Passenger car battery temperature control intelligent regulation and control method and system based on multi-source sensor data

Abstract

The invention relates to the field of navigation signal processing, and discloses a passenger car battery temperature control intelligent regulation and control method and system based on multi-source sensor data, environment data and battery operation state data are acquired by constructing a multi-source sensing array, and fusion processing and feature extraction are performed on the acquired data by using an edge calculation method; analyzing and predicting the charging and discharging behaviors of the battery based on the extracted characteristic parameters, judging the charging and discharging states of the battery, dynamically optimizing the switching threshold value of the refrigerating or heating mode of the battery, switching the mode according to the optimized switching threshold value, and performing matching analysis on the photovoltaic power generation energy and the temperature control energy consumption, so as to determine the temperature control energy consumption of the battery. According to a matching analysis result, an exclusive temperature control matrix is generated through a geographic position and meteorological data, and time-sharing gradient control is performed by using an adaptive PID control algorithm in combination with instantaneous wave cloud characteristics of photovoltaic energy storage, so that customized temperature control strategies can be provided for different environmental conditions, temperature fluctuation is avoided, and the temperature control stability is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of navigation signal processing, and more particularly to an intelligent temperature control method and system for bus batteries based on multi-source sensor data. Background Art

[0002] With the increasing severity of the energy crisis and environmental problems, new energy vehicles have gradually become the development trend of the automotive industry due to their low-carbon and environmental protection characteristics. As an important branch of new energy vehicles, electric vehicles have an increasing market share year by year. The operation of electric vehicles depends on on-vehicle batteries, and the performance and safety of the batteries are directly related to the driving safety of the vehicle and the lives of passengers. Modern electric vehicles are equipped with a variety of sensors, including temperature sensors, current sensors, voltage sensors, etc., which can monitor the status of the battery in real time. However, the data of a single sensor may have limitations and cannot comprehensively reflect the actual condition of the battery. In order to ensure that the battery works within the optimal temperature range, extend the service life of the battery, and prevent the battery from overheating or overcooling, an efficient battery temperature control system is required. Through advanced algorithms, it can analyze and process sensor data in real time, accurately monitor and intelligently control, effectively prevent the battery from overheating or overcooling, reduce battery failures and safety risks, and implement a dynamic and adaptive temperature control strategy;

[0003] However, the above process still has the following disadvantages:

[0004] First, the existing battery temperature control methods mainly switch the control mode through fixed thresholds, and cannot perceive environmental changes in real time and dynamically optimize the switching thresholds, which may lead to energy waste or insufficient temperature control caused by fixed thresholds;

[0005] Second, the existing battery temperature control methods have fixed energy distribution, lack of analysis of the matching relationship between photovoltaic power generation energy and temperature control energy consumption to optimize energy distribution, resulting in low energy utilization efficiency;

[0006] Third, the existing battery temperature control methods have a large fluctuation range for temperature control, lack of combination of geographical location information and meteorological data to control the battery temperature, resulting in poor temperature control stability. Summary of the Invention

[0007] In order to overcome the above-mentioned defects of the prior art, the present invention provides an intelligent temperature control method and system for bus batteries based on multi-source sensor data to solve the problems existing in the above background art.

[0008] The present invention provides the following technical solutions: An intelligent temperature control method for bus batteries based on multi-source sensor data, including:

[0009] S1: Construct a multi-source perception array through multi-source sensors to collect environmental data and battery operating status data in real time;

[0010] S3: Obtain the geographical location information where the vehicle is currently located in real time through the geolocation unit;

[0011] S3: Perform fusion processing on the environmental data and battery operating status data collected by the multi-source sensors through an edge computing method, and extract global environmental characteristic parameters from them;

[0012] S4: Based on the analysis of the extracted global environmental characteristic parameters, predict the charging and discharging behavior of the battery, thereby judging the charging and discharging status of the battery;

[0013] S5: Based on the dynamic optimization of the switching threshold for the cooling or heating mode during battery charging or discharging, perform mode switching according to the optimized switching threshold;

[0014] S6: Based on the switched mode, perform matching analysis on the photovoltaic power generation energy and the temperature control energy consumption, and judge whether further temperature control is required for the current battery housing temperature in the switched mode;

[0015] S7: Based on the judgment result of the battery temperature, make a prediction through geographical location information and meteorological data, and generate an exclusive temperature control matrix;

[0016] S8: By combining the exclusive temperature control matrix with the instantaneous wave cloud characteristics of photovoltaic energy storage, use an adaptive PID control algorithm for time-sharing gradient control.

[0017] Preferably, in S1, by deploying a temperature sensor, a humidity sensor, a light irradiance sensor, a thermal imaging spectrometer sensor, and a micro-current sensor at different positions on the battery housing respectively, setting each sensor to the same data acquisition frequency, and using cables to connect each sensor to a data acquisition unit, configuring the data acquisition unit to regularly receive the data collected by each sensor, thereby constructing a multi-source perception array to collect environmental data and battery operating status data in real time. The environmental data includes environmental temperature, environmental humidity, and light amplitude, and the battery operating status data includes battery temperature and battery charging and discharging current.

[0018] Preferably, the geolocation unit installs a GPS locator on the vehicle, connects the power supply, and connects it to the data acquisition unit, performs initialization configuration and calibration on the connected GPS locator, starts the GPS locator, regularly sends location requests, and receives satellite signals to obtain and calculate the position of the vehicle.

[0019] Preferably, in S3, an edge computing node is deployed in the BMS main control unit of the vehicle, which is connected to the vehicle network through the CAN bus or Ethernet and maintains a real-time communication state with the data acquisition unit. Thus, the collected environmental data and battery operation status data are directly input into the edge computing node by the data acquisition unit, and the data is synchronized, cleaned, and normalized. By using the historical environmental data and historical battery operation status data to train the LSTM neural network model, the trained model is converted into the TensorFlowLite format and deployed to the edge computing node to perform fusion processing and feature extraction on the real-time collected data. The fusion processing includes time-domain fusion, frequency-domain fusion, feature-level fusion, and decision-level fusion. Then, the application feature selection and dimensionality reduction technology are used to extract features from the fused data, and the statistics and high-order features of the feature vectors are calculated to comprehensively extract the global environmental feature parameters. Thus, the extracted global environmental feature parameters are output for managing the health of the battery.

[0020] Preferably, in S4, the logistic regression model is used as the prediction model for the battery charge and discharge behavior, and the prediction model is trained using the historical data set. Among them, the historical data set includes feature parameters and corresponding charge and discharge labels, that is, charging is 1 and discharging is 0. The feature parameters include the surface temperature distribution of the battery housing, the current intensity flowing through the battery, and the light intensity. The real-time extracted feature parameters are input into the trained model for charge and discharge behavior prediction. The specific model prediction formula is where P(Y = 1) represents the charging probability predicted by the model, e represents the natural logarithm base, T surface represents the surface temperature distribution of the battery housing, I battery represents the current intensity flowing through the battery, L light represents the light intensity, β0 represents the bias term, and β1, β2, and β3 represent the model parameters respectively. By setting a charging probability threshold θ, the predicted charging probability is compared with the charging probability threshold to judge the charge and discharge state of the battery. If P(Y = 1) ≥ θ, it indicates that the current state of the battery is in the charging state. If P(Y = 1) < θ, it indicates that the current state of the battery should not be in the charging state, specifically including the battery being in the discharging state and the battery being in the idle state. Further analyze the direction and magnitude of the current flowing through the battery to judge whether the battery is currently in the discharging state. When it is detected that there is current flowing out of the battery, it is judged that the battery is in the discharging state. When no current is detected flowing through the battery, it is judged that the battery is in the idle state, and there is no need to execute S5 for the dynamic optimization of the cooling or heating mode switching threshold.

[0021] Preferably, S5 analyzes the current temperature and charge-discharge state of the battery by collecting the temperature changes, environmental conditions, and heat generated by the battery under different charge-discharge states, and calculates the temperature change rate evaluation coefficient. The specific calculation formula is where represents the rate of change of the battery temperature with time, I represents the current flowing through the battery, R represents the internal resistance of the battery, h represents the convective heat transfer coefficient, A represents the surface area of the battery, H represents the current temperature of the battery, and H a represents the current ambient temperature, m represents the mass of the battery, and C p represents the specific heat capacity of the battery;

[0022] According to the type and specifications of the battery, set the initial cooling or heating mode switching threshold. Based on the battery state data, perform real-time dynamic optimization analysis on the initial switching threshold, and calculate the optimized cooling mode switching threshold and the optimized heating mode switching threshold respectively. The specific calculation formula of the optimized cooling mode switching threshold is T cool_new = T cool + k1×ΔH, where T cool represents the threshold for initial cooling mode switching, k1 represents the cooling threshold adjustment coefficient, and ΔH represents the temperature change rate evaluation coefficient. The specific calculation formula of the optimized heating mode switching threshold is T heat_new = T heat + k2×ΔH, where T heat represents the threshold for initial heating mode switching, and k2 represents the heating threshold adjustment coefficient;

[0023] By comparing and judging the current battery temperature with the optimized cooling mode switching threshold and the optimized heating mode switching threshold, evaluate whether the current cooling or heating mode needs to be switched. If the current battery temperature is higher than the optimized cooling mode switching threshold and the temperature change rate evaluation coefficient indicates that the battery temperature continues to rise, it indicates that the cooling mode needs to be started. If the current battery temperature is lower than the optimized heating mode switching threshold and the temperature change rate evaluation coefficient indicates that the temperature continues to drop, it indicates that the heating mode needs to be started. If the current battery temperature remains between the optimized cooling mode switching threshold and the optimized heating mode switching threshold, no mode switching is performed.

[0024] Preferably, S6 calculates the matching degree of the photovoltaic power generation energy and the energy consumption of the temperature control system by performing matching analysis on the collected photovoltaic power generation power of the battery and the energy consumption of the temperature control system as where P pv represents the photovoltaic power generation power, and P tcRepresents the energy consumption of the temperature control system; if M≥1, it indicates that the photovoltaic power generation is sufficient to support the temperature control system and no additional regulation is required. If M<1, it indicates that the photovoltaic power generation is insufficient to support the temperature control system and further regulation analysis is needed to maintain the battery temperature.

[0025] Preferably, S7 predicts the impact of the environment on battery temperature control by using the geographical location coordinates (south / north) of the vehicle and meteorological data, establishes a relationship model between environmental factors and battery temperature control requirements using a machine learning algorithm, and uses historical geographical location coordinates, meteorological data, and corresponding battery temperature changes to train and validate the relationship model. By inputting the currently collected geographical location coordinates and meteorological data of the vehicle into the trained relationship model, a dedicated temperature control matrix is output.

[0026] Preferably, S8 analyzes the photovoltaic energy storage state by real-time monitoring of the instantaneous wave cloud characteristics of the photovoltaic energy storage output by the photovoltaic energy storage system, uses the dedicated temperature control matrix to determine the ideal operating temperature range of the battery under different environmental conditions and photovoltaic energy storage states, and then combines the photovoltaic energy storage state to set the target value of battery temperature control. Using the adaptive PID control algorithm, the difference between the current battery temperature and the target value is calculated, and based on the difference between the current battery temperature and the target value output by the PID control, the reflectivity of the heat-absorbing coating, the energization waveform of the semiconductor chip, and the wind cavity vector diversion efficiency are synchronously adjusted.

[0027] To achieve the above object, the present invention provides the following technical solution: A bus battery temperature control intelligent regulation system based on multi-source sensor data, implementing the above-mentioned bus battery temperature control intelligent regulation method based on multi-source sensor data, including:

[0028] Multi-source data acquisition module: Constructs a multi-source perception array through multi-source sensors to collect environmental data and battery operating state data in real time;

[0029] Geographical location acquisition module: Real-time obtains the geographical location information where the vehicle is currently located through a geolocation unit;

[0030] Feature extraction module: Performs fusion processing on the environmental data and battery operating state data collected by the multi-source sensors through an edge computing method, and extracts global environmental characteristic parameters therefrom;

[0031] Battery behavior prediction module: Based on the analysis of the extracted global environmental characteristic parameters, predicts the charging and discharging behavior of the battery, thereby judging the charging and discharging state of the battery;

[0032] Threshold switching module: Dynamically optimizes the cooling or heating mode switching threshold for battery charging or discharging, and performs mode switching according to the optimized switching threshold;

[0033] Temperature control judgment module: Based on the switched mode, match and analyze the photovoltaic power generation energy and temperature control energy consumption to determine whether further temperature control is required for the current battery housing temperature under the switched mode;

[0034] Temperature control matrix generation module: Based on the judgment result of the battery temperature, make a prediction through geographical location information and meteorological data, and generate a dedicated temperature control matrix;

[0035] Temperature control module: By combining the dedicated temperature control matrix with the instantaneous wave cloud characteristics of photovoltaic energy storage, use the adaptive PID control algorithm for time-sharing gradient control.

[0036] Technical effects and advantages of the present invention:

[0037] (1) Based on a multi-source sensor to construct a sensing array, combined with geographical positioning and meteorological data, the system can real-time sense environmental changes, dynamically optimize the switching threshold of the cooling or heating mode, and avoid energy waste or insufficient temperature control caused by fixed thresholds.

[0038] (2) Based on the analysis of the matching relationship between photovoltaic power generation energy and temperature control energy consumption, determine whether further regulation of the current battery housing temperature is required, thereby optimizing energy distribution, reducing dependence on the power grid, and combined with the instantaneous wave cloud characteristics of photovoltaic energy storage, being able to more accurately predict photovoltaic power generation and adjust the temperature control strategy to improve energy utilization efficiency.

[0039] (3) Based on geographical location information and meteorological data, the system generates a dedicated temperature control matrix, provides a customized temperature control strategy for different environmental conditions, improves temperature control accuracy, and realizes time-sharing gradient control through the adaptive PID control algorithm to avoid temperature fluctuations and improve temperature control stability. Description of the drawings

[0040] Figure 1 It is a method step diagram of the present invention.

[0041] Figure 2 It is a system structure block diagram of the present invention. Detailed implementation manners

[0042] Next, the technical solutions in the present invention will be clearly and completely described in conjunction with the drawings in the present invention. In addition, the forms of each structure described in the following embodiments are only examples, and the intelligent temperature control method and system for bus batteries based on multi-source sensor data involved in the present invention are not limited to the structures described in the following embodiments. All other embodiments obtained by those of ordinary skill in the art without creative efforts fall within the protection scope of the present invention.

[0043] Such as Figure 1The present embodiment provides an intelligent temperature control method for bus batteries based on multi-source sensor data, including:

[0044] S1: Construct a multi-source perception array through multi-source sensors to collect environmental data and battery operating status data in real time.

[0045] In this embodiment, in S1, a temperature sensor, a humidity sensor, a light irradiance sensor, a thermal imaging spectrometer sensor, and a micro-current sensor are respectively deployed at different positions on the battery housing. Each sensor is set to the same data acquisition frequency, and each sensor is connected to the data acquisition unit using a cable. The data acquisition unit is configured to regularly receive the data collected by each sensor, thereby constructing a multi-source perception array to collect environmental data and battery operating status data in real time. The environmental data includes environmental temperature, environmental humidity, and light amplitude, and the battery operating status data includes battery temperature and battery charge and discharge current.

[0046] Specifically, according to the parameters to be monitored, such as battery temperature, environmental temperature, humidity, light intensity, charge and discharge current, etc., suitable sensors are selected, including temperature sensors, humidity sensors, light sensors, micro-current sensors, thermal imaging sensors, etc. Based on the structure and heat distribution characteristics of the battery housing, the number of sensors required and the installation position of each sensor are determined. The sensors are fixed at the predetermined positions on the battery housing to ensure firmness and without interfering with the normal operation of the battery, and the monitoring area is fully covered to obtain the best monitoring effect. All sensors are connected to the data acquisition unit using a cable; for example, temperature sensors are deployed on the surface and inside of the battery housing to monitor the temperature distribution of the battery, humidity sensors are installed inside the battery housing to monitor the environmental humidity to evaluate its impact on battery performance, light irradiance sensors are installed on the top of the battery housing to measure the sunlight intensity received by the photovoltaic panel, thermal imaging spectrometer sensors are installed on the surface of the battery housing to obtain the thermal distribution image of the battery surface, and micro-current sensors are integrated into the battery management system (BMS) to monitor the charge and discharge current of the battery.

[0047] S2: Real-time obtain the geographical location information of the vehicle where it is currently located through the geolocation unit.

[0048] In this embodiment, the geolocation unit installs a GPS locator on the vehicle, connects the power supply, and connects it to the data acquisition unit. The connected GPS locator is initialized and calibrated, the GPS locator is started, a positioning request is sent regularly, and satellite signals are received to obtain and calculate the position of the vehicle.

[0049] Specifically, install the GPS locator at a position where the vehicle can receive satellite signals well, connect the power cord of the GPS locator to the vehicle's power system to ensure that the locator can be continuously powered, set the parameters of the GPS locator, including time zone setting, satellite system selection, update rate setting, and coordinate system selection, calibrate it, then start the GPS locator, start searching for satellite signals, send location requests regularly according to the preset update rate, receive signals from at least four satellites for calculating the position. For each satellite, calculate the time difference τ = t - t′ between the time (t) when the GPS receiver receives the signal and the time (t′) when the satellite sends the signal, and use the time difference and the speed of light (c) to calculate the distance ρ = c × τ from the GPS receiver to each satellite;

[0050] Exemplarily, assume that the coordinates of four satellites are (x1, y1, z1), (x2, y2, z2), (x3, y3, z3), (x4, y4, z4) respectively, and the distances from them to the GPS receiver are ρ1, ρ2, ρ3, ρ4 respectively. Then the current geographical location of the vehicle is

[0051]

[0052] S3: Perform fusion processing on the environmental data and battery operation status data collected by multi-source sensors through edge computing methods, and extract global environmental characteristic parameters from them.

[0053] In this embodiment, S3 deploys edge computing nodes in the BMS main control unit of the vehicle, connects it to the vehicle network through CAN bus or Ethernet, and maintains a real-time communication state with the data acquisition unit. Thus, the collected environmental data and battery operation status data are directly input into the edge computing nodes by the data acquisition unit, and the data is synchronized, cleaned, and normalized. By using historical environmental data and historical battery operation status data to train the LSTM neural network model, convert the trained model into the TensorFlowLite format, and deploy it to the edge computing nodes to perform fusion processing and feature extraction on the real-time collected data. The fusion processing includes time-domain fusion, frequency-domain fusion, feature-level fusion, and decision-level fusion. Then use application feature selection and dimensionality reduction techniques to extract features from the fused data, calculate the statistics and high-order features of the feature vectors, and comprehensively extract global environmental characteristic parameters, so as to output the extracted global environmental characteristic parameters for managing the health of the battery.

[0054] Specifically, the historical environmental data and historical battery operating status data are used to train the LSTM neural network model. The training process includes forward propagation and backward propagation to optimize the loss function. Then, the trained LSTM model is converted into the TensorFlowLite format and deployed to the edge computing node for fusion processing. The execution steps of the fusion processing include time-domain fusion, frequency-domain fusion, feature-level fusion, and decision-level fusion. Through time-domain fusion, data at different time points are directly merged. Through frequency-domain fusion, the data is subjected to Fourier transform to merge the frequency components. Through feature-level fusion, features are extracted from each data source and merged into a feature vector. Through decision-level fusion, the preliminary decision results are merged. Then, feature selection is applied to calculate the statistical features of the feature vector, such as mean, median, standard deviation, variance, maximum, and minimum. The PCA algorithm is used to calculate the high-order features of the feature vector, such as peak value, skewness, entropy, and autocorrelation function. The statistical features and high-order features are combined to form a set of global environmental feature parameters.

[0055] S4: Based on the analysis of the extracted global environmental feature parameters, predict the charge and discharge behavior of the battery, thereby judging the charge and discharge state of the battery.

[0056] In this embodiment, in S4, the logistic regression model is used as the prediction model for the charge and discharge behavior of the battery, and the historical data set is used to train the prediction model. The historical data set includes feature parameters and corresponding charge and discharge labels, that is, charging is 1 and discharging is 0. The feature parameters include the surface temperature distribution of the battery housing, the current intensity flowing through the battery, and the illumination amplitude. The real-time extracted feature parameters are input into the trained model for charge and discharge behavior prediction. The specific model prediction formula is where P(Y = 1) represents the charging probability predicted by the model, e represents the natural base, T surface represents the surface temperature distribution of the battery housing, I battery represents the current intensity flowing through the battery, L light represents the illumination amplitude, β0 represents the bias term, and β1, β2, β3 represent the model parameters respectively. By setting a charging probability threshold θ, the predicted charging probability is compared with the charging probability threshold to judge the charge and discharge state of the battery. If P(Y = 1) ≥ θ, it indicates that the current state of the battery is in the charging state. If P(Y = 1) < θ, it indicates that the current state of the battery is in a state where it should not be charged, specifically including the battery being in the discharging state and the battery being in the idle state. Further analyze the direction and magnitude of the current flowing through the battery to judge whether the battery is currently in the discharging state. When it is detected that there is current flowing out of the battery, it is judged that the battery is in the discharging state. When no current is detected flowing through the battery, it is judged that the battery is in the idle state, and there is no need to execute S5 for the dynamic optimization of the cooling or heating mode switching threshold.

[0057] S5: Dynamically optimize the cooling or heating mode switching threshold based on the charging or discharging of the battery, and perform mode switching according to the optimized switching threshold.

[0058] In this embodiment, S5 collects the temperature change, environmental conditions, and heat generated by the battery under different charge and discharge states, analyzes the current temperature and charge and discharge state of the battery, and calculates the temperature change rate evaluation coefficient. The specific calculation formula is Where represents the rate of change of the battery temperature with time, I represents the current flowing through the battery, R represents the internal resistance of the battery, h represents the convective heat transfer coefficient, A represents the surface area of the battery, H represents the current temperature of the battery, H a represents the current ambient temperature, m represents the mass of the battery, C p represents the specific heat capacity of the battery;

[0059] Set the initial cooling or heating mode switching threshold according to the type and specifications of the battery. Based on the battery state data, perform real-time dynamic optimization analysis on the initial switching threshold, and calculate the optimized cooling mode switching threshold and the optimized heating mode switching threshold respectively. The specific calculation formula for the optimized cooling mode switching threshold is T cool_new = T cool + k1×ΔH, where T cool represents the threshold of the initial cooling mode switching, k1 represents the cooling threshold adjustment coefficient, ΔH represents the temperature change rate evaluation coefficient. The specific calculation formula for the optimized heating mode switching threshold is T heat_new = T heat + k2×ΔH, where T heat represents the threshold of the initial heating mode switching, k2 represents the heating threshold adjustment coefficient;

[0060] By comparing and judging the current battery temperature with the optimized cooling mode switching threshold and the optimized heating mode switching threshold, evaluate whether the current cooling or heating mode needs to be switched. If the current battery temperature is higher than the optimized cooling mode switching threshold and the temperature change rate evaluation coefficient indicates that the battery temperature continues to rise, it indicates that the cooling mode needs to be started. If the current battery temperature is lower than the optimized heating mode switching threshold and the temperature change rate evaluation coefficient indicates that the temperature continues to drop, it indicates that the heating mode needs to be started. If the current battery temperature remains between the optimized cooling mode switching threshold and the optimized heating mode switching threshold, no mode switching is performed.

[0061] S6: Based on the switched mode, match and analyze the photovoltaic power generation energy and the temperature control energy consumption, and judge whether further temperature control is required for the current battery housing temperature in the switched mode.

[0062] In this embodiment, S6 calculates the matching degree between the photovoltaic power generation energy of the battery and the energy consumption of the temperature control system by performing matching analysis on the collected photovoltaic power generation power of the battery and the energy consumption of the temperature control system, and the matching degree is where P pv represents the photovoltaic power generation power, and P tc represents the energy consumption of the temperature control system; if M≥1, it indicates that the photovoltaic power generation is sufficient to support the temperature control system and no additional regulation is required. If M<1, it indicates that the photovoltaic power generation is insufficient to support the temperature control system and further regulation analysis is needed to maintain the battery temperature.

[0063] S7: Based on the judgment result of the battery temperature, perform prediction through geographical location information and meteorological data, and generate a dedicated temperature control matrix.

[0064] In this embodiment, S7 predicts the influence of the environment on battery temperature control by using the geographical location coordinates (south / north) of the vehicle and meteorological data, establishes a relationship model between environmental factors and battery temperature control requirements by using a machine learning algorithm to establish a model, and uses historical geographical location coordinates, meteorological data, and corresponding battery temperature changes to train and verify the relationship model. The dedicated temperature control matrix is output by inputting the currently collected geographical location coordinates and meteorological data of the vehicle into the trained relationship model.

[0065] Specifically, the collected historical data, including the geographical location coordinates (south / north) of the vehicle, meteorological data (such as temperature, humidity, wind speed, solar radiation, etc.), and corresponding battery temperature changes, are divided into a training set, a validation set, and a test set. Features related to battery temperature control requirements, such as environmental temperature and solar radiation intensity, are selected to create new features, such as time series features and geographical area coding. A suitable machine learning algorithm is selected to establish a model, and the input features and output targets are defined. The model is trained using the training set data, the model parameters are adjusted, and cross-validation is performed using the validation set to optimize the model performance. Then, the accuracy, generalization ability, and robustness of the model are evaluated using the test set data. If the model performance does not meet the requirements, return to adjust the model input features, output targets, and model parameters.

[0066] S8: By combining the dedicated temperature control matrix with the instantaneous wave cloud characteristics of the photovoltaic energy storage, use the adaptive PID control algorithm for time-sharing gradient control.

[0067] In this embodiment, S8 analyzes the photovoltaic energy storage state by monitoring the instantaneous wave cloud characteristics of the photovoltaic energy storage output by the photovoltaic energy storage system in real time. A dedicated temperature control matrix is used to determine the ideal operating temperature range of the battery under different environmental conditions and photovoltaic energy storage states. Then, combined with the photovoltaic energy storage state, the target value of battery temperature control is set. The adaptive PID control algorithm is used to calculate the difference between the current battery temperature and the target value. Based on the difference between the current battery temperature and the target value output by the PID control, the reflectivity of the heat-absorbing coating, the energization waveform of the semiconductor chip, and the wind cavity vector diversion efficiency are synchronously adjusted.

[0068] As Figure 2 shown, the implementation system corresponding to the intelligent temperature control method for bus batteries based on multi-source sensor data provided in this embodiment includes a multi-source data acquisition module, a geographical location acquisition module, a feature extraction module, a battery behavior prediction module, a threshold switching module, a temperature control judgment module, a temperature control matrix generation module, and a temperature control module. The multi-source data acquisition module is connected to the feature extraction module, the feature extraction module is connected to the battery behavior prediction module, the battery behavior prediction module is connected to the threshold switching module, the threshold switching module is connected to the temperature control judgment module, the temperature control judgment module is connected to the temperature control matrix generation module, the geographical location acquisition module is connected to the temperature control matrix generation module, and the temperature control matrix generation module is connected to the temperature control module.

[0069] The multi-source data acquisition module constructs a multi-source perception array through multi-source sensors to collect environmental data and battery operation state data in real time;

[0070] The geographical location acquisition module obtains the geographical location information where the vehicle is currently located in real time through a geolocation unit;

[0071] The feature extraction module performs fusion processing on the environmental data and battery operation state data collected by the multi-source sensors through an edge computing method, and extracts global environmental feature parameters therefrom;

[0072] The battery behavior prediction module analyzes the extracted global environmental feature parameters to predict the charging and discharging behavior of the battery, thereby judging the charging and discharging state of the battery;

[0073] The threshold switching module dynamically optimizes the cooling or heating mode switching threshold for battery charging or discharging, and performs mode switching according to the optimized switching threshold;

[0074] The temperature control judgment module performs matching analysis on the photovoltaic power generation energy and the temperature control energy consumption based on the switched mode, and judges whether further temperature control is required for the current battery housing temperature in the switched mode;

[0075] Based on the judgment result of the battery temperature, the temperature control matrix generation module makes a prediction through geographical location information and meteorological data, and generates an exclusive temperature control matrix;

[0076] The temperature control module combines the exclusive temperature control matrix with the instantaneous wave cloud characteristics of the photovoltaic energy storage, and uses an adaptive PID control algorithm for time-sharing gradient control.

[0077] Finally: The above are only the preferred embodiments of the present invention and are not used to limit the present invention. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principles of the present invention shall be included in the protection scope of the present invention.

[0078] The above is only the specific implementation manner of this application, but the protection scope of this application is not limited thereto. Any person skilled in the art within the technical scope disclosed in this application can easily think of changes or replacements, which should all be covered by the protection scope of this application. Therefore, the protection scope of this application shall be subject to the protection scope of the claims.

Claims

1. An intelligent temperature control method for bus batteries based on multi-source sensor data, characterized in that, Including: S1: Construct a multi-source perception array through multi-source sensors to collect environmental data and battery operation status data in real time; S2: Obtain the geographical location information of the vehicle in real time through the geolocation unit; S3: Use edge computing methods to fuse the environmental data and battery operation status data collected by the multi-source sensors, and extract global environmental characteristic parameters from them; S4: Based on the analysis of the extracted global environmental characteristic parameters, predict the charge and discharge behavior of the battery, and thus judge the charge and discharge status of the battery; S5: Dynamically optimize the cooling or heating mode switching threshold for battery charging or discharging, and perform mode switching according to the optimized switching threshold; S6: Based on the switched mode, perform a matching analysis of the photovoltaic power generation energy and the temperature control energy consumption, and judge whether further temperature control is required for the current battery housing temperature in the switched mode; S7: Based on the judgment result of the battery temperature, make a prediction through the geographical location information and meteorological data, and generate an exclusive temperature control matrix; S8: Combine the exclusive temperature control matrix with the instantaneous wave cloud characteristics of the photovoltaic energy storage, and use the adaptive PID control algorithm for time-sharing gradient control.

2. The intelligent temperature control method for bus batteries based on multi-source sensor data according to claim 1, characterized in that In S1, the temperature sensor, humidity sensor, light irradiance sensor, thermal imaging spectral sensor, and micro-current sensor are respectively deployed at different positions on the battery housing, each sensor is set to the same data acquisition frequency, and each sensor is connected to the data acquisition unit using a cable. The data acquisition unit is configured to regularly receive the data collected by each sensor, thereby constructing a multi-source perception array to collect environmental data and battery operation status data in real time. The environmental data includes environmental temperature, environmental humidity, and light amplitude, and the battery operation status data includes battery temperature and battery charge and discharge current.

3. The intelligent temperature control method for bus batteries based on multi-source sensor data according to claim 1, wherein The geolocation unit installs a GPS locator on the vehicle, connects the power supply, and connects it to the data acquisition unit. The connected GPS locator is initialized and calibrated, the GPS locator is started, a positioning request is sent regularly, and satellite signals are received to obtain and calculate the position of the vehicle.

4. The intelligent temperature control method for bus batteries based on multi-source sensor data according to claim 1, characterized in that, S3 deploys edge computing nodes in the BMS master control unit of the vehicle, connects it to the vehicle network through the CAN bus or Ethernet, and maintains a real-time communication state with the data acquisition unit. Thus, the collected environmental data and battery operation status data are directly input into the edge computing nodes by the data acquisition unit, and the data is synchronized, cleaned, and normalized. By using historical environmental data and historical battery operation status data to train the LSTM neural network model, the trained model is converted into the TensorFlowLite format and deployed to the edge computing nodes to perform fusion processing and feature extraction on the real-time collected data. The fusion processing includes time-domain fusion, frequency-domain fusion, feature-level fusion, and decision-level fusion. Then, application feature selection and dimensionality reduction techniques are used to extract features from the fused data, and the statistics and high-order features of the feature vectors are calculated to comprehensively extract the global environmental feature parameters. Thus, the extracted global environmental feature parameters are output for managing the health of the battery.

5. The intelligent temperature control method for bus batteries based on multi-source sensor data according to claim 1, characterized in that S4 uses a logistic regression model as a prediction model for the battery charging and discharging behavior, and trains the prediction model using a historical dataset. The historical dataset includes feature parameters and corresponding charging and discharging labels, where charging is 1 and discharging is 0. The feature parameters include the surface temperature distribution of the battery housing, the current intensity flowing through the battery, and the light intensity. The feature parameters extracted in real time are input into the trained model for charging and discharging behavior prediction. The specific model prediction formula is where P(Y = 1) represents the charging probability predicted by the model, e represents the natural logarithm base, T surface represents the surface temperature distribution of the battery housing, I battery represents the current intensity flowing through the battery, L light represents the light intensity, β0 represents the bias term, and β1, β2, β3 represent the model parameters respectively. By setting a charging probability threshold θ, the predicted charging probability is compared with the charging probability threshold to determine the charging and discharging state of the battery. If P(Y = 1) ≥ θ, it indicates that the current state of the battery is in the charging state. If P(Y = 1) < θ, it indicates that the current state of the battery should not be in the charging state, specifically including the battery being in the discharging state and the battery being in the idle state. Further analyze the direction and magnitude of the current flowing through the battery to determine whether the battery is currently in the discharging state. When it is detected that there is current flowing out of the battery, it is determined that the battery is in the discharging state. When no current is detected flowing through the battery, it is determined that the battery is in the idle state, and there is no need to execute S5 for the dynamic optimization of the cooling or heating mode switching threshold.

6. The intelligent temperature control method for bus batteries based on multi-source sensor data according to claim 1, characterized in that, The S5 analyzes the current temperature and charge-discharge state of the battery by collecting the temperature changes, environmental conditions, and heat generated by the battery under different charge-discharge states, and calculates the temperature change rate evaluation coefficient. The specific calculation formula is Wherein, represents the rate of change of the battery temperature with time, I represents the current flowing through the battery, R represents the internal resistance of the battery, h represents the convective heat transfer coefficient, A represents the surface area of the battery, H represents the current temperature of the battery, and H a represents the current ambient temperature, m represents the mass of the battery, and C p represents the specific heat capacity of the battery; Set an initial cooling or heating mode switching threshold according to the type and specifications of the battery. Based on the battery status data, perform real-time dynamic optimization analysis on the initial switching threshold, and calculate the optimized cooling mode switching threshold and the optimized heating mode switching threshold respectively. The specific calculation formula for the optimized cooling mode switching threshold is T cool_new = T cool + k1×ΔH, where T cool represents the threshold for initial cooling mode switching, k1 represents the cooling threshold adjustment coefficient, and ΔH represents the temperature change rate evaluation coefficient. The specific calculation formula for the optimized heating mode switching threshold is T heat_new = T heat + k2×ΔH, where T heat represents the threshold for initial heating mode switching, and k2 represents the heating threshold adjustment coefficient; By comparing and judging the current battery temperature with the optimized cooling mode switching threshold and the optimized heating mode switching threshold, it is evaluated whether the current cooling or heating mode needs to be switched. If the current battery temperature is higher than the optimized cooling mode switching threshold and the temperature change rate evaluation coefficient indicates that the battery temperature continues to rise, it indicates that the cooling mode needs to be started. If the current battery temperature is lower than the optimized heating mode switching threshold and the temperature change rate evaluation coefficient indicates that the temperature continues to drop, it indicates that the heating mode needs to be started. If the current battery temperature remains between the optimized cooling mode switching threshold and the optimized heating mode switching threshold, no mode switching is performed.

7. The intelligent temperature control method for bus batteries based on multi-source sensor data according to claim 1, characterized in that The above S6 calculates the matching degree between the photovoltaic power generation energy of the battery and the energy consumption of the temperature control system by performing a matching analysis on the collected photovoltaic power generation power of the battery and the energy consumption of the temperature control system as Among them, P pv represents the photovoltaic power generation power, and P tc represents the energy consumption of the temperature control system; if M≥1, it indicates that the photovoltaic power generation is sufficient to support the temperature control system and no additional regulation is required. If M<1, it indicates that the photovoltaic power generation is insufficient to support the temperature control system and further regulation analysis is needed to maintain the battery temperature.

8. The intelligent temperature control method for bus batteries based on multi-source sensor data according to claim 1, characterized in that, S7 predicts the impact of the environment on battery temperature control by using the geographical location coordinates (south / north) of the vehicle and meteorological data, establishes a relationship model between environmental factors and battery temperature control requirements by using machine learning algorithms to build a model, and uses historical geographical location coordinates, meteorological data, and the corresponding battery temperature changes to train and verify the relationship model. By inputting the currently collected geographical location coordinates and meteorological data of the vehicle into the trained relationship model, a dedicated temperature control matrix is output.

9. The intelligent temperature control method for bus batteries based on multi-source sensor data according to claim 1, characterized in that, S8 analyzes the photovoltaic energy storage state by real-time monitoring of the instantaneous wave cloud characteristics of the photovoltaic energy storage output by the photovoltaic energy storage system, uses the dedicated temperature control matrix to determine the ideal operating temperature range of the battery under different environmental conditions and photovoltaic energy storage states, and then combines the photovoltaic energy storage state to set the target value of battery temperature control. Using the adaptive PID control algorithm, the difference between the current battery temperature and the target value is calculated. Based on the difference between the current battery temperature and the target value output by the PID control, the reflectivity of the heat absorption coating, the energization waveform of the semiconductor chip, and the wind cavity vector diversion efficiency are synchronously adjusted.

10. An intelligent temperature control system for bus batteries based on multi-source sensor data, which implements the intelligent temperature control method for bus batteries based on multi-source sensor data according to any one of claims 1-9, characterized in that Including: Multi-source data acquisition module: Constructs a multi-source perception array through multi-source sensors for real-time acquisition of environmental data and battery operation status data; Geographical location acquisition module: Real-time obtains the geographical location information where the vehicle is currently located through the geolocation unit; Feature extraction module: Through edge computing methods, it fuses the environmental data and battery operation status data collected by multi-source sensors, and extracts global environmental characteristic parameters from them; Battery behavior prediction module: Based on the analysis of the extracted global environmental characteristic parameters, it predicts the charging and discharging behavior of the battery, thereby judging the charging and discharging status of the battery; Threshold switching module: Based on the dynamic optimization of the cooling or heating mode switching threshold for battery charging or discharging, it performs mode switching according to the optimized switching threshold; Temperature control judgment module: Based on the switched mode, it matches and analyzes the photovoltaic power generation energy and temperature control energy consumption, and judges whether further temperature control is required for the current battery housing temperature in the switched mode; Thermostatic matrix generation module: Based on the judgment result of the battery temperature, it makes a prediction through geographical location information and meteorological data, and generates an exclusive thermostatic matrix; Temperature control module: By combining the exclusive thermostatic matrix with the instantaneous wave cloud characteristics of photovoltaic energy storage, it uses an adaptive PID control algorithm for time-sharing gradient control.

Images