Multi-source data fused deburring equipment service life monitoring and transmission system and method

Abstract

The invention discloses a multi-source data fused deburring equipment service life monitoring and transmission system and method, and relates to the technical field of equipment state monitoring, and the method comprises the following steps: S100, collecting operation data of deburring equipment, and carrying out the operation data integration; s200, extracting characteristic parameters related to the service life of the deburring equipment from the integrated data, constructing a service life prediction model based on exponential smoothing by using the characteristic parameters, and optimizing a model coefficient through batch gradient descent; s300, collecting characteristic parameter data in real time, performing service life prediction by using the service life prediction model, and analyzing a service life attenuation rate in combination with historical characteristic parameter data; and S400, the operation stability of the deburring equipment is analyzed, an early warning threshold value is set, the stability and the service life of the deburring equipment are monitored, monitoring data and a prediction result are transmitted through the industrial Internet of Things technology, and the service life of the deburring equipment is monitored.

Description

Technical Field

[0001] The present invention relates to the technical field of equipment status monitoring, and specifically to a deburring equipment life monitoring and transmission system and method that integrates multi-source data. Background Technique

[0002] With the continuous development of industrial automation and precision manufacturing technologies, the application of deburring equipment in various manufacturing enterprises has gradually increased. Especially in industries with high-precision requirements such as automobiles, aerospace, and electronic products, the performance and stability of deburring equipment directly affect product quality and production efficiency. As a core link in modern manufacturing, the operating status, maintenance management, and life prediction of deburring equipment are of great significance for enterprises' production decisions and equipment management.

[0003] However, traditional deburring equipment life monitoring and management methods often face the following problems when dealing with the complexity of equipment operation data, multi-dimensional data analysis, and maintenance management: First, the monitoring efficiency is low. A large amount of data generated during the operation of the equipment needs to be manually collected, sorted, and analyzed, resulting in a time-consuming and laborious data processing process. Especially in the case of high-frequency production, manual monitoring not only has a slow processing speed but also easily ignores key data, leading to the failure to detect equipment problems in a timely manner, affecting production efficiency and the health status of the equipment. Second, traditional methods lack intelligent means and are difficult to quickly identify potential risks of the equipment. Traditional equipment monitoring usually relies on manual or simple rules for judgment and cannot identify abnormal situations of the equipment from multi-dimensional and complex operation data. Especially when facing complex problems such as equipment failures and life attenuation, the effectiveness of existing methods is limited, and it is difficult to detect potential risks in a timely manner. In addition, equipment life prediction and maintenance management lack global intelligent analysis, lacking comprehensive analysis of the mutual influence between multi-dimensional operating states of the equipment, and unable to provide accurate equipment health prediction for enterprises. Summary of the Invention

[0004] The purpose of the present invention is to provide a deburring equipment life monitoring and transmission system and method that integrates multi-source data to solve the problems raised in the prior art.

[0005] To achieve the above purpose, the present invention provides the following technical solution: A deburring equipment life monitoring and transmission method that integrates multi-source data, the method includes the following steps:

[0006] S100. Collect the operation data of the deburring equipment and perform operation data integration;

[0007] S200. Extract characteristic parameters related to the life of the deburring equipment from the integrated data, construct a life prediction model based on exponential smoothing using the characteristic parameters, and optimize the model coefficients through batch gradient descent;

[0008] S300, collecting characteristic parameter data in real time, using the service life prediction model to predict the service life, and analyzing the service life decay rate in combination with historical characteristic parameter data;

[0009] S400: Analyze the operational stability of deburring equipment, set early warning thresholds, monitor equipment stability and lifespan, and transmit monitoring data and prediction results through industrial Internet of Things technology.

[0010] In step S100, the specific steps include:

[0011] S101, obtaining the tool force, vibration frequency and amplitude, and the temperature of the tool, motor, transmission mechanism and motor working current involved in the operation of the deburring equipment through sensors, and arranging the collected data according to the time axis to form a multi-dimensional time series data stream;

[0012] S102, converting the data collected by different sensors into a unified time series format, using Min-Max to perform normalization processing, using an anomaly detection model to identify and remove outliers and missing values ​​in the data, calibrating and aligning the timestamps, and sorting out the deburring equipment operation status data in different time periods.

[0013] In step S200, the specific steps include:

[0014] S201, extract characteristic parameters related to the life of the deburring equipment from the integrated data, extract the pressure peak value, pressure fluctuation amplitude and pressure mean value from the tool force data, extract the maximum value, root mean square value and vibration frequency amplitude of the vibration amplitude from the vibration data, extract the highest temperature of the tool, motor and transmission mechanism from the temperature data, extract the motor starting current peak value and working current stability index from the current data, and record the extracted characteristic parameters as the characteristic vector X=[x1,x2...x n ], where X represents the characteristic parameter set, n represents the total number of characteristic parameters extracted from the deburring equipment operation data, x1, x2, ..., x n The 1st, 2nd, ..., nth characteristic parameters respectively;

[0015] S202, construct a deburring equipment life prediction model based on exponential smoothing, which is defined as follows: L t+1 =α*L t +(1-α)*(β0+β1*x 1,t +β2*x 2,t +...+β n *x n,t ), where L t+1 It represents the remaining life of the deburring equipment predicted at the next moment, L tRepresents the estimated remaining life of the deburring device at the current moment. The initial device life is set to the average life of the deburring device, x 1,t 、x 2,t 、...、x n,t Represent the 1st, 2nd,..., nth characteristic parameters collected at time t respectively. t represents the time when the operating characteristic parameters of the deburring device are collected. α represents the smoothing coefficient, and the value range of α is from 0 to 1. β0, β1,..., β n Represent the coefficients to be determined.

[0016] In step S202, the batch gradient descent algorithm is used to determine the coefficient β i , and the loss function is defined as follows:

[0017]

[0018] Among them, m represents the number of samples, and L j Represents the actual remaining life of the jth sample, Represents the remaining life of the jth sample predicted according to the model, and J(β) represents the loss function;

[0019] Calculate the gradient, and the formula is as follows:

[0020]

[0021] Among them, i represents the data subscript. For When i = 0, it is (1 - α), and when i ≠ 0, it is (1 - α) * x i,j ;

[0022] Update the coefficient in each iteration, and the calculation formula is as follows:

[0023]

[0024] Among them, η represents the learning rate. Through multiple iterations until the loss function converges, the optimized coefficient β i is obtained, and the construction of the deburring device life prediction model based on exponential smoothing is completed.

[0025] In step S300, the specific steps include:

[0026] S301. Use the multi-layer perceptron MLP deep neural network to perform feature learning and pattern recognition on the operating data of the deburring device. Take the multi-source data features extracted above as the input of the deburring device life prediction model, and use the remaining life of the corresponding deburring device in the historical data as the output label to train the deburring device life prediction model based on exponential smoothing;

[0027] S302. Input the operation data of the deburring device collected in real time into the trained life prediction model of the deburring device to obtain the predicted value of the current remaining life of the deburring device. Analyze the change range of the device operation data in different time periods, and calculate the decay rate of the life of the deburring device. The formula is as follows:

[0028]

[0029] Among them, R decay represents the decay rate of the life of the deburring device, ΔF i represents the change range of the i-th characteristic parameter within the time period T, w i represents the weight of the i-th characteristic, and ΔT represents the average sampling time interval within the time period T. Generate a device life evaluation report based on the predicted remaining life value, life decay rate, and device operation conditions and maintenance record information.

[0030] In step S400, the specific steps include:

[0031] S401. Calculate the operation stability index of the deburring device according to the number of failures and the duration of failures of the deburring device within a time period. The calculation formula of the operation stability index of the deburring device is as follows:

[0032]

[0033] Among them, k represents the failure number, and the range of k is k = 1, 2,..., F. T represents the total length of the time period, t k represents the duration of the k-th failure, F represents the number of failures of the deburring device, and S represents the operation stability index of the deburring device;

[0034] S402. Set the stability warning threshold S w . When S is less than S w , trigger the warning mechanism, mark that the stability of the current deburring device is abnormal, and there is a positive correlation between the device operation stability index and the life of the deburring device. Monitor the service life of the deburring device according to the positive correlation between the device operation stability index and the life of the deburring device. Among them, the stability warning threshold is obtained according to the service life specified by the factory of the deburring device combined with the historical data since the deburring device was put into use;

[0035] S403. Use the Industrial Internet of Things (IIoT) technology to transmit the device monitoring data and life prediction results to the remote monitoring platform through the wireless communication protocol MQTT.

[0036] Deburring equipment life monitoring and transmission system integrating multi-source data, the system includes: a data acquisition module, an equipment health status analysis module, a life prediction module, and an equipment stability analysis module. The data acquisition module is used to collect multi-source operation data of the deburring equipment and perform preprocessing. The equipment health status analysis module is used to extract key features of the equipment operation from the collected original data, fuse the multi-source features into a multi-dimensional feature vector, and input the multi-dimensional feature vector into the life prediction model. The life prediction module is used to train a deburring equipment life prediction model based on exponential smoothing and calculate the life decay rate of the deburring equipment. The equipment stability analysis module is used to analyze the fault situation of the equipment within a time period and evaluate the stability of the equipment. The output end of the data acquisition module is connected to the input end of the equipment health status analysis module, the output end of the equipment health status analysis module is connected to the input end of the life prediction module, and the output end of the life prediction module is connected to the input end of the equipment stability analysis module.

[0037] The data acquisition module includes an operation data acquisition unit and a data preprocessing unit. The operation data acquisition unit uses installed sensors to obtain the operation data of the deburring equipment in real time, including tool force, vibration frequency and amplitude, temperatures of the tool, motor, and transmission mechanism, and the working current of the motor. All the collected data is arranged in chronological order to form a multi-dimensional time series data stream. The data preprocessing unit is used to uniformly process the original data from different sensors, convert it into a standardized time series format, normalize the data using the Min-Max normalization method, and identify and remove outliers and missing values through an algorithm based on an anomaly detection model. At the same time, the timestamps are calibrated and aligned.

[0038] The equipment health status analysis module includes a feature extraction unit and a life prediction model establishment unit. The feature extraction unit is used to extract feature parameters related to the life of the deburring equipment from the integrated data, extract the pressure peak value, pressure fluctuation amplitude, and pressure mean value from the tool force data, extract the maximum value, root mean square value, and vibration frequency amplitude of the vibration amplitude from the vibration data, extract the highest temperatures of the tool, motor, and transmission mechanism parts from the temperature data, and extract the motor starting current peak value and the stability index of the working current from the current data. The life prediction model establishment unit is used to construct a deburring equipment life prediction model based on exponential smoothing according to the extracted features, and optimize the model coefficients through batch gradient descent to complete life estimation and quantitative analysis.

[0039] The life prediction module includes a life prediction model training unit and a life decay rate calculation unit. The life prediction model training unit is used to perform feature learning and pattern recognition on the operation data of the deburring device using a multi-layer perceptron (MLP) deep neural network, taking the extracted multi-source data features as input and the remaining life of the device in historical data as the output label to train the device life prediction model. The life decay rate calculation unit is used to analyze the change amplitude of the device operation data in different time periods and calculate the decay rate of the deburring device life. The device stability analysis module includes a fault statistics unit and a stability calculation unit. The fault statistics unit is used to count the number of faults and the duration of the faults that occur to the device within a time period. The stability calculation unit is used to calculate the stability index of the device based on the number of faults and the duration, and use the industrial Internet of Things (IIoT) technology to transmit the monitoring data and life prediction results of the device to a remote monitoring platform through the wireless communication protocol MQTT.

[0040] Compared with the prior art, the beneficial effects of the present invention are as follows:

[0041] 1. By collecting multi-source data in real time and performing automated data processing and analysis, the device monitoring efficiency is improved, data is obtained in real time at each link of the device operation, the time and labor costs of manual monitoring are reduced, and at the same time, the omissions in manual review are avoided;

[0042] 2. By integrating multi-source data, deep learning, batch gradient descent algorithm and anomaly detection model, potential fault risks and abnormal conditions of the device are identified, and the remaining life of the device is predicted in real time by analyzing the change trend of the key features of the device;

[0043] 3. Through real-time life decay rate calculation and remaining life prediction, it helps managers take timely maintenance or replacement measures before the device reaches the critical state, reducing device downtime and maintenance costs. BRIEF DESCRIPTION OF THE DRAWINGS

[0044] Figure 1 is a schematic flow chart of the method for monitoring and transmitting the life of a deburring device integrating multi-source data according to the present invention;

[0045] Figure 2 is a schematic structural diagram of the system for monitoring and transmitting the life of a deburring device integrating multi-source data according to the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0046] Next, in combination with the accompanying drawings in the embodiments of the present invention, the technical solutions in the embodiments of the present invention will be clearly and completely described. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present invention.

[0047] As Figure 1 - Figure 2 shown, the present invention provides a technical solution, a method for monitoring and transmitting the life of a deburring device that integrates multi-source data, and the method includes the following steps:

[0048] S100. Collect the operation data of the deburring device and integrate the operation data;

[0049] S200. Extract the characteristic parameters related to the life of the deburring device from the integrated data, construct a life prediction model based on exponential smoothing using the characteristic parameters, and optimize the model coefficients through batch gradient descent;

[0050] S300. Real-time collect the characteristic parameter data, use the life prediction model to predict the life, and analyze the life decay rate in combination with the historical characteristic parameter data;

[0051] S400. Analyze the operation stability of the deburring device, set the warning threshold, monitor the device stability and life, and transmit the monitoring data and prediction results through industrial Internet of Things technology.

[0052] In step S100, the specific steps include:

[0053] S101. Obtain the tool force, vibration frequency and amplitude involved in the operation of the deburring device, as well as the temperature of the tool, motor, transmission mechanism and the motor working current through sensors, and arrange the collected data according to the time axis to form a multi-dimensional time series data stream;

[0054] S102. Convert the data collected by different sensors into a unified time series format, perform normalization processing using Min-Max, use an anomaly detection model based on it to identify and remove the outliers and missing values in the data, calibrate and align the timestamps, and sort out the operation status data of the deburring device in different time periods.

[0055] In step S200, the specific steps include:

[0056] S201. Extract the characteristic parameters related to the life of the deburring device from the integrated data. Extract the peak pressure, pressure fluctuation amplitude, and pressure mean value from the tool force data, extract the maximum value of the vibration amplitude, root mean square value, and vibration frequency amplitude from the vibration data, extract the highest temperatures of the tool, motor, and transmission mechanism parts from the temperature data, and extract the peak value of the motor starting current and the stability index of the working current from the current data. Denote the extracted characteristic parameters as the feature vector X = [x1, x2... x n , where X represents the set of characteristic parameters, n represents the total number of characteristic parameters extracted from the operation data of the deburring device, and x1, x2,..., x n are the 1st, 2nd,..., nth characteristic parameters respectively;

[0057] S202. Construct a life prediction model for the deburring device based on exponential smoothing, defined as follows: L t+1 = α * L t + (1 - α) * (β0 + β1 * x 1,t + β2 * x 2,t +... + β n * x n,t ); where L t+1 represents the remaining life of the deburring device predicted at the next moment, L t represents the estimated value of the remaining life of the deburring device at the current moment. The initial device life is set as the average life of the deburring device. x 1,t , x 2,t ,..., x n,t are the 1st, 2nd,..., nth characteristic parameters collected at time t respectively, t represents the time when the operation characteristic parameters of the deburring device are collected, α represents the smoothing coefficient, and the value range of α is from 0 to 1. β0, β1,..., β n represent the coefficients to be determined.

[0058] In step S202, the batch gradient descent algorithm is used to determine the coefficient β i , and the loss function is defined as follows:

[0059]

[0060] where m represents the number of samples, L j represents the actual remaining life of the jth sample, represents the remaining life of the jth sample predicted according to the model, and J(β) represents the loss function;

[0061] Calculate the gradient, and the formula is as follows:

[0062]

[0063] where \(i\) represents the data subscript. For it is \((1 - \alpha)\) when \(i = 0\), and \((1 - \alpha) \times x\) when \(i \neq 0\). i,j ;

[0064] Update the coefficients in each iteration, and the calculation formula is as follows:

[0065]

[0066] where \(\eta\) represents the learning rate. Through multiple iterations until the loss function converges, the optimized coefficient \(\beta\) is obtained. i , and the construction of the burr removal equipment life prediction model based on exponential smoothing is completed.

[0067] In step S300, the specific steps include:

[0068] S301. Use the multi-layer perceptron MLP deep neural network to perform feature learning and pattern recognition on the operation data of the burr removal equipment. Take the multi-source data features extracted above as the input of the burr removal equipment life prediction model, and take the remaining life of the corresponding burr removal equipment in the historical data as the output label to train the burr removal equipment life prediction model based on exponential smoothing;

[0069] S302. Input the operation data of the burr removal equipment collected in real time into the trained burr removal equipment life prediction model to obtain the predicted value of the current remaining life of the burr removal equipment. Analyze the change range of the equipment operation data in different time periods, and calculate the attenuation rate of the burr removal equipment life. The formula is as follows:

[0070]

[0071] where \(R\) decay represents the attenuation rate of the burr removal equipment life, \(\Delta F\) i represents the change range of the \(i\)-th characteristic parameter within the time period \(T\), \(w\) i represents the weight of the \(i\)-th feature, and \(\Delta T\) represents the average sampling time interval within the time period \(T\). Generate an equipment life evaluation report according to the predicted value of the remaining life, the life attenuation rate, and the equipment operation conditions and maintenance record information.

[0072] In step S400, the specific steps include:

[0073] S401. Calculate the operation stability index of the burr removal equipment according to the number of faults and the duration of faults of the burr removal equipment within the time period. The calculation formula of the operation stability index of the burr removal equipment is as follows:

[0074]

[0075] Among them, k represents the fault number, and the range of k is k = 1, 2,..., F. T represents the total length of the time period, and t k represents the duration of the k-th fault. F represents the number of deburring equipment faults, and S represents the operation stability index of the deburring equipment;

[0076] S402. Set the stability warning threshold S w , when S is less than S w , trigger the warning mechanism, mark that there is an abnormality in the stability of the current deburring equipment, and there is a positive correlation between the equipment operation stability index and the deburring equipment life. Monitor the service life of the deburring equipment according to the positive correlation between the equipment operation stability index and the deburring equipment life. Among them, the stability warning threshold is obtained based on the service life specified by the deburring equipment manufacturer combined with the historical data since the deburring equipment was put into use;

[0077] S403. Use the Industrial Internet of Things (IIoT) technology to transmit the equipment monitoring data and life prediction results to the remote monitoring platform through the wireless communication protocol MQTT.

[0078] A deburring equipment life monitoring and transmission system that integrates multi-source data. The system includes: a data acquisition module, an equipment health status analysis module, a life prediction module, and an equipment stability analysis module. The data acquisition module is used to collect multi-source operation data of the deburring equipment and perform preprocessing. The equipment health status analysis module is used to extract the key features of the equipment operation from the collected original data, fuse the multi-source features into a multi-dimensional feature vector, and input the multi-dimensional feature vector into the life prediction model. The life prediction module is used to train a deburring equipment life prediction model based on exponential smoothing and calculate the life decay rate of the deburring equipment. The equipment stability analysis module is used to analyze the fault situation of the equipment within the time period and evaluate the stability of the equipment. The output end of the data acquisition module is connected to the input end of the equipment health status analysis module, the output end of the equipment health status analysis module is connected to the input end of the life prediction module, and the output end of the life prediction module is connected to the input end of the equipment stability analysis module.

[0079] The data acquisition module includes an operating data acquisition unit and a data preprocessing unit. The operating data acquisition unit uses installed sensors to obtain the operating data of the deburring device in real time, including tool force, vibration frequency and amplitude, the temperatures of the tool, motor, and transmission mechanism, and the working current of the motor. All the collected data is arranged in chronological order to form a multi-dimensional time series data stream. The data preprocessing unit is used to uniformly process the raw data from different sensors, convert it into a standardized time series format, normalize the data using the Min-Max normalization method, and identify and remove outliers and missing values through an algorithm based on an anomaly detection model. At the same time, the timestamps are calibrated and aligned.

[0080] The device health status analysis module includes a feature extraction unit and a remaining life prediction model establishment unit. The feature extraction unit is used to extract feature parameters related to the remaining life of the deburring device from the integrated data. It extracts the peak pressure, pressure fluctuation amplitude, and pressure mean from the tool force data, the maximum vibration amplitude, root mean square value, and vibration frequency amplitude from the vibration data, the highest temperatures of the tool, motor, and transmission mechanism parts from the temperature data, and the peak starting current of the motor and the stability index of the working current from the current data. The remaining life prediction model establishment unit is used to construct a remaining life prediction model of the deburring device based on exponential smoothing according to the extracted features, and optimize the model coefficients through batch gradient descent to complete life estimation and quantitative analysis.

[0081] The remaining life prediction module includes a remaining life prediction model training unit and a remaining life decay rate calculation unit. The remaining life prediction model training unit is used to perform feature learning and pattern recognition on the operating data of the deburring device using a multi-layer perceptron (MLP) deep neural network. It takes the multi-source data features extracted as input and the remaining life of the device in historical data as the output label to train the device remaining life prediction model. The remaining life decay rate calculation unit is used to analyze the change amplitude of the device operating data in different time periods and calculate the remaining life decay rate of the deburring device. The device stability analysis module includes a fault statistics unit and a stability calculation unit. The fault statistics unit is used to count the number of faults and the duration of faults that occur in the device within a time period. The stability calculation unit is used to calculate the stability index of the device based on the number of faults and the duration, and use industrial Internet of Things (IIoT) technology to transmit the device monitoring data and remaining life prediction results to a remote monitoring platform through the wireless communication protocol MQTT.

[0082] In an embodiment: Tool force data, vibration data, temperature data, and current data of the deburring device are obtained through sensors. The pressure sensor samples once per second to record the force change of the tool. The collected data shows that the pressure peak is 500 N, the pressure fluctuation amplitude is 400 N, and the pressure mean value is 300 N. It samples once per second to record the vibration amplitude and frequency of the device. The maximum vibration amplitude collected is 2.5 mm, the root mean square value of vibration is 1.8 mm, and the vibration frequency amplitude is 120 Hz. It samples once per minute to record the temperatures of the tool, motor, and transmission mechanism. The collected tool temperature is 70 °C, the motor temperature is 85 °C, and the transmission mechanism temperature is 80 °C. It samples once per minute to record the working current of the motor. The peak value of the motor starting current collected is 10 A, and the working current stability is 5.2 A. Using the Min - Max normalization method, the range of all sensor data is adjusted to [0, 1]. Through rule - and model - based anomaly detection, outliers and missing values are removed. When the temperature data at a certain sampling point is much higher than the normal operating temperature threshold of 100 °C, it is regarded as an outlier and removed. All sensor data is aligned according to the timestamp. Feature parameters related to the remaining life of the deburring device are extracted from the integrated data. The pressure peak, pressure fluctuation amplitude, and pressure mean value are extracted from the tool force data. The maximum value of the vibration amplitude, root mean square value, and vibration frequency amplitude are extracted from the vibration data. The highest temperatures of the tool, motor, and transmission mechanism parts are extracted from the temperature data. The peak value of the motor starting current and the stability index of the working current are extracted from the current data. Through fusion, a unified feature vector X = [500, 400, 300, 2.5, 1.8, 120, 70, 85, 80, 10, 5.2] is formed. Using a multi - layer perceptron (MLP) neural network, it is trained with the historical data of the device and the remaining life labels. The training dataset includes the feature of the device operation data and the known remaining life. The feature vector of the deburring device is input, and the predicted value of the remaining life of the deburring device is output. The predicted value of the remaining life of the deburring device is 150 hours. According to the change of pressure from 300 N to 500 N, the vibration amplitude from 1.8 mm to 2.5 mm, the temperature data from 80 °C to 85 °C, and the current peak from 5.2 A to 10 A, and the change amplitudes of other extracted feature parameters are relatively small within the time period and the influence on the life decay rate is negligible, so they are not included in the calculation of the change amplitude of the feature parameters. The weights of each feature are set to 0.3, 0.25, 0.2, and 0.25 respectively. The average sampling interval within the time period is 60 s. Using the life decay rate calculation formula, the calculated life decay rate is 1.04. A failure statistics of the deburring device is carried out, and it is obtained that within the past 48 hours, the device has had 3 failures, and the failure durations are 1 hour, 0.5 hour, and 0.3 hour respectively. According to the stability index calculation, the stability index S of the device is obtained as 0.9625. According to the service life specified in the factory regulations of the deburring equipment and the historical data since the deburring equipment was put into use, the stability warning threshold is obtained as 0.95. The stability index of the equipment is greater than the stability warning threshold, indicating that the current operating state of the equipment is good, its life is within the normal range, and there is no need to trigger a warning.

[0083] For those skilled in the art, it is obvious that the present invention is not limited to the details of the above-described exemplary embodiments, and the present invention can be implemented in other specific forms without departing from the spirit or basic characteristics of the present invention. Therefore, from any point of view, the embodiments should be regarded as exemplary and non-limiting. The scope of the present invention is defined by the appended claims rather than the above description. Therefore, all changes falling within the meaning and scope of the equivalent elements of the claims are intended to be embraced by the present invention. Any reference signs in the claims should not be construed as limiting the claims involved.

Claims

1. Deburring equipment life monitoring and transmission method for integrating multi-source data, characterized in that: The method includes the following steps: S100. Collect the operation data of the deburring device and integrate the operation data; S200. Extract the characteristic parameters related to the life of the deburring device from the integrated data, construct a life prediction model based on exponential smoothing using the characteristic parameters, and optimize the model coefficients through batch gradient descent; S300. Collect the characteristic parameter data in real time, perform life prediction using the life prediction model, and analyze the life decay rate in combination with the historical characteristic parameter data; S400. Analyze the operation stability of the deburring device, set a warning threshold, monitor the device stability and life, and transmit the monitoring data and prediction results through industrial Internet of Things technology.

2. The method for monitoring and transmitting the life of the deburring device that integrates multi-source data according to claim 1, characterized in that: In step S100, the specific steps include: S101. Obtain the tool force, vibration frequency and amplitude involved in the operation of the deburring device, as well as the temperatures of the tool, motor, and transmission mechanism and the motor working current through sensors, and arrange the collected data according to the time axis to form a multi-dimensional time series data stream; S102. Convert the data collected by different sensors into a unified time series format, perform normalization processing using Min-Max, use an anomaly detection model based on it to identify and remove the outliers and missing values in the data, calibrate and align the timestamps, and sort out the operation status data of the deburring device in different time periods.

3. The method for monitoring and transmitting the lifespan of the deburring device that fuses multi-source data according to claim 2, wherein: In step S200, the specific steps include: S201. Extract the characteristic parameters related to the life of the deburring device from the integrated data. Extract the peak pressure, pressure fluctuation amplitude, and pressure mean value from the tool force data. Extract the maximum value of the vibration amplitude, root mean square value, and vibration frequency amplitude from the vibration data. Extract the highest temperatures of the tool, motor, and transmission mechanism parts from the temperature data. Extract the peak value of the motor starting current and the stability index of the working current from the current data. Denote the extracted characteristic parameters as the feature vector X = [x1, x2... x n , where X represents the set of characteristic parameters, n represents the total number of characteristic parameters extracted from the operation data of the deburring device, and x1, x2,..., x n are the 1st, 2nd,..., nth characteristic parameters respectively; S202. Construct a deburring equipment life prediction model based on exponential smoothing, which is defined as follows: L t+1 = α * L t + (1 - α) * (β0 + β1 * x 1,t + β2 * x 2,t +... + β n * x n,t ); where, L t+1 represents the remaining life of the deburring equipment predicted at the next moment, L t represents the estimated value of the remaining life of the deburring equipment at the current moment. The initial equipment life is set as the average life of the deburring equipment. x 1,t , x 2,t ,..., x n,t respectively represent the 1st, 2nd,..., nth characteristic parameters collected at time t. t represents the time when the operating characteristic parameters of the deburring equipment are collected. α represents the smoothing coefficient, and the value range of α is from 0 to 1. β0, β1,..., β n represent the coefficients to be determined.

4. The method for monitoring and transmitting the lifespan of the deburring device that integrates multi-source data according to claim 3, wherein: In step S202, the batch gradient descent algorithm is used to determine the coefficient β i , and the loss function is defined as follows: where m represents the number of samples, and L j represents the actual remaining life of the j-th sample, represents the remaining life of the j-th sample predicted according to the model, and J(β) represents the loss function; Calculate the gradient, and the formula is as follows: where i represents the data subscript. For it is (1 - α) when i = 0 and (1 - α) * x when i ≠ 0 i,j ; Update the coefficient in each iteration, and the calculation formula is as follows: Among them, η represents the learning rate. Through multiple iterations until the loss function converges, the optimized coefficient β is obtained. i , and the construction of the deburring equipment life prediction model based on exponential smoothing is completed.

5. The method for monitoring and transmitting the lifespan of a deburring device that fuses multi-source data according to claim 4, characterized in that: In step S300, the specific steps include: S301. Use a multi-layer perceptron MLP deep neural network to perform feature learning and pattern recognition on the operation data of the deburring device, use the multi-source data features extracted above as the input of the deburring device life prediction model, and use the remaining life of the corresponding deburring device in the historical data as the output label to train the deburring device life prediction model based on exponential smoothing; S302. Input the operation data of the deburring device collected in real time into the trained deburring device life prediction model to obtain the predicted value of the current remaining life of the deburring device, analyze the change range of the device operation data in different time periods, and calculate the decay rate of the deburring device life. The formula is as follows: Among them, R decay represents the deburring equipment life attenuation rate, ΔF i represents the change amplitude of the i-th characteristic parameter within the time period T, w i represents the weight of the i-th characteristic, ΔT represents the average sampling time interval within the time period T, and an equipment life evaluation report is generated based on the remaining life prediction value, life attenuation rate, and equipment operating conditions and maintenance record information.

6. The method for monitoring and transmitting the lifespan of the deburring device that fuses multi-source data according to claim 5, characterized in that: In step S400, specifically: S401. Calculate the operation stability index of the deburring device according to the number of failures and the duration of the failures of the deburring device within a time period. The calculation formula of the operation stability index of the deburring device is as follows: Among them, k represents the fault number, and the range of k is k = 1, 2,..., F. T represents the total length of the time period, and t k represents the duration of the k-th fault, F represents the number of burr removal equipment faults, and S represents the operation stability index of the burr removal equipment; S402. Set the stability warning threshold S w , when S is less than S w , trigger the warning mechanism, mark that there is an abnormality in the stability of the current deburring device, and there is a positive correlation between the device operation stability index and the deburring device life. Monitor the service life of the deburring device according to the positive correlation between the device operation stability index and the deburring device life; S403. Use industrial Internet of Things IIoT technology to transmit the monitoring data and life prediction results of the device to the remote monitoring platform through the wireless communication protocol MQTT.

7. A deburring equipment life monitoring and transmission system that integrates multi-source data, characterized in that: The system includes: a data acquisition module, a device health status analysis module, a life prediction module, and a device stability analysis module. The data acquisition module is used to collect multi-source operation data of the deburring device and perform preprocessing. The device health status analysis module is used to extract key features of the device operation from the collected raw data, fuse the multi-source features into a multi-dimensional feature vector, and input the multi-dimensional feature vector into the life prediction model. The life prediction module is used to train a life prediction model of the deburring device based on exponential smoothing and calculate the life decay rate of the deburring device. The device stability analysis module is used to analyze the fault situation of the device within a time period and evaluate the stability of the device. The output end of the data acquisition module is connected to the input end of the device health status analysis module, the output end of the device health status analysis module is connected to the input end of the life prediction module, and the output end of the life prediction module is connected to the input end of the device stability analysis module.

8. The burr removal equipment life monitoring and transmission system integrating multi-source data according to claim 7, characterized in that: The data acquisition module includes an operation data acquisition unit and a data preprocessing unit. The operation data acquisition unit obtains the operation data of the deburring device in real time through the installed sensors, including the tool force, vibration frequency and amplitude, the temperatures of the tool, motor, and transmission mechanism, and the working current of the motor, and arranges all the collected data in chronological order to form a multi-dimensional time series data stream. The data preprocessing unit is used to uniformly process the raw data from different sensors, convert it into a standardized time series format, normalize the data using the Min-Max normalization method, identify and remove outliers and missing values through an algorithm based on an anomaly detection model, and at the same time calibrate and align the timestamps.

9. The deburring equipment life monitoring and transmission system for fusing multi-source data according to claim 8, characterized in that: The device health status analysis module includes a feature extraction unit and a life prediction model establishment unit. The feature extraction unit is used to extract feature parameters related to the life of the deburring device from the integrated data, extract the pressure peak value, pressure fluctuation amplitude, and pressure mean value from the tool force data, extract the maximum value of the vibration amplitude, root mean square value, and vibration frequency amplitude from the vibration data, extract the highest temperatures of the tool, motor, and transmission mechanism parts from the temperature data, and extract the motor starting current peak value and the stability index of the working current from the current data. The life prediction model establishment unit is used to construct a life prediction model of the deburring device based on exponential smoothing according to the extracted features and optimize the model coefficients through batch gradient descent.

10. The burr removal equipment life monitoring and transmission system integrating multi-source data according to claim 9, characterized in that: The life prediction module includes a life prediction model training unit and a life decay rate calculation unit. The life prediction model training unit is used to perform feature learning and pattern recognition on the operation data of the deburring device using a multi-layer perceptron (MLP) deep neural network, taking the extracted multi-source data features as input and the remaining life of the device in historical data as the output label to train the device life prediction model. The life decay rate calculation unit is used to analyze the change amplitude of the device operation data in different time periods and calculate the decay rate of the deburring device life. The device stability analysis module includes a fault statistics unit and a stability calculation unit. The fault statistics unit is used to count the number of faults and the duration of faults that occur in the device within a time period. The stability calculation unit is used to calculate the stability index of the device based on the number of faults and the duration, and use the industrial Internet of Things (IIoT) technology to transmit the monitoring data and life prediction results of the device to the remote monitoring platform through the wireless communication protocol MQTT.

Images