A smart thermal management control system and control method for power batteries

Abstract

This invention discloses an intelligent thermal management control system and method for a power battery, including a battery module, a temperature monitoring and sensing module, an industrial control computer, and a dynamic cooling module. Each battery module is equipped with a sensing optical fiber, and the sensing optical fiber integrates multiple optical fiber sensors arranged in an array. Each optical fiber sensor in the same sensing optical fiber corresponds to a different wavelength. The industrial control computer receives the temperature from the temperature monitoring and sensing module, uses a trained temperature analysis model to predict and analyze the temperature state of the battery, makes corresponding control decisions according to preset logic, and uses a genetic algorithm to obtain the optimal liquid cooling flow rate, and controls the dynamic cooling module to adjust the coolant flow rate to prevent the battery from overheating.

Description

Technical Field

[0001] This invention belongs to the field of power energy, specifically relating to an intelligent thermal management control system and control method for power batteries. Background Technology

[0002] With the development of marine electrification, the performance and safety of power batteries, as core components of marine power systems, are receiving increasing attention. Marine lithium batteries operate in high-temperature and humid environments, and temperature fluctuations significantly impact their charge-discharge efficiency, cycle life, and safety. Currently, most marine lithium battery thermal management systems rely on passive cooling or traditional air-cooling and liquid-cooling technologies. These systems typically suffer from uneven temperature control, lag in response, and high energy consumption, especially under conditions of high-speed navigation or significant load variations, making it difficult to effectively manage the heat generated by the batteries. Furthermore, due to the limited space in marine battery cabinets, traditional cooling methods struggle to achieve efficient heat dissipation in compact environments, leading to localized overheating and increasing the risk of battery thermal runaway, potentially causing safety accidents. Existing technologies fail to adequately consider environmental adaptability and energy efficiency optimization under the special operating conditions of ships.

[0003] Existing power battery thermal management systems mostly employ fixed cooling strategies, failing to dynamically adjust according to the ship's operating conditions or actual battery temperature changes. This results in low cooling efficiency and high energy consumption. Furthermore, traditional thermal management systems have slow response times when battery temperatures are abnormal, lack intelligent safety protection mechanisms, and cannot effectively prevent thermal runaway.

[0004] CN 202310208836X discloses a battery module, a power battery pack, and a cooling control method. When the battery experiences thermal runaway, the module's flow guiding device guides the thermal runaway gas to a specific area of ​​the gas collector and discharges the gas through the exhaust pipe, thereby effectively controlling the flow of thermal runaway gas in the battery, preventing gas accumulation from causing an accident, and also dissipating heat to reduce the battery temperature.

[0005] Currently, distributed temperature monitoring using fiber optic sensors is being implemented. Fiber optic sensors monitor battery temperature using FBG (wavelength variation-based) technology, which allows multiple gratings to be placed within a single fiber, enabling multi-point, distributed temperature monitoring. Therefore, there is an urgent need for an intelligent thermal management and control system for marine lithium batteries. This system should possess temperature monitoring and sensing capabilities, dynamically adjusted cooling strategies, energy efficiency optimization, and safety protection mechanisms to ensure the safe and efficient operation of the battery under various working environments. Summary of the Invention

[0006] The purpose of this invention is to provide an intelligent thermal management control system and method for marine power batteries. By introducing an intelligent temperature monitoring and dynamic adjustment mechanism, the system can monitor the battery's operating temperature in real time and automatically adjust the cooling system according to the ship's operating status, ensuring that each battery cell is always within its optimal temperature range. Furthermore, this invention aims to improve the system's energy efficiency by optimizing cooling strategies to reduce unnecessary energy consumption and taking timely safety protection measures when the battery temperature is abnormal, thereby preventing thermal runaway and ensuring the safe operation of the battery in complex marine environments, thus improving the overall performance and reliability of the ship's power system.

[0007] The first aspect of the present invention is to provide an intelligent thermal management control system for a power battery, comprising a battery module, a temperature monitoring and sensing module, an industrial control computer, and a dynamic cooling module;

[0008] The thermal management control system is applied to a multi-layer battery cabinet, with one battery module in each layer. The battery module includes multiple battery cells arranged in an array.

[0009] The temperature monitoring and sensing module is used to detect the wavelength light information of each battery cell in real time to obtain the temperature change inside the battery cell. The obtained reflection spectrum is sequentially converted into electrical signals and then digital signals, which are then amplified and filtered. The temperature monitoring and sensing module includes a light source, an optical fiber splitter, multiple sensing optical fibers, a spectrometer, a photodetector, and a data acquisition and preprocessing unit connected in sequence. Each battery cell is equipped with one of the sensing optical fibers, and each sensing optical fiber integrates multiple optical fiber sensors arranged in an array. Each optical fiber sensor within the same sensing optical fiber corresponds to a different wavelength. The number of optical fiber sensors corresponds to the number of battery cells in the battery cell, and the optical fiber sensors are located inside the battery cells. The data acquisition and preprocessing unit acquires the electrical signals from the photodetector, amplifies and filters the electrical signals, digitizes the analog signals, and transmits them to the industrial control computer.

[0010] The industrial control computer receives temperature data from the temperature monitoring and sensing module, as well as data transmitted from the temperature sensor and the flow sensor. It uses a trained temperature analysis model to predict and analyze the temperature state of the battery and makes corresponding control decisions based on preset logic. The temperature sensor is located at the liquid inlet and outlet in the liquid cooling plate, and the flow sensor is located at the liquid inlet.

[0011] The dynamic cooling module is used to adjust the cooling pump according to the instructions of the industrial control computer, thereby regulating the flow rate of the coolant in the liquid cooling plate. The dynamic cooling module includes a cooling pump, a control valve, a liquid cooling plate, and a radiator connected in sequence. The top and bottom surfaces of the liquid cooling plate can contact adjacent battery cells. The liquid cooling plate has S-shaped liquid cooling pipes inside, which contain coolant. The inlet of the liquid cooling pipe is connected to the cooling pump through the control valve, and its outlet is connected to the radiator. The coolant circulates through the cooling pump, carrying heat from the battery module to the radiator.

[0012] Furthermore, a heat insulation plate is provided between the sidewalls of adjacent battery cells within the battery module.

[0013] Furthermore, the light source is a superfluorescent light source, the fiber optic splitter is used to distribute the light signal of the light source to each of the sensing fibers; the spectrometer is used to detect the reflection spectrum of the fiber optic sensor; the photodetector is used to read the spectral information, convert the light signal into an electrical signal, and then receive it by the data acquisition and preprocessing unit; the data acquisition and preprocessing unit includes a data acquisition card, an amplifier, and a filter.

[0014] Furthermore, the industrial control computer includes a demodulation submodule, a historical temperature storage submodule, and a temperature prediction and analysis submodule. The demodulation submodule receives digital signals from the data acquisition and preprocessing unit, runs a demodulation algorithm based on the reflectance spectrum signal, and calculates the temperature values ​​detected by each fiber optic sensor. The historical temperature storage submodule stores the temperature data obtained from the demodulation submodule as historical temperature data and forms a time-stamped temperature curve. The temperature prediction and analysis submodule controls the dynamic cooling module to adjust the coolant flow rate based on the temperature change trend obtained from the historical temperature storage submodule and the data transmitted by the temperature sensor and flow sensor.

[0015] Furthermore, the top wall of the liquid cooling plate is in close contact with the bottom surface of the battery cell, and the bottom wall of the liquid cooling plate 4 is in close contact with the tab of the battery cell.

[0016] Furthermore, the industrial computer can also adjust the fan speed of the radiator based on the prediction and analysis results.

[0017] Furthermore, the adjustment of the cooling pump is achieved by controlling the speed of the cooling pump or adjusting the opening of the control valve.

[0018] Furthermore, the intelligent thermal management control system for the power battery also includes a safety protection module. When the temperature or voltage of a single battery cell continuously exceeds a preset safety range, the industrial control computer automatically reduces the battery load and limits the current output. If the temperature continues to rise and reaches a dangerous level, the industrial control computer will quickly cut off the battery power supply to prevent thermal runaway or battery damage.

[0019] A second aspect of the present invention is to provide a control method for the intelligent thermal management control system of the power battery, comprising:

[0020] Step 1: The temperature monitoring and sensing module emits a light source, which is reflected in the fiber optic sensor in the sensing fiber. The spectrometer detects the reflection spectrum of the fiber optic sensor, and then the light is converted into an electrical signal, amplified, filtered, and then converted into a digital signal by a photodetector and a data acquisition and preprocessing unit before being sent to the industrial control computer for data processing and analysis.

[0021] Step 2: Based on the digital signal of the received reflection spectrum, run the demodulation algorithm to calculate the temperature value detected by each fiber optic sensor, and form the obtained temperature value into a battery temperature change curve;

[0022] Step 3: Determine whether the battery cell is in a charging / discharging state. If so, input the temperature change curve, battery status, and battery cell temperature difference into the trained convolutional neural network temperature analysis model for prediction and analysis. When it is determined that the dynamic cooling module needs to be activated, proceed to Step 4. The temperature analysis model is trained to obtain the charging safety temperature threshold, discharging safety temperature threshold, discharging warning temperature threshold, and extreme temperature of the battery cell.

[0023] Step 4: When Step 3 determines that the dynamic cooling module needs to be activated, based on the battery charging and discharging state and the corresponding temperature threshold determined in Step 3, a dynamic adjustment cooling strategy based on a genetic algorithm is used to obtain the optimal liquid cooling flow rate. The dynamic adjustment cooling strategy is then applied to adjust the dynamic cooling module to keep the battery temperature within a reasonable range.

[0024] Furthermore, step three specifically includes:

[0025] S31: First, determine whether the battery cell is in a charging or discharging state; if yes, execute S32; otherwise, assume the battery cell is in standby or low load state and continue monitoring.

[0026] S32: When it is determined that the battery module is in a charging state and the charging temperature of any of the battery cells is within the charging safety temperature range determined by the temperature prediction and analysis submodule, the coolant flow rate is kept constant; when the charging temperature of any of the battery cells exceeds the above-mentioned safety temperature range, step four is executed.

[0027] When it is determined that the battery module is in a discharging state, and the discharge temperature of any of the battery cells is within the safe discharge temperature range determined by the temperature prediction and analysis submodule, step four is executed to reduce the coolant flow rate; when the discharge temperature of any battery cell reaches the discharge warning temperature threshold, a warning is issued to the user to remind the operator to pay attention to the temperature change; when the discharge temperature reaches the set limit temperature, the system will automatically cut off the power supply to protect the battery.

[0028] When the battery BMS detects that the discharge rate of a single battery cell is 1C or higher, proceed to step four.

[0029] Furthermore, the dynamic adjustment of the cooling strategy in step four employs a genetic algorithm to model the problem as a multi-objective optimization problem, encoding the coolant flow rate as a binary string or real number vector; initializing a population, with each individual corresponding to a different combination of cooling strategies; comprehensively considering the three aspects of safe temperature range, energy consumption, and response speed obtained in step two, using a tournament selection method to select individuals with high fitness as the parents of the next generation, iterating to output the optimal cooling strategy;

[0030] When step three determines that the battery module is in a discharge state and the discharge temperature of any of the battery cells is within the safe discharge temperature range, the coolant flow rate is reduced until the optimal liquid cooling flow rate is obtained by the genetic algorithm.

[0031] When step three determines that the battery module is in a charging state and exceeds the above-mentioned safe temperature range, a genetic algorithm is used to determine the optimal liquid cooling flow rate, and the cooling pump is adjusted in real time according to temperature changes to increase the coolant flow rate.

[0032] When step three determines that the battery module discharge rate is 1C or higher, the dynamic cooling module is activated within 10 seconds after the load discharge, the cooling pump is adjusted to increase the coolant flow rate, and a genetic algorithm is used to determine the optimal liquid cooling flow rate to prevent the battery from overheating.

[0033] Compared with the prior art, the beneficial effects of the technical solution of the present invention are:

[0034] The present invention discloses an intelligent thermal management control system for power batteries, which effectively prevents battery overheating and reduces the risk of thermal runaway by monitoring battery temperature in real time and dynamically adjusting the dynamic cooling module, ensuring that the battery pack is always within a safe operating temperature range during navigation.

[0035] Secondly, the intelligent thermal management control system for power batteries of this invention employs an intelligent optimization algorithm to dynamically adjust the cooling strategy based on the actual operating state of the battery and environmental conditions, significantly reducing energy consumption. This optimization not only extends the battery's lifespan but also reduces ship operating costs.

[0036] Third, the intelligent thermal management control system for power batteries of this invention integrates multiple safety protection mechanisms. When abnormal temperature or fault is detected, the system can promptly issue an alarm and take corresponding measures. This design enhances the reliability of the system and reduces the potential risk of equipment damage. Attached Figure Description

[0037] Figure 1This is an overall structural diagram of the battery cabinet of a power battery intelligent thermal management control system according to the present invention;

[0038] Figure 2 Top view of the dynamic cooling module;

[0039] Figure 3 This is a schematic diagram of the working principle of the battery thermal management system;

[0040] Figure 4 This is a flowchart of the control method of the intelligent thermal management control system for power batteries.

[0041] In the picture:

[0042] 1: Tab; 2: Battery cell; 3: Heat shield; 4: Liquid cooling plate; 5: Liquid inlet; 6: Fiber optic sensor;

[0043] 7: Liquid cooling pipe; 8: Liquid outlet; 9: Amplifier; 10: Filter; 11: Industrial control computer;

[0044] 12: Cooling pump; 13: Radiator; 14: Control valve. Detailed Implementation

[0045] To make the objectives, technical solutions, beneficial effects, and significant advancements of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings provided in the examples of the present invention. Obviously, all the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0046] like Figure 3 As shown, a power battery intelligent thermal management control system includes a battery module, a temperature monitoring and sensing module, an industrial control computer 11, a dynamic cooling module, and a safety protection module.

[0047] The thermal management control system is applied to, for example Figure 1 The battery cabinet shown is a 5-layer battery cabinet, with one battery module on each layer. The battery module includes 9 battery cells 2 arranged in an array, and heat insulation plates 3 are provided between the side walls of adjacent battery cells 2.

[0048] The temperature monitoring and sensing module is used to collect temperature data of each battery cell in real time. Its internal FBG fiber optic sensor continuously detects the battery temperature at a preset sampling frequency. By measuring the internal temperature change of the battery, the temperature value is converted into an electrical signal and sent to the industrial control computer 11. In this embodiment, the temperature monitoring and sensing module includes a light source, a fiber optic splitter, five sensing fibers, a spectrometer / interferometer, a photodetector, and a data acquisition and preprocessing unit connected in sequence. The light source is connected to the fiber optic splitter, which is connected to one end of each sensing fiber. The other end of the sensing fiber is connected to the spectrometer, and the other end of the spectrometer is connected to the photodetector. The photodetector is connected to the data acquisition and preprocessing unit, which is electrically connected to the preprocessing module. The light source is a superfluorescent light source (SLED) emitting a continuous spectrum. The fiber optic splitter is used to distribute the light signal from the light source into different fibers, reducing the number of light sources in the system, thereby reducing costs and simplifying the system architecture. Depending on actual needs, the fiber optic splitter can be replaced with an optical switch. When the superfluorescent light source passes through the sensing fiber, the FBG fiber optic sensor 6 reflects light of a specific wavelength that matches its Bragg wavelength. The reflection spectrum of the FBG fiber optic sensor 6 is detected by the spectrometer. The photodetector is used to read the spectral information and convert the optical signal into an electrical signal, which is then received by the data acquisition and preprocessing unit. The data acquisition and preprocessing unit includes a data acquisition card, an amplifier 9, and a filter 10. The data acquisition card uses a 12-bit or higher resolution ADC (analog-to-digital converter). The amplified analog signal is converted into a digital signal by the ADC to ensure accurate digitization. It also needs to have multi-channel functionality to simultaneously acquire multiple signals to monitor multiple FBG sensors simultaneously. The digital signal can be directly processed by the central control unit (industrial computer) for real-time analysis. This step is crucial for the system's real-time performance, ensuring that the data accurately reflects the current temperature state of the battery. The data acquisition card acquires the electrical signal from the photodetector, amplifies the weak electrical signal from the fiber optic sensor 6 through amplifier 9 to ensure that the sensor data will not be attenuated during long-distance transmission; then, a filter is used to remove high-frequency or low-frequency interference noise, which can significantly improve the reliability of the signal, especially in the complex electrical environment of a ship; then, the amplified and filtered analog signal is digitized and transmitted to the industrial control computer 11.

[0049] Each sensing fiber integrates nine high-precision FBG (Fiber Bragg Grating) fiber optic sensors arranged in an array, with each FBG fiber optic sensor 6 corresponding to a different Bragg wavelength. One of these FBG fiber optic sensors 6 is located inside each battery cell 2 (near the cell interior or active material region) to monitor the overall and local temperature of the battery module. The FBG fiber optic sensors 6 in each layer are integrated into a single sensing fiber.

[0050] The industrial control computer 11 is the core component of the marine lithium battery intelligent thermal management control system. It is responsible for receiving data from the temperature monitoring and sensing module, analyzing the battery's temperature state, and making corresponding control decisions based on preset logic. The industrial control computer 11 is connected to both the dynamic cooling module and the temperature monitoring and sensing module, and includes a historical temperature storage submodule, a demodulation submodule, and a temperature prediction and analysis submodule. The demodulation submodule receives digital signals from the data acquisition and preprocessing unit, runs a demodulation algorithm for the FBG signal based on the reflectance spectrum signal, and accurately calculates the temperature values ​​detected by each FBG fiber optic sensor. The historical temperature storage submodule stores the temperature data obtained from the demodulation submodule as historical temperature data, and the acquired signals change over time to form a temperature curve, thus reflecting the actual thermal condition of the lithium battery during operation. The temperature prediction and analysis submodule executes the control algorithm based on the temperature change trend obtained from the historical temperature storage submodule and the data transmitted by the temperature sensor and flow sensor. The control algorithm is optimized using a convolutional neural network (CNN) and utilizes a deep learning model to predict and analyze the temperature data. Meanwhile, the temperature data of each battery cell 2 corresponding to each FBG fiber optic sensor is displayed in real time on the screen connected to the industrial control computer 11. The industrial control computer 11 can also transmit the data to a remote server via a network, facilitating remote monitoring and maintenance. The industrial control computer is electrically connected to the temperature sensor and flow sensor, enabling real-time transmission of the temperature and flow rate of the coolant in the dynamic cooling module.

[0051] The dynamic cooling module is used to adjust the operating frequency of the water pump and regulate the flow rate of the coolant according to the instructions of the industrial control computer (the industrial control computer adjusts the operating frequency of the water pump and regulates the flow rate of the coolant according to the feedback of the temperature sensor and the flow sensor). It includes a cooling pump 12, a control valve 14, a liquid cooling plate 4 and a radiator 13 connected in sequence.

[0052] Horizontally arranged liquid cooling plates 4 are provided between adjacent battery modules and below the bottommost battery module, totaling 5 layers of horizontally arranged liquid cooling plates 4. The distance between adjacent liquid cooling plates 4 is equal to the height of the power battery, so that the top wall of the liquid cooling plate 4 is in close contact with the bottom surface of the battery cell 2, and the bottom wall of the liquid cooling plate 4 is in close contact with the tab 1 of the battery cell 2. Figure 2 As shown, the liquid cooling plate 4 has a hollow structure, including a liquid cooling pipe 7, a liquid inlet 5, and a liquid outlet 8, with the liquid inlet 5 and liquid outlet 8 located at opposite ends of the liquid cooling pipe. The liquid cooling pipe 7 is S-shaped and laid inside the liquid cooling plate 4, with water as the internal working fluid. The bottom of the liquid cooling plate 4 has a groove that matches the outer contour of the tab and electrical connector. When the liquid cooling plate 4 is fastened onto the tab and electrical connector as a cover, it allows the liquid cooling plate 4 to fully contact the tab and electrical connector located on top of the battery cell 2, facilitating cooling and heat dissipation of the tab.

[0053] The inlet of the liquid cooling pipe 7 is connected to the cooling pump 12 via a control valve 14, and the outlet is connected to the radiator 13. The coolant inside circulates through the cooling pump 12, carrying heat from the battery module to the radiator 13. In the radiator 13, the heat from the coolant is transferred to the surface of the radiator through heat-conducting fins or pipes. Temperature sensors are installed at the coolant inlet and outlet to monitor the temperature difference of the coolant and understand the temperature change of the coolant flowing under the battery after heat exchange. A flow sensor is located at the coolant inlet to monitor the actual flow rate of the coolant entering the liquid cooling plate. The radiator 13 transfers heat from the coolant to the air. The radiator can be cooled by air or water. The industrial control computer 11 dynamically adjusts the radiator fan speed or the flow rate of the external cooling medium based on the battery temperature and the coolant temperature.

[0054] The fan speed of the radiator 13 is controlled by a PWM (Pulse Width Modulation) signal. The industrial computer 11 adjusts the fan motor speed via the PWM signal based on real-time temperature sensor feedback. The common fan speed adjustment range is 0-100%. The flow rate of the cooling pump 12 is adjusted by controlling the speed of the cooling pump 12 or adjusting the valve opening through a frequency converter. The industrial computer adjusts the pump's operating frequency based on temperature and flow sensor feedback to regulate the coolant flow rate.

[0055] The safety protection module monitors the battery module's operating status (such as temperature, voltage, and current) through a temperature monitoring sensor module, detects potential safety risks, and triggers a safety protection mechanism by the industrial control computer when an anomaly is detected. Its purpose is to ensure the battery operates within its normal operating range, avoiding overheating, overcharging, over-discharging, and other abnormalities to prevent safety accidents. If the temperature of a single battery cell or the voltage detected by the BMS inside the battery pack continuously exceeds a preset safety range, the industrial control computer automatically reduces the battery load and limits current output to reduce heat generation. If the temperature continues to rise and reaches a dangerous level, the industrial control computer quickly cuts off the battery power to prevent thermal runaway or battery damage, and simultaneously alarms the user. The alarm can be transmitted to the main control panel via the BMS system through sound and light to alert the user.

[0056] like Figure 4 As shown, the working process of the intelligent thermal management control system for power batteries is as follows:

[0057] Step 1: Initially, the control valve of the dynamic cooling module is opened, and the coolant flow rate can be slightly slow or even zero when the temperature does not rise sharply;

[0058] The temperature monitoring and sensing module emits an SLED light source, which is reflected by the fiber optic sensor 6 in the sensing fiber. The spectrometer detects the reflection spectrum of the FBG fiber optic sensor 6, and then the light is converted into an electrical signal, amplified, and filtered by a photodetector and a data acquisition and preprocessing unit. This process eliminates errors caused by sensor deviation, environmental noise, etc., and removes short-term abnormal noise or transient temperature spikes to prevent data instability from affecting subsequent decisions. The signal is then converted into a digital signal by an analog-to-digital converter and sent to the industrial control computer 11 for data processing and analysis. The industrial control computer 11 can perform time-synchronized processing of data from multiple sensors to ensure that the data reflects the temperature status of various parts of the battery pack at the same time, facilitating subsequent comprehensive analysis.

[0059] Step 2: After receiving the digital signal of the reflection spectrum, the industrial control computer runs the demodulation algorithm of the FBG signal to accurately calculate the temperature value detected by each FBG fiber optic sensor.

[0060] Based on the fundamental principles of fiber Bragg gratings (FBGs), the correspondence between temperature and the reflection spectrum of a FBG is determined. Specifically, temperature changes lead to changes in the Bragg wavelength, which is achieved by utilizing the periodic refractive index variation in the fiber to cause reflection of light at a specific wavelength. As temperature changes, the wavelength of the reflected light also changes, allowing temperature to be calculated by measuring this wavelength change. The relationship between temperature changes and Bragg wavelength changes can be expressed by the formula Δλ. B =α·λ B (T0)·ΔT;Δλ B This refers to the Bragg wavelength variation, α, the temperature sensitivity coefficient (usually measured in nm / ℃), which is determined by the fiber material and grating properties, and is typically a known constant. B (T0) is the Bragg wavelength at the reference temperature T0, and ΔT is the temperature change. Using the above formula, the temperature can be calculated from the change in Bragg wavelength: T0 is a known reference temperature.

[0061] The obtained temperature values ​​form a temperature curve, and the collected current temperature changes are input into a convolutional neural network temperature analysis model for prediction and analysis.

[0062] Step 3: The trained and validated convolutional neural network can be deployed to the industrial control computer 11 to process the temperature data from the FBG fiber optic sensor in real time and analyze the battery temperature change trend, battery status, and temperature differences between individual battery cells.

[0063] Before performing temperature analysis, the convolutional neural network model needs to be trained. First, historical temperature data from the FBG fiber optic sensors (6) needs to be collected and prepared, including the temperature value and timestamp for each FBG sensor, the time series of temperature changes, and external environmental data such as ambient temperature and humidity. Data preprocessing requires normalizing the temperature data and converting the time series temperature data into a format suitable for convolutional neural network processing, typically a three-dimensional tensor, i.e., the number of samples, time step, and number of features. The collected temperature data is then divided into training and testing sets according to a certain ratio: 80% training set and 20% testing set, to ensure the model has sufficient data for learning and validation.

[0064] This design employs a Convolutional Neural Network (CNN) model to extract features from time series data and predict battery temperature trends. The input layer takes data in the form of a three-dimensional tensor. The convolutional layers use multiple kernels to perform sliding convolution operations on the input time series data, extracting patterns and trends in temperature changes over time. A non-linear activation function (such as ReLU) is applied after the convolutional layers to introduce non-linearity and enhance the model's expressive power. Pooling layers reduce the size of the convolutional layer outputs, thereby reducing computation and model complexity. After feature extraction by the convolutional and pooling layers, the data is typically flattened and fed into a fully connected layer. The fully connected layer integrates the extracted features and predicts temperature trends. The output layer is a single-valued output representing the battery temperature at a future point in time, helping to provide early warning of battery overheating.

[0065] Model training defines a loss function. To minimize this loss function, gradient descent-type optimization algorithms are used to adjust the network weights. The Adam optimizer converges quickly and is suitable for handling large-scale time series data. Preprocessed temperature data is input into the CNN model and trained through multiple epochs. In each iteration, the model extracts time series features through convolutional operations and adjusts the weights to minimize the loss function. After each training iteration, the model is validated using a test set to evaluate its generalization ability and prediction accuracy.

[0066] After model training is complete, a final evaluation is performed using a test set. Mean squared error (MSE) is used to measure the difference between predicted and actual values; R is used... 2 The value measures how well the model fits the data, with 1 being the best and 0 indicating that the model has no explanatory power.

[0067] According to the temperature prediction and analysis submodule, the safe charging temperature range of each battery cell in the battery module is 0℃ to 45℃, the safe discharging temperature is -20℃ to 45℃, the discharge warning temperature is 45℃ to 60℃, and the extreme temperature is 70℃.

[0068] S31: First, determine whether the battery cell is in a charging / discharging state. The current charging / discharging state of the battery cell is monitored in the Battery Management System (BMS) by a voltage sensor to estimate the battery's State of Charge (SOC). By monitoring the battery's terminal voltage and internal resistance, the load on the battery cell is estimated. The BMS can periodically measure the battery's internal resistance using the pulse current method or AC impedance method. All sensor data (such as voltage, current, and temperature) are periodically collected by the industrial control computer of the BMS. The collection frequency is once per second, ensuring that the data collection accuracy of each sensor is within 0.1%. The system dynamically adjusts the collection cycle and accuracy according to requirements to meet the real-time requirements of different application scenarios. A suitable cooling strategy is selected based on the actual operating state of the battery, such as charging, discharging, or stationary.

[0069] When the battery is determined to be in a static state, it indicates that the individual battery cells are in standby, low-load, or dormant mode. The dynamic cooling module will automatically enter energy-saving mode to reduce unnecessary energy consumption. The low-load or dormant state referred to in this article means that the battery module is supplying power to low-power devices, such as navigation systems, communication equipment, and ship monitoring equipment, which have low power requirements from the battery.

[0070] S32: When a battery cell is determined to be in a charging / discharging state, if the temperature value of any battery cell at the current timestamp exceeds a preset threshold, the trained temperature prediction and analysis submodule analyzes the temperature change trends at multiple time points based on the severity of the temperature, the rate of increase, and the load condition. It then determines whether the temperature trend of each battery cell is steadily rising or falling, and activates or deactivates the dynamic cooling module in advance or delays its shutdown. This adjusts the flow rate or volume of the coolant to remove internal heat from the lithium battery through circulating coolant, achieving a more efficient thermal management effect. The dynamic cooling module is typically activated when the battery load is high or the ambient temperature is too high.

[0071] Charging status: When the battery module is determined to be in a charging state and the charging temperature of any of the battery cells is within the safe charging temperature range of 0°C to 45°C, the lithium battery is considered to be safe to charge. The highest charging efficiency should be achieved by considering the analysis of the temperature prediction and analysis submodule to maintain the coolant flow rate constant. When the charging temperature of any battery cell exceeds the above-mentioned safe temperature range, step four is executed to activate the dynamic cooling module to prevent the accumulation of heat generated during charging.

[0072] Discharge status: When the battery module is determined to be in a discharge state and the discharge temperature is within the safe discharge temperature range of -20℃ to 45℃, the lithium battery can be safely discharged, but the discharge performance will be affected by temperature. At low temperatures, the capacity and discharge efficiency may decrease. At this time, step four is executed to start the dynamic cooling module to reduce the flow rate.

[0073] When the discharge temperature reaches the discharge warning temperature range (45℃ to 60℃), the system will trigger the warning mechanism to remind the operator to pay attention to the temperature change and take appropriate measures; when the discharge temperature reaches the set limit temperature of 70℃, the system will automatically cut off the power supply to protect the battery.

[0074] When the battery BMS detects that a single battery cell is discharging under heavy load (i.e., the discharge rate is 1C or higher), the battery temperature rises faster. Step four (i.e., the dynamic cooling module is activated) is executed to adjust the flow control valve to regulate the flow rate of the coolant to avoid overheating; and to reduce the battery load and limit the current output to reduce heat generation. If the temperature still rises and reaches a dangerous level, the system will quickly cut off the battery power to prevent thermal runaway or battery damage.

[0075] Battery idle state: When the battery is under low load or in a dormant state, the cooling system will automatically enter energy-saving mode to reduce unnecessary energy consumption.

[0076] Step 4: When Step 3 determines that the dynamic cooling module needs to be activated, the control algorithm of the dynamic cooling module of the battery module is designed using a genetic algorithm (GA). The optimization goal of the dynamic cooling module is to transmit the temperature and flow values ​​of the coolant in the dynamic cooling module to the industrial control computer in real time based on the battery's working state (charging, discharging, and resting) and temperature changes obtained in Step 3.

[0077] The cooling strategy is dynamically adjusted to keep the battery temperature within a reasonable range, while optimizing energy consumption and cooling efficiency.

[0078] The problem is modeled as a multi-objective optimization problem: maintaining the battery temperature within a safe range (20℃-40℃); minimizing the energy consumption of the cooling system; and responding quickly to temperature changes to avoid overheating or overcooling. The coolant flow rate is encoded as a binary string or real-number vector: Liquid cooling flow rate (0-100%): controls the flow speed of the coolant; Cooling start-up temperature threshold: when the battery temperature exceeds the threshold obtained from the temperature analysis model in step two, the dynamic cooling module is activated; Energy-saving mode temperature threshold: when the temperature is lower than the preset energy-saving temperature threshold, the cooling system enters energy-saving mode, i.e., the flow rate is adjusted to 1L / min.

[0079] Initialize a population, with each individual corresponding to a different combination of cooling strategies. Individuals in the population are randomly generated to cover a large search space, ensuring the genetic algorithm can effectively search for the global optimum. Define a fitness function to evaluate the effectiveness of the cooling strategies, considering temperature maintenance range, energy consumption, and response speed. Based on the fitness value, use Tournament Selection to select individuals with high fitness as parents for the next generation. Generate new offspring through single-point crossover, and randomly select some parameters (such as temperature) for mutation in the offspring. Mutation helps prevent the algorithm from getting trapped in local optima and maintains population diversity. Replace the old population with the new offspring, repeating the above steps until the algorithm converges or reaches the maximum number of iterations. When the population fitness no longer significantly improves, or the set number of iterations is reached, the genetic algorithm stops and outputs the optimal cooling strategy.

[0080] After the genetic algorithm generates the optimal cooling strategy, the industrial control computer can dynamically apply these strategies based on the battery's operating state (charging, discharging, and stationary). When step S31 detects that the battery is charging and is within the warning threshold, the genetic algorithm is used to determine the optimal liquid cooling flow rate, which is then adjusted in real time according to temperature changes. When step S31 detects that the battery is discharging and is within the safe discharge temperature range, the genetic algorithm is used to determine the optimal liquid cooling flow rate, and the coolant flow rate is reduced. When battery discharge is detected and a heavy load is applied, the dynamic cooling module is activated within 10 seconds after the load discharge, and an optimized cooling strategy is used to prevent the battery from overheating. In stationary or low-load conditions, the cooling system operates according to an optimized energy-saving strategy to avoid unnecessary energy consumption.

[0081] According to the control method of the present invention, when the temperature of a single battery cell exceeds 60°C and the rate of temperature increase exceeds 1°C / min, especially when the temperature is close to or above 50°C and the battery discharge current exceeds 1C, it means that the battery is overheating too quickly. Therefore, the optimal liquid cooling flow rate is determined by a genetic algorithm, and the frequency of the cooling pump 12 is rapidly adjusted to increase the coolant flow rate to 5-8 L / min. When the temperature of a single battery cell is in the 45-60°C range, which is not considered severe, the optimal liquid cooling flow rate is determined by the genetic algorithm, and the flow rate of the dynamic cooling module is increased to 2-4 L / min to enhance the heat dissipation effect. When the battery module is charging and the temperature is within the safe temperature range (0-45°C), the flow rate of the dynamic cooling module is kept constant at 1-2 L / min.

[0082] This invention enables precise control by adjusting the workload of the cooling pump 12 and the coolant flow rate according to changes in battery heat generation and temperature. For example, when the temperature rises rapidly, the system increases the coolant flow rate to improve heat dissipation efficiency; conversely, when the temperature drops, the coolant flow rate decreases accordingly. The industrial control computer predicts the temperature changes of the battery system in real time and dynamically adjusts the cooling strategy or triggers alarms based on the model prediction results, ensuring the safe operation of the power battery system.

[0083] Furthermore, since the coolant in the system flows through the radiator after passing through the battery module, the dynamic liquid cooling module can further enhance the cooling effect by controlling the radiator fan speed or the flow rate of the external cooling medium.

[0084] The above embodiments are only used to illustrate the technical solutions of the present invention, and are not intended to limit them. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some or all of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present invention. Non-essential improvements, adjustments or substitutions made by those skilled in the art based on the content of this specification are all within the scope of protection claimed by the present invention.

Claims

1. A power battery intelligent thermal management control system, characterized in that... It includes a battery module, a temperature monitoring and sensing module, an industrial computer (11), and a dynamic cooling module; The thermal management control system is applied to a multi-layer battery cabinet, with a battery module set in each layer. The battery module includes multiple battery cells arranged in an array (2). The temperature monitoring and sensing module is used to detect the wavelength light information of each battery cell in real time to obtain the temperature change inside the battery cell, and converts the obtained reflection spectrum into electrical signals and digital signals in sequence before amplification and filtering. The temperature monitoring and sensing module includes a light source, an optical fiber splitter, multiple sensing optical fibers, a spectrometer, a photodetector, and a data acquisition and preprocessing unit connected in sequence. Each battery module is equipped with one of the sensing optical fibers, and the sensing optical fiber integrates multiple optical fiber sensors arranged in an array. Each optical fiber sensor in the same sensing optical fiber corresponds to a different wavelength. The number of fiber optic sensors corresponds to the number of battery cells in the battery module, and the fiber optic sensors are located inside the battery cells; the data acquisition and preprocessing unit acquires the electrical signals of the photodetector, and after amplifying and filtering the electrical signals, digitizes the analog signals and transmits them to the industrial control computer (11). The industrial control computer (11) receives temperature data from the temperature monitoring and sensing module, as well as data transmitted from the temperature sensor and the flow sensor. It uses a trained temperature analysis model to predict and analyze the temperature state of the battery and makes corresponding control decisions based on preset logic. The temperature sensor is set at the liquid inlet and liquid outlet in the liquid cooling plate, and the flow sensor is located at the liquid inlet. The dynamic cooling module is used to adjust the flow rate of the coolant in the liquid cooling plate (4) by adjusting the cooling pump (12) according to the instructions of the industrial control computer. The dynamic cooling module includes a cooling pump (12), a control valve (14), a liquid cooling plate (4) and a radiator (13) connected in sequence. The top and bottom surfaces of the liquid cooling plate can contact the adjacent battery cells. The liquid cooling plate (4) has an S-shaped liquid cooling pipe (7) inside. The liquid cooling pipe (7) contains coolant, and the inlet of the liquid cooling pipe (7) is connected to the cooling pump (12) through the control valve (14), and its outlet is connected to the radiator (13). The coolant circulates through the cooling pump (12) to carry heat from the battery module to the radiator (13).

2. The intelligent thermal management control system for power batteries according to claim 1, characterized in that, A heat insulation plate (3) is provided between the side walls of adjacent battery cells (2) within the battery module.

3. The intelligent thermal management control system for power batteries according to claim 1, characterized in that, The light source is a superfluorescent light source, and the fiber optic splitter is used to distribute the light signal of the light source to each of the sensing fibers; the spectrometer is used to detect the reflection spectrum of the fiber optic sensor (6); the photodetector is used to read the spectral information and convert the light signal into an electrical signal, which is then received by the data acquisition and preprocessing unit; the data acquisition and preprocessing unit includes a data acquisition card, an amplifier (9), and a filter (10).

4. The intelligent thermal management control system for power batteries according to claim 1, characterized in that, The industrial control computer (11) includes a demodulation submodule, a historical temperature storage submodule, and a temperature prediction and analysis submodule. The demodulation submodule receives digital signals from the data acquisition and preprocessing unit, runs a demodulation algorithm based on the signal of the reflection spectrum, and calculates the temperature values ​​detected by each fiber optic sensor. The historical temperature storage submodule stores the temperature data obtained from the demodulation submodule as historical temperature data and forms a time-stamped temperature curve; the temperature prediction and analysis submodule controls the dynamic cooling module to adjust the coolant flow rate based on the temperature change trend obtained from the historical temperature storage submodule and the data transmitted by the temperature sensor and flow sensor.

5. The intelligent thermal management control system for power batteries according to claim 4, characterized in that, The industrial computer (11) can also adjust the speed of the fan of the heat sink 13 according to the prediction and analysis results.

6. The intelligent thermal management control system for power batteries according to claim 1, characterized in that, The adjustment of the cooling pump (12) is achieved by controlling the speed of the cooling pump (12) or adjusting the opening of the control valve.

7. The intelligent thermal management control system for power batteries according to claim 1, characterized in that, The intelligent thermal management control system for power batteries also includes a safety protection module, which is used to reduce the load on the battery and limit the current output when the temperature or voltage of a single battery cell continues to exceed the preset safety range; if the temperature continues to rise and reaches the preset limit temperature, the industrial control computer will quickly cut off the battery power.

8. The control method of the intelligent thermal management control system for power batteries according to claim 1, comprising: Step 1: The temperature monitoring sensing module emits a light source, which is reflected in the fiber optic sensor (6) in the sensing fiber. The spectrometer detects the reflection spectrum of the fiber optic sensor (6), which is then converted into an electrical signal, amplified, filtered, and then converted into a digital signal by a photodetector and a data acquisition and preprocessing unit before being sent to the industrial control computer (11) for data processing and analysis. Step 2: Based on the digital signal of the received reflection spectrum, run the demodulation algorithm to calculate the temperature value detected by each fiber optic sensor, and form the obtained temperature value into a battery temperature change curve; Step 3: Determine whether the battery module is in a charging / discharging state. If so, input the temperature change curve, battery status, and individual battery cell temperature differences into the trained convolutional neural network temperature analysis model for prediction and analysis. When it is determined that the dynamic cooling module needs to be activated, proceed to Step 4. The temperature analysis model is trained to obtain the charging safety temperature threshold, discharging safety temperature threshold, discharging warning temperature threshold, and extreme temperature of the individual battery cells. Step 4: When Step 3 determines that the dynamic cooling module needs to be activated, based on the battery charging and discharging state and the corresponding temperature threshold determined in Step 3, a dynamic adjustment cooling strategy based on a genetic algorithm is used to obtain the optimal liquid cooling flow rate. The dynamic adjustment cooling strategy is then applied to adjust the dynamic cooling module to keep the battery temperature within a reasonable range.

9. The control method according to claim 8, characterized in that... Step three specifically includes: S31: First, determine whether the battery cell is in a charging or discharging state; if yes, execute S32; otherwise, assume the battery cell is in standby or low load state and continue monitoring. S32: When it is determined that the battery module is in a charging state and the charging temperature of any of the battery cells is within the charging safety temperature range determined by the temperature prediction and analysis submodule, the coolant flow rate is kept constant; when the charging temperature of any of the battery cells exceeds the above-mentioned safety temperature range, step four is executed. When it is determined that the battery module is in a discharging state, and the discharge temperature of any of the battery cells is within the discharge safety temperature range determined by the temperature prediction and analysis submodule, step four is executed to reduce the coolant flow rate; when the discharge temperature of any battery cell reaches the discharge warning temperature threshold, a warning is issued to the user to remind the operator to pay attention to the temperature change; when the discharge temperature reaches the set limit temperature, the power is cut off to protect the battery. When the battery BMS detects that the discharge rate of a single battery cell is 1C or higher, proceed to step four.

10. The control method according to claim 8, characterized in that... The fourth step specifically includes: the dynamic adjustment of the cooling strategy in the fourth step adopts a genetic algorithm to model the problem as a multi-objective optimization problem and encode the coolant flow rate as a binary string or real number vector; taking into account the three aspects of safe temperature range, energy consumption and response speed obtained in the second step, a tournament selection method is used to select individuals with high fitness as the parents of the next generation, and the process is iterated to output the optimal cooling strategy; When step three determines that the battery module is in a discharge state and the discharge temperature of any of the battery cells is within the safe discharge temperature range, the coolant flow rate is reduced until the optimal liquid cooling flow rate is obtained by the genetic algorithm. When step three determines that the battery module is in a charging state and exceeds the above-mentioned safe temperature range, the optimal liquid cooling flow rate is obtained by using a genetic algorithm, and the cooling pump (12) is adjusted in real time according to the temperature change to increase the coolant flow rate. When step three determines that the battery module discharge rate is 1C or higher, the dynamic cooling module is started in advance within 10 seconds after the load discharge, the cooling pump (12) is adjusted to increase the coolant flow rate, and the optimal liquid cooling flow rate is obtained by using a genetic algorithm to prevent the battery from overheating.

Images