An online capacity balancing method and system for a vanadium flow battery system

Abstract

The application relates to an online capacity balancing method and system for a vanadium flow battery system. The method part mainly comprises the following steps: detecting the liquid level of a positive electrolyte tank and a negative electrolyte tank in real time; calculating the liquid level difference and the liquid level deviation rate of the positive electrolyte tank and the negative electrolyte tank; generating an electromagnetic valve control signal according to the liquid level difference and the liquid level deviation rate, adjusting the duty cycle of the electromagnetic valve between the positive electrolyte tank and the negative electrolyte tank, and controlling the electrolyte flow between the positive electrolyte tank and the negative electrolyte tank to realize capacity balancing. The application can realize online capacity balancing during system operation, reduce downtime, simplify operation process and improve overall system operation efficiency, thereby solving the problems of obvious limitations in efficiency, automation degree and economy of the prior art.

Description

Technical Field

[0001] This application relates to the field of vanadium redox flow battery technology, and in particular to an online capacity balancing method and system for vanadium redox flow battery systems. Background Technology

[0002] Vanadium redox flow batteries (VRFBs) are an energy storage technology characterized by high safety, long lifespan, and good scalability, widely used in renewable energy grid integration, power peak shaving, and backup power. These batteries store electrolytes in separate positive and negative electrolyte tanks, and achieve energy storage and release through the transfer of electrons and ions via ion exchange membranes during charging and discharging. With the continuous growth in energy storage demand, the application scope of VRFBs is expanding, and their importance in the energy sector is increasingly prominent.

[0003] During long-term operation, the usable capacity of the positive and negative electrolytes in a vanadium redox flow battery gradually becomes unbalanced, manifesting as increased liquid level differences and decreased energy utilization. To restore the system's capacity, existing technologies typically employ the following capacity balancing methods when the system detects a liquid level difference exceeding a preset threshold or a significant decrease in usable capacity: First, the positive and negative electrolyte mixing method, which involves mixing a portion of the positive electrolyte with the negative electrolyte via pipeline valves or manual means, thereby restoring the overall concentration of vanadium ions of different valence states within the system to a more uniform level; second, the circulation transfer method, which utilizes an external circulation pump to transfer a portion of the electrolyte from the side with the higher liquid level to the side with the lower liquid level, thus achieving a rebalancing of volume and capacity.

[0004] Existing capacity balancing methods are complex and time-consuming. Before balancing, the system typically needs to be shut down to check for liquid levels or capacity differences. The balancing process requires transferring a portion of the electrolyte from one side to the other until the electrolyte states on both sides are nearly identical, a process that takes considerable time. Furthermore, additional post-processing is required. After balancing, the concentrations of vanadium ions in different valence states are disrupted, necessitating further electrochemical activation or cyclic adjustment to restore the electrochemical activity and balance stability of each valence state, increasing operational complexity. Moreover, the economic costs are high; frequent mixing and activation operations not only consume system operating time but also lead to increased energy consumption, equipment wear and tear, and labor maintenance costs.

[0005] Therefore, overcoming the significant limitations of existing capacity balancing methods in terms of efficiency, automation, and economy is a problem that needs to be solved in this technical field. Summary of the Invention

[0006] In view of the above-mentioned defects or improvement needs of the existing technology, and in order to solve the problem that the existing capacity balancing methods have obvious limitations in terms of efficiency, automation and economy, this application provides an online capacity balancing method and system for vanadium redox flow battery systems. It can achieve online capacity balancing during system operation, thereby reducing downtime, simplifying operation procedures and improving the overall operating efficiency of the system, thus solving the problem that the existing technology has obvious limitations in terms of efficiency, automation and economy.

[0007] The embodiments of this application adopt the following technical solutions:

[0008] In a first aspect, this application provides an online capacity balancing method for vanadium redox flow battery systems, comprising:

[0009] Real-time monitoring of the liquid level heights in the positive and negative electrolyte tanks;

[0010] Calculate the liquid level difference and liquid level drift rate between the positive electrode electrolyte tank and the negative electrode electrolyte tank;

[0011] Based on the liquid level difference and liquid level deviation rate, a solenoid valve control signal is generated to adjust the duty cycle of the solenoid valve between the positive and negative electrolyte tanks, thereby controlling the electrolyte flow between the positive and negative electrolyte tanks and achieving capacity balance.

[0012] By adopting the above technical solutions, early dynamic suppression of capacity imbalance is achieved, avoiding serious liquid level deviation; the balancing process is fully automated, requiring no manual intervention or shutdown for mixing; balancing time and electrolyte loss are significantly reduced, improving the system's available capacity and operating economy.

[0013] In some embodiments, calculating the liquid level difference and liquid level shift rate between the positive electrode electrolyte tank and the negative electrode electrolyte tank specifically includes:

[0014] Set the calculation time window;

[0015] At each time point in the calculation time window, the liquid level difference between the positive electrode electrolyte tank and the negative electrode electrolyte tank is calculated to obtain the liquid level difference sequence;

[0016] Linear regression was performed on the liquid level difference sequence over a period of time to calculate the liquid level shift rate.

[0017] By adopting the above technical solution, the liquid level difference and liquid level deviation rate between the positive electrode electrolyte tank and the negative electrode electrolyte tank can be accurately calculated. This provides accurate data support for generating solenoid valve control signals based on the liquid level difference and liquid level deviation rate to achieve capacity balance. It also helps to achieve early dynamic suppression of capacity imbalance and avoid serious liquid level deviation.

[0018] In some embodiments, the step of generating a solenoid valve control signal based on the liquid level difference and liquid level shift rate, and adjusting the duty cycle of the solenoid valve between the positive and negative electrolyte tanks, specifically includes:

[0019] Using the liquid level difference as the core indicator for control optimization, a corresponding loss function is constructed.

[0020] A proportional-derivative control law is introduced, with the liquid level difference and liquid level deviation rate as feedback variables;

[0021] Adjust the proportional coefficient and derivative coefficient to make the solenoid valve opening control meet the condition of minimizing the loss function.

[0022] By adopting the above technical solution, in order to achieve the optimal balance of system capacity, the liquid level difference is used as the core indicator for control optimization and a corresponding loss function is constructed. In order to achieve the real-time solution of this objective, a proportional-derivative control law is introduced, and the liquid level difference and liquid level deviation rate are used as feedback variables. By adjusting the proportional coefficient and the derivative coefficient, the opening control of the solenoid valve is made to approximately satisfy the condition of minimizing the loss function, thereby achieving the optimal balance effect under the condition of limited computation.

[0023] In some embodiments, the solenoid valve is opened when the liquid level difference is greater than a first set threshold, and closed when the liquid level difference is less than a second set threshold.

[0024] By adopting the above technical solution, the opening and closing of the solenoid valve can be automatically controlled based on the comparison between the liquid level difference and the set threshold, making the balancing process fully automated, without manual intervention or downtime for mixing, significantly reducing balancing time and electrolyte loss, and improving the system's available capacity and operating economy.

[0025] In some embodiments, the proportional coefficient and differential coefficient are periodically updated using recursive least squares or Bayesian estimation based on historical operating data.

[0026] By adopting the above technical solution, adaptive control is achieved, and by updating the proportional coefficient and derivative coefficient, deviations caused by changes in electrolyte concentration and valve resistance are compensated, thereby achieving the goal of maintaining high precision and stability of capacity balance control over a long period of time.

[0027] In some embodiments, the relationship between the balanced flow rate and the liquid level difference after the solenoid valve is opened is determined through offline testing. Based on the test results, the corresponding linear or nonlinear model is selected, and then the corresponding parameters are identified.

[0028] By adopting the above technical solution, the relationship between the balanced flow rate and the liquid level difference after the solenoid valve is opened can be accurately determined. This provides more precise parameter basis for generating solenoid valve control signals and adjusting the solenoid valve duty cycle based on the liquid level difference and liquid level deviation rate. It helps to achieve precise control of capacity balance, further improves the efficiency and accuracy of system capacity balance, reduces balancing time and electrolyte loss, and improves the available capacity and operating economy of the system.

[0029] Secondly, this application provides an online capacity balancing system for vanadium redox flow battery systems, employing the online capacity balancing method for vanadium redox flow battery systems as described in the first aspect, including a liquid level detection module, a balancing pipeline, a solenoid valve assembly, and a control and calculation module, wherein:

[0030] The liquid level detection module is installed in the positive electrolyte tank and the negative electrolyte tank respectively, and is used to detect the liquid level height signal in real time;

[0031] The balancing pipeline connects the positive electrolyte tank and the negative electrolyte tank, and is used to realize the automatic flow of electrolyte on both sides when the capacity is unbalanced;

[0032] The solenoid valve assembly is installed on the balance pipeline and is used to control the flow of electrolyte between the positive electrolyte tank and the negative electrolyte tank;

[0033] The control and calculation module is electrically connected to the liquid level detection module and the solenoid valve assembly, respectively, and is used to collect liquid level difference data, calculate liquid level offset rate, identify system parameters and output solenoid valve control signals.

[0034] By adopting the above technical solution, the system can detect the liquid level height of the positive and negative electrolyte tanks in real time, calculate the liquid level difference and offset rate, and adjust the duty cycle of the solenoid valve to control the electrolyte flow, thereby achieving capacity balance. This realizes early dynamic suppression of capacity imbalance and avoids serious liquid level offset. The balancing process is fully automated, requiring no manual intervention or shutdown for mixing. It significantly reduces balancing time and electrolyte loss, and improves the system's available capacity and operating economy.

[0035] In some embodiments, the liquid level detection module includes one or more of an ultrasonic level gauge, a differential pressure level gauge, and a magnetic float level gauge.

[0036] By adopting the above technical solutions, using ultrasonic level gauges, differential pressure level gauges, or magnetic float level gauges as level detection modules, the liquid level height of the positive and negative electrolyte tanks can be detected in real time, providing a data basis for subsequent calculation of liquid level difference and liquid level deviation rate.

[0037] In some embodiments, the balancing pipeline is provided with a bidirectional flow path, and the solenoid valve assembly includes two directional control valves, which respectively control the flow of electrolyte to the positive electrode or the negative electrode tank.

[0038] By adopting the above technical solution, the balance pipeline can achieve bidirectional flow, and the two directional control valves can control the electrolyte to flow to the positive or negative electrode tank respectively, further improving the flexibility and accuracy of electrolyte flow control, and more effectively achieving capacity balance between the positive and negative electrode electrolyte tanks.

[0039] In summary, this application includes at least the following beneficial technical effects:

[0040] 1. Achieve early dynamic suppression of capacity imbalance to avoid severe liquid level deviation;

[0041] 2. The balancing process is fully automated, requiring no manual intervention or downtime for mixing;

[0042] 3. Equilibrium time and electrolyte loss are significantly reduced, improving system capacity and operating economy;

[0043] 4. It maintains high accuracy and stability over a long period of time through an adaptive update mechanism. Attached Figure Description

[0044] To more clearly illustrate the technical solutions of the embodiments of this application, the drawings used in the embodiments of this application will be briefly described below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0045] Figure 1 A flowchart of an online capacity balancing method for a vanadium redox flow battery system provided in this application embodiment;

[0046] Figure 2 An online capacity balancing system architecture diagram for a vanadium redox flow battery system is provided for embodiments of this application;

[0047] Figure 3 The flowchart illustrates the operation of the control and calculation module provided in the embodiments of this application. Detailed Implementation

[0048] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the scope of this application. Furthermore, the technical features involved in the various embodiments described below can be combined with each other as long as they do not conflict with each other. The application will now be described in detail with reference to the accompanying drawings and embodiments. Example 1

[0049] like Figure 1As shown in the figure, this application provides an online capacity balancing method for a vanadium redox flow battery system, and the specific steps are as follows.

[0050] Step 101: Real-time monitoring of the liquid level in the positive and negative electrolyte tanks. In this step, the liquid level data of the positive and negative electrolyte tanks can be collected separately using a liquid level detection module.

[0051] Step 102: Calculate the level difference and level shift rate between the positive and negative electrolyte tanks. Specifically, a calculation time window can be set; at each time point within the calculation time window, the level difference between the positive and negative electrolyte tanks is calculated to obtain a level difference sequence; a linear regression is performed on the level difference sequence over a period of time to calculate the level shift rate. This step can accurately calculate the level difference and level shift rate between the positive and negative electrolyte tanks, providing precise data support for subsequently generating solenoid valve control signals based on the level difference and level shift rate to achieve capacity balance. This helps to achieve early dynamic suppression of capacity imbalance and avoid severe level shifts.

[0052] Step 103: Generate solenoid valve control signals based on the liquid level difference and liquid level deviation rate. Adjust the duty cycle of the solenoid valves between the positive and negative electrolyte tanks to control the electrolyte flow between them and achieve capacity balance. Specifically, the liquid level difference can be used as the core indicator for control optimization, and a corresponding loss function can be constructed. A proportional-derivative control law is introduced, with the liquid level difference and liquid level deviation rate used as feedback variables. The proportional and derivative coefficients are adjusted to make the solenoid valve opening control satisfy the condition of minimizing the loss function. This step aims to achieve optimal system capacity balance by using the liquid level difference as the core indicator for control optimization and constructing a corresponding loss function. To achieve real-time solution of this objective, a proportional-derivative control law is introduced, with the liquid level difference and liquid level deviation rate used as feedback variables. By adjusting the proportional and derivative coefficients, the solenoid valve opening control approximately satisfies the condition of minimizing the loss function, thereby achieving optimal balance under computational constraints.

[0053] The above technical solutions achieve early dynamic suppression of capacity imbalance, avoiding severe liquid level deviation; the balancing process is fully automated, requiring no manual intervention or shutdown for mixing; balancing time and electrolyte loss are significantly reduced, improving the system's available capacity and operating economy.

[0054] In some implementations, the solenoid valve is opened when the liquid level difference is greater than a first set threshold, and closed when the liquid level difference is less than a second set threshold. Automatically controlling the opening and closing of the solenoid valve based on the comparison between the liquid level difference and the set threshold automates the balancing process, eliminating the need for manual intervention or system shutdown for mixing. This significantly reduces balancing time and electrolyte loss, improving system capacity and operational economy.

[0055] Furthermore, in some implementations, the proportional and differential coefficients can be periodically updated using recursive least squares or Bayesian estimation based on historical operating data. Through these technical solutions, adaptive control is achieved, and by updating the proportional and differential coefficients, deviations caused by changes in electrolyte concentration, valve resistance, etc., are compensated for, thus maintaining high precision and stability of capacity balance control over the long term.

[0056] Furthermore, in some implementations, the relationship between the equilibrium flow rate and the liquid level difference after the solenoid valve opens is determined through offline testing. Based on the test results, a corresponding linear or nonlinear model is selected, and then the relevant parameters are identified. This technical solution accurately determines the relationship between the equilibrium flow rate and the liquid level difference after the solenoid valve opens, providing more precise parameter data for generating solenoid valve control signals and adjusting the solenoid valve duty cycle based on the liquid level difference and liquid level deviation rate. This helps achieve precise control of capacity balance, further improving the efficiency and accuracy of system capacity balance, reducing balancing time and electrolyte loss, and increasing the system's available capacity and operational economy. Example 2

[0057] Based on the online capacity balancing method for vanadium redox flow battery systems provided in Example 1, this Example 2 provides an online capacity balancing system for vanadium redox flow battery systems, applying the online capacity balancing method for vanadium redox flow battery systems as described in Example 1.

[0058] First, the vanadium redox flow battery system includes a positive electrolyte tank and a negative electrolyte tank, and both the positive electrolyte tank and the negative electrolyte tank are connected to the battery stack.

[0059] refer to Figure 2 As shown, the online capacity balancing system for vanadium redox flow battery systems in this application includes a liquid level detection module, a balancing pipeline, a solenoid valve assembly, and a control and calculation module.

[0060] The liquid level detection modules are installed in both the positive and negative electrolyte tanks to detect the liquid level height signal in real time. Each liquid level detection module is one or more of an ultrasonic level gauge, a differential pressure level gauge, and a magnetic level gauge. Using an ultrasonic level gauge, a differential pressure level gauge, or a magnetic level gauge as the liquid level detection module allows for real-time detection of the liquid level height in both the positive and negative electrolyte tanks, providing a data basis for subsequent calculations of the liquid level difference and liquid level shift rate.

[0061] The balancing pipeline connects the positive and negative electrolyte tanks, enabling automatic flow of electrolyte between the two sides when there is a capacity imbalance. The balancing pipeline is designed with a bidirectional flow path, allowing for two-way flow.

[0062] The solenoid valve assembly is installed on the balancing pipeline and is used to control the flow of electrolyte between the positive and negative electrolyte tanks. The solenoid valve assembly includes two directional control valves, which respectively control the flow of electrolyte to the positive or negative tank. The ability to control the flow of electrolyte to either the positive or negative tank through these two directional control valves further enhances the flexibility and precision of electrolyte flow control, and more effectively achieves capacity balance between the positive and negative electrolyte tanks.

[0063] The control and calculation module is electrically connected to the liquid level detection module and the solenoid valve assembly, respectively, and is used to collect liquid level difference data, calculate the liquid level offset rate, identify system parameters, and output solenoid valve control signals. (Reference) Figure 3 As shown, the control and calculation module acquires the detected liquid level in real time, calculates the liquid level deviation and the rate of liquid level change. The liquid level deviation is the difference in liquid level between the positive and negative electrolyte tanks, and the rate of liquid level change is the liquid level deviation rate. Next, the control and calculation module determines whether the liquid level difference reaches the set opening threshold. If not, it continues to detect the liquid level; if it does, it performs PD (proportional-derivative) control on the solenoid valve. Then, it judges whether the predicted liquid level is close to the actual detection. If so, it further determines whether the liquid level difference reaches the set closing threshold; otherwise, it closes the solenoid valve. After updating the parameters (proportional coefficient, derivative coefficient), the solenoid valve is then subjected to PD control; to determine whether the liquid level difference has reached the set closing threshold, if so, the solenoid valve is closed, otherwise the solenoid valve is subjected to PD control.

[0064] Through the above technical solution, the system can detect the liquid level height of the positive and negative electrolyte tanks in real time, calculate the liquid level difference and offset rate, and adjust the duty cycle of the solenoid valve to control the electrolyte flow, thereby achieving capacity balance. This realizes early dynamic suppression of capacity imbalance and avoids serious liquid level offset. The balancing process is fully automated, requiring no manual intervention or shutdown for mixing. It significantly reduces balancing time and electrolyte loss, and improves the system's available capacity and operating economy. Example 3

[0065] Based on the online capacity balancing method for vanadium redox flow battery systems provided in Example 1 and the online capacity balancing system for vanadium redox flow battery systems provided in Example 2, this Example 3 further describes the method flow provided in this application through more specific process examples.

[0066] 1. Real-time liquid level detection and liquid level deviation rate calculation:

[0067] During the system's charging and discharging process, the liquid level detection module collects liquid level data from both the positive and negative electrolyte tanks. The liquid level detection module can be one or more of the following: ultrasonic level gauge, differential pressure level gauge, magnetic level gauge, etc. The liquid level detection module sends the analog signal corresponding to the liquid level to the control and calculation module in real time. Because the liquid level changes relatively slowly, the sampling period used for liquid level deviation calculation is... The setting can be determined based on the tank capacity and project experience.

[0068] Set the calculation time window (k=1,2……N);

[0069] In each Calculate at a point in time The liquid level difference sequence was obtained, where This represents the liquid level in the positive electrode storage tank at a given time point. This refers to the liquid level height of the negative electrode storage tank at the same time point.

[0070] For time points with a window length of N Perform linear regression to estimate the rate of change of liquid level:

[0071] ;

[0072] in These are the average values ​​of the time interval and the liquid level difference within the window, respectively.

[0073] 2. Balance control calculation and solenoid valve opening strategy:

[0074] Before determining the solenoid valve opening strategy, the relationship between the equilibrium flow rate and the liquid level difference after the solenoid valve opens is first determined through offline experiments. The following uses a linear model relationship as an example. In practical applications, the corresponding linear or nonlinear model can be selected based on the test results, and then the corresponding parameters can be identified:

[0075] ;

[0076] in, To balance traffic, For liquid level difference, This is a proportionality coefficient determined by the system structure and valve characteristics. For example, under a fixed liquid level difference condition, the change in liquid level difference per unit time is measured to determine the correspondence between the solenoid valve opening duration and the liquid level balancing rate. The control and calculation module can control the actual flow rate by controlling the solenoid valve's opening duty cycle. ;

[0077] Where D(t) is the duty cycle of the solenoid valve controlled by the control and calculation module, and 0≤D(t)≤1.

[0078] This function serves as the basis for calibrating the system's control sensitivity. It allows the control and calculation modules to estimate the trend of equilibrium flow rate changes based on the current liquid level difference during operation, and adjust the solenoid valve opening time and control gain accordingly.

[0079] 3. Set a balancing threshold, enable balancing, and update parameters based on the balancing results:

[0080] To achieve optimal balance of system capacity, this application embodiment uses liquid level deviation as the core indicator for control optimization and constructs the following loss function:

[0081] ;

[0082] in:

[0083] : Reflects the squared error of the liquid level difference;

[0084] : Reflects the smoothness of the rate of change of liquid level difference, avoiding drastic fluctuations;

[0085] : This is a weighting coefficient used to balance response speed and stability.

[0086] The optimization objective is: ;

[0087] The constraints are: ;

[0088] To achieve real-time solution for this objective, the control module introduces a proportional-derivative (PD) control law, using the liquid level deviation and its rate of change as feedback variables. This is achieved by adjusting the proportional coefficient. With differential coefficients This allows the solenoid valve opening control to approximately satisfy the loss function minimization condition, thereby achieving the optimal balance effect under limited computational constraints.

[0089] 4. Control of solenoid valves during operation:

[0090] The controller employs a time-sharing discrete control strategy and adjusts the solenoid valve duty cycle in real time according to the following logic:

[0091] ;

[0092] in, These are the proportional coefficient and the derivative coefficient, respectively, when the liquid level difference... Less than the set threshold At this time, the control module automatically closes the solenoid valve, terminating the balancing operation.

[0093] 5. Parameter update and adaptation:

[0094] To maintain long-term accuracy, the control module periodically adjusts the model parameters based on historical operating data. Recursive least squares or Bayesian estimation updates are performed to compensate for deviations caused by changes in electrolyte concentration, valve resistance, etc., to achieve adaptive control.

[0095] During the update cycle, the system records the deviation between the predicted and actual liquid level difference values ​​in real time:

[0096] ;

[0097] in, To account for the actual measured liquid level error, To determine the equilibrium flow rate velocity relative to liquid level deviation Based on the predicted liquid level deviation using the actual cross-sectional area of ​​the tank, when the predicted liquid level difference deviates from the actual value... When the set threshold is exceeded, the model parameter correction process is triggered, and the parameters are adjusted according to a proportional recursive algorithm. :

[0098] ;

[0099] ;

[0100] in, This is the learning rate coefficient, used to limit the update speed and ensure system stability.

[0101] In summary, by employing a closed-loop control strategy with the objective function of minimizing liquid level deviation, this application achieves: early dynamic suppression of capacity imbalance, avoiding severe liquid level deviation; full automation of the balancing process, eliminating the need for manual intervention or shutdown for mixing; significant reduction in balancing time and electrolyte loss, improving the system's available capacity and operational economy; and long-term maintenance of high accuracy and stability through an adaptive update mechanism.

[0102] The above description is merely a preferred embodiment of this application and is not intended to limit this application. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of this application should be included within the protection scope of this application.

Claims

1. An online capacity balancing method for vanadium redox flow battery systems, characterized in that, include: Real-time monitoring of the liquid level heights in the positive and negative electrolyte tanks; Calculate the liquid level difference and liquid level shift rate between the positive and negative electrolyte tanks; specifically, set a calculation time window; at each time point of the calculation time window, calculate the liquid level difference between the positive and negative electrolyte tanks to obtain the liquid level difference sequence; perform linear regression on the liquid level difference sequence over a period of time to calculate the liquid level shift rate. The solenoid valve control signal is generated based on the liquid level difference and liquid level deviation rate. The duty cycle of the solenoid valve between the positive and negative electrolyte tanks is adjusted to control the electrolyte flow between the two tanks and achieve capacity balance. Specifically, the liquid level difference is used as the core indicator for control optimization, and a corresponding loss function is constructed. A proportional-derivative control law is introduced, with the liquid level difference and liquid level deviation rate used as feedback variables. The proportional coefficient and derivative coefficient are adjusted to ensure that the solenoid valve opening control meets the condition of minimizing the loss function. When the liquid level difference is greater than the first set threshold, the solenoid valve is opened; when the liquid level difference is less than the second set threshold, the solenoid valve is closed. The proportional coefficient and derivative coefficient are periodically updated using recursive least squares or Bayesian estimation based on historical operating data. The relationship between the equilibrium flow rate and the liquid level difference after the solenoid valve is opened is determined through offline experiments. Based on the test results, the corresponding linear or nonlinear model is selected, and then the corresponding parameters are identified.

2. An online capacity balancing system for vanadium redox flow battery systems, employing the online capacity balancing method for vanadium redox flow battery systems as described in claim 1, characterized in that, It includes a liquid level detection module, a balancing pipeline, a solenoid valve assembly, and a control and calculation module, among which: The liquid level detection module is installed in the positive electrolyte tank and the negative electrolyte tank respectively, and is used to detect the liquid level height signal in real time; The balancing pipeline connects the positive electrolyte tank and the negative electrolyte tank, and is used to realize the automatic flow of electrolyte on both sides when the capacity is unbalanced; The solenoid valve assembly is installed on the balance pipeline and is used to control the flow of electrolyte between the positive electrolyte tank and the negative electrolyte tank; The control and calculation module is electrically connected to the liquid level detection module and the solenoid valve assembly, respectively, and is used to collect liquid level difference data, calculate liquid level offset rate, identify system parameters and output solenoid valve control signals.

3. The online capacity balancing system for vanadium redox flow battery systems according to claim 2, characterized in that, The liquid level detection module includes one or more of the following: ultrasonic liquid level gauge, differential pressure liquid level gauge, and magnetic float liquid level gauge.

4. The online capacity balancing system for vanadium redox flow battery systems according to claim 2, characterized in that, The balancing pipeline is provided with a bidirectional flow path, and the solenoid valve assembly includes two directional control valves, which respectively control the flow of electrolyte to the positive electrode or the negative electrode tank.

Images