Photovoltaic power distribution network hybrid energy storage capacity optimization method and system based on VMD

Abstract

The invention discloses a VMD-based photovoltaic power distribution network hybrid energy storage capacity optimization method and system. The method comprises the steps of obtaining hybrid energy storage reference power according to a difference value between photovoltaic field output power and grid-connected power; determining the number of decomposition layers through an optimized variational mode decomposition algorithm, and decomposing the hybrid energy storage reference power according to the determined number of decomposition layers to obtain a plurality of mode components; and determining the lowest cost of the whole life cycle of the hybrid energy storage system, determining the rated power and rated capacity of the hybrid energy storage system under the condition of the lowest cost according to the plurality of modal components based on the charging and discharging consistency of the power type energy storage and the energy type energy storage and the energy storage working characteristics of the energy type energy storage, and obtaining an optimal configuration scheme. According to the method, the number of modal decomposition layers can be more accurately determined, and better overall performance is achieved, so that the energy storage capacity is balanced, the energy storage service life is prolonged, the energy utilization efficiency is improved, the energy storage replacement frequency and replacement cost are reduced, and the economical efficiency is further improved.

Description

Technical Field

[0001] The present invention relates to the technical field of photovoltaic distribution network hybrid energy storage capacity optimization configuration, and in particular to a method and system for optimizing photovoltaic distribution network hybrid energy storage capacity based on VMD. Background Art

[0002] As the proportion of photovoltaic power in the distribution network gradually increases, the impact of its power fluctuation on the safe operation of the distribution network is also increasing. Configuring a single energy storage system for photovoltaic power generation sites is not only difficult to meet the technical index requirements of its flexible grid connection, but also has higher investment costs. Therefore, in recent years, HESS (Hybrid Energy Storage System) composed of power and energy types has been widely used and is considered to be a potential solution that can simultaneously meet the flexible technical index requirements and economic requirements of photovoltaic site grid connection. By rationally configuring HESS hybrid energy storage to smooth the fluctuation of photovoltaic output, photovoltaic grid connection can be promoted, its grid connection income can be increased, and the cost of energy storage configuration can be reduced.

[0003] Most of the existing hybrid energy storage capacity configuration methods use the VMD (Variational Mode Decomposition) algorithm to decompose the hybrid energy storage output power, and allocate the decomposed components to power-type energy storage and energy-type energy storage according to the frequency, and obtain the capacity and power configuration of the HESS based on the operating cost and constraints. However, when allocating power to power-type energy storage and energy-type energy storage, the inconsistent charging and discharging states of power-type energy storage and energy-type energy storage at the same time and the working characteristics of energy storage are not fully considered in actual operation, resulting in an additional increase in the number of charging and discharging conversions of energy storage, affecting the life of energy storage, and further leading to an increase in the frequency of energy storage replacement and overall cost. Summary of the invention

[0004] The technical problem to be solved by the present invention is as follows: In view of the above-mentioned problems in the prior art, a method and system for optimizing the hybrid energy storage capacity of a photovoltaic distribution network based on VMD is provided, which comprehensively considers the charging and discharging state and the energy storage working characteristics, so as to extend the energy storage life, reduce the overall cost, and improve the economy and overall performance.

[0005] In order to solve the above technical problems, the technical solution adopted by the present invention is: A hybrid energy storage capacity optimization method for photovoltaic distribution network based on VMD, comprising: Step S1: Calculate the hybrid energy storage reference power; obtain the hybrid energy storage reference power according to the difference between the photovoltaic field output power and the grid-connected power; Step S2: variational modal decomposition; determining the number of decomposition layers by an optimized variational modal decomposition algorithm, and decomposing the hybrid energy storage reference power according to the determined number of decomposition layers to obtain a plurality of modal components; Step S3: Determine the optimal configuration; determine the minimum cost of the hybrid energy storage system over its entire life cycle, determine the rated power and rated capacity of the hybrid energy storage system under the minimum cost condition based on the charge and discharge consistency of the power-type energy storage and the energy-type energy storage and the energy storage working characteristics of the energy-type energy storage according to the multiple modal components, and obtain the optimal configuration solution.

[0006] Furthermore, the step S1 comprises: Monitor the maximum photovoltaic fluctuation of the photovoltaic field according to the output power of the photovoltaic field; When it is monitored that the maximum photovoltaic fluctuation of the photovoltaic field exceeds the grid-connected standard limit, the Gaussian filtering algorithm is used to obtain the grid-connected power, and the hybrid energy storage reference power is calculated based on the difference between the photovoltaic field output power and the grid-connected power.

[0007] Furthermore, the step S2 comprises: The FTTA-VMD algorithm is used to optimize the mode number K and the quadratic penalty term α in the variational mode decomposition algorithm by taking the minimum value of envelope entropy as the fitness function for optimizing VMD parameters. The hybrid energy storage reference power is subjected to variational modal decomposition according to the found optimal combination of the modal number K and the quadratic penalty term α to obtain K modal components.

[0008] Furthermore, the step S3 comprises: Step S31: Establishing a life cycle cost model for a hybrid energy storage system, calculating the life cycle cost of the hybrid energy storage system under different high and low frequency cutoff points of modal components, and obtaining the corresponding high and low frequency cutoff points under the optimal cost; Step S32: according to the high-low frequency dividing point under the optimal cost, the modal components of the multiple modal components higher than the high-low frequency dividing point are allocated to the power type energy storage, and the modal components lower than the high-low frequency dividing point are allocated to the energy type energy storage, so as to obtain the initial power allocation result; Step S33: optimizing the power type energy storage and energy type energy storage after the initial power distribution based on the charge and discharge consistency of the power type energy storage and the energy type energy storage and the energy storage working characteristics of the energy type energy storage to obtain a secondary power distribution result; Step S34: Determine the rated power and rated capacity of the hybrid energy storage system according to the secondary power allocation result to obtain an optimal configuration solution.

[0009] Furthermore, in step S31, the full life cycle cost model of the hybrid energy storage system is calculated based on the sum of investment cost, operation and maintenance cost, decommissioning and disposal cost, renewal and replacement cost, and auxiliary cost minus the recovered residual value.

[0010] Furthermore, the step S33 includes: Step S331: adjusting the initial power allocation result according to the charge and discharge states of the power type energy storage and the energy type energy storage. If the charge and discharge states of the two are inconsistent, the total power is allocated to the power type energy storage and the energy type energy storage with the larger power value to obtain an intermediate power allocation result; Step S332: adjusting the intermediate power allocation result according to the power value of the energy storage in the intermediate power allocation result; if the power value of the energy storage is lower than a preset threshold, allocating the total power to the power storage to obtain the secondary power allocation result.

[0011] Furthermore, the step S331 includes: The initial power allocation result is judged and adjusted according to the charge-discharge consistency coefficient, and the expression is:

[0012]

[0013] In the above formula, and are the powers of energy storage and power storage at time t after charge-discharge consistency optimization, μ is the charge-discharge consistency coefficient, e is a natural constant, and They are respectively the powers of energy type energy storage and power type energy storage at time t after the initial distribution.

[0014] Furthermore, the expression of step S332 is:

[0015] In the above formula, and They are respectively the power of energy type energy storage and power type energy storage at time t after adjustment considering the energy storage working characteristics, and d is the preset threshold.

[0016] Furthermore, the step S34 includes: According to the configured rated power, the maximum excess power or maximum power shortage of the energy storage power during the assessment period should be absorbed or supplemented. The rated power is calculated by combining the converter efficiency and the charging and discharging efficiency of the energy storage element. The expression is:

[0017] In the above formula, and are the rated powers of lithium batteries and supercapacitors respectively, , They are the discharge power and charging power of the lithium battery at time t after the secondary power distribution is optimized through charge and discharge consistency and energy storage working characteristics. , They are the discharge power and charging power of the supercapacitor at time t after the secondary power distribution is optimized through charge-discharge consistency and energy storage working characteristics. is the initial moment, are the efficiencies of DC-DC and DC-AC respectively, , are the charging and discharging efficiencies of lithium batteries, , are the charging and discharging efficiencies of lithium batteries, respectively, and T is the operating cycle; The rated capacity is calculated based on the state of charge of the energy storage itself, and the expression is:

[0018] In the above formula, and are the rated capacities of lithium batteries and supercapacitors, respectively. , The energy storage reference state of charge of lithium batteries and supercapacitors respectively, , The upper and lower limits of the state of charge of the lithium battery. , The upper and lower limits of the supercapacitor's state of charge.

[0019] The present invention further provides a photovoltaic distribution network hybrid energy storage capacity optimization system based on VMD, comprising an interconnected microprocessor and a memory, wherein the microprocessor is programmed or configured to execute the above-mentioned photovoltaic distribution network hybrid energy storage capacity optimization method based on VMD.

[0020] Compared with the prior art, the advantages of the present invention are: The present invention decomposes the hybrid energy storage reference power through an optimized variational modal decomposition algorithm to obtain multiple modal components. Compared with the conventional VMD algorithm, the degree of aliasing between the modal components can be further reduced, making the determination of the modal decomposition layer number more accurate and the overall performance better; the rated power and rated capacity configuration scheme of the hybrid energy storage system under the lowest cost condition is determined through a two-layer power optimization allocation strategy based on charge and discharge consistency and energy storage working characteristics, which can balance the energy storage capacity, extend the energy storage life, improve energy utilization efficiency, reduce the number of energy storage replacements and replacement costs, and further improve economy. BRIEF DESCRIPTION OF THE DRAWINGS

[0021] Figure 1 The figure is a flow chart of a hybrid energy storage capacity optimization method for a photovoltaic distribution network based on VMD according to an embodiment of the present invention.

[0022] Figure 2 It is a specific schematic diagram of the system structure and optimization process of the hybrid energy storage capacity optimization method for photovoltaic distribution network based on VMD in an embodiment of the present invention.

[0023] Figure 3 It is a curve diagram of actual output power on a typical day in an embodiment of the present invention.

[0024] Figure 4 This is a comparison diagram of photovoltaic power before and after using Gaussian filtering and other filtering methods in an embodiment of the present invention.

[0025] Figure 5 It is the probability distribution of 1-minute volatility on a typical day in the embodiment of the present invention.

[0026] Figure 6 : is the probability distribution of the 10-minute volatility on a typical day in an embodiment of the present invention.

[0027] Figure 7 This is a specific flow chart of optimizing VMD parameters using the FTTA-VMD algorithm in an embodiment of the present invention.

[0028] Figure 8 It is a comparison chart of algorithm iteration results of the FTTA-VMD algorithm and other optimization algorithms in an embodiment of the present invention.

[0029] Fig. 9 This is the Hilbert marginal spectrum using the EMD algorithm in the embodiment of the present invention.

[0030] Fig.10 This is the Hilbert marginal spectrum using the FTTA-VMD algorithm in the embodiment of the present invention. DETAILED DESCRIPTION

[0031] In order to better understand the above technical solution, the above technical solution will be described in detail below in conjunction with the accompanying drawings and specific implementation methods.

[0032] like Figure 1 and Figure 2 As shown, an embodiment of the present invention provides a method for optimizing hybrid energy storage capacity of a photovoltaic distribution network based on VMD, comprising: Step S1: Calculate the hybrid energy storage reference power; obtain the hybrid energy storage reference power according to the difference between the photovoltaic field output power and the grid-connected power; Step S2: variational modal decomposition; determining the number of decomposition layers by an optimized variational modal decomposition algorithm, and decomposing the hybrid energy storage reference power according to the determined number of decomposition layers to obtain multiple modal components; Step S3: Determine the optimal configuration; determine the minimum cost of the hybrid energy storage system over its entire life cycle, determine the rated power and rated capacity of the hybrid energy storage system under the lowest cost condition based on multiple modal components based on the charging and discharging consistency of power-type energy storage and energy-type energy storage and the energy storage working characteristics of energy-type energy storage, and obtain the optimal configuration solution.

[0033] It can be understood that this embodiment decomposes the hybrid energy storage reference power through the optimized variational modal decomposition algorithm to obtain multiple modal components. Compared with the conventional VMD algorithm, the degree of aliasing between the modal components can be further reduced, so that the determination of the number of modal decomposition layers is more accurate and the overall performance is better; the rated power and rated capacity configuration scheme of the hybrid energy storage system under the lowest cost condition is determined through the two-layer power optimization allocation strategy of charge and discharge consistency and energy storage working characteristics, which can balance the energy storage capacity, extend the energy storage life, improve energy utilization efficiency, reduce the number of energy storage replacements and replacement costs, and further improve economy.

[0034] In this embodiment, step S1 includes: Monitor the maximum photovoltaic fluctuation of the photovoltaic field according to the output power of the photovoltaic field; When it is monitored that the maximum photovoltaic fluctuation of the photovoltaic field exceeds the grid-connected standard limit, a Gaussian filter algorithm (preferably an adaptive Gaussian filter algorithm AGF) is used to obtain the grid-connected power, and the hybrid energy storage reference power is calculated based on the difference between the photovoltaic field output power and the grid-connected power.

[0035] Specifically, the expression of the maximum photovoltaic fluctuation is: (1) In the above formula, is the maximum photovoltaic fluctuation, is the photovoltaic field output power at time t, is the photovoltaic field output power at time (t+1), and max is the maximum value.

[0036] The expression of grid-connected power is: (2) In the above formula, is the grid-connected power at time t, i is the original data number, j is the data number in the window, L is the sliding window length at time i, is the AGF weight coefficient (preferably set to 1).

[0037] The expression of hybrid energy storage power is: (3) In the above formula, is the reference power of hybrid energy storage at time t.

[0038] In an exemplary application, the actual Figure 3 The output power data shown in the figure (sampling time is 1 minute) are filtered by Gaussian filtering, sliding average and exponential average algorithms to obtain the following: Figure 4 The PV power and original PV power curves after filtering by various algorithms are shown. It can be seen that the grid-connected power after AGF filtering is very smooth and the power fluctuation rate is greatly reduced, indicating that AGF can effectively assist the safe operation of high-proportion PV distribution networks, and verify that the use of adaptive Gaussian filtering algorithm can accurately extract the reference power of the HESS hybrid energy storage system, significantly improve the efficiency of PV power tracking, reduce energy storage pressure, and enhance economy.

[0039] Subsequently, the Gaussian filtered data was simulated and analyzed in Matlab, and the probability distribution of power fluctuation rate for 10 minutes and 1 minute was calculated as follows: Figure 5 and Figure 6 As shown, the data shown in Table 1 are obtained through analysis:

[0040] Table 1 From the data in Table 1, we can see that the maximum fluctuation rate of photovoltaic power in 1 minute is 10.49%, and the maximum fluctuation rate of photovoltaic power in 10 minutes is 40.05%, both of which are lower than the grid-connected standards of my country's photovoltaic power stations of 20% (upper limit of maximum fluctuation rate in 1 minute) and 100% (upper limit of maximum fluctuation rate in 10 minutes), meeting the requirements of the grid-connected standards.

[0041] In this embodiment, step S2 includes: The FTTA-VMD algorithm is used to optimize the mode number K and the quadratic penalty term α in the variational mode decomposition algorithm by taking the minimum value of envelope entropy as the fitness function for optimizing VMD parameters. According to the optimal combination of the found mode number K and the quadratic penalty term α, the hybrid energy storage reference power is subjected to variational modal decomposition to obtain K modal components.

[0042] Specifically, in order to charge and discharge the operating characteristics of lithium batteries and supercapacitors in HESS, the FTTA-VMD (Football Team Training Algorithm) algorithm is preferably used to decompose the HESS power obtained after the AGF filter is smoothed. FTTA is a meta-heuristic algorithm that simulates the behavior of players in high-level football team training. The practical application ability of FTTA is far superior to other optimization algorithms, which can effectively improve the utilization efficiency of renewable energy, and FTTA can be used to optimize variational mode decomposition (VMD) to further improve the data denoising effect. FTTA simulates three parts of a football training class: collective training, group training, and individual additional training. The traditional VMD algorithm requires the number of modes K and the quadratic penalty term α to be set in advance. The values ​​of K and α determine whether the original signal is fully decomposed, and the envelope entropy is an important indicator of the sparsity of the original signal. The smaller the value of the envelope entropy, the lower the degree of aliasing between the decomposed modal components. FTTA is applied to the optimization of the VMD parameter combination [K, α], which can greatly improve the objectivity of VMD parameter selection.

[0043] The FTTA optimization algorithm can be divided into 4 stages: 1. Initialization phase Set population size N ′, maximum number of iterations , the dimension of the search space , search the initial set of individuals and calculate the fitness value.

[0044] 2. Group training stage FTTA simulates the process of players training collectively under the guidance of a coach. Individuals are divided into: followers, discoverers, thinkers, and volatiles. In each iteration, the individual will randomly change the type. The mathematical model of the four types of individuals is as follows: (1) Followers Followers perform moderately and improve their performance by imitating the best performers. The follower mathematical model is as follows: (4) In the above formula, k is the number of iterations, is the value of the current best player, is the value of the i-th player on the current dimension j, is the state of the i-th player in dimension j before training, is the state of the i-th player in dimension j after training.

[0045] (2) Discoverer The discoverer has the potential to surpass the current best player by exploring new strategies. The mathematical model of the discoverer is as follows: (5) In the above formula, The worst player currently.

[0046] (3) Thinker Thinkers excel in theory and strategy, and improve performance by analyzing the key differences between the current best solution and the worst solution. The mathematical model of Thinkers is as follows: (6) (4) Fluctuators Volatility performance is poor, and random search in the solution space brings innovation and unexpected possibilities to the entire team. The mathematical model of the volatile player is as follows: (7) In the above formula, t(k) is a random number with t distribution, and its degree of freedom is the current number of iterations. As the degree of freedom increases, the probability of t distribution approaching the middle value (0) becomes higher and higher, and the distribution at both ends gradually decreases, becoming closer and closer to the normal distribution. Therefore, as the number of iterations increases, the degree of fluctuation will become smaller and smaller, and gradually change from global search to local search.

[0047] 3. Group training stage FTTA simulates the process of players entering group training after collective training. The individuals are divided into forwards, midfielders, defenders and goalkeepers. In group training, FTTA uses the MGEM adaptive clustering method (MixGaussEM) to divide individuals into four groups. The specific classification form is as follows: (8) The number of people in each group must be greater than or equal to 2, otherwise group training cannot be implemented. At this time, the coach will conduct a second grouping, using a uniform random grouping strategy to randomly and evenly divide the players into four groups.

[0048] After the grouping is completed, the players in the group learn or communicate, and group training is defined as three states: optimal learning, random learning, and random communication. The learning probability is defined as , the communication probability is , the player randomly selects the state in each iteration, and the mathematical model of the three states is as follows: (1) Optimal Learning In each dimension, a player has a certain probability of directly learning the ability of the best player in the group. The mathematical model is as follows: (9) In the above formula, Is the best player in group i. It is the state in which players have undergone optimal learning.

[0049] (2) Stochastic Learning In each dimension, a player has a certain probability of directly learning the ability of any random player in the group. The mathematical model is as follows: (10) In the above formula, is a random player of dimension j in group i, It is the state of the player after random learning.

[0050] (3) Random communication In training, learning is only part of it, and communication between two players is more important for improving ability. In each dimension, a player has a certain probability of communicating with any player in the group, and the mathematical model is as follows: (11) In the above formula, is a random player in j dimensions in group i, and the players exchanged their abilities in j dimensions; Random is a normally distributed random number, multiplied by Represents the two players' understanding of the others' abilities.

[0051] (4) Random error During training, there is a certain probability that an error will occur, that is, a player accidentally learns the content of another player in another dimension. The probability of this happening is low, but it is real and objective. Define the error probability as , the mathematical model is as follows: (12) 4. Personal additional training FTTA simulates the process of recalculating new physical fitness values ​​after group training and updating the player's status with better physical fitness values. The mathematical model is as follows: (13) In the above formula, (Cauchy) and (Gaussian) joint variation is used to describe individual additional training. The reason why Gaussian Cauchy distribution is chosen is that in the early stages of training, it can effectively provide a large range of improvements for players, which is beneficial to global search; as the number of iterations increases, the Gaussian distribution accounts for a larger proportion, which is more conducive to local search.

[0052] The minimum value of envelope entropy is used as the fitness function of FTTA to optimize VMD parameters. The flowchart of FTTA to optimize VMD parameters is as follows: Figure 7 As shown: The first step is to initialize the FTTA optimization parameters and randomly generate the [K, α] combination as the initial input of the optimization process; the second step is to calculate the fitness of the current parameter combination to measure the quality of the current solution and provide a reference for subsequent optimization; the third step is to enter the main loop, and the loop variable i starts from 1 until the maximum number of iterations In each iteration, find the best and worst solutions in the current population, improve the fitness of all individuals by training the population as a whole, and group the population to form multiple clusters for subsequent collaboration and optimization; the fourth step is to calculate the population size and determine whether the current population size is less than the number of teams. number, if the population is smaller than the number of teams, random error, random communication, random learning, learning optimization and other steps are executed, and after each training and learning, the fitness is recalculated, and the [K, α] combination with better fitness is updated to keep the population evolving in a better direction; if the population is larger than the number of teams, it directly enters the stage of finding the optimal additional training; the fifth step is to find the optimal additional training. The system tries to find a better solution and conducts additional training for individuals with higher fitness to further improve the overall optimization effect; the sixth step is to determine whether the new solution is better than the current solution. If the new solution is better than the current solution, the current population is updated to the new solution, and the population is continued to be updated and optimized; if the new solution is not better than the current solution, the current solution is kept unchanged to avoid degradation; the seventh step is that when the maximum number of iterations is reached or the termination condition is met, the algorithm ends and outputs the optimized [K, α] combination as the final optimal solution.

[0053] In an exemplary application, the initial parameters of the FTTA algorithm are set as follows: the dimension of the search space Dim = 2, the population size N' = 10, the maximum number of iterations , and its iterative results are compared with other algorithms. Figure 8 As shown. It can be seen that other optimization algorithms need to be iterated at least twice to get the minimum fitness function value of 0.0779543, and there is a phenomenon of falling into the local optimum, while FTTA only needs to be iterated once to get the minimum fitness function value, the optimal K value is 10, and the optimal α value is 2000. That is, the optimization speed and accuracy of the FTTA algorithm are better than other algorithms, and the global optimization ability is stronger. In addition, please refer to Fig. 9 and Fig.10 , for hybrid energy storage power The traditional EMD method and the FTTA-VMD method are used for decomposition and comparative analysis. From the Hilbert marginal spectrum after decomposition, it can be seen that the results obtained by EMD decomposition have serious modal aliasing, while the use of FTTA-VMD decomposition effectively avoids this problem, further demonstrating that the FTTA-VMD method has better performance.

[0054] Subsequently, the hybrid energy storage power is decomposed according to the optimal solution to obtain the corresponding number of sub-components IMF, and the IMF is Hilbert transformed. According to the minimum allocation of adjacent mode aliasing of the Hilbert marginal spectrum, the low-frequency power component is allocated to the lithium battery, and the high-frequency power component and the remainder are allocated to the supercapacitor according to the difference in the characteristics of power-type energy storage and energy-type energy storage. The initial power allocation result is shown in the following formula (14): (14) In the above formula, and The power of the lithium battery and supercapacitor after allocation at time t, represents the high and low frequency critical points of the decomposition (refer to step S31 for the specific determination method), is the Xth modal component of the decomposition, K is the total number of decomposed modes, and res is the decomposed modal residual.

[0055] In this embodiment, step S3 includes: Step S31: Establishing a life cycle cost model for a hybrid energy storage system, calculating the life cycle cost of the hybrid energy storage system under different high and low frequency cutoff points of modal components, and obtaining the corresponding high and low frequency cutoff points under the optimal cost; Step S32: according to the high-low frequency dividing point under the optimal cost, the modal components higher than the high-low frequency dividing point among the multiple modal components are allocated to the power type energy storage, and the modal components lower than the high-low frequency dividing point are allocated to the energy type energy storage, so as to obtain the initial power allocation result; Step S33: optimizing the power type energy storage and energy type energy storage after the initial power distribution based on the charge and discharge consistency of the power type energy storage and the energy type energy storage and the energy storage working characteristics of the energy type energy storage to obtain a secondary power distribution result; Step S34: Determine the rated power and rated capacity of the hybrid energy storage system according to the secondary power allocation result to obtain the optimal configuration solution.

[0056] In step S31 of this embodiment, the life cycle cost model of the hybrid energy storage system is obtained by subtracting the recovery residual value from the sum of the investment cost, operation and maintenance cost, decommissioning disposal cost, replacement cost, and auxiliary cost. The specific expression is: (15) In the above formula, is the HESS life cycle cost, is the investment cost incurred during the entire life cycle, For operation and maintenance costs, is the decommissioning cost, To update the replacement cost, For auxiliary costs, To recover the residual value.

[0057] The expression of investment cost is as follows: (16) In the above formula, is the investment cost of HESS, , are the investment costs of lithium batteries and supercapacitors, , are the unit power investment costs of lithium batteries and supercapacitors, , are the unit capacity investment costs of lithium batteries and supercapacitors, r is the base discount rate, k is the kth year in the life cycle, and n is the number of replacements of the energy storage equipment ( The number of times the lithium battery is replaced. is the number of times the supercapacitor is replaced), The total life cycle period, is the rated power of the lithium battery in HESS, is the rated power of the supercapacitor in HESS, is the rated capacity of the lithium battery in HESS, is the rated capacity of the supercapacitor in HESS.

[0058] The expression of operation and maintenance cost is as follows: (17) In the above formula, is the operation and maintenance cost of HESS, , are the operation and maintenance costs of lithium batteries and supercapacitors, , They are the unit power operation and maintenance costs of lithium batteries and supercapacitors, , They are the unit capacity operation and maintenance costs of lithium batteries and supercapacitors, , They are the annual charge and discharge capacity of lithium batteries and supercapacitors respectively.

[0059] The expression of decommissioning disposal cost is as follows: (18) In the above formula, is the HESS decommissioning and disposal cost, , are the decommissioning costs of lithium batteries and supercapacitors, , are the unit power retirement disposal costs of lithium batteries and supercapacitors, , are the unit capacity retirement and disposal costs of lithium batteries and supercapacitors respectively, and j is the jth year in the entire life cycle.

[0060] The expression for updating the replacement cost is as follows: (19) In the above formula, is the replacement cost of HESS, , They are the replacement costs of lithium batteries and supercapacitors respectively; , are the unit power replacement costs of lithium batteries and supercapacitors, , They are the unit capacity renewal and replacement costs of lithium batteries and supercapacitors respectively.

[0061] The expression for auxiliary equipment cost is as follows: (20) In the above formula, is the cost of auxiliary equipment for HESS, , are the auxiliary equipment costs of lithium batteries and supercapacitors, , are the unit power auxiliary equipment costs of lithium batteries and supercapacitors, , They are the unit capacity auxiliary equipment costs of lithium batteries and supercapacitors respectively.

[0062] The expression of the recovery residual value is as follows: (twenty one) In the above formula, is the residual value recovery rate.

[0063] In addition, the following constraints need to be considered to calculate the optimal value of the life cycle cost of the hybrid energy storage system: (1) Energy storage power constraint, the specific expression is as follows: (twenty two) In the above formula, is the rated power of the lithium battery, is the rated power of the supercapacitor, , They are the powers of lithium battery and supercapacitor at time t after considering the charging and discharging efficiency and converter efficiency.

[0064] (2) Energy storage SOC constraint, the specific expression is as follows: (twenty three) In the above formula, , are the charge states of the lithium battery and supercapacitor at time t, , They are the upper and lower limits of the state of charge of the lithium battery, , They are the upper and lower limits of the state of charge of the supercapacitor respectively.

[0065] In this embodiment, step S33 includes: Step S331, adjusting the initial power allocation result according to the charge and discharge states of the power type energy storage and the energy type energy storage. If the charge and discharge states of the two are inconsistent, the total power is allocated to the power type energy storage and the energy type energy storage with the larger power value to obtain an intermediate power allocation result; Step S332: adjust the intermediate power allocation result according to the power value of the energy storage in the intermediate power allocation result. If the power value of the energy storage is lower than the preset threshold, the total power is allocated to the power storage to obtain a secondary power allocation result.

[0066] Specifically, after the initial power distribution, there may be inconsistencies in the charging and discharging states of lithium batteries and supercapacitors, unreasonable distribution of small high-frequency power fluctuations, and battery energy storage operating under extreme charge states SoC. These problems will affect the life of battery energy storage. Based on the above problems, the initially allocated hybrid energy storage power needs to be optimized again, which is specifically divided into two levels of optimization strategies, namely, optimization strategies that consider charging and discharging consistency and optimization strategies that consider energy storage working characteristics.

[0067] Since the charging and discharging states of lithium batteries and supercapacitors are inconsistent, the number of times the hybrid energy storage is charged and discharged will increase, which will reduce the life of the hybrid energy storage, increase the number of replacements, increase the replacement and renewal costs, and deteriorate the economy. Based on the above problems, a strategy considering the consistency of charging and discharging is proposed. The expression of the HESS optimization strategy considering the consistency of charging and discharging is as follows: (twenty four) In the above formula, and are the power of lithium battery and supercapacitor at time t after charge and discharge consistency optimization, μ is the charge and discharge consistency coefficient, and its expression is as follows: (25) In the above formula, e is a natural constant. If 1≤μ≤e, it means that the charging and discharging of the two are the same at this time; otherwise, it means that the charging and discharging states of the lithium battery and the supercapacitor are opposite. If μ<1, it means that the absolute value of the supercapacitor power is greater than that of the lithium battery, so that the supercapacitor bears the total power and the lithium battery does not work; similarly, if μ>e, the lithium battery bears the total power and the supercapacitor does not work.

[0068] Since the initial distribution of lithium battery power contains some high-frequency and small-amplitude components, the lithium battery is frequently charged and discharged, which affects the battery service life. Based on the above problems, a strategy considering the working characteristics of energy storage is proposed. The expression of the HESS optimization strategy considering the working characteristics of energy storage is as follows: (26) In the above formula, and are the power of lithium battery and supercapacitor at time t after adjustment considering the energy storage working characteristics, and d is the set threshold, which can be selected according to the actual situation. ,in is the rated power of the lithium battery.

[0069] It can be understood that by optimizing the power after the initial power distribution through the consistency of charge and discharge and the working characteristics of energy storage, it is possible to balance the energy storage capacity, extend the energy storage life, improve energy utilization efficiency, reduce the number of energy storage replacements and replacement costs, and further improve economic efficiency.

[0070] In this embodiment, the maximum excess power or maximum power shortage of the energy storage power that should be absorbed or supplemented by the configured rated power during the assessment period is calculated by combining the converter efficiency and the charging and discharging efficiency of the energy storage element (including power-type energy storage and energy-type energy storage), and the rated power is expressed as follows: (27) In the above formula, and are the rated powers of lithium batteries and supercapacitors respectively, , They are the discharge power and charging power of the lithium battery at time t after the secondary power distribution is optimized through charge and discharge consistency and energy storage working characteristics. , They are the discharge power and charging power of the supercapacitor at time t after the secondary power distribution is optimized through charge-discharge consistency and energy storage working characteristics. is the initial moment, are the efficiencies of DC-DC and DC-AC respectively, , are the charging and discharging efficiencies of lithium batteries, , are the charging and discharging efficiencies of lithium batteries respectively, and T is the operating cycle.

[0071] The rated capacity is calculated based on the state of charge of the energy storage itself, and its expression is: (28) In the above formula, and are the rated capacities of lithium batteries and supercapacitors, respectively. , The energy storage reference state of charge of lithium batteries and supercapacitors respectively, , The upper and lower limits of the state of charge of the lithium battery. , The upper and lower limits of the supercapacitor's state of charge.

[0072] In the calculation of the full life cycle cost of the hybrid energy storage system, the number of replacements of the energy storage equipment is a key point to consider. Since the traditional battery life estimation model only considers the cycle life, it ignores the impact of the SoC overcharge and over-discharge state on the life, and at the same time increases the estimation of the battery life loss caused by micro-circulation under the SoC floating charge state. Based on the above problems, a battery life estimation model that considers cycle life and calendar life is proposed. This model considers the effects of solid-electrolyte interface (SEI) film formation, calendar aging, cycle aging and temperature on the chemical composition of many lithium-ion batteries. Taking lithium batteries as an example, the model analyzes the energy storage capacity attenuation characteristics of the battery as shown in the following formula (9): (29) In the above formula, and is the fitting parameter of lithium battery (NMC battery), f is the function of cycle life and calendar life, is the cycle life function, is the calendar life function; It is the SoC factor. is the temperature factor based on the Arrhenius equation, where: (30) In the above formula, 3000 and 0.8 mean that the cycle life is 3000 cycles at 80% dod; is the calendar life fitting coefficient; is the SOC influencing factor coefficient; σ is the average SOC value; is the reference SOC value; is the temperature influence factor coefficient; T is the actual temperature; is the reference temperature. The preferred reference SOC value is 0.5, reference temperature is 25°C, the reference condition is not affected by the decrease, at this time , , kσ=1.039.

[0073] Based on the battery life calculation model, the HESS life prediction steps are as follows: The first step is to calculate the capacity attenuation: Input the SoC sequence and use the rain flow counting method to calculate the battery charge and discharge depth dod and the corresponding number of cycles for each cycle , the dod of each cycle and the corresponding Substituting into formula (29) (10) we get the function f about cycle life and calendar life. Substituting f through formula (29) (9) we get the capacity attenuation of the energy storage battery. .

[0074] The second step is to calculate the operating life of the HESS: Calculate the annual capacity attenuation of battery energy storage operation based on the annual photovoltaic fluctuation data , substituting into formula (31) (11) we can get the operating life of the configured energy storage battery : (31) In the above formula, is the capacity attenuation degree of battery failure, and its preferred value is 20%.

[0075] Since supercapacitors are less affected by DOD, the number of charge and discharge times is the main factor affecting their lifespan. Their service life can be estimated based on their actual daily charge and discharge times. for: (32) In the above formula, is the maximum number of charge and discharge times in the life cycle of the supercapacitor, is the actual number of times the supercapacitor is charged and discharged per day, and Day is the number of days the supercapacitor is in operation in a year.

[0076] The third step is to calculate the number of replacements during the entire life cycle of the HESS: (33) In the above formula, is the number of battery replacements, is the number of supercapacitor replacements, is the full life cycle of HESS, For battery operating life, is the operating life of the supercapacitor.

[0077] The number of replacements between lithium batteries and supercapacitors is calculated using the life prediction model and After that, substitute it into the life cycle cost model of the hybrid energy storage system for subsequent calculations.

[0078] In an exemplary application, the life cycle cost values ​​of the hybrid energy storage system calculated at different frequency cutoff points (i.e., high and low frequency cutoff points) are shown in Table 2:

[0079] Table 2 From the data in Table 2, we can see that the hybrid energy storage system has the lowest life cycle cost at the frequency demarcation point 1. Therefore, taking the modal component IMF1 as the lithium battery power, the modal components IMF2-IMF9 and res as the supercapacitor power, the optimal cost is 8.02×10 7 Yuan.

[0080] Subsequently, the power is distributed according to steps S32 and S33, and the rated power and rated capacity of the hybrid energy storage system are calculated as shown in Table 3:

[0081] Table 3 In Table 3, Scheme 1 is the configuration result after FTTA-VMD decomposition, Scheme 2 is the configuration result after further considering the optimization of charge-discharge consistency, and Scheme 3 is the configuration result after further considering the optimization of charge-discharge consistency and energy storage working characteristics. b is the cost of lithium battery, C lcc is the hybrid energy storage cost (HESS full life cycle cost), which includes the lithium battery cost and the supercapacitor cost. As can be seen from Table 3, considering the optimization of charge and discharge consistency and the optimization of energy storage working characteristics can effectively reduce the cost of lithium batteries, thereby reducing the cost of the hybrid energy storage system, which has economic improvement.

[0082] The present invention further provides a photovoltaic distribution network hybrid energy storage capacity optimization system based on VMD, comprising an interconnected microprocessor and a memory, wherein the microprocessor is programmed or configured to execute the above-mentioned photovoltaic distribution network hybrid energy storage capacity optimization method based on VMD.

[0083] The system of the present invention corresponds to the above method and also has the advantages of the above method, which will not be described in detail here.

[0084] The present invention implements all or part of the processes in the above-mentioned embodiment method, and can also be completed by instructing related hardware through a computer program. The computer program can be stored in a computer-readable storage medium. When the computer program is executed by the processor, the steps of the above-mentioned method embodiment can be implemented. Among them, the computer program includes computer program code, and the computer program code can be in source code form, object code form, executable file or some intermediate form. Computer-readable media include: any entity or device that can carry computer program code, recording medium, U disk, mobile hard disk, disk, optical disk, computer memory, read-only memory (ROM, Read-Only Memory), random access memory (RAM, RandomAccess Memory), electric carrier signal, telecommunication signal and software distribution medium. The memory is used to store computer programs and / or modules. The processor implements various functions by running or executing computer programs and / or modules stored in the memory, and calling data stored in the memory. The memory may include a high-speed random access memory and may also include a non-volatile memory, such as a hard disk, a memory, a plug-in hard disk, a smart media card (SMC), a secure digital (SD) card, a flash card (Flash Card), at least one disk storage device, a flash memory device, or other volatile solid-state storage devices.

[0085] The above is only a preferred embodiment of the present invention, and the protection scope of the present invention is not limited to the above embodiments. All technical solutions under the concept of the present invention belong to the protection scope of the present invention. It should be pointed out that for ordinary technicians in this technical field, some improvements and modifications without departing from the principle of the present invention should also be regarded as the protection scope of the present invention.

Claims

1. A hybrid energy storage capacity optimization method for photovoltaic distribution network based on VMD, characterized in that: include: Step S1, calculating the hybrid energy storage reference power: obtaining the hybrid energy storage reference power according to the difference between the photovoltaic field output power and the grid-connected power; Step S2, variational modal decomposition: determining the number of decomposition layers by using an optimized variational modal decomposition algorithm, and decomposing the hybrid energy storage reference power according to the determined number of decomposition layers to obtain a plurality of modal components; Step S3, determine the optimal configuration: determine the minimum cost of the hybrid energy storage system over its entire life cycle, determine the rated power and rated capacity of the hybrid energy storage system under the minimum cost condition based on the charge and discharge consistency of the power-type energy storage and the energy-type energy storage and the energy storage working characteristics of the energy-type energy storage according to the multiple modal components, and obtain the optimal configuration solution.

2. The method for optimizing hybrid energy storage capacity of photovoltaic distribution network based on VMD according to claim 1, characterized in that: The step S1 comprises: Monitor the maximum photovoltaic fluctuation of the photovoltaic field according to the output power of the photovoltaic field; When it is monitored that the maximum photovoltaic fluctuation of the photovoltaic field exceeds the grid-connected standard limit, the Gaussian filtering algorithm is used to obtain the grid-connected power, and the hybrid energy storage reference power is calculated based on the difference between the photovoltaic field output power and the grid-connected power.

3. The method for optimizing hybrid energy storage capacity of photovoltaic distribution network based on VMD according to claim 1, characterized in that: The step S2 comprises: The FTTA-VMD algorithm is used to optimize the mode number K and the quadratic penalty term α in the variational mode decomposition algorithm by taking the minimum value of envelope entropy as the fitness function for optimizing VMD parameters. The hybrid energy storage reference power is subjected to variational modal decomposition according to the found optimal combination of the modal number K and the quadratic penalty term α to obtain K modal components.

4. The method for optimizing hybrid energy storage capacity of photovoltaic distribution network based on VMD according to claim 1, characterized in that: The step S3 comprises: Step S31: Establishing a life cycle cost model for a hybrid energy storage system, calculating the life cycle cost of the hybrid energy storage system under different high and low frequency cutoff points of modal components, and obtaining the corresponding high and low frequency cutoff points under the optimal cost; Step S32: according to the high-low frequency dividing point under the optimal cost, the modal components of the multiple modal components higher than the high-low frequency dividing point are allocated to the power type energy storage, and the modal components lower than the high-low frequency dividing point are allocated to the energy type energy storage, so as to obtain the initial power allocation result; Step S33: optimizing the power type energy storage and energy type energy storage after the initial power distribution based on the charge and discharge consistency of the power type energy storage and the energy type energy storage and the energy storage working characteristics of the energy type energy storage to obtain a secondary power distribution result; Step S34: Determine the rated power and rated capacity of the hybrid energy storage system according to the secondary power allocation result to obtain an optimal configuration solution.

5. The method for optimizing hybrid energy storage capacity of photovoltaic distribution network based on VMD according to claim 4 is characterized in that: In step S31, the full life cycle cost model of the hybrid energy storage system is calculated based on the sum of investment cost, operation and maintenance cost, decommissioning cost, renewal and replacement cost, and auxiliary cost minus the recovered residual value.

6. The method for optimizing hybrid energy storage capacity of photovoltaic distribution network based on VMD according to claim 4, characterized in that: The step S33 comprises: Step S331: adjusting the initial power allocation result according to the charge and discharge states of the power type energy storage and the energy type energy storage. If the charge and discharge states of the two are inconsistent, the total power is allocated to the power type energy storage and the energy type energy storage with the larger power value to obtain an intermediate power allocation result; Step S332: adjusting the intermediate power allocation result according to the power value of the energy storage in the intermediate power allocation result; if the power value of the energy storage is lower than a preset threshold, allocating the total power to the power storage to obtain the secondary power allocation result.

7. The method for optimizing hybrid energy storage capacity of photovoltaic distribution network based on VMD according to claim 6, characterized in that: The step S331 includes: The initial power allocation result is judged and adjusted according to the charge-discharge consistency coefficient, and the expression is: In the above formula, and are the powers of energy storage and power storage at time t after charge-discharge consistency optimization, μ is the charge-discharge consistency coefficient, e is a natural constant, and They are respectively the powers of energy type energy storage and power type energy storage at time t after the initial distribution.

8. The method for optimizing hybrid energy storage capacity of photovoltaic distribution network based on VMD according to claim 6, characterized in that: The expression of step S332 is: In the above formula, and They are respectively the power of energy type energy storage and power type energy storage at time t after adjustment considering the energy storage working characteristics, and d is the preset threshold.

9. The method for optimizing hybrid energy storage capacity of photovoltaic distribution network based on VMD according to claim 5, characterized in that: The step S34 comprises: According to the configured rated power, the maximum excess power or maximum power shortage of the energy storage power during the assessment period should be absorbed or supplemented. The rated power is calculated by combining the converter efficiency and the charging and discharging efficiency of the energy storage element. The expression is: In the above formula, and are the rated powers of lithium batteries and supercapacitors respectively, , They are the discharge power and charging power of the lithium battery at time t after the secondary power distribution is optimized through charge and discharge consistency and energy storage working characteristics. , They are the discharge power and charging power of the supercapacitor at time t after the secondary power distribution is optimized through charge-discharge consistency and energy storage working characteristics. is the initial moment, are the efficiencies of DC-DC and DC-AC respectively, , are the charging and discharging efficiencies of lithium batteries, , are the charging and discharging efficiencies of lithium batteries, respectively, and T is the operating cycle; The rated capacity is calculated based on the state of charge of the energy storage itself, and the expression is: In the above formula, and are the rated capacities of lithium batteries and supercapacitors, respectively. , Energy storage reference state of charge for lithium batteries and supercapacitors, respectively. , The upper and lower limits of the state of charge of the lithium battery. , The upper and lower limits of the supercapacitor's state of charge.

10. A photovoltaic distribution network hybrid energy storage capacity optimization system based on VMD, comprising an interconnected microprocessor and a memory, characterized in that: The microprocessor is programmed or configured to execute the VMD-based hybrid energy storage capacity optimization method for photovoltaic distribution network as described in any one of claims 1 to 9.

Images