Simulation method and system for water effect test of squatting pan flush valve

Abstract

The invention discloses a simulation method and system for a water effect test of a squatting pan flush valve, and relates to the technical field of intelligent bathroom fluid simulation, and the method comprises the steps: constructing a multi-dimensional parameter matrix, generating an initial sampling matrix through a space filling optimization algorithm, executing fluid dynamic simulation, and forming a training data source; inputting a training data source into the time sequence prediction model, and generating a scouring force three-dimensional distribution field and a transient flow change curve through Fourley Z operator measurement and time convolutional network decoding; on the basis of the scouring force three-dimensional distribution field, calculating a surface cleaning coverage rate value and a minimum shearing force of a key area, establishing a water efficiency evaluation index in combination with a transient flow change curve, and obtaining a parameterized design space of geometric parameters of the groove; furthermore, space-time diagram network nodes are constructed through vortex mark tensors, and high-precision modeling of a scouring force three-dimensional distribution field and a transient flow change curve is achieved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of intelligent bathroom fluid simulation, in particular to a simulation method and system for water efficiency testing of a squat toilet flush valve. BACKGROUND

[0002] With the increasing severity of global water resource shortage, the development of water-saving sanitary appliances and the establishment of standardized testing methods have become the focus of water efficiency management. As a widely used sanitary equipment in public and household toilets, the water efficiency performance of the flush valve of a squat toilet directly affects the overall water efficiency. In recent years, the evaluation of flush valve water efficiency has gradually shifted from traditional qualitative experience-based judgment to quantitative simulation analysis, with significant progress in fluid dynamics modeling, scouring force distribution prediction, and transient flow control.

[0003] However, the existing technology still has obvious deficiencies in achieving high-precision and real-time response water efficiency evaluation. First, the computational overhead of traditional computational fluid dynamics simulation is large, making it difficult to complete iterative optimization under a large number of parameter combinations in a short time, limiting its application efficiency in actual engineering design. Second, although existing research has attempted to introduce deep learning models to predict the time series data of the flushing process, due to the lack of effective extraction mechanism for key flow characteristics such as vortex structure, the model has weak generalization ability under non-steady flow conditions, affecting the accuracy of the prediction results. SUMMARY

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

[0005] Therefore, the present application provides a simulation method for water efficiency testing of a squat toilet flush valve to solve the problems of low computational fluid dynamics simulation efficiency, weak generalization ability of deep learning models, and insufficient structural optimization in the prior art.

[0006] To solve the above technical problems, the present application provides the following technical solutions:

[0007] In a first aspect, the present application provides a simulation method for water efficiency test of a squat toilet flush valve, which comprises: constructing a multi-dimensional parameter matrix, generating an initial sampling matrix through a space-filling optimization algorithm, and performing fluid dynamics simulation to form a training data source; inputting the training data source into a time series prediction model, generating a scouring force three-dimensional distribution field and a transient flow variation curve through Pauli Z operator measurement and time convolution network decoding; based on the scouring force three-dimensional distribution field, calculating a surface cleaning coverage rate value and a minimum shear force in a key area, combining the transient flow variation curve to establish a water efficiency evaluation index, and obtaining a parameterized design space of the groove geometric parameters; in the parameterized design space, a quantum optimization objective function is constructed by using the water efficiency evaluation index, and a constraint condition of the surface cleaning coverage rate value is met, the groove geometric parameters are optimized, and an optimal drainage structure feature is generated; the optimal drainage structure feature is input into a transient fluid dynamics model for verification, and when a deviation between a simulation flow and a predicted value of the transient flow variation curve exceeds a set tolerance, the time series prediction model is updated.

[0008] As a preferred scheme of the simulation method for water efficiency test of the squat toilet flush valve, the method comprises the following steps:

[0009] The original working condition data of the water supply pressure, valve opening and closing movement characteristics and water temperature are collected, and a multi-dimensional parameter matrix is generated through a Latin hypercube sampling algorithm;

[0010] The multi-dimensional parameter matrix is input into a space-filling optimization algorithm for processing to generate an initial sampling matrix with equidistant distribution characteristics;

[0011] The initial sampling matrix is input into a transient multiphase flow solver to perform fluid dynamics simulation and output transient flow sequence and scouring force field data;

[0012] The scouring force field data are processed by dimension reduction to extract a scouring feature distribution, and a training data set is formed in combination with the transient flow sequence.

[0013] As a preferred scheme of the simulation method for water efficiency test of the squat toilet flush valve, the method comprises the following steps:

[0014] The time series prediction model loads the scouring feature distribution and the transient flow sequence in the training data source, and generates a vortex label tensor through a vortex intensity detection function;

[0015] Quantum feature screening is performed based on the vortex label tensor, and a key vortex feature point set is extracted by minimizing a Hamiltonian equation.

[0016] constructing space-time graph network nodes through the key vortex feature point set, performing energy gradient-aware message passing in a three-dimensional flow field neighborhood, and updating the space-time graph network nodes;

[0017] inputting the updated space-time graph network nodes into a quantum measurement unit, and outputting a three-dimensional distribution field of flushing force through Pauli Z operator measurement;

[0018] decoding the residual connection between the space-time graph network nodes and the transient flow sequence through a time convolution network, and generating a transient flow variation curve.

[0019] As a preferred scheme of the simulation method for testing the water efficiency of the squatting pan flushing valve, wherein: based on the three-dimensional distribution field of flushing force, the surface cleaning coverage value and the minimum shear force of the key area are calculated, and the specific steps are as follows,

[0020] Based on the three-dimensional distribution field of flushing force, a differential manifold model of the squatting pan surface is established, and a principal curvature tensor is calculated;

[0021] According to the principal curvature tensor, and combined with the rheological properties of the dirt, a dynamic critical shear force is calculated;

[0022] Based on the vector projection of the dynamic critical shear force and the three-dimensional distribution field of flushing force, a surface cleaning coverage value is calculated through a curvature weight function;

[0023] According to the surface cleaning coverage value, the boundary of the key area is determined, and the minimum observed value of the vector projection shear force in the key area boundary is taken as the minimum shear force of the key area.

[0024] As a preferred scheme of the simulation method for testing the water efficiency of the squatting pan flushing valve, wherein: the water efficiency evaluation index is established combined with the transient flow variation curve, and a parameterized design space of the groove geometric parameters is obtained, and the specific steps are as follows,

[0025] Based on the transient flow variation curve, a transient flow acceleration is calculated, and a pulse energy function curve with a sharp peak characteristic is extracted;

[0026] The pulse energy function curve, the surface cleaning coverage value and the minimum shear force of the key area are input into a neuron excitation response function to generate a water efficiency evaluation index;

[0027] Based on the water efficiency evaluation index, a human portal vein fractal structure is taken as a biological prototype to obtain a parameterized design space of the groove geometric parameters.

[0028] As a preferred scheme of the simulation method for testing the water efficiency of the squatting toilet flushing valve, wherein: the quantum optimization objective function is constructed by using the water efficiency evaluation index in the parameterized design space, and the constraint condition of the surface cleaning coverage value is met, the groove geometric parameters are optimized, and the optimal drainage structure characteristics are generated, and the specific steps are as follows,

[0029] The water efficiency evaluation index is converted into the constraint condition of the surface cleaning coverage value, and the feasible region range is constructed in the parameterized design space;

[0030] The quantum optimization objective function is constructed based on the groove geometric parameters in the feasible region range, and the quantum optimization objective function is processed by quantum parameter coding to generate the optimized parameter combination;

[0031] The biological calibration is performed on the optimized parameter combination, and the deviation of the optimized parameter combination is corrected by the biological fluid gradient characteristics to generate the optimal drainage structure characteristics.

[0032] As a preferred scheme of the simulation method for testing the water efficiency of the squatting toilet flushing valve, wherein: the optimal drainage structure characteristics are input into the transient fluid dynamics model for verification, when the deviation between the simulated flow and the predicted value of the transient flow change curve exceeds the set tolerance, the time series prediction model is updated, and the specific steps are as follows,

[0033] The optimal drainage structure characteristics are input into the transient fluid dynamics model, the quantumized fluid field is constructed, and the simulated transient flow data are generated;

[0034] The hybrid loss value between the simulated transient flow data and the predicted flow value output by the time series prediction model is calculated;

[0035] When the hybrid loss value exceeds the set tolerance, the weight updating mechanism of the time series prediction model is triggered, and the time series prediction model is updated by the quantum-fluid hybrid gradient.

[0036] In a second aspect, the present application provides a simulation system for water efficiency test of a squat toilet flushing valve, comprising a parameter sampling module, a time sequence prediction module, a water efficiency evaluation module, a quantum optimization module and a model verification module. The parameter sampling module is configured to construct a multi-dimensional parameter matrix, generate an initial sampling matrix by a space filling optimization algorithm, and perform fluid dynamics simulation to form a training data source. The time sequence prediction module is configured to input the training data source into a time sequence prediction model, generate a scouring force three-dimensional distribution field and a transient flow change curve by a Pauli Z operator measurement and a time convolution network decoding. The water efficiency evaluation module is configured to calculate a surface cleaning coverage value and a minimum shear force of a key area based on the scouring force three-dimensional distribution field, establish a water efficiency evaluation index in combination with the transient flow change curve, and define a parameterized design space of a groove geometric parameter. The quantum optimization module is configured to construct a quantum optimization objective function by using the water efficiency evaluation index in the parameterized design space, and optimize the groove geometric parameter to generate optimal drainage structure characteristics under the constraint condition of the surface cleaning coverage value. The model verification module is configured to input the optimal drainage structure characteristics into a transient fluid dynamics model for verification, and update the time sequence prediction model when a deviation between a simulation flow and a predicted value of the transient flow change curve exceeds a set tolerance.

[0037] In a third aspect, the present application provides a computer device comprising a memory and a processor, wherein the memory stores a computer program, and when the computer program is executed by the processor, any step of the simulation method for water efficiency test of a squat toilet flushing valve according to the first aspect of the present application is implemented.

[0038] In a fourth aspect, the present application provides a computer readable storage medium having a computer program stored thereon, wherein when the computer program is executed by a processor, any step of the simulation method for water efficiency test of a squat toilet flushing valve according to the first aspect of the present application is implemented.

[0039] The present application has the following beneficial effects: by constructing a multi-dimensional parameter matrix and generating an initial sampling matrix by a space filling optimization algorithm, the distribution quality and calculation efficiency of simulation samples are improved, thereby effectively reducing the calculation cost of traditional computational fluid dynamics simulation in large-scale parameter iteration; further, by constructing a space-time graph network node by a vortex marker tensor, high-precision modeling of the scouring force three-dimensional distribution field and the transient flow change curve is realized, and the problem of insufficient generalization ability of existing deep learning models under non-steady flow conditions is solved. BRIEF DESCRIPTION OF DRAWINGS

[0040] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings needed in the embodiment description will be briefly introduced as follows. Obviously, the drawings in the following description are only some embodiments of the present application, and other drawings can be obtained by those skilled in the art without creative labor.

[0041] Fig. 1 Flow chart for a simulation method for a urinal flush valve water efficiency test.

[0042] Fig. 2 Flow chart for generating a material rheology parameter set.

[0043] Fig. 3 Flow chart for generating a print instruction set.

[0044] Fig. 4 Flow chart for generating a global optimization parameter set. DETAILED DESCRIPTION

[0045] In order to make the above objectives, features and advantages of the present application more apparent, the specific embodiments of the present application will be described in detail below with reference to the accompanying drawings.

[0046] In the following description, numerous specific details are set forth in order to provide a thorough understanding of the present application. However, it will be apparent to one skilled in the art that the present application can be practiced without the specific details given herein. In other instances, well-known methods have not been described in detail in order to avoid unnecessarily obscuring the present application. Accordingly, it will be appreciated that the present application can be practiced with modification and alteration, and that the present application disclosed herein will be limited only by the scope of the appended claims.

[0047] Secondly, the "one embodiment" or "embodiment" referred to herein means that a specific feature, structure or characteristic can be included in at least one implementation of the present application. The "in one embodiment" appearing in different places in the specification does not mean the same embodiment, nor is it an embodiment that is independent or alternative to other embodiments.

[0048] REFERENCE Figs. 1-4 For one embodiment of the present application, the embodiment provides a simulation method for a urinal flush valve water efficiency test, comprising the following steps:

[0049] S1: Construct a multi-dimensional parameter matrix, generate an initial sampling matrix through a space-filling optimization algorithm, and perform a fluid dynamics simulation to form a training data source.

[0050] S1.1: Collect original working condition data of water supply pressure, valve opening and closing motion characteristics and water temperature, and generate a multi-dimensional parameter matrix through a Latin hypercube sampling algorithm.

[0051] The specific process includes that the original working condition data of water supply pressure, valve opening and closing motion characteristics and water temperature are processed through the Latin hypercube sampling algorithm to ensure that the parameter combinations in the multi-dimensional parameter space have uniform distribution characteristics, and a multi-dimensional parameter matrix is generated.

[0052] S1.2: Input the multi-dimensional parameter matrix into a space-filling optimization algorithm for processing to generate an initial sampling matrix with equidistant distribution characteristics.

[0053] The specific process includes that the multi-dimensional parameter matrix is iteratively optimized by a space filling optimization algorithm, the spatial distribution of the multi-dimensional parameter combination is adjusted to make the Euclidean distance of any two sampling points in the multi-dimensional space consistent, and finally an initial sampling matrix satisfying the equidistant distribution characteristic is output, ensuring that the input parameters of the transient multiphase flow solver have global uniform coverage.

[0054] S1.3: input the initial sampling matrix into the transient multiphase flow solver, execute the fluid dynamics simulation and output the transient flow sequence and the scouring force field data.

[0055] The specific process includes that the initial sampling matrix is input into the transient multiphase flow solver as a boundary condition, the transient multiphase flow solver obtains the motion state of the fluid during the flushing process based on the Navier-Stokes equation, and the flow field distribution at each time step is obtained by discrete solving through the finite volume method, and finally the transient flow sequence and the three-dimensional spatial distribution of the scouring force field data are output.

[0056] S1.4: dimensionality reduction processing is performed on the scouring force field data, the scouring feature distribution is extracted, and the transient flow sequence is combined to form a training data set.

[0057] The specific process includes that the scouring force field data is reduced in spatial dimension by principal component analysis method, the scouring feature distribution in the main variation direction is retained, and at the same time the transient flow sequence is time-aligned and normalized, and finally the processed scouring feature distribution and the transient flow sequence are combined according to the time step to form a training data set for training the time sequence prediction model.

[0058] S2: input the training data source into the time sequence prediction model, generate the scouring force three-dimensional distribution field and the transient flow change curve through the Pauli Z operator measurement and the time convolution network decoding.

[0059] S2.1: the time sequence prediction model loads the scouring feature distribution and the transient flow sequence in the training data source, and generates the vortex marked tensor through the vortex intensity detection function.

[0060] The specific process includes that the time sequence prediction model receives the scouring feature distribution and the transient flow sequence in the training data source as input, obtains the vorticity amplitude and the helicity feature at each spatial position through the vortex intensity detection function, judges the vortex core existing area according to the dynamic adaptive discrimination condition and generates the binary marked vortex marked tensor, and the three-dimensional vortex marked tensor is used to identify the vortex core distribution position in the scouring force field.

[0061] The dynamic adaptive discrimination condition is automatically set based on the statistical analysis of the vorticity amplitude gradient change rate and the local helicity extreme value, and the discrimination condition is dynamically adjusted by sliding window analysis of the statistical distribution characteristics of the vortex features in the historical data.

[0062] Further, the pre-training process of the time series prediction model first aligns the scour feature distribution with the transient flow sequence in the time dimension, extracts the time series features through the long short-term memory network, optimizes the prediction accuracy of the transient flow change using the mean square error loss function, and simultaneously combines the vortex intensity information provided by the vortex signature tensor as an auxiliary supervision signal. Through a multi-task learning framework, the flow prediction and vortex feature identification tasks are jointly optimized to ultimately obtain a time series prediction model that can accurately predict the transient flow change curve.

[0063] S2.2: Quantum feature screening based on vortex signature tensor, key vortex feature point set is extracted by minimizing Hamiltonian equation.

[0064] The specific process includes that the vortex signature tensor is processed by the quantum feature screening algorithm, the vortex position information is encoded into the superposition state of quantum bits, the energy function containing the vorticity amplitude and helicity features is constructed based on the Hamiltonian equation, and the minimum value of the energy function is obtained through the quantum variational optimization algorithm. Finally, the spatial coordinate set corresponding to the key vortex feature points is output, and the key vortex feature points represent the core area of the dominant flow structure in the scour force field.

[0065] S2.3: Constructing space-time graph network nodes through key vortex feature point set, performing energy gradient-aware message passing in three-dimensional flow field neighborhood, and updating space-time graph network nodes.

[0066] The specific process includes that the key vortex feature point set is used as the space-time graph network node, each space-time graph network node establishes a connection edge with the adjacent feature points in the three-dimensional flow field, the vorticity transport strength and pressure gradient direction between the space-time graph network nodes are obtained through the energy gradient-aware function, the flow field feature information of the neighborhood space-time graph network nodes is aggregated using the graph attention mechanism, and the hidden state vector of each node in the space-time graph network is iteratively updated. Finally, the space-time graph network node representation containing the dynamic evolution characteristics of the flow field is obtained.

[0067] S2.4: Input the updated space-time graph network node into the quantum measurement unit, and output the three-dimensional distribution field of the scour force through the Pauli Z operator measurement.

[0068] The specific process includes that the updated space-time graph network node is converted into a quantum state representation through the quantum measurement unit, the Pauli Z operator measurement is applied to each space-time graph network node to obtain the expected value of the quantum bit, and the measurement result is mapped back to the three-dimensional spatial coordinates and interpolated to generate a continuous distribution of scour force field intensity values. Finally, the three-dimensional distribution field of the scour force covering the entire flow field area is output.

[0069] S2.5: Decoding the residual connection between the space-time graph network node and the transient flow sequence through the time convolution network to generate the transient flow change curve.

[0070] The specific process includes merging features between spatiotemporal graph network nodes and transient flow sequences through residual connections, extracting multi-scale time patterns layer by layer using dilated causal convolutional layers in the temporal convolutional network, filtering key time step features through gated activation functions, and finally generating transient flow change curves aligned with the input time steps through the linear transformation layer of the output channel.

[0071] S3: Based on the three-dimensional distribution field of scouring force, calculate the surface cleaning coverage rate and the minimum shear force in key areas, establish water efficiency evaluation index by combining transient flow rate change curves, and obtain the parametric design space of trench geometric parameters.

[0072] S3.1: Based on the three-dimensional distribution field of the scouring force, establish a differential manifold model of the squat toilet surface and calculate the principal curvature tensor, the expression of which is:

[0073]

[0074] Where K represents the principal curvature tensor, Let I represent the gradient tensor of the surface normal vector, and let n represent the unit tensor of the surface. This represents the tensor product operator.

[0075] The specific process includes establishing a parameterized surface representation on the surface of the squat toilet using a differential geometry processing method for the three-dimensional distribution field of the scouring force, calculating the first-order and second-order basic form coefficient matrices at each point on the surface, solving for the characteristic directions and eigenvalues ​​of the coefficient matrices through eigenvalue decomposition, and obtaining the principal curvature tensor describing the local geometric properties of the surface. The principal curvature tensor contains the curvature characteristics of the vortex structure formed on the surface by the fluid under the action of the scouring force field.

[0076] Furthermore, the pre-training process of the differential manifold model first maps the three-dimensional distribution field of the scouring force onto the parameterized grid of the squat toilet surface, establishes the differential structure through local coordinate system transformation, uses the Riemannian geometry method to obtain the Gaussian curvature and average curvature of each grid point, constructs the curvature feature field using the principal curvature tensor, and optimizes the manifold parameters by combining the gradient descent algorithm to minimize the error between the curvature feature field and the measured scouring force field distribution, and finally obtains a differential manifold model that can accurately characterize the fluid-surface interaction properties.

[0077] S3.2: Based on the principal curvature tensor and combined with the rheological characteristics of the wastewater, calculate the dynamic critical shear force. The expression is as follows:

[0078]

[0079] Where x represents the three-dimensional spatial coordinates of the toilet surface, t represents the time variable of the flushing process, τ(x,t) represents the dynamic critical shear force at position x on the toilet surface and time t of the flushing process, τ0 represents the minimum shear force required for dirt to begin to peel off at standard temperature, α represents the temperature sensitivity coefficient of dirt yield stress, T represents the current fluid temperature, β represents the acceleration coupling coefficient, u represents the fluid velocity vector field, and γ K K represents the correction coefficient of the principal curvature tensor. max This represents the maximum principal curvature tensor.

[0080] The specific process includes: the principal curvature tensor provides the surface geometric curvature characteristics; the rheological properties of the contaminant describe the deformation response characteristics of the material under shear; the shear stress field distribution relationship is established by coupling the curvature characteristics with the rheological constitutive equation; the critical shear stress threshold is solved based on the rheological yield condition; and finally, dynamic critical shear force distribution data that varies with surface position and time are output.

[0081] The rheological properties of the waste were obtained by measuring the stress-strain relationship curves of the waste samples under the shear rate gradient using a rotational rheometer, specifically manifested as constitutive parameters such as shear thinning index and yield stress.

[0082] S3.3: Based on the vector projection of the three-dimensional distribution field of dynamic critical shear force and scouring force, the surface cleaning coverage rate is calculated using a curvature weighting function. The expression is as follows:

[0083]

[0084] Where η represents the surface cleaning coverage rate, Ω represents the entire surface area of ​​the squat toilet, and K max (x) represents the maximum principal curvature tensor at position x on the surface of the squat toilet, I represents the cleanliness indicator function, τ1 represents the effective shear force, and A represents the infinitesimal area of ​​the squat toilet surface.

[0085] in:

[0086]

[0087] It should be noted that when the effective shear force τ1 is greater than the dynamic critical shear force τ, The function value is 1, indicating that the surface dirt of the squat toilet has been effectively removed. When the effective shear force τ1 is less than or equal to the dynamic critical shear force τ, then... The function value is 1, indicating that the surface dirt of the squat toilet has not been removed.

[0088] The specific process includes obtaining the local effective scouring force components by vector dot product operation of the three-dimensional distribution field of dynamic critical shear force and scouring force, adjusting the force efficiency of different surface regions by combining the curvature weight function constructed by the principal curvature tensor, statistically analyzing the proportion of surface area where the effective scouring force exceeds the dynamic critical shear force, and finally outputting the surface cleaning coverage rate value that characterizes the cleaning effect.

[0089] S3.4: Delineate the boundary of the critical area based on the surface cleanliness coverage rate, and take the minimum observed value of the vector projected shear force within the boundary of the critical area as the minimum shear force of the critical area.

[0090] The specific process includes determining the transition boundary between compliant and non-compliant cleaning effects using contour line segmentation, marking non-compliant areas within the transition boundary as key areas, extracting the vector projection shear force observation values ​​of all locations within the key areas, finding the minimum value through sorting and comparison, and finally outputting the minimum shear force of the key areas that characterizes the degree of scouring force weakness in the key areas.

[0091] S3.5: Calculate the transient flow acceleration based on the transient flow rate change curve, and extract the pulse energy function curve with peak characteristics. The expression is:

[0092]

[0093] Where a(t) represents the transient flow acceleration during the flushing process at time t, and Q represents the transient flow rate value. Let λ represent the second derivative of the transient flow rate Q with respect to time t, and let λ represent the nonlinear gain coefficient (0.015 ≤ λ ≤ 0.025). This represents the rate of change of the transient flow rate Q at time t during the flushing process.

[0094] The specific process includes: calculating the rate of change of flow at each time point using the central difference method to obtain the transient flow acceleration sequence; using wavelet transform to detect high-frequency abrupt changes in the acceleration sequence; extracting the envelope of the abrupt changes as the pulse energy function using Hilbert transform; and finally outputting the pulse energy function curve characterizing the intensity of the siphon effect.

[0095] S3.6: Input the pulse energy function curve, the surface cleanliness coverage rate, and the minimum shear force in the key area into the neuron excitation response function to generate water efficiency evaluation indicators.

[0096] The specific process includes converting the pulse energy function curve, surface cleaning coverage rate, and minimum shear force in the key area into a uniform-dimensional input vector through normalization. The input vector is then nonlinearly mapped through the sigmoid activation layer of the neuron's excitation response function. Finally, a weighted summation layer is used to fuse the contribution of the pulse energy function curve, the surface cleaning coverage rate, and the minimum shear force in the key area, resulting in a comprehensive scalar water efficiency evaluation index that characterizes flushing effectiveness.

[0097] S3.7: Based on water efficiency evaluation indicators, and using the fractal structure of the human portal vein as a biological prototype, obtain the parametric design space for the geometric parameters of the trench.

[0098] The specific process includes establishing a correlation mapping relationship between water efficiency evaluation indicators and the geometric characteristic parameters of the human portal vein fractal structure, determining the benchmark range of the trench fractal iteration factor through the fractal dimension, constraining the feasible range of trench inclination angle and width-to-depth ratio by combining the portal vein branch angle and diameter ratio, and finally generating a parametric design space containing feasible combinations of trench geometric parameters.

[0099] S4: Within the parametric design space, a quantum optimization objective function is constructed using water efficiency evaluation indicators, while satisfying the constraint condition of surface clean coverage rate. The trench geometric parameters are optimized to generate the optimal drainage structure features.

[0100] S4.1: Transform water efficiency evaluation indicators into constraints for surface cleanliness coverage rates, and construct the feasible domain in the parametric design space.

[0101] The specific process includes establishing a mapping relationship between water efficiency evaluation indicators and surface clean coverage rate values ​​through logistic regression, transforming the surface clean coverage rate requirements into boundary constraints of trench geometric parameters in the parametric design space, and using a hyperplane segmentation algorithm to divide the feasible domain range that satisfies the constraints in the parametric design space.

[0102] S4.2: Construct a quantum optimization objective function based on the trench geometry parameters within the feasible region, and generate an optimal parameter combination by processing the quantum optimization objective function through quantum parameter encoding.

[0103] The specific process includes: the trench geometric parameters within the feasible region are mathematically expressed through a quantum optimization objective function; the quantum optimization objective function includes a weighted combination of constraints such as water efficiency evaluation index and surface clean coverage rate; the trench geometric parameters are converted into a superposition state representation of qubits through quantum parameter encoding; the minimum value of the quantum optimization objective function is obtained by using a quantum annealing algorithm; and the lowest energy state output by the quantum processor is decoded to obtain the optimized parameter combination.

[0104] S4.3: Perform biomimetic calibration on the optimized parameter combination and correct the deviation of the optimized parameter combination through the biofluid gradient characteristics to generate the optimal drainage structure characteristics.

[0105] The specific process includes: optimizing the parameter combination by comparing it with the fractal structure characteristics of the human portal vein through a biomimetic calibration process; describing the flow field distribution law at the branch of the portal vein using biofluid gradient characteristics; adjusting the groove inclination angle and fractal iteration factor in the optimized parameter combination using a gradient descent algorithm to maximize the matching degree between the drainage structure and the fractal characteristics of the biological prototype; and finally outputting the optimal drainage structure characteristics that meet the dual standards of fluid performance and biological similarity.

[0106] S5: Input the optimal drainage structure features into the transient fluid dynamics model for verification. When the deviation between the simulated flow rate and the predicted value of the transient flow rate change curve exceeds the set tolerance, update the time series prediction model.

[0107] S5.1: Input the optimal drainage structure features into the transient fluid dynamics model, perform quantized fluid field construction, and generate simulated transient flow data.

[0108] The specific process includes using the optimal drainage structure features as boundary conditions to input the transient fluid dynamics model, coupling the classical grid and quantum state superposition through the quantized fluid field construction method, obtaining the Navier-Stokes equations to obtain the fluid velocity field and pressure field distribution, integrating to obtain the relationship between the volumetric flow rate of the outlet section and time, and finally outputting the simulated transient flow rate data corresponding to the time series.

[0109] Furthermore, the training process of the transient fluid dynamics model first uses the optimal drainage structure features as geometric boundary conditions input to the solver, discretizes the computational domain using the finite volume method and initializes a quantum-classical hybrid grid, uses a projection algorithm to solve the coupled Navier-Stokes equations and quantum pressure terms, uses an adaptive time step to advance the computation of transient flow field evolution, synchronously collects outlet cross-sectional flow data at each time step to form a simulated transient flow sequence, optimizes the turbulence model parameters by comparing experimental measurement data, and finally obtains a transient fluid dynamics model that can accurately predict flow changes.

[0110] S5.2: Calculate the mixed loss value between the simulated transient flow data and the predicted flow value output by the time series prediction model. The expression is:

[0111]

[0112] Where L represents the mixed loss value, B represents the total duration of the flushing process, Φ represents the double exponential filter function, Q0(t) represents the simulated flow rate output by the transient fluid dynamics model at time t, Q1(t) represents the predicted flow rate output by the time series prediction model at time t, and σQ The standard deviation of transient flow data is represented by N, the total number of quantum state measurements is represented by i, and the discrete time point number of the quantum state sampling is represented by φ. i Let ψ represent the theoretical ideal quantum state at the i-th sampling time. i This represents the quantum state actually measured by the quantum sensor at the i-th sampling time.

[0113] The specific process includes standardizing the simulated transient flow data and the predicted flow values ​​output by the time series prediction model through a double exponential filter function, calculating the absolute value of the flow deviation at each time point, and combining it with the quantum fidelity loss term for weighted fusion, finally outputting a mixed loss value that simultaneously reflects classical flow error and quantum state stability.

[0114] S5.3: When the mixed loss value exceeds the preset tolerance threshold, the weight update mechanism of the time series prediction model is triggered, and the time series prediction model is updated through quantum-fluid mixed gradient.

[0115] The specific process includes: comparing the mixed loss value with a preset tolerance threshold to determine whether to trigger the update process; the weight update mechanism of the time series prediction model uses quantum-fluid hybrid gradient to calculate the partial derivative of the loss function with respect to the network parameters; adjusting the connection weight matrix of the long short-term memory network through the gradient descent algorithm; updating the variational parameters of the quantum coding layer; and finally outputting the retrained time series prediction model.

[0116] The preset tolerance threshold is based on the statistical distribution characteristics of mixed loss values ​​in historical simulation data, and is dynamically set by obtaining the mean and three times the standard deviation through a sliding window.

[0117] This embodiment also provides a simulation system for testing the water efficiency of a squat toilet flush valve, including: a parameter sampling module, a time series prediction module, a water efficiency evaluation module, a quantum optimization module, and a model verification module.

[0118] The system comprises four modules: a parameter sampling module, a time series prediction module, and a model. The parameter sampling module constructs a multidimensional parameter matrix, generates an initial sampling matrix using a space-filling optimization algorithm, and performs fluid dynamics simulation to form a training data source. The time series prediction module inputs the training data source into the time series prediction model, generates a three-dimensional scouring force distribution field and transient flow rate change curves through Pauli Z-operator measurement and temporal convolutional network decoding. The water efficiency evaluation module calculates the surface cleanliness coverage rate and the minimum shear force in key areas based on the three-dimensional scouring force distribution field, establishes water efficiency evaluation indicators based on the transient flow rate change curves, and defines a parameterized design space for the trench geometry. The quantum optimization module constructs a quantum optimization objective function within the parameterized design space using the water efficiency evaluation indicators, while satisfying the constraint of the surface cleanliness coverage rate, optimizes the trench geometry, and generates the optimal drainage structure features. The model validation module inputs the optimal drainage structure features into the transient fluid dynamics model for validation. When the deviation between the simulated flow rate and the predicted value of the transient flow rate change curve exceeds a set tolerance, the time series prediction model is updated.

[0119] This embodiment also provides a computer device suitable for simulating the water efficiency test of a squat toilet flush valve, comprising: a memory and a processor; the memory is used to store computer-executable instructions, and the processor is used to execute the computer-executable instructions to implement the simulation method for testing the water efficiency of a squat toilet flush valve as proposed in the above embodiment.

[0120] The computer device can be a terminal, comprising a processor, memory, communication interface, display screen, and input devices connected via a system bus. The processor provides computing and control capabilities. The memory includes non-volatile storage media and internal memory. The non-volatile storage media stores the operating system and computer programs. The internal memory provides an environment for the operation of the operating system and computer programs stored in the non-volatile storage media. The communication interface is used for wired or wireless communication with external terminals; wireless communication can be achieved through Wi-Fi, carrier networks, NFC (Near Field Communication), or other technologies. The display screen can be an LCD screen or an e-ink screen. The input devices can be a touch layer covering the display screen, buttons, a trackball, or a touchpad on the computer device's casing, or an external keyboard, touchpad, or mouse.

[0121] This embodiment also provides a storage medium storing a computer program that, when executed by a processor, implements the simulation method for testing the water efficiency of a squat toilet flush valve as proposed in the above embodiment. The storage medium can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as Static Random Access Memory (SRAM), Electrically Erasable Programmable Read-Only Memory (EEPROM), Erasable Programmable Read Only Memory (EPROM), Programmable Red-Only Memory (PROM), Read-Only Memory (ROM), magnetic storage, flash memory, magnetic disk, or optical disk.

[0122] In summary, this invention improves the distribution quality and computational efficiency of simulation samples by constructing a multidimensional parameter matrix and using a space-filling optimization algorithm to generate an initial sampling matrix, thereby effectively reducing the computational overhead of traditional computational fluid dynamics simulations in large-scale parameter iterations. Furthermore, by constructing spatiotemporal graph network nodes using vortex-labeled tensors, it achieves high-precision modeling of the three-dimensional distribution field of scouring force and transient flow rate change curves, solving the problem of insufficient generalization ability of existing deep learning models under unsteady flow conditions.

[0123] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention, and all such modifications or substitutions should be covered within the scope of the claims of the present invention.

Claims

1. A simulation method for testing the water efficiency of a squat toilet flush valve, characterized in that: include, A multidimensional parameter matrix is ​​constructed, an initial sampling matrix is ​​generated through a space-filling optimization algorithm, and fluid dynamics simulation is performed to form a training data source; The training data source is input into the time series prediction model, and the three-dimensional distribution field of scouring force and transient flow change curve are generated by measuring with the Pauli Z operator and decoding with a temporal convolutional network. Based on the three-dimensional distribution field of scouring force, the surface cleaning coverage rate and the minimum shear force in key areas are calculated. Water efficiency evaluation index is established by combining transient flow rate change curves, and the parametric design space of trench geometry parameters is obtained. Within the parametric design space, a quantum optimization objective function is constructed using water efficiency evaluation indicators, while satisfying the constraint condition of surface clean coverage rate. This optimizes the trench geometric parameters and generates the optimal drainage structure features. The optimal drainage structure features are input into the transient fluid dynamics model for verification. When the deviation between the simulated flow rate and the predicted value of the transient flow rate change curve exceeds the set tolerance, the time series prediction model is updated.

2. The simulation method for testing the water efficiency of a squat toilet flush valve as described in claim 1, characterized in that: The process of constructing a multidimensional parameter matrix, generating an initial sampling matrix through a space-filling optimization algorithm, and performing fluid dynamics simulations to form a training data source is as follows. Raw operating condition data such as water supply pressure, valve opening and closing motion characteristics, and water temperature are collected and multi-dimensional parameter matrices are generated using the Latin hypercube sampling algorithm. The multidimensional parameter matrix is ​​processed by the input space filling optimization algorithm to generate an initial sampling matrix with equidistant distribution characteristics; Input the initial sampling matrix into the transient multiphase flow solver, perform fluid dynamics simulation, and output transient flow sequence and scour force field data; The scour force field data is dimensionality reduced to extract the scour feature distribution, and combined with the transient flow sequence to form a training dataset.

3. The simulation method for testing the water efficiency of a squat toilet flush valve as described in claim 2, characterized in that: The process involves inputting the training data source into the time series prediction model, measuring with the Pauli Z operator, and decoding with a temporal convolutional network to generate a three-dimensional distribution field of scouring force and transient flow change curves. The specific steps are as follows: The time series prediction model loads the scour feature distribution and transient flow sequence from the training data source, and generates a vortex label tensor through the vortex intensity detection function; Quantum feature screening is performed based on vortex labeling tensors, and key vortex feature point sets are extracted by minimizing the Hamiltonian equation. Spatiotemporal graph network nodes are constructed by using key vortex feature point sets, energy gradient sensing message passing is performed in the neighborhood of the three-dimensional flow field, and the spatiotemporal graph network nodes are updated. The updated spacetime graph network nodes are input into the quantum measurement unit, and the three-dimensional distribution field of the scouring force is measured by the Pauli Z operator. The transient flow change curve is generated by decoding the residual connections between the spatiotemporal graph network nodes and the transient flow sequence through a temporal convolutional network.

4. The simulation method for testing the water efficiency of a squat toilet flush valve as described in claim 3, characterized in that: The specific steps for calculating the surface cleaning coverage rate and the minimum shear force in key areas based on the three-dimensional distribution field of scouring force are as follows. A differential manifold model of the squat toilet surface is established based on the three-dimensional distribution field of the scouring force, and the principal curvature tensor is calculated. Based on the principal curvature tensor and combined with the rheological characteristics of the pollutants, the dynamic critical shear force is calculated. Based on the vector projection of the three-dimensional distribution field of dynamic critical shear force and scouring force, the surface cleaning coverage rate is calculated by the curvature weighting function. The critical area boundary is delineated based on the surface cleanliness coverage value, and the minimum observed value of the vector projected shear force within the critical area boundary is taken as the minimum shear force of the critical area.

5. The simulation method for testing the water efficiency of a squat toilet flush valve as described in claim 4, characterized in that: The specific steps for establishing water efficiency evaluation indicators by combining transient flow rate change curves and obtaining the parametric design space of trench geometric parameters are as follows: The transient flow acceleration is calculated based on the transient flow change curve, and the pulse energy function curve with peak characteristics is extracted; The pulse energy function curve, along with the surface cleaning coverage rate and the minimum shear force in the key area, are input into the neuron excitation response function to generate water efficiency evaluation indicators. Based on water efficiency evaluation indicators, and using the fractal structure of the human portal vein as a biological prototype, a parametric design space for the geometric parameters of the trench is obtained.

6. The simulation method for testing the water efficiency of a squat toilet flush valve as described in claim 5, characterized in that: Within the parametric design space, a quantum optimization objective function is constructed using water efficiency evaluation indicators, satisfying the constraint condition of surface cleanliness coverage rate. This optimizes the trench geometric parameters and generates the optimal drainage structure features. The specific steps are as follows: Water efficiency evaluation indicators are transformed into constraints on surface cleanliness coverage rates, and a feasible domain is constructed in the parametric design space. A quantum optimization objective function is constructed based on the trench geometry parameters within the feasible region, and the quantum optimization objective function is processed by quantum parameter encoding to generate a combination of optimization parameters; Biomimetic calibration is performed on the optimized parameter combination, and the deviation of the optimized parameter combination is corrected by the biofluid gradient characteristics to generate the optimal drainage structure characteristics.

7. The simulation method for testing the water efficiency of a squat toilet flush valve as described in claim 6, characterized in that: The optimal drainage structure characteristics are input into the transient fluid dynamics model for verification. When the deviation between the simulated flow rate and the predicted value of the transient flow rate change curve exceeds the set tolerance, the time series prediction model is updated. The specific steps are as follows: The optimal drainage structure features are input into the transient fluid dynamics model to perform quantized fluid field construction and generate simulated transient flow data; Calculate the mixed loss value between simulated transient flow data and the predicted flow value output by the time series prediction model; When the mixed loss value exceeds the set tolerance, the weight update mechanism of the time series prediction model is triggered, and the time series prediction model is updated through quantum-fluid mixed gradient.

8. A simulation system for testing the water efficiency of a squat toilet flush valve, based on the simulation method for testing the water efficiency of a squat toilet flush valve according to any one of claims 1 to 7, characterized in that: It includes a parameter sampling module, a time series prediction module, a water efficiency assessment module, a quantum optimization module, and a model validation module. The parameter sampling module is used to construct a multidimensional parameter matrix, generate an initial sampling matrix through a space filling optimization algorithm, and perform fluid dynamics simulation to form a training data source. The time series prediction module is used to input the training data source into the time series prediction model, and generate the three-dimensional distribution field of scouring force and transient flow change curve through Pauli Z operator measurement and temporal convolutional network decoding. The water efficiency assessment module is used to calculate the surface cleaning coverage rate and the minimum shear force in key areas based on the three-dimensional distribution field of scouring force, establish water efficiency evaluation indicators by combining transient flow rate change curves, and define the parametric design space of trench geometry parameters. The quantum optimization module is used to construct a quantum optimization objective function within the parameterized design space using water efficiency evaluation indicators, while satisfying the constraint condition of surface clean coverage rate, optimizing trench geometric parameters, and generating optimal drainage structure features. The model validation module is used to input the optimal drainage structure features into the transient fluid dynamics model for validation. When the deviation between the simulated flow rate and the predicted value of the transient flow rate change curve exceeds the set tolerance, the time series prediction model is updated.

9. 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 simulation method for testing the water efficiency of the squat toilet flush valve as described in any one of claims 1 to 7.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by the processor, it implements the steps of the simulation method for testing the water efficiency of the squat toilet flush valve as described in any one of claims 1 to 7.

Images