Method and device for accurately controlling flow of cooling liquid
By acquiring and processing temperature, pressure, and flow velocity signals, and utilizing support vector machines and convolutional neural networks, precise regulation of coolant flow rate was achieved, solving the problem of insufficient accuracy in flow control under complex operating conditions and improving the system's thermal balance and safety.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- SHENZHEN HAIJIE PRECISION MOULD & PLASTIC CO LTD
- Filing Date
- 2025-11-24
- Publication Date
- 2026-04-17
AI Technical Summary
Existing coolant flow control methods struggle to monitor coolant flow patterns in real time and dynamically adjust flow rates based on changes in conditions under complex operating conditions. This results in insufficient flow control accuracy, impacting system thermal balance and safety.
By acquiring temperature data, pressure data, and flow velocity signals, noise reduction and anomaly correction are performed. Support vector machines are used to classify the flow pattern of the coolant. Combined with the judgment of supercooling degree and pressure fluctuation data, the final control parameters are generated to achieve precise regulation of coolant flow.
It enables precise control of coolant flow under complex operating conditions, improves the system's thermal balance control capability, avoids system failures caused by insufficient or excessive cooling, and enhances the safety and stability of equipment operation.
Smart Images
Figure CN121879435A_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of flow control technology, and in particular to a method and apparatus for precise control of coolant flow. Background Technology
[0002] Currently, in the field of industrial production and equipment operation, the cooling system is an important component of advanced process control. Its efficiency and stability affect the performance and safety of the entire system. Especially in high temperature and high pressure environments, the coolant, as the medium for heat conversion, has its flow control directly determining the efficiency of heat transfer and whether the system can avoid failures caused by overheating or overcooling.
[0003] In existing technologies, coolant flow control methods often rely on simple feedback mechanisms based on fixed thresholds or classic PID control algorithms. These methods primarily focus on flow setting under a single operating condition, lacking real-time insight and adaptive adjustment to the dynamic changes in the coolant's own state. When the system faces complex operating environments, such as when parameters like temperature and pressure fluctuate drastically, existing control strategies struggle to adapt to these complex conditions, exhibiting control lag or inaccuracies, leading to system thermal imbalance.
[0004] In summary, existing technologies are insufficient to monitor the flow pattern of coolant in real time under complex operating conditions and dynamically adjust the flow rate according to changes in state, resulting in insufficient precision in coolant flow control. Summary of the Invention
[0005] This invention provides a method and apparatus for precise control of coolant flow rate to enable real-time monitoring of coolant flow pattern under complex operating conditions and dynamic adjustment of flow rate according to state changes.
[0006] In a first aspect, to solve the above-mentioned technical problems, the present invention provides a method for precise control of coolant flow rate, comprising: Temperature data, pressure data, and flow velocity signals are acquired and subjected to noise reduction and anomaly correction to obtain the initial dataset; Based on the initial dataset, the flow patterns of the coolant are classified using a support vector machine to determine the flow pattern type; If the flow pattern is boiling, then the temperature difference value is extracted from the initial dataset, and the supercooling degree value is obtained by combining the temperature difference value with the preset supercooling degree judgment rule. The temperature distribution characteristics and fluid velocity parameters are obtained and integrated with the supercooling value and the flow pattern type into an initial feature set. The flow pattern parameters are obtained and corrected according to the initial feature set. The predicted value of heat transfer efficiency is calculated based on the corrected parameters and compared with the actual value to obtain the transfer deviation value. A deviation identifier is generated based on the transmitted deviation value, environmental parameters are obtained, and flow adjustment requirements are analyzed based on the deviation identifier and the environmental parameters. Initial control parameters are then generated based on the requirements. Acquire pressure fluctuation data, and adjust the initial control parameters based on the pressure fluctuation data to obtain the final control parameters; The coolant flow rate is adjusted based on the final control parameters to obtain the coolant volume adjustment result.
[0007] In one optional implementation, the acquisition of temperature data, pressure data, and flow velocity signals, followed by noise reduction and anomaly correction to obtain an initial dataset, includes: Temperature data, pressure data, and flow velocity signals are collected and noise reduction is performed to obtain the first parameter set; The first parameter set is smoothed to obtain the second parameter set; The correlation between the temperature data and the pressure data in the second parameter set is analyzed, and abnormal correlations are corrected to obtain the initial dataset.
[0008] In one optional implementation, the step of classifying the flow patterns of the coolant using a support vector machine based on the initial dataset to determine the flow pattern type includes: The correlation between flow rate changes and temperature fluctuations in the initial dataset was analyzed, and abnormal correlations were corrected to obtain a correlated dataset. Obtain a first correlation index between coolant properties and flow patterns, and correct the associated dataset based on the first correlation index to obtain an optimized dataset; The optimized dataset is classified using a support vector machine. Based on the classification results and the preset rules for bubble generation and boiling state, the flow pattern type is determined.
[0009] In one optional implementation, the step of extracting temperature difference values from the initial dataset and determining the degree of supercooling based on these temperature difference values and a preset supercooling determination rule includes: Identify and mark points in the initial dataset with abnormal temperature differences to obtain the first set of differences; The first set of differences is smoothed to obtain the second set of differences; Based on the second set of differences, a second correlation index between the temperature difference and the boiling state is calculated. The second set of differences is then corrected based on the second correlation index to obtain a third set of differences. The third set of differences is classified and processed, and combined with the preset supercooling degree judgment rules, the supercooling degree value is obtained.
[0010] In one optional implementation, the acquisition of temperature distribution characteristics and fluid velocity parameters, along with the supercooling value and the flow pattern type, is integrated into an initial feature set. Flow mode parameters are obtained and corrected based on the initial feature set. A predicted heat transfer efficiency value is calculated based on the corrected parameters and compared with the actual value to obtain a transfer deviation value, including: The temperature distribution characteristics and fluid velocity parameters are obtained, and the supercooling value, the flow pattern type, the temperature distribution characteristics, and the fluid velocity parameters are integrated to obtain an initial feature set; Data on the influence of ambient temperature and cooling properties are acquired, and the initial feature set is modified based on the data to obtain an optimized feature set. Obtain boundary condition data and heat transfer paths, and correlate them with the optimized feature set to obtain dynamic flow mode parameters. Then, correct the dynamic flow mode parameters to obtain a flow feature set. Based on the set of flow characteristics, a predicted value of heat transfer efficiency is calculated using a convolutional neural network. The difference between the predicted value and the actual value collected in real time is analyzed to obtain the transfer deviation value.
[0011] In one optional implementation, the step involves generating a deviation identifier based on the transmission deviation value, obtaining environmental parameters, analyzing flow adjustment requirements based on the deviation identifier and the environmental parameters, and generating initial control parameters for the requirements. Based on the transmitted deviation value, a deviation identifier is generated; Obtain environmental parameters, associate the environmental parameters with the deviation identifier, analyze the demand type for flow adjustment, and obtain the first control demand; Obtain historical data of flow control, analyze the first degree of matching between the historical data and the first control requirement, determine the flow adjustment instruction based on the first degree of matching, and obtain the first adjustment instruction; Based on the first adjustment instruction, the corresponding amplitude adjustment data is determined to obtain the initial control parameters.
[0012] In one optional implementation, the step of acquiring pressure fluctuation data and adjusting the initial control parameters based on the pressure fluctuation data to obtain the final control parameters includes: Acquire pressure fluctuation data, and generate fluctuation identifiers based on the pressure fluctuation data; The fluctuation identifier and the initial control parameters are correlated to determine the pressure compensation requirement and obtain the second control requirement. Obtain historical pressure fluctuation and flow control matching records, analyze the matching records and the second degree of matching with the second control requirement, obtain the pressure adjustment requirement based on the second degree of matching, and determine the second adjustment instruction; According to the second adjustment instruction, the corresponding pressure adjustment data is obtained and the pressure adjustment data is optimized to obtain the final control parameters.
[0013] In a second aspect, the present invention provides a device for precise control of coolant flow rate, comprising: The data acquisition and processing module is used to acquire temperature data, pressure data, and flow velocity signals, and perform noise reduction and anomaly correction to obtain the initial dataset. The flow pattern classification module is used to classify the flow pattern of the coolant using a support vector machine based on the initial dataset, and to determine the flow pattern type. The supercooling degree analysis module is used to extract the temperature difference value from the initial dataset if the flow mode type is boiling, and to make a judgment based on the temperature difference value and a preset supercooling degree judgment rule to obtain the supercooling degree value. The deviation prediction module is used to acquire temperature distribution characteristics and fluid velocity parameters, and integrate them with the supercooling degree value and the flow pattern type into an initial feature set. The flow mode parameters are acquired and corrected based on the initial feature set. The predicted value of heat transfer efficiency is calculated based on the corrected parameters and compared with the actual value to obtain the transfer deviation value. The control parameter generation module is used to generate a deviation identifier based on the transmission deviation value, obtain environmental parameters, analyze the flow adjustment requirements based on the deviation identifier and the environmental parameters, and generate initial control parameters for the requirements. The control parameter optimization module is used to acquire pressure fluctuation data and adjust the initial control parameters according to the pressure fluctuation data to obtain the final control parameters; The flow control module is used to adjust the coolant flow rate according to the final control parameters to obtain the coolant quantity adjustment result.
[0014] Compared with the prior art, the present invention has the following beneficial effects: (1) This invention obtains an initial dataset by acquiring temperature data, pressure data, and flow velocity signals and performing noise reduction and anomaly correction. This series of operations effectively reduces the interference of environmental noise and abnormal fluctuations on data acquisition, improves the accuracy and reliability of monitoring parameters, and provides a high-quality data foundation for subsequent flow pattern identification and flow control, thereby solving the problem of inaccurate data acquisition under complex working conditions in existing technologies.
[0015] (2) After acquiring the initial dataset, this invention classifies the flow pattern of the coolant using a support vector machine to identify the boiling state, and extracts the temperature difference value from the initial dataset. The supercooling value is obtained by combining the temperature difference value analysis. Through this multi-step, multi-modal analysis mechanism, the dynamic flow characteristics of the coolant can be judged in real time and accurately, improving the system's ability to perceive complex working conditions and providing a key basis for subsequent adaptive control of coolant flow.
[0016] (3) This invention combines the degree of subcooling and the flow pattern type to predict the heat transfer efficiency, calculate the difference between the predicted and actual values to obtain the transfer deviation value, and then integrates environmental parameters and pressure fluctuation data to generate the final control parameters. This process fully utilizes the advantages of advanced control software (APC) conforming to G06F17 in dynamic modeling and multivariate optimization, achieving precise and adaptive adjustment of coolant flow rate and significantly improving the system's thermal balance control capability.
[0017] (4) The present invention adjusts the coolant flow rate according to the final control parameters to obtain the coolant quantity adjustment result, thereby realizing real-time maintenance of the system's thermal state. This flow control method not only responds quickly, but also operates stably under complex environments such as high temperature and high pressure, solving the problem of insufficient accuracy in coolant flow control, effectively avoiding system failures caused by insufficient or excessive cooling, and improving the safety, stability, and energy efficiency of equipment operation. Attached Figure Description
[0018] Figure 1 This is a flowchart illustrating an embodiment of the precise coolant flow control method provided by the present invention; Figure 2 This is a flowchart illustrating an embodiment of the initial control parameter acquisition provided by the present invention.
[0019] Figure 3 This is a schematic diagram of an embodiment of the coolant flow precision control device provided by the present invention. Detailed Implementation
[0020] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, 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.
[0021] Reference Figure 1 The first embodiment of the present invention provides a method for precise control of coolant flow rate, comprising the following steps: S11: Acquire temperature data, pressure data, and flow velocity signals, and perform noise reduction and anomaly correction to obtain the initial dataset; S12, Based on the initial dataset, the flow patterns of the coolant are classified using a support vector machine to determine the flow pattern type; S13, If the flow pattern is boiling, then extract the temperature difference value from the initial dataset, and make a judgment based on the temperature difference value and the preset supercooling degree judgment rule to obtain the supercooling degree value. S14. Obtain temperature distribution characteristics and fluid velocity parameters, and integrate them with the supercooling degree value and the flow pattern type into an initial feature set. Obtain flow mode parameters based on the initial feature set and make corrections. Calculate the predicted value of heat transfer efficiency based on the corrected parameters and compare it with the actual value to obtain the transfer deviation value. S15, generate a deviation identifier based on the transmission deviation value, obtain environmental parameters, analyze the flow adjustment requirements based on the deviation identifier and the environmental parameters, and generate initial control parameters for the requirements; S16, acquire pressure fluctuation data, adjust the initial control parameters according to the pressure fluctuation data, and obtain the final control parameters; S17, adjust the coolant flow rate according to the final control parameters to obtain the coolant quantity adjustment result.
[0022] In step S11, temperature data, pressure data, and flow velocity signals are acquired and subjected to noise reduction and anomaly correction to obtain an initial dataset, including: Temperature data, pressure data, and flow velocity signals are collected and noise reduction is performed to obtain the first parameter set; The first parameter set is smoothed to obtain the second parameter set; The correlation between the temperature data and the pressure data in the second parameter set is analyzed, and abnormal correlations are corrected to obtain the initial dataset.
[0023] First, temperature data, pressure data, and flow velocity signals are collected and noise-reduced to obtain the first parameter set. It should be noted that the temperature data is acquired by a temperature sensor; the pressure data is detected by a pressure sensor; and the flow velocity signal can be detected in real time by a flow meter. The collected data are integrated and noise-reduced using a low-pass filter to obtain the first parameter set. The cutoff frequency of the low-pass filter is set based on the sampling frequency and the Nyquist sampling theorem, typically set to 0.2 to 0.25 times the sampling frequency. For example, when the sampling frequency is 10Hz, the cutoff frequency can be set to 2Hz to 2.5Hz.
[0024] Subsequently, the first parameter set is smoothed using the sliding window averaging method to obtain the second parameter set. It should be noted that the sliding window averaging method is used to smooth all data in the first parameter set. For example, the window size is set to 5. For the data in the first parameter set, the two data points before and after it are taken, plus the data itself, for a total of 5 data points. The average value is calculated to replace the current data, thus completing the smoothing process.
[0025] Finally, the correlation between the temperature data and the pressure data in the second parameter set is analyzed, and the first Pearson correlation coefficient between temperature and pressure is calculated. If the calculated first Pearson correlation coefficient is lower than the preset first correlation coefficient threshold, it is marked as an abnormal data point. The temperature data and pressure data are smoothed and corrected by the same sliding window averaging method as in the previous step. The corrected data and the original data are integrated to obtain the initial dataset.
[0026] It should be noted that the preset first correlation coefficient threshold is determined by analyzing a large amount of historical data and based on the lower quantile of the temperature-pressure correlation coefficient distribution. For example, if a cooling system analyzes 1000 sets of historical data and the 5th quantile of its temperature-pressure correlation coefficient is 0.62, then the preset threshold is set to 0.6.
[0027] In step S12, based on the initial dataset, the flow patterns of the coolant are classified using a support vector machine to determine the flow pattern type, including: The correlation between flow rate changes and temperature fluctuations in the initial dataset was analyzed, and abnormal correlations were corrected to obtain a correlated dataset. Obtain a first correlation index between coolant properties and flow patterns, and correct the associated dataset based on the first correlation index to obtain an optimized dataset; The optimized dataset is classified using a support vector machine. Based on the classification results and the preset rules for bubble generation and boiling state, the flow pattern type is determined.
[0028] First, the Pearson correlation coefficient between flow rate and temperature is calculated. If the calculated Pearson correlation coefficient is lower than the preset second correlation coefficient threshold (the threshold is determined in a similar way to the preset first correlation coefficient threshold in step S11, and will not be repeated here), it is marked as an abnormal data point. The flow rate and temperature data of the abnormal data point are corrected by a sliding window averaging method similar to that in step S11 to obtain the associated dataset.
[0029] Subsequently, the first correlation index between coolant properties and flow patterns is obtained. It should be noted that a coolant property parameter vector is first established. ,in For density, For dynamic viscosity, Thermal conductivity, To determine the specific heat capacity, a multiple linear regression was then used, with coolant characteristic parameters as independent variables and flow pattern-related features as dependent variables. The regression coefficients were obtained using the least squares method. The standardized absolute value of the regression coefficients is used as the first correlation index, where the Reynolds number, a relevant characteristic of the flow pattern, is obtained experimentally. Based on the first correlation index, the associated dataset is corrected to obtain an optimized dataset. The correction formula is as follows: ,in, For the corrected data, For the data in the associated dataset, The regression coefficient is denoted as .
[0030] Finally, the optimized dataset is classified using a support vector machine (SVM). Based on the classification results and pre-defined rules for bubble generation and boiling states, the flow pattern type is determined. Specifically, the SVM model uses a radial basis function as the kernel function, with a penalty parameter C of 2 and a kernel parameter γ of 0.2. This model automatically classifies the data into two categories: "bubble generation state" and "boiling state." Based on the classification results and pre-defined rules for bubble generation and boiling states, the flow pattern type is further determined. It should be noted that the pre-defined rules are as follows: when... and At that time, the flow pattern was determined to be boiling, in which... The bubble formation rate is calculated by counting the number of bubbles generated per unit area per unit time using pressure sensor signals. The boiling temperature difference is obtained by measuring the difference between the temperature of the heating surface and the temperature of the coolant during the boiling state. To determine the bubble formation rate threshold, the number of bubbles generated per unit area was measured using a high-frequency pressure sensor, and the minimum bubble formation rate at which boiling occurred was used as the threshold. To determine the boiling temperature difference threshold, we analyze the distribution of temperature differences corresponding to boiling states in historical data and select a specific quantile (such as the 95th quantile) as the threshold.
[0031] It should be noted that the pre-training process of the support vector machine model includes: simultaneously collecting the optimized dataset obtained from the aforementioned steps (S11 and the first half of S12) as input features under various typical cooling systems and operating conditions; simultaneously, accurately identifying and labeling the flow morphology (bubble generation state and boiling state) of the coolant using high-frequency pressure sensors combined with pressure pulsation analysis and other observation methods, obtaining the corresponding flow morphology type as the target label for model training, thereby completing the construction of the training dataset; the dataset needs to cover different coolant types, operating temperature and pressure ranges, flow rate ranges, and system load conditions to ensure the model's generalization ability. Subsequently, the dataset is randomly divided into training and test sets in a 7:3 ratio, with an optimal range given for the hyperparameters of the support vector machine, classification accuracy as the evaluation index, and model performance evaluated through 5-fold cross-validation; grid search is performed to determine the optimal hyperparameter combination, finally completing the training and validation of the model.
[0032] In step S13, if the flow pattern is boiling, the temperature difference value is extracted from the initial dataset. Based on the temperature difference value and a preset supercooling determination rule, a supercooling value is obtained. Specifically, temperature difference values are extracted from the initial dataset, and a supercooling degree value is obtained by judging the supercooling degree based on the temperature difference values and a preset supercooling degree judgment rule, including: Identify and mark points in the initial dataset with abnormal temperature differences to obtain the first set of differences; The first set of differences is smoothed to obtain the second set of differences; Based on the second set of differences, a second correlation index between the temperature difference and the boiling state is calculated. The second set of differences is then corrected based on the second correlation index to obtain a third set of differences. The third set of differences is classified and processed, and combined with the preset supercooling degree judgment rules, the supercooling degree value is obtained.
[0033] First, if the flow morphology type judgment result in step S12 is boiling state, then temperature data is extracted from the initial dataset and the temperature difference is calculated to obtain a temperature difference sequence. The temperature difference sequence is converted into a normal distribution through Box-Cox and the mean and standard deviation are calculated. Then, outliers are identified and marked based on the 3σ rule to obtain the first difference set.
[0034] Subsequently, the first difference set is processed using a moving average method similar to that in step S11 to obtain the second difference set.
[0035] Next, based on the second set of differences, a second correlation index between the temperature difference and the boiling state is calculated. It should be noted that the boiling state is quantified by the bubble formation rate. The bubble formation rate is calculated by statistically analyzing the number of bubbles generated per unit area per unit time using pressure sensor signals. Then, the Pearson correlation coefficient between the temperature difference and the bubble formation rate is calculated to obtain the second correlation index. Subsequently, the formula is used... The second set of differences is then modified to obtain the third set of differences. For the corrected data, The preset threshold was determined by calculating the correlation distribution from multiple sets of boiling experimental data. The minimum acceptable correlation that could stably reflect the bubble formation pattern was selected as the threshold. The second correlation index obtained is calculated. The empirical adjustment coefficient, obtained through cross-validation based on a large amount of historical data, has an optimal value range of [0.1, 0.3]. For example, suppose the difference at a certain moment... Preset threshold =0.8, the second correlation index =0.7, empirical coefficient If we take 0.2, then .
[0036] Finally, the third set of differences is categorized and, combined with a preset rule for judging supercooling, the supercooling degree is obtained. It should be noted that this step requires calculating the average temperature difference of the third set of differences. and standard deviation Based on the correlation between temperature difference and bubble formation characteristics in the experiment, the critical threshold for distinguishing between mild, moderate, and severe supercooling was determined by statistically analyzing the temperature difference distribution under different operating conditions. The supercooling judgment rule was then set as follows: ,in and To preset the threshold, probability density curves are plotted by statistically analyzing the temperature difference distribution range under different coolant flow conditions, and the threshold is determined based on the critical points of mild, moderate, and severe overcooling. For example, under a certain coolant flow condition, the temperature difference distribution was obtained through 50 sets of experiments. Mild overcooling was mainly concentrated in the range of 3–6℃, moderate overcooling in the range of 6–9℃, and severe overcooling greater than 9℃. Therefore, the following threshold can be set: , The final output supercooling level is labeled as slightly supercooled = 1, moderately supercooled = 2, and severely supercooled = 3. The third difference set is then classified according to the supercooling level judgment rule to obtain the supercooling level value.
[0037] In step S14, temperature distribution characteristics and fluid velocity parameters are acquired and integrated with the supercooling value and flow pattern type into an initial feature set. Flow mode parameters are obtained and corrected based on the initial feature set. The predicted heat transfer efficiency is calculated based on the corrected parameters and compared with the actual value to obtain the transfer deviation value, including: The temperature distribution characteristics and fluid velocity parameters are obtained, and the supercooling value, the flow pattern type, the temperature distribution characteristics, and the fluid velocity parameters are integrated to obtain an initial feature set; Data on the influence of ambient temperature and cooling properties are acquired, and the initial feature set is modified based on the data to obtain an optimized feature set. Obtain boundary condition data and heat transfer paths, and correlate them with the optimized feature set to obtain dynamic flow mode parameters. Then, correct the dynamic flow mode parameters to obtain a flow feature set. Based on the set of flow characteristics, a predicted value of heat transfer efficiency is calculated using a convolutional neural network. The difference between the predicted value and the actual value collected in real time is analyzed to obtain the transfer deviation value.
[0038] First, obtain the temperature distribution characteristics and fluid velocity parameters, and collect the temperature along the fluid channel using a temperature sensor. Then through the formula The average temperature gradient is extracted as the temperature distribution feature. For spatial coordinates, Temperature along the fluid passage The calculated temperature distribution characteristics are used to obtain the fluid velocity field distribution through the flow velocity signal. The average flow velocity and turbulence intensity are calculated as the fluid velocity parameters. The temperature distribution characteristics and the fluid velocity parameters are integrated with the supercooling value and the flow pattern type to obtain an initial feature set. , }.in, The supercooling value is... The flow pattern type, The temperature distribution characteristics, The average flow velocity is... The turbulence intensity is given.
[0039] Subsequently, ambient temperature was collected using a temperature sensor, and coolant property data, including viscosity, thermal conductivity, specific heat capacity, and other physical properties, were experimentally determined. Then, the data was analyzed using formulas... , The initial feature set is modified, wherein, The supercooling value is... The flow pattern type, The corrected temperature distribution characteristics, The corrected average flow velocity, For turbulence intensity, For ambient temperature, The reference ambient temperature is used to define the ideal operating environment based on the coolant design. This refers to the viscosity of the coolant. and These are the temperature correction factor and the viscosity correction factor, determined through regression analysis and error minimization fitting of coolant characteristic data under different ambient temperatures. The preferred value range is [0.01, 0.1], resulting in the optimized feature set. .
[0040] Next, boundary condition data and heat transfer paths are acquired and correlated with the optimized feature set to obtain dynamic flow mode parameters. These parameters are then corrected to obtain the flow feature set. It should be noted that the boundary condition data includes the thermal conductivity of the pipe material. Wall thickness and dirt coefficient This can be found in industry standard manuals, through formulas. Calculate the overall heat transfer coefficient, where h is the convective heat transfer coefficient. The thermal conductivity of the pipe material is given. The wall thickness is [value missing]. The fouling coefficient is defined. A heat transfer path feature vector is established. ,in To transmit the path length, For the cross-sectional area of circulation, and The inlet and outlet heat flow rates are measured by a flow meter. This is obtained through the formula... Calculate the dynamic flow model parameters, where, The optimized feature set is obtained by principal component analysis, which decomposes the feature matrix formed by a large amount of historical data, calculates the contribution rate of each principal component, and selects the top few principal components with a cumulative contribution rate of 85%-95%. The weight matrix is obtained by constructing a sample matrix from a large amount of historical data and performing principal component analysis on the sample matrix to extract the principal component loading matrix. This is the transpose of the feature matrix formed by the feature vectors of the heat transfer path. (Using the formula...) The dynamic flow mode parameters are corrected to obtain a set of flow characteristics, wherein, The time decay coefficient is determined by applying a step disturbance to the system and recording its output response curve. The coefficient is determined by fitting a response model of a first- or second-order system, with a preferred value range of [0.9, 1.0]. The system time constant is obtained from the step response curve. As a reference time point, it is usually set to the start time of the current calculation window. These are the parameters of the dynamic flow mode.
[0041] Finally, based on the set of flow characteristics, a convolutional neural network is used to calculate the predicted heat transfer efficiency. The difference between the predicted heat transfer efficiency and the actual value collected in real time is analyzed to obtain the transfer deviation value. It should be noted that the predicted heat transfer efficiency is calculated using a convolutional neural network by measuring the heat changes at the inlet and outlet of the coolant. The actual heat transfer efficiency is calculated by the ratio of the inlet / outlet heat difference to the inlet heat. The difference between the predicted and actual values is then calculated to obtain the transfer deviation value.
[0042] It should be noted that the establishment and pre-training process of the convolutional neural network includes: constructing a training dataset based on historical operating data. This dataset contains a set of flow features obtained from the aforementioned steps under various operating conditions as input features, and the actual heat transfer efficiency calculated by a high-precision heat flow sensor as a label; the network structure adopts a one-dimensional convolutional neural network, including an input layer, two convolutional layers (using 32 and 64 3×1 convolutional kernels respectively, with ReLU activation function), corresponding max pooling layers, flattening layers, and three fully connected layers (with 128, 64, and 1 neurons respectively), and the final output layer uses a linear activation function to predict the efficiency value; during training, the dataset is divided into training and test sets in a 7:3 ratio, the mean squared error is used as the loss function, the Adam optimizer (initial learning rate 0.001) is used to update the model parameters, and early stopping is used to prevent overfitting; the trained network model is used to predict heat transfer efficiency in real time. For example, during a certain operation, the system obtains the coolant inlet temperature as 92.4℃, the outlet temperature as 79.1℃, and the corresponding flow rate as 105L / min in real time. The system inputs the flow feature set obtained from the preceding steps into a pre-trained convolutional neural network model. After two layers of 1D convolution and max pooling, the network extracts local patterns, which are then flattened and mapped through multiple fully connected layers to obtain the predicted output. Finally, the predicted heat transfer efficiency under the current operating conditions is calculated to be 68.7%.
[0043] like Figure 2 As shown, in step S15, a deviation identifier is generated based on the transmission deviation value, environmental parameters are obtained, flow adjustment requirements are analyzed based on the deviation identifier and the environmental parameters, and initial control parameters are generated for the requirements, including: S151, Generate a deviation identifier based on the transmitted deviation value; S152, acquire environmental parameters, and perform correlation processing on the environmental parameters and the deviation identifier to analyze the demand type of flow adjustment and obtain the first control demand; S153 acquires historical data of flow control, analyzes the first degree of matching between the historical data and the first control requirement, determines the flow adjustment instruction based on the first degree of matching, and obtains the first adjustment instruction; S154, Based on the first adjustment instruction, determine the corresponding amplitude adjustment data and obtain the initial control parameters.
[0044] First, based on the transmission deviation value, using the formula... Calculate the relative deviation rate, where The calculated relative deviation rate, , The heat transfer efficiency deviation value and the actual value obtained in the previous steps are respectively used. Based on the numerical range of the relative deviation rate, multi-level deviation classification intervals are set to obtain corresponding deviation labels. The threshold settings of the classification intervals can be adjusted according to the response characteristics and accuracy requirements of different cooling systems. For example, when When the deviation is P1+ (slightly too high); when When the deviation is P1- (slightly too low); when When the deviation is P2+ (moderately high); when When the deviation is P2+ (moderately low); when When the deviation is P3+ (weight too high); when When the deviation is too low, it is marked as P3- (weight too low).
[0045] Subsequently, ambient temperature was collected as an environmental parameter using a temperature sensor, and then calculated using the formula... Calculate the correlation between the environmental parameters and the transmission deviation value, where, To calculate the correlation degree, For the environmental parameters, The transmission deviation value is used. Based on the correlation and the deviation identifier, the flow adjustment demand type is determined, resulting in a first control demand. For example, during a monitoring process, the system detects the transmission deviation rate. =+6.3%, corresponding to a deviation marked P2+ (moderately high), indicating that the predicted heat transfer efficiency is significantly higher than the actual value, suggesting insufficient heat dissipation. Simultaneously, the collected ambient temperature... =36.5℃, and the correlation coefficient between ambient temperature and transmission deviation was obtained through correlation analysis. =0.82, indicating a strong positive correlation between the two, meaning that an increase in ambient temperature leads to a decrease in heat transfer efficiency. Based on this result, the system combines the deviation label type P2+ with the strong positive correlation. The deviation is >0.8, indicating that the current deviation is mainly caused by the increase in ambient temperature, which is a passive cooling insufficiency type deviation. Therefore, the first control requirement of the system output is to increase the coolant flow rate by approximately 8% to enhance heat dissipation capacity.
[0046] Next, historical data of flow control is acquired, including flow setpoint, response time, and corresponding deviation changes. This historical data is Z-score standardized, and the matching degree between the first control requirement and the historical data is calculated using Euclidean distance. The historical control instruction with the highest matching degree is selected as the first matching degree. The first control requirement is then corrected based on this first matching degree to obtain the first adjustment instruction. For example, when a system generates a first control requirement of "increase coolant flow by approximately 8%" under current operating conditions, it calls the historical flow control database for matching analysis. It finds that under similar past operating conditions (ambient temperature 35–37°C, deviation indicator P2+), the optimal response of "increasing flow by 10%" was achieved, reducing the deviation to 2% within 30 seconds. After normalization, the feature vector currently collected by the system is denoted as X=[0.82,0.40,0.33], and the feature vector of the historical data is T=[0.90,0.35,0.30]. The difference between the two is calculated using Euclidean distance to obtain the distance value. The distance is then mapped to a matching degree according to a predetermined normalization rule. The maximum distance in the historical data is 1.4, so the matching degree is calculated as follows: The system determines the first adjustment command to increase the coolant flow rate by approximately 9.3% based on a 1% adjustment amount corresponding to a 0.1 increase in the matching degree, with the adjustment duration set to 25 seconds.
[0047] Finally, based on the first adjustment instruction, the corresponding amplitude adjustment data is determined to obtain the initial control parameters. It should be noted that this is achieved through the formula... Determine the amplitude adjustment data, wherein For the calculated amplitude adjustment data, To adjust the gain coefficient, a linear regression is performed on historical deviation data and its corresponding transmission deviation, and the optimal estimate is obtained through cross-validation as the adjusted gain coefficient. The amplitude adjustment data calculated based on the transmission deviation value is used as the initial control parameter.
[0048] In step S16, pressure fluctuation data is acquired, and the initial control parameters are adjusted based on the pressure fluctuation data to obtain the final control parameters, including: Acquire pressure fluctuation data, and generate fluctuation identifiers based on the pressure fluctuation data; The fluctuation identifier and the initial control parameters are correlated to determine the pressure compensation requirement and obtain the second control requirement. Obtain historical pressure fluctuation and flow control matching records, analyze the matching records and the second degree of matching with the second control requirement, obtain the pressure adjustment requirement based on the second degree of matching, and determine the second adjustment instruction; According to the second adjustment instruction, the corresponding pressure adjustment data is obtained and the pressure adjustment data is optimized to obtain the final control parameters.
[0049] First, pressure fluctuation data is collected using a pressure sensor, with the sampling frequency set to 1-5kHz. Then, the data is processed using the formula... The pressure fluctuation time series was obtained, where The average pressure over a certain period of time. This is the collected real-time pressure sequence. The calculated pressure fluctuation time series is used. A fluctuation identifier is determined by comparing the pressure fluctuation time series with the pressure fluctuation threshold range. It should be noted that the pressure fluctuation threshold range is obtained by analyzing historical pressure data of the system under stable operating conditions, statistically analyzing the mean and standard deviation of its fluctuation amplitude, and combining this with the equipment's allowable pressure upper limit and safety margin to determine the acceptable relative pressure fluctuation percentage under normal operating conditions (e.g., ±5% to ±8%). For example, the threshold range and corresponding fluctuation identifier are set as follows: when... When the fluctuation is W0, it indicates that the fluctuation is within the normal range; when When the fluctuation is W1, it indicates a slight fluctuation; when When the volatility is W2, it indicates moderate volatility; when At this time, the fluctuation is identified as W3, indicating severe fluctuation. The threshold range can be adjusted according to actual operating conditions to adapt to changes in the dynamic characteristics of the fluid.
[0050] Subsequently, the fluctuation identifier and the initial control parameters are correlated to determine the pressure compensation requirement, thus obtaining the second control requirement. It should be noted that the correspondence between the fluctuation identifier and the pressure compensation direction is as follows: if the fluctuation identifier is W1 or W2, it indicates the existence of low-to-medium amplitude periodic fluctuations, in which case the control output is reduced to decrease system stress; if the fluctuation identifier is W3, it indicates that the system experiences high-amplitude, persistent pressure instability, in which case a pressure buffer is temporarily added or the flow response time is adjusted; if the fluctuation identifier is W0, the initial control parameters remain unchanged. A pressure compensation coefficient Kc is established based on the fluctuation indicator. When the fluctuation indicator is W0, Kc=1; when the fluctuation indicator is W1, Kc=0.85; when the fluctuation indicator is W2, Kc=0.7; and when the fluctuation indicator is W3, Kc=0.55. The specific value of the pressure compensation coefficient Kc is determined by conducting step response tests and stability analyses on the cooling system. Different control quantities are applied when the system is at different fluctuation levels, and the system's overshoot, settling time, and other indicators are observed. Finally, the compensation coefficient that enables the system to quickly and smoothly reach the target state is selected as the pressure compensation coefficient. Then, the initial control parameters are multiplied by the pressure compensation coefficient to obtain the second control requirement. For example, if the initial control parameter requires "increasing the flow rate by 9%", and the fluctuation indicator is W2, then Kc=0.7, and the second control requirement is "increasing the flow rate by 6.3%".
[0051] Next, historical pressure fluctuation and flow control matching records are obtained, including historical fluctuation identifiers, flow control commands, and corresponding pressure stabilization effect data. The second matching degree between the second control requirement and the historical matching records is calculated using the Euclidean distance method, and the historical record with the highest matching degree is selected as a reference. Based on the second matching degree, [the system] selects [the appropriate data]. The minimum historical record is analyzed to determine its pressure adjustment needs, and the current second control need is modified to obtain the second adjustment instruction. For example, the current second control need is "increase flow rate by 6.3%", the fluctuation identifier is W2 (numerical code 2), there is a record in the historical record: fluctuation identifier W2 (numerical code 2), the historical control instruction is "increase flow rate by 5%", and the pressure stabilization effect is good. The normalized distance upper limit is determined to be 20 based on the historical flow rate adjustment range and the fluctuation identifier encoding range. Therefore, the second matching degree... The matching degree is relatively high. Based on this, the system modifies the second control requirement to... The second control requirement was modified to "increase flow rate by 6.2%", and the response time was set to 20 seconds, thus obtaining the second adjustment instruction.
[0052] Finally, based on the second adjustment instruction, the corresponding pressure adjustment data is obtained, and then adjusted using the formula. Calculate the pressure adjustment data, wherein Adjust data for pressure. The pressure regulation gain coefficient is determined by establishing a linear regression model of historical pressure fluctuation data and corresponding flow adjustment values, solving for the coefficient using the least squares method, and selecting the optimal coefficient value as the pressure regulation gain coefficient through cross-validation. This represents the average of the absolute values of the pressure fluctuation time series. This is the flow adjustment value in the second adjustment instruction. The pressure adjustment data is optimized by using a low-pass filter algorithm to smooth fluctuations; the formula is as follows: in For the optimized final control parameters, These are the filter coefficients, determined through cross-validation of experimental data, and are typically taken as 0.2-0.5. This is the control parameter value from the previous moment. For example, the second adjustment command is "increase flow rate by 5.5%", which is calculated as follows: , ,but During optimization, take =0.3, control parameter at the previous moment =5.0%, then The system ultimately outputs the control parameter as "increase flow rate by approximately 3.6%".
[0053] In step S17, the coolant flow rate is adjusted according to the final control parameters to obtain the coolant quantity adjustment result. It should be noted that the system executes the control signal through a flow regulation execution unit (such as a variable frequency pump or electric valve), and corrects the coolant flow rate in real time according to the target flow rate setpoint and adjustment range in the final control parameters. During the adjustment process, the system continuously monitors the temperature, pressure, and flow rate changes in the cooling loop, calculates the real-time deviation using a closed-loop feedback control strategy, and corrects the flow output through a PID control algorithm to make the flow rate stably approach the target value, thereby obtaining the coolant quantity adjustment result.
[0054] In summary, this invention discloses a method for precise control of coolant flow rate, comprising: acquiring temperature data, pressure data, and flow velocity signals, and performing noise reduction and anomaly correction to obtain an initial dataset; classifying the flow pattern of the coolant using a support vector machine based on the initial dataset to determine the flow pattern type; if the flow pattern type is boiling, extracting the temperature difference value from the initial dataset, and judging the supercooling degree value based on the temperature difference value and a preset supercooling degree judgment rule; acquiring temperature distribution features and fluid velocity parameters, and integrating them with the supercooling degree value and the flow pattern type to form an initial feature set; acquiring and correcting flow mode parameters based on the initial feature set; calculating a predicted heat transfer efficiency value based on the corrected parameters and comparing it with the actual value to obtain a transfer deviation value; generating a deviation identifier based on the transfer deviation value; acquiring environmental parameters; analyzing flow rate adjustment requirements based on the deviation identifier and the environmental parameters; generating initial control parameters based on the requirements; acquiring pressure fluctuation data; adjusting the initial control parameters based on the pressure fluctuation data to obtain final control parameters; and adjusting the coolant flow rate based on the final control parameters to obtain a coolant flow rate adjustment result. This invention enables real-time monitoring of coolant flow patterns under complex operating conditions and dynamic adjustment of flow rate based on state changes through flow control.
[0055] Reference Figure 3 The second embodiment of the present invention provides a coolant flow rate precision control device, comprising: The data acquisition and processing module is used to acquire temperature data, pressure data, and flow velocity signals, and perform noise reduction and anomaly correction to obtain the initial dataset. The flow pattern classification module is used to classify the flow pattern of the coolant using a support vector machine based on the initial dataset, and to determine the flow pattern type. The supercooling degree analysis module is used to extract the temperature difference value from the initial dataset if the flow mode type is boiling, and to make a judgment based on the temperature difference value and a preset supercooling degree judgment rule to obtain the supercooling degree value. The deviation prediction module is used to acquire temperature distribution characteristics and fluid velocity parameters, and integrate them with the supercooling degree value and the flow pattern type into an initial feature set. The flow mode parameters are acquired and corrected based on the initial feature set. The predicted value of heat transfer efficiency is calculated based on the corrected parameters and compared with the actual value to obtain the transfer deviation value. The control parameter generation module is used to generate a deviation identifier based on the transmission deviation value, obtain environmental parameters, analyze the flow adjustment requirements based on the deviation identifier and the environmental parameters, and generate initial control parameters for the requirements. The control parameter optimization module is used to acquire pressure fluctuation data and adjust the initial control parameters according to the pressure fluctuation data to obtain the final control parameters; The flow control module is used to adjust the coolant flow rate according to the final control parameters to obtain the coolant quantity adjustment result.
[0056] It should be noted that the coolant flow precise control device provided in this embodiment of the invention is used to execute all the process steps of the coolant flow precise control method in the above embodiment. The working principle and beneficial effect of the two are one-to-one, so they will not be described again.
[0057] This invention also provides an electronic device. The electronic device includes a processor, a memory, and a computer program stored in the memory and executable on the processor, such as a data acquisition program. When the processor executes the computer program, it implements the steps in the various embodiments of the precise coolant flow control method described above, for example... Figure 1 Step S11 is shown. Alternatively, when the processor executes the computer program, it implements the functions of each module / unit in the above-described device embodiments, such as the data acquisition and processing module.
[0058] For example, the computer program may be divided into one or more modules / units, which are stored in the memory and executed by the processor to complete the present invention. The one or more modules / units may be a series of computer program instruction segments capable of performing a specific function, which describe the execution process of the computer program in the electronic device.
[0059] The electronic device may be a desktop computer, laptop, handheld computer, or smart tablet, etc. The electronic device may include, but is not limited to, a processor and memory. Those skilled in the art will understand that the above components are merely examples of electronic devices and do not constitute a limitation on the electronic device. It may include more or fewer components than described above, or combine certain components, or different components. For example, the electronic device may also include input / output devices, network access devices, buses, etc.
[0060] The processor can be a Central Processing Unit (CPU), or other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. A general-purpose processor can be a microprocessor or any conventional processor. The processor is the control center of the electronic device, connecting all parts of the electronic device via various interfaces and lines.
[0061] The memory can be used to store the computer programs and / or modules. The processor implements various functions of the electronic device by running or executing the computer programs and / or modules stored in the memory and by calling data stored in the memory. The memory may mainly include a program storage area and a data storage area. The program storage area may store the operating system, at least one application program required for a function (such as sound playback function, image playback function, etc.), etc.; the data storage area may store data created according to the use of the mobile phone (such as audio data, phonebook, etc.). In addition, the memory may include high-speed random access memory, and may also include non-volatile memory, such as hard disk, memory, plug-in hard disk, smart media card (SMC), secure digital (SD) card, flash card, at least one disk storage device, flash memory device, or other volatile solid-state storage device.
[0062] Wherein, if the modules / units integrated in the electronic device are implemented as software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, all or part of the processes in the methods of the above embodiments of the present invention can also be implemented by a computer program instructing related hardware. The computer program can be stored in a computer-readable storage medium, and when executed by a processor, it can implement the steps of the various method embodiments described above. The computer program includes computer program code, which can be in the form of source code, object code, executable files, or certain intermediate forms. The computer-readable medium can include: any entity or device capable of carrying the computer program code, recording media, USB flash drives, portable hard drives, magnetic disks, optical disks, computer memory, read-only memory (ROM), random access memory (RAM), electrical carrier signals, telecommunication signals, and software distribution media, etc. It should be noted that the content included in the computer-readable medium can be appropriately added or removed according to the requirements of legislation and patent practice in the jurisdiction. For example, in some jurisdictions, according to legislation and patent practice, computer-readable media do not include electrical carrier signals and telecommunication signals.
[0063] It should be noted that the device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate, and the components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs. Furthermore, in the accompanying drawings of the device embodiments provided by this invention, the connection relationships between modules indicate that they have communication connections, which can be specifically implemented as one or more communication buses or signal lines. Those skilled in the art can understand and implement this without any creative effort.
[0064] The specific embodiments described above further illustrate the purpose, technical solution, and beneficial effects of the present invention. It should be understood that the above descriptions are merely specific embodiments of the present invention and are not intended to limit the scope of protection of the present invention. In particular, it should be noted that any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention for those skilled in the art.
Claims
1. A method for precise control of coolant flow rate, characterized in that, include: Temperature data, pressure data, and flow velocity signals are acquired and subjected to noise reduction and anomaly correction to obtain the initial dataset; Based on the initial dataset, the flow patterns of the coolant are classified using a support vector machine to determine the flow pattern type; If the flow pattern is boiling, then the temperature difference value is extracted from the initial dataset, and the supercooling degree value is obtained by combining the temperature difference value with the preset supercooling degree judgment rule. The temperature distribution characteristics and fluid velocity parameters are obtained and integrated with the supercooling value and the flow pattern type into an initial feature set. The flow pattern parameters are obtained and corrected according to the initial feature set. The predicted value of heat transfer efficiency is calculated based on the corrected parameters and compared with the actual value to obtain the transfer deviation value. A deviation identifier is generated based on the transmitted deviation value, environmental parameters are obtained, and flow adjustment requirements are analyzed based on the deviation identifier and the environmental parameters. Initial control parameters are then generated based on the requirements. Acquire pressure fluctuation data, and adjust the initial control parameters based on the pressure fluctuation data to obtain the final control parameters; The coolant flow rate is adjusted based on the final control parameters to obtain the coolant volume adjustment result.
2. The method for precise control of coolant flow rate according to claim 1, characterized in that, The acquisition of temperature data, pressure data, and flow velocity signals, followed by noise reduction and anomaly correction, yields an initial dataset, including: Temperature data, pressure data, and flow velocity signals are collected and noise reduction is performed to obtain the first parameter set; The first parameter set is smoothed to obtain the second parameter set; The correlation between the temperature data and the pressure data in the second parameter set is analyzed, and abnormal correlations are corrected to obtain the initial dataset.
3. The method for precise control of coolant flow rate according to claim 1, characterized in that, The step of classifying the flow patterns of the coolant using a support vector machine based on the initial dataset to determine the flow pattern type includes: The correlation between flow rate and temperature in the initial dataset was analyzed, and abnormal correlations were corrected to obtain a correlated dataset. The characteristics and flow patterns of the coolant are obtained, and a first correlation index between the coolant characteristics and the flow patterns is calculated. The associated dataset is then corrected based on the first correlation index to obtain an optimized dataset. The optimized dataset is classified using a support vector machine. Based on the classification results and the preset rules for bubble generation and boiling state, the flow pattern type is determined.
4. The method for precise control of coolant flow rate according to claim 1, characterized in that, The step of extracting temperature difference values from the initial dataset and determining the degree of supercooling based on these temperature difference values and a preset supercooling determination rule includes: Identify and mark points in the initial dataset with abnormal temperature differences to obtain the first set of differences; The first set of differences is smoothed to obtain the second set of differences; Based on the second set of differences, a second correlation index between the temperature difference and the boiling state is calculated. The second set of differences is then corrected based on the second correlation index to obtain a third set of differences. The third set of differences is classified and processed, and combined with the preset supercooling degree judgment rules, the supercooling degree value is obtained.
5. The method for precise control of coolant flow rate according to claim 1, characterized in that, The process involves acquiring temperature distribution characteristics and fluid velocity parameters, integrating them with the supercooling degree value and the flow pattern type to form an initial feature set, obtaining and correcting flow mode parameters based on the initial feature set, calculating a predicted heat transfer efficiency value based on the corrected parameters, and comparing it with the actual value to obtain a transfer deviation value, including: The temperature distribution characteristics and fluid velocity parameters are obtained, and the supercooling value, the flow pattern type, the temperature distribution characteristics, and the fluid velocity parameters are integrated to obtain an initial feature set; Data on the influence of ambient temperature and cooling properties are acquired, and the initial feature set is modified based on the data to obtain an optimized feature set. Obtain boundary condition data and heat transfer paths, and correlate them with the optimized feature set to obtain dynamic flow mode parameters. Then, correct the dynamic flow mode parameters to obtain a flow feature set. Based on the set of flow characteristics, a predicted value of heat transfer efficiency is calculated using a convolutional neural network. The difference between the predicted value and the actual value collected in real time is analyzed to obtain the transfer deviation value.
6. The method for precise control of coolant flow rate according to claim 1, characterized in that, The process of acquiring environmental parameters, generating a deviation identifier based on the transmission deviation value, acquiring environmental parameters, analyzing flow adjustment requirements based on the deviation identifier and the environmental parameters, and generating initial control parameters for the requirements includes: Based on the transmitted deviation value, a deviation identifier is generated; Obtain environmental parameters, associate the environmental parameters with the deviation identifier, analyze the demand type for flow adjustment, and obtain the first control demand; Obtain historical data of flow control, analyze the first degree of matching between the historical data and the first control requirement, determine the flow adjustment instruction based on the first degree of matching, and obtain the first adjustment instruction; Based on the first adjustment instruction, the corresponding amplitude adjustment data is determined to obtain the initial control parameters.
7. The method for precise control of coolant flow rate according to claim 1, characterized in that, The process of acquiring pressure fluctuation data and adjusting the initial control parameters based on the pressure fluctuation data to obtain the final control parameters includes: Acquire pressure fluctuation data, and generate fluctuation identifiers based on the pressure fluctuation data; The fluctuation identifier and the initial control parameters are correlated to determine the pressure compensation requirement and obtain the second control requirement. Obtain historical pressure fluctuation and flow control matching records, analyze the matching records and the second degree of matching with the second control requirement, obtain the pressure adjustment requirement based on the second degree of matching, and determine the second adjustment instruction; According to the second adjustment instruction, the corresponding pressure adjustment data is obtained and the pressure adjustment data is optimized to obtain the final control parameters.
8. A device for precise control of coolant flow rate, characterized in that, include: The data acquisition and processing module is used to acquire temperature data, pressure data, and flow velocity signals, and perform noise reduction and anomaly correction to obtain the initial dataset. The flow pattern classification module is used to classify the flow pattern of the coolant using a support vector machine based on the initial dataset, and to determine the flow pattern type. The supercooling degree analysis module is used to extract the temperature difference value from the initial dataset if the flow mode type is boiling, and to make a judgment based on the temperature difference value and a preset supercooling degree judgment rule to obtain the supercooling degree value. The deviation prediction module is used to acquire temperature distribution characteristics and fluid velocity parameters, and integrate them with the supercooling degree value and the flow pattern type into an initial feature set. The flow mode parameters are acquired and corrected based on the initial feature set. The predicted value of heat transfer efficiency is calculated based on the corrected parameters and compared with the actual value to obtain the transfer deviation value. The control parameter generation module is used to generate a deviation identifier based on the transmission deviation value, obtain environmental parameters, analyze the flow adjustment requirements based on the deviation identifier and the environmental parameters, and generate initial control parameters for the requirements. The control parameter optimization module is used to acquire pressure fluctuation data and adjust the initial control parameters according to the pressure fluctuation data to obtain the final control parameters; The flow control module is used to adjust the coolant flow rate according to the final control parameters to obtain the coolant quantity adjustment result.