Measurement system, calculation method and detection method of plunger pump flow distribution pair friction coefficient

Abstract

The invention discloses a measuring system and a calculating method for the friction coefficient of a plunger pump flow distribution pair, and relates to the technical field of fluid machinery and intelligent measurement and control. The system comprises the following components: a multi-dimensional data acquisition module, a data processing and preprocessing module, an intelligent calculation model construction module, a friction coefficient prediction and optimization module and a visualization and report generation module. According to the invention, through the multi-dimensional data acquisition module and the intelligent calculation model construction module, accurate measurement and prediction of the friction coefficient of the plunger pump flow distribution pair are realized, the nonlinear relationship among the parameters is fully considered during design of the intelligent calculation model, and training is carried out through a large amount of experimental data, so that the accuracy and the accuracy of the friction coefficient of the plunger pump flow distribution pair are improved. According to the method, the complex influence mechanism of each parameter on the friction coefficient can be automatically learned and identified, the prediction precision of the friction coefficient is improved, and the generalization ability of the model is remarkably enhanced through wide-coverage operation conditions and design variants, so that the prediction result is more reliable and stable.

Description

Technical Field

[0001] The present invention relates to the technical field of fluid machinery and intelligent measurement and control, and in particular to a measurement system, a calculation method and a detection method for a friction coefficient of a flow distribution pair of a plunger pump. Background Art

[0002] In the field of fluid machinery, the performance of the plunger pump, as an important power transmission equipment, directly affects the efficiency and reliability of the entire system. Among them, the friction coefficient of the valve plate / cylinder body is one of the important indicators to measure the performance of the pump body. The friction coefficient not only affects the energy consumption and efficiency of the pump, but is also directly related to the life and stability of the pump. Therefore, accurately measuring and optimizing the friction coefficient of the valve plate / cylinder body of the plunger pump is of great significance to improving the overall performance of the pump.

[0003] Traditional prediction and optimization methods for the friction coefficient of the piston pump distribution pair mainly rely on empirical formulas and experimental tests. Although these methods can reflect the changing trend of the friction coefficient to a certain extent, they have many shortcomings. Traditional friction coefficient measurement methods often rely on direct measurement of a single parameter or simple empirical formula estimation, and cannot quickly predict the friction coefficient of the distribution pair. According to the prediction results, it is important to analyze which parameters have the greatest impact on the friction coefficient. Therefore, it is particularly important to develop a measurement system and calculation method for the friction coefficient of the piston pump distribution pair. Summary of the invention

[0004] The purpose of the present invention is to make up for the deficiencies of the prior art and to provide a measurement system, calculation method and detection method for the friction coefficient of the flow distribution pair of a plunger pump. The system realizes comprehensive, accurate and intelligent measurement and optimization of the friction coefficient of the flow distribution pair of a plunger pump by introducing high-precision sensor networks and intelligent calculation model technologies.

[0005] In order to solve the above technical problems, the present invention provides the following technical solutions: a measurement system and calculation method for the friction coefficient of a piston pump flow distribution pair, the system comprising a multi-dimensional data acquisition module, a data processing and preprocessing module, an intelligent calculation model construction module, a friction coefficient prediction and optimization module, and a visualization and report generation module;

[0006] The multi-dimensional data acquisition module: adopts a high-precision sensor network to collect the multi-dimensional data of the geometric dimensions, material properties, and working conditions of the piston pump flow distribution pair in real time. The data acquisition system has the characteristics of high sampling rate and low noise;

[0007] The data processing and preprocessing module: cleans, denoises and standardizes the collected raw data to improve the quality of subsequent analysis, introduces data compression and feature extraction technology, and extracts key features that have a significant impact on the friction coefficient;

[0008] The intelligent computing model building module is used to build an intelligent computing model. The model design fully considers the nonlinear relationship between various parameters and trains the model through a large amount of experimental data so that it can automatically learn and identify the complex influence mechanism of various parameters on the friction coefficient.

[0009] The friction coefficient prediction and optimization module: collects various historical data on the distribution pair, including but not limited to geometric dimensions, material properties, working conditions and actually measured friction coefficients, ensures that the data covers a wide range of operating conditions and design variations to improve the generalization ability of the model, cleans the data, and encodes or converts the data as needed to adapt to the model input requirements;

[0010] Analyze the data, select parameters that have a significant impact on the friction coefficient as features, select an intelligent computing model, consider the model's prediction accuracy, training time, and interpretability factors, and adjust the model structure and parameters according to the verification results until a satisfactory prediction accuracy is achieved. Among them, F represents the friction coefficient of the piston pump flow pair, λ is a dynamic adjustment coefficient, Φ is a nonlinear activation function, m represents the number of different submodules or feature combinations in the model, τ j is the weight of each submodule or feature combination, G(X j ,θ j ) represents the function of the jth submodule or feature combination, X j is the corresponding input feature, θ j is the parameter set of this submodule;

[0011] Input new multi-dimensional parameters into the trained model to quickly predict the friction coefficient of the distribution pair. Among them, F represents the friction coefficient of the piston pump flow pair, Ω is the comprehensive adjustment factor, ψ is a nonlinear transformation function with adaptive ability, s represents the number of different functional modules in the model, ρ k is the weight coefficient of each functional module, H(Y k , μ k ,∈) represents the function of the kth functional module, Y k is the corresponding input parameter set, μ kis the parameter vector of the module, ∈ is a parameter representing uncertainty, which evaluates the confidence and uncertainty of the prediction results and provides a basis for subsequent optimization. According to the prediction results, analyze which parameters have the greatest impact on the friction coefficient, provide specific optimization design suggestions, adjust the geometric dimensions to reduce the contact area or improve the lubrication conditions, replace materials with better lubrication performance, optimize the working conditions to reduce friction and wear, implement the optimization suggestions, and re-measure the friction coefficient to verify the effect. Adjust the optimization strategy according to the measured results, and perform multiple rounds of iterative optimization until the expected goals of reducing friction and improving energy efficiency are achieved. Organize the data, models, prediction results and optimization suggestions in the optimization process, write a detailed optimization report, analyze the performance comparison before and after optimization, and quantify the improvement effect;

[0012] The visualization and report generation module provides a user-friendly interface, displays the predicted results and optimization suggestions of the friction coefficient in real time, and automatically generates a detailed measurement report.

[0013] Furthermore, the multi-dimensional data acquisition module selects laser displacement, roughness meter, pressure, flow, rotation speed and temperature sensors to monitor geometric dimensions, material properties and working conditions by clarifying the acquisition requirements. Subsequently, the sensor network is installed to ensure comprehensive coverage and stable and reliable performance. A high sampling rate and low noise data acquisition system is used, and the configuration optimizes parameters and isolates signal interference. After calibration and verification, multi-dimensional data is collected in real time and safely stored, laying a solid foundation for subsequent analysis.

[0014] Furthermore, the data cleaning of the data processing and preprocessing module involves removing outliers, processing missing values ​​and duplicate data, data denoising reduces noise through filtering technology and smoothing, data standardization includes normalization and feature scaling to unify dimensions and optimize algorithm performance, and feature extraction combines statistical analysis, domain knowledge, dimensionality reduction technology and feature selection to extract key features that are crucial to friction coefficient prediction from the data.

[0015] Furthermore, the demand analysis of the intelligent computing model building module clearly defines the modeling goal as predicting the friction coefficient of the piston pump distribution pair, selects the intelligent computing model after analyzing the parameter relationship, and then collects, preprocesses and divides the data set to build the model structure. After initializing the parameters, the training set data is used for training, and the performance of the validation set is monitored to avoid overfitting. The model performance is evaluated through the test set, the deficiencies are analyzed and optimized, and finally, the optimized model is deployed in actual applications for real-time monitoring and continuous optimization based on feedback.

[0016] Furthermore, the intelligent computing model construction module constructs an intelligent computing model: Among them, F represents the friction coefficient of the piston pump distribution pair, ω and η are global adjustment coefficients, n represents the number of different influencing factors considered, and A i , B i , C i Represent different influencing parameters, p i ,q i 、r i is the nonlinear influence index corresponding to each parameter, δ i is the normalization factor, σ i is the weight coefficient of each factor.

[0017] Furthermore, the friction coefficient prediction and optimization module analyzes which parameters have the greatest impact on the friction coefficient based on the prediction results: Where i represents different parameter indexes, I i It is an indicator to measure the influence of the i-th parameter on the friction coefficient. i It indicates the corresponding change value of the friction coefficient predicted by the model when the i-th parameter changes within its value range. is the average value of the friction coefficient predicted by the model when all parameters are at their baseline values, σ F is the population standard deviation of the friction coefficient predicted by the model, w i is the weight factor of the i-th parameter, log represents the natural logarithm, p i is the correlation coefficient between the ith parameter and the friction coefficient.

[0018] Furthermore, the visualization and report generation module requirements analysis clarified the prediction result display, optimization suggestion type and report content structure, data processing and model construction ensured the accuracy of the data and the reliability of the prediction model, the user interface development integrated real-time data visualization and optimization suggestion display, and enhanced user interactivity, and the report generation module realized automated report production and multi-format export. After iterative optimization through comprehensive testing and user feedback, the system was finally deployed on the server and equipped with detailed documentation to support ongoing maintenance.

[0019] On the other hand, a method for measuring and calculating the friction coefficient of a piston pump distribution pair is characterized in that the method comprises the following specific steps:

[0020] S100, multi-dimensional data acquisition: using a high-precision sensor network to collect real-time multi-dimensional data on the geometric dimensions, material properties, and working conditions of the piston pump flow pair. The data acquisition system has high sampling rate and low noise characteristics;

[0021] S200, data processing and preprocessing: cleaning, denoising and standardizing the collected raw data to improve the quality of subsequent analysis, introducing data compression and feature extraction technology to extract key features that have a significant impact on the friction coefficient;

[0022] S300, intelligent computing model construction: construct an intelligent computing model. The model design fully considers the nonlinear relationship between various parameters. The model is trained with a large amount of experimental data so that it can automatically learn and identify the complex influence mechanism of various parameters on the friction coefficient.

[0023] S400, Friction Coefficient Prediction and Optimization: Utilizes the trained intelligent computing model to comprehensively analyze the input multi-dimensional parameters, quickly and accurately predict the friction coefficient of the distribution pair, provide optimization design suggestions, and adjust the geometric design, material selection or working conditions of the distribution pair according to the prediction results to achieve the purpose of reducing friction and improving energy efficiency;

[0024] S500, Visualization and Report Generation: Provides a user-friendly interface, real-time display of friction coefficient prediction results and optimization suggestions, and automatic generation of detailed measurement reports.

[0025] Furthermore, the specific steps of S400 are:

[0026] S401, collect various historical data about the distribution pair, including but not limited to geometric dimensions, material properties, working conditions, and actual measured friction coefficients, ensure that the data covers a wide range of operating conditions and design variations to improve the generalization ability of the model, clean the data, and encode or convert the data as needed to meet the model input requirements;

[0027] S402, analyzing the data, selecting parameters that have a significant impact on the friction coefficient as features, selecting an intelligent computing model, considering the prediction accuracy, training time, and interpretability factors of the model, and adjusting the model structure and parameters according to the verification results until a satisfactory prediction accuracy is achieved;

[0028] S403, input the new multi-dimensional parameters into the trained model, quickly predict the friction coefficient of the distribution pair, evaluate the confidence and uncertainty of the prediction results, provide a basis for subsequent optimization, analyze which parameters have the greatest impact on the friction coefficient based on the prediction results, provide specific optimization design suggestions, adjust the geometric dimensions to reduce the contact area or improve the lubrication conditions, replace materials with better lubrication properties, optimize the working conditions to reduce friction and wear, implement the optimization suggestions, and re-measure the friction coefficient to verify the effect. Adjust the optimization strategy according to the measured results, perform multiple rounds of iterative optimization until the expected goal of reducing friction and improving energy efficiency is achieved, organize the data, models, prediction results and optimization suggestions in the optimization process, write a detailed optimization report, analyze the performance comparison before and after optimization, and quantify the improvement effect.

[0029] Compared with the prior art, a measurement system, calculation method and detection method for the friction coefficient of the flow distribution pair of a plunger pump have the following beneficial effects:

[0030] 1. The present invention realizes the accurate measurement and prediction of the friction coefficient of the piston pump distribution pair through a multi-dimensional data acquisition module and an intelligent computing model construction module. It uses a high-precision sensor network to collect multi-dimensional data of the geometric dimensions, material properties, and working conditions of the distribution pair in real time, and cleans, denoises, and standardizes the original data through a data processing and preprocessing module, which effectively improves the quality of subsequent analysis. The intelligent computing model fully considers the nonlinear relationship between various parameters when designing, and is trained through a large amount of experimental data. It can automatically learn and identify the complex influence mechanism of various parameters on the friction coefficient. This method not only improves the prediction accuracy of the friction coefficient, but also significantly enhances the generalization ability of the model by covering a wide range of operating conditions and design variants, making the prediction results more reliable and stable.

[0031] 2. The present invention further provides detailed optimization design suggestions based on the predicted friction coefficient through the friction coefficient prediction and optimization module. The system can analyze which parameters have the greatest impact on the friction coefficient based on the prediction results, thereby guiding users to perform targeted optimization design. The implementation of these optimization suggestions helps to significantly reduce the friction coefficient of the piston pump distribution pair, improve the overall energy efficiency of the pump, reduce energy consumption and wear, thereby extending the service life of the pump and reducing operating costs. In addition, the system also provides the ability of multiple rounds of iterative optimization. By continuously adjusting the optimization strategy and re-measuring and verifying, it ensures that the expected goals of reducing friction and improving energy efficiency are ultimately achieved. BRIEF DESCRIPTION OF THE DRAWINGS

[0032] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the prior art descriptions are briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention, and for ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0033] Figure 1 An operation flow chart of a system for measuring the friction coefficient of a piston pump distribution pair.

[0034] Figure 2 An operation flow chart of a method for measuring and calculating the friction coefficient of a piston pump distribution pair. DETAILED DESCRIPTION

[0035] The technical solutions in the embodiments of the present invention are described clearly and completely below. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of them. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0036] Embodiment 1

[0037] List the data items that need to be collected in detail, including geometric dimensions, material properties, and working conditions. Select appropriate sensors based on the characteristics of each data item. For example, use a laser displacement sensor to measure the gap size, use an optical or contact roughness meter to measure the surface roughness, use a pressure sensor to measure the working pressure, use a flow sensor to measure the flow, use a speed sensor to measure the speed, and use a temperature sensor to measure the temperature. For material properties, it may be necessary to combine laboratory test data input into the system. Based on the structure and working principle of the plunger pump flow pair, design a layout diagram of the sensor network to ensure that it can fully cover the area that needs to be measured. Install the sensors in the corresponding positions according to the layout diagram, pay attention to the installation accuracy and stability of the sensors, and avoid the influence of interference factors on the measurement results;

[0038] Select a data acquisition card or data acquisition module with high sampling rate and low noise characteristics to ensure that the sensor signal can be captured in real time and accurately. Connect the sensor to the data acquisition system, configure the data acquisition parameters to meet the accuracy and real-time requirements of data acquisition, provide a stable and reliable power supply for the sensor and data acquisition system, and use signal isolation technology to reduce noise interference. Use standard equipment to calibrate the sensor to ensure the accuracy and reliability of the measurement results. Verify the data acquisition system in the laboratory or under actual working conditions, check the integrity and accuracy of data acquisition, and adjust system parameters to optimize performance;

[0039] Start the data acquisition system and begin to collect multi-dimensional data of the piston pump flow distribution pair in real time. Store the collected data in a high-performance database for subsequent data processing and analysis, ensure the security and reliability of the data storage system, prevent data loss or damage, remove outliers, missing values ​​and noise data, ensure data quality, and format and standardize the data for subsequent data analysis and calculation model use.

[0040] Check the data for abnormal or unreasonable values, such as values ​​that exceed the physical range or deviate significantly from other data points, and remove them from the data set or replace them with reasonable values. Analyze the causes of missing values, select a filling strategy based on the specific circumstances of the data, or directly delete records with a large number of missing values. Check and delete duplicate records in the data set to avoid introducing bias in subsequent analysis. Use digital filtering technology to remove high-frequency noise or interference signals in the data. Use sliding average and median filtering methods to smooth the data to reduce the impact of random errors.

[0041] Convert data of different dimensions to the same dimension for comparison and analysis. For some machine learning algorithms, it can speed up convergence and improve algorithm performance. Use lossy compression algorithm to reduce data size while allowing a certain degree of information loss. It is suitable for scenarios with extremely large data volumes and not requiring very high precision. For scenarios where data integrity needs to be maintained, use lossless compression algorithm to reduce data storage space without losing any information. Calculate data statistics as part of the feature. According to the physical properties of the piston pump distribution pair and the relevant knowledge of the friction coefficient, select or design a specific feature extraction method. Apply principal component analysis and linear discriminant analysis dimensionality reduction techniques to extract key features that have a significant impact on the friction coefficient from the original data. At the same time, reduce the data dimension and improve the efficiency of subsequent analysis. Use correlation analysis and mutual information methods to evaluate the predictive ability of each feature on the friction coefficient, and select the most important features for subsequent analysis.

[0042] Deeply understand the purpose of modeling, that is, to accurately predict or evaluate the friction coefficient of the piston pump flow pair, identify and analyze the key parameters that affect the friction coefficient, and preliminarily evaluate the nonlinear relationship between them. According to the analysis results, select the intelligent calculation model. Among them, F represents the friction coefficient of the piston pump distribution pair, ω and η are global adjustment coefficients, n represents the number of different influencing factors considered, and A i , B i , C i Represent different influencing parameters, p i ,q i 、r i is the nonlinear influence index corresponding to each parameter, δ i is the normalization factor, σ i are the weight coefficients of each factor. These models have strong ability to handle nonlinear relationships. Experimental data containing key parameters and corresponding friction coefficients are collected to ensure the diversity and representativeness of the data. The collected data are cleaned, denoised, and standardized to improve data quality. The preprocessed data are divided into training set, validation set, and test set, usually in a ratio of 70%, 15%, 15%, or similar ratios, for model training and evaluation;

[0043] According to the selected model type, design the specific structure of the model, including but not limited to the number of layers, number of neurons, activation function, key parameters of the optimization algorithm, assign initial values ​​to the weights and bias parameters in the model, which has a certain impact on the training effect of the model, use appropriate programming languages ​​and libraries to build the model, and perform necessary debugging and verification;

[0044] Define the key parameters of learning rate, batch size, and number of iterations during the training process, use the training set data to train the model, continuously optimize the model parameters through the back propagation algorithm to reduce the prediction error, regularly evaluate the performance of the model on the validation set, observe the changes in the loss function and accuracy indicators to avoid overfitting or underfitting, use the test set data to evaluate the performance of the model, analyze the problems and deficiencies of the model based on the evaluation results, and adjust the model structure or training parameters based on the analysis results for further optimization training.

[0045] Collect various historical data on the valve pair, including but not limited to geometric dimensions, material properties, working conditions, and actual measured friction coefficients, ensure that the data covers a wide range of operating conditions and design variations to improve the generalization ability of the model, clean the data, and encode or convert the data as needed to adapt to the model input requirements;

[0046] Analyze the data, select parameters that have a significant impact on the friction coefficient as features, select an intelligent computing model, consider the model's prediction accuracy, training time, and interpretability factors, and adjust the model structure and parameters according to the verification results until a satisfactory prediction accuracy is achieved. Among them, F represents the friction coefficient of the piston pump flow pair, λ is a dynamic adjustment coefficient, Φ is a nonlinear activation function, m represents the number of different submodules or feature combinations in the model, τ j is the weight of each submodule or feature combination, G(X j ,θ j ) represents the function of the jth submodule or feature combination, X j is the corresponding input feature, θ j is the parameter set of this submodule;

[0047] Input new multi-dimensional parameters into the trained model to quickly predict the friction coefficient of the distribution pair. Among them, F represents the friction coefficient of the piston pump flow pair, Ω is the comprehensive adjustment factor, A is a nonlinear transformation function with adaptive ability, s represents the number of different functional modules in the model, ρ k is the weight coefficient of each functional module, H(Y k , μ k ,∈) represents the function of the kth functional module, Y k is the corresponding input parameter set, μ k is the parameter vector of the module, ∈ is a parameter representing uncertainty, which evaluates the confidence and uncertainty of the prediction results and provides a basis for subsequent optimization. According to the prediction results, it analyzes which parameters have the greatest impact on the friction coefficient. Where i represents different parameter indexes, I i It is an indicator to measure the influence of the i-th parameter on the friction coefficient.i It indicates the corresponding change value of the friction coefficient predicted by the model when the i-th parameter changes within its value range. is the average value of the friction coefficient predicted by the model when all parameters are at their baseline values, σ F is the population standard deviation of the friction coefficient predicted by the model, w i is the weight factor of the i-th parameter, log represents the natural logarithm, p i It is the correlation coefficient between the i-th parameter and the friction coefficient. Provide specific optimization design suggestions, adjust the geometric dimensions to reduce the contact area or improve the lubrication conditions, replace materials with better lubrication properties, optimize the working conditions to reduce friction and wear, implement the optimization suggestions, and re-measure the friction coefficient to verify the effect. Adjust the optimization strategy according to the measured results, and perform multiple rounds of iterative optimization until the expected goals of reducing friction and improving energy efficiency are achieved. Organize the data, models, prediction results and optimization suggestions in the optimization process, write a detailed optimization report, analyze the performance comparison before and after optimization, and quantify the improvement effect.

[0048] Determine which forecast results and optimization suggestions need to be displayed, design a user-friendly interface to ensure that information is intuitive and easy to understand, and that interactions are convenient; determine which parts of the report should include; integrate data sources to ensure data accuracy; and perform necessary cleaning and preprocessing; build a friction coefficient prediction model based on historical data using machine learning or statistical methods; validate the model to ensure its accuracy and reliability, and make adjustments and optimizations when necessary;

[0049] Integrate data visualization library to display the predicted results and dynamic changes of friction coefficient in real time. Dynamically display optimization suggestions on the interface based on the predicted results and preset rules. Add user interaction elements to allow users to adjust parameters or view more details. Design report templates to ensure that the content structure is clear and easy to understand. Obtain relevant data from the database and organize it. Write scripts or functions to automatically generate complete measurement reports based on templates and data. Support export in multiple formats to facilitate user sharing and archiving. Conduct comprehensive tests on the visualization interface and report generation module to ensure the normal operation of various functions. Invite target user groups to test and collect feedback. Iteratively optimize the system based on test results and user feedback.

[0050] Embodiment 2

[0051] A high-precision sensor network is installed at key locations on the production line, including laser displacement sensors, roughness meters, pressure sensors, flow sensors, speed sensors and temperature sensors. These sensors are carefully arranged to ensure that the multi-dimensional data of the geometric dimensions, material properties and working conditions of the piston pump distribution pair can be fully and accurately collected. A high-sampling rate, low-noise data acquisition system is used to ensure the real-time and accuracy of the data. At the same time, the system configuration optimizes parameters and isolates signal interference. After strict calibration and verification, real-time collection and secure storage of multi-dimensional data are achieved.

[0052] The data processing module is used to thoroughly clean the collected raw data, remove outliers, process missing values ​​and duplicate data, ensure data quality, reduce noise interference in the data through advanced filtering technology and smoothing processing, improve data purity, normalize the data and perform feature scaling processing, unify the dimensions and optimize algorithm performance, laying a solid foundation for subsequent analysis. By combining statistical analysis, domain knowledge, dimensionality reduction technology and feature selection methods, key features that are crucial to the prediction of friction coefficient are extracted from the data.

[0053] The modeling goal is to predict the friction coefficient of the piston pump flow pair, analyze the nonlinear relationship between the parameters, and build an intelligent calculation model. Among them, F represents the friction coefficient of the piston pump distribution pair, ω and η are global adjustment coefficients, n represents the number of different influencing factors considered, and A i , B i , C i Represent different influencing parameters, p i ,q i 、r i is the nonlinear influence index corresponding to each parameter, δ i is the normalization factor, σ i The model design fully considers the complex interactions among various parameters to accurately capture the dynamic changes of the friction coefficient. A large amount of historical experimental data is used to train the model so that it can automatically learn and identify the complex influence mechanism of various parameters on the friction coefficient. At the same time, the performance of the validation set is monitored to avoid overfitting, and the model performance is evaluated through the test set to continuously optimize the model structure and parameters.

[0054] Input the multi-dimensional data of the newly produced plunger pump flow pair into the trained intelligent calculation model to quickly predict its friction coefficient. Among them, F represents the friction coefficient of the piston pump distribution pair, which is the final result predicted by this module, Ω is the comprehensive adjustment factor, ψ is a nonlinear transformation function with adaptive ability, s represents the number of different functional modules in the model, ρ k is the weight coefficient of each functional module, H(Y k, μ k ,∈) represents the function of the kth functional module, Y k is the corresponding input parameter set, μ k is the parameter vector of the module, ∈ is a parameter representing uncertainty, which evaluates the confidence and uncertainty of the prediction results and analyzes which parameters have the greatest impact on the friction coefficient. Where i represents different parameter indexes, I i It is an indicator to measure the influence of the i-th parameter on the friction coefficient. i It indicates the corresponding change value of the friction coefficient predicted by the model when the i-th parameter changes within its value range. is the average value of the friction coefficient predicted by the model when all parameters are at their baseline values, σ F is the population standard deviation of the friction coefficient predicted by the model, w i is the weight factor of the i-th parameter, log represents the natural logarithm, p i It is the correlation coefficient between the i-th parameter and the friction coefficient. The prediction results and parameter sensitivity analysis are intuitively displayed through visualization tools. Specific optimization design suggestions are put forward based on the prediction results, such as adjusting the geometric dimensions to reduce the contact area, replacing materials with better lubrication properties, and optimizing working conditions to reduce friction and wear. The optimization suggestions are implemented, and the friction coefficient is re-measured to verify the effect. The optimization strategy is adjusted according to the measured results, and multiple rounds of iterative optimization are carried out until the expected goals of reducing friction and improving energy efficiency are achieved.

[0055] It provides a friendly user interface, displays the predicted results and optimization suggestions of the friction coefficient in real time, and facilitates production personnel to quickly obtain key information. It automatically generates detailed measurement reports, including data collection, processing, model prediction results, optimization suggestions and implementation effects. The report supports multi-format export for subsequent analysis and archiving.

[0056] It will be apparent to those skilled in the art that the invention is not limited to the details of the exemplary embodiments described above and that the invention can be implemented in other specific forms without departing from the spirit or essential features of the invention. Therefore, the embodiments should be considered exemplary and non-limiting in all respects, and the scope of the invention is defined by the appended claims rather than the foregoing description, and it is intended that all variations falling within the meaning and scope of the equivalent elements of the claims be included in the invention. Any reference numeral in a claim should not be considered as limiting the claim to which it relates.

Claims

1. A measuring system for the friction coefficient of a piston pump flow distribution pair, characterized in that: The system includes a multi-dimensional data acquisition module, a data processing and preprocessing module, an intelligent computing model building module, a friction coefficient prediction and optimization module, and a visualization and report generation module; The multi-dimensional data acquisition module: adopts a high-precision sensor network to collect the multi-dimensional data of the geometric dimensions, material properties, and working conditions of the piston pump flow distribution pair in real time. The data acquisition system has the characteristics of high sampling rate and low noise; The data processing and preprocessing module: cleans, denoises and standardizes the collected raw data to improve the quality of subsequent analysis, introduces data compression and feature extraction technology, and extracts key features that have a significant impact on the friction coefficient; The intelligent computing model building module is used to build an intelligent computing model. The model design fully considers the nonlinear relationship between various parameters and trains the model through a large amount of experimental data so that it can automatically learn and identify the complex influence mechanism of various parameters on the friction coefficient. The friction coefficient prediction and optimization module: collects various historical data about the distribution pair, including but not limited to geometric dimensions, material properties, working conditions and actually measured friction coefficients, to improve the generalization ability of the model, cleans the data, and encodes or converts the data as needed; Analyze the data, select parameters that have a significant impact on the friction coefficient as features, select an intelligent computing model, consider the model's prediction accuracy, training time, and interpretability factors, and adjust the model structure and parameters according to the verification results until a satisfactory prediction accuracy is achieved. Among them, F represents the friction coefficient of the piston pump flow pair, λ is a dynamic adjustment coefficient, Φ is a nonlinear activation function, m represents the number of different submodules or feature combinations in the model, τ j is the weight of each submodule or feature combination, G(X j ,θ j ) represents the function of the jth submodule or feature combination, X j is the corresponding input feature, θ j is the parameter set of this submodule; Input new multi-dimensional parameters into the trained model to quickly predict the friction coefficient of the distribution pair. Among them, F represents the friction coefficient of the piston pump flow pair, Ω is the comprehensive adjustment factor, ψ is a nonlinear transformation function with adaptive ability, s represents the number of different functional modules in the model, ρ k is the weight coefficient of each functional module, H(Y k , μ k ,∈) represents the function of the kth functional module, Y k is the corresponding input parameter set, μ k is the parameter vector of the module, ∈ is a parameter representing uncertainty, which evaluates the confidence and uncertainty of the prediction results and provides a basis for subsequent optimization. According to the prediction results, analyze which parameters have the greatest impact on the friction coefficient, provide specific optimization design suggestions, adjust the geometric dimensions to reduce the contact area or improve the lubrication conditions, replace materials with better lubrication performance, optimize the working conditions to reduce friction and wear, implement the optimization suggestions, and re-measure the friction coefficient to verify the effect. Adjust the optimization strategy according to the measured results, and perform multiple rounds of iterative optimization until the expected goals of reducing friction and improving energy efficiency are achieved. Organize the data, models, prediction results and optimization suggestions in the optimization process, write a detailed optimization report, analyze the performance comparison before and after optimization, and quantify the improvement effect; The visualization and report generation module provides a user-friendly interface, displays the predicted results and optimization suggestions of the friction coefficient in real time, and automatically generates a detailed measurement report.

2. A system for measuring the friction coefficient of a piston pump flow distribution pair according to claim 1, characterized in that: The multi-dimensional data acquisition module clarifies the acquisition requirements, selects laser displacement, roughness meter, pressure, flow, rotation speed and temperature sensors, monitors geometric dimensions, material properties and working conditions, then installs a sensor network, adopts a high sampling rate, low noise data acquisition system, configures optimized parameters and isolates signal interference, and after calibration and verification, collects and securely stores multi-dimensional data in real time.

3. A system for measuring the friction coefficient of a piston pump flow distribution pair according to claim 1, characterized in that: The data cleaning of the data processing and preprocessing module involves removing outliers, processing missing values ​​and duplicate data, data denoising reduces noise through filtering technology and smoothing, data standardization includes normalization and feature scaling to unify dimensions and optimize algorithm performance, and feature extraction combines statistical analysis, domain knowledge, dimensionality reduction technology and feature selection to extract key features that are crucial to friction coefficient prediction from the data.

4. A system for measuring the friction coefficient of a piston pump flow distribution pair according to claim 1, characterized in that: The intelligent computing model building module requirement analysis clearly defines the modeling goal as predicting the friction coefficient of the piston pump distribution pair. After analyzing the parameter relationship, the intelligent computing model is selected, and then data collection, preprocessing and data set division are carried out to build the model structure. After initializing the parameters, the training set data is used for training. At the same time, the performance of the validation set is monitored to avoid overfitting, and the model performance is evaluated through the test set. The deficiencies are analyzed and optimized. Finally, the optimized model is deployed in actual applications, real-time monitoring is carried out, and continuous optimization is carried out based on feedback.

5. A system for measuring the friction coefficient of a piston pump flow distribution pair according to claim 4, characterized in that: The intelligent computing model building module builds an intelligent computing model: Among them, F represents the friction coefficient of the piston pump flow pair, ω and η are global adjustment coefficients, n represents the number of different influencing factors considered, and A i , B i , C i Represent different influencing parameters, p i ,q i 、r i is the nonlinear influence index corresponding to each parameter, δ i is the normalization factor, σ i is the weight coefficient of each factor.

6. A system for measuring the friction coefficient of a piston pump flow distribution pair according to claim 1, characterized in that: The friction coefficient prediction and optimization module analyzes which parameters have the greatest impact on the friction coefficient based on the prediction results: Where i represents different parameter indexes, I i It is an indicator to measure the influence of the i-th parameter on the friction coefficient. i It indicates the corresponding change value of the friction coefficient predicted by the model when the i-th parameter changes within its value range. is the average value of the friction coefficient predicted by the model when all parameters are at their baseline values, σ F is the population standard deviation of the friction coefficient predicted by the model, w i is the weight factor of the i-th parameter, log represents the natural logarithm, p i is the correlation coefficient between the ith parameter and the friction coefficient.

7. A system for measuring the friction coefficient of a piston pump flow distribution pair according to claim 1, characterized in that: The visualization and report generation module requirement analysis clarified the prediction result display, optimization suggestion type and report content structure. Data processing and model construction ensured the accuracy of data and the reliability of prediction model. User interface development integrated real-time data visualization and optimization suggestion display to enhance user interactivity. The report generation module realized automated report production and multi-format export. After iterative optimization through comprehensive testing and user feedback, the system was finally deployed on the server and equipped with detailed documentation to support ongoing maintenance.

8. A method for measuring and calculating the friction coefficient of a piston pump distribution pair according to any one of claims 1 to 7, characterized in that: The specific steps of this method are: S100, multi-dimensional data acquisition: using a high-precision sensor network to collect real-time multi-dimensional data on the geometric dimensions, material properties, and working conditions of the piston pump flow pair. The data acquisition system has high sampling rate and low noise characteristics; S200, data processing and preprocessing: cleaning, denoising and standardizing the collected raw data to improve the quality of subsequent analysis, introducing data compression and feature extraction technology to extract key features that have a significant impact on the friction coefficient; S300, intelligent computing model construction: construct an intelligent computing model. The model design fully considers the nonlinear relationship between various parameters. The model is trained with a large amount of experimental data so that it can automatically learn and identify the complex influence mechanism of various parameters on the friction coefficient. S400, Friction Coefficient Prediction and Optimization: Utilizes the trained intelligent computing model to comprehensively analyze the input multi-dimensional parameters, quickly and accurately predict the friction coefficient of the distribution pair, provide optimization design suggestions, and adjust the geometric design, material selection or working conditions of the distribution pair according to the prediction results to achieve the purpose of reducing friction and improving energy efficiency; S500, Visualization and Report Generation: Provides a user-friendly interface, real-time display of friction coefficient prediction results and optimization suggestions, and automatic generation of detailed measurement reports.

9. The method for measuring and calculating the friction coefficient of the piston pump distribution pair according to claim 8, characterized in that: The specific steps of S400 are: S401, collecting various historical data about the flow distribution pair, including but not limited to geometric dimensions, material properties, working conditions, and actually measured friction coefficients, to improve the generalization ability of the model, clean the data, and encode or convert the data as needed; S402, analyzing the data, selecting parameters that have a significant impact on the friction coefficient as features, selecting an intelligent computing model, considering the prediction accuracy, training time, and interpretability factors of the model, and adjusting the model structure and parameters according to the verification results until a satisfactory prediction accuracy is achieved; S403, input the new multi-dimensional parameters into the trained model, quickly predict the friction coefficient of the distribution pair, evaluate the confidence and uncertainty of the prediction results, provide a basis for subsequent optimization, analyze which parameters have the greatest impact on the friction coefficient based on the prediction results, provide specific optimization design suggestions, adjust the geometric dimensions to reduce the contact area or improve the lubrication conditions, replace materials with better lubrication properties, optimize the working conditions to reduce friction and wear, implement the optimization suggestions, and re-measure the friction coefficient to verify the effect. Adjust the optimization strategy according to the measured results, perform multiple rounds of iterative optimization until the expected goal of reducing friction and improving energy efficiency is achieved, organize the data, models, prediction results and optimization suggestions in the optimization process, write a detailed optimization report, analyze the performance comparison before and after optimization, and quantify the improvement effect.

Images