High-adaptability lead-acid-modified lithium battery power monitoring method and system

Abstract

The invention relates to the technical field of automobile battery performance monitoring, discloses a high-adaptability lead-acid-to-lithium battery power monitoring method and system, and aims to solve the problem of power monitoring after a lead-acid battery of a battery car is replaced with a lithium battery. The method and the system comprise a data acquisition layer, a data processing layer, a machine learning layer, a monitoring layer and a display interaction layer, voltage and current data of an electric group are acquired in real time, cleaning, denoising and storage processing are carried out, a voltage and current prediction model is trained by utilizing a machine learning technology, and the real-time monitoring of the electric group is realized. The curve of the output voltage and the current of the battery pack along with time change is predicted. And comparing the prediction curve with the real-time curve to generate a difference curve and mark abnormal points, finally performing visual display on a user interface, and generating an alarm signal according to abnormal point information. Accuracy and timeliness of battery car battery monitoring are improved, adaptability is high, and battery car battery monitoring requirements of different models and specifications can be met.

Description

Technical Field

[0001] The present invention relates to the technical field of battery performance monitoring, and in particular to a highly adaptable lead-acid-to-lithium battery power monitoring method and system. Background Art

[0002] In the field of electric vehicles, with the continuous development and popularization of electric vehicle technology, lead-acid batteries are widely used in electric vehicles due to their low cost and mature technology. However, as users' requirements for electric vehicle performance increase, the disadvantages of lead-acid batteries are becoming increasingly prominent, such as heavy weight, short driving range, and limited cycle life. Therefore, more and more users tend to replace lead-acid batteries in electric vehicles with lithium batteries to improve the performance of electric vehicles and user experience.

[0003] In the process of converting lead-acid batteries to lithium batteries, power monitoring has become a key link. Traditional battery monitoring of electric vehicles mostly relies on manual regular inspections or simple sensor monitoring, which cannot achieve real-time collection and analysis of key parameters of battery voltage and current, and it is difficult to detect potential performance problems in a timely manner. Therefore, we propose a highly adaptable lead-acid to lithium battery power monitoring method and system. Summary of the invention

[0004] The purpose of the present invention is to provide a highly adaptable lead-acid to lithium battery power monitoring method and system to solve the problems raised in the above background technology.

[0005] To achieve the above-mentioned purpose, the present invention provides the following technical solutions: a highly adaptable lead-acid to lithium battery power monitoring method and system, comprising a data acquisition layer, a data processing layer, a machine learning layer, a monitoring layer and a display interaction layer;

[0006] The data acquisition layer includes a sensor module and a data transmission module;

[0007] The sensor module is used to collect voltage and current data of the electric group in real time; the data transmission module is used to send the data collected by the sensor module to the data processing layer;

[0008] The data processing layer includes a data preprocessing module and a data storage module;

[0009] The data preprocessing module is used to clean and denoise the raw data collected by the sensor module, and the data storage module is used to persistently store the preprocessed data;

[0010] The machine learning layer includes a model building module;

[0011] The model building module uses machine learning technology to train two prediction models, namely a voltage change prediction model and a current prediction model. The voltage change prediction model is used to predict the curve of the output voltage changing with time according to the working conditions of the battery pack, and the current change prediction model is used to predict the curve of the output current changing with time according to the working conditions of the battery pack, and the working conditions include battery structure design, basic battery parameters and usage condition data;

[0012] The monitoring layer includes a real-time curve drawing module, a prediction model execution module and a difference analysis module; the real-time curve drawing module is used to receive the battery pack voltage and current data transmitted by the data processing layer, and draw the real-time received data into a curve that changes over time; the prediction model execution module is used to call the voltage change prediction model and the current change prediction model, and the specific steps are: i) obtaining model input parameters, including battery structure design, battery basic parameters and usage condition data; ii) passing the input parameters to the voltage change prediction model and the current change prediction model respectively; iii) receiving the predicted change curves of the output voltage and output current of the battery pack under the current input parameters; the difference analysis module is used to compare the predicted change curves output by the voltage change prediction model and the current change prediction model with the real-time change curves output by the data curve drawing module The difference between the two curve data is calculated to generate a difference curve, and the specific steps are: i) obtaining the predicted voltage change curve data and the predicted current change curve data output by the prediction model execution module, and the real-time voltage change curve data and the real-time current change curve data output by the data curve drawing module; ii) aligning the timestamps of the predicted curve data and the real-time curve data to make the data correspondence at each time point consistent; iii) calculating the difference between the predicted voltage and the real-time voltage and the difference between the predicted current and the real-time current at the corresponding time point, and generating a voltage difference curve and a current difference curve respectively; iv) judging whether each point on the voltage difference curve and the current difference curve exceeds the normal range according to a preset threshold, and if so, marking it as an abnormal point; v) transmitting the voltage difference curve, the current difference curve and the abnormal point information to the display interaction layer;

[0013] The display interaction layer includes a user interface module and an alarm processing module; the user interface module is used to receive the voltage difference curve, current difference curve and abnormal point information transmitted by the monitoring layer, and display them visually on the user interface; the alarm processing module is used to generate an alarm signal based on the received abnormal point information, and send the alarm signal to a designated alarm output device.

[0014] Preferably, the battery structure design includes battery size, battery internal structure and battery module layout, the basic battery parameters include rated voltage, rated capacity, internal resistance, charge and discharge cut-off voltage, energy density, power density, cycle life and self-discharge rate, and the usage condition data includes vehicle speed setting and motor parameters.

[0015] Preferably, the steps of the model building module training the voltage change prediction model and the current change prediction model include:

[0016] Step 1, collecting historical data, wherein the historical data is time series data of voltage and current changes of the battery pack under specific working conditions, wherein the specific working conditions include battery size, battery internal structure, battery module layout, rated voltage, rated capacity, internal resistance, charge and discharge cut-off voltage, energy density, power density, cycle life, self-discharge rate, vehicle speed setting and motor parameters;

[0017] Step 2: Clean the collected historical data to deal with missing values ​​and outliers; standardize the cleaned feature data to form a training data set; divide the training data set into a training set and a validation set;

[0018] Step 3: Select the long short-term memory network LSTM model to construct the voltage change prediction model and the current change prediction model respectively, and initialize the parameters of the model;

[0019] Step 4: Use the training set data to iteratively train the LSTM models of the voltage change prediction model and the current change prediction model. During the training process, regularly use the validation set to evaluate the model performance until the model performance meets the requirements.

[0020] Step 5: After the training is completed, the weights and biases of the LSTM models of the voltage change prediction model and the current change prediction model are saved respectively, and the models are deployed for subsequent voltage change and current change prediction.

[0021] Preferably, in the above step 2, for missing values, a linear interpolation algorithm is used to fill them; for outliers, a threshold judgment method is used to regard data points that exceed a preset threshold range as outliers and replace them with the mean of their neighboring points;

[0022] Using the Z-score standardization method, each numerical data is subtracted from its mean and divided by its standard deviation so that the processed data conforms to the standard normal distribution. The standardization formula is: Z = (X-μ) / σ, where X is the original data, μ is the mean of the original data, σ is the standard deviation of the original data, and Z is the standardized data.

[0023] Preferably, the mean square error MSE algorithm is used to evaluate the performance of the prediction model. The specific algorithm is:

[0024]

[0025] Where n is the number of data points, Y pred,i is the predicted value of the model, Y actual,i is the actual monitoring value; the prediction accuracy of the model is evaluated according to the MSE value. The smaller the MSE value, the higher the prediction accuracy of the model.

[0026] Preferably, the user interface module includes a real-time data monitoring area, a predicted data display area and a difference data comparison area; the user interface module receives the real-time battery pack voltage and current data transmitted by the data sending module, and plots it into a curve that changes with time, and displays it in the real-time data monitoring area; obtains the predicted voltage change and current change curves by calling the voltage change prediction model and the current change prediction model, and displays them in the predicted data display area; calculates the difference between the real-time monitoring data and the model predicted data by comparing the two, and plots the difference data into a curve, and displays it in the difference data comparison area.

[0027] Preferably, the user interface module includes a threshold setting function for the user to set normal range thresholds of the voltage difference curve and the current difference curve.

[0028] Preferably, the alarm processing module receives abnormal point information transmitted by the monitoring layer, which includes points on the voltage difference curve and the current difference curve that exceed a preset threshold; generates an alarm signal based on the received abnormal point information, and the alarm signal contains specific information of the abnormal point; sends the generated alarm signal to a designated alarm output device to promptly notify relevant personnel for processing.

[0029] Preferably, a highly adaptable lead-acid to lithium battery power monitoring method comprises:

[0030] Step 1: Collect the voltage and current data of the battery pack in real time, clean and denoise the collected raw data, and then store the pre-processed data persistently;

[0031] Step 2: using machine learning technology to train a voltage change prediction model and a current change prediction model, wherein the voltage change prediction model is used to predict a curve of an output voltage changing over time according to the working conditions of the battery pack, and the current change prediction model is used to predict a curve of an output current changing over time according to the working conditions of the battery pack, wherein the working conditions include battery structure design, basic battery parameters and usage condition data;

[0032] Step 3, receiving the stored battery pack voltage and current data, and plotting the real-time received data into a curve that changes with time;

[0033] Step 4: Call the voltage change prediction model and the current change prediction model to obtain model input parameters, including battery structure design, basic battery parameters and usage condition data, and pass the input parameters into the voltage change prediction model and the current change prediction model respectively, and receive the predicted change curves of the output voltage and output current of the battery pack under the current input parameters;

[0034] Step 5, comparing the difference between the predicted change curve and the real-time change curve, and generating a difference curve by calculating the difference between the two curve data, the specific steps are: obtaining the predicted voltage change curve data and the predicted current change curve data, as well as the real-time voltage change curve data and the real-time current change curve data; aligning the timestamps of the predicted curve data and the real-time curve data to make the data correspondence at each time point consistent; calculating the difference between the predicted voltage and the real-time voltage and the difference between the predicted current and the real-time current at the corresponding time points, and generating a voltage difference curve and a current difference curve respectively; judging whether each point on the voltage difference curve and the current difference curve exceeds the normal range according to a preset threshold value, and if so, marking it as an abnormal point;

[0035] Step 6: receiving the voltage difference curve, the current difference curve and the abnormal point information, and visually displaying them on the user interface;

[0036] Step 7: Generate an alarm signal based on the received abnormal point information, and send the alarm signal to a designated alarm output device.

[0037] Compared with the prior art, the present invention has the following beneficial effects:

[0038] The sensor module collects the voltage and current data of the electric group in real time, realizing continuous monitoring of key battery parameters and improving the accuracy and timeliness of the data. The machine learning layer uses machine learning technology to train voltage change prediction models and current prediction models, which can accurately predict the curves of output voltage and output current changes over time according to the working conditions of the battery pack. By comparing the predicted change curve with the real-time change curve, the monitoring layer can promptly discover potential problems in battery performance, providing a scientific basis for battery maintenance and replacement.

[0039] The present invention is suitable for power monitoring of electric battery vehicles with lead-acid batteries converted to lithium batteries, has high adaptability, and can meet the battery monitoring requirements of electric battery vehicles of different models and specifications. BRIEF DESCRIPTION OF THE DRAWINGS

[0040] Figure 1 It is the overall structure diagram of the present invention;

[0041] Figure 2 This is the workflow diagram of the difference analysis module. DETAILED DESCRIPTION

[0042] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0043] See also Figure 1-2 The present invention provides a technical solution: a highly adaptable lead-acid to lithium battery power monitoring method and system, comprising a data acquisition layer, a data processing layer, a machine learning layer, a monitoring layer and a display interaction layer.

[0044] The data acquisition layer includes the sensor module and the data transmission module. The sensor module is responsible for collecting the voltage and current data of the lithium battery pack of the electric vehicle in real time. The voltage and current data are key indicators reflecting the battery performance; the data transmission module sends the data collected by the sensor module to the data processing layer for further processing and analysis.

[0045] The data processing layer includes a data preprocessing module and a data storage module. The data preprocessing module cleans and denoises the raw data collected by the sensor module to eliminate outliers and noise in the data and improve the accuracy and reliability of the data. The data storage module stores the preprocessed data persistently for subsequent analysis and query. For missing values, a linear interpolation algorithm is used to fill them; for outliers, a threshold judgment method is used to treat data points that exceed the preset threshold range as outliers and replace them with the mean of their neighboring points; using the Z-score standardization method, each numerical data is subtracted from its mean and divided by its standard deviation, so that the processed data conforms to the standard normal distribution. The formula for standardization is: Z = (X-μ) / σ, where X is the original data, μ is the mean of the original data, σ is the standard deviation of the original data, and Z is the standardized data.

[0046] The machine learning layer includes a model building module. This module uses machine learning technology to train two prediction models: a voltage change prediction model and a current prediction model. These two models predict the curves of the output voltage and output current over time based on the working conditions of the battery pack (including battery structure design, basic battery parameters and usage condition data). Through these two models, the future performance of the battery pack can be predicted and evaluated.

[0047] The monitoring layer includes a real-time curve drawing module, a prediction model execution module and a difference analysis module. The real-time curve drawing module receives the battery pack voltage and current data transmitted by the data processing layer, and draws the real-time received data into a curve that changes over time, so as to intuitively observe the changing trend of battery performance. The prediction model execution module is responsible for calling the voltage change prediction model and the current prediction model to obtain the predicted change curve of the output voltage and output current of the battery pack. The difference analysis module compares the difference between the predicted change curve and the real-time change curve, generates a difference curve by calculating the difference between the two curve data, and judges whether each point on the difference curve exceeds the normal range according to the preset threshold. If it exceeds, it is marked as an abnormal point.

[0048] The display interaction layer includes a user interface module and an alarm processing module. The user interface module receives the voltage difference curve, current difference curve and abnormal point information transmitted by the monitoring layer, and displays them visually on the user interface so that users can intuitively understand the changes in battery performance and potential problems. The alarm processing module generates an alarm signal based on the received abnormal point information and sends the alarm signal to the designated alarm output device (such as a mobile phone, computer, etc.) so that users can take timely countermeasures.

[0049] The highly adaptable lead-acid-to-lithium battery power monitoring system described in the present invention has the advantages of strong real-time performance, strong predictive ability, high monitoring accuracy and strong applicability. It can be widely used in different types of electric vehicles to achieve comprehensive, real-time and accurate monitoring of the performance of lead-acid-to-lithium batteries.

[0050] The present invention will be further described below in conjunction with Examples 1 to 3:

[0051] Embodiment 1:

[0052] The steps of training the voltage change prediction model and the current change prediction model by the model building module include:

[0053] Step 1: Collect historical data. Historical data is the time series data of the voltage and current changes of the battery pack under specific working conditions. The specific working conditions include battery structure design (such as battery size, battery internal structure, battery module layout) and basic battery parameters (such as rated voltage, rated capacity, internal resistance, charge and discharge cut-off voltage, energy density, power density, cycle life, self-discharge rate), as well as usage condition data (such as vehicle speed setting and motor parameters). The above data is the basis for training the voltage change prediction model and the current change prediction model.

[0054] Step 2: Preprocess the collected historical data. First, clean the data to handle missing values ​​and outliers to ensure the integrity and accuracy of the data. Then, standardize the cleaned feature data to eliminate the dimensional differences between different features and improve the training efficiency of the model. Finally, divide the processed data into training set and validation set for subsequent model training and validation.

[0055] Step 3: Model initialization phase. In the model initialization phase, the LSTM model needs to be carefully parameterized to ensure that the model can be trained smoothly in the future. The specific steps include:

[0056] i) Initialize the structural parameters of the LSTM model. The parameters include the feature dimension of the input layer, which matches the number of input features to ensure that the model can receive and process all relevant input information; set the number of hidden layers and the number of hidden units in each layer, which determine the complexity and learning ability of the model; set the number of nodes in the output layer to ensure that the output of the model can meet the prediction requirements.

[0057] ii) Initialize the model weights and biases. In order to avoid the problem of gradient vanishing or exploding in the early stage of model training, small random numbers are used to initialize the model weights and biases.

[0058] iii) Set the activation function of the LSTM model. In order to control the flow and update of information, the Sigmoid function is set for the gating mechanism. In order to calculate the candidate memory cells and hidden states, the Tanh function is set as the activation function.

[0059] iv) Configure the optimization algorithm and set the learning rate parameters. The selection of the optimization algorithm and the setting of the learning rate have a crucial impact on the training effect and speed of the model. Therefore, it is necessary to make reasonable configurations based on the characteristics of the specific task and data set.

[0060] Step 4: Iterative training phase. In the iterative training phase, the LSTM model is trained using the training set data. The specific steps include:

[0061] In each iteration, a batch of input data is fed into the LSTM model. By calculating the values ​​of the input gate, forget gate, and output gate, the model can control the inflow, outflow, and update of information; then update the memory cells and hidden states to save and transmit long-term dependency information; finally, the model outputs the predicted output sequence.

[0062] The loss function value is calculated based on the predicted output and the true label to evaluate the prediction performance of the model. The gradient of the loss function with respect to the model parameters is calculated through the back propagation algorithm to update the model.

[0063] Use an optimization algorithm to update the model weights and biases based on the gradient to improve the model's predictive performance. During training, regularly use the validation set to evaluate model performance to ensure that the model has good generalization capabilities. Adjust the learning rate or stop training early as needed to prevent overfitting.

[0064] Step 5: Model saving and deployment stage. In the model saving and deployment stage, after training is completed, the weights and biases of the LSTM models of the voltage change prediction model and the current change prediction model are saved respectively. The weights and biases represent the knowledge and experience learned by the model during the training process and are the basis for the model to make predictions.

[0065] The trained model is deployed to predict voltage and current changes in actual applications. The deployed model can output the corresponding voltage and current change prediction results based on the input battery pack working condition data, providing support for performance monitoring and early warning of electric battery vehicles.

[0066] Embodiment 2:

[0067] The user interface module is responsible for displaying real-time data, predicted data and the difference between the two to the user, and provides a threshold setting function. The following is a detailed description of the implementation of the user interface module:

[0068] Interface layout design:

[0069] The user interface module consists of three main areas: real-time data monitoring area, predicted data display area and difference data comparison area. Through clear layout and intuitive graphical display, these three areas enable users to easily monitor the real-time status of the battery pack, view predicted data and compare the difference between real-time monitoring data and model predicted data.

[0070] Real-time data monitoring and display:

[0071] The user interface module receives the real-time battery pack voltage and current data transmitted by the data transmission module, and uses the graphics drawing technology to draw these data into a curve that changes over time, and updates it in real-time in the real-time data monitoring area. In this way, the user can intuitively see the real-time voltage and current status of the battery pack, as well as their changing trends.

[0072] Forecast data acquisition and display:

[0073] The user interface module obtains the predicted voltage change and current change curves in the future by calling the voltage change prediction model and the current change prediction model. These predicted data are also drawn into curves and displayed in the prediction data display area. Users can view the predicted data to understand the possible state of the battery pack in the future and make corresponding decisions.

[0074] Comparison and display of difference data:

[0075] The user interface module is also responsible for comparing the difference between the real-time monitoring data and the model prediction data. By calculating the difference between the two and plotting the difference data into a curve, the difference data is displayed in the difference data comparison area. In this way, the user can intuitively see the difference between the real-time monitoring data and the prediction data, and thus judge the prediction accuracy of the model and whether the actual status of the battery pack meets expectations.

[0076] Threshold setting function:

[0077] The user interface module includes a threshold setting function, which allows users to set the normal range thresholds of the voltage difference curve and the current difference curve according to actual needs. When the difference data exceeds the threshold set by the user, the user interface module will give corresponding prompts or alarm information so that users can take timely measures to deal with abnormal situations.

[0078] The alarm processing module is responsible for receiving the abnormal point information transmitted by the monitoring layer, generating an alarm signal accordingly, and notifying relevant personnel to handle it in a timely manner. The following is a detailed description of the implementation method of the alarm processing module:

[0079] Receive abnormal point information:

[0080] The alarm processing module first receives abnormal point information from the monitoring layer, which includes points on the voltage difference curve and the current difference curve that exceed preset thresholds, representing a significant deviation between the battery pack status and expectations.

[0081] Generate an alarm signal:

[0082] Once the abnormal point information is received, the alarm processing module will immediately generate an alarm signal based on the information. The alarm signal contains the specific information of the abnormal point, such as the location, time, voltage or current difference of the abnormal point, so that relevant personnel can accurately understand the details of the abnormal situation.

[0083] Sending an alert signal:

[0084] After the alarm signal is generated, the alarm processing module will send it to the designated alarm output device. The device may include an audible and visual alarm, a display screen, a mobile phone APP, etc., which can immediately issue an obvious alarm prompt after receiving the alarm signal to attract the attention of relevant personnel.

[0085] Notify relevant personnel promptly

[0086] Through the alarm prompts issued by the alarm output device, relevant personnel can promptly learn about the abnormal conditions of the battery pack. Users can quickly locate the problem and take corresponding treatment measures based on the specific information of the abnormal point provided in the alarm signal to ensure the normal operation of the battery pack and the safe use of the battery vehicle.

[0087] Embodiment 3:

[0088] A highly adaptable lead-acid to lithium battery power monitoring method, the specific steps comprising:

[0089] Step 1: Collect the voltage and current data of the battery pack in real time, clean and denoise the collected raw data, and then store the pre-processed data persistently;

[0090] Step 2: using machine learning technology to train a voltage change prediction model and a current change prediction model, wherein the voltage change prediction model is used to predict a curve of an output voltage changing over time according to the working conditions of the battery pack, and the current change prediction model is used to predict a curve of an output current changing over time according to the working conditions of the battery pack, wherein the working conditions include battery structure design, basic battery parameters and usage condition data;

[0091] Step 3, receiving the stored battery pack voltage and current data, and plotting the real-time received data into a curve that changes with time;

[0092] Step 4: Call the voltage change prediction model and the current change prediction model to obtain model input parameters, including battery structure design, basic battery parameters and usage condition data, and pass the input parameters into the voltage change prediction model and the current change prediction model respectively, and receive the predicted change curves of the output voltage and output current of the battery pack under the current input parameters;

[0093] Step 5, comparing the difference between the predicted change curve and the real-time change curve, and generating a difference curve by calculating the difference between the two curve data, the specific steps are: obtaining the predicted voltage change curve data and the predicted current change curve data, as well as the real-time voltage change curve data and the real-time current change curve data; aligning the timestamps of the predicted curve data and the real-time curve data to make the data correspondence at each time point consistent; calculating the difference between the predicted voltage and the real-time voltage and the difference between the predicted current and the real-time current at the corresponding time points, and generating a voltage difference curve and a current difference curve respectively; judging whether each point on the voltage difference curve and the current difference curve exceeds the normal range according to a preset threshold value, and if so, marking it as an abnormal point;

[0094] Step 6: receiving the voltage difference curve, the current difference curve and the abnormal point information, and visually displaying them on the user interface;

[0095] Step 7: Generate an alarm signal based on the received abnormal point information, and send the alarm signal to a designated alarm output device.

[0096] It should be noted that, in this article, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device.

[0097] Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present invention, and that the scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. A highly adaptable lead-acid to lithium battery power monitoring system, characterized in that: It includes data collection layer, data processing layer, machine learning layer, monitoring layer and display interaction layer; The data acquisition layer includes a sensor module and a data transmission module; The sensor module is used to collect voltage and current data of the electric group in real time; the data transmission module is used to send the data collected by the sensor module to the data processing layer; The data processing layer includes a data preprocessing module and a data storage module; The data preprocessing module is used to clean and denoise the raw data collected by the sensor module, and the data storage module is used to persistently store the preprocessed data; The machine learning layer includes a model building module; The model building module uses machine learning technology to train two prediction models, namely a voltage change prediction model and a current prediction model. The voltage change prediction model is used to predict the curve of the output voltage changing with time according to the working conditions of the battery pack, and the current change prediction model is used to predict the curve of the output current changing with time according to the working conditions of the battery pack, and the working conditions include battery structure design, basic battery parameters and usage condition data; The monitoring layer includes a real-time curve drawing module, a prediction model execution module and a difference analysis module; the real-time curve drawing module is used to receive the battery pack voltage and current data transmitted by the data processing layer, and draw the real-time received data into a curve that changes with time; the prediction model execution module is used to call the voltage change prediction model and the current change prediction model, and the specific steps are: i) obtaining model input parameters, including battery structure design, battery basic parameters and usage condition data; ii) passing the input parameters into the voltage change prediction model and the current change prediction model respectively; iii) receiving the predicted change curves of the output voltage and output current of the battery pack under the current input parameters; the difference analysis module is used to compare the difference between the predicted change curves output by the voltage change prediction model and the current change prediction model and the real-time change curve output by the data curve drawing module, and generate a difference curve by calculating the difference between the two curve data, and the specific steps are: i) obtaining the predicted voltage change curve data and the predicted current change curve data output by the prediction model execution module, and the real-time voltage change curve data and the real-time current change curve data output by the data curve drawing module; ii) aligning the timestamps of the predicted curve data and the real-time curve data so that the data at each time point corresponds to each other; iii) calculating the difference between the predicted voltage and the real-time voltage and the difference between the predicted current and the real-time current at the corresponding time point, and generating a voltage difference curve and a current difference curve respectively; iv) judging whether each point on the voltage difference curve and the current difference curve exceeds the normal range according to a preset threshold value, and if so, marking it as an abnormal point; v) transmitting the voltage difference curve, current difference curve and abnormal point information to the display interaction layer; The display interaction layer includes a user interface module and an alarm processing module; the user interface module is used to receive the voltage difference curve, current difference curve and abnormal point information transmitted by the monitoring layer, and display them visually on the user interface; the alarm processing module is used to generate an alarm signal based on the received abnormal point information, and send the alarm signal to a designated alarm output device.

2. A highly adaptable lead-acid to lithium battery power monitoring system according to claim 1, characterized in that: The battery structure design includes battery size, battery internal structure and battery module layout; the basic battery parameters include rated voltage, rated capacity, internal resistance, charge and discharge cut-off voltage, energy density, power density, cycle life and self-discharge rate; the usage condition data includes vehicle speed setting and motor parameters.

3. A highly adaptable lead-acid to lithium battery power monitoring system according to claim 2, characterized in that: The steps of training the voltage change prediction model and the current change prediction model by the model building module include: Step 1, collecting historical data, wherein the historical data is time series data of voltage and current changes of the battery pack under specific working conditions, wherein the specific working conditions include battery size, battery internal structure, battery module layout, rated voltage, rated capacity, internal resistance, charge and discharge cut-off voltage, energy density, power density, cycle life, self-discharge rate, vehicle speed setting and motor parameters; Step 2: Clean the collected historical data to deal with missing values ​​and outliers; standardize the cleaned feature data to form a training data set; divide the training data set into a training set and a validation set; Step 3: Select the long short-term memory network LSTM model to construct the voltage change prediction model and the current change prediction model respectively, and initialize the parameters of the model; Step 4: Use the training set data to iteratively train the LSTM models of the voltage change prediction model and the current change prediction model. During the training process, regularly use the validation set to evaluate the model performance until the model performance meets the requirements. Step 5: After the training is completed, the weights and biases of the LSTM models of the voltage change prediction model and the current change prediction model are saved respectively, and the models are deployed for subsequent voltage change and current change prediction.

4. A highly adaptable lead-acid to lithium battery power monitoring method according to claim 3, characterized in that: In the above step 2, for missing values, linear interpolation algorithm is used to fill them; for outliers, threshold judgment method is used to regard data points beyond the preset threshold range as outliers and replace them with the mean of their neighboring points; Using the Z-score standardization method, each numerical data is subtracted from its mean and divided by its standard deviation so that the processed data conforms to the standard normal distribution. The standardization formula is: Z = (X-μ) / σ, where X is the original data, μ is the mean of the original data, σ is the standard deviation of the original data, and Z is the standardized data.

5. A highly adaptable lead-acid to lithium battery power monitoring method according to claim 4, characterized in that: The mean square error MSE algorithm is used to evaluate the performance of the prediction model. The specific algorithm is: Where n is the number of data points, Y pred,i is the predicted value of the model, Y actual,i is the actual monitoring value; the prediction accuracy of the model is evaluated according to the MSE value. The smaller the MSE value, the higher the prediction accuracy of the model.

6. A highly adaptable lead-acid to lithium battery power monitoring system according to claim 5, characterized in that: The user interface module includes a real-time data monitoring area, a predicted data display area and a difference data comparison area; the user interface module receives the real-time battery pack voltage and current data transmitted by the data sending module, and draws it into a curve that changes with time, and displays it in the real-time data monitoring area; obtains the predicted voltage change and current change curves by calling the voltage change prediction model and the current change prediction model, and displays them in the predicted data display area; calculates the difference between the real-time monitoring data and the model predicted data by comparing the two, and draws the difference data into a curve, and displays it in the difference data comparison area.

7. A highly adaptable lead-acid to lithium battery power monitoring system according to claim 6, characterized in that: The user interface module includes a threshold setting function, which is used for the user to set the normal range thresholds of the voltage difference curve and the current difference curve.

8. A highly adaptable lead-acid to lithium battery power monitoring system according to claim 7, characterized in that: The alarm processing module receives abnormal point information transmitted by the monitoring layer, which includes points on the voltage difference curve and the current difference curve that exceed a preset threshold; generates an alarm signal based on the received abnormal point information, and the alarm signal includes specific information of the abnormal point; sends the generated alarm signal to a designated alarm output device to promptly notify relevant personnel for processing.

9. A highly adaptable lead-acid to lithium battery power monitoring method, characterized in that: include: Step 1: Collect the voltage and current data of the battery pack in real time, clean and denoise the collected raw data, and then store the pre-processed data persistently; Step 2: using machine learning technology to train a voltage change prediction model and a current change prediction model, wherein the voltage change prediction model is used to predict a curve of an output voltage changing over time according to the working conditions of the battery pack, and the current change prediction model is used to predict a curve of an output current changing over time according to the working conditions of the battery pack, wherein the working conditions include battery structure design, basic battery parameters and usage condition data; Step 3, receiving the stored battery pack voltage and current data, and plotting the real-time received data into a curve that changes with time; Step 4: Call the voltage change prediction model and the current change prediction model to obtain model input parameters, including battery structure design, basic battery parameters and usage condition data, and pass the input parameters into the voltage change prediction model and the current change prediction model respectively, and receive the predicted change curves of the output voltage and output current of the battery pack under the current input parameters; Step 5: Compare the difference between the predicted change curve and the real-time change curve, and generate a difference curve by calculating the difference between the two curve data. The specific steps are: obtain the predicted voltage change curve data and the predicted current change curve data, as well as the real-time voltage change curve data and the real-time current change curve data; align the timestamps of the predicted curve data and the real-time curve data so that the data at each time point corresponds to each other; Calculate the difference between the predicted voltage and the real-time voltage and the difference between the predicted current and the real-time current at the corresponding time point, and generate a voltage difference curve and a current difference curve respectively; Determine whether each point on the voltage difference curve and the current difference curve exceeds the normal range according to the preset threshold value, and if so, mark it as an abnormal point; Step 6: receiving the voltage difference curve, the current difference curve and the abnormal point information, and visually displaying them on the user interface; Step 7: Generate an alarm signal based on the received abnormal point information, and send the alarm signal to a designated alarm output device.

Images