Lithium battery SOC prediction method and device
By collecting and processing lithium battery data in real time, and building an SOC prediction model with algorithms such as Kalman filtering and neural networks, the problems of deviation and low accuracy of SOC prediction results in the existing technology are solved, high-precision and real-time SOC prediction are achieved, and battery management and use are optimized.
Patent Information
- Application Number
- CN202411721465.6
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-11-28
- Publication Date
- 2025-03-11
- Estimated Expiration
- Not applicable · inactive patent
AI Technical Summary
In the prior art, the prediction results of the lithium battery SOC are deviated from the actual value, the open circuit voltage method is affected by temperature and discharge rate, and the Kalman filter method has low prediction accuracy when the noise is too high or the characteristics are unknown.
The data acquisition module is used to collect the voltage, current and temperature data of lithium batteries in real time, and filter, denoising and calibration are performed through the data processing module. The SOC prediction model is constructed with algorithms such as Kalman filtering and neural networks to improve prediction accuracy.
It improves the accuracy of SOC prediction and system reliability, enhances real-time performance, meets the timely demand for SOC information of the battery management system, optimizes battery usage and extends battery life.
Abstract
Description
Technical Field
[0001] The present invention relates to the field of lithium battery technology, and in particular to a lithium battery SOC prediction method and device. Background Art
[0002] Lithium battery SOC is the state of charge of the battery, also known as the remaining capacity. It is a key parameter in the battery management system (BMS). SOC refers to the ratio of the current remaining capacity of the battery to the total capacity of the battery, usually expressed as a percentage. Its value range is 0-100%. When SOC=0, it means that the battery is fully discharged, and when SOC=100%, it means that the battery is fully charged. SOC reflects the actual availability of the battery, similar to the fuel gauge in a traditional car. Accurate SOC prediction is crucial for the safe operation, effective management and optimized energy use of the battery. The estimation of SOC is affected by many factors, including the size of the discharge current, temperature, battery pack cycle life and self-discharge. These factors will cause changes in the internal impedance of the battery and measurement errors, thereby affecting the accurate estimation of SOC. Since the SOC of lithium batteries cannot be measured directly, it is usually necessary to estimate its size through parameters such as battery terminal voltage, charge and discharge current and internal resistance. Commonly used SOC estimation methods include open circuit voltage method, ampere-hour integration method, Kalman filter method, neural network method, etc. These methods have their own advantages and disadvantages and are suitable for different application scenarios and needs. SOC prediction is widely used in electric vehicles, energy storage systems, portable electronic devices and other fields. In electric vehicles, SOC prediction can help drivers understand the remaining mileage and optimize energy use; in energy storage systems, SOC prediction helps to achieve effective scheduling and management of electricity.
[0003] The ampere-hour integration method used in the prior art does not take into account the self-discharge phenomenon of the battery, which may cause the SOC prediction result to deviate from the actual value after long-term static state. At the same time, the open circuit voltage method is used. The open circuit voltage is closely related to the temperature and discharge rate of the battery. The open circuit voltage SOC curves at different temperatures and discharge rates may be different, resulting in inaccurate prediction results. Although the Kalman filter method can suppress the influence of noise on the observation signal, if the noise is too large or the characteristics are unknown, it may affect the filtering effect, thereby affecting the prediction accuracy of the SOC, which is extremely inconvenient. Therefore, a lithium battery SOC prediction method and device are urgently needed to solve the above problems. Summary of the invention
[0004] The purpose of the present invention is to provide a lithium battery SOC prediction method and device, which solves the problem that the SOC prediction result in the prior art deviates from the actual value. At the same time, the open circuit voltage method is adopted. The open circuit voltage is closely related to the temperature and discharge rate of the battery. The open circuit voltage SOC curves at different temperatures and discharge rates may be different, resulting in inaccurate prediction results. Although the Kalman filtering method can suppress the influence of noise on the observation signal.
[0005] In order to achieve the above object, the present invention adopts the following technical solutions:
[0006] The lithium battery SOC prediction method and device include a data acquisition module: responsible for collecting relevant data of the lithium battery, voltage, current, temperature, which are the basis of SOC prediction, a data processing module: pre-processing, filtering, denoising, and calibration of the collected data to improve data quality, a prediction model module: a built-in SOC prediction model, used to calculate the SOC value based on the processed data, the prediction model may be based on a variety of algorithms, Kalman filtering, neural network, and an output module: outputting the predicted SOC value to the user or other systems.
[0007] Preferably, the method of use uses the lithium battery SOC prediction device as claimed in claim 1 to predict SOC, including data collection, data preprocessing, model selection and loading, SOC prediction calculation, and result output. The data collection module starts working to collect key parameters such as voltage, current, and temperature of the lithium battery in real time. The data processing module receives the collected raw data and performs preprocessing, including filtering noise, calibrating sensor errors, and data smoothing to ensure the quality of data input to the prediction model. According to actual needs and application scenarios, a suitable SOC prediction model is selected. The model may be based on algorithms such as Kalman filtering, neural networks, and machine learning. The pre-trained prediction model is loaded into the device to prepare for SOC prediction. The pre-processed data is input into the prediction model. The model calculates the SOC value based on the input data. The model may involve complex mathematical operations and logical judgments to accurately estimate the current battery SOC. The output module outputs the predicted SOC value to the user or other system, including displaying the SOC value on the display screen and transmitting the data to other devices through the communication interface.
[0008] Preferably, the sensor assembly includes a voltage sensor, a current sensor and a temperature sensor, which are respectively used to measure the terminal voltage, charge and discharge current and battery temperature of the lithium battery. The sensor needs to have high precision and fast response capabilities to ensure the accuracy of the data. The signal conditioning circuit amplifies, filters and calibrates the original signal output by the sensor to eliminate noise and errors and improve the stability and reliability of the signal. The data acquisition unit digitally samples the processed signal and converts it into a digital signal for subsequent data processing and analysis.
[0009] Preferably, the data processing module includes a preprocessing unit: further filtering, denoising and calibrating the collected digital signals to eliminate outliers and noise in the data and improve data quality; a feature extraction unit: extracting characteristic parameters related to SOC prediction, such as voltage change rate and current integral value, from the preprocessed data.
[0010] Preferably, the voltage, current and temperature data of the lithium battery are collected in real time through the sensor component, the collected signal is amplified, filtered and calibrated to eliminate noise and errors, the processed signal is digitally sampled to obtain a discrete digital signal, the voltage average value, change rate and other characteristic parameters are calculated, the current is integrated to obtain the battery charge and discharge capacity; the current average value, change rate and other characteristic parameters are calculated, and the average temperature, maximum temperature, minimum temperature and other characteristic parameters of the battery are recorded.
[0011] Preferably, for model selection and loading, a suitable SOC prediction model is selected according to the type, characteristics and prediction accuracy requirements of the battery. For batteries with obvious nonlinear characteristics, a neural network model can be selected; for systems that require real-time updates, a Kalman filter model can be selected, and a pre-trained prediction model is loaded from a model storage unit into a model inference unit, and the extracted feature parameters are input into the prediction model. The prediction model is used to calculate the SOC, which may involve complex mathematical operations and logical judgments, to obtain a predicted SOC value, which is output to a display unit or a communication interface.
[0012] The present invention has at least the following beneficial effects:
[0013] The present invention improves the accuracy of SOC prediction. It collects multi-dimensional data such as battery voltage, current, temperature, etc. in real time through high-precision sensors, providing a comprehensive and accurate information basis for SOC prediction. It uses advanced algorithms such as Kalman filtering and neural networks to build a prediction model, which can effectively capture the nonlinear characteristics and dynamic changes of the battery, thereby improving the accuracy of SOC prediction and enhancing system reliability. It filters, denoises and extracts features from the original data, eliminates outliers and noise in the data, improves the stability and reliability of the input data, and further enhances the stability of the prediction system. It regularly updates and maintains the prediction model to adapt to battery aging and changes in the use environment, ensuring that the prediction system maintains high accuracy and reliability for a long time. , improve real-time performance, adopt efficient algorithms and hardware platforms, realize rapid prediction of SOC, meet real-time requirements, and provide timely SOC information for the battery management system. Through real-time data collection and processing, as well as fast model reasoning, the prediction system can quickly respond to changes in battery status, providing strong support for system control and optimization, and promoting intelligent management. Through the communication interface, SOC data is transmitted to other systems or devices, realizing the sharing and integration of battery status information, and providing a data basis for intelligent management. Accurate SOC information provides an important basis for battery charging and discharging strategies, energy management strategies, etc., which helps to optimize battery use, extend battery life, and reduce system costs.
[0014] Improve user experience. The predicted SOC value is displayed in real time on the display unit, allowing users to intuitively understand the battery status, enhancing the user experience. Based on the SOC prediction results, functions such as low battery warning and overcharge and over-discharge protection can be realized, improving the safety of battery use. DETAILED DESCRIPTION
[0015] In order to make the purpose, technical solution and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not used to limit the present invention.
[0016] Embodiment 1
[0017] It includes data acquisition module: responsible for collecting relevant data of lithium batteries, voltage, current, temperature, which are the basis of SOC prediction; data processing module: pre-processing the collected data, such as filtering, denoising, and calibration to improve data quality; prediction model module: built-in SOC prediction model, used to calculate the SOC value based on the processed data. The prediction model may be based on multiple algorithms, Kalman filtering, neural network, output module: output the predicted SOC value to the user or other systems.
[0018] Embodiment 2
[0019] It includes data acquisition, data preprocessing, model selection and loading, SOC prediction calculation, and result output. The data acquisition module starts working to collect key parameters such as voltage, current, and temperature of the lithium battery in real time. The data processing module receives the collected raw data and performs preprocessing, including filtering out noise, calibrating sensor errors, and data smoothing to ensure the quality of data input into the prediction model. According to actual needs and application scenarios, a suitable SOC prediction model is selected. The model may be based on Kalman filtering, neural networks, machine learning and other algorithms. The pre-trained prediction model is loaded into the device to prepare for SOC prediction. The preprocessed data is input into the prediction model. The model calculates the SOC value based on the input data. Complex mathematical operations and logical judgments may be involved in the model to accurately estimate the current battery SOC. The output module outputs the predicted SOC value to the user or other systems, including displaying the SOC value on the display and transmitting the data to other devices through the communication interface.
[0020] Embodiment 3
[0021] It includes voltage sensor, current sensor and temperature sensor, which are used to measure the terminal voltage, charge and discharge current and battery temperature of lithium battery respectively. The sensor needs to have high precision and fast response capability to ensure the accuracy of data. Signal conditioning circuit: amplifies, filters and calibrates the original signal output by the sensor to eliminate noise and error and improve the stability and reliability of the signal. Data acquisition unit: digitally samples the processed signal and converts it into a digital signal for subsequent data processing and analysis.
[0022] Embodiment 4
[0023] The data processing module includes a preprocessing unit: further filtering, denoising and calibrating the collected digital signals to eliminate outliers and noise in the data and improve data quality; a feature extraction unit: extracting characteristic parameters related to SOC prediction from the preprocessed data, such as voltage change rate and current integral value.
[0024] Embodiment 5
[0025] The voltage, current and temperature data of the lithium battery are collected in real time through the sensor component, and the collected signals are amplified, filtered and calibrated to eliminate noise and errors. The processed signals are digitally sampled to obtain discrete digital signals, and the characteristic parameters such as the average value and change rate of the voltage are calculated. The current is integrated to obtain the charge and discharge capacity of the battery; the characteristic parameters such as the average value and change rate of the current are calculated, and the characteristic parameters such as the average temperature, maximum temperature, and minimum temperature of the battery are recorded.
[0026] Embodiment 6
[0027] Model selection and loading: select a suitable SOC prediction model based on the battery type, characteristics and prediction accuracy requirements. For batteries with obvious nonlinear characteristics, a neural network model can be selected. For systems that require real-time updates, a Kalman filter model can be selected. Load the pre-trained prediction model from the model storage unit to the model inference unit, input the extracted feature parameters into the prediction model, and use the prediction model to calculate the SOC, which may involve complex mathematical operations and logical judgments to obtain the predicted SOC value and output it to the display unit or communication interface.
[0028] The lithium battery SOC prediction device and detailed prediction method have brought significant beneficial effects in improving prediction accuracy, enhancing system reliability, improving real-time performance, promoting intelligent management, and improving user experience. These effects help optimize battery use, extend battery life, reduce system costs, and improve the performance and reliability of the entire system.
[0029] The above shows and describes the basic principles, main features and advantages of the present invention. Those skilled in the art should understand that the present invention is not limited to the above embodiments. The above embodiments and descriptions only describe the principles of the present invention. The present invention may be subject to various changes and improvements without departing from the spirit and scope of the present invention. These changes and improvements fall within the scope of the present invention. The scope of protection claimed by the present invention is defined by the attached claims and their equivalents.
Claims
1. A lithium battery SOC prediction device includes a data acquisition module: responsible for collecting relevant data of the lithium battery, such as voltage, current, and temperature, which are the basis for SOC prediction; a data processing module: preprocessing, filtering, denoising, and calibrating the collected data to improve data quality; a prediction model module: a built-in SOC prediction model, which is used to calculate the SOC value based on the processed data. The prediction model may be based on a variety of algorithms, such as Kalman filtering and neural networks; an output module: outputting the predicted SOC value to the user or other systems.
2. A lithium battery SOC prediction method, characterized in that: The method of use uses the lithium battery SOC prediction device as claimed in claim 1 to predict SOC, including data collection, data preprocessing, model selection and loading, SOC prediction calculation, and result output. The data collection module starts working to collect key parameters such as voltage, current, and temperature of the lithium battery in real time. The data processing module receives the collected raw data and performs preprocessing, including filtering noise, calibrating sensor errors, and data smoothing to ensure the quality of data input to the prediction model. According to actual needs and application scenarios, a suitable SOC prediction model is selected. The model may be based on algorithms such as Kalman filtering, neural networks, and machine learning. The pre-trained prediction model is loaded into the device to prepare for SOC prediction. The pre-processed data is input into the prediction model. The model calculates the SOC value based on the input data. The model may involve complex mathematical operations and logical judgments to accurately estimate the current battery SOC. The output module outputs the predicted SOC value to the user or other system, including displaying the SOC value on the display screen and transmitting the data to other devices through the communication interface.
3. The lithium battery SOC prediction device according to claim 1, characterized in that: The sensor assembly includes a voltage sensor, a current sensor and a temperature sensor, which are respectively used to measure the terminal voltage, charge and discharge current and battery temperature of the lithium battery. The sensor needs to have high precision and fast response capabilities to ensure the accuracy of the data. The signal conditioning circuit amplifies, filters and calibrates the original signal output by the sensor to eliminate noise and errors and improve the stability and reliability of the signal. The data acquisition unit digitally samples the processed signal and converts it into a digital signal for subsequent data processing and analysis.
4. The lithium battery SOC prediction device according to claim 1, characterized in that: The data processing module includes a preprocessing unit: further filtering, denoising and calibrating the collected digital signals to eliminate outliers and noise in the data and improve data quality; a feature extraction unit: extracting characteristic parameters related to SOC prediction, such as voltage change rate and current integral value, from the preprocessed data.
5. The lithium battery SOC prediction method according to claim 2, characterized in that: The voltage, current and temperature data of the lithium battery are collected in real time through the sensor component, the collected signals are amplified, filtered and calibrated to eliminate noise and errors, the processed signals are digitally sampled to obtain discrete digital signals, the characteristic parameters such as the average value and change rate of the voltage are calculated, the current is integrated to obtain the charge and discharge capacity of the battery; the characteristic parameters such as the average value and change rate of the current are calculated, and the characteristic parameters such as the average temperature, maximum temperature, minimum temperature and the like of the battery are recorded.
6. The lithium battery SOC prediction method according to claim 2, characterized in that: The model selection and loading selects a suitable SOC prediction model according to the type, characteristics and prediction accuracy requirements of the battery. For batteries with obvious nonlinear characteristics, a neural network model can be selected; for systems that require real-time updates, a Kalman filter model can be selected, and a pre-trained prediction model is loaded from a model storage unit into a model reasoning unit, and the extracted feature parameters are input into the prediction model. The prediction model is used to calculate the SOC, which may involve complex mathematical operations and logical judgments, to obtain a predicted SOC value, which is output to a display unit or a communication interface.