Water-cooled valve train pressure control method, system, and storage medium

Abstract

The application relates to the technical field of industrial automation control, and in particular to a water-cooled valve group pressure control method and system and a storage medium. The method comprises the following steps: acquiring and standardizing pressure data to generate a fluctuation sequence; identifying correlation between loops and potential conflict points; determining an optimal intervention timing based on a threshold judgment to construct a speed model and an optimization target; generating and verifying an adjustment instruction set to ensure system balance; updating parameters through real-time data to determine a timing synchronization protocol; driving the valve group according to the protocol and monitoring pressure changes to calculate a system balance index; analyzing interaction and noise characteristics based on the index, triggering noise filtering when conditions are met, and generating optimization iteration information. The application effectively improves the stability and response synergy of water-cooled valve group pressure control.

Description

Technical Field

[0001] This application relates to the field of industrial automation control technology, and in particular to a water cooling valve group pressure control method, system and storage medium. Background Technology

[0002] Water-cooled valve assemblies are widely used in industrial systems, particularly in chemical, energy dispatch, and smart manufacturing fields. The core task of a water-cooling system is to ensure stable equipment operation by regulating water flow and temperature. However, with increasing system complexity, traditional water-cooled valve assembly pressure control methods often struggle to cope with the interactions between loops in multi-loop systems, especially under conditions of pressure fluctuations and valve response delays, which can easily lead to regulation conflicts and system instability.

[0003] Traditional control methods typically focus on the independent regulation of a single loop, neglecting the coupling effects between loops. For example, when multiple valve groups are regulated simultaneously, pressure fluctuations in one loop may trigger a chain reaction in other loops, leading to overall system instability. Furthermore, the delay in valve group response and dynamic changes in the system make it difficult to maintain system balance, especially under high loads or sudden conditions, where traditional methods cannot effectively predict and regulate pressure fluctuations.

[0004] Therefore, how to monitor system pressure in real time in a dynamic environment and accurately identify regulation conflicts caused by pressure fluctuations and response delays has become a key issue in water cooling valve group pressure control technology. An innovative control method is urgently needed to dynamically optimize valve group response and ensure the balance and stability of multi-loop systems. Summary of the Invention

[0005] This application provides a water cooling valve group pressure control method, system, and storage medium to improve the stability and responsiveness of water cooling valve group pressure control.

[0006] In a first aspect, this application provides a method for controlling the pressure of a water cooling valve assembly, the method comprising:

[0007] S1. Obtain real-time pressure data and valve group status information from the multi-loop system, process the real-time pressure data using time series analysis methods, and combine it with the valve group status information to obtain a standardized pressure peak fluctuation sequence.

[0008] S2. Based on the standardized pressure peak fluctuation sequence, determine the sequence correlation between each loop and identify potential regulation conflict points;

[0009] S3. If the speed value of the potential adjustment conflict point exceeds the preset speed threshold, adjust the valve group response characteristic parameters to obtain a preliminary speed calculation model, and construct an optimization objective function based on historical pressure data to determine the optimal intervention time.

[0010] S4. Generate an initial adjustment instruction sequence based on the optimal intervention timing, determine whether the influence of the instruction sequence on the system balance exceeds a preset balance threshold, and obtain a verified instruction set.

[0011] S5. Extract key action parameters from the verified instruction set, update the key action parameters through real-time monitoring data, and determine the final timing action synchronization protocol.

[0012] S6. Drive the valve group to perform actions according to the timing action synchronization protocol, and monitor the change in valve group pressure state after the action is performed to obtain the system balance index.

[0013] S7. Based on the system balance index, integrate the interaction characteristics between multiple loops, determine whether the noise distribution characteristics meet the preset noise distribution conditions, and if so, generate a noise filtering trigger signal to obtain optimized timing parameter optimization iteration information.

[0014] Secondly, this application provides a water cooling valve group pressure control system, the system comprising:

[0015] The time series analysis preprocessing module is used to acquire real-time pressure data and valve group status information from the multi-loop system, process the real-time pressure data using time series analysis methods, and combine it with the valve group status information to obtain a standardized pressure peak fluctuation sequence.

[0016] The sequence correlation identification module is used to determine the sequence correlation between each loop based on the standardized pressure peak fluctuation sequence and to identify potential regulation conflict points.

[0017] The intervention timing calculation module is used to adjust the valve group response characteristic parameters if the speed value of the potential adjustment conflict point exceeds the preset speed threshold, obtain a preliminary speed calculation model, and construct an optimization objective function based on historical pressure data to determine the optimal intervention timing.

[0018] The instruction set verification module is used to generate an initial adjustment instruction sequence based on the optimal intervention timing, determine whether the influence of the instruction sequence on the system balance exceeds a preset balance threshold, and obtain the verified instruction set.

[0019] The synchronization protocol determination module is used to extract key action parameters from the verified instruction set, update the key action parameters through real-time monitoring data, and determine the final timing action synchronization protocol.

[0020] The balance index monitoring module is used to drive the valve group to perform actions according to the timing action synchronization protocol, and monitor the change in valve group pressure state after the action is performed to obtain the system balance index.

[0021] The parameter optimization iteration module is used to integrate the interaction characteristics between multiple loops based on the system balance index, determine whether the noise distribution characteristics meet the preset noise distribution conditions, and if so, generate a noise filtering trigger signal to obtain optimized timing parameter optimization iteration information.

[0022] Thirdly, this application provides a computer-readable storage medium storing instructions that, when executed on a computer, cause the computer to perform the aforementioned water cooling valve group pressure control method.

[0023] Compared with the prior art, the beneficial effects of the present invention are at least as follows:

[0024] The technical solution provided in this application, by acquiring pressure data of a multi-loop system in real time and combining it with time-series analysis methods, can accurately identify potential regulation conflicts caused by pressure fluctuations and valve group response delays, thereby achieving dynamic optimization of valve group response characteristics and improving system stability and response speed. Unlike traditional methods that focus on the independent regulation of a single loop, this application's solution comprehensively considers the mutual influence between multi-loop systems, optimizes the coordinated action of valve groups, and avoids system instability caused by inconsistent responses between loops. By adapting to dynamic environments and responding to pressure fluctuations and equipment load changes in real time, this invention ensures that the system can operate quickly and stably under high loads or sudden conditions, improving the overall coordination and efficiency of the system. Simultaneously, it reduces unnecessary regulation and pressure fluctuations, thereby effectively reducing system energy consumption and equipment wear, extending equipment lifespan, and lowering maintenance costs. In summary, this invention provides a precise, efficient, and stable water-cooled valve group pressure control method with certain technical advantages and broad application prospects. Attached Figure Description

[0025] To more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings used in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0026] Figure 1 This is a flowchart of a water cooling valve group pressure control method according to this application;

[0027] Figure 2 Flowchart for the optimized pressure timing synchronization protocol of the water cooling valve assembly in this application;

[0028] Figure 3 This is a graph showing the results of the stress regression analysis in this application;

[0029] Figure 4This is a schematic diagram of the structure of a water cooling valve group pressure control system according to this application. Detailed Implementation

[0030] This application provides a water cooling valve group pressure control method, system, and storage medium. The terms "first," "second," "third," "fourth," etc. (if present) in the specification, claims, and accompanying drawings are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments described herein can be implemented in a sequence other than that illustrated or described herein. Furthermore, the terms "comprising" or "having" and any variations thereof are intended to cover a non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0031] For ease of understanding, the specific process of the embodiments of this application is described below. Please refer to [link / reference]. Figure 1 One embodiment of a water cooling valve group pressure control method in this application includes:

[0032] Step S1: Obtain real-time pressure data and valve group status information from the multi-loop system, process the real-time pressure data using time series analysis, and combine it with the valve group status information to obtain a standardized pressure peak fluctuation sequence.

[0033] In one specific embodiment, the process of performing step S1 may specifically include the following steps:

[0034] Real-time pressure data and valve group status information of each loop are acquired through a sensor network;

[0035] Perform time-series analysis on real-time pressure data to extract the pressure peak sequence of each loop;

[0036] Based on the pressure peak sequence, calculate the pressure fluctuation frequency and pressure fluctuation amplitude between each loop;

[0037] Based on the pressure fluctuation frequency and pressure fluctuation amplitude, combined with the valve group status information, the time delay distribution between each loop is determined;

[0038] Based on the pressure peak sequence and combined with the delay time distribution, a pressure peak fluctuation sequence between each loop is generated;

[0039] The pressure peak fluctuation sequence is standardized to obtain a standardized pressure peak fluctuation sequence.

[0040] Specifically, real-time pressure data is obtained through pressure sensors installed in each loop. These sensors continuously monitor the fluid pressure in each loop of the system and transmit the data to the control system in real time. The pressure data for each loop changes continuously in the time domain, reflecting the internal pressure status of the system in real time. Meanwhile, valve group status information is obtained through valve status sensors, including at least the valve's opening degree, closing status, and response time. For example, valve status sensors can detect indicators such as the valve's opening angle (e.g., 0% to 100%) and the time required for the valve to respond to a predetermined state.

[0041] Furthermore, time-series analysis identifies significant fluctuations in system pressure at a specific moment or within a given time period by analyzing the trends and patterns of pressure data over time. Since sensor data can be affected by noise, real-time pressure data is typically smoothed before time-series analysis, for example, using moving averages or filtering algorithms (such as low-pass filters) to reduce noise. Then, peak detection algorithms are used to identify the pressure peaks of each loop from the time-series pressure data. For each loop's pressure data sequence... The existence of a local maximum is determined by comparing pressure data at adjacent time points. If at time... The pressure data is greater than the pressure data at the two consecutive time points, i.e. and If the value is not found at a given moment, then that moment is considered the pressure peak. After peak detection, a pressure peak sequence can be obtained. Each element Indicates at a point in time The pressure peaks detected at the location. Using this method, the pressure peak sequence for each loop can be extracted.

[0042] Pressure fluctuation amplitude and frequency are calculated based on the pressure peak sequence. Specifically, the difference between the maximum and minimum values ​​in the pressure peak sequence is the pressure fluctuation amplitude. For example, if the pressure peak sequence of loop 1 is 50 kPa, 65 kPa, 60 kPa, 80 kPa, etc., then the pressure fluctuation amplitude of this loop is the difference between the maximum pressure (80 kPa) and the minimum pressure (50 kPa) in the peak sequence, which is 30 kPa. The pressure fluctuation frequency describes the frequency at which pressure fluctuations occur, reflecting the rate of pressure change. The time interval between adjacent pressure peaks can be calculated from the pressure peak sequence. If adjacent peaks in the pressure peak sequence of loop 1 occur within a short period, the fluctuation frequency is high, indicating that the pressure fluctuations in this loop are more frequent; conversely, a longer time interval indicates a lower frequency. By calculating the time intervals between these pressure peaks and extrapolating the average period based on these intervals, the pressure fluctuation frequency of this loop can be obtained. A higher frequency indicates more frequent pressure changes, which may mean that the system faces a greater fluctuation burden.

[0043] Specifically, based on the frequency and amplitude of pressure fluctuations, the time delay distribution of each loop can be inferred. The frequency and amplitude of pressure fluctuations reflect the severity of pressure changes in the loop. Generally, loops with larger pressure fluctuations require faster valve responses to maintain system balance, thus their valve response times are shorter and their time delays are smaller. Conversely, loops with smaller pressure fluctuations may require longer responses to pressure changes, resulting in larger time delays. By calculating the frequency and amplitude of pressure fluctuations in each loop, the time delay distribution of each loop can be preliminarily estimated. For example, if loop H1 has a high frequency and large amplitude of pressure fluctuations, it indicates that the pressure changes in this loop are more drastic, requiring a rapid response to maintain balance, and its time delay is smaller. Conversely, loop H2 has smaller pressure fluctuations, a slower response, and a larger time delay. Simultaneously, by combining valve group status information, the time delay of each loop can be determined. The valve response time refers to the time difference between receiving a control command and the actual valve action. By analyzing the valve group status information, the system can obtain the valve response characteristics of each loop. Specifically, if a valve can execute a control command quickly, the response time of that loop is short and the delay time is small; if the valve reacts slowly, resulting in a long time interval between the execution of the control command, the delay time of that loop is large. Based on these data, the resulting delay time distribution can reflect the time delay characteristics of the valve response in each loop and reveal the cooperative relationship between these loops in the system.

[0044] The method for obtaining pressure peak fluctuation sequences involves real-time acquisition of pressure data from each loop in the system. This data is typically monitored and transmitted to the control system via a sensor network installed in the loops. Assume that pressure data for loops F1 and F2 are acquired from the system within a certain time period. Time series analysis is used to process this real-time pressure data and extract the key characteristics of pressure fluctuations in each loop. Specifically, the pressure peak refers to the maximum pressure value of a loop within a certain time range, representing the intensity of the pressure fluctuation. For example, if the pressure in loop F1 reaches 100 kPa at time t = 3 seconds, and then the pressure begins to decrease, then 100 kPa at t = 3 seconds is the pressure peak of loop F1. Subsequently, based on the time delay distribution between loops, the system infers how the pressure fluctuations of each loop influence each other. Due to the different response times of the valves in each loop, when the pressure fluctuation in loop F1 reaches its peak, the pressure fluctuation in loop F2 will appear with a delay, the time of which is determined by the time delay distribution between loops. If the valve response in loop F2 has a 2-second delay, then the pressure peak of loop F1 occurs at t = 3 seconds, while the pressure peak of loop F2 may not appear until t = 5 seconds. For example, suppose the pressure in loop F1 reaches 100 kPa at t=3 seconds, and the delay time of loop F2 is 2 seconds. Then the pressure in loop F2 will reach its peak at t=5 seconds. In this way, by combining the delay time of each loop, the system can generate the pressure fluctuation sequence of loop F2 based on the pressure fluctuation of loop F1, and thus obtain the pressure peak fluctuation sequence of each loop.

[0045] The purpose of standardizing pressure peak fluctuation sequences is to transform pressure fluctuation sequences from different loops to a unified dimension and range, thereby eliminating the influence of pressure value differences between different loops. The typical steps are: first, calculate the mean of the sequence (i.e., the average pressure peak); second, calculate the standard deviation of the pressure peak sequence for that loop; third, normalize the original pressure peak fluctuation sequence by subtracting the mean and dividing by the standard deviation. After standardization, the resulting standardized pressure peak fluctuation sequence removes potential dimensional differences from the original data, allowing pressure fluctuations from different loops to be compared and processed under the same standard.

[0046] Step S2: Based on the standardized pressure peak fluctuation sequence, determine the sequence correlation between each loop and identify potential regulation conflict points.

[0047] In one specific embodiment, the process of performing step S2 may specifically include the following steps:

[0048] Based on the standardized peak pressure fluctuation sequence, the sequence correlation between each loop was determined;

[0049] The sequence correlation is vectorized to obtain the feature vector;

[0050] The anomaly propagation path is simulated based on feature vectors, and the nodes and time points of anomaly propagation are determined based on the anomaly propagation path.

[0051] Analyze the nodes and time points of abnormal propagation to identify potential regulatory conflict points.

[0052] Specifically, when calculating the sequence correlation between loops, the correlation coefficient is typically used to measure the linear relationship between standardized pressure peak fluctuation sequences of different loops. The most commonly used correlation coefficient is the Pearson correlation coefficient, which quantifies the relationship between the pressure fluctuation sequences of two loops. The correlation coefficient ranges from -1 to 1: if the correlation coefficient is close to 1, it indicates that the pressure fluctuations of the two loops are highly positively correlated, meaning that when the pressure in one loop rises, the pressure in the other loop will also rise; if the correlation coefficient is close to -1, it indicates that the pressure fluctuations of the two loops are negatively correlated, meaning that when the pressure in one loop rises, the pressure in the other loop will fall; if the correlation coefficient is close to 0, it indicates that there is almost no linear relationship between the pressure fluctuations of the two loops.

[0053] Based on the previously calculated correlations of the pressure peak fluctuation sequences between each loop, these correlations are transformed into a feature vector. During vectorization, the correlation between each loop is treated as a dimension, forming a vector containing all relationships between loops. Assuming there are n loops, the final feature vector will be an n-dimensional vector, with each dimension representing the correlation between two loops. For example, the first element in the feature vector might represent the correlation between loop 5 and loop 6, the second element the correlation between loop 5 and loop 7, and so on.

[0054] Furthermore, it is necessary to identify the source loop of the anomaly (i.e., the anomaly source). For example, if the pressure in a loop exceeds the normal fluctuation range, triggering an anomaly alarm, this loop is assumed to be the starting point of the anomaly, i.e., where the anomaly initially occurred. The correlation between loops can be analyzed using eigenvectors. Eigenvectors represent the linear relationship between different loops; in other words, they reflect the degree to which the fluctuation (anomaly) in one loop might affect other loops. For example, a high correlation between loops 11 and 12 means that a pressure change in loop 11 might trigger a pressure fluctuation in loop 12. Therefore, by calculating the correlation coefficient in the eigenvectors, we can determine which loops have a strong correlation with the anomaly source, thereby predicting the potential path of anomaly propagation. Based on the correlation analysis between loops, it is assumed that anomaly propagation between loops is transmitted through correlation. The specific propagation mechanism can be set as follows: if loop A and loop B have a high correlation, and loop A experiences anomaly fluctuations, loop B is also likely to experience similar fluctuations in the following period; anomaly propagation is usually accompanied by a time delay, the specific delay time of which can be determined by the time delay distribution between loops. The time delay between loops reflects the time delay of valve response, that is, the time lag in the propagation of anomaly fluctuations in the system.

[0055] Based on the correlation and time delay between loops, the path of anomaly propagation can be simulated. The simulation steps include: starting from the anomaly source, calculating the correlation between the loop and other loops, and judging the next loop that the anomaly may propagate to; for each loop with high correlation, predicting how long the loop may be affected (i.e., at what time point it will exhibit abnormal fluctuations) based on the time delay distribution; and gradually simulating the propagation of the anomaly from one loop to other loops with high correlation over time, forming a propagation path.

[0056] During the simulation of anomaly propagation, the loops to which the anomaly propagates (i.e., anomaly propagation nodes) can be identified, and the time points at which the anomaly occurs at these nodes can be recorded. Assuming loop 21 is the anomaly source, loops 22 and 23 have a high correlation with loop 21, and based on the time delay between loops, the pressure in loop 22 may be affected by the anomaly in loop 21 after 10 seconds, while loop 23 will be affected after 15 seconds. In this way, the system not only identifies the path of anomaly propagation but also determines the time point at which the anomaly occurs at each node.

[0057] The identification of potential regulation conflict points is based on the nodes and time points of anomaly propagation. It is assumed that when an anomaly occurs in one loop, pressure fluctuations in that loop will affect other loops, especially those highly correlated with it. Regulation conflicts may arise if the regulatory actions between loops are not coordinated in a timely manner, or if multiple loops experience pressure fluctuations and are affected by the anomaly within a similar timeframe. Regulation conflicts typically manifest as multiple loops attempting to adjust valve openings or flow rates at similar times, which can further exacerbate system pressure fluctuations and even lead to instability. The core of identifying potential regulation conflict points is identifying which loops experience pressure fluctuations within a similar timeframe and require regulation. If multiple loops attempt to adjust their valves or flow rates within a short period, and their regulatory actions are not coordinated, regulation conflicts may occur. For example, suppose the pressure in loop T1 peaks at t=3 seconds, the valve response in loop T2 may need to adjust at t=3.5 seconds, and the pressure in loop T3 will also fluctuate and require adjustment at t=4 seconds. If the regulatory actions of loops T2 and T3 are close together and there is insufficient time interval, this can lead to system instability because they may compete for the same system resources (such as flow control or valve adjustment), thus triggering a regulation conflict.

[0058] Step S3: If the speed value of the potential adjustment conflict point exceeds the preset speed threshold, adjust the valve group response characteristic parameters to obtain a preliminary speed calculation model, and construct an optimization objective function based on historical pressure data to determine the optimal intervention time.

[0059] In one specific embodiment, the process of performing step S3 may specifically include the following steps:

[0060] Obtain the speed value of potential adjustment conflict points and determine whether the speed value exceeds the preset speed threshold. If so, extract the valve group response characteristic parameters from the valve group status information.

[0061] The gradient descent algorithm is used to adjust the valve group response characteristic parameters, and an integral formula for velocity calculation is constructed based on the adjusted valve group response characteristic parameters.

[0062] Determine whether the rate of change of the integral formula for velocity calculation is less than a preset threshold. If so, generate a preliminary velocity calculation model.

[0063] Historical pressure data stored in the sensor network is read, and historical fluctuation patterns of each loop are constructed based on the historical pressure data, and further optimization objective functions are established.

[0064] By optimizing the objective function, candidate time points for intervention are calculated;

[0065] The system state characteristics of each candidate time point are extracted. Based on the system state characteristics and the preliminary velocity calculation model, the intervention effect of each candidate time point is evaluated.

[0066] Determine the optimal timing for intervention based on the intervention's effectiveness.

[0067] Specifically, by analyzing potential regulation conflict points, the rate of pressure change in each loop is calculated, i.e., the rate value of the potential regulation conflict point. This rate value can be estimated by changes in pressure data and usually reflects the severity of pressure fluctuations. A large rate of pressure change means that the system may be in an unstable state, with a risk of regulation conflict. The preset rate threshold is set according to the system's stability requirements and is used to distinguish between normal and abnormal fluctuations. If the rate of pressure change in a loop exceeds this threshold, it is determined that the loop has a potential regulation conflict point, and intervention may be necessary. Valve group response characteristic parameters related to the loop are extracted from the valve group status information. These parameters include valve response time, sensitivity, and regulation accuracy, which describe the valve group's ability to respond to pressure changes.

[0068] A loss function is defined to measure the difference between the current valve group's response characteristics and the ideal state. The loss function can typically be expressed as the squared error of the pressure fluctuation, i.e., the sum of the squares of the differences between the current pressure fluctuation and the target pressure fluctuation. Using a gradient descent algorithm, the valve group's response characteristics are iteratively updated, gradually reducing the value of the loss function and thus optimizing the valve group's response capability. Assume the valve group's response characteristics are... loss function This can be expressed as the difference between the current pressure fluctuation and the expected pressure fluctuation of the system, and can be written in the following form:

[0069]

[0070] in, For the current moment The actual pressure value, The target pressure value is given by n, which represents the number of samples. The valve group response characteristic parameters are obtained using the gradient descent algorithm. The gradient will be adjusted according to the loss function, that is:

[0071]

[0072] in, For learning rate, These are the parameter values ​​updated using the gradient descent algorithm. These are the parameter values ​​from the previous iteration. It is the loss function with respect to parameters The partial derivatives. In this way, the valve group response characteristic parameters are continuously adjusted until the system pressure fluctuations are minimized, achieving the optimal response.

[0073] After adjusting the valve group response characteristic parameters, a velocity calculation integral formula is constructed based on these parameters. This formula is used to simulate the pressure change process in the loop. Specifically, it can predict the rate of change of loop pressure at a certain moment, reflecting the system's response speed to external disturbances or internal adjustments. Assuming the system's pressure change rate is... The rate of change can be obtained by integration:

[0074]

[0075] in, The rate of change of pressure over time is the integral formula for velocity calculation. The rate of change can be expressed as the degree of change of velocity over time. If the calculated rate of change is less than a preset threshold, it means that the system has stabilized and the pressure fluctuation is no longer large. This indicates that the system's adjustment parameters have reached an equilibrium state, and no further adjustment is needed. The integral formula for velocity calculation can then be used as the initial model for velocity calculation.

[0076] By reading historical pressure data from the sensor network, which reflects pressure changes in each loop over a period of time, the system can identify historical fluctuation patterns in each loop through analysis. These patterns reveal the pressure fluctuation characteristics of each loop over different time periods; for example, some loops experience larger pressure fluctuations at specific times, while others maintain relatively stable fluctuations. These historical fluctuation patterns are crucial for understanding the long-term operating behavior of the system and identifying potential regulation trends. Based on these patterns, an optimization objective function is further established, typically by minimizing the error in pressure fluctuations across each loop in the system. For example, the optimization objective might be to minimize pressure fluctuations in the loops or maintain coordination between loops to reduce the transmission effect of pressure fluctuations. The design of the optimization objective function is usually determined based on actual control requirements, with the goal of finding a regulation strategy that reduces system instability or conflict.

[0077] The optimal time points for intervention are determined by calculating and optimizing the objective function. These points typically occur when system pressure fluctuations exceed a certain threshold. Candidate time points are those moments when system pressure fluctuations reach a level that may affect system stability or cause regulatory conflicts between loops. After obtaining the candidate time points for intervention, system state characteristics are extracted from each time point. These characteristics include system pressure, flow rate, valve opening, system load, etc., comprehensively reflecting the system's operating status at these moments. For example, a candidate time point might be an opportunity to intervene when loop pressure fluctuations are significant. A preliminary model, combined with velocity calculations, is used to assess whether the intervention can effectively reduce pressure fluctuations.

[0078] At multiple candidate intervention time points, various system state data, such as pressure, flow rate, valve opening, and system load, are extracted. These state characteristics reflect the overall operating state of the system and the specific circumstances of pressure fluctuations. For example, at a specific moment, the system may experience significant pressure fluctuations, while at other moments, it may remain relatively stable. A preliminary velocity calculation model helps predict future pressure changes, particularly pressure fluctuations at the intervention time. If the system state characteristics at a candidate intervention time point indicate rapid pressure fluctuations, and the velocity model predicts that intervention at this time can effectively mitigate these fluctuations, then the intervention at that moment is likely to be more effective. The intervention effect at each candidate time point is evaluated. Based on the system state characteristics and the velocity calculation model, the system can simulate pressure changes at different intervention time points. This includes predicting valve response, the magnitude of pressure fluctuation reduction, and the impact on system equilibrium. By statistically analyzing the frequency of different scenarios, an effectiveness score is calculated to determine the optimal intervention timing and avoid regulatory conflicts.

[0079] Step S4: Generate an initial adjustment instruction sequence based on the optimal intervention timing, determine whether the influence of the instruction sequence on the system balance exceeds the preset balance threshold, and obtain the verified instruction set.

[0080] In one specific embodiment, the process of performing step S4 may specifically include the following steps:

[0081] The initial adjustment command sequence is generated based on the optimal intervention timing, and the initial adjustment command sequence is calibrated using a synchronous clock.

[0082] Generate an action trigger sequence based on the calibrated initial adjustment command sequence;

[0083] Simulate the system equilibrium state based on the action trigger sequence to obtain the state results;

[0084] Based on the state results, quantify the impact of the instruction sequence on the system balance;

[0085] Determine if the impact exceeds the preset balance threshold. If so, use a genetic algorithm to adjust the action trigger sequence and generate a verified instruction set.

[0086] Specifically, the optimal intervention timing is organized into commands to obtain an initial control command sequence, which mainly includes adjustment commands for valve opening, flow rate, or pressure settings in each loop. The command sequence typically consists of multiple steps, each representing a loop adjustment at a specific moment or time window. For example, if the pressure in loop 41 exceeds a threshold, it may be necessary to adjust the valve opening to 50% at t=10 seconds and adjust the flow rate at t=12 seconds. Simultaneously, synchronization clock calibration ensures that all loop control commands are executed at the same time step. This means that the control commands for each loop will be executed within the same time window, preventing any loop from adjusting too early or too late. For example, assuming the valve in loop 42 needs adjustment at t=10 seconds and the valve in loop 43 needs adjustment at t=10.5 seconds, the synchronization clock unifies these two time points to t=10 seconds, ensuring they are adjusted within the same time frame. Based on the calibrated initial control command sequence, an action trigger sequence is further generated, which determines the timing of the control operation for each loop at different time points. The generated action trigger sequence is used to simulate the equilibrium state of the entire system, and the simulation results are used to evaluate the system's performance after the execution of regulation operations in each loop. The core of this process is to execute the loop regulation actions in a time sequence using computer simulation or numerical simulation, simulating the system's dynamic response under these regulation operations, and ultimately obtaining the system's state results. The state results represent the performance of key parameters such as pressure and flow rate after the system executes these regulation commands. Next, based on the state results, the impact of the regulation commands on system pressure fluctuations is quantified, typically by calculating the difference in pressure fluctuations before and after regulation to determine the degree of impact. For example, how much the peak difference or standard deviation of loop pressure fluctuations decreased after the regulation command was executed can be used to evaluate the effectiveness of the command.

[0087] Based on the current control command sequence and the quantified impact, a balance threshold is set. This threshold represents the tolerance range for system pressure fluctuations or instability. If the impact exceeds this threshold, it indicates that the current control command sequence has failed to effectively reduce pressure fluctuations, and the system remains in an unstable state. At this point, a genetic algorithm is triggered to adjust the action trigger sequence, aiming to optimize the timing, amplitude, and control order in the command sequence to improve the system's balance. The application process of the genetic algorithm includes the following steps:

[0088] (1) Initializing the population: The first step of the genetic algorithm is to randomly generate an initial population, where each individual represents a different action trigger sequence, including loop adjustment timing, valve opening adjustment range, adjustment direction, etc. The selection of the initial population is random, with the aim of exploring the potential optimization space through a diverse population.

[0089] (2) Fitness Assessment: Each individual (action trigger sequence) is assessed based on the system's regulatory effect. The assessment criteria are usually based on the quantitative results of influence, with lower influence indicating better regulatory effect. Therefore, each individual's fitness value is related to its ability to effectively reduce system instability.

[0090] (3) Selection operation: Based on fitness, the genetic algorithm selects the individual with the highest fitness as the parent to generate the next generation of individuals. The selection method can be implemented by roulette wheel selection, tournament selection, etc., to ensure that individuals who can better regulate the system balance occupy a higher proportion in the next generation.

[0091] (4) Crossover and mutation: After the selection operation, the genetic algorithm performs crossover and mutation operations. The crossover operation exchanges part of the genes (i.e., part of the action trigger sequence) of two parent individuals to generate new individuals, while the mutation operation randomly changes some parameters of some individuals (such as adjusting the timing or magnitude of the regulation of a certain loop). Crossover and mutation operations help to explore new solution spaces and find possible optimal solutions.

[0092] (5) Iterative optimization: Through repeated selection, crossover and mutation operations, the genetic algorithm continuously optimizes the individuals in the population until it finds an optimal action trigger sequence. In each generation, the fitness of the individuals will become higher and higher, and the regulatory effect of the system will gradually improve.

[0093] After iterative adjustments, a verified instruction set was generated.

[0094] Step S5: Extract key action parameters from the verified instruction set, update the key action parameters through real-time monitoring data, and determine the final timing action synchronization protocol.

[0095] In one specific embodiment, the process of performing step S5 may specifically include the following steps:

[0096] Extract key action parameters from the verified instruction set;

[0097] Collect real-time monitoring data and update key action parameters based on the monitoring data;

[0098] Based on the updated key action parameters, an initial timing action synchronization protocol is generated, and a delay compensation mechanism is used to adjust the initial timing action synchronization protocol.

[0099] Determine whether the adjusted initial timing action synchronization protocol meets the executable conditions. If so, generate the final timing action synchronization protocol.

[0100] Specifically, key action parameters are extracted from the validated instruction set. This process involves parsing the instruction set to extract crucial information such as the specific timing, magnitude, and direction of each loop's adjustment, ensuring effective execution in actual control. For example, assuming the optimized instruction set determines specific times for loop adjustment, if loop 51 needs to adjust the valve opening at t=10 seconds, and loop 52 needs to adjust the flow rate at t=12 seconds, then these time points are the adjustment timings. Furthermore, the adjustment magnitude refers to the degree of adjustment for each loop. For instance, loop 51 needs to adjust the valve opening from 50% to 70%, while loop 52 needs to increase the flow rate from 10 units to 15 units. The direction of adjustment is also important; for example, loop 51 may need to increase the valve opening, while loop 52 may need to increase the flow rate. Real-time monitoring data acquisition is typically accomplished through sensors and data acquisition systems, periodically or continuously acquiring data such as pressure, flow rate, and valve opening for each loop. This data is then transmitted to the control system as input information through a feedback mechanism. After analyzing this data, the control system can obtain suggestions for optimizing and adjusting parameters, thereby dynamically adjusting the previously set key action parameters (such as valve opening, adjustment range, adjustment direction, etc.).

[0101] By updating the key action parameters, the regulatory actions that each loop needs to perform at specific times can be obtained, such as adjusting the valve opening or changing the flow rate. Using these parameters, the system can determine the timing of regulation for each loop and generate an initial timing synchronization protocol. For example, suppose the valve in loop L1 needs to be adjusted to a certain opening at t=10 seconds, and the flow rate in loop L2 needs to increase at t=12 seconds. These timings will be included as part of the initial timing synchronization protocol to ensure that regulation occurs at these specific times.

[0102] However, in real-world systems, the adjustment action is delayed due to valve response time, sensor feedback time, and coupling effects between loops. Assuming the adjustment action of loop L1 affects loop L2, the adjustment of loop L2 may not occur immediately at t=12 seconds due to the valve response delay. Without considering this delay, loop L2 might adjust at an inappropriate time, leading to adjustment conflicts between loops and even affecting the stability of the entire system. To avoid this, a delay compensation mechanism is used. Delay compensation estimates the delay time of loop L2 and adjusts its adjustment timing accordingly. For example, if the delay of loop L2 is 2 seconds, adjusting its adjustment timing from t=12 seconds to t=10 seconds ensures that when loop L1 adjusts its valve at t=10 seconds, loop L2's adjustment action will also occur synchronously, ensuring coordination between loops.

[0103] Furthermore, it is necessary to verify whether the adjusted initial timing synchronization protocol meets the executable condition. The executable condition refers to whether the adjustment instructions can be executed smoothly according to the adjusted timing without causing system instability due to time conflicts or delays. For example, suppose the adjustment actions between loops L3 and L4 should be executed synchronously, but due to the response delay of loop L4, the adjustment time of loop L4 is adjusted to be two seconds later than that of loop L3. If this adjustment causes the adjustment action of loop L4 to affect the adjustment result of loop L3, or causes a conflict in the system, then this adjusted timing synchronization protocol does not meet the executable condition.

[0104] In real-world scenarios, the adjusted protocol is comprehensively evaluated based on factors such as adjustment timing, delay compensation, valve response, and system load to determine its feasibility. If overlapping, conflicting, or uncoordinated adjustment actions in certain loops are found, the adjustment timing is further fine-tuned to ensure that all loop adjustments are executed smoothly at the predetermined time without causing system instability or misalignment.

[0105] Once the adjusted timing synchronization protocol meets all executable conditions, the system can generate the final timing synchronization protocol. This protocol is a verified and optimized adjustment timing arrangement that ensures that during actual adjustment, all loop adjustment actions are executed according to precise timing and appropriate amplitude, without causing conflicts, and can effectively coordinate adjustment operations between loops. (Reference) Figure 2 The figure illustrates the optimization process of the water cooling valve group pressure timing action synchronization protocol.

[0106] Step S6: Drive the valve group to perform actions according to the timing action synchronization protocol, and monitor the change in valve group pressure state after the action is performed to obtain the system balance index.

[0107] In one specific embodiment, the process of performing step S6 may specifically include the following steps:

[0108] Based on the final timing action synchronization protocol, valve group execution instructions are generated, and the valve group is driven to perform actions according to the valve group execution instructions;

[0109] The sensor network is used to monitor changes in valve group pressure after the action is performed.

[0110] Extract system balance characteristics based on changes in valve group pressure.

[0111] The system equilibrium characteristics are processed using a data packet format definition method to obtain the system equilibrium characteristic matrix;

[0112] Based on the system equilibrium characteristic matrix, generate system equilibrium indices.

[0113] Specifically, based on the adjustment timing and magnitude determined in the final synchronization protocol, specific valve assembly execution commands are generated. These commands instruct the valves how to adjust their opening degree or other control parameters at specified times, thereby enabling the pressure, flow rate, and other parameters of the loop system to reach predetermined targets. These valve assembly execution commands are then translated into specific control signals, driving valves, flow control devices, or other equipment to perform corresponding actions.

[0114] Once the valve assembly execution command is generated and issued, the valve assembly begins to execute its actions. After receiving the command, the valve assembly adjusts the valve opening, flow rate, or other relevant regulating parameters according to the predetermined control signals to ensure that the system completes the regulation tasks of each loop according to the predetermined sequence. The valve opening will change according to the command value, or parameters such as flow rate will be adjusted according to the command set value to achieve the regulation target.

[0115] The sensor network monitors changes in valve group pressure after actions are executed. It collects real-time data on changes in key parameters such as pressure and flow rate in each loop after adjustment, providing fundamental information for subsequent system evaluation and optimization. The sensors can monitor pressure changes after valve group actions and provide feedback on the effectiveness of valve group regulation. For example, if the valve opening is executed as instructed, but the system pressure does not change or changes only slightly, the sensor will provide real-time pressure data to indicate whether the regulation effect has met the target.

[0116] By analyzing this real-time monitoring data, system equilibrium characteristics are extracted. These characteristics are key parameters describing the current stable state of the system and typically include indicators such as loop pressure fluctuations and flow rate changes. These characteristics reflect whether the system has entered a stable state, whether the pressure has stabilized, and whether the flow rate is within the expected range. Extracting these equilibrium characteristics can help evaluate the effectiveness of current regulation strategies and whether further optimization is needed.

[0117] The extracted system balance characteristics are processed using a data packet format definition method. This method organizes these characteristics in a standardized and structured manner, enabling efficient storage and processing of balance data from different time points and different loops. After processing, these system balance characteristics form a system balance characteristic matrix, where each item represents the state characteristics of a different loop at a specific time. This matrix reflects the interaction state between loops, pressure fluctuations, and overall system balance after the execution of control commands.

[0118] Finally, based on the generated system balance characteristic matrix, a system balance index is calculated. The system balance index is a comprehensive indicator used to measure whether the system has reached a stable and coordinated state after regulation. Typically, the system balance index includes multiple dimensions, such as the amplitude of pressure fluctuations, the coordination between loops, and the stability of pressure and flow. If the balance index indicates that the system has stabilized and the coordination between loops is good, then the regulation effect is successful, and the system can continue to operate. If the balance index does not meet expectations, it indicates that the system has not yet stabilized, and further optimization of the regulation strategy or adjustment of the intervention timing may be necessary.

[0119] Step S7: Based on the system balance index, integrate the interaction characteristics between multiple loops, determine whether the noise distribution characteristics meet the preset noise distribution conditions, and if so, generate a noise filtering trigger signal to obtain optimized timing parameter optimization iteration information.

[0120] In one specific embodiment, the process of performing step S7 may specifically include the following steps:

[0121] Extract the interaction characteristics between multiple loops from the system balance index;

[0122] Based on the node coordination rules, integrate the interaction impact features to obtain the interaction impact feature set;

[0123] An interaction effect model is generated based on the interaction effect feature set, and the distribution characteristics of sequence noise are calculated based on the interaction effect model.

[0124] Determine whether the distribution characteristics meet the preset noise distribution conditions; if so, generate a noise filtering trigger signal.

[0125] Based on the noise filtering trigger signal, start the sequence noise filtering program and obtain the filtering data;

[0126] Based on the filtered data, initial time series parameters are generated, and then optimized using the gradient descent algorithm.

[0127] Based on the optimized initial timing parameters, optimize timing parameter optimization iteration information is generated.

[0128] Specifically, the interaction characteristics between loops can be determined by comparing the time synchronicity and mutual influence between pressure fluctuations, flow changes, and regulation actions of the loops. For example, if the regulation action of loop 71 causes pressure fluctuations or adjustments in loop 72, and the flow change in loop 72 has a significant time correlation with the fluctuation in loop 71, then there is an interaction between loops 71 and 72. By calculating the correlation between these loops, the system can quantify the interaction characteristics between each loop. Interaction characteristics typically include the transmission effect of pressure fluctuations, the feedback effect of flow changes, and the synchronicity of loop regulation actions.

[0129] Based on node coordination rules, the system integrates interaction influence characteristics to obtain an interaction influence characteristic set. Node coordination rules refer to the coordination strategies between loops determined by the interaction influence and adjustment timing when the system is adjusting multiple loops. By analyzing the interaction influence characteristics between loops, the system can integrate these characteristics to form a feature set containing all loop interactions. Specifically, assuming there is a strong negative correlation between loops 81 and 82 (such as the reverse change in pressure fluctuations), and a positive correlation between the flow rates of loop 83 and loop 81, the system will integrate these characteristics to create an interaction influence characteristic set, which includes various mutual influence relationships between loops 1, 82, and 83.

[0130] An interaction model is generated based on a set of interaction characteristics. This model is used to simulate the interaction effects between loops and predict their performance in actual regulation. Interaction models are typically built using statistical regression analysis. For example, assuming a linear relationship between pressure fluctuations in loops 91 and 92, a regression model can be used to represent the impact of pressure changes in loop 91 on loop 92. The model uses historical data between loops to calculate the inter-loop transmission coefficient, i.e., the degree of influence of pressure changes in loop 91 on pressure changes in loop 92. This transmission coefficient is incorporated into the interaction model to help the system accurately predict inter-loop effects in future regulation.

[0131] In the above example, the basic formula of the regression model is as follows: in: This indicates that loop 92 is at time point The pressure. This indicates that loop 91 is at time point The pressure. It is the intercept term, representing the reference value of the pressure in loop 92. It is a regression coefficient, representing the degree of influence of pressure change in loop 91 on pressure change in loop 92. It is a point in time. The error term represents the portion of the pressure fluctuation in loop 92 that cannot be explained by the pressure fluctuation in loop 91. Using historical pressure data for loop 91 and loop 92 as model input, the regression model aims to predict the pressure in loop 92. . refer to Figure 3 The figure shows the results of the stress regression analysis. The left figure shows the stress regression analysis, and the right figure shows the distribution of the regression residuals.

[0132] Furthermore, noise distribution characteristics reflect fluctuations between loops in the system caused by interference or instability. Through an interaction model, the system can predict how noise (e.g., pressure or flow fluctuations) will propagate between loops during regulation and calculate the noise distribution characteristics. Noise distribution characteristics can be quantified using statistical methods such as standard deviation and root mean square error (RMSE). For example, assuming the pressure fluctuation range of loop 01 is [80 kPa, 100 kPa], and the flow fluctuation range of loop 02 is [10 units, 12 units], by calculating their standard deviations and correlation, the system can obtain the standard error of the noise distribution, thus providing a basis for noise filtering.

[0133] The system determines whether the noise distribution characteristics meet preset noise distribution conditions. If so, a noise filtering trigger signal is generated. In this step, the system checks whether the noise distribution characteristics meet the set conditions, such as whether the noise fluctuation range is too large or exceeds the system's tolerance range. If the noise distribution characteristics meet the preset conditions (e.g., the standard deviation of the noise is below a certain threshold), the noise filtering signal is triggered. This trigger signal indicates that the system should start the noise filtering procedure, that is, to eliminate or reduce the impact of noise on system stability by adjusting the control strategy.

[0134] Based on the noise filtering trigger signal, a sequential noise filtering program is initiated to acquire filtering data. The noise filtering program reduces system noise by optimizing loop regulation strategies or adjusting parameters such as valve opening and flow rate. For example, the system can avoid fluctuations caused by interference between loops by adjusting the regulation amplitude of a certain loop or changing the timing of regulation. The filtering program dynamically adjusts based on real-time monitoring data until the noise is effectively controlled and the system's balance is ensured.

[0135] Based on the filtered data, initial timing parameters are generated and optimized using a gradient descent algorithm. By analyzing the filtered data, the system generates new initial timing parameters. These parameters reflect the optimal timing, magnitude, and sequence of adjustments for each loop in the control command. To further optimize the timing parameters, the system employs gradient descent to adjust the timing parameters of each loop, achieving the best overall system control effect. The gradient descent algorithm gradually adjusts the timing parameters according to the loss function until the error in the system is minimized. For example, if the pressure fluctuations in the system are still large, the gradient descent algorithm will gradually adjust the control timing until the pressure fluctuations are minimized.

[0136] Based on the optimized initial timing parameters, optimized timing parameter iteration information is generated. The timing parameters optimized by gradient descent can be used as input parameters for the next round of adjustment. The system records these optimized parameters and generates timing parameter optimization iteration information as a reference for future adjustments. In each iteration, the system continuously updates the timing parameters and adjusts the adjustment behavior of the loop according to the system's balance requirements to ensure the stability and efficiency of the system in long-term operation.

[0137] Through these steps, the system can continuously optimize the adjustment strategy based on the interaction between loops, noise distribution, and adjustment feedback, thereby improving the operating efficiency and stability of the multi-loop system.

[0138] It is understood that the executing entity of this application can be a water cooling valve group pressure control system, a terminal, or a server; no specific limitation is made here. This application's embodiments use a server as an example for illustration.

[0139] The above describes a water cooling valve group pressure control method according to an embodiment of this application. The following describes a water cooling valve group pressure control system according to an embodiment of this application. Please refer to [link / reference]. Figure 4 One embodiment of a water cooling valve group pressure control system in this application includes:

[0140] The time series analysis preprocessing module is used to acquire real-time pressure data and valve group status information from the multi-loop system, process the real-time pressure data using time series analysis methods, and combine it with the valve group status information to obtain a standardized pressure peak fluctuation sequence.

[0141] The sequence correlation identification module is used to determine the sequence correlation between loops based on the standardized peak pressure fluctuation sequence and to identify potential regulation conflict points.

[0142] The intervention timing calculation module is used to adjust the valve group response characteristic parameters if the speed value of the potential regulation conflict point exceeds the preset speed threshold, obtain a preliminary speed calculation model, and construct an optimization objective function based on historical pressure data to determine the optimal intervention timing.

[0143] The instruction set verification module is used to generate an initial adjustment instruction sequence based on the optimal intervention timing, determine whether the influence of the instruction sequence on the system balance exceeds a preset balance threshold, and obtain the verified instruction set.

[0144] The synchronization protocol determination module is used to extract key action parameters from the verified instruction set, update the key action parameters through real-time monitoring data, and determine the final timing action synchronization protocol.

[0145] The balance index monitoring module is used to drive the valve group to perform actions according to the timing action synchronization protocol, and monitor the changes in the valve group pressure state after the actions are performed to obtain the system balance index.

[0146] The parameter optimization iteration module is used to integrate the interaction characteristics between multiple loops based on the system balance index, determine whether the noise distribution characteristics meet the preset noise distribution conditions, and if so, generate a noise filtering trigger signal to obtain optimized timing parameter optimization iteration information.

[0147] Through the collaborative efforts of the aforementioned components, the system achieves efficient and precise regulation and optimization. The time-series analysis preprocessing module acquires real-time pressure data and valve group status information, processes the data using time-series analysis methods, and obtains a standardized pressure peak fluctuation sequence, providing reliable foundational data for subsequent analysis. The sequence correlation identification module identifies the correlation between loops based on the pressure peak fluctuation sequence, discovering potential regulation conflict points and providing a basis for subsequent intervention decisions. The intervention timing calculation module, based on the identified potential regulation conflict points, calculates the system's intervention timing and constructs an optimization objective function by adjusting valve group response characteristic parameters, ensuring the system can intervene at the optimal time. The instruction set verification module generates an initial regulation instruction sequence and verifies its impact on system balance. If the impact exceeds a preset balance threshold, it further optimizes and obtains a verified instruction set. The synchronization protocol determination module extracts key action parameters from the instruction set, updates these parameters through real-time monitoring data, and ultimately determines the time-series action synchronization protocol, ensuring precise synchronous execution of regulation actions between loops. The balance index monitoring module drives the valve group to execute corresponding regulation actions according to the time-series action synchronization protocol, monitors the changes in valve group pressure status after execution, and extracts the system's balance index. The parameter optimization iteration module analyzes the interaction characteristics between loops based on the balance index, determines the noise distribution characteristics, and if the preset noise distribution conditions are met, generates a noise filtering trigger signal and optimizes the timing parameters, forming timing parameter optimization iteration information. The collaborative work of each module enables the system to dynamically adjust the regulation strategy of each loop in real time, identify and solve potential problems, and ensure that the system maintains optimal performance in complex and variable environments.

[0148] The present invention also provides a computer-readable storage medium, which can be a non-volatile computer-readable storage medium or a volatile computer-readable storage medium, wherein the computer-readable storage medium stores instructions that, when the instructions are executed on a computer, cause the computer to perform the steps of the water cooling valve group pressure control method.

[0149] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of this application. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0150] The above-described embodiments are only used to illustrate the technical solutions of this application, and are not intended to limit them. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of this application.

Claims

1. A method for controlling the pressure of a water cooling valve assembly, characterized in that, The method includes: S1. Obtain real-time pressure data and valve group status information from the multi-loop system, process the real-time pressure data using time series analysis methods, and combine it with the valve group status information to obtain a standardized pressure peak fluctuation sequence. S2. Based on the standardized pressure peak fluctuation sequence, determine the sequence correlation between each loop and identify potential regulation conflict points; S3. If the speed value of the potential adjustment conflict point exceeds the preset speed threshold, adjust the valve group response characteristic parameters to obtain a preliminary speed calculation model, and construct an optimization objective function based on historical pressure data to determine the optimal intervention time. S4. Generate an initial adjustment instruction sequence based on the optimal intervention timing, determine whether the influence of the instruction sequence on the system balance exceeds a preset balance threshold, and obtain a verified instruction set. S5. Extract key action parameters from the verified instruction set, update the key action parameters through real-time monitoring data, and determine the final timing action synchronization protocol. S6. Drive the valve group to perform actions according to the timing action synchronization protocol, and monitor the change in valve group pressure state after the action is performed to obtain the system balance index. S7. Based on the system balance index, integrate the interaction characteristics between multiple loops, determine whether the noise distribution characteristics meet the preset noise distribution conditions, and if so, generate a noise filtering trigger signal to obtain optimized timing parameter optimization iteration information.

2. The water cooling valve group pressure control method according to claim 1, characterized in that, S1 includes: Real-time pressure data and valve group status information of each loop are acquired through a sensor network; Time-series analysis was performed on the real-time pressure data to extract the pressure peak sequence of each loop; Based on the pressure peak sequence, calculate the pressure fluctuation frequency and pressure fluctuation amplitude between each loop; Based on the pressure fluctuation frequency and the pressure fluctuation amplitude, and combined with the valve group status information, the delay time distribution between each loop is determined; Based on the pressure peak sequence and the delay time distribution, a pressure peak fluctuation sequence between each loop is generated; The pressure peak fluctuation sequence is standardized to obtain a standardized pressure peak fluctuation sequence.

3. The water cooling valve assembly pressure control method according to claim 2, characterized in that, S2 includes: Based on the standardized pressure peak fluctuation sequence, the sequence correlation between each loop is determined; The sequence correlation is vectorized to obtain a feature vector; Based on the feature vector, simulate the anomaly propagation path, and based on the anomaly propagation path, determine the nodes and time points of anomaly propagation; Analyze the nodes and time points of the abnormal propagation to identify potential regulatory conflict points.

4. The water cooling valve assembly pressure control method according to claim 3, characterized in that, S3 includes: Obtain the speed value of the potential adjustment conflict point and determine whether the speed value exceeds the preset speed threshold. If so, extract the valve group response characteristic parameters from the valve group status information. The gradient descent algorithm is used to adjust the response characteristic parameters of the valve group, and an integral formula for calculating the velocity is constructed based on the adjusted response characteristic parameters of the valve group. Determine whether the rate of change of the velocity calculation integral formula is less than a preset change threshold; if so, generate a preliminary velocity calculation model. Historical pressure data stored in the sensor network is read, and historical fluctuation patterns of each loop are constructed based on the historical pressure data, and an optimization objective function is further established. Candidate time points for intervention are calculated using the aforementioned optimization objective function; The system state characteristics of each candidate time point are extracted from the candidate time points. Based on the system state characteristics and the preliminary velocity calculation model, the intervention effect of each candidate time point is evaluated. Based on the intervention effect, determine the optimal timing for intervention.

5. The water cooling valve assembly pressure control method according to claim 1, characterized in that, S4 includes: An initial adjustment command sequence is generated based on the optimal intervention timing, and the initial adjustment command sequence is calibrated using a synchronous clock. Generate an action trigger sequence based on the calibrated initial adjustment command sequence; The system equilibrium state is simulated based on the action trigger sequence to obtain the state result; Based on the state results, the impact of the instruction sequence on the system balance is quantified; Determine whether the influence exceeds a preset balance threshold. If so, use a genetic algorithm to adjust the action trigger sequence and generate a verified instruction set.

6. The water cooling valve assembly pressure control method according to claim 1, characterized in that, S5 includes: Extract key action parameters from the verified instruction set; Collect real-time monitoring data and update the key action parameters based on the monitoring data; Based on the updated key action parameters, an initial timing action synchronization protocol is generated, and a delay compensation mechanism is used to adjust the initial timing action synchronization protocol. Determine whether the adjusted initial timing action synchronization protocol meets the executable conditions. If so, generate the final timing action synchronization protocol.

7. The water cooling valve assembly pressure control method according to claim 6, characterized in that, S6 includes: Based on the final timing action synchronization protocol, valve group execution instructions are generated, and the valve group is driven to perform actions according to the valve group execution instructions; The sensor network is used to monitor changes in valve group pressure after the action is performed. Based on the changes in the pressure state of the valve group, extract the system balance characteristics; The system balance characteristics are processed using a data packet format definition method to obtain the system balance characteristic matrix; Based on the system balance characteristic matrix, a system balance index is generated.

8. The water cooling valve assembly pressure control method according to claim 1, characterized in that, S7 includes: Extract the interaction characteristics between multiple loops from the system balance index; Based on the node coordination rules, the interaction impact features are integrated to obtain the interaction impact feature set; An interaction influence model is generated based on the set of interaction influence features, and the distribution characteristics of sequence noise are calculated based on the interaction influence model. Determine whether the distribution characteristics meet the preset noise distribution conditions; if so, generate a noise filtering trigger signal. Based on the noise filtering trigger signal, start the sequence noise filtering program and obtain the filtering data; Based on the filtered data, initial time series parameters are generated, and the initial time series parameters are optimized using a gradient descent algorithm. Based on the optimized initial timing parameters, optimize timing parameter optimization iteration information is generated.

9. A water cooling valve group pressure control system, used to implement the water cooling valve group pressure control method as described in any one of claims 1 to 8, characterized in that, The water cooling valve group pressure control system includes: The time series analysis preprocessing module is used to acquire real-time pressure data and valve group status information from the multi-loop system, process the real-time pressure data using time series analysis methods, and combine it with the valve group status information to obtain a standardized pressure peak fluctuation sequence. The sequence correlation identification module is used to determine the sequence correlation between each loop based on the standardized pressure peak fluctuation sequence and to identify potential regulation conflict points. The intervention timing calculation module is used to adjust the valve group response characteristic parameters if the speed value of the potential adjustment conflict point exceeds the preset speed threshold, obtain a preliminary speed calculation model, and construct an optimization objective function based on historical pressure data to determine the optimal intervention timing. The instruction set verification module is used to generate an initial adjustment instruction sequence based on the optimal intervention timing, determine whether the influence of the instruction sequence on the system balance exceeds a preset balance threshold, and obtain the verified instruction set. The synchronization protocol determination module is used to extract key action parameters from the verified instruction set, update the key action parameters through real-time monitoring data, and determine the final timing action synchronization protocol. The balance index monitoring module is used to drive the valve group to perform actions according to the timing action synchronization protocol, and monitor the change in valve group pressure state after the action is performed to obtain the system balance index. The parameter optimization iteration module is used to integrate the interaction characteristics between multiple loops based on the system balance index, determine whether the noise distribution characteristics meet the preset noise distribution conditions, and if so, generate a noise filtering trigger signal to obtain optimized timing parameter optimization iteration information.

10. A computer-readable storage medium storing instructions thereon, characterized in that, When the instruction is executed by the processor, it implements a water cooling valve group pressure control method as described in any one of claims 1 to 8.

Images