Intelligent capacity management system for all-vanadium redox flow battery

Abstract

The invention provides an intelligent capacity management system for an all-vanadium redox flow battery, which comprises a control device and an electrolyte capacity recovery device, and is characterized in that the control device is used for acquiring the discharge capacity and charge-discharge cycle data of the all-vanadium redox flow battery after single discharge cycle; based on the charge-discharge cycle data, performing prediction through an electrolyte average valence state monitoring model to obtain an electrolyte average valence state of the current electrolyte; when the discharge capacity is smaller than a capacity threshold value, the electrolyte capacity recovery device is controlled to carry out liquid mixing operation; when the discharge capacity is larger than or equal to a capacity threshold value and the average valence state of the electrolyte is larger than an average valence state threshold value, the electrolyte capacity recovery device is controlled to carry out electroreduction capacity recovery operation; and automatic liquid mixing and electroreduction capacity recovery operations of the all-vanadium redox flow battery are realized.

Description

Technical Field

[0001] The present application relates to the technical field of all-vanadium liquid flow battery management, and in particular to an intelligent capacity management system for all-vanadium liquid flow batteries. Background Art

[0002] During the continuous operation of the all-vanadium liquid flow battery, the cross-membrane contamination of vanadium ions and the migration of water molecules across the membrane will cause self-discharge and imbalance of the positive and negative electrolytes, thereby causing the discharge capacity to decay. The charge and discharge capacity is an important indicator for measuring the battery's energy storage capacity.

[0003] The active substances in the positive and negative electrolytes of the all-vanadium flow battery are vanadium ions of different valence states, and the supporting electrolyte is sulfuric acid, which is highly acidic and stable. During the charge and discharge process, the vanadium ions and sulfuric acid present in the aqueous solution will not precipitate, volatilize or decompose. Therefore, the charge and discharge capacity can be restored by mixing the liquid, activating and other methods, thereby effectively extending the battery life.

[0004] In the prior art, all-vanadium liquid flow batteries use a mixed liquid method to recover capacity. The existing mixed liquid method has the following problems:

[0005] (1) During the operation of the all-vanadium liquid flow battery, there is a problem of hydrogen evolution, which reduces the number of electrons in the system, increases the average valence state, and leads to a decrease in the available capacity. This part of the capacity loss cannot be recovered by mixing the liquid.

[0006] (2) The existing method is to add gaseous reducing agents or liquid reducing agents to the electrolyte of the all-vanadium flow battery to perform chemical reduction recovery. However, the former, such as ammonia, hydrogen, carbon monoxide, etc., all have pollution or explosion risks; the latter, such as oxalic acid, propanol, etc., will produce water during the reaction, affecting the concentration of the electrolyte; and the existing flow battery monitoring device cannot monitor the average valence state of the electrolyte after mixing, and the number of electrons lost by the electrolyte cannot be known, so it is impossible to control the amount of reducing agent added.

[0007] (3) The current mixing method uses a mixing device that requires manual operation of valves and circulating pumps, which increases the labor cost of the operation and maintenance process. At the same time, the compactness of the pipeline network needs to be sacrificed in the design to provide sufficient space for manual operation. Summary of the invention

[0008] The present application aims to solve one of the technical problems in the related art at least to some extent.

[0009] To this end, the purpose of this application is to propose an intelligent capacity management system for an all-vanadium liquid flow battery, wherein the all-vanadium liquid flow battery includes a positive electrode side liquid storage tank and a negative electrode side liquid storage tank, and the system includes a control device and an electrolyte capacity recovery device, wherein the control device is used to:

[0010] Obtaining the discharge capacity and charge-discharge cycle data of the all-vanadium liquid flow battery after completing a single discharge cycle;

[0011] Based on the charge-discharge cycle data, an average valence state of the electrolyte is predicted by an average valence state monitoring model to obtain an average valence state of the electrolyte of the current electrolyte;

[0012] When the discharge capacity is less than a capacity threshold and the average valence state of the electrolyte is less than or equal to an average valence state threshold, controlling the electrolyte capacity recovery device to perform a liquid mixing operation;

[0013] When the discharge capacity is greater than or equal to a capacity threshold and the average valence state of the electrolyte is greater than an average valence state threshold, controlling the electrolyte capacity recovery device to perform an electric reduction capacity recovery operation;

[0014] When the discharge capacity is less than a capacity threshold and the average valence state of the electrolyte is greater than an average valence state threshold, the electrolyte capacity recovery device is controlled to first perform a liquid mixing operation and then perform an electric reduction capacity recovery operation.

[0015] In some implementations, the control device, when controlling the electrolyte capacity recovery device to perform an electric reduction capacity recovery operation, is used to:

[0016] Obtaining an electric reduction quantity based on an average valence state of the electrolyte;

[0017] The electrolyte capacity recovery device is controlled to perform an electro-reduction capacity recovery operation based on the electro-reduction quantity.

[0018] In some implementations, the control device obtains the electroreduction quantity based on the average valence state of the electrolyte by the following formula:

[0019]

[0020] Wherein, Q is the electroreduction quantity, x is the average valence state of the electrolyte predicted by the average valence state monitoring model of the electrolyte, a is the total amount of vanadium ions in the electrolyte, which is determined according to the vanadium ion concentration and volume of the all-vanadium liquid flow battery; and F is the Faraday constant.

[0021] In some implementations, the electrolyte capacity recovery device includes a first open-circuit primary battery, a second open-circuit primary battery, an electric reduction fluid battery, and a third open electrolyte container, wherein the first open-circuit primary battery and the second open-circuit primary battery each include an open electrolyte container and a closed container; the electric reduction fluid battery includes a first half-cell and a second half-cell;

[0022] The first opening of the open electrolyte container of the first open circuit primary battery is connected to the positive electrode side liquid storage tank through a first magnetic pump, and the second opening of the open electrolyte container of the first open circuit primary battery is connected to the negative electrode side liquid storage tank through a first pipeline, and a solenoid valve is provided on the first pipeline;

[0023] The first opening of the open electrolyte container of the second open-circuit primary battery is connected to the first magnetic pump, the first opening of the closed container of the second open-circuit primary battery is connected to the negative electrode side liquid storage tank through the second magnetic pump, a second pipeline with a solenoid valve is connected between the positive electrode side liquid storage tank and the negative electrode side liquid storage tank, and the open electrolyte container and the closed container of the second open-circuit primary battery are both connected to a third pipeline between the first half cell and the positive electrode side liquid storage tank, and a solenoid valve is provided on the third pipeline;

[0024] The first opening of the first half-cell is connected to the closed container of the second open-circuit primary cell, the first opening of the second half-cell is connected to the first opening of the third open electrolyte container, the second opening of the third open electrolyte container is connected to the second opening of the second half-cell through a fourth pipeline, and the fourth pipeline is provided with a third magnetic pump and a first solenoid valve.

[0025] In some implementations, the electric reduction fluid battery includes a diaphragm and electrodes, a sealing gasket, a bipolar plate, a current collecting plate, a metal end plate and an insulating end plate that are symmetrically arranged on both sides of the diaphragm. A flow channel groove is provided on the side of the bipolar plate facing the sealing gasket, and a reinforcing rib is provided on the metal end plate.

[0026] In some implementations, the control device, when controlling the electrolyte capacity recovery device to perform an electro-reduction capacity recovery operation based on the electro-reduction quantity, is used to:

[0027] Charging the all-vanadium liquid flow battery at a preset ratio of the rated power to a cut-off voltage condition;

[0028] The electric reduction fluid battery is turned on to discharge the electric reduction fluid battery at a constant current to the electric reduction amount; during the discharge process, the third magnetic pump and the first solenoid valve are turned on to enable the third open electrolyte container to continuously provide the reducing agent to the second half-cell.

[0029] In some implementations, the method for obtaining the average valence state monitoring model of the electrolyte includes:

[0030] Acquire multiple charge and discharge cycle data of the all-vanadium liquid flow battery and the average valence state of the electrolyte after each charge and discharge cycle; wherein the multiple charge and discharge cycle data include monitoring data collected during multiple charge and discharge cycles and characteristic data calculated, and the monitoring data is time series data;

[0031] Based on the multiple charge and discharge cycle data and the average valence state of the electrolyte after each charge and discharge cycle, a dual-channel deep neural network prediction model is trained to obtain an electrolyte average valence state monitoring model; wherein the monitoring data and the characteristic data are respectively used as input data of two channels of the dual-channel deep neural network prediction model, and the average valence state of the electrolyte after each charge and discharge cycle is used as target data of the dual-channel deep neural network prediction model.

[0032] In some implementations, the monitoring data include charging and discharging status, positive electrode side temperature, positive electrode side flow, positive electrode side liquid level, negative electrode side temperature, negative electrode side flow, negative electrode side liquid level, liquid storage tank pressure, battery system total voltage and battery system total current, and the characteristic data include the starting charging voltage, starting charging current, starting discharging voltage, starting discharging current, integrated average charging voltage, integrated average charging voltage, integrated average discharging voltage, integrated average discharging current, total charging charge, total discharging charge, total charging electrical energy, total discharging electrical energy, coulomb efficiency, voltage efficiency and energy efficiency of each charging and discharging cycle.

[0033] In some implementations, the dual-channel deep neural network prediction model includes a first channel, a second channel, a merging layer, a flattening layer, a second fully connected layer, and an output layer. The first channel includes a first input layer, a long short-term memory layer, and a first discard layer in sequence according to the order of data flow. The second channel includes a second input layer, a first fully connected layer, and a second discard layer in sequence according to the order of data flow. The data of the first discard layer and the second discard layer flow to the merging layer, the flattening layer, the second fully connected layer, and the output layer in sequence; the first channel receives two-dimensional time series data, and the first channel receives one-dimensional feature data.

[0034] In some implementations, the method for obtaining the average valence state of the electrolyte after each charge and discharge cycle is completed includes:

[0035] After each charge and discharge is completed, an electrolyte sample is taken out from the liquid storage tank, and the vanadium ion concentration in the electrolyte sample is titrated by potentiometric titration to obtain the average valence state of the electrolyte corresponding to the completion of the cycle.

[0036] The intelligent capacity management system for all-vanadium liquid flow batteries provided in the present application monitors the average valence state of the electrolyte during the charge and discharge cycle through an average valence state monitoring model of the electrolyte, and automatically controls the reduction of the electrolyte based on the obtained average valence state of the electrolyte and the electrolyte capacity recovery device; the control device fully automates the mixing of the liquid, valence state monitoring, and electro-reduction processes involved in the entire capacity recovery process, without the need for manual on-site inspections and operations, which not only improves the reliability of the system, but also saves labor and maintenance costs; the electrolyte capacity recovery device introduces an electro-reduction fluid battery on the positive electrode side of the all-vanadium liquid flow battery to achieve electrochemical reduction of the electrolyte, replenish the number of electrons lost due to the hydrogen evolution reaction during the operation of the all-vanadium liquid flow battery, and achieve 100% recovery of the electrolyte capacity; and the electrolyte capacity recovery device separates the reducing agent, the reduction product and the electrolyte, which will not affect the composition and concentration of the electrolyte; each pipeline of the electrolyte capacity recovery device is opened and closed by a solenoid valve to achieve automatic control, saving labor and maintenance costs; and there is no need to consider the manual operation space when designing the pipeline, which can further improve the compactness of the pipeline.

[0037] Additional aspects and advantages of the present application will be given in part in the description below, and in part will become apparent from the description below, or will be learned through the practice of the present application. BRIEF DESCRIPTION OF THE DRAWINGS

[0038] The above and / or additional aspects and advantages of the present application will become apparent and easily understood from the following description of the embodiments in conjunction with the accompanying drawings, in which:

[0039] Figure 1 A schematic diagram of the structure of an electrolyte capacity recovery device provided in an embodiment of the present application;

[0040] Figure 2 An exploded diagram of the structure of the electric reduction fluid battery provided in an embodiment of the present application;

[0041] Figure 3 A block diagram of a dual-channel deep neural network prediction model provided in an embodiment of the present application. DETAILED DESCRIPTION

[0042] Embodiments of the present application are described in detail below, and examples of the embodiments are shown in the accompanying drawings, wherein the same or similar reference numerals throughout represent the same or similar elements or elements having the same or similar functions. The embodiments described below with reference to the accompanying drawings are exemplary and are intended to be used to explain the present application, and should not be construed as limiting the present application.

[0043] The following describes an intelligent capacity management system for all-vanadium liquid flow batteries according to an embodiment of the present application with reference to the accompanying drawings.

[0044] The embodiment of the present application provides an intelligent capacity management system for an all-vanadium liquid flow battery, wherein the all-vanadium liquid flow battery includes a positive electrode side liquid storage tank and a negative electrode side liquid storage tank, and the intelligent capacity management system for the all-vanadium liquid flow battery includes: a control device and an electrolyte capacity recovery device, wherein the control device is used to:

[0045] Obtaining the discharge capacity and charge-discharge cycle data of the all-vanadium liquid flow battery after completing a single discharge cycle;

[0046] Based on the charge-discharge cycle data, an average valence state of the electrolyte is predicted by an average valence state monitoring model to obtain an average valence state of the electrolyte of the current electrolyte;

[0047] When the discharge capacity is less than a capacity threshold and the average valence state of the electrolyte is less than or equal to an average valence state threshold, controlling the electrolyte capacity recovery device to perform a liquid mixing operation;

[0048] When the discharge capacity is greater than or equal to a capacity threshold and the average valence state of the electrolyte is greater than an average valence state threshold, controlling the electrolyte capacity recovery device to perform an electric reduction capacity recovery operation;

[0049] When the discharge capacity is less than a capacity threshold and the average valence state of the electrolyte is greater than an average valence state threshold, the electrolyte capacity recovery device is controlled to first perform a liquid mixing operation and then perform an electric reduction capacity recovery operation.

[0050] As an implementation method, the discharge capacity after a single cycle is completed can be obtained through an electric energy meter.

[0051] Exemplarily, the capacity threshold is 80% of the rated nominal capacity, and the average valence threshold is 3.7. The overall logic of capacity management of the all-vanadium liquid flow battery is: if the discharge capacity is lower than 80% of the rated nominal capacity, the mixing operation is started; if the average valence of the electrolyte is higher than 3.7, the electro-reduction capacity recovery operation is started; when both mixing and electro-reduction are required, mixing is performed first and then electro-reduction.

[0052] In some embodiments, when controlling the electrolyte capacity recovery device to perform an electro-reduction capacity recovery operation, the control device is used to: obtain an electro-reduction amount based on the average valence state of the electrolyte; and control the electrolyte capacity recovery device to perform an electro-reduction capacity recovery operation based on the electro-reduction amount.

[0053] In some embodiments, the control device obtains the electroreduction quantity based on the average valence state of the electrolyte by the following formula:

[0054]

[0055] Among them, Q is the electroreduction quantity, x is the average valence state of the electrolyte predicted by the average valence state monitoring model of the electrolyte, and the unit is 1; a is the total amount of vanadium ions in the electrolyte, and the unit is mol, which is determined according to the vanadium ion concentration and volume of the all-vanadium liquid flow battery; F is the Faraday constant, which can be 96485, and the unit is C / mol.

[0056] In some embodiments, Figure 1 As shown, the electrolyte capacity recovery device includes a first open-circuit primary battery, a second open-circuit primary battery, an electric reduction fluid battery and a third open electrolyte container, wherein the first open-circuit primary battery and the second open-circuit primary battery both include an open electrolyte container and a closed container; the electric reduction fluid battery includes a first half cell 5 and a second half cell 6;

[0057] The first opening of the open electrolyte container of the first open circuit primary battery is connected to the positive electrode side liquid storage tank 1 through the first magnetic pump 18, and the second opening of the open electrolyte container of the first open circuit primary battery is connected to the negative electrode side liquid storage tank 2 through the first pipeline, and the first pipeline is provided with a solenoid valve;

[0058] The first opening of the open electrolyte container of the second open-circuit galvanic cell is connected to the first magnetic pump 18, the first opening of the closed container of the second open-circuit galvanic cell is connected to the negative electrode side liquid storage tank 2 through the second magnetic pump 19, a second pipeline with a solenoid valve is connected between the positive electrode side liquid storage tank 1 and the negative electrode side liquid storage tank 2, and the open electrolyte container and the closed container of the second open-circuit galvanic cell are both connected to a third pipeline between the first half cell 5 and the positive electrode side liquid storage tank 1, and a solenoid valve is provided on the third pipeline;

[0059] The first opening of the first half-cell 5 is connected to the closed container of the second open-circuit primary cell, the first opening of the second half-cell 6 is connected to the first opening of the third open electrolyte container, the second opening of the third open electrolyte container is connected to the second opening of the second half-cell 6 through a fourth pipeline, and the fourth pipeline is provided with a third magnetic pump 20 and a first solenoid valve 17.

[0060] like Figure 1As shown, the positive electrode side liquid storage tank 1 and the first half cell 5 are both connected to the positive electrode liquid inlet mother pipe of the liquid flow battery, and the positive electrode liquid inlet mother pipe of the liquid flow battery is provided with a second solenoid valve 16; the negative electrode side liquid storage tank 2 and the first open circuit primary battery are connected to the negative electrode liquid inlet mother pipe of the liquid flow battery, and the negative electrode liquid inlet mother pipe of the liquid flow battery is provided with a third solenoid valve 15. The first open circuit primary battery includes a first open electrolyte container 3 and a first closed container 4, and the second open circuit primary battery includes a second open electrolyte container 8 and a second closed container 9, wherein the first open electrolyte container 3 has a liquid inlet and a liquid outlet, and its material can be selected from corrosion-resistant materials such as PPH, PE, PP, and quartz glass, and the container is equipped with a graphite electrode. The first closed container 4 is a closed container for long-term storage of liquids. A V3.5+ electrolyte is placed in the container with a concentration range of 1.5-1.7 mol / L, preferably 1.7 mol / L. The material can be selected from corrosion-resistant materials such as PPH, PE, PP, and quartz glass, and the container is equipped with a graphite electrode. The graphite electrodes on the first open electrolyte container 3 and the first closed container 4 are connected through a voltage sensor to form an open-circuit primary cell. The first open electrolyte container 3 and the first closed container 4 are separated by an insulating and impermeable material. The first half-cell 5 and the second half-cell 6 are both half-cells with a liquid inlet and a liquid outlet. The first half-cell 5 and the second half-cell 6 are connected to form an electrolytic cell. The third open electrolyte container 7 is an open electrolyte container with a liquid inlet and a liquid outlet. The material can be selected from corrosion-resistant materials such as PPH, PE, PP, and quartz glass. The third open electrolyte container 7 provides the reducing agent required for electric reduction for the second half-cell 6. The second open electrolyte container 8 is an open electrolyte container with a liquid inlet and a liquid outlet. The material can be selected from corrosion-resistant materials such as PPH, PE, PP, and quartz glass. The container is equipped with graphite electrodes. The graphite electrodes on the second open electrolyte container 8 and the second closed container 9 are connected through a voltage sensor to form an open-circuit primary battery. The second open electrolyte container 8 and the second closed container 9 are separated by insulating and impermeable materials. The solenoid valves 10-17 and the magnetic pumps 18-20 are connected to the electronic control system of the control device through communication lines and are controlled by the capacity management software configured on the control device.

[0061] In some implementations, such as Figure 2 As shown, the electric reduction fluid battery includes a diaphragm and electrodes 26, a sealing gasket 25, a bipolar plate 24, a current collecting plate 23, a metal end plate 22 and an insulating end plate 21 which are symmetrically arranged on both sides of the diaphragm 27 as the center. A flow channel groove 29 is provided on the side of the bipolar plate 24 facing the sealing gasket 25, and a reinforcing rib 30 is provided on the metal end plate 22.

[0062] Among them, the insulating end plate 21 can be made of conventional engineering plastics; the metal end plate 22 can be made of aluminum alloy or stainless steel; the current collecting plate 23 can be made of copper plate or gold-plated copper plate; the bipolar plate 24 can be made of graphite; the sealing gasket 25 can be made of fluororubber, EPDM rubber or polytetrafluoroethylene; the electrode 25 can be made of graphite felt, carbon cloth, carbon paper and the like; the diaphragm 27 is preferably a perfluorosulfonic acid proton exchange membrane; the flow channel 29 on the bipolar plate 24 is preferably various types of bionic flow channels, Figure 2 The middle is an embodiment of the alveolar type flow channel; the sealing gasket 25, the bipolar plate 24, the current collecting plate 23, the metal end plate 22 and the insulating end plate 21 are all provided with bolt holes 28 for fastening by bolts.

[0063] It should be noted that the electric reduction flow battery is connected to the positive electrode side pipeline of the all-vanadium liquid flow battery, and the reducing agent can be selected from active substances such as sodium sulfite and ferrous chloride that do not generate gas in the negative electrode reaction. The collecting plate of the electric reduction flow battery is connected to the discharge controller of the control device through a wire, and the discharge of the electric reduction flow battery is controlled by the discharge controller.

[0064] In some implementations, when the control device controls the electrolyte capacity recovery device to perform an electric reduction capacity recovery operation based on the electric reduction amount, it is used to: charge the all-vanadium liquid flow battery at a preset ratio of the rated power to a cut-off voltage condition; turn on the electric reduction fluid battery to discharge the electric reduction fluid battery at a constant current to the electric reduction amount; during the discharge process, turn on the third magnetic pump and the first solenoid valve to allow the third open electrolyte container to continuously supply the reducing agent to the second half-cell.

[0065] The product of discharge time and discharge current is the discharge amount, which is the electro-reduction amount Q. The discharge time can be obtained by dividing the calculated electro-reduction amount by the constant current when the electro-reduction fluid battery is discharged. The discharge time of the electro-reduction fluid battery is controlled according to the discharge time, thereby achieving control of the amount of reducing agent based on the average valence state of the electrolyte.

[0066] The following describes respectively the liquid mixing operation and the electric reduction capacity recovery operation of the control device controlling the electrolyte capacity recovery device.

[0067] Mixing operation: The control device sends instructions to the solenoid valve and magnetic pump of the electrolyte capacity recovery device to control the electrolyte flow direction in the pipeline of the electrolyte capacity recovery device. The specific control logic is: after the mixing starts, the third solenoid valve 15 and the second solenoid valve 16 are automatically closed, and the fourth solenoid valve 10 to the eighth solenoid valve 14 (i.e. Figure 110, 11, 12, 13, 14) are turned on, the first magnetic pump 18 and the second magnetic pump 19 are turned on, and feedback adjustment is performed according to the liquid level sensor data on the positive side liquid storage tank 1 and the negative side liquid storage tank 2 to ensure that the liquid levels in the positive side liquid storage tank 1 and the negative side liquid storage tank 2 are equal. The open circuit voltage of the primary battery composed of the second open electrolyte container 8 and the second closed container 9 is collected and monitored by a voltage sensor. When the open circuit voltage is less than or equal to 0.01V, it means that the electrolytes in the positive side liquid storage tank 1 and the negative side liquid storage tank 2 are basically the same, and the mixing is sufficient.

[0068] Electric reduction capacity recovery operation: When the need for electric reduction is detected after a certain charge and discharge cycle, the all-vanadium liquid flow battery is first charged at 10% of the rated power to the cut-off voltage condition. For example, if the system rated power is 50kW, the charging power at this time is 5kW. Then, the electric reduction fluid battery composed of the first half cell 5 and the second half cell 6 is turned on by the discharge controller, so that the electric reduction fluid battery is discharged at a constant current. During the discharge process, the third magnetic pump 20 and the first solenoid valve 17 are continuously turned on to ensure the circulation of the electrolyte in the first half cell 5 and to continuously provide the reducing agent to the second half cell 6. The discharge time of the electric reduction fluid battery is determined by dividing the calculated discharge amount Q by the constant current. Alternatively, the current average valence state of the electrolyte monitored by the average valence state monitoring model of the electrolyte determines whether the average valence state of the electrolyte has been restored. If it has been restored, the electric reduction process is stopped.

[0069] In some embodiments, the method for obtaining the average valence state monitoring model of the electrolyte includes:

[0070] Acquire multiple charge and discharge cycle data of the all-vanadium liquid flow battery and the average valence state of the electrolyte after each charge and discharge cycle; wherein the multiple charge and discharge cycle data include monitoring data collected during multiple charge and discharge cycles and characteristic data calculated, and the monitoring data is time series data;

[0071] Based on the multiple charge and discharge cycle data and the average valence state of the electrolyte after each charge and discharge cycle, a dual-channel deep neural network prediction model is trained to obtain an electrolyte average valence state monitoring model; wherein the monitoring data and the characteristic data are respectively used as input data of two channels of the dual-channel deep neural network prediction model, and the average valence state of the electrolyte after each charge and discharge cycle is used as target data of the dual-channel deep neural network prediction model.

[0072] In some embodiments, the monitoring data include charging and discharging status, positive electrode side temperature, positive electrode side flow, positive electrode side liquid level, negative electrode side temperature, negative electrode side flow, negative electrode side liquid level, liquid storage tank air pressure, battery system total voltage and battery system total current, and the characteristic data include the starting charging voltage, starting charging current, starting discharge voltage, starting discharge current, integrated average charging voltage, integrated average charging voltage, integrated average discharge voltage, integrated average discharge current, total charging charge, total discharge charge, total charging electrical energy, total discharge electrical energy, coulomb efficiency, voltage efficiency and energy efficiency of each charging and discharging cycle.

[0073] In some embodiments, Figure 3 As shown, the dual-channel deep neural network prediction model includes a first channel, a second channel, a merging layer, a flattening layer, a second fully connected layer, and an output layer. The first channel includes a first input layer, a long short-term memory layer, and a first discard layer in sequence according to the order of data flow. The second channel includes a second input layer, a first fully connected layer, and a second discard layer in sequence according to the order of data flow. The data of the first discard layer and the second discard layer flow to the merging layer, the flattening layer, the second fully connected layer, and the output layer in sequence; the first channel receives two-dimensional time series data, and the first channel receives one-dimensional feature data.

[0074] In some embodiments, the method for obtaining the average valence state of the electrolyte after each charge and discharge cycle is completed includes:

[0075] After each charge and discharge is completed, an electrolyte sample is taken out from the liquid storage tank, and the vanadium ion concentration in the electrolyte sample is titrated by potentiometric titration to obtain the average valence state of the electrolyte corresponding to the completion of the cycle.

[0076] For example, taking a vanadium liquid flow battery product with a storage time of 4 hours as an example, the model training method includes the following steps:

[0077] The first step is data preparation.

[0078] The data of the first 10 charge and discharge cycles of the battery are used as training data. The data can be collected by the battery management system of the battery, and the sampling interval of the time series data is 60 seconds. Preprocess this part of the data. The preprocessing method is to separate the time series of monitoring data such as temperature, flow, liquid level, voltage, current, and air pressure obtained in each cycle into sequence samples containing 20 sampling points. In this way, each charge and discharge cycle will contain several two-dimensional array samples of size 11×20 as the input of the first channel. In addition, the starting charge voltage, starting charge current, starting discharge voltage, starting discharge current, integrated average charge voltage, integrated average charge voltage, integrated average discharge voltage, integrated average discharge current, total charge charge, total discharge charge, total charge energy, total discharge energy, coulomb efficiency, voltage efficiency, and energy efficiency of each charge and discharge cycle are calculated as the input of the second channel. These data can be calculated in a conventional way based on the voltage and current time series, and will not be repeated here. In this way, each charge and discharge cycle contains a set of 15×1 one-dimensional arrays composed of the above features as the input of channel 2. In addition, after each charge and discharge, a small amount of electrolyte sample is taken from the storage tank, and the vanadium ion concentration is titrated by potentiometric titration to obtain the average valence state of the electrolyte corresponding to the completion of the cycle, and this value is used as the training target. In this way, the training data is ready.

[0079] The second step is data preprocessing.

[0080] The data obtained in the first step are preprocessed by conventional data cleaning and normalization methods.

[0081] The third step is model training.

[0082] The dual-channel deep neural network prediction model is constructed using the preprocessed data training. During the training process, particle swarm optimization can be used to optimize hyperparameters such as training algorithm, learning rate, number of training steps, and number of neurons.

[0083] The first channel receives a 2D array input of size 11×20 (2D time series data), where the first dimension is the feature dimension, representing monitoring data such as charge and discharge status, positive electrode side temperature, positive electrode side flow, positive electrode side liquid level, negative electrode side temperature, negative electrode side flow, negative electrode side liquid level, tank pressure, battery system total voltage, battery system total current, etc.; the second dimension is the time dimension, representing 20 sampling point data with a sampling interval of 60 seconds. The first channel is divided into the first input layer, long short-term memory layer (containing 128 neurons) and the first discard layer.

[0084] The second channel receives 15×1 one-dimensional feature data. The selected features include starting charge voltage, starting charge current, starting discharge voltage, starting discharge current, integrated average charge voltage, integrated average charge voltage, integrated average discharge voltage, integrated average discharge current, total charge charge, total discharge charge, total charge energy, total discharge energy, Coulomb efficiency, voltage efficiency, and energy efficiency. The second channel is divided into the second input layer, the first fully connected layer (containing 128 neurons), and the second drop layer. The data streams of the two channels converge to the merging layer after the first drop layer and the second drop layer and merge into the same two-dimensional array. The merging layer is followed by the flattening layer, which converts the two-dimensional array into a one-dimensional array, passes through the second fully connected layer (containing 128 neurons), and is finally mapped to the output layer. The output layer contains 1 neuron, which is used to output the average valence state of the electrolyte.

[0085] At this point, the dual-channel neural network training for predicting the average valence state of the electrolyte is completed, and a trained average valence state monitoring model of the electrolyte is obtained. The average valence state monitoring model of the electrolyte can predict the current average valence state of the electrolyte based on a set of charge and discharge cycle data of the battery.

[0086] The intelligent capacity management system of the all-vanadium liquid flow battery of the embodiment of the present application monitors the average valence state of the electrolyte in the charge and discharge cycle through the average valence state monitoring model of the electrolyte, and automatically controls the reduction of the electrolyte based on the obtained average valence state of the electrolyte and the electrolyte capacity recovery device; the control device fully automates the mixing, valence state monitoring, and electro-reduction processes involved in the entire capacity recovery process, without the need for manual on-site inspections and operations, which not only improves the reliability of the system, but also saves labor and maintenance costs; the electrolyte capacity recovery device introduces an electro-reduction fluid battery on the positive electrode side of the all-vanadium liquid flow battery to achieve electrochemical reduction of the electrolyte, replenish the number of electrons lost due to the hydrogen evolution reaction during the operation of the all-vanadium liquid flow battery, and achieve 100% recovery of the electrolyte capacity; and the electrolyte capacity recovery device separates the reducing agent, the reduction product and the electrolyte, which will not affect the composition and concentration of the electrolyte; each pipeline of the electrolyte capacity recovery device is opened and closed by a solenoid valve to achieve automatic control, saving labor and maintenance costs; and there is no need to consider the manual operation space when designing the pipeline, which can further improve the compactness of the pipeline.

[0087] In the description of the aforementioned embodiments, the description with reference to the terms "one embodiment", "some embodiments", "example", "specific example", or "some examples" etc. means that the specific features, structures, materials or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present application. In this specification, the schematic representations of the above terms do not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described may be combined in any one or more embodiments or examples in a suitable manner. In addition, those skilled in the art may combine and combine the different embodiments or examples described in this specification and the features of the different embodiments or examples, without contradiction.

[0088] In addition, the terms "first" and "second" are used for descriptive purposes only and should not be understood as indicating or implying relative importance or implicitly indicating the number of the indicated technical features. Therefore, the features defined as "first" and "second" may explicitly or implicitly include at least one of the features. In the description of this application, the meaning of "plurality" is at least two, such as two, three, etc., unless otherwise clearly and specifically defined.

[0089] Any process or method description in a flowchart or otherwise described herein may be understood to represent a module, fragment or portion of code comprising one or more executable instructions for implementing the steps of a custom logical function or process, and the scope of the preferred embodiments of the present application includes alternative implementations in which functions may not be performed in the order shown or discussed, including performing functions in a substantially simultaneous manner or in the reverse order depending on the functions involved, which should be understood by technicians in the technical field to which the embodiments of the present application belong.

[0090] A person skilled in the art may understand that all or part of the steps in the method for implementing the above-mentioned embodiment may be completed by instructing related hardware through a program, and the program may be stored in a computer-readable storage medium, which, when executed, includes one or a combination of the steps of the method embodiment.

[0091] In addition, each functional unit in each embodiment of the present application may be integrated into a processing module, or each unit may exist physically separately, or two or more units may be integrated into one module. The above-mentioned integrated module may be implemented in the form of hardware or in the form of a software functional module. If the integrated module is implemented in the form of a software functional module and sold or used as an independent product, it may also be stored in a computer-readable storage medium.

[0092] Although the embodiments of the present application have been shown and described above, it can be understood that the above embodiments are exemplary and cannot be understood as limitations on the present application. Ordinary technicians in this field can change, modify, replace and modify the above embodiments within the scope of the present application.

Claims

1. An intelligent capacity management system for an all-vanadium liquid flow battery, the all-vanadium liquid flow battery comprising a positive electrode side liquid storage tank and a negative electrode side liquid storage tank, characterized in that: The system comprises a control device and an electrolyte capacity recovery device, wherein the control device is used to: Obtaining the discharge capacity and charge-discharge cycle data of the all-vanadium liquid flow battery after completing a single discharge cycle; Based on the charge-discharge cycle data, an average valence state of the electrolyte is predicted by an average valence state monitoring model to obtain an average valence state of the electrolyte of the current electrolyte; When the discharge capacity is less than a capacity threshold and the average valence state of the electrolyte is less than or equal to an average valence state threshold, controlling the electrolyte capacity recovery device to perform a liquid mixing operation; When the discharge capacity is greater than or equal to a capacity threshold and the average valence state of the electrolyte is greater than an average valence state threshold, controlling the electrolyte capacity recovery device to perform an electric reduction capacity recovery operation; When the discharge capacity is less than a capacity threshold and the average valence state of the electrolyte is greater than an average valence state threshold, the electrolyte capacity recovery device is controlled to first perform a liquid mixing operation and then perform an electric reduction capacity recovery operation.

2. The intelligent capacity management system for all-vanadium redox flow batteries according to claim 1 is characterized in that: When the control device controls the electrolyte capacity recovery device to perform an electric reduction capacity recovery operation, the control device is used to: Obtaining an electric reduction quantity based on an average valence state of the electrolyte; The electrolyte capacity recovery device is controlled to perform an electro-reduction capacity recovery operation based on the electro-reduction quantity.

3. The intelligent capacity management system for all-vanadium redox flow batteries according to claim 2 is characterized in that: The control device obtains the electric reduction quantity based on the average valence state of the electrolyte by the following formula: Wherein, Q is the electroreduction quantity, x is the average valence state of the electrolyte predicted by the average valence state monitoring model of the electrolyte, a is the total amount of vanadium ions in the electrolyte, which is determined according to the vanadium ion concentration and volume of the all-vanadium liquid flow battery; and F is the Faraday constant.

4. The intelligent capacity management system for all-vanadium redox flow batteries according to claim 2 is characterized in that: The electrolyte capacity recovery device comprises a first open circuit primary battery, a second open circuit primary battery, an electric reduction fluid battery and a third open electrolyte container, wherein the first open circuit primary battery and the second open circuit primary battery both comprise an open electrolyte container and a closed container; the electric reduction fluid battery comprises a first half cell and a second half cell; The first opening of the open electrolyte container of the first open circuit primary battery is connected to the positive electrode side liquid storage tank through a first magnetic pump, and the second opening of the open electrolyte container of the first open circuit primary battery is connected to the negative electrode side liquid storage tank through a first pipeline, and a solenoid valve is provided on the first pipeline; The first opening of the open electrolyte container of the second open-circuit primary battery is connected to the first magnetic pump, the first opening of the closed container of the second open-circuit primary battery is connected to the negative electrode side liquid storage tank through the second magnetic pump, a second pipeline with a solenoid valve is connected between the positive electrode side liquid storage tank and the negative electrode side liquid storage tank, and the open electrolyte container and the closed container of the second open-circuit primary battery are both connected to a third pipeline between the first half cell and the positive electrode side liquid storage tank, and a solenoid valve is provided on the third pipeline; The first opening of the first half-cell is connected to the closed container of the second open-circuit primary cell, the first opening of the second half-cell is connected to the first opening of the third open electrolyte container, the second opening of the third open electrolyte container is connected to the second opening of the second half-cell through a fourth pipeline, and the fourth pipeline is provided with a third magnetic pump and a first solenoid valve.

5. The intelligent capacity management system for all-vanadium redox flow batteries according to claim 4 is characterized in that: The electric reduction fluid battery includes a diaphragm and electrodes, a sealing gasket, a bipolar plate, a current collecting plate, a metal end plate and an insulating end plate symmetrically arranged on both sides of the diaphragm. A flow channel groove is arranged on the side of the bipolar plate facing the sealing gasket, and a reinforcing rib is arranged on the metal end plate.

6. The intelligent capacity management system for all-vanadium redox flow batteries according to claim 4 is characterized in that: When the control device controls the electrolyte capacity recovery device to perform an electro-reduction capacity recovery operation based on the electro-reduction quantity, the control device is used to: Charging the all-vanadium liquid flow battery at a preset ratio of the rated power to a cut-off voltage condition; Turning on the electric reduction fluid battery to discharge the electric reduction fluid battery to the electric reduction amount at a constant current; During the discharge process, the third magnetic pump and the first electromagnetic valve are turned on so that the third open electrolyte container continuously provides the reducing agent to the second half-cell.

7. The intelligent capacity management system for all-vanadium redox flow batteries according to claim 1 is characterized in that: The method for obtaining the average valence state monitoring model of the electrolyte comprises: Acquire multiple charge and discharge cycle data of the all-vanadium liquid flow battery and the average valence state of the electrolyte after each charge and discharge cycle; wherein the multiple charge and discharge cycle data include monitoring data collected during multiple charge and discharge cycles and characteristic data calculated, and the monitoring data is time series data; Based on the multiple charge and discharge cycle data and the average valence state of the electrolyte after each charge and discharge cycle, a dual-channel deep neural network prediction model is trained to obtain an electrolyte average valence state monitoring model; wherein the monitoring data and the characteristic data are respectively used as input data of two channels of the dual-channel deep neural network prediction model, and the average valence state of the electrolyte after each charge and discharge cycle is used as target data of the dual-channel deep neural network prediction model.

8. The intelligent capacity management system for all-vanadium redox flow batteries according to claim 7 is characterized in that: The monitoring data include charging and discharging status, positive electrode side temperature, positive electrode side flow, positive electrode side liquid level, negative electrode side temperature, negative electrode side flow, negative electrode side liquid level, liquid storage tank pressure, battery system total voltage and battery system total current; the characteristic data include the starting charging voltage, starting charging current, starting discharging voltage, starting discharging current, integrated average charging voltage, integrated average charging voltage, integrated average discharging voltage, integrated average discharging current, total charging charge, total discharging charge, total charging electrical energy, total discharging electrical energy, coulomb efficiency, voltage efficiency and energy efficiency of each charging and discharging cycle.

9. The intelligent capacity management system for all-vanadium redox flow batteries according to claim 7 is characterized in that: The dual-channel deep neural network prediction model includes a first channel, a second channel, a merging layer, a flattening layer, a second fully connected layer, and an output layer. The first channel includes a first input layer, a long short-term memory layer, and a first discard layer in sequence according to the order of data flow. The second channel includes a second input layer, a first fully connected layer, and a second discard layer in sequence according to the order of data flow. The data of the first discard layer and the second discard layer flow to the merging layer, the flattening layer, the second fully connected layer, and the output layer in sequence; the first channel receives two-dimensional time series data, and the first channel receives one-dimensional feature data.

10. The intelligent capacity management system for all-vanadium redox flow batteries according to claim 7, characterized in that: The method for obtaining the average valence state of the electrolyte after each charge and discharge cycle is completed comprises: After each charge and discharge is completed, an electrolyte sample is taken out from the liquid storage tank, and the vanadium ion concentration in the electrolyte sample is titrated by potentiometric titration to obtain the average valence state of the electrolyte corresponding to the completion of the cycle.

Images