Optimal configuration method of energy storage system based on fluctuation smoothing strategy under multi-granularity mechanism

Abstract

The present invention discloses an energy storage system optimization configuration method based on a fluctuation suppression strategy under a multi-granularity mechanism, specifically: under a multi-granularity mechanism, a wind-solar power decomposition and reconstruction method based on time-of-use electricity prices is designed, and the wind-solar hybrid output power is decomposed and reconstructed to obtain the grid-connected component and the absorption compensation component; a high-proportion hydrogen storage power allocation strategy is designed considering multiple electrolysis types; a capacity optimization configuration model for a wind-solar hybrid energy storage system with minimum total economic cost and optimal suppression effect as the objective function is established, and the absorption compensation component and the power allocation strategy are input into the optimization configuration model, and the solution is obtained to obtain the rated power of the electrolyzer, the rated power of the fuel cell, the rated capacity of the hydrogen storage tank, the rated power and capacity of the electrochemical energy storage, the economic cost and the suppression effect. The method of the present invention improves the suppression effect, reduces the capacity and power of the electrochemical energy storage, and exerts the main absorption compensation effect of hydrogen energy storage.

Description

Technical Field

[0001] The present invention belongs to the technical field of energy storage configuration, and specifically relates to an energy storage system optimization configuration method based on a fluctuation suppression strategy under a multi-granularity mechanism. Background Art

[0002] The hybrid energy storage system can alleviate the fluctuations and randomness caused by the grid connection of new energy. When new energy is running for a long time, the electrochemical energy storage for short-term fluctuations cannot match it. Therefore, in order to achieve long-term and large-capacity energy storage, seasonal energy storage can be used to effectively store electrical energy. If necessary, it can even be moved within a space, which can solve the problem of unbalanced electricity.

[0003] As a new clean energy, hydrogen can achieve large-capacity energy storage and peak-shaving. Its power-to-gas technology converts surplus electricity into hydrogen to solve the problem of power imbalance. Hydrogen energy storage is used to smooth out power fluctuations caused by wind power and photovoltaic grid connection, providing strong support for new power systems and increasing the penetration rate of renewable energy, thereby achieving the "dual carbon" goal. The hybrid electrolysis energy storage system mainly aims to smooth out wind and solar fluctuations, using hydrogen as an intermediate energy storage to adjust wind and solar fluctuations.

[0004] Water electrolysis hydrogen production technology includes alkaline electrolysis, proton membrane electrolysis and solid oxide electrolysis. At present, alkaline electrolyzers and proton membrane electrolyzers are mostly used to consume new energy electricity, analyze the capacity configuration or optimize the scheduling of their combined operation, and their main goal is to maximize hydrogen production and minimize economic costs. However, there are few studies on the use of multiple electrolysis combined with wind and solar grid connection, and they are highly dependent on electrochemical energy storage. The working characteristics of the electrolyzer are not highlighted, and hydrogen energy storage and water electrolysis hydrogen production technology have not really played their characteristics of seasonal energy storage with a long time span. Summary of the invention

[0005] The purpose of the present invention is to provide an energy storage system optimization configuration method based on a fluctuation smoothing strategy under a multi-granularity mechanism, thereby improving the wind-solar penetration rate and enhancing the hybrid energy storage system's effect of smoothing wind and solar power.

[0006] The technical solution adopted by the present invention is an energy storage system optimization configuration method based on a fluctuation suppression strategy under a multi-granularity mechanism, specifically:

[0007] Step 1: Under the multi-granularity mechanism, a wind-solar power decomposition and reconstruction method based on time-of-use electricity prices is designed, and the wind-solar hybrid output power is decomposed and reconstructed to obtain the grid-connected component and the consumption compensation component;

[0008] Step 2: Design a high-proportion hydrogen storage power allocation strategy considering multiple electrolysis types;

[0009] Step 3: Establish a capacity optimization configuration model for the wind-solar hybrid energy storage system with the minimum total economic cost and the best suppression effect as the objective function.

[0010] Step 4: Input the consumption compensation component obtained in step 1 and the power allocation strategy obtained in step 2 into the capacity optimization configuration model of the wind-solar hybrid energy storage system, and use the NSGA-II algorithm to solve the model to obtain the rated power of the two electrolyzers, the rated power of the fuel cell, the rated capacity of the hydrogen storage tank, the rated power and rated capacity of the electrochemical energy storage, the economic cost and the smoothing effect.

[0011] The present invention is also characterized in that:

[0012] In step 1, specifically:

[0013] Step 1.1: Set the photovoltaic output power P pv and wind power output P w Add them together to get the wind-solar hybrid power P M , as shown in formula (1);

[0014] P M =P pv +P w (1);

[0015] Step 1.2, calculate the decomposition and reconstruction standard score of the fine-grained layer, as shown in formula (2);

[0016] Score = Fine-Grained (PRICE, P pv ,P w ) (2)

[0017] PRICE is the time-of-use electricity price, which is divided into peak electricity price, valley electricity price and flat electricity price;

[0018] Normalize the standard score to get the reconstruction standard coefficient K t , as shown in formula (3);

[0019] K t =Min-Max(Score) (3)

[0020] Step 1.3, the grid access limit of wind farms is divided into two time scales of 1 minute and 10 minutes, and the grid access limit of photovoltaic power stations is only 1 minute. Based on this, the Coarse-Grained function is established, as shown in equations (4) to (6);

[0021] [ΔD 1 ,ΔD 10 ]=Coarse-Grained(P T ) (4);

[0022]

[0023] ΔD 1 and ΔD 10 are the fluctuation limits of wind-solar hybrid power for 1 minute and 10 minutes respectively, P T The total installed capacity of renewable energy connected to the system, in MW;

[0024] Step 1.4: Based on step 1.2 and step 1.3, construct multi-granularity grid-connected reconstruction limit conditions and establish wind-solar hybrid power fluctuation limit conditions, as shown in equations (7) and (8);

[0025]

[0026] is the wind-solar hybrid power output at time t, and is the wind-solar hybrid power output at time t-1 and t-10;

[0027] Step 1.5: Decompose the wind-solar hybrid power in step 1.1 into various intrinsic modal components and residual components by using the empirical mode decomposition (EMD) method, and reconstruct it according to the hourly fluctuation constraint conditions established in step 1.4 to obtain the grid-connected component and the absorption compensation component.

[0028] In step 2, specifically: the high proportion hydrogen storage power allocation strategy is divided into two parts:

[0029] 1) Hydrogen energy storage power compensation part

[0030] Hydrogen fuel cells are used as the main compensation output, and electrochemical energy storage is used as an auxiliary. There are two working states according to the content of the hydrogen storage tank;

[0031] Working mode 1, as shown in formula (11):

[0032] SOH<0.4 (11);

[0033] Considering the power output of electrochemical energy storage, there are two cases according to the state of electrochemical energy storage capacity SOC, as shown in equations (12) and (13):

[0034]

[0035] It is the power output of the electrochemical energy storage at time t, which has positive and negative conditions, representing the charging and discharging states of the electrochemical energy storage in the entire energy storage system respectively;

[0036] is the compensation power output of the external power grid at time t; is the compensation component at time t, P FCmin It is the minimum output power of the fuel cell;

[0037] Working mode 2: As shown in formula (14):

[0038] SOH ≥ 0.4 (14);

[0039] When there is sufficient hydrogen reserve and the fuel cell operates at rated power, if the electrochemical energy storage SOC is low, the excess will be used for electrochemical charging to maintain the energy storage state of the electrochemical energy storage and prepare good power storage to cope with the extreme conditions of the hybrid energy storage system; otherwise, the fuel cell operating power will be updated and adjusted.

[0040] 2) Hydrogen energy storage power consumption

[0041] Working mode 3: As shown in formula (15):

[0042] SOH≤0.6 (15);

[0043] Working mode 4: When power is needed, the two electrolyzers are given a start-up signal. The first cold start time of AEL is 60 minutes, while that of PEM is only 5 minutes. During the electrolyzer start-up time, the power is mainly consumed by the electrochemical energy storage output. When PEM is available but AEL is not, PEM works alone, as shown in formula (17):

[0044] SOC≤0.75 (17);

[0045] 0.75 is selected as the electrochemical energy storage output judgment condition to avoid overcharging of the electrochemical energy storage. If the SOC is not exceeded, the electrochemical energy storage is charged. If it exceeds the limit, it is directly discarded, as shown in formula (18):

[0046]

[0047] is the part of power discarded at time t.

[0048] In step 3, specifically:

[0049] Step 3.1, establish the objective function;

[0050] 1. Best economic performance

[0051] The economic cost of the hybrid electrolysis energy storage system mainly includes the construction cost and the cost incurred during system operation, as shown in formula (19);

[0052] min F=F c +F o +F g +F d (19)

[0053] Where: F is the total economic cost, F c is the construction cost, F o is the operation and maintenance cost, F g is the compensation cost of connecting to the grid for power shortage, F d is the cost of discarding excess renewable energy;

[0054] 2. Suppression effect indicators

[0055] The suppression effect index considers the processing of the hybrid energy storage system for the absorption and compensation components, including absorption and compensation, and calculates the grid compensation and discarded electric energy, thereby obtaining the efficiency of grid-connected suppression within one day, as shown in formula (28);

[0056]

[0057] Where: E is the suppression effect index, is the compensation component absorbed at time t, is the power that needs to be compensated or discarded at time t;

[0058] Step 3.2, constraints;

[0059] 1. Hydrogen energy storage constraints

[0060] 1) Electrolyzer Constraints

[0061] According to the analysis of the electrolytic cell working conditions and the simplified working mode of the alkaline electrolytic cell, the electrolytic cell power constraint is formulated as shown in equations (29) and (30):

[0062]

[0063] 2) Fuel cell constraints, as shown in equation (31):

[0064]

[0065] Where: P FCmin is the minimum power limit for fuel cell operation;

[0066] 3) Hydrogen storage tank constraints

[0067] SOH min ≤SOH t ≤SOH max (32)

[0068]

[0069] Where: SOH min , SOH max are the minimum and maximum limits of SOH, respectively. tis the remaining hydrogen content in the hydrogen storage tank at time t, SOH t-1 is the remaining hydrogen content in the hydrogen storage tank at time t-1, M HST is the maximum hydrogen storage capacity in the hydrogen storage tank, is the hydrogen production rate of the alkaline electrolyzer at time t, is the hydrogen production rate of the proton membrane electrolyzer at time t, is the rate at which the fuel cell consumes hydrogen at time t;

[0070] 2. Electrochemical energy storage constraints, as shown in equations (34) and (35):

[0071]

[0072] SOC min ≤SOC t ≤SOC max (35)

[0073] Where: SOC min , SOC max They are the minimum and maximum limits of electrochemical energy storage SOC respectively;

[0074] SOH t and SOH t-1 is the electrochemical energy storage charge state at time t and t-1, as shown in formula (36);

[0075]

[0076] μ is the charge and discharge coefficient, as shown in formula (37);

[0077]

[0078] e BAT is the charge and discharge efficiency of electrochemical energy storage;

[0079] In step 3.1, the construction cost and operation and maintenance cost both include hydrogen energy storage and electrochemical energy storage, as shown in equations (20) and (21); the hydrogen energy storage part consists of two electrolyzers, fuel cells and hydrogen storage tanks, as shown in equations (22) and (23);

[0080] F c =F h,con +F bat,con (20);

[0081] F o =F h,ope +F bat,ope (twenty one);

[0082]

[0083] Where: F h,con 、F bat,con are the construction costs of hydrogen energy storage and electrochemical energy storage, respectively, h,ope 、F bat,ope are the operation and maintenance costs of hydrogen energy storage and electrochemical energy storage, P AEL,R , P PEM,R , P FC,R 、V HST,R , P BAT,R 、E BAT They are the rated power and capacity of alkaline electrolyzer, proton membrane electrolyzer, fuel cell, hydrogen storage tank, and electrochemical energy storage, respectively. AEL , K PEM , K FC , K HST are the power and capacity investment coefficients of the alkaline electrolyzer, proton membrane electrolyzer, fuel cell, and hydrogen storage tank respectively; the investment recovery coefficient of the system equipment is calculated based on the discount rate r and the system operation time y;

[0084] The operation and maintenance cost of the energy storage system is converted according to the proportion of the investment cost, as shown in equations (24) and (25);

[0085] F h,ope =α h F h,con (twenty four);

[0086] F bat,ope =α bat F bat,con (25);

[0087] Compensation cost for additional grid connection due to power shortage F g , as shown in formula (26);

[0088]

[0089] Cost of discarding excess renewable energy F d , as shown in formula (27);

[0090]

[0091] The beneficial effects of the present invention are as follows: the method of the present invention is based on a system architecture in which hydrogen energy storage is the main component and electrochemical energy storage is the auxiliary component. In order to enhance the applicability of the reconstruction method, the source-side output changes and load levels are considered, Coarse-Grained and Fine-Grained functions are established, and a wind-solar power decomposition method based on time-of-use electricity prices under a multi-granularity mechanism is proposed to achieve reconstruction at coarse and fine granularities, thereby obtaining multi-granularity grid-connected reconstruction limit conditions; secondly, the working characteristics of alkaline electrolyzers and proton membrane electrolyzers are analyzed, and a high-proportion hydrogen storage power allocation strategy considering multiple electrolysis types is proposed to increase the proportion of hydrogen energy storage in the energy storage system and reduce dependence on electrochemical energy storage; thereafter, considering the two indicators of economy and smoothing effect, a capacity optimization configuration model for a hybrid energy storage system with the goal of smoothing wind-solar fluctuations is established; compared with the existing power allocation strategy, the capacity configuration method of the present invention reduces the electrochemical energy storage capacity and power while improving the smoothing effect, thereby effectively improving the operating economy of the energy storage system. Moreover, during the operation of the system, the electrochemical energy storage is slightly charged and discharged, and the overall change in hydrogen storage content is small, thus playing the main role of hydrogen energy storage in absorption and compensation. BRIEF DESCRIPTION OF THE DRAWINGS

[0092] Figure 1 It is a flow chart of the energy storage system optimization configuration method of the present invention.

[0093] Figure 2 It is a diagram of the architecture of the hybrid electrolysis energy storage system of the present invention;

[0094] Figure 3 It is a diagram of the multi-granularity wind and solar power decomposition and reconstruction process in the method of the present invention;

[0095] Figure 4 It is a diagram of the calculation process of the fine-grained layer coefficient in the method of the present invention;

[0096] Figure 5 It is a diagram of the calculation process of the coarse-grained layer in the method of the present invention;

[0097] Figure 6 It is a diagram of power compensation allocation strategy for high-proportion hydrogen storage under multiple electrolysis types;

[0098] Figure 7 It is a diagram of the power consumption allocation strategy for high-proportion hydrogen storage under multiple electrolysis types;

[0099] Figure 8 This is a typical daily effort curve;

[0100] Fig. 9 This is the decomposition result diagram at different time scales in method one;

[0101] Fig.10 This is the decomposition result diagram at different time scales in method 2;

[0102] Fig.11a This is the decomposition result diagram at the 1-min scale in method three;

[0103] Fig.11b This is the decomposition result diagram at the 10-min scale in method three;

[0104] Fig.12a It is the grid connection and consumption compensation component diagram after decomposition and reconstruction in method one;

[0105] Figure 12b It is the grid connection and consumption compensation component diagram after decomposition and reconstruction in method 2;

[0106] Fig.12c It is the grid connection and consumption compensation component diagram after decomposition and reconstruction in method three;

[0107] Fig.13a This is a comparison chart before and after the consumption compensation in Scheme 2;

[0108] Fig.13b This is a comparison chart before and after the consumption compensation in Scheme 3;

[0109] Fig.14a This is a comparison chart of the changes in SOH and SOC in Scheme 2;

[0110] Fig.14b This is a comparison chart of the changes in SOH and SOC in Scheme 3;

[0111] Fig.15a This is a comparison chart of hydrogen energy storage output and SOH changes in Option 2;

[0112] Fig.15b This is a comparison chart of hydrogen energy storage output and SOH changes in Option 3;

[0113] Fig.16a This is a comparison chart of the output of each part of hydrogen energy storage in Option 2;

[0114] Fig.16b This is a comparison chart of the output of each part of hydrogen energy storage in Option 3;

[0115] Fig.17a This is a comparison chart of electrochemical energy storage output and SOC changes in Scheme 2;

[0116] Fig.17b This is a comparison chart of electrochemical energy storage output and SOC changes in Scheme 3. DETAILED DESCRIPTION

[0117] The present invention is described in detail below with reference to the accompanying drawings and specific embodiments.

[0118] The energy storage system optimization configuration method based on the fluctuation suppression strategy under the multi-granularity mechanism of the present invention is as follows: Figure 1As shown, based on the establishment of a hybrid electrolysis energy storage system, the architecture is as follows Figure 2 As shown; specifically:

[0119] Step 1: Under the multi-granularity mechanism, a wind-solar power decomposition and reconstruction method based on time-of-use electricity prices is designed. The wind-solar hybrid output power is decomposed and reconstructed to obtain the grid-connected component and the consumption compensation component, such as Figure 3 As shown, specifically:

[0120] Step 1.1: Set the photovoltaic output power P pv and wind power output P w Add them together to get the wind-solar hybrid power P M , as shown in formula (1);

[0121] P M =P pv +P w (1);

[0122] Step 1.2, calculate the decomposition and reconstruction standard score of the fine-grained layer, as shown in formula (2);

[0123] Score = Fine-Grained (PRICE, P pv ,P w ) (2)

[0124] PRICE is the time-of-use electricity price, which is divided into peak electricity price, valley electricity price and flat electricity price.

[0125] Time time <![CDATA[Electricity price / (yuan · kWh -1 )]]> Peak 09:00～16:00、19:00～23:00 0.83 Valley 01:00～08:00 0.50 Flat section Other time periods 0.66

[0126] like Figure 4 As shown, the standard score is normalized to obtain the reconstruction standard coefficient K t , as shown in formula (3);

[0127] K t =Min-Max(Score) (3)

[0128] Step 1.3: The wind farm access grid limit is divided into two time scales of 1 minute and 10 minutes, and the photovoltaic power station access grid limit is only one time scale of 1 minute, such as Figure 5 As shown, based on this, the Coarse-Grained function is established, as shown in formula (4) to formula (6);

[0129] [ΔD 1 ,ΔD 10 ]=Coarse-Grained(P T ) (4);

[0130]

[0131] ΔD 1 and ΔD 10 are the fluctuation limits of wind-solar hybrid power for 1 minute and 10 minutes respectively, P T It is the total installed capacity of renewable energy connected to the system, in MW.

[0132] Step 1.4: Based on step 1.2 and step 1.3, construct multi-granularity grid-connected reconstruction limit conditions and establish wind-solar hybrid power fluctuation limit conditions, as shown in equations (7) and (8);

[0133]

[0134] in, is the wind-solar hybrid power output at time t, and is the wind-solar hybrid power output at time t-1 and t-10;

[0135] Step 1.5: Decompose the wind-solar hybrid power in step 1.1 into various intrinsic modal components and residual components by using the empirical mode decomposition (EMD) method, and reconstruct it according to the hourly fluctuation constraint conditions established in step 1.4 to obtain the grid-connected component and the absorption compensation component.

[0136] Step 2: Design a high-proportion hydrogen storage power allocation strategy considering multiple electrolysis types, specifically:

[0137] Establishing a hydrogen energy storage power allocation strategy that takes into account multiple electrolysis types has the advantages of increasing the startup time of the electrolyzer, matching the working mode of multiple types of electrolysis, reducing the electrochemical energy storage capacity and proportion, and widening the working range of the electrolyzer.

[0138] The hydrogen energy storage power allocation strategy is divided into two parts:

[0139] 1) Hydrogen energy storage power compensation part

[0140] Hydrogen fuel cells are used as the main compensation output, and electrochemical energy storage is used as an auxiliary. There are two working states according to the content of the hydrogen storage tank. The specific power distribution is as follows: Figure 6 As shown;

[0141] Working mode 1, as shown in formula (11):

[0142] SOH<0.4 (11);

[0143] The hydrogen storage capacity SOH (State of Hydrogen) is too small. The fuel cell operates at the lowest power and consumes the least hydrogen, but still maintains its minimum operating requirements. It can no longer meet the compensation task requirements. Considering the power output of electrochemical energy storage, as shown in equations (12) and (13):

[0144]

[0145] It is the power output of the electrochemical energy storage at time t, which has positive and negative conditions, representing the charging and discharging states of the electrochemical energy storage in the entire energy storage system respectively; is the compensation power output of the external power grid at time t; is the compensation component at time t, P FCmin It is the minimum output power of the fuel cell;

[0146] There are two cases for the state of electrochemical energy storage capacity SOC. The judgment condition is set to 0.3 to avoid excessive discharge of electrochemical energy storage. If it is too high, electrochemical energy storage can be used for compensation, as shown in formula (12); if it is too low, an external power grid is used to compensate for the power shortage, as shown in formula (13); at the same time, the fuel cell can also extend the operating time to avoid excess energy consumption and life degradation caused by starting and stopping.

[0147] Working mode 2: As shown in formula (14):

[0148] SOH ≥ 0.4 (14);

[0149] When there is sufficient hydrogen reserve and the fuel cell operates at rated power, if the electrochemical energy storage SOC is low, the excess will be used for electrochemical charging to maintain the energy storage state of the electrochemical energy storage and prepare good power storage to cope with the extreme conditions of the hybrid energy storage system; otherwise, the fuel cell operating power will be updated and adjusted.

[0150] 2) Hydrogen energy storage power consumption

[0151] Relying on two types of electrolyzers to absorb excess renewable energy power generation, establish an allocation strategy based on the hydrogen storage content SOH and the working status of the alkaline electrolyzer, give full play to the advantages of hydrogen energy storage, use the characteristics of each electrolyzer, and achieve the power consumption target in an economical way. The specific strategies are as follows: Figure 7 As shown;

[0152] Working mode 3: As shown in formula (15):

[0153] SOH≤0.6 (15);

[0154] When the remaining hydrogen storage SOH is slightly low, AEL and PEM output at the same time. According to the size of the power required to be consumed, the output of AEL is divided into two types: half load and full load. While stabilizing the output of AEL, PEM supplements the output. If the hydrogen storage content exceeds 0.6, the electrolyzer needs to be shut down and the electrochemical energy storage is used to complete the consumption. The power output of the proton membrane electrolyzer is shown in formula (16):

[0155]

[0156] is the power output of the proton membrane electrolyzer at time t; is the working power of the alkaline electrolyzer at time t;

[0157] This working mode not only simplifies the output mode of AEL, but also operates with two electrolyzers in combination, which has stronger adjustment ability, faster dynamic response speed, wider operating range, and also has a positive impact on the healthy operation status of the electrolyzer.

[0158] Working mode 4: When power is needed, the two electrolyzers are given a start-up signal. The first cold start time of AEL is 60 minutes, while that of PEM is only 5 minutes. During the electrolyzer start-up time, the power is mainly consumed by the electrochemical energy storage output. When PEM is available but AEL is not, PEM works alone, as shown in formula (17):

[0159] SOC≤0.75 (17);

[0160] As shown in formula (17), 0.75 is selected as the electrochemical energy storage output judgment condition to avoid overcharging of the electrochemical energy storage. If the SOC is not over the limit, the electrochemical energy storage is charged. If it exceeds the limit, it is directly discarded, as shown in formula (18):

[0161]

[0162] is the part of power discarded at time t;

[0163] The hydrogen energy storage power allocation strategy established by the present invention takes into account multiple electrolysis types. The absorption task is mainly undertaken by multiple types of electrolyzers, which overall reduces the electrochemical energy storage capacity and usage scenarios, enhances the absorption of hydrogen storage, and increases the proportion of hydrogen energy storage in the hybrid energy storage system. The compensation task uses electrochemical energy storage to compensate for the charging of fuel cells, extend their operating time, and reduce the life impact caused by power fluctuations.

[0164] Step 3: Establish a capacity optimization configuration model for the wind-solar hybrid energy storage system with the minimum total economic cost and the best suppression effect as the objective function, specifically:

[0165] Step 3.1, establish the objective function;

[0166] 1. Best economic performance

[0167] The economic cost of the hybrid electrolysis energy storage system mainly includes the construction cost and the cost incurred during system operation, as shown in formula (19);

[0168] min F=F c +F o +Fg +F d (19)

[0169] Where: F is the total economic cost, F c is the construction investment cost, F o is the operation and maintenance cost, F g is the compensation cost of connecting to the grid for power shortage, F d is the cost of discarding excess renewable energy;

[0170] Both the construction cost and the operation and maintenance cost include hydrogen energy storage and electrochemical energy storage, as shown in equations (20) and (21); the hydrogen energy storage part consists of two electrolyzers, fuel cells and hydrogen storage tanks, as shown in equations (22) and (23);

[0171] F c =F h,con +F bat,con (20);

[0172] F o =F h,ope +F bat,ope (twenty one);

[0173]

[0174] Where: F h,con 、F bat,con are the investment and construction costs of hydrogen energy storage and electrochemical energy storage, respectively. h,ope 、F bat,ope are the operation and maintenance costs of hydrogen energy storage and electrochemical energy storage, P AEL,R , P PEM,R , P FC,R 、V HST,R , P BAT,R 、E BAT They are the rated power and capacity of alkaline electrolyzer, proton membrane electrolyzer, fuel cell, hydrogen storage tank, and electrochemical energy storage, respectively. AEL , K PEM , K FC , K HST are the power and capacity investment coefficients of alkaline electrolyzer, proton membrane electrolyzer, fuel cell and hydrogen storage tank respectively. Based on the discount rate r and system operation time y, the investment recovery coefficient of system equipment is calculated.

[0175] The operation and maintenance cost of the energy storage system is converted according to the proportion of the investment cost, as shown in equations (24) and (25);

[0176] F h,ope =α h F h,con (twenty four);

[0177] F bat,ope =α bat F bat,con (25);

[0178] Where: α h is the conversion ratio of hydrogen energy storage, α bat is the conversion ratio of electrochemical energy storage;

[0179] Compensation cost for additional grid connection due to power shortage F g , as shown in formula (26);

[0180]

[0181] Where: K GRID is the grid compensation cost coefficient, is the power shortage compensation at the grid end at time t, and Δt is the time interval;

[0182] Cost of discarding excess renewable energy F d , as shown in formula (27);

[0183]

[0184] Where: K discard is the power abandonment penalty coefficient, is the power discarded by the system at time t;

[0185] 2. Suppression effect indicators

[0186] The suppression effect index considers the processing of the hybrid energy storage system for the absorption and compensation components, including absorption and compensation, and calculates the grid compensation and discarded electric energy, thereby obtaining the efficiency of grid-connected suppression within one day, as shown in formula (28);

[0187]

[0188] Where: E is the suppression effect index, is the compensation component absorbed at time t, is the power that needs to be compensated or discarded at time t.

[0189] Step 3.2, constraints;

[0190] 1. Hydrogen energy storage constraints

[0191] 1) Electrolyzer Constraints

[0192] According to the analysis of the electrolytic cell working conditions and the simplified working mode of the alkaline electrolytic cell, the electrolytic cell power constraint is formulated as shown in equations (29) and (30):

[0193]

[0194] 2) Fuel cell constraints, as shown in equation (31):

[0195]

[0196] Where: P FCmin It is the minimum power limit for fuel cell operation.

[0197] 3) Hydrogen storage tank constraints

[0198] SOH min ≤SOH t ≤SOH max (32)

[0199]

[0200] Where: SOH min , SOH max are the minimum and maximum limits of SOH, respectively. t is the remaining hydrogen content in the hydrogen storage tank at time t, SOH t-1 is the remaining hydrogen content in the hydrogen storage tank at time t-1, M HST is the maximum hydrogen storage capacity in the hydrogen storage tank, is the hydrogen production rate of the alkaline electrolyzer at time t, is the hydrogen production rate of the proton membrane electrolyzer at time t, is the rate at which the fuel cell consumes hydrogen at time t.

[0201] 2. Electrochemical energy storage constraints, as shown in equations (34) and (35):

[0202]

[0203] SOC min ≤SOC t ≤SOC max (35)

[0204] Where: SOC min , SOC max They are the minimum and maximum limits of electrochemical energy storage SOC respectively;

[0205] SOH t and SOH t-1 is the electrochemical energy storage charge state at time t and t-1, as shown in formula (36);

[0206]

[0207] μ is the charge and discharge coefficient, as shown in formula (37);

[0208]

[0209] e BAT It is the charge and discharge efficiency of electrochemical energy storage.

[0210] The capacity configuration model takes into account the two objective functions of economy and stabilization effect, and establishes constraints based on the hydrogen energy storage and electrochemical energy storage in the hybrid energy storage system, so as to determine the optimal capacity of alkaline electrolyzer, proton membrane electrolyzer, fuel cell, hydrogen storage tank and electrochemical energy storage.

[0211] Step 4: Input the consumption compensation component obtained in step 1 and the power allocation strategy obtained in step 2 into the capacity optimization configuration model of the wind-solar hybrid energy storage system, and use the NSGA-II algorithm to solve the model to obtain the rated power of the two electrolyzers, the rated power of the fuel cell, the rated capacity of the hydrogen storage tank, the rated power and rated capacity of the electrochemical energy storage, the economic cost and the smoothing effect.

[0212] Example 1

[0213] The energy storage system optimization configuration method based on the fluctuation suppression strategy under the multi-granularity mechanism of the present invention is specifically as follows:

[0214] Step 1: Using the original regional output data, the K-means clustering method is used to obtain five typical days, and decomposition and reconstruction are performed based on the typical days. Compared with the existing methods, the method of the present invention has superiority.

[0215] According to the actual annual wind power and photovoltaic output data of a certain place in Northwest my country, the corresponding hybrid electrolysis energy storage system capacity is equipped. The total installed capacity of wind power and photovoltaic power is 350MW, and the parameters required for the configuration model are shown in Table 1.

[0216] Table 1 Parameters required for configuring the model

[0217] parameter value Electrochemical energy storage power cost coefficient / (10,000 yuan / MW) 270 Electrochemical energy storage capacity cost coefficient / (10,000 yuan / MWh) 64 Power cost coefficient of alkaline electrolyzer / (10,000 yuan / MW) 300 Proton membrane electrolyzer power cost coefficient / (10,000 yuan / MW) 450 Fuel cell power cost coefficient / (10,000 yuan / MW) 500 <![CDATA[Investment cost coefficient of hydrogen storage tank / (10,000 yuan / m 3 )]]> 1.5 Hydrogen energy storage equipment life cycle / year 25 Electrochemical energy storage life cycle / year 20 discount rate / % 5 Electrochemical energy storage operation and maintenance investment proportion / % 2 Hydrogen energy storage operation and maintenance investment proportion / % 2 Electrochemical energy storage charge and discharge efficiency / % 80 SOC Range 0.2～0.85 SOH Range 0.1～0.9

[0218] Taking the mixed power of wind power and photovoltaic power generation throughout the year as the object, the K-Means clustering method is used to analyze the power data and obtain five typical daily scenarios. The specific days and results are shown in Table 2. The typical daily power curve is shown in Figure 8 As shown;

[0219] Table 2 Corresponding day probabilities for five typical day scenarios

[0220] Clustering scenario 1 2 3 4 5 Corresponding days 90 70 86 58 61 Probability / % 24.6 19.2 23.6 15.9 16.7

[0221] Taking typical day 1 as an example, three methods are used to decompose the mixed power and compare their reconstruction effects:

[0222] Method 1: EMD power decomposition and reconstruction method.

[0223] Method 2: A decomposition and reconstruction method based on method 1 that considers the fluctuation characteristics of wind and solar power in different periods of time.

[0224] Method 3: The wind-solar power decomposition method based on time-of-use electricity price under the multi-granularity mechanism proposed in the present invention.

[0225] Fig. 9 The calculation results obtained by method 1 are based on the reconstruction standard of the fluctuation limit within 1 minute and 10 minutes. The reconstruction results under two different time scales are integrated, and finally the low-frequency sixth-order reconstruction component is selected for direct grid connection. Fig.12a It is the final result of decomposition and reconstruction of method one.

[0226] The reconstruction standard is further refined to the hourly time scale, and the reconstruction order that meets the grid connection standard is selected based on the comparison between the maximum fluctuation in each time period and the grid connection fluctuation limit, and the grid connection and consumption compensation components obtained by method 2 are obtained. Compared with method 1, Figure 12b Method 2 does not require the energy storage system to smooth out the load at night and in the early morning, but it does require smoothing during load peaks, which greatly reduces the output of the energy storage system.

[0227] On the basis of method 2, time-of-use electricity price and output level are added to quantify the reconstruction standard, that is, Fig.12c Method 3: Wind-solar power decomposition method based on time-of-use electricity price under multi-granularity mechanism, which updates the existing method based on multi-granularity mechanism, and its reconstruction result is more improved than before the update;

[0228] Fig.11a This is the result of the 1-minute limit decomposition and reconstruction of method three. The seventh-order low-frequency grid-connected component is completely below the fluctuation limit, while the eighth-order low-frequency grid-connected component exceeds the limit in several time periods such as 9, 11, and 19. Method three further improves the grid-connected standard for load peaks, enhances the smoothing effect in the exceeded period, and increases the output task of the energy storage system. Therefore, it is too rough to reconstruct the grid-connected component by comparing the maximum fluctuation amount in one day according to method one, and method two is too absolute, resulting in insufficient utilization of the energy storage system. Frequent shutdowns will also reduce the life of the system, and the load level cannot be reasonably evaluated. There are situations where efficiency is reduced and complexity is increased. Method three is based on the standard limit of multi-period grid connection. The eighth-order low-frequency grid-connected component maintains the standard limit in the load valley section and increases the grid-connected limit in the load peak section. The output of new energy is no longer aimed at the maximum output, so that the fluctuation of wind and solar is minimized, and more new energy output is allocated to the consumption compensation component to achieve reasonable use of energy. Fig.11bThe 10-minute limit decomposition and reconstruction result diagram in Method 3 reflects the same problem. According to the standard of Method 1, the sixth-order low-frequency grid-connected components are all below the limit, but the seventh-order low-frequency grid-connected components only exceed the limit in two periods. From the comparison chart, it can be found that the fluctuations reflected by the same sixth-order low-frequency grid-connected components at the 1-minute and 10-minute time scales are quite different. The former meets the limit requirements, while the latter exceeds the limit at some moments, which fully reflects the uncertainty and randomness of the output of new energy.

[0229] Taking into account the reconstruction results of 1min and 10min, the optimal grid-connected component order is selected in each period. Through comparison, it is found that under the premise of meeting the regulations for access to the power system, the reconstruction effect of method 3 is significantly better than that of method 1. Its absorption and compensation task is reduced by half, focusing on the absorption and compensation component; compared with method 2, the number of system shutdowns is reduced and the system absorption range is widened. Although the grid-connected component decomposed by method 1 does not change frequently, the grid-connected component and the absorption and compensation component are mutually exclusive. Therefore, it not only has a large amount of absorption and compensation tasks, but also a long system operation time. The multi-granularity mechanism uses time-of-use electricity prices to consider the load demand and energy output in different periods, and combines coarse and fine granularity, which is better than the result of decomposition and reconstruction using the empirical mode decomposition method.

[0230] Example 2

[0231] Step 2: Based on the capacity configuration model proposed by the present invention and established by the high-proportion hydrogen storage power allocation strategy under multiple electrolysis types, the annual consumption compensation component is input and the optimization model is solved to obtain the optimal capacity configuration plan for each part of the energy storage system.

[0232] The NSGA-II algorithm is used to solve the capacity configuration results. In order to highlight the superiority of the power allocation strategy proposed in the present invention, the hybrid electrolysis energy storage system proposed in the present invention, the single hydrogen energy storage system, and the mainstream hybrid energy storage system are mainly considered for comparison.

[0233] Option 1: Single hydrogen energy storage.

[0234] Option 2: A hybrid energy storage system based on a single type of alkaline electrolyzer and electrochemical energy storage.

[0235] Solution 3: A hybrid energy storage system based on the power allocation strategy proposed in the present invention, which uses an alkaline electrolyzer and a proton membrane electrolyzer for operation.

[0236] Table 3 Final capacity solution for configuration optimization

[0237]

[0238]

[0239] The optimal solution was obtained from the Pareto solution set using the entropy weight method. The weights of economic cost and smoothing effect were set to 0.62 and 0.38 respectively. The final capacity solution results of the configuration optimization are shown in Table 3. Compared with Option 1 with only hydrogen energy storage, the delay in the start-up of hydrogen energy storage and the inability to cope with rapid power changes resulted in high comprehensive costs and poor smoothing effects. Compared with Option 2 with the same electric-hydrogen hybrid energy storage, Option 3 has a 7.38% higher smoothing effect index, and also reduces the comprehensive cost by 6.8428 million yuan, a decrease of 27.7%, and reduces the wind and solar abandonment and compensation costs by 67.8%, a decrease of 6.5354 million yuan.

[0240] The main advantage of the third scheme proposed in the present invention is that it considers the high-proportion hydrogen storage power allocation strategy under multiple electrolysis types, giving full play to the advantages of hydrogen energy storage, and electrochemistry only serves as a supplement. In the configuration results, compared with the second scheme, the electrochemical energy storage power is reduced by 40.17% and the capacity is reduced by 39.34%. However, the second scheme relies too much on electrochemical energy storage, and the role of hydrogen energy storage is not brought into play. With a small difference in investment cost, the power allocation strategy designed by the present invention plays a greater role in achieving economic stabilization of wind and solar grid connection.

[0241] Example 3

[0242] Step 3: Select a typical day to analyze the stabilization effect and compare it with the existing capacity configuration scheme to prove the superiority of the scheme proposed in the present invention.

[0243] Taking the historical data of wind and solar power on a typical day 1 as an example, the processing effects of the hybrid energy storage system absorption and compensation components of Scheme 2 and Scheme 3 are obtained. Fig.13a The solid line curve in the middle is the power curve that has not been processed by the hybrid energy storage system. After absorption and compensation, it can be seen that the component of the dotted line part has not been properly processed and is mainly compensated by the grid side. This is also the source of the high economic cost of this part. Fig.13b It is obtained by processing the scheme three of the present invention through the multi-electrolysis type power allocation strategy. Compared with scheme two, it can be seen that the dotted line part is greatly reduced, the compensation part and the abandoned part on the grid side are less than those in scheme two, and the percentage of the smoothing task increases from 89.79% to 96.25%.

[0244] In order to further refine the changes in the energy storage conditions of hydrogen energy storage and electrochemical energy storage, SOH and SOC are used to analyze them respectively, and the changes in hydrogen storage SOH and battery SOC within a day are obtained. Fig.14aThe SOH fluctuates between 0.38 and 0.52, and the change is not frequent. Finally, it stabilizes at about 0.4, which means that in the strategy of Option 2, the output of the alkaline electrolyzer is not much. Although hydrogen is still produced, the overall trend is downward. The hydrogen storage content gradually decreases, but the SOC trend continues to decline. The frequency of change is relatively frequent, down to 0.4. Relatively speaking, such a situation has a negative impact on the life of electrochemical energy storage, greatly reducing its long-term operation possibility. The electrochemical energy storage state is low, and it is difficult to cope with the subsequent compensation and consumption tasks, thereby generating more compensation and discarding costs, and the economic cost is further increased.

[0245] Fig.14b The SOH and SOC of the solution proposed in the present invention are compared. The overall trend of SOC is decreasing, with a variation range of 0.37 to 0.66, a small frequency of change, and basically within the normal variation range of about 0.4. This also reflects from the side that the electrochemical energy storage output is only an auxiliary energy storage function. Although SOH shows a fluctuating trend, it is still near the initial capacity of 0.5. In subsequent operations, the corresponding power can continue to be allocated to it based on the power allocation strategy, so as to perform hydrogen production stabilization operations and keep the hydrogen storage within a safe and economic range.

[0246] Comparing the hydrogen energy storage output and SOH changes, Fig.15a In the second scheme, the output time of the alkaline electrolyzer is very short, which makes it difficult to cope with the change of power. The alkaline electrolyzer has a long startup time, which brings certain difficulties to the continuous operation and response speed of the electrolyzer. Fig.15b In the case of scheme three of the present invention, the operation time of the electrolytic cell is greatly extended, and the number of shutdowns and starts of the electrolytic cell is reduced, making a major contribution to stabilizing the situation.

[0247] There are two main types of hydrogen energy storage output: electrolyzer and fuel cell. Within the two options, there are single-type electrolyzer and dual-type electrolyzer. Fig.16a The output of a single type of electrolyzer has the advantages of simplicity and low cost, but the disadvantage is that the power variation is large, and the power variation and frequent start and stop will cause the rapid decline of the electrolyzer life, resulting in a reduction in service time. The present invention introduces a proton membrane electrolyzer, and the working output is as follows Fig.16b As shown in the figure, although the construction cost increases, while reducing the capacity of the alkaline electrolyzer, the power allocation strategy simplifies the working mode, ensures the working life of the AEL, supplements the AEL power consumption gap when the consumption task is small and changes quickly, increases the system's consumption and hydrogen storage capacity, fuel cells and electrochemical energy storage output, maintains the low-power operation of the electrolyzer, and ensures the response speed of the electrolyzer. Electrochemical energy storage plays different roles in the two schemes. Fig.17aThe electrochemical energy storage output is frequent and plays a major role in the energy storage system. It can quickly respond to changes in power output, but the final SOC stabilizes at around 0.4. Such energy storage content cannot continue to rely on the electrochemical energy storage output. In subsequent work, the electrochemical energy storage may be unable to output, and hydrogen energy storage will not be able to cope with fluctuations, thereby affecting the compensation task under smoothing fluctuations, increasing the compensation cost on the grid side, and significantly reducing the smoothing effect.

[0248] The SOC of the third scheme proposed in the present invention is mainly between 0.37 and 0.66. Fig.17b As shown in the figure, the SOC changes are simple and stable, ensuring that the frequency of electrochemical energy storage changes is small. The main output is at the start-up time of the two electrolyzers. After the electrolyzers are successfully started, they play a role in power compensation. If there is an extreme situation with low hydrogen content, the electrochemical energy storage can compensate the output in time at the required time to ensure the stabilization effect, and the system's grid-connected capacity will not be affected in a short time.

Claims

1. An energy storage system optimization configuration method based on a fluctuation suppression strategy under a multi-granularity mechanism, characterized in that: Specifically: Step 1: Under the multi-granularity mechanism, a wind-solar power decomposition and reconstruction method based on time-of-use electricity prices is designed, and the wind-solar hybrid output power is decomposed and reconstructed to obtain the grid-connected component and the consumption compensation component; Step 2: Design a high-proportion hydrogen storage power allocation strategy considering multiple electrolysis types; Step 3: Establish a capacity optimization configuration model for the wind-solar hybrid energy storage system with minimum total economic cost and optimal smoothing effect as the objective function; Step 4: Input the consumption compensation component obtained in step 1 and the power allocation strategy obtained in step 2 into the capacity optimization configuration model of the wind-solar hybrid energy storage system, and use the NSGA-II algorithm to solve the model to obtain the rated power of the two electrolyzers, the rated power of the fuel cell, the rated capacity of the hydrogen storage tank, the rated power and rated capacity of the electrochemical energy storage, the economic cost and the smoothing effect.

2. The energy storage system optimization configuration method based on the fluctuation suppression strategy under the multi-granularity mechanism according to claim 1 is characterized in that: In the step 1, specifically: Step 1.1: Set the photovoltaic output power P pv and wind power output P w Add them together to get the wind-solar hybrid power P M , as shown in formula (1); P M =P pv +P w (1); Step 1.2, calculate the decomposition and reconstruction standard score of the fine-grained layer, as shown in formula (2); Score=Fine-Grained(PRICE,P pv ,P w ) (2) PRICE is the time-of-use electricity price, which is divided into peak electricity price, valley electricity price and flat electricity price; Normalize the standard score to get the reconstruction standard coefficient K t , as shown in formula (3); K t =Min-Max(Score) (3) Step 1.3, the grid access limit of wind farms is divided into two time scales of 1 minute and 10 minutes, and the grid access limit of photovoltaic power stations is only 1 minute. Based on this, the Coarse-Grained function is established, as shown in equations (4) to (6); [ΔD1,ΔD 10 ]=Coarse-Grained(P T ) (4); ΔD1 and ΔD 10 are the fluctuation limits of wind-solar hybrid power for 1 minute and 10 minutes respectively, P T The total installed capacity of renewable energy connected to the system, in MW; Step 1.4: Based on step 1.2 and step 1.3, construct multi-granularity grid-connected reconstruction limit conditions and establish wind-solar hybrid power fluctuation limit conditions, as shown in equations (7) and (8); in, is the wind-solar hybrid power output at time t, and is the wind-solar hybrid power output at time t-1 and t-10; Step 1.5: Decompose the wind-solar hybrid power in step 1.1 into various intrinsic modal components and residual components by using the empirical mode decomposition (EMD) method, and reconstruct it according to the hourly fluctuation constraint conditions established in step 1.4 to obtain the grid-connected component and the absorption compensation component.

3. The energy storage system optimization configuration method based on the fluctuation suppression strategy under the multi-granularity mechanism according to claim 2 is characterized in that: In step 2, specifically: the high proportion hydrogen storage power allocation strategy is divided into two parts: 1) Hydrogen energy storage power compensation part Hydrogen fuel cells are used as the main compensation output, and electrochemical energy storage is used as an auxiliary. There are two working states according to the content of the hydrogen storage tank; Working mode 1, as shown in formula (11): SOH < 0.4 (11); Consider the power output of electrochemical energy storage, as shown in equations (12) and (13): It is the power output of the electrochemical energy storage at time t, which has positive and negative conditions, representing the charging and discharging states of the electrochemical energy storage in the entire energy storage system respectively; is the compensation power output of the external power grid at time t; is the compensation component at time t, P FCmin It is the minimum output power of the fuel cell; Working mode 2: As shown in formula (14): SOH ≥ 0.4 (14); There is sufficient hydrogen storage and the fuel cell is running at rated power. If the electrochemical energy storage SOC is low, that is, SOC ≤ 0.3, the excess is charged for electrochemical energy storage, that is, On the contrary, the fuel cell operating power is updated and adjusted, that is, 2) Hydrogen energy storage power consumption Working mode 3: As shown in formula (15): SOH≤0.6 (15); If the hydrogen storage content exceeds 0.6, the electrolyzer needs to be shut down and the electrochemical energy storage is used to absorb the hydrogen. The power output of the proton membrane electrolyzer is shown in formula (16): is the power output of the proton membrane electrolyzer at time t; is the working power of the alkaline electrolyzer at time t; Working mode 4: When power is needed, the two electrolyzers are given a start-up signal. The first cold start time of AEL is 60 minutes, while that of PEM is only 5 minutes. During the electrolyzer start-up time, the power is mainly consumed by the electrochemical energy storage output. When PEM is available but AEL is not, PEM works alone, as shown in formula (17): SOC≤0.75 (17); 0.75 is selected as the electrochemical energy storage output judgment condition to avoid overcharging of the electrochemical energy storage. If the SOC is not exceeded, the electrochemical energy storage is charged. If it exceeds the limit, it is directly discarded, as shown in formula (18): is the part of power discarded at time t.

4. The energy storage system optimization configuration method based on the fluctuation suppression strategy under the multi-granularity mechanism according to claim 3 is characterized in that: In the step 3, specifically: Step 3.1, establish the objective function; Best economic performance The economic cost of the hybrid electrolysis energy storage system mainly includes the construction cost and the cost incurred during system operation, as shown in formula (19); minF=F c +F o +F g +F d (19) Where: F is the total economic cost, F c is the construction cost, F o is the operation and maintenance cost, F g is the compensation cost of connecting to the grid for power shortage, F d is the cost of discarding excess renewable energy; Suppression effect index The suppression effect index considers the processing of the hybrid energy storage system for the absorption and compensation components, including absorption and compensation, and calculates the grid compensation and discarded electric energy, thereby obtaining the efficiency of grid-connected suppression within one day, as shown in formula (28); Where: E is the suppression effect index, is the compensation component absorbed at time t, is the power that needs to be compensated or discarded at time t; Step 3.2, constraints; Hydrogen storage constraints 1) Electrolyzer Constraints According to the analysis of the electrolytic cell working conditions and the simplified working mode of the alkaline electrolytic cell, the electrolytic cell power constraint is formulated as shown in equations (29) and (30): 2) Fuel cell constraints, as shown in equation (31): Where: P FCmin is the minimum power limit for fuel cell operation; is the power output of the proton membrane electrolyzer at time t; is the working power of the alkaline electrolytic cell at time t; P AEL,R is the rated power of the alkaline electrolyzer, P FC,R is the rated power of the fuel cell, P PEM,R is the rated power of the proton membrane electrolyzer, 3) Hydrogen storage tank constraints SOH min ≤SOH t ≤SOH max (32) Where: SOH min , SOH max are the minimum and maximum limits of SOH, respectively. t is the remaining hydrogen content in the hydrogen storage tank at time t, SOH t-1 is the remaining hydrogen content in the hydrogen storage tank at time t-1, M HST is the maximum hydrogen storage capacity in the hydrogen storage tank, is the hydrogen production rate of the alkaline electrolyzer at time t, is the hydrogen production rate of the proton membrane electrolyzer at time t, is the rate at which the fuel cell consumes hydrogen at time t; The constraints of electrochemical energy storage are shown in equations (34) and (35): SOC min ≤SOC t ≤SOC max (35) Where: SOC min , SOC max are the minimum and maximum limits of electrochemical energy storage SOC respectively; P BAT,R is the rated power of the electrochemical energy storage; SOC t and SOC t-1 is the electrochemical energy storage charge state at time t and t-1, as shown in formula (36); E BAT is the rated capacity of the electrochemical energy storage; μ is the charge and discharge coefficient, as shown in formula (37); e BAT is the charge and discharge efficiency of electrochemical energy storage; The capacity configuration model of the wind-solar hybrid energy storage system takes into account the two objective functions of economy and smoothing effect, and establishes constraints based on the hydrogen energy storage and electrochemical energy storage in the hybrid energy storage system, so as to determine the optimal capacity of the alkaline electrolyzer, proton membrane electrolyzer, fuel cell, hydrogen storage tank and electrochemical energy storage.

5. The method for optimizing the configuration of an energy storage system based on a fluctuation suppression strategy under a multi-granularity mechanism according to claim 4, characterized in that: In step 3.1, the construction cost and operation and maintenance cost both include hydrogen energy storage and electrochemical energy storage, as shown in formula (20) and formula (21); the hydrogen energy storage part is composed of two electrolyzers, a fuel cell and a hydrogen storage tank, as shown in formula (22) and formula (23); F c =F h,con +F bat,con (20); F o =F h,ope +F bat,ope (21); Where: F h,con 、F bat,con are the construction costs of hydrogen energy storage and electrochemical energy storage, respectively, h,ope 、F bat,ope are the operation and maintenance costs of hydrogen energy storage and electrochemical energy storage, P AEL,R is the rated power of the alkaline electrolyzer, P PEM,R is the rated power of the proton membrane electrolyzer, P FC,R is the rated power of the fuel cell, V HST,R is the rated capacity of the hydrogen storage tank, P BAT,R is the rated power of electrochemical energy storage, E BAT is the rated capacity of electrochemical energy storage; K AEL is the power investment factor of the alkaline electrolyzer, K PEM is the power investment factor of the proton membrane electrolyzer, K FC is the power investment factor of the fuel cell, K HST is the capacity investment coefficient of the hydrogen storage tank; according to the discount rate r and the system operation time y, the investment recovery coefficient of the system equipment is calculated; The operation and maintenance cost of the energy storage system is converted according to the proportion of the investment cost, as shown in equations (24) and (25); F h,ope =a h F h,con (24); F bat,ope =a bat F bat,con (25); Where: α h is the conversion ratio of hydrogen energy storage, α bat is the conversion ratio of electrochemical energy storage; Compensation cost for additional grid connection due to power shortage F g , as shown in formula (26); Where: K GRID is the grid compensation cost coefficient, is the power shortage compensation at the grid end at time t, and Δt is the time interval; Cost of discarding excess renewable energy F d , as shown in formula (27); Where: K discard is the power abandonment penalty coefficient, is the amount of power discarded by the system at time t.

Images