Fully-immersed liquid cooling energy storage system and control method thereof

Abstract

The invention discloses a full-immersion type liquid cooling energy storage system and a control method thereof, and relates to the technical field of energy storage control, and the method comprises the steps: installing multiple sensors in a liquid immersion tank, carrying out the calibration, collecting the state data of a battery module and cooling liquid in real time through the calibrated multiple sensors, and obtaining the state data of the battery module and the cooling liquid; the state data of the battery module and the cooling liquid are preprocessed and uploaded to a control center, a Transform model is constructed in the control center, historical state data of the battery module and the cooling liquid are used for training, the trained Transform model is obtained, the preprocessed state data of the battery module and the cooling liquid are input into the trained Transform model, and the state data of the battery module and the cooling liquid are obtained. And obtaining prediction results of the high-risk area position and the expected temperature change value. The invention provides an intelligent control strategy and a multi-sensor network based on a Transform model, and local cooling is realized by monitoring the temperature change of the liquid cooling energy storage equipment in real time and adjusting the flow and path of the cooling liquid in real time.

Description

Technical Field

[0001] The present invention relates to the technical field of energy storage control, and particularly to a fully immersed liquid-cooled energy storage system and its control method. Background Art

[0002] As an emerging efficient cooling solution, fully immersed liquid-cooled energy storage has been widely applied in recent years. Its working principle is to completely immerse the battery module in the coolant and utilize the high heat conductivity of the liquid to effectively remove heat, thereby maintaining the working temperature of the battery within a safe range. However, there are still some deficiencies in its existing technology.

[0003] On the one hand, the control strategy of traditional fully immersed liquid-cooled energy storage is relatively simple and lacks adaptability, making it difficult to accurately adjust the flow path and flow rate of the coolant according to real-time changing load conditions. On the other hand, the data collection in the existing technology is not comprehensive enough. The traditional collection method can only obtain limited state information and lacks comprehensive monitoring of the temperature distribution inside the entire battery module. Summary of the Invention

[0004] In view of the above existing problems, the present invention is proposed.

[0005] Therefore, the present invention provides a fully immersed liquid-cooled energy storage system and its control method to solve the problems of insufficiently intelligent cooling control strategy and incomplete data collection.

[0006] To solve the above technical problems, the present invention provides the following technical solutions: In the first aspect, the present invention provides a fully immersed liquid-cooled energy storage control method, which includes: Install multiple sensors in the liquid immersion tank and calibrate them; Utilize the calibrated multiple sensors to collect the state data of the battery module and the cooling liquid in real time; Preprocess the state data of the battery module and the cooling liquid and upload it to the control center; Build a Transformer model in the control center and train it using the historical state data of the battery module and the cooling liquid to obtain a trained Transformer model; Input the preprocessed state data of the battery module and the cooling liquid into the trained Transformer model to obtain the prediction results of the high-risk area position and the expected temperature change value; Based on the prediction results of the high-risk area position and the expected temperature change value, the control center automatically adjusts the working states of the micro pump and the micro valve, changes the flow path and flow rate of the cooling liquid, and realizes local cooling.

[0007] As a preferred solution of the fully immersed liquid cooling energy storage control method of the present invention, wherein: installing multiple sensors in the liquid immersion tank and calibrating them specifically includes the following steps, Place the temperature sensor in a thermostat, measure the readings, and compare them with the temperature of the thermostat. Use linear regression to calibrate the temperature sensor; Place the current sensor under a standard current source with a known current, measure the readings, and compare them with the known current value. Use the least squares method to calibrate the current sensor; Place the flow rate sensor under a standard fluid with a known flow rate, measure the readings, and compare them with the known flow rate value. Use linear interpolation to calibrate the flow rate sensor; The standard current source is defined based on a known current value, and the standard fluid is defined based on a known flow rate value and constant temperature and pressure conditions.

[0008] As a preferred solution of the fully immersed liquid cooling energy storage control method of the present invention, wherein: using the calibrated multiple sensors to collect the status data of the battery module and the cooling liquid specifically includes the following steps, Use the calibrated temperature sensor and the method of arranging embedded temperature monitoring points and timing sampling to collect the temperature data of the battery module and the cooling liquid; Use the calibrated current sensor and the method of combining Coulomb counting with temperature compensation to collect the charging status data of the battery module; Install the calibrated flow rate sensor on the coolant flow path to collect the flow rate data of the cooling liquid.

[0009] As a preferred solution of the fully immersed liquid cooling energy storage control method of the present invention, wherein: the preprocessing includes outlier processing, normalization processing, and difference processing; Convert the format of the status data after preprocessing, encrypt it using TLS, establish a secure connection with the control center through the HTTP protocol, and upload the status data.

[0010] As a preferred solution of the fully immersed liquid cooling energy storage control method of the present invention, wherein: constructing a Transformer model in the control center and training it using the historical status data of the battery module and the cooling liquid to obtain a trained Transformer model specifically includes the following steps, Select the TensorFlow deep learning framework in the control center and construct a Transformer model using the multi-head self-attention mechanism, feed-forward neural network, residual connection, and layer normalization; Input the historical status data of the battery module and the cooling liquid into the Transformer model for training; During the training process, batch gradient descent is used for iterative training, and L2 regularization and early stopping are used to prevent overfitting. Forward propagation is used to calculate the predicted values and losses, and backward propagation is used to calculate the gradients and update the model parameters using the Adam optimizer.

[0011] As a preferred embodiment of the fully immersed liquid-cooled energy storage control method of the present invention, wherein: the state data of the preprocessed battery module and the cooling liquid are input into the trained Transformer model to obtain the prediction results of the high-risk area position and the expected temperature change value, which specifically includes the following steps. Convert the state data of the preprocessed battery module and the cooling liquid into a three-dimensional tensor structure. The three-dimensional tensor structure includes the number of samples, the number of time steps, and the number of feature dimensions. Use the multi-head self-attention mechanism for the three-dimensional tensor, calculate the attention weights, and perform weighted summation to obtain the high-risk area position and the expected temperature change value.

[0012] As a preferred embodiment of the fully immersed liquid-cooled energy storage control method of the present invention, wherein: based on the prediction results of the high-risk area position and the expected temperature change value, the control center automatically adjusts the working states of the micro-pump and the micro-valve, changes the flow path and flow rate of the cooling liquid, and realizes local cooling, which specifically includes the following steps. Based on the obtained high-risk area position information, automatically identify the positions of the battery modules that need to be cooled preferentially, and combine the change value of the expected temperature to adjust the working states of the micro-pump and the micro-valve. By adjusting the output power of the micro-pump and using the pulse width modulation and PID control algorithms, control the flow rate of the coolant flowing to the high-risk area position. By controlling the opening degree of the micro-valve and using the stepping motor control and closed-loop feedback, change the flow path of the coolant, and cover the coolant to the positions of the battery modules that need to be cooled, so as to realize local cooling.

[0013] In a second aspect, the present invention provides a fully immersed liquid-cooled energy storage control system, including a calibration module, a data acquisition module, a preprocessing module, a model training module, a result prediction module, and an adjustment module. The calibration module is used to install multiple sensors in the liquid immersion tank and perform calibration. The data acquisition module is used to use the calibrated multiple sensors to collect the state data of the battery module and the cooling liquid in real time. The preprocessing module is used to preprocess the state data of the battery module and the cooling liquid and upload it to the control center. A model training module, configured to build a Transformer model in a control center and train it using the state data of historical battery modules and cooling liquid, so as to obtain a trained Transformer model; A result prediction module, configured to input the preprocessed state data of the battery module and the cooling liquid into the trained Transformer model to obtain prediction results of the high-risk area location and the expected temperature change value; An adjustment module, configured to input the preprocessed state data of the battery module and the cooling liquid into the trained Transformer model to obtain prediction results of the high-risk area location and the expected temperature change value.

[0014] In a third aspect, the present invention provides a computer device, including a memory and a processor, where the memory stores a computer program, and: when the computer program is executed by the processor, any step of the full-immersion liquid-cooled energy storage control method described in the first aspect of the present invention is implemented.

[0015] In a fourth aspect, the present invention provides a computer-readable storage medium, on which a computer program is stored, and: when the computer program is executed by the processor, any step of the full-immersion liquid-cooled energy storage control method described in the first aspect of the present invention is implemented.

[0016] The beneficial effects of the present invention are as follows: An intelligent control strategy based on a Transformer model is proposed, which predicts future temperature changes according to real-time monitoring data and automatically adjusts the coolant flow path and flow rate, significantly improving the dynamic response ability and self-adaptability, ensuring accurate temperature control under complex working conditions, and improving the safety and service life of the battery. By using a multi-sensor network, comprehensive monitoring of the internal temperature distribution of the entire battery module is realized, and data quality is improved through data preprocessing technology, which not only provides richer state information but also enhances the ability to identify potential high-temperature risk areas, thereby effectively preventing faults from occurring. Description of the Drawings

[0017] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following will briefly introduce the drawings required for the description of the embodiments. Obviously, the following drawings are only some embodiments of the present invention. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative efforts.

[0018] Figure 1 It is a flowchart of the full-immersion liquid-cooled energy storage control method in Embodiment 1.

[0019] Figure 2 It is a schematic diagram of the full-immersion liquid-cooled energy storage system in Embodiment 1. Detailed Embodiments

[0020] To make the above objects, features, and advantages of the present invention more apparent and understandable, the following provides a detailed description of the specific embodiments of the present invention in conjunction with the accompanying drawings of the specification.

[0021] In the following description, many specific details are set forth to facilitate a full understanding of the present invention. However, the present invention may also be implemented in other ways different from those described herein. Those skilled in the art may make similar generalizations without departing from the spirit of the present invention. Therefore, the present invention is not limited by the specific embodiments disclosed below.

[0022] Secondly, the so-called "one embodiment" or "embodiment" herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation manner of the present invention. The appearances of "in one embodiment" in different places in this specification do not all refer to the same embodiment, nor are they separate or alternative embodiments that exclude each other with other embodiments.

[0023] Embodiment 1, referring to Figure 1 and Figure 2 , which is the first embodiment of the present invention. This embodiment provides a fully immersed liquid-cooled energy storage system and its control method, including the following steps: S1. Install multiple sensors in the liquid immersion tank and calibrate them.

[0024] Specifically, the following operations are included. When calibrating the temperature sensor, select a high-precision constant temperature bath as a reference. Place the temperature sensor to be calibrated in the constant temperature bath, set different temperature points, record the readings of the sensor at each temperature point, and compare them with the actual temperature value of the constant temperature bath. Use the linear regression analysis method to determine the relationship between the sensor output and the actual temperature, and find the best-fit line by minimizing the sum of squared errors. Adjust the output of the sensor according to the slope and intercept of the line for calibration. When calibrating the current sensor, use a standard current source with a known current as a reference device. Connect the current sensor to the standard current source, record the readings of the sensor at each set current value, compare them with the known current value, and perform analysis using the least squares method to obtain the best-fit relationship between the sensor output and the actual current value. Based on this fitting relationship, adjust the gain and offset of the sensor for calibration. When calibrating the flow rate sensor, use a standard fluid device, place the flow rate sensor therein, test it under different preset flow rate conditions, select multiple flow rate points, record the readings of the sensor, compare them with the known flow rate values, and use the method of linear interpolation to adjust the output of the sensor for calibration. The above standard current sources are defined based on known current values. They are devices used to provide stable current outputs and can accurately generate the required current intensity according to the set values. The standard fluids are defined based on known flow rate values ​​and constant temperature and pressure conditions. They are designed for calibrating flow rate sensors and are intended to eliminate interference from external factors in the calibration process.

[0025] S2. Use calibrated multi-sensors to collect status data of battery modules and cooling liquid in real time.

[0026] The specific operations include the following: When collecting temperature data of battery modules and cooling liquid, an embedded design is used inside the battery module and the coolant flow path, and calibrated temperature sensors are arranged to directly monitor local temperature changes. At the same time, a timed sampling method is used to ensure the continuity and accuracy of the data. For example, a temperature reading is set every 5 seconds or 1 minute. This method can not only capture instantaneous temperature changes, but also analyze its change trend over time, providing accurate data support for subsequent thermal management; When collecting the charging status data of the battery module, the coulomb counting method is used to estimate the remaining capacity of the battery. The coulomb counting method calculates the charging status based on measuring the total charge flowing through the battery. In specific operations, the current sensor continuously monitors the current flowing into and out of the battery, and accumulates this data to calculate the charging status of the battery module. At the same time, the charging status estimation value is adjusted in combination with the temperature compensation technology, and the current measurement value is adjusted according to the operating temperature of the battery, thereby improving the accuracy of the collected charging status data; When collecting the flow rate data of the cooling liquid, the flow rate sensor is installed on the coolant flow path, usually at the pump outlet and the main branch point of the cooling circuit. Commonly used flow rate sensors include ultrasonic flow meters and turbine flow meters, which can provide high-precision flow rate measurement. At the same time, in order to ensure the accuracy of the data, a pulse output mode can be used, that is, a pulse signal is generated each time the fluid passes through the sensor, and the flow rate is calculated by counting the number of pulses. In addition, the data from the temperature sensor can be combined to comprehensively analyze the operating conditions of the liquid-cooled energy storage to ensure the accuracy of the collected flow rate data.

[0027] S3. Pre-process the status data of the battery module and cooling liquid and upload them to the control center.

[0028] The specific steps include: First, perform data denoising and outlier removal. In the specific operation, a low-pass filter is used to remove high-frequency noise. The low-pass filter allows frequencies below the set filtering threshold to pass through while attenuating frequency components above this filtering threshold. First-order and second-order infinite impulse response filters can be used to achieve this. Then, the Z-Score method is adopted to detect and remove outliers. Z-Score calculates the standard score of each data point, that is, the difference between this point and the mean divided by the standard deviation. If the absolute value of the Z-Score of a certain data point exceeds the outlier removal threshold, it is considered an outlier and is removed, thus ensuring that the data quality for subsequent analysis is not affected by outliers; Subsequently, perform data standardization. The Min-Max standardization algorithm is used to map the collected data features to the same scale. Min-Max standardization scales the data to a specific interval, usually [0,1], through linear transformation. This not only makes the data of different features comparable but also facilitates the training and optimization of subsequent machine learning models; Next, perform time synchronization and interpolation processing. Synchronize the collected data according to the timestamps and use the interpolation method to fill in the missing time point data. Since the sensors may work at different sampling frequencies, it is necessary to ensure that the time axes of all data are consistent. By checking the timestamps of each data point, discontinuous time periods are found. For linear interpolation, the value of the missing point can be predicted by calculating the linear equation between adjacent two points. For spline interpolation, a piecewise polynomial function is used to fit the data curve. The specific operation can be performed using the Pandas library in Python to ensure the coherence and integrity of the time series data; Finally, convert the format of the preprocessed status data. Format conversion usually involves converting the status data into a format that is easy to parse in JSON. For example, use the json module in Python to achieve data serialization. Once the data format conversion is completed, use TLS encryption to ensure that the data will not be eavesdropped or tampered with during network transmission. TLS encryption can be automatically achieved through the HTTPS protocol to ensure end-to-end security. For example, in Python, use the requests library to send an HTTP request with TLS encryption to upload the data to the control center; Through the above operations, the process of preprocessing from denoising, standardization to time synchronization and encrypted upload is realized. These high-quality preprocessed data provide a solid foundation for the subsequent training of the Transformer model, improving the overall performance and reliability of the energy storage device.

[0029] S4. Build a Transformer model in the control center and use the historical status data of the battery module and the cooling liquid for training to obtain a trained Transformer model.

[0030] Specifically, it includes the following steps: Select TensorFlow as the deep learning framework in the control center, and build a Transformer model based on TensorFlow. The core components of this Transformer model include multi-head self-attention mechanism, feed-forward neural network, residual connection, and layer normalization. The multi-head self-attention mechanism captures the dependencies between different positions in the input sequence through multiple independent attention heads. Each attention head calculates the attention weights through the dot product of query, key, and value matrices, thereby highlighting the important time-step features. The feed-forward neural network is used to further extract deep features. It consists of two layers of linear transformation and activation functions, and can capture complex patterns in the data. Residual connection and layer normalization are used to alleviate the vanishing gradient problem in deep networks respectively, and ensure that the output of each layer has a stable distribution, which helps to accelerate convergence and improve model performance; Input the preprocessed state data of the battery module and the cooling liquid into the Transformer model for training. These state data are organized in the form of time series. Each input sample contains data of multiple time steps, and each time step contains multiple feature dimensions, such as temperature, flow rate, providing rich information for the Transformer model; During the training process, the batch gradient descent method is used for iterative training. In each iteration, a small batch of data is randomly sampled from the dataset to calculate the gradient. This method improves the computational efficiency and helps the model to jump out of local optima. To prevent overfitting, L2 regularization and early stopping method are adopted. L2 regularization limits the model complexity by adding a weight decay term to the loss function, thereby avoiding overfitting the training data. The early stopping method terminates the training in advance when the performance on the validation set no longer improves, preventing the model from overfitting; The specific steps of training include forward propagation to calculate the predicted value and loss, and backward propagation to calculate the gradient. The specific mathematical formulas are: ; Among them, represents the loss value, represents the number of samples, represents the sample index, represents the true value of the th sample, represents the predicted value of the th sample; ; Among them, represents the loss value the gradient of the weight , true value vector, represents the predicted value vector, represents the input data; The mathematical formula for calculating the bias gradient is: ; where, represents the loss value the gradient of the bias of, the true value vector, represents the predicted value vector; At the same time, during the iteration of the Adam optimizer, based on the calculated gradients, it updates the first - moment estimate and the second - moment estimate. The first - moment estimate helps the model accelerate the learning rate along the steep direction, while the second - moment estimate adaptively adjusts according to the historical gradient magnitudes of the parameters, enabling the learning rate to vary dynamically for each parameter. The Adam optimizer uses the updated first - moment estimate, second - moment estimate, and the bias - corrected values to calculate a new adaptive learning rate for each parameter and adjusts the parameter values accordingly, gradually optimizing the model performance; Through the above operations, a Transformer model was successfully constructed and trained in the control center, ensuring the accuracy and generalization ability of the model, providing a solid foundation for further analysis and decision - making.

[0031] S5. Input the pre - processed status data of the battery module and the cooling liquid into the trained Transformer model to obtain the prediction results of the high - risk area location and the expected temperature change value.

[0032] Specifically, it includes the following steps: Convert the pre - processed status data of the battery module and the cooling liquid into a three - dimensional tensor structure. This process includes the number of samples, the number of time steps, and the number of feature dimensions. For example, there are 100 samples, each sample contains 50 time steps, and each time step has 10 features such as temperature and flow rate. These data can be organized into a tensor with a shape of (100, 50, 10). This step is implemented in the deep - learning framework TensorFlow, using the view function to adjust the shape of the data to ensure it meets the input requirements of the Transformer model; Next, the multi - head self - attention mechanism is used to process the three - dimensional tensor. The core of the multi - head self - attention mechanism lies in calculating the attention weights for each time step. For each time step, query matrix, key matrix, and value matrix are generated through linear transformation, and then attention scores are calculated. This process involves dot - product operations and the softmax function to ensure that the sum of attention weights is 1. In this way, the model can automatically identify which time - step information is most important for prediction. For example, when detecting battery overheating, this mechanism can automatically focus on those time periods when the temperature rises rapidly, thus improving the accuracy of prediction; Based on the attention weights obtained from the above operations, the position of the high - risk area is further calculated. The specific formula is: ; where, represents the coordinates of the position of the high - risk area, represents the position of the high - risk area on the axis in three - dimensional space, represents the position of the high - risk area on the axis in three - dimensional space, represents the position of the high - risk area on the axis in three - dimensional space, represents the weight matrix of dimension 3×d, represents the number of samples, represents the sample index, represents the th attention weight at the th time step, represents the feature vector at the th time step, represents the bias term used to adjust the output position; The output value coordinates provide a specific three - dimensional coordinate for identifying the position of the high - risk area. Due to the use of the attention mechanism, the model can focus on those time periods and features that are most critical for prediction, thus improving the positioning accuracy; The specific formula for calculating the expected temperature change value is: ; where represents the expected temperature change value, represents the number of samples, represents the sample index, represents the temperature value of the th sample at time point , represents the temperature value of the th sample at time point , represents the time - point index; The output value provides a standardized temperature change index to help evaluate the trend of temperature change. If the expected temperature change value is close to 1, it indicates a general increase in temperature; if it is close to -1, it indicates a general decrease in temperature; if it is close to 0, it means there is no significant temperature change trend.

[0033] Through the above operations, not only can the location of high-risk areas be accurately identified, but also the future temperature change situation can be predicted. This information provides key data support for the control center, thereby effectively managing the temperature of energy storage devices, extending the device life and improving safety.

[0034] S6. Based on the prediction results of the high-risk area location and the expected temperature change value, the control center automatically adjusts the working states of the micropump and the microvalve, changes the flow path and flow rate of the cooling liquid, and realizes local cooling.

[0035] Specifically, it includes the following steps. Based on the obtained prediction results of the high-risk area location, automatically identify the positions of the battery modules that need to be cooled preferentially, and combine the expected temperature change trend to determine which areas are most likely to have overheating problems, and then immediately take measures for cooling operations. To adjust the flow rate of the coolant flowing to the high-risk area, the control center adjusts the output power of the micropump, and uses pulse width modulation technology and PID control algorithm to accurately control the flow rate of the coolant. Pulse width modulation adjusts the speed of the micropump by changing the duty cycle of the signal, thereby flexibly controlling the flow rate of the coolant, while the PID control algorithm dynamically adjusts the working parameters of the micropump according to the real-time feedback data, such as the difference between the current temperature and the set temperature, to ensure that the coolant flow rate is always maintained at the optimal level. This method not only improves the cooling efficiency but also effectively avoids energy waste caused by overcooling. At the same time, by controlling the opening degree of the microvalve, change the flow path of the coolant so that it accurately covers the positions of the battery modules that need to be cooled. The opening degree of the microvalve is controlled by a stepper motor, which has high action precision and fast response speed and can quickly adapt to changes. In addition, a closed-loop feedback mechanism is used to monitor the actual flow direction and temperature distribution of the coolant to ensure that the coolant can accurately reach the target area and achieve the expected cooling effect. This precise control method ensures the efficient operation and safety of the energy storage device even under complex and changeable working conditions. Through the above steps, the fully immersed liquid-cooled energy storage control device can intelligently adjust the working states of the micropump and the microvalve according to the real-time prediction results, optimize the flow path and flow rate of the cooling liquid, thereby effectively managing the temperature of the energy storage device and providing a solid guarantee for long-term stable operation.

[0036] This embodiment also provides a fully immersed liquid-cooled energy storage control system, including: a calibration module, a data acquisition module, a preprocessing module, a model training module, a result prediction module, and an adjustment module; The calibration module is used to install multiple sensors in the liquid immersion tank and perform calibration; The data acquisition module is used to use the calibrated multiple sensors to collect the status data of the battery module and the cooling liquid in real time; The preprocessing module is used to preprocess the status data of the battery module and the cooling liquid and upload it to the control center; The model training module is used to build a Transformer model in the control center and use the historical status data of the battery module and the cooling liquid for training to obtain a trained Transformer model; The result prediction module is used to input the preprocessed status data of the battery module and the cooling liquid into the trained Transformer model to obtain the prediction results of the high-risk area location and the expected temperature change value; The adjustment module is used to input the preprocessed status data of the battery module and the cooling liquid into the trained Transformer model to obtain the prediction results of the high-risk area location and the expected temperature change value.

[0037] This embodiment also provides a computer device applicable to the case of the fully immersed liquid-cooled energy storage control method, including: a memory and a processor; the memory is used to store computer-executable instructions, and the processor is used to execute the computer-executable instructions to implement the fully immersed liquid-cooled energy storage control method proposed in the above embodiment.

[0038] The computer device can be a terminal. The computer device includes a processor, a memory, a communication interface, a display screen, and an input device connected through a system bus. Among them, the processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The communication interface of the computer device is used to communicate with an external terminal in a wired or wireless manner. The wireless manner can be achieved through WIFI, a carrier network, NFC (Near Field Communication), or other technologies. The display screen of the computer device can be a liquid crystal display screen or an electronic ink display screen. The input device of the computer device can be a touch layer covering the display screen, or a button, a trackball, or a touchpad provided on the housing of the computer device, or an external keyboard, a touchpad, or a mouse, etc.

[0039] This embodiment also provides a storage medium, on which a computer program is stored. When the program is executed by a processor, it implements the full-immersion liquid-cooled energy storage control method proposed in the above embodiment. The storage medium can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (SRAM for short), electrically erasable programmable read-only memory (EEPROM for short), erasable programmable read-only memory (EPROM for short), programmable read-only memory (PROM for short), read-only memory (ROM for short), magnetic memory, flash memory, magnetic disk or optical disc.

[0040] In summary, the present invention: proposes an intelligent control strategy based on the Transformer model, predicts future temperature changes according to real-time monitoring data, and automatically adjusts the coolant flow path and flow rate, significantly improving the dynamic response ability and self-adaptability, ensuring accurate temperature control even under complex working conditions, and improving the safety and service life of the battery. Utilizing a multi-sensor network, it realizes comprehensive monitoring of the temperature distribution inside the entire battery module, and improves the data quality through data preprocessing technology, not only providing richer status information, but also enhancing the device's ability to identify potential high-temperature risk areas, thereby effectively preventing failures from occurring.

[0041] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not restrictive. Although the present invention has been described in detail with reference to the preferred embodiments, those of ordinary skill in the art should understand that the technical solutions of the present invention can be modified or equivalently replaced without departing from the spirit and scope of the technical solutions of the present invention, and they should all be covered by the scope of the claims of the present invention.

Claims

1. A full-immersion liquid-cooled energy storage control method, characterized in that: including installing multiple sensors in the liquid immersion tank and calibrating them; using the calibrated multiple sensors to collect the status data of the battery module and the cooling liquid in real time; preprocessing the status data of the battery module and the cooling liquid and uploading it to the control center; building a Transformer model in the control center and training it using the historical status data of the battery module and the cooling liquid to obtain a trained Transformer model; inputting the preprocessed status data of the battery module and the cooling liquid into the trained Transformer model to obtain the prediction results of the high-risk area location and the expected temperature change value; based on the prediction results of the high-risk area location and the expected temperature change value, the control center automatically adjusts the working states of the micropump and the microvalve to change the flow path and flow rate of the cooling liquid to achieve local cooling.

2. The fully immersed liquid-cooled energy storage control method according to claim 1, characterized in that: The installing multiple sensors in the liquid immersion tank and calibrating them specifically includes the following steps placing the temperature sensor in a thermostat, measuring the reading, comparing it with the thermostat temperature, and calibrating the temperature sensor using linear regression; placing the current sensor under a standard current source with a known current, measuring the reading, comparing it with the known current value, and calibrating the current sensor using the least squares method; placing the flow rate sensor under a standard fluid with a known flow rate, measuring the reading, comparing it with the known flow rate value, and calibrating the flow rate sensor using linear interpolation; The standard current source is defined based on a known current value, and the standard fluid is defined based on a known flow rate value and constant temperature and pressure conditions.

3. The fully immersed liquid cooling energy storage control method according to claim 2, characterized in that: The using the calibrated multiple sensors to collect the status data of the battery module and the cooling liquid in real time specifically includes the following steps using the calibrated temperature sensor and the method of arranging embedded temperature monitoring points and timed sampling to collect the temperature data of the battery module and the cooling liquid; using the calibrated current sensor and the method of combining coulomb counting with temperature compensation to collect the charging status data of the battery module; installing the calibrated flow rate sensor on the coolant flow path to collect the flow rate data of the cooling liquid.

4. The fully-immersed liquid-cooled energy storage control method according to claim 3, wherein: The preprocessing includes outlier processing, normalization processing and difference processing; converting the format of the preprocessed status data, encrypting it using TLS, establishing a secure connection with the control center through the HTTP protocol, and uploading the status data.

5. The fully-immersed liquid-cooled energy storage control method according to claim 4, characterized in that: The building a Transformer model in the control center and training it using the historical status data of the battery module and the cooling liquid to obtain a trained Transformer model specifically includes the following steps selecting the TensorFlow deep learning framework in the control center and building a Transformer model using the multi-head self-attention mechanism, feed-forward neural network, residual connection and layer normalization; inputting the historical status data of the battery module and the cooling liquid into the Transformer model for training; using batch gradient descent for iterative training during the training process and using L2 regularization and early stopping to prevent overfitting; The training process includes forward propagation to calculate prediction values and losses, backward propagation to calculate gradients, and using the Adam optimizer to update model parameters.

6. The fully submerged liquid cooling energy storage control method according to claim 5, characterized in that: Inputting the preprocessed state data of the battery module and the cooling liquid into the trained Transformer model to obtain the prediction results of the high-risk area location and the expected temperature change value, specifically including the following steps: Converting the preprocessed state data of the battery module and the cooling liquid into a three-dimensional tensor structure; The three-dimensional tensor structure includes the number of samples, the number of time steps, and the number of feature dimensions; Using the multi-head self-attention mechanism for the three-dimensional tensor to calculate the attention weights, perform weighted summation, and obtain the high-risk area location and the expected temperature change value.

7. The fully immersed liquid cooling energy storage control method according to claim 6, characterized in that: Based on the prediction results of the high-risk area location and the expected temperature change value, the control center automatically adjusts the working states of the micropump and the microvalve, changes the flow path and flow rate of the cooling liquid, and realizes local cooling. Specifically, it includes the following steps: Based on the obtained high-risk area location information, automatically identify the positions of the battery modules that need to be cooled preferentially, and combine the change value of the expected temperature to adjust the working states of the micropump and the microvalve; By adjusting the output power of the micropump and using the pulse width modulation and PID control algorithms, control the flow rate of the coolant flowing to the high-risk area location; By controlling the opening degree of the microvalve and using the stepper motor control and closed-loop feedback, change the flow path of the coolant, and cover the coolant to the positions of the battery modules that need to be cooled, thereby realizing local cooling.

8. A fully immersed liquid-cooled energy storage system, based on the fully immersed liquid-cooled energy storage control method according to any one of claims 1 to 7, characterized in that: Including: A calibration module, a data acquisition module, a preprocessing module, a model training module, a result prediction module, and an adjustment module; The calibration module is used to install multiple sensors in the liquid immersion tank and perform calibration; The data acquisition module is used to use the calibrated multiple sensors to collect the state data of the battery module and the cooling liquid in real time; The preprocessing module is used to preprocess the state data of the battery module and the cooling liquid and upload it to the control center; The model training module is used to build a Transformer model in the control center and use the historical state data of the battery module and the cooling liquid for training to obtain a trained Transformer model; The result prediction module is used to input the preprocessed state data of the battery module and the cooling liquid into the trained Transformer model to obtain the prediction results of the high-risk area location and the expected temperature change value; The adjustment module is used to input the preprocessed state data of the battery module and the cooling liquid into the trained Transformer model to obtain the prediction results of the high-risk area location and the expected temperature change value.

9. A computer device, comprising a memory and a processor, the memory storing a computer program, characterized in that: When the processor executes the computer program, it implements the steps of the fully immersed liquid cooling energy storage control method according to any one of claims 1 to 7.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by the processor, it implements the steps of the fully immersed liquid cooling energy storage control method according to any one of claims 1 to 7.

Images