A method for assessing the operational life loss of a battery energy storage system based on the rotating door algorithm

Abstract

This invention relates to a method for assessing the operational lifespan loss of a battery energy storage system (BESS) based on the rotating gate algorithm, belonging to the field of power energy storage technology. The technical solution involves: improving the Whale Optimization Algorithm; using the improved Whale Optimization Algorithm to search for the globally optimal gate width of the rotating gate algorithm based on the state-of-charge (SOC) change curve of the battery cells; using the globally optimal gate width again, extracting feature data points from the SOC change curve of the battery cells using the rotating gate algorithm; and using these feature data points to obtain the depth of discharge of each battery cell in each half-cycle, thereby calculating the operational lifespan loss of the BESS. This invention can quickly and accurately assess the operational lifespan loss of a BESS, accelerating the assessment speed and reducing the time required for assessment. It combines accuracy and speed, promoting the better application of BESS in various aspects of the power system, and also providing a reference for BESS investment and control.

Description

Technical Field

[0001] This invention relates to a method for assessing the operational lifespan loss of a battery energy storage system based on the rotating door algorithm, belonging to the field of power energy storage technology. Background Technology

[0002] Resource scarcity and environmental pollution are two major challenges facing my country that urgently need to be addressed. To this end, the Chinese government has proposed a "dual-carbon" target, implemented a series of measures including building a new power system primarily based on new energy sources, and accelerating energy transition. Battery energy storage systems (BESS), due to their ability to store electrical energy and their rapid charging and discharging characteristics, can spatially or temporally shift electrical energy. They are widely used in various power system scenarios, such as reducing peak-valley load differences, regulating system frequency, smoothing power fluctuations from new energy sources, and black start, becoming a crucial and high-quality control resource in the power system.

[0003] However, BESS (Battery Safe Energy) has high investment costs and a limited lifespan, making it crucial to quickly and accurately assess its operational lifespan degradation. Currently, there are two main methods for assessing BESS operational lifespan degradation: one is based on battery health status, but this method has poor practicality as it requires shutting down the BESS, which can affect the safe operation of the entire power system and is therefore difficult to use in practical engineering; the second method uses rainflow counting to assess BESS operational lifespan degradation. While this method has some practical value, its complexity is high, resulting in a long assessment time, which is insufficient to meet the rapid requirements of power systems. Summary of the Invention

[0004] The purpose of this invention is to provide a method for assessing the operational life loss of a battery energy storage system (BESS) based on the rotating door algorithm. By using an improved Whale Optimization Algorithm and the rotating door algorithm, the method effectively extracts SOC feature data points from the SOC change curve of battery cells, enabling rapid and accurate assessment of BESS operational life loss. This method has significant practical value, accelerates the assessment speed, reduces the time required for operational life loss assessment, meets the speed requirements of power systems, and combines accuracy and speed. It can promote the better application of BESS in various aspects of power systems and also provides a reference for BESS investment and control, solving the aforementioned technical problems existing in existing technologies.

[0005] The technical solution of this invention is:

[0006] A method for assessing the operational lifespan loss of a battery energy storage system based on the rotating door algorithm includes the following steps:

[0007] (1) Improved beluga whale optimization (IBWO) algorithm uses Logistic chaotic initialization method and Tent chaotic initialization method to determine the initial candidate solution of the optimization individual to improve the diversity of the initial candidate solution; a balance factor calculation method is designed based on the Sigmoid function to improve the optimization algorithm's ability to balance between global exploration and local development.

[0008] (2) Design the value function for searching the globally optimal door width E in the Swing Door Trending (SDT) algorithm. Based on the SOC change curve of the battery cells in the battery energy storage system, the improved Whale Optimization Algorithm is used to search for the globally optimal door width E of the Swing Door Trending algorithm.

[0009] (3) Based on the global optimal gate width E, the rotating gate algorithm is used to process the SOC change curve of the battery cell in order to extract the feature data points in the SOC change curve.

[0010] (4) Based on the characteristic data points, the discharge depth of each battery cell in each half-cycle is obtained, and then the operating life loss of the battery energy storage system is calculated.

[0011] In step (1), the Logistic chaotic initialization method and the Tent chaotic initialization method are used to determine the initial candidate solutions of the optimization individual. The calculation formula for the initial candidate solutions of the optimization individual is as follows:

[0012] x i =x lb +α1Z Li ·(x ub -x lb )+α2Z Ti ·(x ub -x lb (1)

[0013] In the formula: x i Let Z be the initial candidate solution for the i-th optimization individual, α1 and α2 are chaotic weight coefficients with values ​​of 0.5 and 0.5 respectively. Li and Z Ti These are Logistic chaotic sequences and Tent chaotic sequences, respectively, x ub and x lb These are the upper and lower boundaries of the optimization variable, respectively.

[0014] In step (1), a balance factor B is designed based on the Sigmoid function. f The calculation method is as follows:

[0015]

[0016] In the formula: B0 is the initial value of the balance factor, and β is the balance factor B. f After numerous experiments, the adjusted parameters were determined, with β set to -0.03, and n and n max These are the current iteration number and the maximum iteration number, respectively.

[0017] In step (3), based on the optimization results of the global optimal gate width, the rotating gate algorithm is used to process the SOC operating curve of the battery cell, thereby extracting feature data points that characterize the changing trend of the SOC operating curve of the battery cell.

[0018] In step (4), the depth of discharge of each battery cell in each half-cycle is obtained based on the SOC characteristic data points, and then the operating life loss of the battery energy storage system is calculated. The calculation method for the operating life loss of the battery energy storage system is as follows:

[0019] The duration between two adjacent SOC feature data points is considered as one half-cycle, and the depth of charge / discharge is calculated:

[0020] D i =|S F,i+1 -S F,i | (3)

[0021] In the formula: D i S represents the depth of charge / discharge during the i-th half-cycle. F,i+1 and S F,i This represents two adjacent SOC feature data points within the i-th half-cycle period;

[0022] Calculate the depth of charge / discharge D during the i-th half-cycle. i The corresponding maximum number of loops N max (D i ):

[0023]

[0024] In the formula: a m (m∈0,1,2,3,4) are the polynomial fitting coefficients, with values ​​of 164000, -872500, 1827000, -1687000, and 572900, respectively.

[0025] Calculate the equivalent number of iterations within the i-th half-cycle:

[0026] N eq,i =0.5·N max (100%) / N max (D i (5)

[0027] Where: N max (100%) represents the maximum number of cycles per battery cell at 100% charge / discharge depth, N. eq,i The equivalent number of cycles lost by the battery cell during the i-th half-cycle period;

[0028] The equivalent number of cycles N lost by the battery energy storage system during the entire operation Bcy The calculation formula is as follows:

[0029]

[0030] Where: N B N represents the number of battery cells within the BESS. bcy,j Let N be the equivalent number of cycles lost by the j-th battery cell during operation. c N represents the number of half-cycles that a battery cell experiences during operation. eq,i,j The equivalent number of cycles lost by the j-th battery cell during the i-th half-cycle period;

[0031] The formula for calculating the percentage of battery energy storage system lifespan loss is as follows:

[0032] L B =N Bcy / N max (100%). (7)

[0033] This invention not only improves the traditional beluga whale optimization algorithm, but also uses the improved beluga whale optimization (IBWO) algorithm and the swing door trending (SDT) algorithm to extract the SOC feature data points in the state of charge (SOC) change curve of the battery cell. Based on this, a BESS service life loss assessment method is designed. Finally, the speed and accuracy of the method are verified by simulation.

[0034] The beneficial effects of this invention are as follows: By using the improved Whale Optimization Algorithm and Rotating Gate Algorithm, the SOC feature data points in the SOC change curve of the battery cell can be effectively extracted, enabling rapid and accurate assessment of BESS service life loss. This has significant practical value, while also accelerating the assessment speed of service life loss and reducing the time required for assessment, meeting the speed requirements of the power system. It combines accuracy and speed, promoting the better application of BESS in various aspects of the power system, and also providing a reference for BESS investment and control. Attached Figure Description

[0035] Figure 1This is a flowchart of an embodiment of the present invention;

[0036] Figure 2 This is a flowchart illustrating the optimization process of the improved beluga optimization algorithm according to an embodiment of the present invention.

[0037] Figure 3 This is a flowchart of the BESS service life loss evaluation based on SOC feature data points according to an embodiment of the present invention.

[0038] Figure 4 The result of extracting SOC feature data points for battery cell 1 in this embodiment of the invention;

[0039] Figure 5 The result of extracting SOC feature data points for battery cell 2 in this embodiment of the invention;

[0040] Figure 6 The results of SOC feature data point extraction for battery cell 3 in this embodiment of the invention;

[0041] Figure 7 The results of SOC feature data point extraction for battery cell 4 in this embodiment of the invention;

[0042] Figure 8 The results of SOC feature data point extraction for battery cell 5 in this embodiment of the invention;

[0043] Figure 9 The results of SOC feature data point extraction for battery cell 6 in this embodiment of the invention;

[0044] Figure 10 The results of SOC feature data point extraction for battery cell 7 in this embodiment of the invention;

[0045] Figure 11 The results of SOC feature data point extraction for battery cell 8 in this embodiment of the invention;

[0046] Figure 12 The results of SOC feature data point extraction for battery cell 9 in this embodiment of the invention;

[0047] Figure 13 This is the result of extracting SOC feature data points for battery cell 10 in an embodiment of the present invention. Detailed Implementation

[0048] The present invention will be further described below with reference to the accompanying drawings and embodiments.

[0049] A method for assessing the operational lifespan loss of a battery energy storage system based on the rotating door algorithm includes the following steps:

[0050] (1) Improved beluga whale optimization (IBWO) algorithm uses Logistic chaotic initialization method and Tent chaotic initialization method to determine the initial candidate solution of the optimization individual to improve the diversity of the initial candidate solution; a balance factor calculation method is designed based on the Sigmoid function to improve the optimization algorithm's ability to balance between global exploration and local development.

[0051] (2) Design the value function for searching the globally optimal door width E in the Swing Door Trending (SDT) algorithm. Based on the SOC change curve of the battery cells in the battery energy storage system, the improved Whale Optimization Algorithm is used to search for the globally optimal door width E of the Swing Door Trending algorithm.

[0052] (3) Based on the global optimal gate width E, the rotating gate algorithm is used to process the SOC change curve of the battery cell in order to extract the feature data points in the SOC change curve.

[0053] (4) Based on the characteristic data points, the discharge depth of each battery cell in each half-cycle is obtained, and then the operating life loss of the battery energy storage system is calculated.

[0054] This invention proposes a method for assessing the operational lifespan loss of a battery energy storage system based on the rotating door algorithm, with appended... Figure 1 The flowchart of an embodiment of the present invention includes the following detailed steps.

[0055] Step 1 designs an improved beluga optimization algorithm, using Logistic chaotic initialization and Tent chaotic initialization methods to determine the initial candidate solutions for the optimization individual, thereby improving the diversity of the initial candidate solutions; and designs a method for calculating the balance factor based on the Sigmoid function to improve the optimization algorithm's ability to balance global exploration and local exploitation.

[0056] The designed IBWO optimization flowchart is attached. Figure 2 As shown, the specific optimization process is as follows:

[0057] 1) Define the IBWO parameter

[0058] Set the population size M and the maximum number of iterations n for IBWO. max ;

[0059] 2) The Logistic chaotic initialization method and the Tent chaotic initialization method are used to determine the initial candidate solutions for the optimization individual;

[0060] The initial candidate solutions X = [x1, x2, ..., x] form the IBWO population. M ], where xi (i∈1,2,...M) represents the initial candidate solution for the i-th optimization individual, and its calculation formula is as follows:

[0061] x i =x lb +α1Z Li ·(x ub -x lb )+α2Z Ti ·(x ub -x lb (8)

[0062] In the formula: α1 and α2 are chaotic weighting coefficients, with values ​​of 0.5 and 0.5 respectively, x ub and x lb To find the upper and lower boundaries of the optimal variable, Z Li and Z Ti These are Logistic chaotic sequences and Tent chaotic sequences, respectively, Z Li and Z Ti The calculation formula is as follows:

[0063]

[0064] In the formula: is the Logistic chaos parameter, whose value is a random number in the range (0,4).

[0065]

[0066] In the formula: The Tent chaos parameter has a value that is a random number in the range (0, 0.5) ∪ (0.5, 1).

[0067] 3) Determine the optimization process

[0068] IBWO is based on the balance factor B f To determine the optimization process of an individual beluga whale, when B... f When B > 0.5, the beluga whale is in the overall exploration phase. f When the value is ≤0.5, the beluga whale is in a stage of partial development; the designed balance factor B f The calculation formula is as follows:

[0069]

[0070] In the formula: B0 is the initial value of the balance factor, and β is the balance factor B. f After several experiments, the value of β was set to -0.03, and n was the current iteration number.

[0071] 4) Global exploration process

[0072] The method for updating the location of individual beluga whales during global exploration is as follows:

[0073]

[0074] In the formula: x i,j (n+1) represents the updated position of the i-th beluga whale in the j-dimensional space; p j (j = 1, 2, ..., D) are randomly selected integers; For the i-th beluga whale in p j Position in 3D space; Let r be the position of the r-th beluga whale in p1-dimensional space (r is a random integer between 1 and N); r1 and r2 are random numbers, both ranging from (0,1).

[0075] 5) Local Development Process

[0076] The method for updating the location of individual beluga whales during partial development is as follows:

[0077] x i (n+1)=r3x b (n)-r4x i (n)+c f ·L f ·(x r (n)-x i (n)) (13)

[0078] In the formula: x i (n+1) represents the updated position of the i-th beluga whale; r3 and r4 are random numbers, both ranging from (0,1); x b (n) represents the globally optimal solution for all beluga whale individuals in the nth iteration; x i (n) and x r (n) represents the positions of the i-th beluga whale and the randomly selected r-th beluga whale during the nth iteration, respectively; c f and L f These are the random jump strength and the Levy flight function, which measure the Levy flight force, respectively, and their calculation formulas are as follows:

[0079] c f =2r4·(1-n / n) max (14)

[0080] L f =0.05·u·σ / υ 1 / β (15)

[0081]

[0082] In the formula: u and υ are random numbers that conform to a normal distribution; β1 is a constant, which is taken as 1.5 in this study.

[0083] 6) The process of the beluga whale falling

[0084] The beluga whale's position during the fall was calculated as follows:

[0085] x i (n+1)=r5x i (n)-r6x r (n)+r7σ s (17)

[0086] In the formula: r5, r6, and r7 are random numbers, and their values ​​are all in the range (0,1); σ s The formula for calculating the step length of a beluga whale when it falls is as follows:

[0087]

[0088] In the formula: u b l b c1 and c2 represent the upper and lower boundaries of the optimization variable, respectively; c2 represents the step factor, the value of which is related to the probability of falling behavior and the population size, and is calculated as follows:

[0089] c2 = 2W f ·N (19)

[0090] In the formula: W f The probability of a beluga whale falling to its death is expressed by the following formula:

[0091] W f =0.1-0.05·n / n max (20)

[0092] Step 2: Design the value function for searching the globally optimal gate width E in the rotating door algorithm. Based on the SOC change curve of the battery cells in the battery energy storage system, the improved white whale optimization algorithm is used to search for the globally optimal gate width E of the rotating door algorithm.

[0093] The calculation steps of the revolving door algorithm are as follows:

[0094] 1) Initialization

[0095]

[0096] In the formula: t0 and x0 are the initial time and their corresponding data values, respectively; t1 and x1 are the first time and their corresponding data values, respectively; k 1d and k 2d These are the initial values ​​of the slopes of the upper and lower fulcrum gates, respectively; E is the compression offset.

[0097] 2) Calculate the slope

[0098]

[0099] In the formula: t j and x j These represent the j-th time point and its corresponding data value; t k and x k These represent the k-th time point and its corresponding data value;

[0100] 3) Slope Update

[0101]

[0102] 4) Data Extraction

[0103] k 1d ≥k 2d (24)

[0104] If equation (9) is satisfied, then the previous time t will be... j-1 Data value x j-1 Record it as feature data and return to step 2); otherwise return to step 3.

[0105] The value function for searching the globally optimal door width E in the revolving door algorithm is as follows:

[0106]

[0107] In the formula: λ1 and λ2 are weights, with values ​​of 10 and 1.75 respectively; N1 and N2 are the number of SOC data points of the battery cell and the number of SOC feature data points extracted by the SDT algorithm, respectively; s i For SOC data values, y i This represents the SOC trend value after linear interpolation of the SOC feature data points.

[0108] Step 3: Based on the improved beluga optimization algorithm, the globally optimal gate width E of the rotating door algorithm is found. The rotating door algorithm is then used again to process the SOC change curve of the battery cell, thereby extracting the feature data points in the SOC change curve.

[0109] Step 4: Based on the feature data points, obtain the depth of discharge of each battery cell in each half-cycle, and then calculate the operating life loss of the battery energy storage system. The process for calculating the operating life loss of BESS based on the SOC feature data points is attached. Figure 3 As shown, the detailed calculation process is as follows:

[0110] 1) Consider the time between two adjacent SOC feature data points as one half-cycle, and calculate the depth of charge / discharge:

[0111] Di =|S F,i+1 -S F,i | (26)

[0112] In the formula: D i S represents the depth of charge / discharge during the i-th half-cycle. F,i+1 and S F,i This represents two adjacent SOC feature data points within the i-th half-cycle period;

[0113] 2) Calculate the depth of charge / discharge D during the i-th half-cycle. i The corresponding maximum number of loops N max (D i ):

[0114]

[0115] In the formula: a m (m∈0,1,2,3,4) are the polynomial fitting coefficients, with values ​​of 164000, -872500, 1827000, -1687000, and 572900, respectively.

[0116] 3) Calculate the equivalent number of iterations within the i-th half-cycle:

[0117] N eq,i =0.5·N max (100%) / N max (D i (28)

[0118] Where: N max (100%) represents the maximum number of cycles per battery cell at 100% charge / discharge depth, N. eq,i The equivalent number of cycles lost by the battery cell during the i-th half-cycle period;

[0119] 4) The equivalent number of cycles N lost by the battery energy storage system during the entire operation process. Bcy The calculation formula is as follows:

[0120]

[0121] Where: N B N represents the number of battery cells within the BESS. bcy,j Let N be the equivalent number of cycles lost by the j-th battery cell during operation. c N represents the number of half-cycles that a battery cell experiences during operation. eq,i,j The equivalent number of cycles lost by the j-th battery cell during the i-th half-cycle period;

[0122] 5) The formula for calculating the percentage of battery energy storage system lifespan loss is as follows:

[0123] L B =N Bcy / N max (100%) (30)

[0124] To further understand this invention and verify the effectiveness of the proposed method for assessing the operational lifespan loss of a battery energy storage system based on the rotating door algorithm, simulations were performed using the SOC change curves of the battery cells in the battery energy storage system on a given day. The battery energy storage system has a scale of 10MW / 10MWh and contains 10 battery cells, each with a scale of 1MW / 1MWh.

[0125] Based on the SOC variation curves of 10 battery cells, the IBWO and SDT algorithms were used to extract the SOC feature data points for each battery cell. The extraction results of the feature data points for the 10 battery cells are shown in the attached figure. Figure 4-13 As shown in Table 1, the globally optimal gate width, compression error, and compression ratio during the extraction of SOC feature data points from the 10 battery cells are all relatively small. This verifies that the extracted SOC feature data points can characterize the changing trend of the battery cell's SOC curve.

[0126] Table 1. Global optimal gate width, compression error, and compression ratio during the extraction of SOC feature data points from 10 battery cells.

[0127]

[0128] Based on SOC feature data points, the operational life loss of BESS was evaluated. The life loss evaluation results of 10 battery cells are shown in Table 2 below. It can be calculated that the equivalent cycle loss of BESS is 1.0454 cycles and the life loss percentage is 0.0261%.

[0129] Table 2. Lifetime degradation assessment results of 10 battery cells

[0130]

[0131] To verify the speed of the BESS lifetime loss assessment method designed in this invention, it was compared with the rainflow counting method. The rainflow counting method took 2.561 minutes to assess the lifetime loss of the BESS, while the present invention only took 1.154 seconds. This verifies the speed of the BESS lifetime loss assessment method designed in this invention.

Claims

1. A method for assessing the operational lifespan loss of a battery energy storage system based on the rotating door algorithm, characterized in that, Includes the following steps: (1) The improved white whale optimization algorithm (IBWO) adopts the Logistic chaotic initialization method and the Tent chaotic initialization method to determine the initial candidate solution of the optimization individual, so as to improve the diversity of the initial candidate solution; a method for calculating the balance factor is designed based on the Sigmoid function to improve the optimization algorithm's ability to balance between global exploration and local development. (2) Design the value function for searching the globally optimal gate width E in the rotating door algorithm (SDT). Based on the SOC change curve of the battery cells in the battery energy storage system, use the improved white whale optimization algorithm to search for the globally optimal gate width E of the rotating door algorithm. (3) Based on the globally optimal gate width E, the rotating gate algorithm is used to process the SOC operation curve of the battery cell, thereby extracting feature data points that characterize the changing trend of the SOC operation curve of the battery cell. (4) Based on the feature data points, the depth of discharge of each battery cell in each half-cycle is obtained, and then the operating life loss of the battery energy storage system is calculated. The processing flow includes two steps: calculation of the maximum number of cycles and calculation of the percentage of life loss. The calculation of the maximum number of cycles is as follows: The duration between two adjacent SOC feature data points is considered as one half-cycle, and the charge / discharge depth is calculated; the charge / discharge depth within the i-th half-cycle is calculated. The corresponding maximum number of loops : ； In the formula: The coefficients are polynomial fitting coefficients, with values ​​of 164000, -872500, 1827000, -1687000, and 572900, respectively.

2. The method for assessing the operational lifespan loss of a battery energy storage system based on the rotating door algorithm according to claim 1, characterized in that, In step (1), the Logistic chaotic initialization method and the Tent chaotic initialization method are used to determine the initial candidate solutions of the optimization individual. The calculation formula for the initial candidate solutions of the optimization individual is as follows: ； In the formula: Let be the initial candidate solution for the i-th optimization individual. and These are the chaos weighting coefficients, with values ​​of 0.5 and 0.5 respectively. and These are Logistic chaotic sequences and Tent chaotic sequences, respectively. and These are the upper and lower boundaries of the optimization variable, respectively.

3. The method for assessing the operational lifespan loss of a battery energy storage system based on the rotating door algorithm according to claim 1, characterized in that, In step (1), a balance factor designed based on the Sigmoid function is mentioned. The calculation method is as follows: ； In the formula: This is the initial value of the balance factor. Balance factor The parameters were adjusted after multiple tests. The value is -0.

03. and These are the current iteration number and the maximum iteration number, respectively.

4. The method for assessing the operational lifespan loss of a battery energy storage system based on the rotating door algorithm according to claim 1, characterized in that, In step (4), the percentage of lifespan loss is calculated as follows: ； In the formula: The depth of charge / discharge during the i-th half-cycle. and This represents two adjacent SOC feature data points within the i-th half-cycle period; Calculate the equivalent number of iterations within the i-th half-cycle: ； In the formula: This indicates the maximum number of cycles a battery cell can perform at 100% charge / discharge depth. The equivalent number of cycles lost by the battery cell during the i-th half-cycle period; The equivalent number of cycles lost by the battery energy storage system during the entire operation The calculation formula is as follows: ； In the formula: This refers to the number of battery cells within the BESS. Let be the equivalent number of cycles lost by the j-th battery cell during operation. This refers to the number of half-cycles that a battery cell experiences during operation. The equivalent number of cycles lost by the j-th battery cell during the i-th half-cycle period; The formula for calculating the percentage of battery energy storage system lifespan loss is as follows: 。

Images