A pure electric unmanned ship BMS battery management system

Abstract

The invention relates to the field of BMS battery management, and discloses a pure electric unmanned ship BMS battery management system. The data acquisition module is used for performing simulation experiments on battery cells and constructing experimental attenuation curves of the battery cells, specifically comprising: collecting the total current and total voltage of the unmanned ship at fixed time intervals to determine the operating power of the unmanned ship at each moment; determining the installation quantity of a target battery pack and battery cells in the unmanned ship, and calculating the output current and output voltage of the battery cells at each moment according to the installation quantity; constructing an experimental attenuation curve based on experimental measurement, and each moment of the experimental attenuation curve includes the battery cell capacity, voltage and current; a heat dissipation module is used for constructing a heat dissipation model of multiple battery cells in a target battery pack to obtain a heat dissipation coefficient of each battery cell; and a prediction module is used for predicting the capacity attenuation of the actual work of the battery cell according to the capacity attenuation curve of the battery cell and the heat dissipation coefficient of the battery cell to obtain an actual attenuation curve.

Description

Technical Field

[0001] The present invention relates to the field of BMS battery management, and more specifically, to a pure electric unmanned ship BMS battery management system. Background Art

[0002] The battery management system is the core component of pure electric equipment and is responsible for monitoring and managing each battery cell in the battery pack.

[0003] However, although the existing BMS battery management system can realize the scheduling and management of battery cells to a certain extent, it does not pay enough attention to the differences in battery cell attenuation, resulting in uneven attenuation rates among battery cells. This imbalance will cause the overall performance of the battery pack to decline, shorten its service life, and increase maintenance costs. Summary of the invention

[0004] The present invention provides a pure electric unmanned ship BMS battery management system to solve the technical problems raised in the background technology.

[0005] In a first aspect, the present invention provides a pure electric unmanned ship BMS battery management system, comprising:

[0006] The data acquisition module is used to simulate the battery cell and construct the experimental attenuation curve of the battery cell, including:

[0007] The total current and total voltage of the unmanned ship are collected at fixed time intervals to determine the operating power of the unmanned ship at each moment;

[0008] Determine the number of target battery packs and cells to be installed on the unmanned ship, and calculate the output current and output voltage of the cells at each moment based on the number of cells installed;

[0009] Construct an experimental decay curve based on experimental measurements, where each moment of the experimental decay curve includes cell capacity, voltage, and current;

[0010] The heat dissipation module is used to construct a heat dissipation model of multiple cells in the target battery pack and obtain the heat dissipation coefficient of each cell;

[0011] A prediction module is used to predict the actual working capacity attenuation of the battery cell according to the capacity attenuation curve of the battery cell and the heat dissipation coefficient of the battery cell to obtain an actual attenuation curve;

[0012] Deviation analysis module, used to perform spatiotemporal analysis on the fitting deviation between the experimental attenuation curve and the actual attenuation curve, and determine the deviation factor of the unmanned ship;

[0013] The attenuation balance module is used to perform a second prediction of the battery cell in combination with the deviation factor, and adjust the output of the battery cell based on the second prediction result to balance the attenuation capacity of each battery cell.

[0014] Furthermore, a heat dissipation model of multiple cells in the target battery pack is constructed, and the calculation formula of the heat dissipation model is as follows:

[0015] ;in, Indicates The heat dissipation coefficient of each battery cell, Indicates The number of sub-layers of the thermal conductive plate that each cell contacts, Indicates The index of the number of sub-layers of the thermal conductive plate that each cell contacts, Indicates Thermal conductivity of the heat conducting plate, Indicates The first heat conducting plate and the The contact area of ​​each cell contact, Indicates The thickness of the heat conducting board layer, Indicates The distance between each cell and the geometric center of the target battery pack, Indicates The number of adjacent cells. Indicates The battery cell and The Euclidean distance of the cells, Indicates the wind speed of the heat sink of the target heat sink, represents the weight of the wind speed of the heat sink of the target heat sink, represents the temperature gradient on the heat conducting plate, represents the weight of the temperature gradient on the heat conducting plate, Indicates The Euclidean distance between each battery cell and the heat sink is Indicates The weight parameter of the Euclidean distance between each battery cell and the heat sink.

[0016] Furthermore, the actual working capacity attenuation of the battery cell is predicted based on the capacity attenuation curve of the battery cell and the heat dissipation coefficient of the battery cell, including:

[0017] The capacity decay curve of the battery cell is segmented by a sliding window according to the time interval, and the current and voltage data in the sliding window are used to construct a feature vector as a training sample, and the capacity at the next moment of the sliding window is used as the sample label of the input;

[0018] The preset model is trained through training samples and sample labels to obtain an initial prediction model;

[0019] The initial prediction model is used to independently predict each battery cell in the target battery pack to obtain a prediction result;

[0020] A composite loss function is constructed based on the heat dissipation coefficient of the battery cell. The formula of the composite loss function is as follows:

[0021] ;

[0022] in, represents the composite loss function of the initial prediction model; Indicates The battery cell and The similarity of the cells, and Respectively represent The battery cell and The heat dissipation coefficient of each battery cell; Represents the initial forecast model prediction The moment The capacity of each battery cell; Represents the initial forecast model prediction The moment The capacity of each battery cell; express The moment Sample label of each battery cell, and denote the first loss coefficient and the second loss coefficient respectively;

[0023] The initial prediction model is trained through a composite loss function to obtain an updated prediction model.

[0024] Furthermore, the fitting deviation between the experimental attenuation curve and the actual attenuation curve is analyzed in time and space to determine the deviation factor of the unmanned ship, including:

[0025] Collecting total current and total voltage data of the target unmanned ship during a first preset time period;

[0026] The total current and total voltage data of the target unmanned ship in the first preset time period are averaged according to the number of battery cells, and the average value is input into the updated prediction model to obtain the predicted capacity of each battery cell of the target unmanned ship in the second preset time period;

[0027] Constructing a battery cell experimental attenuation curve of the target unmanned ship according to the predicted capacity of each battery cell of the target unmanned ship in the second preset time period;

[0028] Select some cells of the target unmanned ship for measurement and calculation to obtain the actual attenuation curve of these cells;

[0029] Compare the actual decay curve of the battery cells with the experimental decay curve to determine the fitting deviation between the experimental decay curve and the actual decay curve;

[0030] Performing a spatiotemporal analysis on the target unmanned ship within a second preset time period, determining a mechanical heat source and an operating environment of the target unmanned ship, and collecting a temperature of the mechanical heat source and an operating environment of the target unmanned ship within the second preset time period as a deviation factor;

[0031] Based on the deviation factor, a correction layer is added to the updated prediction model, and the correction layer is used to correct the output of the updated prediction model.

[0032] Furthermore, the formula of the correction layer is as follows:

[0033] ;

[0034] in, Indicates The cell capacity after calibration at a certain moment; Indicates Update the predicted cell capacity output by the prediction model at each moment; Indicates the effect of the temperature of the mechanical heat source on the cell attenuation, represents the weight coefficient of the first external heat source, Indicates that the unmanned ship is The temperature of the mechanical heat source at a certain moment, represents the maximum value of the mechanical heat source temperature within the second preset time period, Indicates that the base is a natural constant The exponential function of Indicates the effect of water temperature on cell attenuation; The standard temperature of seawater. Indicates The temperature of the sea water at that moment, represents the exponential decay coefficient, , represents the natural logarithm; represents the combined effect of mechanical heat source temperature and seawater temperature on cell degradation, and They represent the mechanical heat source index control coefficient and the seawater temperature index control coefficient respectively. , , Indicates the maximum temperature difference between the mechanical heat source and the seawater temperature.

[0035] Furthermore, the output of the battery cell is adjusted based on the second prediction result to balance the attenuated capacity of each battery cell, including:

[0036] Randomly generate an initial population with R individuals that meet the constraints, and initialize the current iteration number Time to 1;

[0037] Constraints:

[0038] Determine and adjust the output current or output voltage of the battery cell according to the series and parallel adjustment of the unmanned boat;

[0039] The total output voltage of the battery cell is equal to the total required voltage of the unmanned ship, and the total output current of the battery cell is equal to the total required current of the unmanned ship;

[0040] The output current or output voltage of the battery cell needs to be adjusted within the preset adjustment range;

[0041] Objective function:

[0042] ;in, It indicates that the correction layer based on the updated prediction model outputs the individual codes to obtain the maximum value of the predicted capacity of the battery cell. It indicates that the correction layer based on the updated prediction model outputs the individual codes to obtain the minimum value of the predicted capacity of the battery cell. It means that the square of the difference between the maximum value and the minimum value in the predicted capacity of the battery cell is minimized by taking the correction layer based on the updated prediction model to output the individual code;

[0043] Output the objective function value for R individuals, retain the individuals whose value is greater than the preset objective function threshold, and update the remaining individuals as parents;

[0044] If the current number of iterations Time ≥ the maximum number of iterations, stop updating the individual with the highest output objective function value as the output solution for the battery cell; otherwise, return to update the parent generation.

[0045] Further, the remaining individuals are updated as parents, including:

[0046] pass Calculate the current individual temperature; where, Indicates the current individual temperature, represents the maximum number of iterations, Indicates the current individual's basal temperature;

[0047] ;in, It represents the average difference between the objective function values ​​of R-1 individuals and the individual with the highest current objective function value. Represents the difference between the current individual and the individual with the highest current objective function value;

[0048] pass Calculate the distance between the current individual and the individual with the highest current objective function value; where, Represents the distance between the current individual and the individual with the highest current objective function value, represents the scaling function, represents the individual with the highest current objective function value, Represents a constant parameter, which is used to adjust the position offset between the current individual and the individual with the highest current objective function value. Indicates the current individual's position, Indicates the absolute value operation;

[0049] ;in, and They represent control parameters, which are used to control the scaling of the scaling function. Represents the natural base;

[0050] pass Determine the position adjustment parameters of the current individual; where, Represents the position adjustment parameter, which is used to control the movement amplitude of the individual. Represents a constant parameter used to control the position adjustment amplitude. Represents the position accuracy adjustment amount, which determines the accuracy of the movement based on the gap between the current individual and the individual with the highest current objective function value. , Represents a random number, used to introduce randomness;

[0051] The position of the current individual is updated according to the current individual temperature, the distance between the current individual and the individual with the highest current objective function value, and the position adjustment parameter. The calculation formula is as follows:

[0052] ;in, Represents the updated individual position.

[0053] In the second aspect, a pure electric unmanned ship BMS battery management method is applied to the pure electric unmanned ship BMS battery management system, comprising:

[0054] Conduct simulation experiments on the battery cells and construct the experimental attenuation curve of the battery cells, including:

[0055] The total current and total voltage of the unmanned ship are collected at fixed time intervals to determine the operating power of the unmanned ship at each moment;

[0056] Determine the number of target battery packs and cells to be installed on the unmanned ship, and calculate the output current and output voltage of the cells at each moment based on the number of cells installed;

[0057] Construct an experimental decay curve based on experimental measurements, where each moment of the experimental decay curve includes cell capacity, voltage, and current;

[0058] Construct a heat dissipation model for multiple cells in the target battery pack and obtain the heat dissipation coefficient of each cell;

[0059] The actual capacity attenuation of the battery cell is predicted based on the capacity attenuation curve of the battery cell and the heat dissipation coefficient of the battery cell to obtain the actual attenuation curve;

[0060] The fitting deviation between the experimental attenuation curve and the actual attenuation curve is analyzed in time and space to determine the deviation factor of the unmanned ship;

[0061] The deviation factor is combined to make a second prediction of the battery cell, and the output of the battery cell is adjusted based on the second prediction result to balance the attenuated capacity of each battery cell.

[0062] According to a third aspect, a computing device includes:

[0063] one or more processors;

[0064] A storage device is used to store one or more programs. When the one or more programs are executed by one or more processors, the one or more processors implement a pure electric unmanned ship BMS battery management system as described above.

[0065] In a fourth aspect, a computer-readable storage medium stores a program, which, when executed by a processor, implements a pure electric unmanned ship BMS battery management system as described above.

[0066] The beneficial effect of the present invention is that through accurate prediction of battery cell attenuation, intelligent heat dissipation management and dynamic environmental correction, efficient battery management of pure electric unmanned ships in complex environments is achieved. By balancing battery cell consumption and optimizing energy consumption, the system not only extends the overall life of the battery pack, but also improves the endurance and operational safety of the unmanned ship, ensuring its stability and reliability in long-term missions. BRIEF DESCRIPTION OF THE DRAWINGS

[0067] Figure 1 This is an architecture diagram of a pure electric unmanned ship BMS battery management system of the present invention;

[0068] Figure 2 It is a flow chart of a BMS battery management method for a pure electric unmanned ship of the present invention. DETAILED DESCRIPTION

[0069] The subject matter described herein will now be discussed with reference to example embodiments. It should be understood that the discussion of these embodiments is only to enable those skilled in the art to better understand and implement the subject matter described herein, and the functions and arrangements of the elements discussed may be changed without departing from the scope of protection of the contents of this specification. Each example may omit, replace or add various processes or components as needed. In addition, the features described relative to some examples may also be combined in other examples.

[0070] like Figure 1-2As shown, a pure electric unmanned ship BMS battery management system includes:

[0071] The data acquisition module is used to simulate the battery cell and construct the experimental attenuation curve of the battery cell, including:

[0072] The total current and total voltage of the unmanned ship are collected at fixed time intervals to determine the operating power of the unmanned ship at each moment;

[0073] Determine the number of target battery packs and cells to be installed on the unmanned ship, and calculate the output current and output voltage of the cells at each moment based on the number of cells installed;

[0074] Construct an experimental decay curve based on experimental measurements, where each moment of the experimental decay curve includes cell capacity, voltage, and current;

[0075] The heat dissipation module is used to construct a heat dissipation model of multiple cells in the target battery pack and obtain the heat dissipation coefficient of each cell;

[0076] A prediction module is used to predict the actual working capacity attenuation of the battery cell according to the capacity attenuation curve of the battery cell and the heat dissipation coefficient of the battery cell to obtain an actual attenuation curve;

[0077] Deviation analysis module, used to perform spatiotemporal analysis on the fitting deviation between the experimental attenuation curve and the actual attenuation curve, and determine the deviation factor of the unmanned ship;

[0078] The attenuation balance module is used to perform a second prediction of the battery cell in combination with the deviation factor, and adjust the output of the battery cell based on the second prediction result to balance the attenuation capacity of each battery cell.

[0079] In one embodiment of the present invention, the remaining capacity of the battery cell is measured by fully charging and discharging the battery cell in a laboratory, including:

[0080] During a preset period of time, the battery cells are subjected to simulated output current and voltage experiments, and the output current and voltage are adjusted every 3 hours;

[0081] At the third hour, record the temperature and power of the battery cell and cool the battery cell;

[0082] Fully charge and discharge the battery cell in a cooling environment, accurately measure the remaining capacity of the battery cell, and use this remaining capacity as the remaining capacity in the third hour;

[0083] After the simulated charge and discharge experiment is completed, the battery cell is replenished in a cooling environment to the recorded power at the third hour. After the replenishment is completed, the temperature of the battery cell is restored to the recorded temperature;

[0084] This step is repeated to obtain the accurate remaining capacity of the battery cell at each moment.

[0085] It should be noted that the remaining capacity measured by the simulated charge and discharge experiment is a relatively accurate method in the prior art, which will not be described in detail here.

[0086] In one embodiment of the present invention, a heat dissipation model of multiple cells in a target battery pack is constructed, and the calculation formula of the heat dissipation model is as follows:

[0087] ;in, Indicates The heat dissipation coefficient of each battery cell, Indicates The number of sub-layers of the thermal conductive plate that each cell contacts, Indicates The index of the number of sub-layers of the thermal conductive plate that each cell contacts, Indicates Thermal conductivity of the heat conducting plate, Indicates The first heat conducting plate and the The contact area of ​​each cell contact, Indicates The thickness of the heat conducting board layer, Indicates The distance between each cell and the geometric center of the target battery pack, Indicates The number of adjacent cells. Indicates The battery cell and The Euclidean distance of the cells, Indicates the wind speed of the heat sink of the target heat sink, represents the weight of the wind speed of the heat sink of the target heat sink, represents the temperature gradient on the heat conducting plate, represents the weight of the temperature gradient on the heat conducting plate, Indicates The Euclidean distance between each battery cell and the heat sink is Indicates The weight parameter of the Euclidean distance between each battery cell and the heat sink.

[0088] It should be noted that for the battery packs installed on unmanned boats, the battery packs are generally exactly the same, and the battery packs are all cooled by a heat dissipation device (generally an air-cooled heat dissipation device). Combined with the internal structure of the battery pack, the heat dissipation coefficient of each battery cell is calculated.

[0089] In one embodiment of the present invention, the capacity attenuation of the actual operation of the battery cell is predicted according to the capacity attenuation curve of the battery cell and the heat dissipation coefficient of the battery cell, including:

[0090] The capacity decay curve of the battery cell is segmented by a sliding window according to the time interval, and the current and voltage data in the sliding window are used to construct a feature vector as a training sample, and the capacity at the next moment of the sliding window is used as the sample label of the input;

[0091] The preset model is trained through training samples and sample labels to obtain an initial prediction model;

[0092] The initial prediction model is used to independently predict each battery cell in the target battery pack to obtain a prediction result;

[0093] A composite loss function is constructed based on the heat dissipation coefficient of the battery cell. The formula of the composite loss function is as follows:

[0094] ;

[0095] in, represents the composite loss function of the initial prediction model; Indicates The battery cell and The similarity of the cells, and Respectively represent The battery cell and The heat dissipation coefficient of each battery cell; Represents the initial forecast model prediction The moment The capacity of each battery cell; Represents the initial forecast model prediction The moment The capacity of each battery cell; express The moment Sample label of each battery cell, and denote the first loss coefficient and the second loss coefficient respectively;

[0096] The initial prediction model is trained through a composite loss function to obtain an updated prediction model.

[0097] In one embodiment of the present invention, the prediction model includes several hidden layers, and the hidden layers are constructed based on LSTM long short time neural network.

[0098] Assume that there are 10 battery cells, and each battery cell is numbered by ASS1-ASS10; there are 10 time series data, and the sliding window is 3 units in length. The sliding window divides the 1st to 3rd data into a group, and the current, voltage, battery cell number and battery cell heat dissipation coefficient in the 1st to 3rd data are used as inputs of the prediction model, and the battery cell capacity in the 4th data is used as the sample label of the 1st to 3rd data; the current, voltage, battery cell number and battery cell heat dissipation coefficient in the 2nd to 4th data are used as inputs of the prediction model, and the battery cell capacity in the 5th data is used as the sample label of the 2nd to 4th data; and this cycle is repeated until the sliding window has completed the division of the 7th to 9th data, and the sliding window stops dividing the time series data.

[0099] It should be noted that, in general, it is difficult to measure the temperature of a single cell in the target battery pack because the temperature between cells is coupled. Therefore, it is unrealistic and unreliable to use the measured cell temperature as the input of the prediction model. The cells are classified by their heat dissipation coefficient. For cells with the same or similar heat dissipation coefficients, their cell attenuation should be very close. Therefore, the prediction model is updated in reverse by using the heat dissipation coefficient of the cell as the loss of the prediction model, thereby making the prediction output of the prediction model more accurate.

[0100] In one embodiment of the present invention, a spatiotemporal analysis is performed on the fitting deviation between the experimental attenuation curve and the actual attenuation curve to determine the deviation factor of the unmanned ship, including:

[0101] Collecting total current and total voltage data of the target unmanned ship during a first preset time period;

[0102] The total current and total voltage data of the target unmanned ship in the first preset time period are averaged according to the number of battery cells, and the average value is input into the updated prediction model to obtain the predicted capacity of each battery cell of the target unmanned ship in the second preset time period;

[0103] Constructing a battery cell experimental attenuation curve of the target unmanned ship according to the predicted capacity of each battery cell of the target unmanned ship in the second preset time period;

[0104] Select some cells of the target unmanned ship for measurement and calculation to obtain the actual attenuation curve of these cells;

[0105] Compare the actual decay curve of the battery cells with the experimental decay curve to determine the fitting deviation between the experimental decay curve and the actual decay curve;

[0106] Performing a spatiotemporal analysis on the target unmanned ship within a second preset time period, determining a mechanical heat source and an operating environment of the target unmanned ship, and collecting a temperature of the mechanical heat source and an operating environment of the target unmanned ship within the second preset time period as a deviation factor;

[0107] Based on the deviation factor, a correction layer is added to the updated prediction model, and the correction layer is used to correct the output of the updated prediction model.

[0108] In one embodiment of the present invention, some cells of the target unmanned ship are selected for measurement and calculation to obtain actual attenuation curves of the cells, including:

[0109] Select several target unmanned ships to perform a unified operation task in the same sea area within a preset time period;

[0110] The same power scheduling is performed on the target unmanned ship;

[0111] The first target unmanned ship executes to the first moment, the second unmanned ship executes to the second moment, and so on, until the last unmanned ship executes to the last moment;

[0112] For each ship, select the battery cell at the same position, such as the 7th battery cell of the 3rd target battery pack, and conduct charge and discharge experiments on the battery cell to determine the remaining capacity;

[0113] Obtain the remaining capacity of the seventh cell of the third target battery pack at each moment within a preset time period, and construct an actual attenuation curve for the remaining capacity at each moment;

[0114] Repeat the experiment to obtain enough samples of actual attenuation curves;

[0115] The actual attenuation curves of these samples are compared with the experimental attenuation curves to obtain the fitting deviation;

[0116] The fitting deviation is caused by the combined influence of seawater temperature and mechanical heat source.

[0117] It should be noted that during the actual use of the unmanned boat, its internal mechanical work will generate additional heat sources, such as mechanical friction, resistive heating, etc. In addition, the temperature of the ocean will also interfere with the temperature of the unmanned boat. These are all external heat sources of the battery pack. The external heat source of the battery pack is the main source of deviation in the output of the prediction model. Therefore, the prediction model needs to be corrected through deviation to make the prediction of each battery cell more in line with reality.

[0118] In one embodiment of the present invention, the formula of the correction layer is as follows:

[0119] ;

[0120] in, Indicates The cell capacity after calibration at a certain moment; Indicates Update the predicted cell capacity output by the prediction model at each moment; Indicates the effect of the temperature of the mechanical heat source on the cell attenuation, represents the weight coefficient of the first external heat source, Indicates that the unmanned ship is The temperature of the mechanical heat source at a certain moment, represents the maximum value of the mechanical heat source temperature within the second preset time period, Indicates that the base is a natural constant The exponential function of Indicates the effect of water temperature on cell attenuation; The standard temperature of seawater. Indicates The temperature of the sea water at that moment, represents the exponential decay coefficient, , represents the natural logarithm; represents the combined effect of mechanical heat source temperature and seawater temperature on cell degradation, and They represent the mechanical heat source index control coefficient and the seawater temperature index control coefficient respectively. , , Indicates the maximum temperature difference between the mechanical heat source and the seawater temperature.

[0121] It should be noted that by adding a correction layer to the prediction model, the prediction of the output of the prediction model is corrected by the deviation of the external heat source, so that the output is more in line with the attenuation of the battery cells at various positions of the actual unmanned ship.

[0122] In one embodiment of the present invention, the output of the battery cell is adjusted based on the secondary prediction result to balance the attenuation capacity of each battery cell, including:

[0123] Randomly generate an initial population with R individuals that meet the constraints, and initialize the current iteration number Time to 1;

[0124] Constraints:

[0125] Determine and adjust the output current or output voltage of the battery cell according to the series and parallel adjustment of the unmanned boat;

[0126] The total output voltage of the battery cell is equal to the total required voltage of the unmanned ship, and the total output current of the battery cell is equal to the total required current of the unmanned ship;

[0127] The output current or output voltage of the battery cell needs to be adjusted within the preset adjustment range;

[0128] Objective function:

[0129] ;in, It indicates that the correction layer based on the updated prediction model outputs the individual codes to obtain the maximum value of the predicted capacity of the battery cell. It indicates that the correction layer based on the updated prediction model outputs the individual codes to obtain the minimum value of the predicted capacity of the battery cell. It means that the square of the difference between the maximum value and the minimum value in the predicted capacity of the battery cell is minimized by taking the correction layer based on the updated prediction model to output the individual code;

[0130] Output the objective function value for R individuals, retain the individuals whose value is greater than the preset objective function threshold, and update the remaining individuals as parents;

[0131] If the current number of iterations Time ≥ the maximum number of iterations, stop updating the individual with the highest output objective function value as the output solution for the battery cell; otherwise, return to update the parent generation.

[0132] It should be noted that by making personalized modifications to the output of each battery cell, the attenuation of each battery cell tends to be consistent, thereby ensuring the service life of the battery pack and reducing the maintenance cost of the battery pack.

[0133] In one embodiment of the present invention, the remaining individuals are updated as parents, including:

[0134] pass Calculate the current individual temperature; where, Indicates the current individual temperature, represents the maximum number of iterations, Indicates the current individual's basal temperature;

[0135] ;in, It represents the average difference between the objective function values ​​of R-1 individuals and the individual with the highest current objective function value. Represents the difference between the current individual and the individual with the highest current objective function value;

[0136] pass Calculate the distance between the current individual and the individual with the highest current objective function value; where, Represents the distance between the current individual and the individual with the highest current objective function value, represents the scaling function, represents the individual with the highest current objective function value, Represents a constant parameter, which is used to adjust the position offset between the current individual and the individual with the highest current objective function value. Indicates the current individual's position, Indicates the absolute value operation;

[0137] ;in, and They represent control parameters, which are used to control the scaling of the scaling function. Represents the natural base;

[0138] pass Determine the position adjustment parameters of the current individual; where, Represents the position adjustment parameter, which is used to control the movement amplitude of the individual. Represents a constant parameter used to control the position adjustment amplitude. Represents the position accuracy adjustment amount, which determines the accuracy of the movement based on the gap between the current individual and the individual with the highest current objective function value. , Represents a random number, used to introduce randomness;

[0139] The position of the current individual is updated according to the current individual temperature, the distance between the current individual and the individual with the highest current objective function value, and the position adjustment parameter. The calculation formula is as follows:

[0140] ;in, Represents the updated individual position.

[0141] A pure electric unmanned ship BMS battery management method, realizing a pure electric unmanned ship BMS battery management system, comprising:

[0142] Conduct simulation experiments on the battery cells and construct the experimental attenuation curve of the battery cells, including:

[0143] The total current and total voltage of the unmanned ship are collected at fixed time intervals to determine the operating power of the unmanned ship at each moment;

[0144] Determine the number of target battery packs and cells to be installed on the unmanned ship, and calculate the output current and output voltage of the cells at each moment based on the number of cells installed;

[0145] Construct an experimental decay curve based on experimental measurements, where each moment of the experimental decay curve includes cell capacity, voltage, and current;

[0146] Construct a heat dissipation model for multiple cells in the target battery pack and obtain the heat dissipation coefficient of each cell;

[0147] The actual capacity attenuation of the battery cell is predicted based on the capacity attenuation curve of the battery cell and the heat dissipation coefficient of the battery cell to obtain the actual attenuation curve;

[0148] The fitting deviation between the experimental attenuation curve and the actual attenuation curve is analyzed in time and space to determine the deviation factor of the unmanned ship;

[0149] The deviation factor is combined to make a second prediction of the battery cell, and the output of the battery cell is adjusted based on the second prediction result to balance the attenuated capacity of each battery cell.

[0150] A computing device comprising:

[0151] one or more processors;

[0152] A storage device is used to store one or more programs. When the one or more programs are executed by one or more processors, the one or more processors implement the BMS battery management system of a pure electric unmanned ship.

[0153] A computer-readable storage medium stores a program, which, when executed by a processor, implements a pure electric unmanned ship BMS battery management system.

[0154] The above describes an embodiment of the present embodiment, but the present embodiment is not limited to the above-mentioned specific implementation methods. The above-mentioned specific implementation methods are merely illustrative and not restrictive. Under the guidance of the present embodiment, ordinary technicians in this field can also make many forms, all of which are within the protection of the present embodiment.

Claims

1. A pure electric unmanned ship BMS battery management system, characterized in that: include: The data acquisition module is used to simulate the battery cell and construct the experimental attenuation curve of the battery cell, including: The total current and total voltage of the unmanned ship are collected at fixed time intervals to determine the operating power of the unmanned ship at each moment; Determine the number of target battery packs and cells to be installed on the unmanned ship, and calculate the output current and output voltage of the cells at each moment based on the number of cells installed; Construct an experimental decay curve based on experimental measurements, where each moment of the experimental decay curve includes cell capacity, voltage, and current; The heat dissipation module is used to construct a heat dissipation model of multiple cells in the target battery pack to obtain the heat dissipation coefficient of each cell. The calculation formula of the heat dissipation model is as follows: ;in, Indicates The heat dissipation coefficient of each battery cell, Indicates The number of sub-layers of the thermal conductive plate that each cell contacts, Indicates The index of the number of sub-layers of the thermal conductive plate that each cell contacts, Indicates Thermal conductivity of the heat conducting plate, Indicates The first heat conducting plate and the The contact area of ​​each cell contact, Indicates The thickness of the heat conducting board layer, Indicates The distance between each cell and the geometric center of the target battery pack, Indicates The number of adjacent cells. Indicates The battery cell and The Euclidean distance of the cells, Indicates the wind speed of the heat sink of the target heat sink, represents the weight of the wind speed of the heat sink of the target heat sink, represents the temperature gradient on the heat conducting plate, represents the weight of the temperature gradient on the heat conducting plate, Indicates The Euclidean distance between each battery cell and the heat sink is Indicates The weight parameter of the Euclidean distance between each battery cell and the heat sink; The prediction module is used to predict the actual working capacity attenuation of the battery cell according to the capacity attenuation curve of the battery cell and the heat dissipation coefficient of the battery cell to obtain the actual attenuation curve, as follows: The capacity decay curve of the battery cell is segmented by a sliding window according to the time interval, and the current and voltage data in the sliding window are used to construct a feature vector as a training sample, and the capacity at the next moment of the sliding window is used as the input sample label; The preset model is trained through training samples and sample labels to obtain an initial prediction model; The initial prediction model is used to independently predict each battery cell in the target battery pack to obtain a prediction result; A composite loss function is constructed based on the heat dissipation coefficient of the battery cell. The formula of the composite loss function is as follows: ； in, represents the composite loss function of the initial prediction model; Indicates The battery cell and The similarity of the cells, and Respectively represent The battery cell and Heat dissipation coefficient of each battery cell; Represents the initial forecast model prediction The moment The capacity of each battery cell; Represents the initial forecast model prediction The moment The capacity of each battery cell; express The moment Sample label of each battery cell, and denote the first loss coefficient and the second loss coefficient respectively; The initial prediction model is trained by a composite loss function to obtain an updated prediction model; Deviation analysis module, used to perform spatiotemporal analysis on the fitting deviation between the experimental attenuation curve and the actual attenuation curve, and determine the deviation factor of the unmanned ship; The attenuation balance module is used to perform a second prediction of the battery cell in combination with the deviation factor, and adjust the output of the battery cell based on the second prediction result to balance the attenuation capacity of each battery cell.

2. A pure electric unmanned ship BMS battery management system according to claim 1, characterized in that: The fitting deviation between the experimental attenuation curve and the actual attenuation curve is analyzed in time and space to determine the deviation factor of the unmanned ship, including: Collecting total current and total voltage data of the target unmanned ship during a first preset time period; The total current and total voltage data of the target unmanned ship in the first preset time period are averaged according to the number of battery cells, and the average value is input into the updated prediction model to obtain the predicted capacity of each battery cell of the target unmanned ship in the second preset time period; Constructing a battery cell experimental attenuation curve of the target unmanned ship according to the predicted capacity of each battery cell of the target unmanned ship in the second preset time period; Select some cells of the target unmanned ship for measurement and calculation to obtain the actual attenuation curve of these cells; Compare the actual decay curve of the battery cells with the experimental decay curve to determine the fitting deviation between the experimental decay curve and the actual decay curve; Performing a spatiotemporal analysis on the target unmanned ship within a second preset time period, determining a mechanical heat source and an operating environment of the target unmanned ship, and collecting a temperature of the mechanical heat source and an operating environment of the target unmanned ship within the second preset time period as a deviation factor; Based on the deviation factor, a correction layer is added to the updated prediction model, and the correction layer is used to correct the output of the updated prediction model.

3. A pure electric unmanned ship BMS battery management system according to claim 2, characterized in that: The formula for the correction layer is as follows: ； in, Indicates The cell capacity after calibration at a certain moment; Indicates Update the predicted cell capacity output by the prediction model at each moment; Indicates the effect of the temperature of the mechanical heat source on the cell attenuation, represents the weight coefficient of the first external heat source, Indicates that the unmanned ship is The temperature of the mechanical heat source at a certain moment, represents the maximum value of the mechanical heat source temperature within the second preset time period, Indicates that the base is a natural constant The exponential function of Indicates the effect of water temperature on cell attenuation; The standard temperature of seawater. Indicates The temperature of the sea water at that moment, represents the exponential decay coefficient, , represents the natural logarithm; represents the combined effect of mechanical heat source temperature and seawater temperature on cell degradation, and They represent the mechanical heat source index control coefficient and the seawater temperature index control coefficient respectively. , , Indicates the maximum temperature difference between the mechanical heat source and the seawater temperature.

4. A pure electric unmanned ship BMS battery management system according to claim 3, characterized in that: Adjust the output of the battery cell based on the second prediction results to balance the attenuation capacity of each battery cell, including: Randomly generate an initial population with R individuals that meet the constraints, and initialize the current iteration number Time to 1; Constraints: According to the series and parallel connection of the unmanned boat, determine and adjust the output current or output voltage of the battery cell; The total output voltage of the battery cell is equal to the total required voltage of the unmanned ship, and the total output current of the battery cell is equal to the total required current of the unmanned ship; The output current or output voltage of the battery cell needs to be adjusted within the preset adjustment range; Objective function: ;in, It indicates that the maximum value of the predicted capacity of the battery cell is obtained by outputting the individual codes based on the correction layer of the updated prediction model. It indicates that the correction layer based on the updated prediction model outputs the individual codes to obtain the minimum value of the predicted capacity of the battery cell. It means that the square of the difference between the maximum value and the minimum value in the predicted capacity of the battery cell is minimized by taking the correction layer based on the updated prediction model to output the individual code; Output the objective function value for R individuals, retain the individuals whose value is greater than the preset objective function threshold, and update the remaining individuals as parents; If the current number of iterations Time ≥ the maximum number of iterations, stop updating the individual with the highest output objective function value as the output solution for the battery cell; otherwise, return to update the parent generation.

5. A pure electric unmanned ship BMS battery management system according to claim 4, characterized in that: Update the remaining individuals as parents, including: pass Calculate the current individual temperature; where, Indicates the current individual temperature, represents the maximum number of iterations, Indicates the current individual's basal temperature; ;in, It represents the average difference between the objective function values ​​of R-1 individuals and the individual with the highest current objective function value. Represents the difference between the current individual and the individual with the highest current objective function value; pass Calculate the distance between the current individual and the individual with the highest current objective function value; where, Represents the distance between the current individual and the individual with the highest current objective function value, represents the scaling function, represents the individual with the highest current objective function value, Represents a constant parameter, which is used to adjust the position offset between the current individual and the individual with the highest current objective function value. Indicates the current individual's position, Indicates the absolute value operation; ;in, and They represent control parameters, which are used to control the scaling of the scaling function. Represents the natural base; pass Determine the position adjustment parameters of the current individual; where, Represents the position adjustment parameter, which is used to control the movement amplitude of the individual. Represents a constant parameter used to control the position adjustment amplitude. Represents the position accuracy adjustment amount, which determines the accuracy of the movement based on the gap between the current individual and the individual with the highest current objective function value. , Represents a random number, used to introduce randomness; The position of the current individual is updated according to the current individual temperature, the distance between the current individual and the individual with the highest current objective function value, and the position adjustment parameter. The calculation formula is as follows: ;in, Represents the updated individual position.

6. A pure electric unmanned ship BMS battery management method, applied to a pure electric unmanned ship BMS battery management system as claimed in any one of claims 1 to 5, characterized in that: include: Conduct simulation experiments on the battery cells and construct the experimental attenuation curve of the battery cells, including: The total current and total voltage of the unmanned ship are collected at fixed time intervals to determine the operating power of the unmanned ship at each moment; Determine the number of target battery packs and cells to be installed on the unmanned ship, and calculate the output current and output voltage of the cells at each moment based on the number of cells installed; Construct an experimental decay curve based on experimental measurements, where each moment of the experimental decay curve includes cell capacity, voltage, and current; Construct a heat dissipation model for multiple cells in the target battery pack and obtain the heat dissipation coefficient of each cell; The actual capacity attenuation of the battery cell is predicted based on the capacity attenuation curve of the battery cell and the heat dissipation coefficient of the battery cell to obtain the actual attenuation curve; The fitting deviation between the experimental attenuation curve and the actual attenuation curve is analyzed in time and space to determine the deviation factor of the unmanned ship; The deviation factor is combined to make a second prediction of the battery cell, and the output of the battery cell is adjusted based on the second prediction result to balance the attenuated capacity of each battery cell.

7. A computing device, characterized in that include: one or more processors; A storage device for storing one or more programs. When the one or more programs are executed by one or more processors, the one or more processors implement a pure electric unmanned ship BMS battery management system as described in any one of claims 1 to 5.

8. A computer-readable storage medium, characterized in that: A program is stored in a computer-readable storage medium, and when the program is executed by a processor, a pure electric unmanned ship BMS battery management system as described in any one of claims 1 to 5 is implemented.

Images