; the probability analysis model is a deep neural network model.

2. The centralized control chain energy storage grid auxiliary service method integrating multiple energy storage technologies according to claim 1 is characterized in that: The M energy storage technology units realize communication and data transmission through a communication protocol stack, which includes communication standards, data transmission protocols and data formats.

3. The centralized control chain energy storage grid auxiliary service method integrating multiple energy storage technologies according to claim 2 is characterized in that: The method for predicting the power grid energy demand value comprises: A collection interval is preset, and N power grid energy demand values ​​are continuously collected according to the collection interval, where N is an integer greater than 1, and the continuously collected N power grid energy demand values ​​are input into the energy prediction model to predict the power grid energy demand values ​​at n future moments, where n is an integer greater than 1, and the predicted power grid energy demand values ​​are marked as predicted energy; The training method of the energy prediction model includes: Pre-collection continuous grid energy demand values ​​and construct a grid energy demand value set, , based on the set of grid energy demand values, training a prediction model for predicting grid energy demand values ​​at future moments; Preset a sliding step size L and a sliding window length; use a sliding window method to convert the grid energy demand values ​​in the grid energy demand value set into multiple training samples, use the training samples as the input of the recurrent neural network model, predict the grid energy demand value after the sliding step size L as the output, use the subsequent grid energy demand value of each training sample as the prediction target, use the prediction accuracy as the training target, use the mean absolute percentage error MAPE to evaluate the model accuracy of the prediction result, and when the calculated MAPE is less than the preset MAPE, the energy prediction model training is completed; generate an energy prediction model that predicts the grid energy demand value at a future moment according to the grid energy demand value; wherein the energy prediction model is an RNN neural network model.

4. The centralized control chain energy storage grid auxiliary service method integrating multiple energy storage technologies according to claim 3 is characterized in that: In step 2, the set label is encoded as X, where X is the chromosome and the range of X is ; Randomly generate G chromosomes to form the initial population ; In step 4, natural selection is carried out by combining the elite method and the rotation method; wherein the elite method produces For a population with a capacity of G, the fitness of the G chromosomes is arranged from large to small, and the one at the front is Each chromosome produces one offspring chromosome; the rotation method produces offspring chromosomes, that is, G chromosomes are generated according to the corresponding rotation probability offspring chromosomes; ; The expression of rotation probability is: ; In the formula, is the rotation probability corresponding to the i-th chromosome.

5. The centralized control chain energy storage grid auxiliary service method integrating multiple energy storage technologies according to claim 4 is characterized in that: In the step 5, U chromosomes are randomly selected from the population for crossover recombination to obtain U new chromosomes; the crossover recombination adopts the PMX method; after the chromosome crossover recombination, the fitness of the U new chromosomes is calculated, the fitness of the U new chromosomes and the fitness of the U chromosomes are sorted from large to small and a second sorting table is generated, and the U chromosomes in the second sorting table are replaced with the U chromosomes in the population for crossover recombination in positive order; In step 6, the mutation probability is preset to H, and the G chromosomes in the population are mutated according to the mutation probability. The mutation method is to randomly select the positions of two genes in the chromosome and exchange the values ​​of the two genes.

6. The centralized control chain energy storage grid auxiliary service method integrating multiple energy storage technologies according to claim 5 is characterized in that: The expression of output energy is: ; In the formula, is the output energy of the kth energy storage technology unit in the jth predicted energy, is the discharge efficiency of the kth energy storage technology unit in the jth predicted energy, is the j-th predicted energy, , , A is the number of energy storage technology units in each predicted energy.

7. The centralized control chain energy storage grid auxiliary service method integrating multiple energy storage technologies according to claim 6 is characterized in that: The interconnection between M types of energy storage technology units and the power grid includes power transmission and data transmission; The grid capacity expansion is to use M types of energy storage technology units to transmit the stored electric energy to the grid, thus expanding the grid load capacity.

8. A centralized control chain energy storage grid auxiliary service system integrating multiple energy storage technologies, implementing a centralized control chain energy storage grid auxiliary service method integrating multiple energy storage technologies as described in any one of claims 1 to 7, characterized in that: include: An energy storage technology unit configuration module, used to configure M different types of energy storage technology units, where M is an integer greater than 1; Centralized control chain communication module, used to connect the communication and data transmission between M types of energy storage technology units; An efficiency calculation module is used to collect energy storage characteristic data of M types of energy storage technology units and calculate the corresponding energy storage efficiency; Energy collaborative management module, used to predict the energy demand value of the power grid and to collaboratively manage M types of energy storage technology units in combination with their energy storage efficiency; The grid auxiliary service module is used to interconnect M types of energy storage technology units with the grid and provide grid auxiliary services; grid auxiliary services include grid capacity expansion, peak shaving and valley filling, demand response and grid frequency regulation.

Images