Numerical control machine tool motor operation monitoring method and system based on data analysis

Abstract

The invention discloses a numerical control machine tool motor operation monitoring method and system based on data analysis, which are used in the field of motor monitoring and fault diagnosis, and the method comprises the steps: collecting working condition data corresponding to the operation process of a numerical control machine tool motor; establishing a temperature change trend model, and predicting temperature change conditions of the motor of the numerical control machine tool under different working conditions; combining the extracted frequency and amplitude with the prediction result of the temperature trend model, and judging the current operation state of the numerical control machine tool motor; and setting a threshold value of the working condition data, comparing a classification analysis result with a preset threshold value, and automatically triggering a multi-stage early warning mechanism when monitoring that the working condition data of the motor of the numerical control machine tool exceeds the threshold value. According to the method, preventative maintenance is realized, sudden faults are avoided, the motor is ensured to operate under efficient and stable conditions by determining the optimal electrical working range and real-time monitoring data, energy consumption is reduced, the service life of equipment is prolonged, and intelligent motor state analysis and prediction are realized.

Description

Technical Field

[0001] The present invention relates to the field of motor monitoring and fault diagnosis, and in particular to a method and system for monitoring the operation of a motor of a numerically controlled machine tool based on data analysis. Background Art

[0002] The motor of a CNC machine tool is one of the key components of a CNC machine tool. It is responsible for providing power to the machine tool and driving it to perform processing tasks. A CNC machine tool (CNC, Computer Numerical Control) is an automated machine tool that controls the operation of mechanical tools through programming instructions. It is widely used in the manufacturing industry, especially in metal processing, automobile manufacturing, aerospace and other fields. The motor is generally used to drive the spindle, feed system and auxiliary system in a CNC machine tool.

[0003] The motor is the core of the machine tool to achieve precision machining. The performance of CNC machine tools (including machining accuracy, speed, efficiency, etc.) is highly dependent on the status and operation quality of the motor. If the motor fails, it will directly cause the machine tool to stop, affect the production progress, and may cause product quality problems.

[0004] The operating environment of CNC machine tool motors is complex. Motors often operate at high speed, high load, and for a long time. This may cause various problems (such as overheating, vibration, overload, etc.) in the long term. Therefore, motor status monitoring and health management have become very necessary. The main goal of monitoring is to achieve preventive maintenance, that is, to detect and deal with problems in advance before they evolve into failures, so as to avoid unnecessary downtime and maintenance costs.

[0005] In the existing technology, the monitoring of CNC machine tool motors often relies on a single parameter, which makes it impossible to fully grasp the overall health status of the motor. Traditional maintenance strategies are usually based on regular maintenance at fixed cycles, which cannot be dynamically adjusted according to the actual health status of the equipment. Unnecessary maintenance is often performed when the equipment has not yet had any problems, or repairs are performed only when the problems are serious. This not only increases maintenance costs, but may also cause equipment failures in unforeseen circumstances. Abnormality identification usually relies on a simple threshold trigger mechanism and lacks intelligent classification and prediction capabilities. Therefore, although motor abnormalities may be detected, the type of abnormality cannot be refined, resulting in non-targeted treatment measures, low problem solving efficiency, and even further damage to the equipment.

[0006] Currently, no effective solution has been proposed for the problems in the related technologies. Summary of the invention

[0007] In order to overcome the above problems, the present invention aims to propose a method and system for monitoring the operation of CNC machine tool motors based on data analysis, with the aim of solving the problem that abnormality identification usually relies on a simple threshold trigger mechanism and lacks intelligent classification and prediction capabilities. Therefore, although motor abnormalities may be detected, the type of abnormality cannot be refined, resulting in non-targeted treatment measures, low problem-solving efficiency, and even further damage to the equipment.

[0008] To this end, the specific technical solution adopted by the present invention is as follows:

[0009] According to one aspect of the present invention, a method for monitoring the operation of a motor of a CNC machine tool based on data analysis is provided, and the method for monitoring the operation of a motor of a CNC machine tool comprises:

[0010] S1. Collecting the working condition data corresponding to the operation process of the CNC machine tool motor, and the working condition data includes electrical data, temperature data, mechanical data and vibration data;

[0011] S2. Based on the collected electrical and mechanical data, determine the optimal fluctuation range and optimal working range of the electrical data, and establish a temperature change trend model by correlating the optimal fluctuation range and optimal working range of the electrical data with the changes in temperature data to predict the temperature changes of CNC machine tool motors under different working conditions; use signal processing technology to extract frequency and amplitude from vibration data;

[0012] S3, according to the preset CNC machine tool motor health status standard, the extracted frequency, amplitude and temperature trend model prediction results are combined to determine the current operating status of the CNC machine tool motor through classification analysis;

[0013] S4. Set the threshold of the working condition data, compare the classification analysis results with the preset threshold, and automatically trigger the multi-level early warning mechanism when the working condition data of the CNC machine tool motor is monitored to exceed the threshold.

[0014] Optionally, collecting operating data corresponding to the operation process of the motor of the CNC machine tool, wherein the operating data includes electrical data, temperature data, mechanical data and vibration data, comprises the following steps:

[0015] S11, using a pre-configured current sensor and voltage sensor to collect current and voltage at a preset frequency to obtain corresponding electrical data;

[0016] S12, using pre-configured thermocouples and platinum resistance temperature detection sensors to monitor the internal and external temperatures of the CNC machine tool motor in real time to obtain corresponding temperature data;

[0017] S13, using a pre-configured rotary encoder and torque sensor to collect the speed and torque of the CNC machine tool motor to obtain corresponding mechanical data;

[0018] S14, using a pre-configured acceleration sensor to monitor the acceleration of the motor vibration of the CNC machine tool in real time, and extracting the frequency and amplitude through signal processing to obtain corresponding vibration data;

[0019] S15, synchronously processing and storing the electrical data, temperature data, mechanical data and vibration data.

[0020] Optionally, based on the collected electrical data, temperature data and mechanical data, the optimal working range of the electrical data is determined, and the changes in the electrical data and temperature are analyzed by correlation, a temperature change trend model is established, and the temperature change of the CNC machine tool motor under different working conditions is predicted; extracting the frequency and amplitude from the vibration data using signal processing technology includes the following steps:

[0021] S21, determining the optimal electrical data range for the operation of the CNC machine tool motor by statistically analyzing the current data and voltage data in the acquired electrical data and comparing the electrical data with historical data;

[0022] S22, performing correlation analysis on the acquired temperature data and the current data and voltage data in the electrical data, and using a multivariate regression analysis method to analyze the impact of electrical data fluctuations on temperature changes;

[0023] S23. Based on the correlation analysis results of electrical data and temperature data, a temperature change trend model is constructed;

[0024] S24, performing spectrum analysis on the acquired vibration data using fast Fourier transform to extract the frequency and amplitude of the vibration;

[0025] S25. Integrate the extracted electrical data, temperature data, frequency and amplitude to form a comprehensive analysis of the working status of the CNC machine tool motor, and combine with the temperature change trend model to predict the health status of the CNC machine tool motor under different working conditions in the future.

[0026] Optionally, correlating the acquired temperature data with the current and voltage data in the electrical data and analyzing the influence of the electrical data fluctuation on the temperature change by using a regression analysis method comprises the following steps:

[0027] S221, using the acquired current data and voltage data as independent variables; using the acquired temperature data as a dependent variable, and arranging the current data, voltage data and temperature data into a time series;

[0028] S222, using a multiple regression model to construct a relationship between the temperature data and the current data and the voltage data;

[0029] S223, fitting the multivariate regression model using the least squares method to calculate the regression coefficients of current and voltage to temperature changes;

[0030] S224, calculating the coefficient of determination, and evaluating the goodness of fit of the multiple regression model based on the coefficient of determination;

[0031] S225. According to the signs and magnitudes of the regression coefficients of the current and voltage to the temperature change, the specific effects of the current and voltage fluctuations on the temperature change are obtained.

[0032] Optionally, calculating the coefficient of determination and evaluating the goodness of fit of the multiple regression model based on the coefficient of determination comprises the following steps:

[0033] S2241. According to the multiple regression model, the values ​​of the independent variables are used to calculate the predicted values ​​of the dependent variable;

[0034] S2242, calculate the difference between the actual value and the predicted value of the dependent variable;

[0035] S2243, calculating the total sum of squares to measure the total variation in temperature, and calculating the regression sum of squares to measure the temperature variation explained by current and voltage;

[0036] S2244, calculate the residual sum of squares to measure the temperature variation not explained by current and voltage;

[0037] S2245. Calculate the coefficient of determination based on the regression sum of squares and the total sum of squares to evaluate the goodness of fit of the multivariate regression model to the temperature data.

[0038] Optionally, the expression for calculating the coefficient of determination is:

[0039]

[0040] In the formula, represents the predicted temperature value of the CNC machine tool motor at the i-th data point;

[0041] Q 2 represents the coefficient of determination;

[0042] Y i Indicates the actual temperature value of the CNC machine tool motor;

[0043] Indicates the average value of temperature;

[0044] n represents the total number of data points;

[0045] i represents a data point.

[0046] Optionally, the extracted electrical data, temperature data, frequency and amplitude are integrated to form a comprehensive analysis of the working state of the CNC machine tool motor, and combined with the temperature change trend model, the future health state of the CNC machine tool motor under different working conditions is predicted, including the following steps:

[0047] S251, integrating the extracted current data, voltage data, temperature data, and the frequency and amplitude extracted from the vibration data;

[0048] S252, extracting the mean, maximum, standard deviation and change rate of electrical, temperature and vibration data;

[0049] S253, performing correlation analysis on the mean, maximum, standard deviation and change rate of the extracted electrical, temperature and vibration data to identify the working state of the CNC machine tool motor;

[0050] S254. Combined with the temperature change trend model, a prediction model for the motor health status is constructed using multivariate statistical analysis;

[0051] S255. Based on the identification of the working status and prediction model of the CNC machine tool motor, the health status of the CNC machine tool motor in the future operation process is simulated, and its temperature, vibration and electrical changes under different working conditions are predicted.

[0052] Optionally, according to a preset CNC machine tool motor health status standard, the extracted frequency, amplitude and prediction results of the temperature trend model are combined, and judging the current operating status of the CNC machine tool motor through classification analysis includes the following steps:

[0053] S31, extracting frequency data, amplitude data and temperature data from the operation data of the CNC machine tool motor, and scaling the extracted frequency data, amplitude data and temperature data respectively;

[0054] S32, designing a first-level classifier using a BP neural network, with the input being the scaled frequency, amplitude and temperature data;

[0055] S33, using known historical operation data and pre-set CNC machine tool motor health status standards to train the primary classifier, and testing the real-time operation data to determine the status of the CNC machine tool motor;

[0056] S34, according to the judgment result of the primary classifier, the preset judgment threshold, and the operation state of the CNC machine tool motor is divided into normal operation or abnormal state. When the judgment result is abnormal state, the abnormal analysis process is triggered and the step S35 is executed;

[0057] S35, selecting feature data sensitive to abnormal motor response from the primary feature extraction, and constructing a secondary feature vector;

[0058] S36, designing a new BP neural network as a secondary classifier, inputting a secondary feature vector, and outputting a specific classification of the abnormal state of the motor;

[0059] S37, using known abnormal data sets to train the secondary classifier and test its classification ability for different abnormal patterns;

[0060] S38. According to the output of the secondary classifier and the preset decision threshold, the abnormal state of the motor is refined into different levels, and corresponding maintenance or repair measures are taken.

[0061] Optionally, designing a new BP neural network as a secondary classifier, inputting a secondary feature vector, and outputting a specific classification of the abnormal state of the motor includes the following steps:

[0062] S361, selecting features that are more sensitive to motor abnormality from the first-level feature extraction, and constructing a second-level feature vector;

[0063] S362, determining a new BP neural network architecture, including an input layer, a hidden layer, and an output layer, wherein the number of input layer nodes matches the dimension of the secondary feature vector, and the number of output layer nodes corresponds to the specific classification number of the motor abnormal state;

[0064] S363, select an activation function and set training parameters;

[0065] S364, using the known abnormal data set to train the new BP neural network, and during the training process, monitoring the loss function and accuracy of the model in real time;

[0066] S365. Evaluate the trained BP neural network using the test data set and adjust the model architecture based on the test results.

[0067] S366. Deploy the trained BP neural network to the CNC machine tool motor operation monitoring to classify the abnormal status of the CNC machine tool motor in real time.

[0068] According to another aspect of the present invention, there is also provided a CNC machine tool motor operation monitoring system based on data analysis, the CNC machine tool motor operation monitoring system comprising: a data acquisition module, a data analysis and modeling module, a state classification and judgment module and an early warning trigger module;

[0069] The data acquisition module is used to collect the working condition data corresponding to the operation process of the CNC machine tool motor, and the working condition data includes electrical data, temperature data, mechanical data and vibration data;

[0070] The data analysis and modeling module is used to determine the optimal fluctuation range and optimal working range of electrical data based on the collected electrical and mechanical data, and to establish a temperature change trend model by correlating the optimal fluctuation range and optimal working range of electrical data with the changes in temperature data, so as to predict the temperature changes of CNC machine tool motors under different working conditions; and to extract frequency and amplitude from vibration data using signal processing technology;

[0071] The state classification and judgment module is used to combine the prediction results of the extracted frequency, amplitude and temperature trend model according to the preset CNC machine tool motor health status standard, and judge the current operation status of the CNC machine tool motor through classification analysis;

[0072] The early warning trigger module is used to set the threshold of the working condition data and compare the classification analysis results with the preset threshold. When the working condition data of the CNC machine tool motor is monitored to exceed the threshold, the multi-level early warning mechanism is automatically triggered.

[0073] Compared with the prior art, this application has the following beneficial effects:

[0074] 1. The present invention can comprehensively grasp the running status of the motor through the comprehensive collection and analysis of electrical, temperature, mechanical and vibration data, avoiding the limitations of single data analysis. Through statistical analysis and multivariate regression models, it can accurately identify the running status of the motor and trigger an early warning when it exceeds the optimal working range to prevent potential failures. Through the temperature change trend model and signal processing of vibration data, it can predict in advance the health status of the motor under different future working conditions, realize preventive maintenance, and avoid sudden failures. By determining the optimal electrical working range and real-time monitoring data, it ensures that the motor operates under efficient and stable conditions, reduces energy consumption and extends the service life of the equipment. By combining historical data, real-time data and statistical models, it realizes intelligent motor status analysis and prediction, and promotes the intelligent transformation of factory equipment maintenance.

[0075] 2. The primary and secondary classifiers implemented by the BP neural network in the present invention can quickly and effectively identify the operating status of the motor and classify abnormal situations, ensure prediction before the motor fails, and reduce the risk of sudden failures; classification can not only detect abnormalities, but also further refine the type of abnormalities (such as overheating, mechanical failure, etc.), which is helpful to formulate targeted processing measures for different types of abnormalities and improve the efficiency of problem solving. Combined with historical data and real-time data, through the training of neural networks, it can gradually learn and optimize the judgment of the motor status. The secondary classifier can accurately judge the abnormal level and specific situation, thereby providing data support for preventive maintenance and optimizing maintenance strategies. Through prediction and classification, factories can plan maintenance in advance to avoid sudden equipment downtime.

[0076] 3. The present invention can timely identify and respond to potential fault problems by setting thresholds and comparing monitoring data. The multi-level early warning mechanism provides flexibility to ensure that faults of different severity can be handled in a timely manner, thereby avoiding the spread or escalation of faults. Through threshold monitoring, protective measures can be automatically taken when the motor is overloaded, overheated or the mechanical vibration is abnormal to prevent serious damage to the equipment, thereby greatly improving the safety of CNC machine tool motor operation. BRIEF DESCRIPTION OF THE DRAWINGS

[0077] The above characteristics, features and advantages of the present invention and their implementation methods and methods will become more clearly understood in conjunction with the following description of the embodiments, which are described in detail in conjunction with the accompanying drawings. Herein, a schematic diagram is shown:

[0078] Figure 1 is a flow chart of a method for monitoring motor operation of a CNC machine tool based on data analysis according to an embodiment of the present invention;

[0079] Figure 2 The invention is a principle block diagram of a CNC machine tool motor operation monitoring system based on data analysis according to an embodiment of the invention.

[0080] In the figure:

[0081] 1. Data acquisition module; 2. Data analysis and modeling module; 3. Status classification and judgment module; 4. Early warning trigger module. DETAILED DESCRIPTION

[0082] In order to enable those skilled in the art to better understand the solution of the present application, the technical solution in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.

[0083] According to an embodiment of the present invention, a method and system for monitoring the operation of a motor of a CNC machine tool based on data analysis are provided.

[0084] The present invention is further described with reference to the accompanying drawings and specific embodiments. Figure 1 As shown, according to one embodiment of the present invention, a method for monitoring the operation of a motor of a CNC machine tool based on data analysis is provided, and the method for monitoring the operation of a motor of a CNC machine tool comprises:

[0085] S1. Collect the working condition data corresponding to the operation process of the CNC machine tool motor, and the working condition data includes electrical data, temperature data, mechanical data and vibration data.

[0086] Preferably, collecting the operating data corresponding to the operation process of the CNC machine tool motor, and the operating data includes electrical data, temperature data, mechanical data and vibration data, comprises the following steps:

[0087] S11, using a pre-configured current sensor and voltage sensor to collect current and voltage at a preset frequency to obtain corresponding electrical data;

[0088] S12, using pre-configured thermocouples and platinum resistance temperature detection sensors to monitor the internal and external temperatures of the CNC machine tool motor in real time to obtain corresponding temperature data;

[0089] S13, using a pre-configured rotary encoder and torque sensor to collect the speed and torque of the CNC machine tool motor to obtain corresponding mechanical data;

[0090] S14, using a pre-configured acceleration sensor to monitor the acceleration of the motor vibration of the CNC machine tool in real time, and extracting the frequency and amplitude through signal processing to obtain corresponding vibration data;

[0091] S15, synchronously processing and storing the electrical data, temperature data, mechanical data and vibration data.

[0092] It is necessary to explain that electrical data: current and voltage are collected at a preset frequency using current sensors and voltage sensors. Through these sensors, the electrical status of the motor can be monitored in real time to ensure that the current and voltage are running within a reasonable range and help diagnose electrical faults.

[0093] Temperature data: Use thermocouples and platinum resistance temperature detection sensors to monitor temperature changes inside and outside the motor. Temperature fluctuations can reflect the health of the motor. Too high or too low temperatures may indicate a motor failure or efficiency problem.

[0094] Mechanical data: Use rotary encoders and torque sensors to collect the motor's speed and torque in real time. These data can reflect the motor's mechanical load. Abnormal speed or torque may indicate mechanical failure or unbalanced operating conditions.

[0095] Vibration data: Use an accelerometer to monitor the acceleration of the motor vibration in real time. Through signal processing, the vibration frequency and amplitude of the motor can be extracted from the acceleration data. Vibration data is an important basis for diagnosing motor mechanical problems (such as imbalance, bearing failure, and mechanical looseness).

[0096] Data synchronization and storage: The collected electrical data, temperature data, mechanical data and vibration data are processed synchronously to ensure that different types of data have the same time base for subsequent analysis. All data are processed and stored uniformly for long-term monitoring and health assessment.

[0097] In a specific embodiment, the CNC machine tool motor monitoring system of a factory needs to monitor the running status of the motor in real time to ensure the efficient operation of the production line. The implementation steps of the CNC machine tool motor monitoring system are as follows:

[0098] Electrical data collection: The factory installed current sensors and voltage sensors on the motor power lines, set the sampling frequency to 1kHz, and collected 1,000 current and voltage samples per second. These sensors can monitor the power consumption of the motor and record the current and voltage changes of the motor under different loads in real time.

[0099] Temperature data collection: Thermocouples and platinum resistance temperature sensors are installed inside and on the motor housing respectively. Thermocouples are responsible for collecting the internal temperature of the motor, and platinum resistance monitors the temperature changes on the surface of the motor. The sampling frequency is set to 10Hz to ensure that temperature fluctuations can be accurately captured to prevent motor failure due to overheating or improper cooling.

[0100] Mechanical data collection: A rotary encoder and torque sensor are installed on the motor's drive shaft to record the motor's speed and torque in real time. The rotary encoder records 1,000 speed data per second, and the torque sensor collects 500 torque samples per second. These data help the factory analyze the load changes of the motor and prevent downtime due to overload or mechanical jamming.

[0101] Vibration data collection: Acceleration sensors are installed at key locations on the motor housing to monitor the vibration acceleration of the motor at a sampling frequency of 1kHz. By analyzing the spectrum of the vibration signal, the vibration frequency and amplitude of the motor are extracted to promptly detect abnormalities in the mechanical structure, such as bearing wear or imbalance.

[0102] Data synchronization and storage: All collected data (electrical, temperature, mechanical, vibration) are synchronized through the central processor. All data is stored in the factory database, and each data sample is accurately timestamped to ensure data consistency during subsequent analysis. Reports are automatically generated every day to analyze the motor operating status, predict potential failures, and arrange maintenance.

[0103] S2. Based on the collected electrical and mechanical data, determine the optimal fluctuation range and optimal working range of the electrical data, and establish a temperature change trend model by correlating the changes in the optimal fluctuation range and optimal working range of the electrical data with the temperature data to predict the temperature changes of the CNC machine tool motor under different working conditions; use signal processing technology to extract frequency and amplitude from vibration data.

[0104] Preferably, based on the collected electrical data, temperature data and mechanical data, the optimal working range of the electrical data is determined, and by correlating and analyzing the changes in the electrical data and temperature, a temperature change trend model is established to predict the temperature changes of the CNC machine tool motor under different working conditions; and extracting the frequency and amplitude from the vibration data using signal processing technology includes the following steps:

[0105] S21. Obtain the current data and voltage data in the electrical data, and determine the optimal electrical data range for the operation of the CNC machine tool motor through statistical analysis and comparison with historical electrical data.

[0106] It should be explained that it is necessary to obtain current data and voltage data in real time or periodically during the operation of the CNC machine tool motor. Usually, it can be obtained through the motor control system, sensor or data acquisition system. Before statistical analysis, the acquired data needs to be preprocessed, including data cleaning (removing outliers, missing values, etc.), data conversion (such as converting analog signals into digital signals) and data normalization (scaling the data to a standard range) and other steps; statistical analysis is performed on the preprocessed current data and voltage data; including calculating statistical quantities such as mean, standard deviation, maximum and minimum values ​​to understand the distribution and characteristics of the data; comparing the currently acquired electrical data with the historical data of the CNC machine tool motor operation; historical data can be collected from past operation records or obtained through experiments; through comparison, the current data can be found The differences and similarities between the current data and the historical data are used to determine the optimal electrical data range for the motor operation, which usually involves finding a balance point so that the motor is both efficient and stable when operating within the range; based on the results of statistical analysis and the comparison with historical data, the optimal electrical data range for the CNC machine tool motor operation can be determined; the optimal electrical data range is an interval that can ensure the normal operation of the motor and the best performance; the optimal electrical data range may vary depending on factors such as motor type, working conditions, and load. Therefore, when determining the range, it is necessary to fully consider these factors and make appropriate adjustments and optimizations. Once the optimal electrical data range is determined, it can be applied to the actual operation of the CNC machine tool motor. By monitoring the current and voltage data of the motor in real time, the situation out of range can be discovered and handled in a timely manner, thereby ensuring the normal operation of the motor and extending its service life.

[0107] S22. Perform correlation analysis on the acquired temperature data with the current data and voltage data in the electrical data, and use multiple regression analysis to analyze the impact of electrical data fluctuations on temperature changes.

[0108] Preferably, the acquired temperature data is correlated with the current and voltage data in the electrical data and the impact of the electrical data fluctuation on the temperature change is analyzed by regression analysis, which includes the following steps:

[0109] S221. Use the acquired current data and voltage data as independent variables; use the acquired temperature data as dependent variable, and organize the current data, voltage data and temperature data into a time series.

[0110] S222. Use a multiple regression model to construct a relationship between temperature data, current data, and voltage data.

[0111] S223. Fit the multivariate regression model using the least squares method to calculate the regression coefficients of current and voltage to temperature changes.

[0112] S224. Calculate the coefficient of determination and evaluate the goodness of fit of the multiple regression model based on the coefficient of determination.

[0113] Preferably, calculating the coefficient of determination and evaluating the goodness of fit of the multiple regression model based on the coefficient of determination comprises the following steps:

[0114] S2241. According to the multiple regression model, the values ​​of the independent variables are used to calculate the predicted values ​​of the dependent variable;

[0115] S2242, calculate the difference between the actual value and the predicted value of the dependent variable;

[0116] S2243, calculating the total sum of squares to measure the total variation in temperature, and calculating the regression sum of squares to measure the temperature variation explained by current and voltage;

[0117] S2244, calculate the residual sum of squares to measure the temperature variation not explained by current and voltage;

[0118] S2245. Calculate the coefficient of determination based on the regression sum of squares and the total sum of squares to evaluate the goodness of fit of the multivariate regression model to the temperature data.

[0119] S225. According to the signs and magnitudes of the regression coefficients of the current and voltage to the temperature change, the specific effects of the current and voltage fluctuations on the temperature change are obtained.

[0120] Preferably, the expression for calculating the coefficient of determination is:

[0121]

[0122] In the formula, represents the predicted temperature value of the CNC machine tool motor at the i-th data point;

[0123] Q 2 represents the coefficient of determination;

[0124] Y i Indicates the actual temperature value of the CNC machine tool motor;

[0125] Indicates the average value of temperature;

[0126] n represents the total number of data points;

[0127] i represents a data point.

[0128] S23. Based on the correlation analysis results of electrical data and temperature data, a temperature change trend model is constructed.

[0129] S24. Perform spectrum analysis on the acquired vibration data using fast Fourier transform to extract the frequency and amplitude of the vibration.

[0130] It should be explained that the current data and voltage data are used as independent variables:

[0131] Independent variables are those factors that are used to predict other variables. In this step, the current data and voltage data are used as independent variables, which means they are the main factors used to explain or predict temperature changes.

[0132] Temperature data as the dependent variable:

[0133] The dependent variable is the variable that is affected by other factors. In this step, the temperature data is used as the dependent variable, and the goal is to explain the change in temperature through the change in current and voltage.

[0134] The data is organized into time series:

[0135] Since the motor operation data is collected continuously, the current, voltage and temperature data need to be arranged in chronological order to form time series data. This ensures that in the subsequent analysis, each set of current, voltage and temperature data is collected at the same time point to avoid data asynchrony affecting the analysis results.

[0136] Implementation:

[0137] Sort the collected current, voltage, and temperature data by timestamp to ensure that all data points are aligned according to the same time base; and use data frame tools in databases or programming languages ​​(such as Pandas in Python) to process these time series data.

[0138] Multiple regression model: Multiple regression model is a statistical method used to study the linear relationship between a dependent variable and multiple independent variables. Through multiple regression, we try to explain how temperature (dependent variable) is affected by current and voltage (independent variables).

[0139] Multiple regression formula: The basic form of the model is: T = β0 + β1I + β2V + ∈, where T is temperature data; I is current data; V is voltage data; β0 is the constant term (intercept); β1 and β2 are regression coefficients, representing the impact of current and voltage on temperature changes; ∈ is the error term.

[0140] Use statistical software or programming tools (such as Python's statsmodels or sklearn library) to implement multiple linear regression, define current and voltage as independent variables, temperature as dependent variable, establish a regression model, and obtain a regression equation.

[0141] Least squares method: Least squares method is an optimization method used for regression analysis. Its goal is to minimize the sum of squares of the error between the predicted value and the actual value of the regression model, so as to find the optimal regression coefficient. By fitting the data with the least squares method, the influence of current and voltage on temperature change can be obtained, that is, the values ​​of β1 and β2 can be calculated.

[0142] The regression coefficients β1 and β2 represent the effect of current and voltage on temperature. If β1 is positive, it means that an increase in current will cause the temperature to rise; if β2 is negative, it means that an increase in voltage will cause the temperature to drop. Use the LinearRegression model in sklearn or the OLS model in statsmodels to apply the least squares method. The fitted model will return the regression coefficients of current and voltage on temperature, through which the specific effect of electrical data on temperature changes can be analyzed.

[0143] As a specific example, the regression coefficient of current is 2.50, which means that for every unit increase in current, the temperature will rise by 2.5 degrees; the regression coefficient of voltage is 0.30, which means that for every unit increase in voltage, the temperature will rise by 0.3 degrees. These coefficients show the specific impact of electrical data on temperature changes.

[0144] S25. Integrate the extracted electrical data, temperature data, frequency and amplitude to form a comprehensive analysis of the working status of the CNC machine tool motor, and combine with the temperature change trend model to predict the health status of the CNC machine tool motor under different working conditions in the future.

[0145] Preferably, the extracted electrical data, temperature data, frequency and amplitude are integrated to form a comprehensive analysis of the working state of the CNC machine tool motor, and combined with the temperature change trend model, the future health state of the CNC machine tool motor under different working conditions is predicted, including the following steps:

[0146] S251, integrating the extracted current data, voltage data, temperature data, and the frequency and amplitude extracted from the vibration data;

[0147] S252, extracting the mean, maximum, standard deviation and change rate of electrical, temperature and vibration data;

[0148] S253, performing correlation analysis on the mean, maximum, standard deviation and change rate of the extracted electrical, temperature and vibration data to identify the working state of the CNC machine tool motor;

[0149] S254. Combined with the temperature change trend model, a prediction model for the motor health status is constructed using multivariate statistical analysis;

[0150] S255. Based on the identification of the working status and prediction model of the CNC machine tool motor, the health status of the CNC machine tool motor in the future operation process is simulated, and its temperature, vibration and electrical changes under different working conditions are predicted.

[0151] It should be explained that electrical data (current and voltage), temperature data, and frequency and amplitude extracted from vibration data are all key operating indicators. By integrating these data, more comprehensive information on the motor's operating status can be obtained. By integrating these data, the mutual influence between the motor's electrical, thermal, and mechanical states during operation can be better captured; all extracted data are synchronously processed in the form of time series to ensure that different data types are integrated at the same time point, and statistical characteristic values ​​such as mean, maximum, standard deviation, and rate of change are extracted from electrical data, temperature data, and vibration data. These values ​​can better characterize the overall working state and volatility of the motor. For example, the rate of change of temperature can reflect the change in the motor's thermal load, and the standard deviation of vibration It can reflect the stability of mechanical vibration, and the extracted features are subjected to multi-dimensional correlation analysis to identify the working state of the motor; through the correlation of these features, it can be determined whether the motor is in normal operation, mild abnormality or severe abnormality. For example, when the vibration frequency and current fluctuate greatly, combined with the temperature increase, it can be inferred that the motor load is too large and there may be abnormalities. Using the previously extracted data and the constructed temperature change trend model, a prediction model for the motor health status can be constructed through multivariate statistical analysis methods (such as regression analysis or machine learning models); this model can predict the future operating state of the motor under different working conditions, especially the changing trends of temperature, vibration and electrical data, and combine the current working state and the prediction model to simulate the health changes that may occur in the motor during future operation. For example, under different load or ambient temperature conditions, predict future temperature increases, increased vibration amplitudes, and changes in electrical parameters.

[0152] background:

[0153] A factory installed multiple CNC machine tools for processing complex workpieces. The motors on the machine tools need to operate stably under various load conditions, and the temperature, vibration, and electrical data of the motors directly affect their health status. In order to improve the operating efficiency and safety of the motors, the factory deployed a health status prediction system based on multivariate analysis.

[0154] Implementation steps:

[0155] 1. Data collection and integration:

[0156] Sensors are installed at various key locations on CNC machine tool motors: current and voltage sensors collect electrical data, temperature sensors monitor internal and external temperatures, and acceleration sensors collect the vibration frequency and amplitude of the motor.

[0157] Data is collected in real time and integrated into a time series database.

[0158] 2. Feature extraction:

[0159] Use programming tools to extract the mean, maximum, standard deviation and change rate of current, temperature and vibration data every minute. Each characteristic value can reflect the overall operation trend of the motor.

[0160] For example, if the standard deviation of temperature suddenly increases during a day's operation and the maximum current value also increases, this may mean that the motor is overloaded.

[0161] 3. Working status identification:

[0162] By analyzing these extracted feature values, the system identifies the current working state of the motor. For example, under normal working conditions, the temperature, vibration and electrical parameters fluctuate slightly, while under abnormal conditions, the parameters fluctuate greatly.

[0163] Use the correlation analysis method to identify whether the current motor is in a normal, slightly abnormal or seriously abnormal state.

[0164] 4. Construction of health status prediction model:

[0165] Based on historical data, a temperature trend model was constructed using multivariate regression analysis. Then, combined with vibration and electrical data, a predictive model was constructed that can predict the health status of the motor under different future load conditions.

[0166] Using a random forest model, the input current, temperature, and vibration signatures are correlated with temperature and vibration predictions for the next few hours.

[0167] 5. Simulation and prediction of future operation status:

[0168] The predictive model was used to simulate the state of the motor under future high load conditions. It was found that as the load increases, the temperature of the motor will increase significantly and the vibration frequency will also tend to increase.

[0169] Based on the prediction results, the factory arranged preventive maintenance in advance to avoid possible equipment overheating failures.

[0170] S3. According to the preset CNC machine tool motor health status standard, the extracted frequency, amplitude and temperature trend model prediction results are combined to determine the current operating status of the CNC machine tool motor through classification analysis.

[0171] Preferably, according to a preset CNC machine tool motor health status standard, the extracted frequency, amplitude and prediction results of the temperature trend model are combined, and judging the current operating status of the CNC machine tool motor through classification analysis includes the following steps:

[0172] S31, extracting frequency data, amplitude data and temperature data from the operation data of the CNC machine tool motor, and scaling the extracted frequency data, amplitude data and temperature data respectively;

[0173] S32, designing a first-level classifier using a BP neural network, with the input being the scaled frequency, amplitude and temperature data;

[0174] S33, using known historical operation data and pre-set CNC machine tool motor health status standards to train the primary classifier, and testing the real-time operation data to determine the status of the CNC machine tool motor;

[0175] S34, according to the judgment result of the primary classifier, the preset judgment threshold, and the operation state of the CNC machine tool motor is divided into normal operation or abnormal state. When the judgment result is abnormal state, the abnormal analysis process is triggered and the step S35 is executed;

[0176] S35, selecting feature data sensitive to abnormal motor response from the primary feature extraction, and constructing a secondary feature vector;

[0177] S36, designing a new BP neural network as a secondary classifier, inputting a secondary feature vector, and outputting a specific classification of the abnormal state of the motor;

[0178] S37, using known abnormal data sets to train the secondary classifier and test its classification ability for different abnormal patterns;

[0179] S38. According to the output of the secondary classifier and the preset decision threshold, the abnormal state of the motor is refined into different levels, and corresponding maintenance or repair measures are taken.

[0180] It should be explained that frequency data, amplitude data and temperature data are extracted from the operating data of the CNC machine tool motor, and these data are scaled to ensure that they are within the same scale range, so as to facilitate subsequent analysis and processing. A first-level classifier is designed using a BP (back propagation) neural network, with the scaled frequency, amplitude and temperature data as input and the preliminary state classification of the motor (such as normal or abnormal) as output; the first-level classifier is trained using known historical operating data and pre-set CNC machine tool motor health status standards, and the real-time operating data is tested. The state of the motor is judged by the first-level classifier, and the operating state of the motor is divided into normal operation or abnormal state according to the judgment result of the first-level classifier and the preset judgment threshold. When the judgment result is an abnormal state, the abnormal analysis process is triggered and executed. In step S35, feature data sensitive to motor abnormality are selected from the primary feature extraction. These feature data may include a specific frequency band, a peak value of amplitude, or a rate of change of temperature, etc. A secondary feature vector is constructed as the input of a secondary classifier. A new BP neural network is designed as a secondary classifier. The input is the secondary feature vector, and the output is a specific classification of the motor abnormal state (such as overheating, overload, abnormal vibration, etc.). The secondary classifier is trained using a known abnormal data set, and the classification ability of the secondary classifier for different abnormal modes is tested to ensure its accuracy and reliability. According to the output of the secondary classifier and a preset decision threshold, the abnormal state of the motor is refined into different levels. According to the level of the abnormal state, corresponding maintenance or repair measures are taken to ensure the normal operation of the motor and extend its service life.

[0181] Assume that there is a CNC machine tool motor in a project, and its operating data includes frequency, amplitude and temperature. Follow the steps below to implement it:

[0182] Frequency data (such as 50 Hz), amplitude data (such as 0.1 mm) and temperature data (such as 40° C.) are extracted from the running data of the motor and scaled.

[0183] A first-level classifier is designed using a BP neural network, with the scaled data as input. Through training, the classifier can preliminarily determine the state of the motor (normal or abnormal).

[0184] The real-time operation data is tested and the state of the motor is judged by the first-level classifier. If it is judged to be an abnormal state, the abnormal analysis process is triggered.

[0185] In the abnormality analysis process, feature data that is sensitive to motor abnormality (such as the frequency fluctuation range, amplitude peak value, etc.) is selected from the primary feature extraction to construct a secondary feature vector.

[0186] A new BP neural network is designed as a secondary classifier, with the secondary feature vector as input. Through training, the classifier can output the specific classification of motor abnormality (such as overheating, abnormal vibration, etc.).

[0187] The secondary classifier is tested to ensure that it can accurately classify different abnormal patterns.

[0188] According to the output of the secondary classifier and the preset decision threshold, the abnormal state of the motor is refined into different levels (such as slight overheating, severe overheating, etc.), and corresponding maintenance or repair measures are taken (such as increasing cooling, reducing load, etc.).

[0189] Preferably, designing a new BP neural network as a secondary classifier, inputting a secondary feature vector, and outputting a specific classification of the abnormal state of the motor includes the following steps:

[0190] S361, selecting features that are more sensitive to motor abnormality from the first-level feature extraction, and constructing a second-level feature vector;

[0191] S362, determining a new BP neural network architecture, including an input layer, a hidden layer, and an output layer, wherein the number of input layer nodes matches the dimension of the secondary feature vector, and the number of output layer nodes corresponds to the specific classification number of the motor abnormal state;

[0192] S363, select an activation function and set training parameters;

[0193] S364, using the known abnormal data set to train the new BP neural network, and during the training process, monitoring the loss function and accuracy of the model in real time;

[0194] S365. Evaluate the trained BP neural network using the test data set and adjust the model architecture based on the test results.

[0195] S366. Deploy the trained BP neural network to the CNC machine tool motor operation monitoring to classify the abnormal status of the CNC machine tool motor in real time.

[0196] It should be explained that the features that are more sensitive to motor abnormalities are selected from the first-level feature extraction. Features are variables that show high importance in the first-level classification or are closely related to the abnormal state of the motor. These selected features are combined into a new feature vector, namely the second-level feature vector. The dimension of this vector depends on the number of selected features.

[0197] The architecture of the BP neural network is determined according to the dimension of the secondary feature vector and the specific classification number of the motor abnormal state.

[0198] The number of input layer nodes should match the dimension of the second-level feature vector to ensure that every feature can be passed as input to the neural network.

[0199] The number of nodes in the output layer should correspond to the specific classification number of the abnormal state of the motor, and each node represents a possible abnormal state.

[0200] The number of hidden layers and number of nodes can be adjusted based on the complexity of the problem and the size of the dataset. Typically, more complex networks may require more hidden layers and nodes to capture complex patterns in the data.

[0201] Select an appropriate activation function, such as ReLU (Rectified Linear Unit), Sigmoid or Tanh, to introduce nonlinear characteristics and enhance the expressiveness of the network.

[0202] Set training parameters such as learning rate, batch size, number of training rounds, etc. These parameters affect the training speed and performance of the network.

[0203] The new BP neural network is trained using the known abnormal data set. During the training process, the network learns based on the input feature vector and the corresponding abnormal state label.

[0204] Monitor the model's loss function and accuracy in real time to evaluate the network's training effect. If the loss function continues to decrease and the accuracy increases, it means that the network is learning effectively.

[0205] The trained BP neural network is evaluated using the test data set, which should contain data samples that are different from the training data set but have similar characteristics.

[0206] Based on the test results, evaluate the performance of the network, such as accuracy, recall, etc. If the performance is poor, you may need to adjust the model architecture, activation function, or training parameters.

[0207] The trained BP neural network is deployed to the motor operation monitoring of CNC machine tools.

[0208] Classify the abnormal status of CNC machine tool motors in real time and take corresponding maintenance or repair measures based on the classification results.

[0209] S4. Set the threshold of the working condition data, compare the classification analysis results with the preset threshold, and automatically trigger the multi-level early warning mechanism when the working condition data of the CNC machine tool motor is monitored to exceed the threshold.

[0210] It needs to be explained that the threshold of the working condition data is set as follows:

[0211] The thresholds of operating condition data are usually based on the normal operating range of the motor and historical data analysis. For different parameters such as current, voltage, temperature and vibration, setting appropriate thresholds can help quickly determine whether the equipment is in an abnormal state.

[0212] The threshold can be static (fixed value set based on experience) or dynamic (automatically adjusted according to different operating or environmental conditions).

[0213] Classification analysis and threshold comparison:

[0214] The current operating status of the motor (such as normal, slightly abnormal, seriously abnormal, etc.) is determined through the classification analysis method, and the analysis result is compared with the preset threshold.

[0215] Once it is detected that one or more operating condition data exceeds the threshold, the abnormal situation will be automatically identified.

[0216] Multi-level warning mechanism: When the operating data exceeds the threshold, the multi-level warning mechanism will be triggered. According to the severity of the abnormality, the warning level can be divided into different levels (such as yellow warning, orange warning and red warning); each warning level corresponds to different response measures. For example, a mild warning may only require the operator to monitor the equipment, while a severe warning may automatically stop the motor operation and notify the maintenance team for maintenance.

[0217] background:

[0218] A factory uses CNC machine tools for precision machining. Due to the heavy processing tasks, the motor needs to operate stably for a long time. In order to avoid sudden failures affecting production, the factory deployed a multi-level early warning system based on working condition data monitoring to monitor the health status of the motor in real time.

[0219] Implementation steps:

[0220] Setting threshold (S4.1):

[0221] The factory's technical team set thresholds for key operating data based on the motor's operating manual and historical data. For example:

[0222] Current threshold: 20A;

[0223] Voltage threshold: 230V;

[0224] Temperature threshold: 75℃;

[0225] Vibration frequency threshold: 50Hz;

[0226] Vibration amplitude threshold: 0.02m / s 2 ;

[0227] The threshold reflects the safe range of the motor under normal working conditions.

[0228] Real-time monitoring and classification analysis:

[0229] Real-time monitoring of the working data of CNC machine tool motors, including current, voltage, temperature, vibration and other parameters. Through classification analysis, the current state of the motor is classified as "normal operation" or "abnormal operation".

[0230] When it is detected that one or more parameters of the motor deviate from the normal value, it is first compared with the preset threshold value.

[0231] Multi-level early warning mechanism triggers:

[0232] Level 1 warning (yellow warning): If the motor temperature slightly exceeds 75°C but does not reach 85°C, a yellow warning is triggered. The operator will receive a notification to remind him to closely monitor the motor operation.

[0233] Level 2 warning (orange warning): If the current exceeds the set 20A threshold and the temperature exceeds 85°C, an orange warning is triggered. The operator needs to take load reduction measures immediately and arrange maintenance.

[0234] Level 3 warning (red warning): If the motor temperature exceeds 90°C and the vibration frequency exceeds 50Hz, a red warning will be triggered, automatically stopping the motor to prevent equipment damage, and notifying the maintenance team to perform repairs immediately.

[0235] Response measures:

[0236] When a yellow or orange warning is triggered, operators can monitor and adjust manually.

[0237] When a red warning is triggered, the machine tool will automatically stop running to avoid further damage. At the same time, a detailed warning report will be generated to analyze the specific cause of the abnormality and assist the maintenance team to make quick repairs.

[0238] According to another embodiment of the present invention, Figure 2 As shown, a CNC machine tool motor operation monitoring system based on data analysis is also provided, and the CNC machine tool motor operation monitoring system includes: a data acquisition module 1, a data analysis and modeling module 2, a state classification and judgment module 3 and an early warning trigger module 4;

[0239] The data acquisition module 1 is used to collect the working condition data corresponding to the operation process of the CNC machine tool motor, and the working condition data includes electrical data, temperature data, mechanical data and vibration data;

[0240] Data analysis and modeling module 2 is used to determine the optimal fluctuation range and optimal working range of electrical data based on the collected electrical data and mechanical data, and to establish a temperature change trend model by correlating the optimal fluctuation range and optimal working range of electrical data with the changes in temperature data, and to predict the temperature changes of CNC machine tool motors under different working conditions; and to extract frequency and amplitude from vibration data using signal processing technology;

[0241] The state classification and judgment module 3 is used to combine the prediction results of the extracted frequency, amplitude and temperature trend model according to the preset CNC machine tool motor health state standard, and judge the current operation state of the CNC machine tool motor through classification analysis;

[0242] The early warning trigger module 4 is used to set the threshold of the working condition data, compare the classification analysis results with the preset threshold, and automatically trigger the multi-level early warning mechanism when the working condition data of the CNC machine tool motor is monitored to exceed the threshold.

[0243] In summary, with the help of the above-mentioned technical scheme of the present invention, the present invention can comprehensively grasp the operating status of the motor through the comprehensive collection and analysis of electrical, temperature, mechanical and vibration data, avoid the limitations of single data analysis, and accurately identify the operating status of the motor through statistical analysis and multivariate regression models, and trigger an early warning when it exceeds the optimal working range to prevent potential failures. Through the temperature change trend model and signal processing of vibration data, the health status of the motor under different future working conditions can be predicted in advance to achieve preventive maintenance and avoid sudden failures. By determining the optimal electrical working range and real-time monitoring data, it is ensured that the motor operates under efficient and stable conditions, reducing energy consumption and extending the service life of the equipment. The combination of historical data, real-time data and statistical models is used to realize intelligent motor status analysis and prediction, and promote the intelligent transformation of factory equipment maintenance. The primary and secondary classifiers realized by the BP neural network of the present invention can quickly and effectively identify the operating status of the motor and classify abnormal conditions, ensuring that the motor is running smoothly and reliably. Predict the machine before it fails and reduce the risk of sudden failures; classification can not only detect anomalies, but also further refine the type of anomalies (such as overheating, mechanical failure, etc.), which is helpful to formulate targeted processing measures for different types of anomalies and improve the efficiency of problem solving. Combined with historical data and real-time data, through the training of neural networks, it can gradually learn and optimize the judgment of the motor state. The secondary classifier can accurately judge the abnormal level and specific situation, thereby providing data support for preventive maintenance and optimizing maintenance strategies. Through prediction and classification, the factory can plan maintenance in advance to avoid sudden equipment shutdowns; the present invention can identify and respond to potential fault problems in a timely manner by setting thresholds and comparing monitoring data. The multi-level early warning mechanism provides flexibility to ensure that faults of different severity can be handled in a timely manner, thereby avoiding the spread or escalation of faults. Through threshold monitoring, protective measures can be automatically taken when the motor is overloaded, overheated or the mechanical vibration is abnormal to prevent serious damage to the equipment, greatly improving the safety of CNC machine tool motor operation.

[0244] Although the present invention has been disclosed as above with preferred embodiments, the embodiments are merely examples for the convenience of description and are not intended to limit the present invention. Those skilled in the art may make several changes and modifications without departing from the spirit and scope of the present invention. The scope of protection claimed by the present invention shall be based on the claims.

Claims

1. A method for monitoring the operation of a CNC machine tool motor based on data analysis, characterized in that: The CNC machine tool motor operation monitoring method comprises: S1. Collecting the working condition data corresponding to the operation process of the CNC machine tool motor, and the working condition data includes electrical data, temperature data, mechanical data and vibration data; S2. Based on the collected electrical and mechanical data, determine the optimal fluctuation range and optimal working range of the electrical data, and establish a temperature change trend model by correlating the optimal fluctuation range and optimal working range of the electrical data with the changes in temperature data to predict the temperature changes of CNC machine tool motors under different working conditions; use signal processing technology to extract frequency and amplitude from vibration data; S3, according to the preset CNC machine tool motor health status standard, the extracted frequency, amplitude and temperature trend model prediction results are combined to determine the current operating status of the CNC machine tool motor through classification analysis; S4. Set the threshold of the working condition data, compare the classification analysis results with the preset threshold, and automatically trigger the multi-level early warning mechanism when the working condition data of the CNC machine tool motor is monitored to exceed the threshold.

2. A method for monitoring the operation of a CNC machine tool motor based on data analysis according to claim 1, characterized in that: The collecting of the working condition data corresponding to the operation process of the CNC machine tool motor, wherein the working condition data includes electrical data, temperature data, mechanical data and vibration data, comprises the following steps: S11, using a pre-configured current sensor and voltage sensor to collect current and voltage at a preset frequency to obtain corresponding electrical data; S12, using pre-configured thermocouples and platinum resistance temperature detection sensors to monitor the internal and external temperatures of the CNC machine tool motor in real time to obtain corresponding temperature data; S13, using a pre-configured rotary encoder and torque sensor to collect the speed and torque of the CNC machine tool motor to obtain corresponding mechanical data; S14, using a pre-configured acceleration sensor to monitor the acceleration of the motor vibration of the CNC machine tool in real time, and extracting the frequency and amplitude through signal processing to obtain corresponding vibration data; S15, synchronously processing and storing the electrical data, temperature data, mechanical data and vibration data.

3. A method for monitoring the operation of a CNC machine tool motor based on data analysis according to claim 2, characterized in that: The method of determining the optimal working range of the electrical data based on the collected electrical data, temperature data and mechanical data, establishing a temperature change trend model by correlating and analyzing the changes of the electrical data and temperature, and predicting the temperature changes of the CNC machine tool motor under different working conditions; and extracting the frequency and amplitude from the vibration data using signal processing technology includes the following steps: S21, determining the optimal electrical data range for the operation of the CNC machine tool motor by statistically analyzing the current data and voltage data in the acquired electrical data and comparing the electrical data with historical data; S22, performing correlation analysis on the acquired temperature data and the current data and voltage data in the electrical data, and using a multivariate regression analysis method to analyze the impact of electrical data fluctuations on temperature changes; S23. Constructing a temperature change trend model based on the correlation analysis results of the electrical data and the temperature data; S24, performing spectrum analysis on the acquired vibration data using fast Fourier transform to extract the frequency and amplitude of the vibration; S25. Integrate the extracted electrical data, temperature data, frequency and amplitude to form a comprehensive analysis of the working status of the CNC machine tool motor, and combine with the temperature change trend model to predict the health status of the CNC machine tool motor under different working conditions in the future.

4. A method for monitoring the operation of a CNC machine tool motor based on data analysis according to claim 3, characterized in that: The method of correlating the acquired temperature data with the current and voltage data in the electrical data and analyzing the influence of the electrical data fluctuation on the temperature change by using the regression analysis method comprises the following steps: S221, using the acquired current data and voltage data as independent variables; using the acquired temperature data as a dependent variable, and arranging the current data, voltage data and temperature data into a time series; S222, using a multiple regression model to construct a relationship between the temperature data and the current data and the voltage data; S223, fitting the multivariate regression model using the least squares method to calculate the regression coefficients of current and voltage to temperature changes; S224, calculating the coefficient of determination, and evaluating the goodness of fit of the multiple regression model based on the coefficient of determination; S225. According to the signs and magnitudes of the regression coefficients of the current and voltage to the temperature change, the specific effects of the current and voltage fluctuations on the temperature change are obtained.

5. A method for monitoring the operation of a CNC machine tool motor based on data analysis according to claim 4, characterized in that: The method of calculating the coefficient of determination and evaluating the goodness of fit of the multiple regression model based on the coefficient of determination includes the following steps: S2241. According to the multiple regression model, the values ​​of the independent variables are used to calculate the predicted values ​​of the dependent variable; S2242, calculate the difference between the actual value and the predicted value of the dependent variable; S2243, calculating the total sum of squares to measure the total variation in temperature, and calculating the regression sum of squares to measure the temperature variation explained by current and voltage; S2244, calculate the residual sum of squares to measure the temperature variation not explained by current and voltage; S2245. Calculate the coefficient of determination based on the regression sum of squares and the total sum of squares to evaluate the goodness of fit of the multivariate regression model to the temperature data.

6. A method for monitoring the operation of a CNC machine tool motor based on data analysis according to claim 5, characterized in that: The expression for calculating the coefficient of determination is: In the formula, represents the predicted temperature value of the CNC machine tool motor at the i-th data point; Q 2 represents the coefficient of determination; Y i Indicates the actual temperature value of the CNC machine tool motor; Indicates the average value of temperature; n represents the total number of data points; i represents a data point.

7. A method for monitoring the operation of a CNC machine tool motor based on data analysis according to claim 6, characterized in that: The extracted electrical data, temperature data, frequency and amplitude are integrated to form a comprehensive analysis of the working state of the CNC machine tool motor, and combined with the temperature change trend model, the health state of the CNC machine tool motor under different working conditions in the future is predicted, including the following steps: S251, integrating the extracted current data, voltage data, temperature data, and the frequency and amplitude extracted from the vibration data; S252, extracting the mean, maximum, standard deviation and change rate of electrical, temperature and vibration data; S253, performing correlation analysis on the mean, maximum, standard deviation and change rate of the extracted electrical, temperature and vibration data to identify the working state of the CNC machine tool motor; S254. Combined with the temperature change trend model, a prediction model for the motor health status is constructed using multivariate statistical analysis; S255. Based on the identification of the working status and prediction model of the CNC machine tool motor, the health status of the CNC machine tool motor in the future operation process is simulated, and its temperature, vibration and electrical changes under different working conditions are predicted.

8. The method for monitoring the operation of a motor of a CNC machine tool based on data analysis according to claim 1, characterized in that: The method of combining the extracted frequency, amplitude and prediction result of the temperature trend model according to the preset CNC machine tool motor health status standard and judging the current operation status of the CNC machine tool motor through classification analysis includes the following steps: S31, extracting frequency data, amplitude data and temperature data from the operation data of the CNC machine tool motor, and scaling the extracted frequency data, amplitude data and temperature data respectively; S32, designing a first-level classifier using a BP neural network, with the input being the scaled frequency, amplitude and temperature data; S33, using known historical operation data and pre-set CNC machine tool motor health status standards to train the primary classifier, and testing the real-time operation data to determine the status of the CNC machine tool motor; S34, according to the judgment result of the primary classifier, the preset judgment threshold, and the operation state of the CNC machine tool motor is divided into normal operation or abnormal state. When the judgment result is abnormal state, the abnormal analysis process is triggered and the step S35 is executed; S35, selecting feature data sensitive to abnormal motor response from the primary feature extraction, and constructing a secondary feature vector; S36, designing a new BP neural network as a secondary classifier, inputting a secondary feature vector, and outputting a specific classification of the abnormal state of the motor; S37, using known abnormal data sets to train the secondary classifier and test its classification ability for different abnormal patterns; S38. According to the output of the secondary classifier and the preset decision threshold, the abnormal state of the motor is refined into different levels, and corresponding maintenance or repair measures are taken.

9. A method for monitoring the operation of a CNC machine tool motor based on data analysis according to claim 8, characterized in that: The design of a new BP neural network as a secondary classifier, inputting a secondary feature vector, and outputting a specific classification of the abnormal state of the motor comprises the following steps: S361, selecting features that are more sensitive to motor abnormality from the first-level feature extraction, and constructing a second-level feature vector; S362, determining a new BP neural network architecture, including an input layer, a hidden layer, and an output layer, wherein the number of input layer nodes matches the dimension of the secondary feature vector, and the number of output layer nodes corresponds to the specific classification number of the motor abnormal state; S363, select an activation function and set training parameters; S364, using the known abnormal data set to train the new BP neural network, and during the training process, monitoring the loss function and accuracy of the model in real time; S365. Evaluate the trained BP neural network using the test data set and adjust the model architecture based on the test results. S366. Deploy the trained BP neural network to the CNC machine tool motor operation monitoring to classify the abnormal status of the CNC machine tool motor in real time.

10. A CNC machine tool motor operation monitoring system based on data analysis, used to implement the CNC machine tool motor operation monitoring method based on data analysis as described in any one of claims 1 to 9, characterized in that: The CNC machine tool motor operation monitoring system includes: a data acquisition module, a data analysis and modeling module, a state classification and judgment module and an early warning trigger module; The data acquisition module is used to collect the working condition data corresponding to the operation process of the CNC machine tool motor, and the working condition data includes electrical data, temperature data, mechanical data and vibration data; The data analysis and modeling module is used to determine the optimal fluctuation range and optimal working range of electrical data based on the collected electrical and mechanical data, and to establish a temperature change trend model by correlating the optimal fluctuation range and optimal working range of electrical data with the changes in temperature data, so as to predict the temperature changes of CNC machine tool motors under different working conditions; and to extract frequency and amplitude from vibration data using signal processing technology; The state classification and judgment module is used to combine the prediction results of the extracted frequency, amplitude and temperature trend model according to the preset CNC machine tool motor health status standard, and judge the current operation status of the CNC machine tool motor through classification analysis; The early warning trigger module is used to set the threshold of the working condition data and compare the classification analysis results with the preset threshold. When the working condition data of the CNC machine tool motor is monitored to exceed the threshold, the multi-level early warning mechanism is automatically triggered.

Images