Water pump operation efficiency testing method, tester, equipment and medium

By monitoring the operating status and environmental parameters of the water pump in real time, and using time series analysis and machine learning models to identify water pump efficiency changes and potential failures, the real-time and accuracy of water pump energy efficiency tests in the existing technology are solved, and the HVAC system is realized Stability and energy efficiency management.

CN118911977AInactive Publication Date: 2024-11-08BEIJING HUAQING ENERGY TECH DEV CO LTD +1
View PDF 0 Cites 5 Cited by

Patent Information

Application Number
CN202411033397.4
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-07-30
Publication Date
2024-11-08
Estimated Expiration
Not applicable · inactive patent

AI Technical Summary

Technical Problem

The existing water pump energy efficiency testing methods cannot monitor the operating status in real time, and lack comprehensive analysis of the water pump operating status and environmental parameters, which makes it difficult to detect and accurately diagnose potential faults in the early stage, affecting the stability and energy efficiency of the HVAC system.

Method used

Real-time acquisition of water pump operating status and environmental parameter information, and through time series analysis and machine learning models, we identify efficiency changes trends and abnormal points, predict potential fault types and their probability, and realize intelligent diagnostic and preventive maintenance.

Benefits of technology

It improves the efficiency of water pump energy efficiency testing, realizes early detection and accurate diagnosis of potential faults, ensures the stability of the HVAC system and the comfort of the indoor environment, and meets the needs of sustainable development and energy conservation and emission reduction.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118911977A_ABST
    Figure CN118911977A_ABST
Patent Text Reader

Abstract

The invention relates to a water pump operation efficiency testing method, a tester, equipment and a medium, and belongs to the technical field of water pumps. The testing method comprises the steps that operation state information and environment parameter information of a to-be-tested water pump are acquired; calculating water pump efficiency data of each time period according to the operation state information; constructing a time sequence according to the water pump efficiency data of each time period, and analyzing and identifying trend components and periodic characteristics in the time sequence; detecting abnormal points in the time sequence according to trend components and periodic characteristics in the time sequence; determining environmental parameter information and water pump efficiency data corresponding to the abnormal points, and inputting the environmental parameter information and the water pump efficiency data to a pre-trained fault prediction model to obtain a fault type prediction result and a corresponding prediction probability; determining a fault test result of the to-be-tested water pump according to the fault type prediction result with the highest prediction probability; and sending the fault test result of the to-be-tested water pump to the management terminal. The water pump energy efficiency testing efficiency can be improved, and potential fault hidden dangers can be recognized in time.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present application relates to the technical field of water pumps, and in particular to a method, a tester, a device and a medium for testing the operating efficiency of a water pump. Background Art

[0002] In the building's heating, ventilation and air conditioning (HVAC) system, water pumps play a vital role. They are responsible for circulating water or other media to ensure that the temperature and humidity of the indoor environment are maintained at an appropriate level. With the continuous rise in energy costs and increasingly stringent environmental protection requirements, improving the energy efficiency of water pumps has become the key to reducing operating costs and achieving sustainable development. The energy efficiency test of water pumps not only helps to evaluate their performance in actual operation, but also has important significance for predicting and preventing potential failure problems.

[0003] At present, the common water pump energy efficiency test method only conducts efficiency evaluation and troubleshooting after the water pump shows obvious performance degradation or stops due to fault. It is impossible to grasp the operating status of the water pump in real time, and lacks comprehensive analysis of the water pump operating status and environmental parameters. There are certain limitations in status analysis and fault prediction, making it difficult to achieve early detection and accurate diagnosis of potential faults, resulting in low accuracy of energy efficiency testing and failure to detect potential faults in a timely manner. Summary of the invention

[0004] In order to improve the efficiency of water pump energy efficiency testing and timely identify potential fault hazards, the present application provides a water pump operation efficiency testing method, tester, equipment and medium.

[0005] In the first aspect, the present application provides a method for testing the operating efficiency of a water pump, which adopts the following technical solution: A method for testing the operating efficiency of a water pump, the method comprising: Obtaining the operating status information and environmental parameter information of the water pump to be tested; Calculate the water pump efficiency data for each time period according to the operating status information; Constructing a time series based on the pump efficiency data for each period, analyzing and identifying trend components and periodic features in the time series; Detecting abnormal points in the time series based on trend components and periodic characteristics in the time series; Determine the environmental parameter information and water pump efficiency data corresponding to the abnormal point, and input them into a pre-trained fault prediction model to obtain a fault type prediction result and a corresponding prediction probability; Determining the fault test result of the water pump to be tested according to the fault type prediction result with the highest prediction probability; The fault test result of the water pump to be tested is sent to the management terminal.

[0006] By adopting the above technical solution, the operating status and environmental parameters of the water pump are monitored in real time, time series analysis is used to identify efficiency change trends, and machine learning models are combined to intelligently diagnose abnormal situations of efficiency decline, predict potential fault types and their probabilities, and finally feed back the diagnosis results to the management terminal, thereby achieving preventive maintenance and optimized operations. This application can better understand and analyze the reasons for changes in water pump efficiency by comprehensively considering environmental parameters and water pump operating status. Through intelligent data processing and decision support, the efficiency of water pump energy efficiency testing is improved, and early detection and accurate diagnosis of potential faults are achieved. At the same time, the stability of the HVAC system and the comfort of the indoor environment are ensured, meeting the modern building needs of sustainable development and energy conservation and emission reduction.

[0007] Optionally, the step of calculating the water pump efficiency data for each time period according to the operating status information includes: According to the operating status information, the pump outlet pressure P in each period is obtained by analysis. out , Pump inlet pressure P in , the height difference H between the inlet and outlet pressure gauges of the water pump, the real-time flow rate G through the water pump and the real-time power W of the water pump; Based on the preset calculation formula, the real-time efficiency η of the water pump in each period is calculated as: η=G*(P out -P in +ρ*g*H*10 -6 ) / (3.6 * W); In the above formula, ρ is the density of cold (hot) water, and g is the acceleration due to gravity.

[0008] By adopting the above technical solution, the preset efficiency calculation formula is converted into a computer-executable algorithm. By accurately calculating the real-time efficiency of the water pump, a data basis is provided for subsequent time series analysis and fault prediction.

[0009] Optionally, the steps of constructing a time series according to the pump efficiency data of each time period, and analyzing and identifying trend components and periodic characteristics in the time series include: Preprocessing the water pump efficiency data of each time period and arranging them in chronological order to obtain a time series data set; Extracting statistical features from the time series data set to obtain a statistical feature set; Calculating rolling statistics in the time series data set according to the statistical feature set to obtain a rolling statistics set; According to the rolling statistics set, the time series data set is decomposed into time series to extract trend components; Correcting the time series data set according to the trend component to identify and remove long-term nonlinear trends or structural changes; The corrected time series data set is identified based on Fourier transform to obtain periodic features.

[0010] By adopting the above technical solution, the time series is constructed after preprocessing, and the trend components and periodic characteristics are separated using statistical analysis and Fourier transform techniques. The long-term patterns and short-term fluctuations of efficiency changes can be effectively identified, which improves the accuracy and practicality of data analysis. This is of great significance to improving the efficiency of energy efficiency management of water pump systems and the scientific nature of maintenance decisions.

[0011] Optionally, the step of detecting abnormal points in the time series according to the trend component and the periodicity characteristics in the time series includes: Establishing a baseline model based on the trend component and periodic characteristics in the time series; Based on the baseline model, determining the deviation of the time series data set to obtain a deviation value set; Identify data points in the deviation value set that exceed a preset deviation threshold to obtain a potential outlier point set; Performing periodic verification on the set of potential abnormal points according to the periodic characteristics, and excluding potential abnormal points consistent with the periodic characteristics; A trend consistency check is performed on the periodically checked potential abnormal point set according to the trend component, and potential abnormal points in the potential abnormal point set that are contrary to the trend component are determined as abnormal points in the time series.

[0012] By adopting the above technical solution, after multi-dimensional analysis of periodicity and trends, the comprehensiveness and accuracy of abnormal point determination are ensured, and abnormal points that deviate from short-term rules and affect long-term trends can be effectively identified, thereby providing strong support for subsequent fault diagnosis, performance optimization and decision-making. This technical solution is particularly suitable for scenarios in this application that require high-reliability monitoring of water pump efficiency, and can promptly detect and respond to potential performance degradation or failures, thereby reducing downtime and improving the overall efficiency of the system.

[0013] Optionally, the testing method further includes a fault prediction model training step, wherein the training step includes: Acquire a historical data set and perform data preprocessing; wherein the historical data set includes multiple sets of historical environmental parameter information and historical water pump efficiency data; Based on the historical fault database, assigning fault type labels to multiple groups of historical environmental parameter information and historical water pump efficiency data in the historical data set after data preprocessing to obtain a sample data set; Perform feature extraction based on the sample data set to identify the feature set that is highly correlated with water pump failure; Divide the sample data set after feature extraction into a training sample set and a validation sample set; Inputting the training sample set into a pre-built machine learning model for initial training, optimizing model parameters, and obtaining the initially trained fault prediction model; The fault prediction model is verified and model parameters are corrected based on the verification sample set until the loss function of the model meets a preset condition or the number of model iterations reaches a preset number, thereby obtaining a trained fault prediction model.

[0014] By adopting the above technical solution, after feature extraction and data set division, the model can learn the most valuable information. Through parameter optimization in the model training and verification process, the generalization ability and prediction accuracy of the model are guaranteed. The final fault prediction model can provide scientific and accurate guidance for the operation and maintenance of the water pump, effectively prevent and reduce the occurrence of faults, reduce maintenance costs, and improve system reliability.

[0015] Optionally, after the step of sending the fault test result of the water pump to be tested to the management terminal, the step further includes: Receiving feedback information sent by the management terminal indicating that the fault test result is incorrect; Extracting environmental parameter information and water pump efficiency data of the abnormal point corresponding to the fault test result according to the feedback information; According to the feedback information, a fault type label is assigned to the environmental parameter information and the water pump efficiency data to generate a feedback sample set; The fault prediction model is iteratively trained according to the feedback sample set, model parameters are corrected, and a corrected fault prediction model is obtained and redeployed.

[0016] By adopting the above technical solution, a feedback mechanism is added to the original fault prediction model training process, so that the model can self-optimize according to the feedback of actual test results. By learning these clearly labeled feedback data, the prediction logic of the model is gradually corrected, which improves the generalization ability and adaptability of the model, making it more accurate in predicting various complex fault conditions.

[0017] In the second aspect, the present application provides a water pump operation efficiency tester, which adopts the following technical solution: A water pump operating efficiency tester, the tester comprising: An acquisition module is used to obtain the operating status information and environmental parameter information of the water pump to be tested; A water pump efficiency calculation module, used to calculate the water pump efficiency data for each time period according to the operating status information; A time series analysis module, used to construct a time series based on the pump efficiency data of each time period, and to analyze and identify trend components and periodic characteristics in the time series; An outlier detection module, used to detect outliers in the time series based on trend components and periodic characteristics in the time series; A fault prediction module is used to determine the environmental parameter information and water pump efficiency data corresponding to the abnormal point, and input them into a pre-trained fault prediction model to obtain a fault type prediction result and a corresponding prediction probability; A fault test result determination module, used to determine the fault test result of the water pump to be tested according to the fault type prediction result with the highest prediction probability; The fault test result sending module is used to send the fault test result of the water pump to be tested to the management terminal.

[0018] Optionally, the tester comprises: A feedback information receiving module, used for receiving feedback information sent by the management terminal indicating that the fault test result is incorrect; An extraction module, used to extract environmental parameter information and water pump efficiency data of the abnormal point corresponding to the fault test result according to the feedback information; A feedback sample set generation module, used to assign fault type labels to the environmental parameter information and water pump efficiency data according to the feedback information, and generate a feedback sample set; The model correction module is used to iteratively train the fault prediction model according to the feedback sample set, correct the model parameters, obtain the corrected fault prediction model and redeploy it.

[0019] In a third aspect, the present application provides a computer device, which adopts the following technical solution: A computer device comprises a memory, a processor and a computer program stored in the memory, wherein the processor executes the computer program to implement the steps of the method as described in the first aspect.

[0020] In a fourth aspect, the present application provides a computer-readable storage medium, which adopts the following technical solution: A computer-readable storage medium stores a computer program that can be loaded by a processor and execute any one of the methods in the first aspect.

[0021] In summary, the present application includes at least one of the following beneficial technical effects: 1. By comprehensively considering environmental parameters and pump operating conditions, we can better understand and analyze the reasons for changes in pump efficiency. Through intelligent data processing and decision support, we can improve the efficiency of pump energy efficiency testing, achieve early detection and accurate diagnosis of potential faults, and ensure the stability of the HVAC system and the comfort of the indoor environment, meeting the modern building needs of sustainable development and energy conservation and emission reduction.

[0022] 2. Integrated data analysis and machine learning algorithms can provide deeper insights. By continuously monitoring and analyzing the operating data of the pump, future maintenance needs and potential failures can be predicted. This can not only take maintenance measures in advance to avoid operation interruptions caused by sudden failures, but also optimize the operation strategy of the pump to further improve energy efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0023] Figure 1 It is a first flow chart of a method for testing the operating efficiency of a water pump according to one of the embodiments of the present application.

[0024] Figure 2 It is a second flow chart of a method for testing the operating efficiency of a water pump according to one of the embodiments of the present application.

[0025] Figure 3 It is a third flow chart of a method for testing the operating efficiency of a water pump according to one of the embodiments of the present application.

[0026] Figure 4 It is a fourth flow chart of a method for testing the operating efficiency of a water pump according to one of the embodiments of the present application.

[0027] Figure 5 It is a fifth flow chart of a method for testing the operating efficiency of a water pump according to one of the embodiments of the present application. DETAILED DESCRIPTION

[0028] In order to make the purpose, technical solutions and advantages of this application more clear, the following Figure 1-5 It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.

[0029] With the development of smart building technology, efficient water pump energy efficiency testing methods can not only ensure that the HVAC system meets the building comfort requirements, but also minimize energy consumption. By monitoring the efficiency of water pumps under different operating conditions, including energy efficiency performance under different loads, different temperatures and humidity conditions, the operation and maintenance team can determine the cause of inefficiency, such as due to failure of the water pump itself, improper system configuration, or external environmental factors.

[0030] Therefore, there is an urgent need for an efficient pump operation efficiency test method to comprehensively monitor and analyze the pump operation status and environmental parameters, improve data utilization efficiency, and have efficient fault prediction capabilities and the ability to provide support for maintenance decisions. The embodiment of the present application discloses a method for testing the operating efficiency of a water pump.

[0031] Reference Figure 1 , a method for testing the operating efficiency of a water pump, the testing method comprising: Step S101, obtaining the operating status information and environmental parameter information of the water pump to be tested; In some embodiments, operating status information such as flow, pressure, power, and environmental parameter information such as temperature and humidity can be collected in real time through a sensor array installed on the water pump system, such as a flow meter, pressure gauge, electric power meter, and temperature and humidity sensor.

[0032] Among them, these data are the basis for subsequent analysis, ensuring the comprehensiveness and accuracy of the water pump test and providing necessary input for subsequent data analysis and fault prediction; Step S102, calculating the pump efficiency data for each time period according to the operation status information; Among them, by processing the collected operating status information, the water pump efficiency is calculated for each preset time period (such as every hour, every day, etc.). This step can be achieved through the efficiency calculation formula, which can dynamically reflect the changes in the working efficiency of the water pump.

[0033] Step S103, constructing a time series based on the pump efficiency data of each time period, analyzing and identifying trend components and periodic characteristics in the time series; Among them, the water pump efficiency data of continuous time periods are arranged to form a time series, and time series analysis methods (such as ARIMA model, seasonal decomposition, etc.) can be used to identify long-term trends in the data (such as the gradual decrease in efficiency over time) and periodic fluctuations (such as the difference in efficiency between weekdays and weekends).

[0034] Understandably, this process helps to understand the pattern of efficiency changes, reveals the inherent laws of pump efficiency changes, and facilitates the distinction between normal fluctuations and abnormal changes.

[0035] Step S104, detecting abnormal points in the time series according to the trend components and periodic characteristics in the time series; Among them, the abnormal point may be a sudden change in efficiency caused by water pump failure or changes in the external environment.

[0036] In some embodiments, anomalies that deviate from normal trends and periodic patterns can be identified by statistical methods (such as Z-score, Grubbs test, etc.) or machine learning algorithms (such as isolation forest, DBSCAN clustering, etc.), which indicate potential efficiency decline or the onset of failure.

[0037] Step S105, determining the environmental parameter information and water pump efficiency data corresponding to the abnormal point, and inputting them into the pre-trained fault prediction model to obtain the fault type prediction result and the corresponding prediction probability; Among them, environmental parameters and efficiency data related to the abnormal points are selected as input features and input into a fault prediction model that has been pre-trained based on machine learning or deep learning technology. The model can predict the fault type (such as mechanical wear, electrical failure, etc.) and the probability of fault occurrence based on these features, providing a quantitative basis for decision-making.

[0038] Step S106, determining the fault test result of the water pump to be tested according to the fault type prediction result with the highest prediction probability; Among them, from the prediction results of multiple fault types, the fault type with the highest prediction probability is selected as the final diagnosis result. This "maximum a posteriori probability" principle reduces the possibility of misjudgment, ensures the reliability of the diagnosis results, and helps to carry out repairs or replacement of parts more targetedly.

[0039] Step S107, sending the fault test result of the water pump to be tested to the management terminal.

[0040] In some embodiments, the fault test results can be transmitted instantly to the terminal device of the maintenance manager, such as a computer or mobile APP, through network communication technology, so that the manager can obtain information in time and take action to arrange maintenance or adjust operation strategies, thereby reducing downtime and maintenance costs.

[0041] In the above implementation, the operating status and environmental parameters of the water pump are monitored in real time, time series analysis is used to identify the efficiency change trend, and combined with the machine learning model, intelligent diagnosis is performed on abnormal situations of efficiency decline, and potential fault types and their probabilities are predicted. Finally, the diagnosis results are fed back to the management terminal, thereby achieving preventive maintenance and optimized operations. This application can better understand and analyze the reasons for changes in water pump efficiency by comprehensively considering environmental parameters and water pump operating status. Through intelligent data processing and decision support, the efficiency of water pump energy efficiency testing is improved, and early detection and accurate diagnosis of potential faults are achieved. At the same time, the stability of the HVAC system and the comfort of the indoor environment are ensured, meeting the modern building needs of sustainable development and energy conservation and emission reduction.

[0042] As an implementation of step S102, the step of calculating the pump efficiency data for each time period according to the operating status information includes: According to the operating status information, the pump outlet pressure P in each period is obtained by analysis. out , Pump inlet pressure P in , the height difference H between the inlet and outlet pressure gauges of the water pump, the real-time flow rate G through the water pump and the real-time power W of the water pump; Based on the preset calculation formula, the real-time efficiency η of the water pump in each period is calculated as: η=G*(P out -P in +ρ*g*H*10 -6 ) / (3.6 * W); In the above formula, ρ is the density of cold (hot) water, and g is the acceleration due to gravity.

[0043] Among them, the pump outlet pressure P out and the pump inlet pressure P in The unit is MPa, the unit of the height difference H of the pump inlet and outlet pressure gauges is m, and the unit of the real-time flow G through the water pump is m 3 / h, the real-time power W of the water pump is in kw, and the density ρ of cold (hot) water is in kg / m 3 , the acceleration due to gravity g is 9.81m / s.

[0044] In the above implementation, the preset efficiency calculation formula is converted into a computer-executable algorithm, and the real-time efficiency of the water pump is accurately calculated to provide a data basis for subsequent time series analysis and fault prediction.

[0045] Reference Figure 2 As an implementation method of step S103, the steps of constructing a time series according to the pump efficiency data of each time period, and analyzing and identifying the trend components and periodic characteristics in the time series include: Step S201, pre-processing the water pump efficiency data of each time period and arranging them in chronological order to obtain a time series data set; In some embodiments, the preprocessing step may include removing invalid or outlier values, such as extreme values ​​due to measurement errors, and filling in missing data, typically using interpolation.

[0046] The data are arranged in chronological order to form an ordered time series data set. The data at each time point contains the pump efficiency value of that period, and the timestamp is accurate to minutes or hours to meet the requirements of time resolution in different scenarios. Both short-term fluctuations and long-term trends can be accurately captured.

[0047] Step S202, extracting statistical features from the time series data set to obtain a statistical feature set; Among them, statistical characteristics such as mean, standard deviation, maximum value, minimum value, etc. reflect the basic characteristics of the data; by calculating these characteristics, we can initially understand the central trend, degree of dispersion and range of the time series, help understand the overall distribution of the data, provide key indicators for subsequent analysis, and facilitate the identification of anomalies and potential patterns in the data.

[0048] Step S203, calculating the rolling statistics in the time series data set according to the statistical feature set to obtain a rolling statistics set; Among them, rolling statistics are statistics calculated by sliding windows on time series, such as moving average, moving standard deviation, etc. This step helps to reveal the changing trend of data over time, reduce the impact of random fluctuations, and more clearly show long-term behavior.

[0049] It is understandable that by smoothing the data, the dynamic changes of the data are highlighted, the trends and periodicity are made more obvious, and a stable basis is provided for decomposition and prediction.

[0050] Step S204, performing time series decomposition on the time series data set according to the rolling statistics set, and extracting trend components; Among them, the time series is decomposed using information such as rolling statistics, usually into trend, seasonal and random components; by separating the trend component, it helps to identify the long-term change trend of system performance without being disturbed by short-term fluctuations, which is crucial for predicting future efficiency changes and formulating maintenance strategies.

[0051] Step S205, correcting the time series data set according to the trend component, identifying and eliminating long-term nonlinear trends or structural changes; Among them, by comparing the original data and trend components, possible nonlinear growth or structural changes, such as efficiency decline caused by equipment aging, can be identified, and then such influences can be corrected to ensure the accuracy of trend analysis and avoid misjudgment due to atypical factors.

[0052] Step S206, identifying the corrected time series data set based on Fourier transform to obtain periodic features.

[0053] Specifically, Fourier transform is a method that converts time domain signals into frequency domain signals, which can decompose complex time series into multiple frequency components. By analyzing the transformed spectrum, periodic features such as daily, weekly or seasonal cycles can be identified. The identification of periodic features helps to understand the repetitive patterns of water pump efficiency data in different time periods, providing a basis for predicting and optimizing water pump performance.

[0054] In the above implementation, a time series is constructed through preprocessing, and trend components and periodic characteristics are separated using statistical analysis and Fourier transform techniques, which effectively identifies long-term patterns and short-term fluctuations in efficiency changes, thereby improving the accuracy and practicality of data analysis. This is of great significance for improving the efficiency of energy efficiency management of water pump systems and the scientific nature of maintenance decisions.

[0055] Reference Figure 3 As an implementation of step S104, the step of detecting abnormal points in the time series according to the trend component and periodic characteristics in the time series includes: Step S301, establishing a baseline model according to the trend component and periodicity characteristics in the time series; Among them, the trend components and periodic characteristics analyzed from the time series data are used to construct a baseline model, which represents the expected behavior of the pump efficiency under normal operating conditions. It serves as a benchmark for subsequent deviation analysis, enabling deviations to be quantitatively identified and compared, providing a clear "normal" boundary for identifying anomalies.

[0056] Step S302, based on the baseline model, determining the deviation of the time series data set to obtain a deviation value set; Among them, by comparing the original time series data with the predicted value of the baseline model, the difference between the actual observation value and the predicted value of each data point, that is, the deviation value, is calculated to quantify the degree of deviation of the data point from the expected behavior, forming a set of deviation values, thereby intuitively showing which time periods have deviated from the normal range predicted by the baseline model, laying the foundation for the subsequent identification of anomalies.

[0057] Step S303, identifying data points in the deviation value set that exceed a preset deviation threshold, and obtaining a potential outlier set; In some embodiments, a reasonable deviation threshold can be set, which is usually a multiple of the standard deviation obtained based on statistical analysis of historical data. This threshold is used to screen out data points in the deviation value set that significantly deviate from the normal range. These points are regarded as potential outliers. By quickly screening out possible outliers, the scope for further detailed analysis is narrowed.

[0058] Step S304, performing periodicity check on the potential outlier set according to the periodic characteristics, and excluding potential outliers consistent with the periodic characteristics; Among them, considering that some seemingly abnormal points may actually be part of a periodic pattern, the periodic features are used to verify the set of potential abnormal points, and points consistent with the known periodic pattern are excluded because they belong to the expected normal fluctuation range. This enhances the accuracy of anomaly detection, avoids false positive anomaly judgments caused by periodic factors, and makes the final anomaly points more reliably represent the real abnormal situation.

[0059] Step S305 , performing trend consistency check on the periodically checked potential outlier point set according to the trend component, and determining the potential outlier points in the potential outlier point set that are contrary to the trend component as outliers in the time series.

[0060] Among them, the trend consistency check is performed on the set of potential abnormal points after periodic verification to check whether these points are contrary to the identified trend components. If a point not only deviates from the current periodic pattern but also violates the long-term trend, it is confirmed as a real abnormal point.

[0061] Understandably, trend analysis further identifies outliers, ensuring that they truly reflect unexpected changes in pump efficiency and are not part of a long-term trend.

[0062] It should be noted that actual time series data may contain nonlinear or non-periodic complex patterns, which may not be fully captured by the baseline model. Using periodicity verification and trend analysis as verification steps ensures the accuracy and applicability of the baseline model and eliminates abnormal situations that may be missed by the model.

[0063] In the above implementation, after multi-dimensional analysis of periodicity and trends, the comprehensiveness and accuracy of abnormal point determination are ensured, and abnormal points that deviate from short-term rules and affect long-term trends can be effectively identified, thereby providing strong support for subsequent fault diagnosis, performance optimization and decision-making. This technical solution is particularly suitable for scenarios in this application that require high-reliability monitoring of water pump efficiency, and can promptly detect and respond to potential performance degradation or failures, thereby reducing downtime and improving the overall efficiency of the system.

[0064] Reference Figure 4 As a further implementation of the test method, the test method also includes a training step of a fault prediction model, and the training step includes: Step S401, obtaining a historical data set and performing data preprocessing; Among them, the historical data set includes multiple sets of historical environmental parameter information and historical water pump efficiency data; In some embodiments, a large amount of records containing historical environmental parameter information (such as temperature, humidity, water quality, etc.) and historical water pump efficiency data (such as flow, pressure, power, etc.) are collected to form a historical data set.

[0065] Specifically, the data preprocessing step includes data cleaning to remove incomplete, erroneous or abnormal data records to ensure the quality of model training; in addition, the data can also be normalized or standardized, such as scaling the numerical range to the same scale, which helps to improve the stability and convergence speed of model training.

[0066] Step S402, based on the historical fault database, assigning fault type labels to multiple groups of historical environmental parameter information and historical water pump efficiency data in the historical data set after data preprocessing, to obtain a sample data set; Among them, according to the pump failure instances recorded in the historical fault database, each set of preprocessed historical environmental parameter information and pump efficiency data is associated with the corresponding fault type, such as "no fault", "bearing wear", "impeller blockage", etc., so as to construct a sample data set with target variables.

[0067] It can be understood that this step enables the model to learn the relationship between different fault modes and corresponding features, which is the basis for the model to perform classification prediction and enhances the pertinence and practicality of fault prediction.

[0068] Step S403, extracting features based on the sample data set to identify a feature set that is highly correlated with the water pump failure; Among them, from the sample data set, statistical analysis, correlation test and other methods are used to identify the feature set most closely related to water pump failure. These features may include specific efficiency parameter change trends, the relationship between pressure and flow, etc., which are used as input variables of the model to streamline the feature set, reduce model complexity, improve training efficiency, and avoid overfitting, so that the model can focus more on key fault indication signals and improve prediction accuracy.

[0069] Step S404, dividing the sample data set after feature extraction into a training sample set and a verification sample set; Among them, the sample data after feature extraction is randomly divided into two parts, one part is used for model training, and the other part is used to verify the model performance, so as to simulate the performance of the model on unknown data and prevent the model from overfitting the training data, so as to objectively evaluate the generalization ability of the model and ensure that the model not only performs well on the training data, but also maintains stable and reliable prediction performance on new data.

[0070] Step S405, inputting the training sample set into a pre-built machine learning model for initial training, optimizing model parameters, and obtaining an initially trained fault prediction model; In one embodiment of the present application, the machine learning model may adopt an integrated learning algorithm, such as a gradient boosted tree (GBM) algorithm, which performs well in processing complex interactions and nonlinear relationships between features and is particularly suitable for anomaly detection and fault prediction tasks.

[0071] During the initial training process, the model parameters (such as learning rate, regularization parameters, etc.) are iteratively adjusted to minimize the loss function and optimize the model performance, so that the model can fit the training data as much as possible while avoiding overfitting, thereby improving the accuracy and stability of the prediction.

[0072] Step S406, verifying the fault prediction model and correcting model parameters based on the verification sample set until the loss function of the model meets a preset condition or the number of model iterations reaches a preset number, thereby obtaining a trained fault prediction model.

[0073] Specifically, the optimized GBM model is applied to the training sample set for training, and an independent validation sample set is used to monitor the model performance. Evaluation indicators include but are not limited to accuracy, recall rate, F1 score, and area under the ROC curve (AUC-ROC). The model parameters are further fine-tuned according to the validation results until the loss function of the model reaches the preset threshold or the number of iterations reaches the upper limit.

[0074] In the above implementation, after feature extraction and data set division, the model can learn the most valuable information. Through parameter optimization in the model training and verification process, the generalization ability and prediction accuracy of the model are guaranteed. The final fault prediction model can provide scientific and accurate guidance for the operation and maintenance of the water pump, effectively prevent and reduce the occurrence of faults, reduce maintenance costs, and improve system reliability.

[0075] Reference Figure 5 As a further implementation of the test method, after the step of sending the fault test result of the water pump to be tested to the management terminal, it also includes: Step S501, receiving feedback information sent by the management terminal indicating that the fault test result is incorrect; Among them, when the management terminal receives the fault test result of the water pump to be tested, if it is found that the test result does not match the actual situation, feedback information can be sent to the tester.

[0076] Understandably, this feedback step reflects the interactivity and dynamic adaptability of the system, allowing manual or automated supervision to correct possible misjudgments, ensuring the accuracy of test results, timely detecting and correcting deviations in model predictions, and improving the reliability of the system.

[0077] Step S502, extracting environmental parameter information and water pump efficiency data of the abnormal point corresponding to the fault test result according to the feedback information; Based on the feedback received, the system can locate specific cases where the test results are incorrect and extract the corresponding environmental parameters and pump efficiency data. By accurately locating the problem, it can analyze why the model misjudged under specific conditions and provide specific and valuable reference data for subsequent model optimization.

[0078] Step S503, assigning fault type labels to the environmental parameter information and the water pump efficiency data according to the feedback information, and generating a feedback sample set; In some embodiments, the correct fault type label is reassigned to the extracted data based on the correct fault type information provided by the management terminal; by adding these "known errors-correct labels" data, the learning samples of the model are enriched, especially for scenarios with previous prediction errors, which helps to improve the judgment accuracy of the model when facing similar situations in the future.

[0079] Step S504, iteratively train the fault prediction model according to the feedback sample set, correct the model parameters, obtain the corrected fault prediction model and redeploy it.

[0080] Among them, the original fault prediction model is re-trained using the feedback sample set, and the internal parameters of the model are adjusted through the algorithm to reduce the prediction error. The training process may include multiple iterations until the model performance indicators reach the optimization standard.

[0081] In the above implementation, a feedback mechanism is added on the basis of the original fault prediction model training process, so that the model can perform self-optimization according to the feedback of actual test results. By learning these clearly labeled feedback data, the prediction logic of the model is gradually corrected, thereby improving the generalization ability and adaptability of the model, making it more able to accurately predict various complex fault conditions.

[0082] This application realizes a closed-loop fault prediction and optimization process by introducing a feedback mechanism. This dynamic adjustment capability enables the system to continuously improve in actual operation and adapt to wider and more complex working conditions, thereby effectively reducing operating costs and maintenance frequency caused by water pump failures, and improving the level of intelligence of equipment maintenance management.

[0083] The embodiment of the present application also discloses a water pump operation efficiency tester.

[0084] A water pump operation efficiency tester, the tester comprising: An acquisition module is used to obtain the operating status information and environmental parameter information of the water pump to be tested; A water pump efficiency calculation module is used to calculate the water pump efficiency data for each period according to the operation status information; The time series analysis module is used to construct a time series based on the pump efficiency data of each period, analyze and identify the trend components and periodic characteristics in the time series; Anomaly detection module, used to detect anomalies in time series based on trend components and periodic characteristics in time series; The fault prediction module is used to determine the environmental parameter information and water pump efficiency data corresponding to the abnormal point, and input them into the pre-trained fault prediction model to obtain the fault type prediction result and the corresponding prediction probability; A fault test result determination module is used to determine the fault test result of the water pump to be tested according to the fault type prediction result with the highest prediction probability; The fault test result sending module is used to send the fault test result of the water pump to be tested to the management terminal.

[0085] In the above implementation, by comprehensively considering environmental parameters and the operating status of the water pump, the reasons for changes in the water pump efficiency can be better understood and analyzed. Through intelligent data processing and decision support, the efficiency of the water pump energy efficiency test is improved, and early detection and accurate diagnosis of potential faults are achieved. At the same time, the stability of the HVAC system and the comfort of the indoor environment are ensured, meeting the modern building needs of sustainable development and energy conservation and emission reduction.

[0086] In some embodiments, the water pump operation efficiency tester of the present application has data testing, calculation, storage and display functions. It can not only display real-time flow, inlet and outlet pressures and real-time efficiency, but also has a remote transmission interface for remote access, allowing managers to monitor the water pump operation efficiency and operating status at any location, thereby providing users with a more convenient and intuitive operation experience.

[0087] As a further embodiment of the tester, the tester further includes: A feedback information receiving module, used to receive feedback information sent by the management terminal indicating that the fault test result is incorrect; An extraction module, used to extract environmental parameter information and water pump efficiency data of abnormal points corresponding to the fault test results according to the feedback information; A feedback sample set generation module is used to assign fault type labels to environmental parameter information and water pump efficiency data according to feedback information to generate a feedback sample set; The model correction module is used to iteratively train the fault prediction model according to the feedback sample set, correct the model parameters, obtain the corrected fault prediction model and redeploy it.

[0088] In the above implementation, the model can self-optimize according to the feedback of actual test results. By learning these clearly labeled feedback data, the prediction logic of the model is gradually corrected, thereby improving the generalization ability and adaptability of the model, making it more accurate in predicting various complex fault conditions.

[0089] A water pump operation efficiency tester in an embodiment of the present application can implement any of the above-mentioned testing methods, and the specific working process of each module in the tester can refer to the corresponding process in the above-mentioned method embodiment.

[0090] In the several embodiments provided in this application, it should be understood that the provided methods and systems can be implemented in other ways. For example, the system embodiments described above are only illustrative; for example, the division of a certain module is only a logical function division, and there may be other division methods in actual implementation, such as multiple modules can be combined or integrated into another system, or some features can be ignored or not executed.

[0091] The embodiment of the present application also discloses a computer device.

[0092] The computer device comprises a memory, a processor and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, the water pump operation efficiency test method as described above is implemented.

[0093] The embodiment of the present application also discloses a computer-readable storage medium.

[0094] A computer-readable storage medium stores a computer program that can be loaded by a processor and execute any one of the above-mentioned water pump operation efficiency testing methods.

[0095] Among them, computer-readable storage media can be any tangible medium that contains or stores a program, which can be used by or in conjunction with an instruction execution system, apparatus or device; the program code contained on the computer-readable medium can be transmitted using any appropriate medium, including but not limited to wireless, wire, optical cable, RF, etc., or any suitable combination of the above.

[0096] It should be noted that in the above embodiments, the description of each embodiment has different emphases, and for parts that are not described in detail in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0097] The above are all preferred embodiments of the present application, and are not intended to limit the protection scope of the present application. Any feature disclosed in this specification (including the abstract and drawings), unless otherwise stated, can be replaced by other equivalent or alternative features with similar purposes. That is, unless otherwise stated, each feature is only an example of a series of equivalent or similar features.

Claims

1. A method for testing the operating efficiency of a water pump, characterized in that: The test method includes: Obtaining the operating status information and environmental parameter information of the water pump to be tested; Calculate the water pump efficiency data for each time period according to the operating status information; Constructing a time series based on the pump efficiency data for each period, analyzing and identifying trend components and periodic features in the time series; Detecting abnormal points in the time series based on trend components and periodic characteristics in the time series; Determine the environmental parameter information and water pump efficiency data corresponding to the abnormal point, and input them into a pre-trained fault prediction model to obtain a fault type prediction result and a corresponding prediction probability; Determining the fault test result of the water pump to be tested according to the fault type prediction result with the highest prediction probability; The fault test result of the water pump to be tested is sent to the management terminal.

2. A method for testing the operating efficiency of a water pump according to claim 1, characterized in that: The step of calculating the pump efficiency data for each time period according to the operating status information comprises: According to the operating status information, the pump outlet pressure P in each period is obtained by analysis. out , Pump inlet pressure P in , the height difference H between the inlet and outlet pressure gauges of the water pump, the real-time flow rate G through the water pump and the real-time power W of the water pump; Based on the preset calculation formula, the real-time efficiency η of the water pump in each period is calculated as: η=G*(P out -P in +p* g * H * 10 -6 ) / (3.6 * W); In the above formula, ρ is the density of cold (hot) water, and g is the acceleration due to gravity.

3. A method for testing the operating efficiency of a water pump according to claim 1, characterized in that: The steps of constructing a time series based on the pump efficiency data of each period, and analyzing and identifying trend components and periodic characteristics in the time series include: Preprocessing the water pump efficiency data of each time period and arranging them in chronological order to obtain a time series data set; Extracting statistical features from the time series data set to obtain a statistical feature set; Calculating rolling statistics in the time series data set according to the statistical feature set to obtain a rolling statistics set; According to the rolling statistics set, the time series data set is decomposed into time series to extract trend components; Correcting the time series data set according to the trend component to identify and remove long-term nonlinear trends or structural changes; The corrected time series data set is identified based on Fourier transform to obtain periodic features.

4. A method for testing the operating efficiency of a water pump according to claim 3, characterized in that: According to the trend component and the periodicity characteristics in the time series, the step of detecting the abnormal point in the time series comprises: Establishing a baseline model based on the trend component and periodic characteristics in the time series; Based on the baseline model, determining the deviation of the time series data set to obtain a deviation value set; Identify data points in the deviation value set that exceed a preset deviation threshold to obtain a potential outlier point set; Performing periodic verification on the set of potential abnormal points according to the periodic characteristics, and excluding potential abnormal points consistent with the periodic characteristics; A trend consistency check is performed on the periodically checked potential abnormal point set according to the trend component, and potential abnormal points in the potential abnormal point set that are contrary to the trend component are determined as abnormal points in the time series.

5. A method for testing the operating efficiency of a water pump according to claim 1, characterized in that: The testing method further comprises a training step of a fault prediction model, wherein the training step comprises: Acquire a historical data set and perform data preprocessing; wherein the historical data set includes multiple sets of historical environmental parameter information and historical water pump efficiency data; Based on the historical fault database, assigning fault type labels to multiple groups of historical environmental parameter information and historical water pump efficiency data in the historical data set after data preprocessing to obtain a sample data set; Perform feature extraction based on the sample data set to identify the feature set that is highly correlated with water pump failure; Divide the sample data set after feature extraction into a training sample set and a validation sample set; Inputting the training sample set into a pre-built machine learning model for initial training, optimizing model parameters, and obtaining the initially trained fault prediction model; The fault prediction model is verified and model parameters are corrected based on the verification sample set until the loss function of the model meets a preset condition or the number of model iterations reaches a preset number, thereby obtaining a trained fault prediction model.

6. A method for testing the operating efficiency of a water pump according to any one of claims 1 to 5, characterized in that: After the step of sending the fault test result of the water pump to be tested to the management terminal, the method further includes: Receiving feedback information sent by the management terminal indicating that the fault test result is incorrect; Extracting environmental parameter information and water pump efficiency data of the abnormal point corresponding to the fault test result according to the feedback information; According to the feedback information, a fault type label is assigned to the environmental parameter information and the water pump efficiency data to generate a feedback sample set; The fault prediction model is iteratively trained according to the feedback sample set, model parameters are corrected, and a corrected fault prediction model is obtained and redeployed.

7. A water pump operating efficiency tester, characterized in that: The tester comprises: An acquisition module is used to obtain the operating status information and environmental parameter information of the water pump to be tested; A water pump efficiency calculation module, used to calculate the water pump efficiency data for each time period according to the operating status information; A time series analysis module, used to construct a time series based on the pump efficiency data of each time period, and to analyze and identify trend components and periodic characteristics in the time series; An abnormal point detection module, used for detecting abnormal points in the time series according to trend components and periodic characteristics in the time series; A fault prediction module is used to determine the environmental parameter information and water pump efficiency data corresponding to the abnormal point, and input them into a pre-trained fault prediction model to obtain a fault type prediction result and a corresponding prediction probability; A fault test result determination module, used to determine the fault test result of the water pump to be tested according to the fault type prediction result with the highest prediction probability; The fault test result sending module is used to send the fault test result of the water pump to be tested to the management terminal.

8. A water pump operating efficiency tester according to claim 7, characterized in that: The tester comprises: A feedback information receiving module, used for receiving feedback information sent by the management terminal indicating that the fault test result is incorrect; An extraction module, used to extract environmental parameter information and water pump efficiency data of the abnormal point corresponding to the fault test result according to the feedback information; A feedback sample set generation module, used to assign fault type labels to the environmental parameter information and water pump efficiency data according to the feedback information, and generate a feedback sample set; The model correction module is used to iteratively train the fault prediction model according to the feedback sample set, correct the model parameters, obtain the corrected fault prediction model and redeploy it.

9. A computer device, characterized in that: The method comprises a memory, a processor and a computer program stored in the memory and executable on the processor, wherein the processor implements the method according to any one of claims 1 to 6 when executing the program.

10. A computer-readable storage medium, characterized in that: A computer program is stored which can be loaded by a processor and execute the method according to any one of claims 1 to 6.

Citation Information

Cited By

  • High-pressure water pump motor remote control method based on data acquisition

    CN119288820A

  • Data center hydraulic analysis method and system

    CN119312594A

  • Predictive maintenance method and system for wind turbine

    CN119379264A

  • A method and system for predictive maintenance of a wind turbine

    CN119379264B