Intelligent water valve self-adaptive flow adjusting method and related equipment

Abstract

The invention relates to an intelligent water valve adaptive flow adjusting method and related equipment, and the method comprises the following steps: carrying out real-time fluid parameter collection on an inlet end of a water valve through a multi-parameter fluid sensor array to obtain multi-dimensional fluid characteristic data; performing wavelet transform and spectral analysis on the multi-dimensional fluid characteristic data to obtain fluid dynamic characteristics; performing estimation and prediction based on the fluid dynamic characteristics to obtain a fluid behavior prediction matrix; the water valve is analyzed and decided based on the fluid behavior prediction matrix, and a water valve adjusting strategy is obtained; adjusting the opening and the flow of the water valve based on the water valve adjusting strategy, and solving the problems that a traditional water valve control method mostly depends on manual setting or a simple feedback mechanism, is difficult to deal with a complex and changeable fluid environment, and is difficult to deal with the conditions of flow sudden change, pressure fluctuation or water quality change and the like. The hysteresis quality and the instability are often shown.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of intelligent water valve, and particularly relates to an intelligent water valve adaptive flow regulation method and related equipment. BACKGROUND

[0002] In the fields of urban water supply systems, industrial fluid control, and smart home, water valves are key devices for regulating water flow, and their control accuracy and response speed directly affect the operation efficiency and energy consumption level of the system. Traditional water valve control methods rely heavily on manual setting or simple feedback mechanisms, which are difficult to cope with complex and variable fluid environments, especially when facing situations such as flow mutation, pressure fluctuation, or water quality change, often showing hysteresis and instability, limiting the overall performance improvement of the system. SUMMARY

[0003] The main purpose of the present application is to provide an intelligent water valve adaptive flow regulation method, which solves the technical problem that traditional water valve control methods rely heavily on manual setting or simple feedback mechanisms, which are difficult to cope with complex and variable fluid environments, especially when facing situations such as flow mutation, pressure fluctuation, or water quality change, often showing hysteresis and instability.

[0004] To achieve the above purpose, the present application provides an intelligent water valve adaptive flow regulation method, comprising the following steps: Real-time fluid parameter acquisition is performed on the inlet end of the water valve by a multi-parameter fluid sensor array to obtain multi-dimensional fluid characteristic data; Wavelet transform and spectral analysis are performed on the multi-dimensional fluid characteristic data to obtain fluid dynamic characteristics; Based on the fluid dynamic characteristics, estimation and prediction are performed to obtain a fluid behavior prediction matrix; Based on the fluid behavior prediction matrix, analysis and decision-making are performed on the water valve to obtain a water valve regulation strategy; Based on the water valve regulation strategy, the opening and flow of the water valve are adjusted.

[0005] Further, the wavelet transform and spectral analysis of the multi-dimensional fluid characteristic data to obtain fluid dynamic characteristics comprises: Multi-scale wavelet decomposition is performed on the multi-dimensional fluid characteristic data to obtain a multi-scale fluid characteristic coefficient matrix, and singular value decomposition is performed on the multi-scale fluid characteristic coefficient matrix to obtain a fluid characteristic principal component matrix; Based on the fluid characteristic principal component matrix, fast Fourier transform is performed to obtain a fluid frequency spectrum distribution map, and power spectral density estimation is performed on the fluid frequency spectrum distribution map to obtain a fluid power spectrum density curve; The fluid dynamic characteristics are obtained in combination with the multi-scale fluid characteristic coefficient matrix and the fluid power spectrum density curve.

[0006] Further, the estimation and prediction based on the fluid dynamic characteristics obtain a fluid behavior prediction matrix, comprising: The time series alignment processing is performed on the fluid dynamic characteristics to obtain a fluid time series alignment matrix, and the autocorrelation function calculation is performed on the fluid time series alignment matrix to obtain a fluid autocorrelation feature sequence; The multivariate state space reconstruction is performed on the fluid dynamic characteristics based on the fluid autocorrelation feature sequence to obtain a fluid state space reconstruction matrix, and the singular spectrum analysis is performed on the fluid state space reconstruction matrix to obtain a fluid singular spectrum feature vector; The fluid singular spectrum feature vector is input into a preset Gaussian process regression model for nonlinear mapping processing to obtain a fluid nonlinear mapping coefficient matrix, and the Bayesian probability prediction is performed based on the fluid nonlinear mapping coefficient matrix to obtain a fluid probability prediction distribution; The Monte Carlo simulation is performed on the fluid dynamic characteristics based on the fluid probability prediction distribution to obtain a fluid behavior prediction matrix.

[0007] Further, the analysis and decision of the water valve based on the fluid behavior prediction matrix obtain a water valve adjustment strategy, comprising: The pressure distribution analysis is performed on the water valve based on the fluid behavior prediction matrix to obtain a pressure gradient field distribution map, and the topological structure of the pressure gradient field distribution map is analyzed to obtain a flow field topological feature vector; The local turbulent flow region identification is performed on the water valve through the flow field topological feature vector to obtain a turbulent core region coordinate, and the boundary layer separation point positioning is performed on the water valve based on the turbulent core region coordinate to obtain a boundary layer separation point position; The spatial gradient operation is performed on the boundary layer separation point position to obtain a boundary layer separation gradient field, and the flow resistance coefficient distribution of the water valve is reconstructed based on the boundary layer separation gradient field to obtain a flow resistance coefficient distribution map; The opening-flow nonlinear mapping is performed on the water valve based on the flow resistance coefficient distribution map to obtain a valve core displacement-flow transfer function, and the optimal control point solving is performed on the water valve based on the valve core displacement-flow transfer function to obtain a water valve adjustment strategy.

[0008] Further, the local turbulent flow region identification of the water valve through the flow field topological feature vector obtains a turbulent core region coordinate, comprising: The curl field calculation is performed on the flow field topological feature vector to obtain a fluid curl distribution map, and the local vorticity density is counted based on the fluid curl distribution map to obtain a vorticity intensity; The singular point set of the flow field is obtained by classifying singular points of the vorticity intensity through critical point theory, and a singular point characteristic classification table is obtained by stability analysis on the singular point set of the flow field; The turbulent core area contour map is obtained by turbulent characteristic structure extraction on the water valve based on the singular point characteristic classification table, and connected domain analysis is performed on the turbulent characteristic marker sequence to obtain the turbulent core area contour map; The initial coordinate sequence is obtained by space coordinate transformation on the turbulent core area contour map, and the core area positioning calibration is performed based on the initial coordinate sequence to obtain the turbulent core area coordinates.

[0009] Further, the opening-flow nonlinear mapping of the water valve based on the flow resistance coefficient distribution map is performed to obtain the valve core displacement-flow transfer function, including: Local extreme points in the flow resistance coefficient distribution map are detected, and the resistance coefficient of the water valve is calculated based on the local extreme points; The valve core displacement sensitivity matrix is obtained by valve core displacement sensitivity analysis of the water valve based on the resistance coefficient, and the nonlinear mapping basis function is constructed based on the valve core displacement sensitivity matrix to obtain the nonlinear mapping basis function; The orthogonalization processing is performed on the nonlinear mapping basis function to obtain the orthogonalization basis function, and the flow response coefficient of the water valve is calculated based on the orthogonalization basis function; The valve core displacement-flow transfer function is solved based on the flow response coefficient of the water valve to obtain the valve core displacement-flow transfer function.

[0010] The application also provides an intelligent water valve adaptive flow regulation device, including: The acquisition module is used for real-time fluid parameter acquisition of the inlet end of the water valve through a multi-parameter fluid sensor array to obtain multi-dimensional fluid feature data; The analysis module is used for wavelet transform and spectrum analysis on the multi-dimensional fluid feature data to obtain fluid dynamic characteristics; The prediction module is used for estimation and prediction based on the fluid dynamic characteristics to obtain a fluid behavior prediction matrix; The decision module is used for analysis and decision on the water valve based on the fluid behavior prediction matrix to obtain a water valve regulation strategy; The regulation module is used for regulating the opening and flow of the water valve based on the water valve regulation strategy.

[0011] The application also provides a computer device including a memory and a processor, the memory stores a computer program, and the processor executes the computer program to realize the steps of the method.

[0012] The present invention also provides a computer-readable storage medium having a computer program stored thereon, wherein the computer program, when executed by a processor, implements the steps of any of the methods described above.

[0013] This invention provides an intelligent water valve adaptive flow regulation method, comprising the following steps: real-time fluid parameter acquisition at the inlet of the water valve using a multi-parameter fluid sensor array to obtain multi-dimensional fluid characteristic data; wavelet transform and spectral analysis of the multi-dimensional fluid characteristic data to obtain fluid dynamic characteristics; estimation and prediction based on the fluid dynamic characteristics to obtain a fluid behavior prediction matrix; analysis and decision-making on the water valve based on the fluid behavior prediction matrix to obtain a water valve regulation strategy; and adjustment of the opening degree and flow rate of the water valve based on the water valve regulation strategy. This method solves the technical problem that traditional water valve control methods rely heavily on manual settings or simple feedback mechanisms, making it difficult to cope with complex and changing fluid environments, especially when facing sudden changes in flow rate, pressure fluctuations, or water quality changes, often exhibiting lag and instability. It achieves the construction of a fluid behavior prediction matrix based on fluid dynamic characteristics, enabling trend estimation and short-term prediction of future fluid states, giving the control system a forward-looking regulation capability, thereby reducing control errors and energy waste caused by response lag. Attached Figure Description

[0014] Figure 1 This is a schematic diagram of the steps of an intelligent water valve adaptive flow regulation method in one embodiment of the present invention; Figure 2 This is a structural block diagram of an intelligent water valve adaptive flow regulation system according to an embodiment of the present invention; Figure 3 This is a schematic block diagram of the structure of a computer device according to an embodiment of the present invention.

[0015] The objectives, features, and advantages of this invention will be further explained in conjunction with the embodiments and with reference to the accompanying drawings. Detailed Implementation

[0016] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the invention.

[0017] like Figure 1 As shown, Figure 1 This invention provides an intelligent water valve adaptive flow regulation method, comprising the following steps: Step S1: Real-time fluid parameters are collected at the inlet of the water valve using a multi-parameter fluid sensor array to obtain multi-dimensional fluid characteristic data.

[0018] Specifically, the inlet end of the water valve is subjected to real-time fluid parameter collection by a multi-parameter fluid sensor array to obtain multi-dimensional fluid characteristic data. This process involves deploying a sensor array composed of multiple sensing units at the inlet end of the water valve. These sensors can simultaneously detect key physical parameters of the fluid such as pressure, temperature, flow rate, and conductivity, and continuously collect data at a high frequency, thereby forming a comprehensive perception of the fluid state. The core of this step lies in utilizing the multi-parameter collaborative perception capability of the sensor array to achieve non-invasive, high-precision, and multi-dimensional data acquisition of the fluid characteristics at the inlet end without changing the fluid transmission path. For example, in a city water supply system, when the water flow passes through the inlet of the water valve, the pressure sensor in the sensor array can monitor the water pressure fluctuations in real time, the flow sensor records the instantaneous flow rate changes, the temperature sensor perceives water temperature anomalies, and the conductivity sensor can be used to determine whether the water quality has changed suddenly, such as pollution or the mixing of impurities. All these parameters are collected and transmitted to the subsequent processing module to provide the original data basis for subsequent wavelet transform and spectral analysis, thereby supporting the implementation of the entire adaptive adjustment process.

[0019] Step S2, wavelet transform and spectral analysis are performed on the multi-dimensional fluid characteristic data to obtain fluid dynamic characteristics.

[0020] Specifically, after obtaining the inlet fluid data collected by the multi-parameter fluid sensor array, wavelet transform technology is used to perform multi-scale decomposition of these non-stationary, high-dimensional data in the time-frequency domain, thereby revealing their local characteristics at different time scales. Combined with spectral analysis methods, the main frequency components and periodicity rules existing in the fluid parameter changes are identified, and a comprehensive characteristic map reflecting the dynamic behavior of the fluid is constructed, i.e., the fluid dynamic characteristics. This process can effectively capture the transient changes and potential fluctuation trends of the fluid state, providing reliable feature basis for subsequent estimation and prediction. For example, in the actual application of a city water supply system, when sudden pressure pulsations or flow oscillations occur in the pipe network, wavelet transform can extract the energy distribution characteristics of these abnormal signals within different time windows, and spectral analysis can identify whether there are periodic disturbances, such as harmonic components caused by water pump start-stop. Through the joint processing of these two methods, the system can more accurately depict the current dynamic behavior pattern of the fluid, laying a solid data foundation for generating a fluid behavior prediction matrix in the next step, thereby ensuring the scientificity and real-time nature of the water valve adjustment strategy.

[0021] Step S3, based on the fluid dynamic characteristics, estimation and prediction are performed to obtain a fluid behavior prediction matrix.

[0022] Specifically, based on the fluid dynamic characteristics for estimation and prediction, the fluid behavior prediction matrix refers to, after completing wavelet transform and spectrum analysis on multi-dimensional fluid characteristic data, using the obtained fluid dynamic characteristics as input features, estimating and predicting the state change trend of the fluid in the future period of time through mathematical modeling and intelligent algorithm, thereby generating a fluid behavior prediction matrix containing multiple prediction variables and their correlation. The core of this step is to combine the historical dynamic characteristics of the fluid with the current state to build a prediction model with time evolution capability, so that the system has forward-looking regulation capability. For example, in the application background of urban water supply system, when the system detects the pressure drop trend caused by the sudden increase of user water consumption, through analysis and modeling of fluid dynamic characteristics, the change amplitude of flow and pressure in the next few minutes can be predicted, and these prediction results are organized into a structured fluid behavior prediction matrix to guide the formulation of subsequent water valve regulation strategy. This process not only improves the response speed and regulation accuracy of the control system, but also provides a scientific decision basis for realizing adaptive regulation of water valves.

[0023] Step S4, analyzing and deciding the water valve based on the fluid behavior prediction matrix to obtain a water valve regulation strategy.

[0024] Specifically, based on the fluid behavior prediction matrix for analyzing and deciding the water valve to obtain a water valve regulation strategy, refers to, after obtaining the fluid behavior prediction matrix generated by the estimation and prediction of fluid dynamic characteristics, using the future fluid state change trend information contained in the matrix, combining the control characteristics and operating constraints of the water valve, optimizing and analyzing the opening regulation path of the water valve through intelligent decision algorithm, thereby generating an optimal regulation strategy suitable for the current and future fluid working conditions. This process emphasizes that from the perspective of data-driven, the fluid behavior prediction result is directly converted into executable control instructions, realizing the closed-loop adaptive regulation mechanism from perception, analysis to control. For example, in the actual operation of urban water supply system, when the fluid behavior prediction matrix shows that the trend of pressure drop at the inlet and flow demand rise due to the upcoming water consumption peak, the system can automatically adjust the valve opening angle to increase the water supply capacity in advance, thereby avoiding the problem of insufficient pipe network pressure or water supply interruption caused by response lag. This active regulation method based on prediction significantly improves the intelligent level of water valve control and the stability of system operation.

[0025] Step S5, adjusting the opening and flow of the water valve based on the water valve regulation strategy.

[0026] Specifically, adjusting the opening degree and flow of the water valve based on the water valve adjustment strategy means that after the analysis and decision of the water valve control behavior are completed, the mechanical opening and closing components of the water valve are accurately driven by the actuator according to the generated water valve adjustment strategy, so as to realize the dynamic adjustment of the opening degree of the water valve, and finally achieve the purpose of real-time and self-adaptive control of the flow through the water valve. This step is the execution terminal of the entire intelligent regulation process, and its core is to convert the control instructions obtained from the multi-dimensional data perception, dynamic characteristic analysis, behavior prediction and strategy generation in the previous steps into actual physical actions, to ensure that the water valve can actively adapt to the changing trend of the inlet fluid state according to the prediction results, and maintain the stability and efficiency of the system operation. For example, in the application of urban water supply system, when the water valve adjustment strategy indicates that the water supply flow needs to be increased to cope with the upcoming water peak, the actuator will accurately adjust the valve opening angle according to the strategy requirements, so that the flow increases and stabilizes in the target range, thereby avoiding the phenomenon of insufficient pressure or flow interruption at the end of the pipe network, and realizing the closed-loop response from data analysis to actual control.

[0027] In specific embodiments, the wavelet transform and spectral analysis of the multi-dimensional fluid characteristic data to obtain fluid dynamic characteristics include: Performing multi-scale wavelet decomposition on the multi-dimensional fluid characteristic data to obtain a multi-scale fluid characteristic coefficient matrix, and performing singular value decomposition on the multi-scale fluid characteristic coefficient matrix to obtain a fluid characteristic principal component matrix; Performing fast Fourier transform based on the fluid characteristic principal component matrix to obtain a fluid frequency spectrum distribution map, and performing power spectral density estimation on the fluid frequency spectrum distribution map to obtain a fluid power spectrum density curve; Combining the multi-scale fluid characteristic coefficient matrix and the fluid power spectrum density curve to obtain fluid dynamic characteristics, wherein the fluid dynamic characteristics include fluid pulsation frequency, turbulence intensity, pressure fluctuation amplitude and flow field structure characteristics.

[0028] Specifically, the process of wavelet transform and spectral analysis on the multi-dimensional fluid characteristic data to obtain the fluid dynamic characteristics is the key link for realizing high-precision fluid state recognition and feature extraction in the whole intelligent water valve adaptive flow regulation method. Firstly, the multi-scale wavelet decomposition is performed on the multi-dimensional fluid characteristic data collected by the multi-parameter fluid sensor array to obtain a multi-scale fluid characteristic coefficient matrix that can reflect the local dynamic behavior of the fluid at different time scales. The core of this step is to effectively capture the non-stationary components in the fluid signal, such as transient fluctuations, periodic disturbances, or sudden events, by using the good time-frequency localization ability of wavelet transform, providing rich detail information for subsequent analysis. Subsequently, singular value decomposition is performed on the multi-scale fluid characteristic coefficient matrix to further compress the data dimension and extract the main energy distribution pattern contained therein, forming a fluid characteristic principal component matrix. This process not only reduces data redundancy but also retains the most representative fluid state characteristics, making the subsequent spectral analysis more targeted and accurate. On this basis, the system performs fast Fourier transform based on the fluid characteristic principal component matrix to identify the distribution rule of each principal component in the frequency domain, and then constructs a fluid spectral distribution map. Further, the power spectral density estimation technique is used to extract the energy intensity of each frequency component from the fluid spectral distribution map to generate a fluid power spectral density curve. Finally, the time scale evolution information revealed by the multi-scale fluid characteristic coefficient matrix and the frequency energy distribution information reflected by the fluid power spectral density curve are combined to comprehensively construct a fluid dynamic characteristic that can fully describe the dynamic behavior of the fluid. This characteristic spectrum not only contains key dynamic parameters such as fluid pulsation frequency, turbulence intensity, and pressure fluctuation amplitude, but also reflects the trend of the change in the flow field structure, providing high-quality input features for subsequent prediction models. For example, in the actual operation of the urban water supply system, when the water pump starts and stops causing frequent disturbance of the water flow in the pipeline, this method can accurately identify the high-frequency pulsation and low-frequency oscillation components in the fluid, helping the system to determine whether it is in an unstable flow state, thereby providing reliable data support and theoretical basis for the formulation of water valve regulation strategies. This technical path that combines multi-scale time-frequency analysis and statistical spectral modeling significantly improves the system's perception ability and response accuracy for complex fluid behavior.

[0029] In specific embodiments, the estimating and predicting based on the fluid dynamic characteristics to obtain a fluid behavior prediction matrix comprises: performing time series alignment processing on the fluid dynamic characteristics to obtain a fluid time series alignment matrix, and performing autocorrelation function calculation on the fluid time series alignment matrix to obtain a fluid autocorrelation feature sequence; Based on the fluid autocorrelation feature sequence, the fluid dynamic characteristics are reconstructed in a multivariate state space to obtain a fluid state space reconstruction matrix. Singular spectrum analysis is then performed on the fluid state space reconstruction matrix to obtain fluid singular spectrum feature vectors. The fluid singular spectrum feature vector is input into a preset Gaussian process regression model for nonlinear mapping processing to obtain the fluid nonlinear mapping coefficient matrix. Based on the fluid nonlinear mapping coefficient matrix, Bayesian probability prediction is performed to obtain the fluid probability prediction distribution. Monte Carlo simulations are performed on the fluid dynamic characteristics based on the fluid probability prediction distribution to obtain a fluid behavior prediction matrix. The fluid behavior prediction matrix includes the flow rate change trend, pressure fluctuation pattern, temperature change curve, and predicted water quality parameters for the next 30 seconds to 5 minutes.

[0030] Specifically, the process of estimating and predicting fluid behavior based on its dynamic characteristics to obtain a fluid behavior prediction matrix is ​​a complex and multi-layered analytical process closely centered around these characteristics. First, time series alignment processing is performed on the fluid dynamic characteristics. This is a crucial step because it ensures that data acquired at different time intervals can be accurately correlated, thus forming a fluid time series alignment matrix. This process eliminates errors caused by asynchronous acquisition devices or inconsistent data recording time intervals. For example, in urban water supply systems, to accurately assess changes in the state of water flow within pipes, data from different sensor nodes need to be time-synchronized so that parameters such as flow rate, pressure, and temperature at each time point accurately reflect the fluid state at the same moment. Next, the autocorrelation function is calculated on the fluid time series alignment matrix. This step aims to reveal the inherent connections and regularities in the changes of fluid characteristics over time, i.e., obtaining the fluid autocorrelation feature sequence. By analyzing these autocorrelation feature sequences, the periodic and non-periodic components of the fluid at different time scales can be identified, which is crucial for understanding the dynamic behavior of fluids. For example, in monitoring a water supply system, autocorrelation analysis might reveal pressure fluctuation peaks occurring every 24 hours, closely related to peak water usage periods in cities. Subsequently, based on the fluid autocorrelation feature sequence, multivariate state-space reconstruction of the fluid dynamics is performed. This process essentially extends the original single-dimensional or low-dimensional fluid characteristic information into a high-dimensional space to more comprehensively capture the dynamic characteristics of fluid behavior, thus obtaining a fluid state-space reconstruction matrix. This step provides a richer data foundation for subsequent analysis. For instance, when analyzing a water supply system, in addition to considering changes in flow rate and pressure, multiple factors such as temperature and water quality should be considered to construct a multi-dimensional state-space model for a more accurate description of the entire system's operating state. Next, singular spectrum analysis is performed on the fluid state-space reconstruction matrix to extract key fluid singular spectrum feature vectors. Singular spectrum analysis is an effective signal processing method that helps extract key components from complex multivariate data, simplifying subsequent predictive model design. In this example, singular spectrum analysis helps identify the main factors affecting the stability of the water supply system and quantifies their respective influence. Then, the fluid singular spectrum eigenvectors are input into a pre-defined Gaussian process regression model for nonlinear mapping processing to obtain the fluid nonlinear mapping coefficient matrix. The Gaussian process regression model is used here because it excels at handling nonlinear relationships and provides uncertainty estimation, which is particularly important for predicting future behavior. In the water supply system case, the Gaussian process regression model allows us to predict flow rate trends and pressure fluctuations over the next 30 seconds to 5 minutes.Finally, Bayesian probabilistic prediction is performed based on the fluid nonlinear mapping coefficient matrix to obtain the fluid probability prediction distribution. Monte Carlo simulations are then conducted on the fluid dynamics based on this distribution, ultimately yielding the fluid behavior prediction matrix. Bayesian probabilistic prediction allows us to consider all possible scenarios and their probabilities, while Monte Carlo simulation uses random sampling to simulate fluid behavior under various possibilities, thus generating a series of possible future scenarios. For water supply systems, this means we can not only predict future flow and pressure changes but also assess the range of water quality parameter variations, providing strong support for system maintenance and optimization. In this way, the entire process not only deepens our understanding of fluid behavior but also provides a scientific basis for decision-making in practical applications.

[0031] In a specific embodiment, the step of analyzing and making decisions about the water valve based on the fluid behavior prediction matrix to obtain a water valve regulation strategy includes: Based on the fluid behavior prediction matrix, pressure distribution analysis is performed on the water valve to obtain a pressure gradient field distribution map, and the topological structure of the pressure gradient field distribution map is analyzed to obtain the flow field topological feature vector. The local turbulent region of the water valve is identified by the flow field topology feature vector to obtain the coordinates of the turbulent core region. Based on the coordinates of the turbulent core region, the boundary layer separation point of the water valve is located to obtain the position of the boundary layer separation point. Spatial gradient calculation is performed on the location of the boundary layer separation point to obtain the boundary layer separation gradient field, and the flow resistance coefficient distribution of the water valve is reconstructed based on the boundary layer separation gradient field to obtain the flow resistance coefficient distribution map; Based on the flow resistance coefficient distribution diagram, the opening-flow nonlinear mapping of the water valve is performed to obtain the valve core displacement-flow transfer function. Based on the valve core displacement-flow transfer function, the optimal control point of the water valve is solved to obtain the water valve regulation strategy.

[0032] Specifically, the process of analyzing and deciding on the water valve based on the fluid behavior prediction matrix to obtain the water valve regulation strategy is a key step in transforming the future fluid state prediction information obtained in the preceding steps into specific control actions. This process first models the dynamic pressure distribution inside and around the water valve based on the fluid behavior prediction matrix, thereby generating a pressure gradient field distribution map. This modeling process relies on pressure field reconstruction algorithms in fluid mechanics and, combined with the water valve structural parameters, maps information such as future flow rate and pressure change trends from the fluid behavior prediction matrix onto the valve cavity, forming a spatial pressure gradient distribution, providing a physical basis for subsequent topological analysis. After obtaining the pressure gradient field distribution map, the system further analyzes its topological structure, extracting pressure extreme points, saddle points, and their connecting paths, thereby obtaining a flow field topological feature vector reflecting the characteristics of the water flow structure. This step, using topological identification methods in computational fluid dynamics, can effectively characterize the mainstream direction, vortex structure, and local pressure anomaly areas flowing through the water valve region, providing a basis for identifying the turbulent core region. For example, in the actual operation of urban water supply systems, when pressure fluctuations or sudden increases in flow occur in the pipeline, local turbulence or backflow may occur inside the water valve. These abnormal areas can be accurately located through the analysis of flow field topological feature vectors. Next, the flow field topological feature vectors are used to identify the local turbulence zone of the water valve, determining the coordinates of the turbulence core region. These coordinates represent the location inside the water valve most prone to energy loss and flow instability. Subsequently, based on these turbulence core region coordinates, the system further analyzes the boundary layer behavior near the water valve wall, identifying boundary layer separation points and constructing their locations. This process integrates boundary layer theory and numerical simulation techniques, accurately capturing boundary layer detachment caused by changes in flow velocity, which in turn affects the valve's flow performance and regulation accuracy. Based on the obtained boundary layer separation point locations, the system performs spatial gradient calculations to quantify the pressure gradient changes between each separation point, thereby constructing a boundary layer separation gradient field. This gradient field reflects the spatial distribution characteristics of fluid resistance inside the water valve and is an important input for subsequent flow resistance coefficient reconstruction. Based on this, the system reconstructs the flow resistance coefficient distribution of the water valve using the boundary layer separation gradient field, generating a flow resistance coefficient distribution map. This map details the flow resistance characteristics of different regions of the water valve under different opening degrees, providing crucial model support for achieving accurate opening-flow rate matching. Finally, the system performs an opening-flow rate nonlinear mapping analysis on the water valve based on the flow resistance coefficient distribution map, establishing the transfer function relationship between valve core displacement and flow rate, i.e., the valve core displacement-flow rate transfer function. This function comprehensively considers the influence of the water valve structure, fluid properties, and boundary conditions, exhibiting highly nonlinear characteristics, and is typically modeled using empirical formulas or neural network fitting methods.Based on this, the system further solves for the optimal control point of the transfer function. Combining the flow demand and pressure constraints in the current fluid behavior prediction matrix, it calculates the optimal water valve opening adjustment scheme, thereby generating a water valve regulation strategy. For example, in an urban water supply system, when the system predicts an upcoming peak water consumption period leading to a decrease in inlet pressure and an increase in flow demand, the above process can identify the turbulent core region and boundary layer separation region that may appear inside the water valve. This allows for the reconstruction of its flow resistance coefficient distribution, evaluation of flow efficiency and energy consumption levels at different opening degrees, and ultimately determination of the optimal valve core displacement adjustment path. This enables the valve opening to be increased in advance to meet the expected flow demand, avoiding insufficient network pressure or flow interruption caused by response lag. This intelligent decision-making mechanism, driven by flow field topology and boundary layer behavior, not only improves the scientific and real-time nature of water valve regulation but also provides a solid guarantee for the stable operation of the entire water supply system.

[0033] In a specific embodiment, the step of performing pressure distribution analysis on the water valve based on the fluid behavior prediction matrix to obtain a pressure gradient field distribution map includes: The future pressure fluctuation pattern in the fluid behavior prediction matrix is ​​coupled temporally and spatially to obtain the transient pressure field evolution sequence. The transient pressure field evolution sequence is then subjected to local weighted non-uniform interpolation to obtain a continuous pressure distribution field. The Navier-Stokes equations in the internal flow channel structure of the water valve are numerically discretized and solved under boundary constraints by the continuous pressure distribution field to obtain the nodal pressure field distribution data. Based on the nodal pressure field distribution data, the differential gradient operator is operated to obtain the three-dimensional pressure gradient tensor matrix. Based on the three-dimensional pressure gradient tensor matrix, the flow channel cross section of the water valve is sliced ​​and projected to obtain multiple cross-sectional pressure gradient vectors. The multiple cross-sectional pressure gradient vectors are then normalized to ensure consistency of the principal direction to obtain a standardized pressure gradient vector field. The standardized pressure gradient vector field is reconstructed in manifold space and subjected to topological adjacency analysis to obtain a cluster of pressure gradient manifold connection paths. Based on the cluster of pressure gradient manifold connection paths, regional connectivity clustering is performed to obtain a distribution map of the pressure gradient field.

[0034] Specifically, the process of analyzing the pressure distribution of the water valve based on the fluid behavior prediction matrix to obtain a pressure gradient field distribution map is a key technical step in transforming the future fluid state prediction information (especially pressure fluctuation patterns) obtained in the preceding steps into the three-dimensional spatial pressure evolution characteristics inside the water valve. This process first performs time-domain-spatial coupling mapping on the future pressure fluctuation patterns in the fluid behavior prediction matrix to establish a mapping relationship between the pressure change trend in the time dimension and the geometric shape of the water valve structure in the spatial dimension, thereby generating a transient pressure field evolution sequence. This step relies on spatiotemporal interpolation algorithms in numerical simulation, which can extend the predicted time series data to the entire internal space of the water valve, realizing the transition from "point prediction" to "field evolution". Subsequently, to improve the spatial continuity and physical rationality of the pressure field, the system performs local weighted non-uniform interpolation on the transient pressure field evolution sequence, fully considering the influence of the internal flow channel geometry, boundary conditions, and material properties on pressure propagation, thereby constructing a high-resolution continuous pressure distribution field. This interpolation process employs a spatial reconstruction strategy based on the finite element method or finite volume method, which can effectively restore the real pressure distribution under complex flow channel structures while ensuring computational efficiency. For example, in the actual operation of urban water supply systems, when pump switching causes sudden changes in pipeline pressure, this method can accurately capture the pressure response time and amplitude differences at various locations inside the water valve, providing high-quality input for subsequent gradient analysis. After completing the construction of the continuous pressure distribution field, the system further uses this pressure field to numerically discretize and solve the Navier-Stokes equations under boundary constraints for the internal flow channel structure in the water valve, thereby obtaining nodal pressure field distribution data. This solution process combines CFD (Computational Fluid Dynamics) modeling methods with a real-time boundary condition update mechanism, which can simulate the flow state of the fluid inside the water valve under different operating conditions and output the pressure value at each grid node. Next, the system performs differential gradient operator operations based on these node pressure data to calculate the pressure gradient components in each direction, thus forming a three-dimensional pressure gradient tensor matrix. This matrix not only reflects the spatial rate of change of fluid pressure inside the water valve but also contains information about the fluid movement direction and energy transfer path. To further extract the structural features of the pressure gradient field, the system performs a slice projection transformation on the flow channel cross-section of the water valve based on the three-dimensional pressure gradient tensor matrix. This involves projecting the three-dimensional pressure gradient tensor into a two-dimensional plane along a specific direction (such as axial, radial, or circumferential), thereby obtaining the pressure gradient vector distribution on multiple cross-sections. This process is similar to tomographic reconstruction in medical imaging, helping to visually display the pressure gradient characteristics of key areas inside the water valve. Subsequently, the system performs principal direction consistency normalization processing on the pressure gradient vectors of the multiple cross-sections to eliminate directional deviations caused by coordinate system differences between different cross-sections, forming a standardized pressure gradient vector field, making subsequent analyses more comparable and consistent.Finally, the system performs manifold spatial reconstruction and topological adjacency analysis on the standardized pressure gradient vector field, identifying the main manifold structures (such as vortices and shear layers) and constructing pressure gradient manifold connection path clusters. Based on this, further regional connectivity clustering analysis is performed on these path clusters to identify spatial regions with similar pressure gradient characteristics, thus generating the final pressure gradient field distribution map. This distribution map not only reveals the spatial distribution law of the pressure gradient inside the water valve but also provides a physical basis for subsequent analyses such as turbulence core region identification and boundary layer separation point location. For example, in practical applications of urban water supply systems, when the system predicts a trend of pressure drop at the inlet due to peak user water consumption, this method can accurately depict the pressure gradient changes in various parts of the water valve cavity, identify regions that may cause local backflow or enhanced turbulence, and provide a scientific basis for optimizing valve opening adjustment strategies. This technical approach, which integrates fluid behavior prediction and three-dimensional pressure field reconstruction, significantly improves the intelligence level and adjustment accuracy of the water valve control system.

[0035] In a specific embodiment, the step of identifying the local turbulent region of the water valve using the flow field topological feature vector to obtain the coordinates of the turbulent core region includes: The topological feature vector of the flow field is used to calculate the curl field to obtain the fluid curl distribution map, and the local vorticity density is statistically calculated based on the fluid curl distribution map to obtain the vorticity intensity. The vorticity intensity is classified into singularities using critical point theory to obtain a set of flow field singularities. Stability analysis is then performed on the set of flow field singularities to obtain a singularity feature classification table. Based on the singularity feature classification table, the turbulence feature structure of the water valve is extracted to obtain a turbulence feature label sequence, and the turbulence feature label sequence is analyzed to obtain a contour map of the turbulence core region. The spatial coordinates of the outline map of the turbulent core region are transformed to obtain an initial coordinate sequence, and the core region positioning and calibration are performed based on the initial coordinate sequence to obtain the coordinates of the turbulent core region.

[0036] Specifically, the process of identifying the local turbulent region of the water valve and obtaining the coordinates of the turbulent core region through the flow field topological feature vector is a key technical step in further transforming the flow field structural features extracted in the previous steps into specific physical region localization. This process first calculates the curl field based on the flow field topological feature vector, that is, using the curl operation in the vector differential operator to extract the spatial distribution of fluid rotation intensity from the pressure gradient field and velocity field distribution, thereby constructing a fluid curl distribution map. This step reveals the existence of vortex structures and their intensity variation trends in the fluid motion inside the water valve, providing basic data for subsequent vorticity density statistics. After obtaining the fluid curl distribution map, the system performs statistical analysis on the vortex intensity in local areas based on the map, calculating the concentration of vorticity per unit volume at each spatial location, and thus generating vorticity intensity. This matrix reflects the flow stability and energy dissipation level in different regions inside the water valve, and is a direct basis for determining whether turbulence has occurred. For example, in the actual operation of urban water supply systems, when valve opening changes abruptly or flow fluctuates significantly, local vortices easily form near the valve core, leading to increased energy loss and potentially causing vibration or noise. These high-vorticity regions can be clearly identified through vorticity intensity. Next, the system performs singularity classification on the vorticity intensity using critical point theory, mathematically identifying and categorizing extreme points (such as source points, sink points, and saddle points) in the vorticity distribution to obtain a set of flow field singularities. This process relies on phase plane analysis methods in nonlinear dynamics, accurately capturing key points with significant structural meaning in complex flow fields. Subsequently, the system performs stability analysis on the set of flow field singularities, evaluating the response characteristics of each singularity under disturbance, classifying them into different types such as stable foci, unstable foci, and center points, and establishing a singularity feature classification table. This classification provides a clear physical basis for subsequent turbulent structure identification. Based on this, the system extracts turbulent feature structures from the water valve using the singularity feature classification table, identifies high vorticity regions dominated by unstable singularities, and marks them as potential turbulence occurrence regions, thus generating a turbulence feature label sequence. This sequence digitally records whether the conditions for turbulence formation exist at various locations inside the water valve, laying the foundation for further spatial clustering analysis. Next, the system performs connected component analysis on the turbulence feature label sequence, identifies continuously existing high vorticity regions, and extracts their boundary contours, ultimately forming a turbulence core region contour map. This contour map visually shows the specific areas inside the water valve where severe flow disturbances may occur, providing geometric support for the next step of coordinate positioning.Finally, to achieve precise positioning of the turbulent core region, the system performs spatial coordinate transformation on the contour map of the turbulent core region, converting the contour information in the image coordinate system into a three-dimensional spatial coordinate sequence in the water valve structure coordinate system, thereby obtaining an initial coordinate sequence. Since image processing may introduce certain errors, the system also needs to perform core region positioning calibration based on the initial coordinate sequence. Combining the geometric model of the water valve and the flow field boundary conditions, the system corrects coordinate offsets caused by interpolation errors or modeling deviations, ultimately outputting accurate coordinates of the turbulent core region. These coordinates can not only be used for subsequent boundary layer separation point identification but also serve as an important reference for optimizing the water valve structure design or adjusting control strategies. For example, in urban water supply systems, when water pumps start and stop frequently or uneven water usage at the user end causes significant fluctuations in pipeline pressure, localized turbulence can easily occur inside the water valve, affecting water supply efficiency and equipment lifespan. Through the above process, the system can identify the location of the turbulent core region within the water valve cavity in real time and optimize the valve opening angle accordingly, reducing energy loss and flow noise, thereby improving the overall system's operational stability and energy efficiency. This intelligent analysis method, which combines flow field topology features with turbulence identification mechanisms, effectively enhances the scientific rigor and engineering practicality of the water valve adaptive control system.

[0037] In a specific embodiment, the step of performing an opening-flow nonlinear mapping on the water valve based on the flow resistance coefficient distribution map to obtain the valve core displacement-flow transfer function includes: Detect local extreme points in the flow resistance coefficient distribution diagram, and calculate the resistance coefficient of the water valve based on the local extreme points; The valve core displacement sensitivity analysis is performed on the water valve using the resistance coefficient to obtain the valve core displacement sensitivity matrix. Based on the valve core displacement sensitivity matrix, a nonlinear mapping basis function is constructed for the water valve to obtain the nonlinear mapping basis function. The nonlinear mapping basis function is orthogonalized to obtain orthogonalized basis functions, and the flow response coefficient of the water valve is calculated based on the orthogonalized basis functions. Based on the flow response coefficient, the valve core displacement-flow transfer function of the water valve is solved to obtain the valve core displacement-flow transfer function.

[0038] Specifically, the process of performing an opening-flow nonlinear mapping on the water valve based on the flow resistance coefficient distribution map to obtain the valve core displacement-flow transfer function is a key step in transforming the spatial distribution information of the internal flow resistance of the water valve constructed in the previous steps into a mathematical model that can be used for control and regulation. This process first detects local extreme points in the flow resistance coefficient distribution map. These extreme points represent the locations of the maximum or minimum resistance to fluid flow in different regions inside the water valve at a specific opening, and have significant physical meaning. Identifying these key points provides an accurate data foundation for subsequent gradient field calculations. Subsequently, the system performs resistance coefficient gradient field calculations on the water valve based on these local extreme points, that is, it uses numerical differentiation methods to analyze the spatial variation trend of the flow resistance coefficient, thereby generating the resistance coefficient. This map not only reveals the differences in the strength of the influence of different regions inside the water valve on fluid flow, but also reflects the abrupt changes in local resistance caused by factors such as structural design, flow channel geometry, or boundary layer behavior, serving as an important basis for evaluating the regulating performance of the water valve. Based on this, the system further performs valve core displacement sensitivity analysis on the water valve using the resistance coefficient. This involves simulating the impact of valve core movement on the overall flow resistance distribution under different opening degrees and extracting its rate of change characteristics to form a valve core displacement sensitivity matrix. This matrix describes the response characteristics of the water valve at different opening degrees and is the core input data for establishing a nonlinear mapping relationship. For example, in the actual operation of an urban water supply system, when the valve is at a small opening degree, changes in flow resistance may have a drastic impact on the flow rate. However, as the opening degree increases, this impact tends to level off. Therefore, sensitivity analysis is needed to capture this nonlinear change pattern. Next, the system constructs nonlinear mapping basis functions for the water valve based on the valve core displacement sensitivity matrix. This involves fitting the sensitivity data into a set of mathematical functions that can reflect the complex relationship between opening degree and flow rate, forming a nonlinear mapping basis function. This set of basis functions typically includes polynomial terms, exponential terms, or radial basis functions, used to approximate the nonlinear mapping relationship between the complex flow behavior and control response within the water valve. To improve mapping accuracy and generalization ability, the system also needs to orthogonalize the nonlinear mapping basis functions, removing redundant or highly correlated components to obtain orthogonalized basis functions. This process helps improve the stability and convergence speed of subsequent parameter estimation. Subsequently, the system calculates the flow response coefficient of the water valve based on the orthogonalized basis functions, i.e., by solving for the weights of each basis function using experimental data or simulation results, enabling the entire mapping model to accurately predict the flow output under different valve core displacements. This coefficient matrix is ​​the core parameter for constructing the final transfer function, directly determining the model's prediction accuracy and control effect.Finally, the system solves for the valve core displacement-flow transfer function based on the flow response coefficient, establishing a mathematical expression with valve core displacement as input and flow rate as output, forming the valve core displacement-flow transfer function. This function comprehensively considers the valve's geometry, fluid properties, boundary conditions, and nonlinear response characteristics, reflecting the actual impact of different opening adjustments on flow rate in real time, and serves as the fundamental model for achieving precise control. For example, in an urban water supply system, when the system predicts an increase in water consumption at the user end and needs to increase the valve opening in advance, the valve core displacement-flow transfer function constructed using the above process can accurately calculate the required adjustment value to ensure a smooth flow transition to the target level, avoiding pressure fluctuations or energy waste caused by over-adjustment or under-adjustment. This intelligent control strategy, based on a combination of flow resistance coefficient distribution and nonlinear mapping modeling, effectively improves the dynamic response capability and control accuracy of the water valve adaptive adjustment system, providing a strong guarantee for the efficient and stable operation of the water supply system.

[0039] The above describes the intelligent water valve adaptive flow regulation method in the embodiments of the present invention. The following describes the intelligent water valve adaptive flow regulation system in the embodiments of the present invention. Please refer to [link / reference]. Figure 2 One embodiment of the intelligent water valve adaptive flow regulation system of the present invention includes: The acquisition module 21 is used to acquire fluid parameters in real time at the inlet of the water valve through a multi-parameter fluid sensor array to obtain multi-dimensional fluid characteristic data. Analysis module 22 is used to perform wavelet transform and spectral analysis on the multi-dimensional fluid feature data to obtain fluid dynamic characteristics; Prediction module 23 is used to estimate and predict based on the fluid dynamic characteristics to obtain a fluid behavior prediction matrix; Decision module 24 is used to analyze and make decisions on the water valve based on the fluid behavior prediction matrix to obtain a water valve regulation strategy; The adjustment module 25 is used to adjust the opening degree and flow rate of the water valve based on the water valve adjustment strategy.

[0040] In this embodiment, the specific implementation of each unit in the above device embodiment is described in the above method embodiment, and will not be repeated here.

[0041] Reference Figure 3 This invention also provides a computer device whose internal structure can be as follows: Figure 3As shown, the computer device includes a processor, memory, display screen, input device, network interface, and database connected via a system bus. The processor provides computing and control capabilities. The memory includes a non-volatile storage medium and internal memory. The non-volatile storage medium stores the operating system, computer programs, and database. The internal memory provides an environment for the operation of the operating system and computer programs in the non-volatile storage medium. The database stores the data corresponding to this embodiment. The network interface is used to communicate with external terminals via a network connection. When the computer program is executed by the processor, it implements the above-described method.

[0042] Those skilled in the art will understand that Figure 3 The structures shown are merely block diagrams of some structures related to the present invention and do not constitute a limitation on the computer devices on which the present invention is applied.

[0043] An embodiment of the present invention also provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the above-described method. It is understood that the computer-readable storage medium in this embodiment can be a volatile readable storage medium or a non-volatile readable storage medium.

[0044] Those skilled in the art will understand that all or part of the processes in the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium. When executed, the computer program can include the processes of the embodiments of the above methods. Any references to memory, storage, databases, or other media used in the present invention and embodiments can include non-volatile and / or volatile memory. Non-volatile memory can include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM), or flash memory. Volatile memory can include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in various forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), dual-rate SDRAM (SSRSDRAM), enhanced SDRAM (ESDRAM), synchronous link DRAM (SLDRAM), Rambus direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM, etc.

[0045] It should be noted that, in this document, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, apparatus, article, or method that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such process, apparatus, article, or method. Unless otherwise specified, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, apparatus, article, or method that includes that element.

[0046] The above description is only a preferred embodiment of the present invention and does not limit the patent scope of the present invention. Any equivalent structural or procedural transformations made based on the content of the present invention specification and drawings, or direct or indirect applications in other related technical fields, are similarly included within the patent protection scope of the present invention.

Claims

1. A method for adaptive flow regulation of an intelligent water valve, characterized in that, Includes the following steps: Real-time fluid parameters are acquired at the inlet of the water valve using a multi-parameter fluid sensor array to obtain multi-dimensional fluid characteristic data. Wavelet transform and spectral analysis are performed on the multi-dimensional fluid characteristic data to obtain the fluid dynamic characteristics; Based on the fluid dynamic characteristics, an estimation and prediction are performed to obtain a fluid behavior prediction matrix; Based on the fluid behavior prediction matrix, the water valve is analyzed and decisions are made to obtain the water valve regulation strategy; The opening degree and flow rate of the water valve are adjusted based on the aforementioned water valve adjustment strategy.

2. The intelligent water valve adaptive flow regulation method according to claim 1, characterized in that, The process of performing wavelet transform and spectral analysis on the multi-dimensional fluid feature data to obtain fluid dynamic characteristics includes: Multi-scale wavelet decomposition is performed on the multi-dimensional fluid feature data to obtain a multi-scale fluid feature coefficient matrix, and singular value decomposition is performed on the multi-scale fluid feature coefficient matrix to obtain a fluid feature principal component matrix. Based on the fluid characteristic principal component matrix, a fast Fourier transform is performed to obtain the fluid spectrum distribution map, and the power spectral density of the fluid spectrum distribution map is estimated to obtain the fluid power spectral density curve. By combining the multi-scale fluid characteristic coefficient matrix and the fluid power spectral density curve, the fluid dynamic characteristics are obtained.

3. The intelligent water valve adaptive flow regulation method according to claim 1, characterized in that, The estimation and prediction based on the fluid dynamic characteristics to obtain the fluid behavior prediction matrix includes: The fluid dynamic characteristics are subjected to time series alignment processing to obtain a fluid time series alignment matrix, and the autocorrelation function is calculated on the fluid time series alignment matrix to obtain a fluid autocorrelation feature sequence; Based on the fluid autocorrelation feature sequence, the fluid dynamic characteristics are reconstructed in a multivariate state space to obtain a fluid state space reconstruction matrix. Singular spectrum analysis is then performed on the fluid state space reconstruction matrix to obtain fluid singular spectrum feature vectors. The fluid singular spectrum feature vector is input into a preset Gaussian process regression model for nonlinear mapping processing to obtain the fluid nonlinear mapping coefficient matrix. Based on the fluid nonlinear mapping coefficient matrix, Bayesian probability prediction is performed to obtain the fluid probability prediction distribution. Based on the fluid probability prediction distribution, Monte Carlo simulations are performed on the fluid dynamic characteristics to obtain the fluid behavior prediction matrix.

4. The intelligent water valve adaptive flow regulation method according to claim 1, characterized in that, The process of analyzing and making decisions about the water valve based on the fluid behavior prediction matrix to obtain a water valve regulation strategy includes: Based on the fluid behavior prediction matrix, pressure distribution analysis is performed on the water valve to obtain a pressure gradient field distribution map, and the topological structure of the pressure gradient field distribution map is analyzed to obtain the flow field topological feature vector. The local turbulent region of the water valve is identified by the flow field topology feature vector to obtain the coordinates of the turbulent core region. Based on the coordinates of the turbulent core region, the boundary layer separation point of the water valve is located to obtain the position of the boundary layer separation point. Spatial gradient calculation is performed on the location of the boundary layer separation point to obtain the boundary layer separation gradient field, and the flow resistance coefficient distribution of the water valve is reconstructed based on the boundary layer separation gradient field to obtain the flow resistance coefficient distribution map; Based on the flow resistance coefficient distribution diagram, the opening-flow nonlinear mapping of the water valve is performed to obtain the valve core displacement-flow transfer function. Based on the valve core displacement-flow transfer function, the optimal control point of the water valve is solved to obtain the water valve regulation strategy.

5. The intelligent water valve adaptive flow regulation method according to claim 4, characterized in that, The step of identifying the local turbulent region of the water valve using the flow field topological feature vector to obtain the coordinates of the turbulent core region includes: The topological feature vector of the flow field is used to calculate the curl field to obtain the fluid curl distribution map, and the local vorticity density is statistically calculated based on the fluid curl distribution map to obtain the vorticity intensity. The vorticity intensity is classified into singularities using critical point theory to obtain a set of flow field singularities. Stability analysis is then performed on the set of flow field singularities to obtain a singularity feature classification table. Based on the singularity feature classification table, the turbulence feature structure of the water valve is extracted to obtain a turbulence feature label sequence, and the turbulence feature label sequence is analyzed to obtain a contour map of the turbulence core region. The spatial coordinates of the outline map of the turbulent core region are transformed to obtain an initial coordinate sequence, and the core region positioning and calibration are performed based on the initial coordinate sequence to obtain the coordinates of the turbulent core region.

6. The intelligent water valve adaptive flow regulation method according to claim 4, characterized in that, The step of performing a nonlinear mapping of the valve opening to flow rate based on the flow resistance coefficient distribution map to obtain the valve core displacement to flow rate transfer function includes: Detect local extreme points in the flow resistance coefficient distribution diagram, and calculate the resistance coefficient of the water valve based on the local extreme points; The valve core displacement sensitivity analysis is performed on the water valve using the resistance coefficient to obtain the valve core displacement sensitivity matrix. Based on the valve core displacement sensitivity matrix, a nonlinear mapping basis function is constructed for the water valve to obtain the nonlinear mapping basis function. The nonlinear mapping basis function is orthogonalized to obtain orthogonalized basis functions, and the flow response coefficient of the water valve is calculated based on the orthogonalized basis functions. Based on the flow response coefficient, the valve core displacement-flow transfer function of the water valve is solved to obtain the valve core displacement-flow transfer function.

7. A smart water valve adaptive flow regulation device, characterized in that, include: The acquisition module is used to acquire real-time fluid parameters at the inlet of the water valve through a multi-parameter fluid sensor array, and obtain multi-dimensional fluid characteristic data. The analysis module is used to perform wavelet transform and spectral analysis on the multi-dimensional fluid feature data to obtain the fluid dynamic characteristics; The prediction module is used to estimate and predict based on the fluid dynamic characteristics to obtain a fluid behavior prediction matrix; The decision module is used to analyze and make decisions about the water valve based on the fluid behavior prediction matrix to obtain a water valve regulation strategy. The adjustment module is used to adjust the opening degree and flow rate of the water valve based on the water valve adjustment strategy.

8. A computer device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that, When the processor executes the computer program, it implements the steps of the method according to any one of claims 1 to 6.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by a processor, it implements the steps of the method according to any one of claims 1 to 6.

Images