Battery output power prediction method and device, equipment and storage medium

Abstract

The invention discloses a battery output power prediction method and device, equipment and a storage medium, and relates to the technical field of batteries, and the method comprises the steps: obtaining real-time battery data under a preset working condition, and the real-time battery data comprises voltage data, current data, temperature data and humidity data; the real-time battery data is input into a preset neural network model, initial battery output power is obtained, and the preset neural network model comprises an input layer, a bidirectional LSTM layer, an attention layer, a residual error connection layer and a full connection layer; and based on Kalman filtering, correcting the initial battery output power to obtain a target battery output power for the preset working condition. According to the invention, high-precision battery output power prediction can be realized.

Description

Technical Field

[0001] The present application relates to the field of battery technology, and in particular to a battery output power prediction method, device, equipment and storage medium. Background Art

[0002] With the widespread popularity of electric vehicles, renewable energy storage systems and portable electronic devices, accurate prediction of battery output power is particularly important as the core energy supply unit of these devices. Accurate battery output power prediction can not only effectively improve the overall efficiency of the energy management system, but also significantly extend the service life of the battery while ensuring the safety and reliability of the system operation. Especially in the field of electric vehicles, accurate power prediction technology can optimize the utilization and distribution of energy, thereby maximizing the driving range, and effectively prevent safety hazards that may be caused by over-discharging or overcharging of the battery. However, existing battery output power prediction methods all have the technical problem of low prediction accuracy.

[0003] The above contents are only used to assist in understanding the technical solution of the present application and do not constitute an admission that the above contents are prior art. Summary of the invention

[0004] The main purpose of the present application is to provide a battery output power prediction method, device, equipment and storage medium, aiming to solve the technical problem of low prediction accuracy in existing battery output power prediction methods.

[0005] To achieve the above objectives, the present application proposes a battery output power prediction method, the battery output power prediction method comprising:

[0006] Acquire real-time battery data under preset working conditions, wherein the real-time battery data includes voltage data, current data, temperature data and humidity data;

[0007] Inputting the real-time battery data into a preset neural network model to obtain an initial battery output power, wherein the preset neural network model includes an input layer, a bidirectional LSTM layer, an attention layer, a residual connection layer, and a fully connected layer;

[0008] Based on Kalman filtering, the initial battery output power is corrected to obtain the target battery output power for the preset working condition.

[0009] In one embodiment, before the step of inputting the real-time battery data into a preset neural network model to obtain the initial battery output power, the step further includes:

[0010] Obtain historical battery data under different operating conditions;

[0011] Preprocessing the historical battery data to obtain target historical battery data;

[0012] Based on a preset division ratio, the target historical battery data is divided into a training set, a validation set, and a test set;

[0013] Based on the training set, the validation set and the test set, an initial neural network model is trained to obtain a preset neural network model.

[0014] In one embodiment, the step of preprocessing the historical battery data to obtain target historical battery data includes:

[0015] Removing abnormal values ​​and noise from the historical battery data to obtain cleaned historical battery data;

[0016] The cleaning historical battery data is normalized to obtain target historical battery data.

[0017] In one embodiment, the step of correcting the initial battery output power based on Kalman filtering to obtain the target battery output power for the preset working condition includes:

[0018] The initial state estimate is set to the median value of the battery initial state;

[0019] Determine an error covariance matrix based on variance estimates of historical battery data;

[0020] Based on the initial value state estimation value, the error covariance matrix, the observation noise and the process noise matrix, the initial battery output power is corrected to obtain the target battery output power for the preset working condition.

[0021] In one embodiment, the method further comprises:

[0022] Determine the battery priority according to the target battery output power and the current battery status, wherein the current battery status includes the remaining battery power and the battery health;

[0023] The distribution of the loads among the batteries is adjusted according to the battery priorities.

[0024] In one embodiment, the method further comprises:

[0025] Determine the current state of the battery according to the target battery output power;

[0026] Determining whether the current charge and discharge state of the battery is a critical full state or a critical empty state;

[0027] When the current state of the battery is a critical full state or a critical empty state, an early warning notification is initiated.

[0028] In one embodiment, the method further comprises:

[0029] When the time interval between the current time and the previous update time is greater than or equal to the model update period, new historical battery data under different working conditions are acquired, and a preset neural network model is updated based on the new historical battery data; and / or,

[0030] Collect user feedback information on the preset neural network model, and optimize the preset neural network model according to the feedback information to improve the adaptability of the preset neural network model under different working conditions.

[0031] In addition, to achieve the above purpose, the present application also proposes a battery output power prediction device, the battery output power prediction device comprising:

[0032] An acquisition module, used to acquire real-time battery data under preset working conditions, wherein the real-time battery data includes voltage data, current data, temperature data and humidity data;

[0033] An input module, used to input the real-time battery data into a preset neural network model to obtain an initial battery output power, wherein the preset neural network model includes an input layer, a bidirectional LSTM layer, an attention layer, a residual connection layer, and a fully connected layer;

[0034] The correction module is used to correct the initial battery output power based on Kalman filtering to obtain the target battery output power for the preset working condition.

[0035] In addition, to achieve the above-mentioned purpose, the present application also proposes a battery output power prediction device, which includes: a memory, a processor, and a computer program stored in the memory and executable on the processor, and the computer program is configured to implement the steps of the battery output power prediction method as described above.

[0036] In addition, to achieve the above-mentioned purpose, the present application also proposes a storage medium, which is a computer-readable storage medium, and a computer program is stored on the storage medium. When the computer program is executed by a processor, the steps of the battery output power prediction method described above are implemented.

[0037] In addition, to achieve the above-mentioned purpose, the present application also provides a computer program product, which includes a computer program, and when the computer program is executed by a processor, the steps of the battery output power prediction method as described above are implemented.

[0038] One or more technical solutions proposed in this application have at least the following technical effects:

[0039] The battery output power prediction method, device, equipment and storage medium proposed in the present application obtain real-time battery data under preset working conditions, wherein the real-time battery data includes voltage data, current data, temperature data and humidity data; input the real-time battery data into a preset neural network model to obtain the initial battery output power, wherein the preset neural network model includes an input layer, a bidirectional LSTM layer, an attention layer, a residual connection layer and a full connection layer; based on Kalman filtering, the initial battery output power is corrected to obtain the target battery output power for the preset working conditions. The technical problem of low prediction accuracy in the existing battery output power prediction methods is solved. Compared with the prior art, the present application collects real-time battery data and performs preliminary prediction through a preset neural network model, and then corrects the preliminary predicted battery output power through Kalman filtering, and finally obtains a more accurate battery output power, realizing high-precision, strong generalization ability and easy-to-maintain battery output power prediction, avoiding potential safety problems such as battery over-discharge and overcharging, and further improving the performance and reliability of the battery management system. BRIEF DESCRIPTION OF THE DRAWINGS

[0040] The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the present application and, together with the description, serve to explain the principles of the present application.

[0041] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, for ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative labor.

[0042] Figure 1 A schematic diagram of a flow chart provided for the first embodiment of the method for predicting battery output power of the present application;

[0043] Figure 2 A schematic diagram of a flow chart provided for the second embodiment of the method for predicting battery output power of the present application;

[0044] Figure 3 This is a schematic diagram of the module structure of the battery output power prediction device according to an embodiment of the present application;

[0045] Figure 4 Schematic diagram of the device structure of the hardware operating environment involved in the battery output power prediction method in the embodiment of the present application.

[0046] The purpose, features and advantages of this application will be further described in conjunction with the embodiments and with reference to the accompanying drawings. DETAILED DESCRIPTION

[0047] It should be understood that the specific embodiments described herein are only used to explain the technical solutions of the present application and are not used to limit the present application.

[0048] In order to better understand the technical solution of the present application, a detailed description will be given below in conjunction with the accompanying drawings and specific implementation methods.

[0049] The main solution of the embodiment of the present application is: obtaining real-time battery data under preset working conditions, wherein the real-time battery data includes voltage data, current data, temperature data and humidity data; inputting the real-time battery data into a preset neural network model to obtain an initial battery output power, wherein the preset neural network model includes an input layer, a bidirectional LSTM layer, an attention layer, a residual connection layer and a fully connected layer; based on Kalman filtering, correcting the initial battery output power to obtain a target battery output power for the preset working conditions.

[0050] It can be seen from the above embodiments that the present application obtains real-time battery data under preset working conditions, wherein the real-time battery data includes voltage data, current data, temperature data and humidity data; inputs the real-time battery data into a preset neural network model to obtain the initial battery output power, wherein the preset neural network model includes an input layer, a bidirectional LSTM layer, an attention layer, a residual connection layer and a fully connected layer; based on Kalman filtering, the initial battery output power is corrected to obtain the target battery output power for the preset working condition. The technical problem of low prediction accuracy in the existing battery output power prediction methods is solved. Compared with the prior art, the present application collects real-time battery data and performs preliminary prediction through a preset neural network model, and then corrects the preliminary predicted battery output power through Kalman filtering, and finally obtains a more accurate battery output power, realizing high-precision, strong generalization ability and easy-to-maintain battery output power prediction, avoiding potential safety problems such as battery over-discharge and overcharging, and further improving the performance and reliability of the battery management system.

[0051] It should be noted that the execution subject of this embodiment can be a computing service device with data processing, network communication and program running functions, such as a tablet computer, a personal computer, a mobile phone, etc., or an electronic device capable of realizing the above functions, a battery output power prediction device, etc. The following takes the battery output power prediction device as an example to illustrate this embodiment and the following embodiments.

[0052] Based on this, the present application embodiment provides a battery output power prediction method, referring to Figure 1 , Figure 1 This is a flow chart of the first embodiment of the battery output power prediction method of the present application.

[0053] In this embodiment, the battery output power prediction method includes steps S10 to S40:

[0054] Step S10, acquiring real-time battery data under a preset working condition, wherein the real-time battery data includes voltage data, current data, temperature data and humidity data;

[0055] It should be noted that real-time battery data can be obtained through sensors, and then the collected real-time battery data is cleaned to remove outliers and noise, and the cleaned real-time battery data is normalized to standardize the real-time battery data to a relatively stable range, so as to adapt to the input requirements of the preset neural network model.

[0056] Step S20, inputting the real-time battery data into a preset neural network model to obtain an initial battery output power, wherein the preset neural network model includes an input layer, a bidirectional LSTM layer, an attention layer, a residual connection layer, and a fully connected layer;

[0057] It should be noted that the preset neural network model is an improved LSTM network; the preset neural network model can capture the time series characteristics of real-time battery data. By analyzing the voltage and current data, the preset neural network model can identify the battery health status and battery performance changes, and can consider the impact of environmental factors such as temperature and humidity on battery performance, and extract these environmental characteristics. The time series characteristics and battery health status characteristics can help predict the future output power change trend of the battery; and considering the environmental impact characteristics can enable the model to maintain high-precision output power prediction under different environmental conditions.

[0058] It should be noted that the preset neural network model has long and short-term memory capabilities to avoid the gradient vanishing problem. It can not only process time series data and effectively capture the dependencies between time steps, but also effectively integrate information from different time steps to provide richer feature representations; it can also process a variety of input data, such as voltage, current, temperature, and humidity, and combine these features for comprehensive analysis to improve the accuracy of predictions.

[0059] It should be noted that the input shape of the input layer is the same as that of the basic LSTM model. The input shape will be a three-dimensional array with a shape of (batch_size, time_steps, num_features), where num_features is 4 (voltage, current, temperature, humidity); the bidirectional LSTM layer can consider both the forward and backward information of the time series, which means that it not only predicts the future based on past information, but also considers future information, thereby improving the model's ability to predict battery output power. By combining the outputs of the forward and backward LSTMs, the preset neural network model can capture richer contextual information; the attention layer allows the model to dynamically focus on more important parts when processing the input sequence through the attention mechanism. For battery output power prediction, This means that the model can learn to find those time points or features in the time series that have the greatest impact on the prediction results. This can be achieved by calculating the attention weights for each time step, allowing the model to more accurately predict the battery output power. In deep networks, residual connections can help alleviate the gradient vanishing problem and promote the effective transmission of information. By adding residual connections between LSTM layers, we can build deeper networks without worrying about performance degradation. For the battery output power prediction model, this means that even if the model becomes very deep, the residual connection can be used to maintain performance stability and avoid performance degradation due to increased network depth. The fully connected layer is the output layer. A fully connected layer is used to convert the output processed by the bidirectional LSTM and attention mechanism into a predicted value of the battery output power.

[0060] In a feasible implementation manner, before the step of inputting the real-time battery data into a preset neural network model to obtain the initial battery output power, the step also includes: acquiring historical battery data under different operating conditions; preprocessing the historical battery data to obtain target historical battery data; based on a preset division ratio, dividing the target historical battery data into a training set, a verification set, and a test set; based on the training set, the verification set, and the test set, training the initial neural network model to obtain a preset neural network model.

[0061] It should be noted that different operating conditions refer to the environmental conditions and usage status of the battery when it is working. Different operating conditions may be different temperatures, humidity, load conditions, etc. Historical battery data may include multiple parameters such as battery voltage, battery current, ambient temperature, ambient humidity, charge and discharge data, and remaining capacity.

[0062] It should be noted that data preprocessing includes cleaning of raw data, missing value processing, outlier detection and processing, data normalization or standardization, etc. The purpose of preprocessing is to make the data neater and more standardized, eliminate noise and redundancy, so as to better use it for subsequent model training.

[0063] It should be noted that during the model training process, the model will continuously adjust its parameters to minimize the loss function on the training set, and use the validation set to monitor the performance of the model so that early stopping or parameter adjustment can be performed when necessary to prevent overfitting. When the performance of the model on the training set and validation set is stable, the test set can be used to evaluate the final performance of the model. Preferably, the division ratio of the training set, validation set, and test set is 70%, 20%, and 10%.

[0064] In a feasible implementation, the step of preprocessing the historical battery data to obtain target historical battery data includes: removing abnormal values ​​and noise from the historical battery data to obtain cleaned historical battery data; and normalizing the cleaned historical battery data to obtain target historical battery data.

[0065] It should be noted that real-time battery data can be obtained through the battery management system, and then the collected historical battery data can be cleaned to remove outliers and noise, and the cleaned historical battery data can be normalized to standardize the historical battery data to a relatively stable range, so as to meet the input requirements of the preset neural network model.

[0066] Step S30: Based on Kalman filtering, the initial battery output power is corrected to obtain a target battery output power for the preset working condition.

[0067] It should be noted that the preset working condition refers to the current working condition of the battery; the target battery output power refers to the output power of the battery in the future obtained by prediction. The output power change trend of the battery in the future can be determined by obtaining the target battery output power of the battery in the future; by predicting the output power of the battery, the required charging time can be estimated more accurately, which helps to avoid unnecessary long-term charging, thereby saving energy and time. During the discharge process, the discharge rate and load can be dynamically adjusted according to the predicted battery output power to ensure that the battery operates in an efficient and safe state. This helps to improve energy utilization efficiency and reduce potential risks caused by battery overheating or overloading; in a multi-battery system, intelligent scheduling and load balancing can be achieved by predicting the output power of each battery. This helps to ensure that each battery operates in its optimal working state, thereby improving the performance and reliability of the entire system.

[0068] In the specific implementation, the Kalman filter parameters are initialized, the initial state estimate is set to the median of the battery's initial state, the error covariance matrix is ​​estimated based on the variance of historical data, and the preliminary prediction results of the neural network (i.e., the initial battery output power) are used as the input of the Kalman filter. According to the difference between the real-time observation value and the predicted value, the Kalman filter is used to update the state and correct the error.

[0069] In a feasible implementation, the method further includes: determining a battery priority based on the target battery output power and a current battery status, wherein the current battery status includes a remaining battery power and a battery health; and adjusting the distribution of loads among batteries based on the battery priority.

[0070] In the specific implementation, more battery power remaining and higher battery health indicate better battery current status. The higher the target battery output power and the better the battery current status, the higher the battery priority. The higher the priority battery can increase the load it bears, and the lower the priority battery can reduce its load. Therefore, the load between batteries can be dynamically adjusted by predicting the target battery output power of each battery in the multi-battery system, which can ensure that each battery operates in its optimal working state to avoid overload or idleness.

[0071] In a feasible implementation, the method further includes: determining the current state of the battery according to the target battery output power; judging whether the current charge and discharge state of the battery is a critical full state or a critical empty state; and initiating an early warning notification when the current state of the battery is a critical full state or a critical empty state.

[0072] It should be noted that accurate battery output power prediction can help determine when the battery is close to being fully charged or discharged (i.e., critically full or critically discharged), thereby avoiding overcharging and overdischarging, which are one of the main reasons for shortening battery life because they cause physical and chemical changes in the battery's internal structure and accelerate battery aging. Therefore, early warnings need to be issued in a timely manner so that timely measures can be taken to prevent failures.

[0073] In a feasible implementation, the method further includes: when the time interval between the current time and the previous update time is greater than or equal to the model update period, acquiring new historical battery data under different operating conditions, and updating the preset neural network model based on the new historical battery data; and / or, collecting user feedback information on the preset neural network model, and optimizing the preset neural network model based on the feedback information to improve the adaptability of the preset neural network model under different operating conditions.

[0074] It should be noted that new data is extracted from electric vehicles or energy storage systems at regular intervals (such as weekly or monthly), the model is updated, and the LSTM model is incrementally trained using online learning methods so that it can quickly adapt to the new data distribution and dynamically adjust the Kalman filter parameters. The prediction error of the real-time updated model on new data continues to remain at a low level; by collecting user feedback, it is found that the performance of the system under extreme temperature conditions can be further optimized, and the model is improved based on this feedback to enhance the adaptability of the system in various environments.

[0075] In this embodiment, an automated update mechanism is used to automatically update model parameters on a regular basis to ensure that the model always maintains optimal performance and that the model operates efficiently.

[0076] This embodiment obtains real-time battery data under preset working conditions, wherein the real-time battery data includes voltage data, current data, temperature data and humidity data; inputs the real-time battery data into a preset neural network model to obtain the initial battery output power, wherein the preset neural network model includes an input layer, a bidirectional LSTM layer, an attention layer, a residual connection layer and a fully connected layer; based on Kalman filtering, the initial battery output power is corrected to obtain the target battery output power for the preset working conditions. The technical problem of low prediction accuracy in the existing battery output power prediction methods is solved. Compared with the prior art, the present application collects real-time battery data and performs preliminary prediction through a preset neural network model, and then corrects the preliminary predicted battery output power through Kalman filtering, and finally obtains a more accurate battery output power, realizing high-precision, strong generalization ability and easy-to-maintain battery output power prediction, avoiding potential safety problems such as over-discharging and overcharging of the battery, and further improving the performance and reliability of the battery management system.

[0077] Based on the first embodiment of the present application, in the second embodiment of the present application, the same or similar contents as those in the above-mentioned embodiment 1 can be referred to the above introduction, and will not be repeated in the following. Figure 2 , step S30 also includes steps S301 to S303:

[0078] Step S301, setting the initial state estimation value as the battery initial state median value;

[0079] It should be noted that the initial state estimate value may be set to the median value of the battery's initial state, which represents the best guess of the battery's initial state.

[0080] Step S302, determining an error covariance matrix based on variance estimation of historical battery data;

[0081] It should be noted that the error covariance matrix is ​​estimated based on the variance of historical battery data, reflecting the uncertainty of the system state.

[0082] Step S303, based on the initial value state estimation value, the error covariance matrix, the observation noise and the process noise matrix, the initial battery output power is corrected to obtain the target battery output power for the preset working condition.

[0083] It should be noted that the observation noise and process noise matrix represent the degree of measurement error and system internal fluctuation, which affects the Kalman filter's confidence in state estimation; the Kalman filter uses the preliminary prediction results generated by the neural network as the input of the Kalman filter to provide a preliminary estimate of the battery state. Combined with real-time observations (such as voltage, current, etc.), the Kalman filter corrects the preliminary prediction by calculating the difference between the real-time observations and the predicted values, corrects the prediction results, and dynamically adjusts the state estimate; the Kalman filter predicts the state at the current moment based on the state estimate and control input at the previous moment. Its update steps are 1. According to the error covariance matrix and observation noise, the Kalman gain is calculated to balance the weight between prediction and observation. 2. Combined with the Kalman gain, the observation value is merged with the preliminary prediction value to update the state estimate. 3 Adjust the error covariance matrix to reflect the uncertainty of the updated state estimate.

[0084] In the specific implementation, determine the system model and observation model: in the correction of battery output power, first need to determine the dynamic model of the battery system, which is usually based on the electrical characteristics and thermal characteristics of the battery; the observation model describes how to obtain the output power data of the battery from actual measurements; initialize Kalman filter parameters: this includes the initial state estimate, the covariance matrix of the state estimate, the process noise covariance matrix and the observation noise covariance matrix; these parameters need to be appropriately set according to the characteristics of the battery system and the characteristics of the measurement data; apply the Kalman filter algorithm: use the initial battery output power as the initial state estimate of the Kalman filter, and predict the state of the next time step according to the system model , use the observed data to update the state estimate, taking into account the uncertainty of prediction (process noise) and the uncertainty of observation (observation noise), and weigh the weights of prediction and observation by calculating the Kalman gain to obtain the optimal state estimate; iterative update: repeat the above process for each time step until all available observation data are processed. In each iteration, the state estimate and covariance matrix are updated to reflect the new observation data and system dynamics; the state estimate corrected by the Kalman filter is the target battery output power for the preset operating conditions. This estimate combines the dynamic characteristics of the system and the actual observation data, so it is more accurate and reliable than the initial battery output power.

[0085] This embodiment introduces Kalman filtering technology and a neural network hybrid model, and the present invention enhances the generalization ability and robustness of the model, so that it can adapt to different types of batteries and changing working conditions and maintain stable prediction performance.

[0086] It should be noted that the above examples are only used to understand the present application and do not constitute a limitation on the battery output power prediction method of the present application. More simple transformations based on this technical concept are all within the scope of protection of the present application.

[0087] This application also provides a battery output power prediction device, please refer to Figure 3 , the battery output power prediction device comprises:

[0088] An acquisition module 10, used to acquire real-time battery data under a preset working condition, wherein the real-time battery data includes voltage data, current data, temperature data and humidity data;

[0089] An input module 20, used to input the real-time battery data into a preset neural network model to obtain an initial battery output power, wherein the preset neural network model includes an input layer, a bidirectional LSTM layer, an attention layer, a residual connection layer, and a fully connected layer;

[0090] The correction module 30 is used to correct the initial battery output power based on Kalman filtering to obtain the target battery output power for the preset working condition.

[0091] The battery output power prediction device provided by the present application adopts the battery output power prediction method in the above embodiment, which can solve the technical problem that the existing battery output power prediction methods have low prediction accuracy. Compared with the prior art, the beneficial effects of the battery output power prediction device provided by the present application are the same as the beneficial effects of the battery output power prediction method provided by the above embodiment, and the other technical features in the battery output power prediction device are the same as the features disclosed in the above embodiment method, which will not be repeated here.

[0092] The present application provides a battery output power prediction device, which includes: at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor so that the at least one processor can execute the battery output power prediction method in the above-mentioned embodiment 1.

[0093] Reference below Figure 4, which shows a schematic diagram of the structure of a battery output power prediction device suitable for implementing the embodiment of the present application. The battery output power prediction device in the embodiment of the present application may include but is not limited to mobile terminals such as mobile phones, laptop computers, digital broadcast receivers, PDAs (Personal Digital Assistants), PADs (Portable Application Descriptions), PMPs (Portable Media Players), vehicle-mounted terminals (such as vehicle-mounted navigation terminals), etc., and fixed terminals such as digital TVs, desktop computers, etc. Figure 4 The battery output power prediction device shown is only an example and should not bring any limitation to the functions and scope of use of the embodiments of the present application.

[0094] like Figure 4 As shown, the battery output power prediction device may include a processing device 1001 (e.g., a central processing unit, a graphics processor, etc.), which can perform various appropriate actions and processes according to a program stored in a read-only memory (ROM: Read Only Memory) 1002 or a program loaded from a storage device 1003 to a random access memory (RAM: Random Access Memory) 1004. In RAM1004, various programs and data required for the operation of the battery output power prediction device are also stored. The processing device 1001, ROM1002, and RAM1004 are connected to each other through a bus 1005. An input / output (I / O) interface 1006 is also connected to the bus. Generally, the following systems can be connected to the I / O interface 1006: an input device 1007 including, for example, a touch screen, a touch pad, a keyboard, a mouse, an image sensor, a microphone, an accelerometer, a gyroscope, etc.; an output device 1008 including, for example, a liquid crystal display (LCD: Liquid Crystal Display), a speaker, a vibrator, etc.; a storage device 1003 including, for example, a magnetic tape, a hard disk, etc.; and a communication device 1009. The communication device 1009 can allow the battery output power prediction device to communicate with other devices wirelessly or by wire to exchange data. Although the figure shows a battery output power prediction device with various systems, it should be understood that it is not required to implement or have all the systems shown. More or fewer systems can be implemented or provided instead.

[0095] In particular, according to the embodiments disclosed in the present application, the process described above with reference to the flowchart can be implemented as a computer software program. For example, the embodiments disclosed in the present application include a computer program product, which includes a computer program carried on a computer-readable medium, and the computer program includes a program code for executing the method shown in the flowchart. In such an embodiment, the computer program can be downloaded and installed from a network through a communication device, or installed from a storage device 1003, or installed from a ROM 1002. When the computer program is executed by the processing device 1001, the above-mentioned functions defined in the method of the embodiment disclosed in the present application are executed.

[0096] The battery output power prediction device provided by the present application adopts the battery output power prediction method in the above embodiment, which can solve the technical problem that the existing battery output power prediction methods have low prediction accuracy. Compared with the prior art, the beneficial effects of the battery output power prediction device provided by the present application are the same as the beneficial effects of the battery output power prediction method provided by the above embodiment, and the other technical features in the battery output power prediction device are the same as the features disclosed in the method of the previous embodiment, which will not be repeated here.

[0097] It should be understood that the various parts disclosed in this application can be implemented by hardware, software, firmware or a combination thereof. In the description of the above embodiments, specific features, structures, materials or characteristics can be combined in any one or more embodiments or examples in a suitable manner.

[0098] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art who is familiar with the present technical field can easily think of changes or substitutions within the technical scope disclosed in the present application, which should be included in the protection scope of the present application. Therefore, the protection scope of the present application should be based on the protection scope of the claims.

[0099] The present application provides a computer-readable storage medium having computer-readable program instructions (ie, computer programs) stored thereon, and the computer-readable program instructions are used to execute the battery output power prediction method in the above-mentioned embodiment.

[0100] The computer-readable storage medium provided in the present application may be, for example, a USB flash drive, but is not limited to electrical, magnetic, optical, electromagnetic, infrared, or semiconductor systems, systems or devices, or any combination of the above. More specific examples of computer-readable storage media may include, but are not limited to: an electrical connection with one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above. In this embodiment, the computer-readable storage medium may be any tangible medium containing or storing a program that can be used by or in combination with an instruction execution system, system or device. The program code contained on the computer-readable storage medium may be transmitted using any appropriate medium, including but not limited to: wires, optical cables, RF (Radio Frequency), etc., or any suitable combination of the above.

[0101] The computer-readable storage medium may be included in the battery output power prediction device; or may exist independently without being assembled into the battery output power prediction device.

[0102] The computer-readable storage medium carries one or more programs. When the one or more programs are executed by the battery output power prediction device, the battery output power prediction device: obtains real-time battery data under preset working conditions, wherein the real-time battery data includes voltage data, current data, temperature data and humidity data; inputs the real-time battery data into a preset neural network model to obtain an initial battery output power, wherein the preset neural network model includes an input layer, a bidirectional LSTM layer, an attention layer, a residual connection layer and a fully connected layer; based on Kalman filtering, corrects the initial battery output power to obtain a target battery output power for the preset working conditions.

[0103] Computer program code for performing the operations of the present application may be written in one or more programming languages ​​or a combination thereof, including object-oriented programming languages ​​such as Java, Smalltalk, C++, and conventional procedural programming languages ​​such as "C" or similar programming languages. The program code may be executed entirely on the user's computer, partially on the user's computer, as a separate software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In the case of a remote computer, the remote computer may be connected to the user's computer through any type of network, including a local area network (LAN) or a wide area network (WAN), or may be connected to an external computer (e.g., via the Internet using an Internet service provider).

[0104] The flow chart and block diagram in the accompanying drawings illustrate the possible architecture, function and operation of the system, method and computer program product according to various embodiments of the present application. In this regard, each square box in the flow chart or block diagram can represent a module, a program segment or a part of a code, and the module, the program segment or a part of the code contains one or more executable instructions for realizing the specified logical function. It should also be noted that in some alternative implementations, the functions marked in the square box can also occur in a sequence different from that marked in the accompanying drawings. For example, two square boxes represented in succession can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each square box in the block diagram and / or flow chart, and the combination of the square boxes in the block diagram and / or flow chart can be implemented with a dedicated hardware-based system that performs a specified function or operation, or can be implemented with a combination of dedicated hardware and computer instructions.

[0105] The modules involved in the embodiments described in this application may be implemented by software or hardware, wherein the name of the module does not constitute a limitation on the unit itself in some cases.

[0106] The readable storage medium provided by the present application is a computer-readable storage medium, which stores computer-readable program instructions (i.e., computer programs) for executing the above-mentioned battery output power prediction method, and can solve the technical problem that the existing battery output power prediction methods all have low prediction accuracy. Compared with the prior art, the beneficial effects of the computer-readable storage medium provided by the present application are the same as the beneficial effects of the battery output power prediction method provided by the above-mentioned embodiment, and will not be repeated here.

[0107] The present application also provides a computer program product, including a computer program, which implements the steps of the battery output power prediction method as described above when executed by a processor.

[0108] The computer program product provided by the present application can solve the technical problem that the existing battery output power prediction methods all have low prediction accuracy. Compared with the prior art, the beneficial effects of the computer program product provided by the present application are the same as the beneficial effects of the battery output power prediction method provided by the above embodiment, which will not be repeated here.

[0109] The above descriptions are only some embodiments of the present application, and are not intended to limit the patent scope of the present application. All equivalent structural changes made using the contents of the present application specification and drawings under the technical concept of the present application, or direct / indirect applications in other related technical fields are included in the patent protection scope of the present application.

Claims

1. A method for predicting battery output power, characterized in that: The method comprises: Acquire real-time battery data under preset working conditions, wherein the real-time battery data includes voltage data, current data, temperature data and humidity data; Inputting the real-time battery data into a preset neural network model to obtain an initial battery output power, wherein the preset neural network model includes an input layer, a bidirectional LSTM layer, an attention layer, a residual connection layer, and a fully connected layer; Based on Kalman filtering, the initial battery output power is corrected to obtain the target battery output power for the preset working condition.

2. The method according to claim 1, characterized in that Before the step of inputting the real-time battery data into a preset neural network model to obtain the initial battery output power, the step further includes: Obtain historical battery data under different operating conditions; Preprocessing the historical battery data to obtain target historical battery data; Based on a preset division ratio, the target historical battery data is divided into a training set, a validation set, and a test set; Based on the training set, the validation set and the test set, an initial neural network model is trained to obtain a preset neural network model.

3. The method according to claim 2, characterized in that The step of preprocessing the historical battery data to obtain target historical battery data includes: Removing abnormal values ​​and noise from the historical battery data to obtain cleaned historical battery data; The cleaning historical battery data is normalized to obtain target historical battery data.

4. The method according to claim 1, characterized in that The step of correcting the initial battery output power based on Kalman filtering to obtain the target battery output power for the preset working condition includes: The initial state estimate is set to the median value of the battery initial state; Determine an error covariance matrix based on variance estimates of historical battery data; Based on the initial value state estimation value, the error covariance matrix, the observation noise and the process noise matrix, the initial battery output power is corrected to obtain the target battery output power for the preset working condition.

5. The method according to claim 1, characterized in that The method further comprises: Determine the battery priority according to the target battery output power and the current battery status, wherein the current battery status includes the remaining battery power and the battery health; The distribution of the loads among the batteries is adjusted according to the battery priorities.

6. The method according to claim 1, characterized in that The method further comprises: Determine the current state of the battery according to the target battery output power; Determining whether the current charge and discharge state of the battery is a critical full state or a critical empty state; When the current state of the battery is a critical full state or a critical empty state, an early warning notification is initiated.

7. The method according to claim 1, characterized in that The method further comprises: When the time interval between the current time and the previous update time is greater than or equal to the model update period, new historical battery data under different working conditions are acquired, and a preset neural network model is updated based on the new historical battery data; and / or, Collect user feedback information on the preset neural network model, and optimize the preset neural network model according to the feedback information to improve the adaptability of the preset neural network model under different working conditions.

8. A battery output power prediction device, characterized in that: The device comprises: An acquisition module, used to acquire real-time battery data under preset working conditions, wherein the real-time battery data includes voltage data, current data, temperature data and humidity data; An input module, used to input the real-time battery data into a preset neural network model to obtain an initial battery output power, wherein the preset neural network model includes an input layer, a bidirectional LSTM layer, an attention layer, a residual connection layer, and a fully connected layer; The correction module is used to correct the initial battery output power based on Kalman filtering to obtain the target battery output power for the preset working condition.

9. A battery output power prediction device, characterized in that: The device comprises: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the computer program is configured to implement the steps of the battery output power prediction method according to any one of claims 1 to 7.

10. A storage medium, characterized in that: The storage medium is a computer-readable storage medium, and a computer program is stored on the storage medium. When the computer program is executed by a processor, the steps of the battery output power prediction method according to any one of claims 1 to 7 are implemented.

Images