An intelligent operation and maintenance method for large-scale energy storage systems

Abstract

The present invention provides an intelligent operation and maintenance method for a large-scale energy storage system, and belongs to the field of intelligent system operation and maintenance technology. The method comprises: step 1, calculating the SOH value of each PACK; step 2, arranging each PACK in order from large to small according to the SOH value of each PACK; step 3, calculating the capacity of the entire battery stack; step 4, determining whether the capacity of the entire battery stack meets the expected capacity requirement, if so, terminating the algorithm, if not, proceeding to step 5; step 5, PACK replacement. The feature of the present invention is to use the algorithm of cell-level SOH to calculate the shortcomings of the energy storage system; and to use the capacity reorganization algorithm and the PACK replacement algorithm to eliminate the shortcomings of the energy storage system, so that the cycle life of the entire energy storage system can be greatly improved and the cost per kilowatt-hour of energy storage can be greatly reduced.

Description

Technical Field

[0001] The present invention belongs to the technical field of intelligent system operation and maintenance, and specifically relates to an intelligent operation and maintenance method for a large-scale energy storage system. Background Art

[0002] Large-scale energy storage systems are made up of millions of battery cells connected in series and parallel. Due to the series-parallel architecture, the operation of the energy storage system has a barrel effect, that is, the energy storage system is limited by the shortest battery cell. During the operation of the entire system, the inconsistency of the operating environment of each battery cell leads to the inconsistency of the battery cell's operating attenuation. This makes the system life significantly shorter than the cycle life of the battery cell.

[0003] At present, the energy storage industry generally over-allocates the capacity of energy storage in the initial stage. Over-allocation can reduce the discharge depth DoD of the energy storage system, thereby increasing the number of cycles of the energy storage system. However, over-allocation in the early stage of project operation increases the system's one-time investment. In addition, since chemical batteries are also limited by calendar life, this causes the value of the over-allocated energy storage system to continue to decay over time, even if this part of the over-allocated battery is not put into operation. Summary of the invention

[0004] The purpose of the present invention is to solve the problems existing in the above-mentioned prior art and to provide an intelligent operation and maintenance method for a large-scale energy storage system. The method can be implanted into the EMS system of the energy storage system, thereby performing intelligent operation and maintenance on the large-scale energy storage system and improving the capacity of the entire energy storage system.

[0005] The present invention is achieved through the following technical solutions:

[0006] An intelligent operation and maintenance method for a large-scale energy storage system specifically includes the following steps:

[0007] Step 1, calculating the remaining life of each PACK in the battery stack, that is, calculating the SOH value of each PACK;

[0008] Step 2: Arrange the PACKs into each cluster in descending order according to their SOH values;

[0009] Step 3, calculating the capacity of the entire battery stack;

[0010] Step 4, determine whether the capacity of the entire battery stack meets the expected capacity requirement, if so, end the algorithm, if not, proceed to step 5;

[0011] Step 5, PACK replacement.

[0012] A further improvement of the present invention is:

[0013] In step 1, the SOH value of each PACK is calculated. The specific operation is:

[0014] Step 1.1, at the rated charge and discharge rate, fully charge and fully discharge each battery stack, and record the cell alpha_max that first reaches the highest voltage value during the full charge process and the cell alpha_min that first reaches the lowest voltage value during the full discharge process of each battery cluster;

[0015] Step 1.2, for other cells in the same cluster, during the full charging process, measure the maximum voltage value of the cell, calculate the difference between the maximum voltage value of the cell and the alpha_max voltage value of the cluster, and calculate the power Power_max value corresponding to the voltage difference;

[0016] During the full discharge process, measure the lowest voltage value of the cell, calculate the difference between the lowest voltage value of the cell and the alpha_min voltage value of the cluster, and calculate the power_min value corresponding to the voltage difference;

[0017] Calculate the SOH value of the battery cell;

[0018] Step 1.3: The SOH value of the PACK is the lowest SOH value of all cells in the PACK multiplied by the number of cells in the PACK.

[0019] A further improvement of the present invention is:

[0020] In step 1.1, alpha_max and alpha_min refer to the highest and lowest voltages in the cluster under the same current conditions, where alpha_max refers to the cell that first reaches the highest voltage threshold during charging, and alpha_min refers to the cell that first reaches the lowest voltage threshold during discharging.

[0021] A further improvement of the present invention is:

[0022] In step 1.2, other cells refer to cells other than alpha_min and alpha_max.

[0023] A further improvement of the present invention is:

[0024] In step 1.2, Power_max=(voltage difference during full charging)×current×time, where the Power_max value is the amount of battery power that is not released due to the short board effect.

[0025] A further improvement of the present invention is:

[0026] In step 1.2, Power_min=(voltage difference during full discharge)×current×time, where the Power_min value is the amount of battery power that is not charged due to the short board effect.

[0027] A further improvement of the present invention is:

[0028] In step 1.2, the SOH value of the battery cell is calculated as follows:

[0029] SOH value = (actual discharge capacity of the battery cell + Power_min + Power_max × battery cell efficiency) / nominal capacity of the battery cell;

[0030] Among them, the actual discharge capacity of the battery cell = actual voltage × current × time.

[0031] A further improvement of the present invention is:

[0032] Step 3, calculate the capacity of the entire battery stack, specifically:

[0033] The capacity of the battery cluster is obtained by multiplying the minimum SOH value in each battery cluster by the number of PACKs in the cluster. The capacity of the entire battery stack is the sum of the capacities of all battery clusters.

[0034] A further improvement of the present invention is:

[0035] Step 4, determining whether the capacity of the entire battery stack meets the expected capacity requirement, the specific operations are:

[0036] Set the rated attenuation curve of the battery stack, that is, the expected attenuation curve of the entire battery stack, and use this expected attenuation curve as the standard for PACK replacement to obtain the attenuation curve of the reorganized battery stack and see whether it intersects with the expected attenuation curve. If there is an intersection, the expected attenuation curve cannot be achieved, that is, the expected capacity requirement is not met.

[0037] A further improvement of the present invention is:

[0038] Step 5: PACK replacement. The specific operations are as follows:

[0039] Calculate the difference Beta between the lowest SOH value and the second-to-last lowest SOH value of each cluster of batteries, and sort the differences from large to small; replace the PACK with the lowest SOH value of the cluster with the largest difference Beta, and the replaced PACK has a 100% SOH value, and then recalculate the capacity of the entire battery stack. If the battery stack capacity meets the requirements, end the algorithm, otherwise calculate the Beta value of each cluster of batteries again, and replace the PACK according to the Beta value until the capacity of the entire battery stack meets the preset capacity value.

[0040] Compared with the prior art, the present invention has the following beneficial effects:

[0041] The present invention discloses an intelligent operation and maintenance method for a large-scale energy storage system. The most significant feature is that the algorithm of cell-level SOH is used to calculate the shortcomings of the energy storage system; the capacity reorganization algorithm and the PACK replacement algorithm are used to eliminate the shortcomings of the energy storage system, thereby greatly improving the cycle life of the entire energy storage system and greatly reducing the cost of energy storage per kilowatt-hour. BRIEF DESCRIPTION OF THE DRAWINGS

[0042] Figure 1 It is the architecture diagram of a large-scale energy storage system;

[0043] Figure 2 It is a flow chart of an intelligent operation and maintenance method for a large-scale energy storage system of the present invention. DETAILED DESCRIPTION

[0044] The present invention is further described in detail below in conjunction with the accompanying drawings:

[0045] The present invention proposes an intelligent operation and maintenance method for a large-scale energy storage system, which is implanted into the EMS system of the energy storage system, so as to perform intelligent operation and maintenance on the large-scale energy storage system. The intelligent operation and maintenance method is divided into two parts. The first part is the capacity reorganization of the energy storage system to increase the capacity of the entire energy storage system; the second part is the intelligent PACK replacement, which increases the capacity of the entire energy storage system by replacing the PACK.

[0046] like Figure 1 As shown in the figure, it is the architecture diagram of the large-scale energy storage system. It is an architecture example. The number of clusters and the number of PACKs can be different. On the DC bus of the entire AC / DC, multiple battery clusters are connected in parallel (on Figure 1 In the figure, five battery clusters are represented by Cluster 1, Cluster 2 to Cluster 5 respectively. These multiple battery clusters are collectively referred to as a battery stack. In a battery stack, the inconsistency of the capacity decay of the PACK leads to circulation between clusters. In the same cluster, the inconsistent capacity decay of the PACK leads to the charge and discharge cycle life of the cluster being much lower than the cycle life of the battery cell (because the entire cluster of batteries is limited by the PACK with the lowest capacity). In order to increase the charge and discharge cycle life of the entire battery stack and make the cycle life of the entire battery stack closer to the cycle life of the single battery cell, it is necessary to make the PACK decay in the single battery cluster tend to be consistent. The method of the present invention is an algorithm that makes all PACKs in a single cluster of batteries tend to be consistent.

[0047] The present invention provides an intelligent operation and maintenance method for a large-scale energy storage system, such as Figure 2 As shown, the specific steps include:

[0048] Step 1: Calculate the remaining life of each PACK in the battery stack, that is, calculate the SOH value of each PACK.

[0049] At the rated charge and discharge rate and rated temperature, the SOH value of the PACK is defined as the charge and discharge amount of the PACK divided by the nominal capacity of the PACK, based on the nominal capacity of the battery.

[0050] The specific calculation method of SOH value is:

[0051] Step 1.1, at the rated charge and discharge rate (temperature deviation is included in the correction factor provided by the manufacturer), fully charge and fully discharge each battery stack, and record the cell alpha_max that first reaches the highest voltage value during the full charge process and the cell alpha_min that first reaches the lowest voltage value during the full discharge process of each battery cluster;

[0052] Each battery cluster is composed of a series of cells connected in series. During the charging process, the current flowing through each cell in the cluster is consistent due to the series connection, but the voltage of each cell is different. So here it refers to the highest and lowest voltages in the cluster under the same current, where alpha_max refers to the cell that first reaches the highest voltage threshold during the charging process, and alpha_min refers to the cell that first reaches the lowest voltage threshold during the discharging process.

[0053] If the battery cluster does not have the conditions for full charge and full discharge (mainly determined by the operation mode of the energy storage system), the released electricity within a certain SOC value range (the SOC value range mainly depends on the operation mode of the system. For example, if the system operation mode makes the energy storage system SOC always run between 60% and 80%, then the range here refers to 60% to 80%) is recorded at the rated charge and discharge rate. The released electricity is compared with the electricity released by the battery cell within the SOC value range to obtain the SOH value.

[0054] For example, for a battery cell with a normal SOH value of 100%, the amount of electricity discharged between 60% and 80% SOC values ​​can be obtained. After running for a period of time, the amount of electricity discharged between 60% and 80% SOC values ​​is measured again and compared with the amount of electricity discharged when the SOH value is 100%, and the SOH value of the battery cell is obtained.

[0055] Step 1.2, for other cells in the same cluster, other cells refer to cells other than alpha_min and alpha_max. During the full charge process, measure the maximum voltage value of the cell (the existing BMS management system measures the voltage value of each cell, which will not be repeated here), calculate the difference between the maximum voltage value of the cell and the voltage value of alpha_max of the cluster, and calculate the power Power_max corresponding to the voltage difference, Power_max = (voltage difference during full charge) × current × time, the Power_max value is the cell power that is not released due to the short board effect of the cell;

[0056] During the full discharge process, measure the lowest voltage value of the battery cell (the existing BMS management system will measure the voltage value of each battery cell, which will not be repeated here), calculate the difference between the lowest voltage value of the battery cell and the voltage value of alpha_min of the cluster, and calculate the power Power_min corresponding to the voltage difference, Power_min = (voltage difference during full discharge) × current × time, the Power_min value is the battery cell power that is not charged due to the short board effect;

[0057] The SOH value of the battery cell is calculated as follows:

[0058] SOH value = (actual discharge capacity of the battery cell + Power_min + Power_max × battery cell efficiency) / nominal capacity of the battery cell

[0059] Among them, the actual discharge capacity of the battery cell = actual voltage × current × time. The BMS will measure and record the current and voltage of each battery cell at each moment;

[0060] The cell efficiency is given by the cell manufacturer and is approximately 90%.

[0061] Step 1.3: The SOH value of the PACK is the lowest SOH value of all cells in the PACK multiplied by the number of cells in the PACK.

[0062] Step 2, capacity reorganization: According to the SOH value of each PACK, arrange them in each cluster in descending order.

[0063] For example Figure 1 In the figure, the 16 PACKs in the first cluster are the PACKs with the largest SOH value to the 16th ranked SOH value in the entire battery stack, the second cluster of batteries are the PACKs with the 17th ranked SOH value to the 32nd ranked SOH value, and so on.

[0064] There is no regulation on how many clusters to divide into and how many PACKs each cluster contains.

[0065] Step 3, calculate the capacity of the entire battery stack.

[0066] The capacity of the battery cluster is obtained by multiplying the minimum SOH value in each battery cluster by the number of PACKs in the cluster. The capacity of the entire battery stack is the sum of the capacities of all battery clusters.

[0067] Step 4, determine whether the capacity of the entire battery stack meets the expected capacity requirement, if so, end the algorithm, if not, proceed to the next step of PACK replacement (step 5);

[0068] The specific operation is: set the rated attenuation curve of the battery stack (artificially set the rated attenuation curve of the battery stack as an indicator for system assessment), that is, the expected attenuation curve of the entire battery stack, and use this expected attenuation curve as the standard for PACK replacement. When the expected attenuation curve cannot be reached after capacity reorganization (obtain the attenuation curve of the reorganized battery stack, and see whether it has an intersection with the expected attenuation curve. If there is an intersection, it means that the expected attenuation curve cannot be reached), it is necessary to add a PACK replacement strategy to increase the capacity of the battery stack to the expected capacity.

[0069] Step 5, PACK replacement: calculate the difference Beta between the lowest SOH value and the second-to-last lowest SOH value of each cluster of batteries, and sort the differences from large to small; replace the PACK with the lowest SOH value of the cluster with the largest difference Beta, and the replaced PACK has a 100% SOH value (i.e. first select the cluster with the largest difference Beta, and then replace the battery cell with the lowest SOH value in that cluster, and the replaced battery cell has a 100% SOH value; a battery cell with 100% SOH is a battery cell with a health level of 100%), and then recalculate the capacity of the entire battery stack (return to step 3). If the battery stack capacity meets the requirements (step 4), end the algorithm, otherwise calculate the Beta value of each cluster of batteries again, and replace the PACK according to the Beta value until the capacity of the entire battery stack meets the preset capacity value.

[0070] Finally, it should be explained that the above technical solution is only one implementation method of the present invention. For those skilled in the art, it is easy to make various types of improvements or modifications based on the application methods and principles disclosed in the present invention, and it is not limited to the method described in the above specific implementation method of the present invention. Therefore, the method described above is only preferred and does not have a restrictive meaning.

Claims

1. An intelligent operation and maintenance method for a large-scale energy storage system, characterized in that: The specific steps include: Step 1, calculating the remaining life of each PACK in the battery stack, that is, calculating the SOH value of each PACK; Step 2: Arrange the PACKs into each cluster in descending order according to their SOH values; Step 3, calculating the capacity of the entire battery stack; Step 4, determine whether the capacity of the entire battery stack meets the expected capacity requirement, if so, end the algorithm, if not, proceed to step 5; Step 5, PACK replacement; In step 1, the SOH value of each PACK is calculated, and the specific operation is as follows: Step 1.1, at the rated charge and discharge rate, fully charge and fully discharge each battery stack, and record the cell alpha_max that first reaches the highest voltage value during the full charge process and the cell alpha_min that first reaches the lowest voltage value during the full discharge process of each battery cluster; Step 1.2, for other cells in the same cluster, during the full charging process, measure the maximum voltage value of the cell, calculate the difference between the maximum voltage value of the cell and the alpha_max voltage value of the cluster, and calculate the power Power_max value corresponding to the voltage difference; During the full discharge process, measure the lowest voltage value of the cell, calculate the difference between the lowest voltage value of the cell and the alpha_min voltage value of the cluster, and calculate the power_min value corresponding to the voltage difference; Calculate the SOH value of the battery cell; Step 1.3, the SOH value of the PACK is the lowest SOH value of all cells in the PACK multiplied by the number of cells in the PACK; Wherein, in the step 1.1, alpha_max and alpha_min refer to the highest voltage and lowest voltage in the cluster under the same current condition, wherein alpha_max refers to the cell that first reaches the highest voltage threshold value during the charging process, and alpha_min refers to the cell that first reaches the lowest voltage threshold value during the discharging process; In the step 1.2, Power_max=(voltage difference during full charging)×current×time, where the Power_max value is the amount of battery power that is not released due to the short board effect; Power_min=(voltage difference during full discharge)×current×time, where the Power_min value is the amount of battery power that is not charged due to the short board effect.

2. The intelligent operation and maintenance method of a large-scale energy storage system according to claim 1, characterized in that: In the step 1.2, other cells refer to cells other than alpha_min and alpha_max.

3. The intelligent operation and maintenance method of a large-scale energy storage system according to claim 1, characterized in that: In step 1.2, the SOH value of the battery cell is calculated as follows: SOH value = (actual discharge capacity of the battery cell + Power_min + Power_max × battery cell efficiency) / nominal capacity of the battery cell; Among them, the actual discharge capacity of the battery cell = actual voltage × current × time.

4. The intelligent operation and maintenance method of a large-scale energy storage system according to claim 1, characterized in that: The step 3 is to calculate the capacity of the entire battery stack, specifically: The capacity of the battery cluster is obtained by multiplying the minimum SOH value in each battery cluster by the number of PACKs in the cluster. The capacity of the entire battery stack is the sum of the capacities of all battery clusters.

5. The intelligent operation and maintenance method of a large-scale energy storage system according to claim 1, characterized in that: The step 4 is to determine whether the capacity of the entire battery stack meets the expected capacity requirement, and the specific operations are as follows: Set the rated attenuation curve of the battery stack, that is, the expected attenuation curve of the entire battery stack, and use this expected attenuation curve as the standard for PACK replacement to obtain the attenuation curve of the reorganized battery stack and see whether it intersects with the expected attenuation curve. If there is an intersection, the expected attenuation curve cannot be achieved, that is, the expected capacity requirement is not met.

6. The intelligent operation and maintenance method of a large-scale energy storage system according to claim 1, characterized in that: The step 5, PACK replacement, specifically includes: Calculate the difference Beta between the lowest SOH value and the second-to-last lowest SOH value of each cluster of batteries, and sort the differences from large to small; replace the PACK with the lowest SOH value of the cluster with the largest difference Beta, and the replaced PACK has a 100% SOH value, and then recalculate the capacity of the entire battery stack. If the battery stack capacity meets the requirements, end the algorithm, otherwise calculate the Beta value of each cluster of batteries again, and replace the PACK according to the Beta value until the capacity of the entire battery stack meets the preset capacity value.

Images