Production equipment management system and method based on multi-source data analysis

Abstract

The invention discloses a production equipment management system and method based on multi-source data analysis, and relates to the technical field of production equipment fault diagnos.The production equipment management system comprises the steps that historical operation data records of production equipment are collected, collection of operation data in the historical operation data records is stopped until a fault occurs, and classification is conducted according to preset fault type labels; calculating a target time period; the method comprises the following steps: acquiring data in a target time period, and calculating temperature, temperature increase rate, vibration frequency, a vibration intensity threshold and a fault evaluation index threshold; the method comprises the following steps: monitoring equipment operation data in real time, judging and sending out a fault early warning prompt according to each threshold value, and adjusting operation parameters within a target duration; the corresponding management system comprises a data acquisition and preprocessing module, a threshold calculation module, a fault evaluation index calculation module and an early warning module. Through multi-source data acquisition and analysis, production equipment faults can be predicted more accurately, and the defects of a traditional single data acquisition mode are effectively overcome.

Description

Technical Field

[0001] This invention relates to the field of production equipment fault diagnosis technology, specifically a production equipment management system and method based on multi-source data analysis. Background Technology

[0002] In industrial production, production equipment plays a crucial role. For example, in the field of modern mechanical manufacturing, milling machines are key pieces of equipment, and accurate monitoring of their operating status and fault diagnosis are of paramount importance for ensuring production efficiency, reducing maintenance costs, and improving product quality.

[0003] Currently, traditional milling machine fault diagnosis techniques mainly rely on collecting and analyzing conventional parameters such as machine speed, current, and voltage. While this single data acquisition mode can reflect the operating status of the equipment to some extent, it has significant limitations. For example, when milling machine tools experience wear or chipping, the initial symptoms may only be minor vibrations or temperature changes. These subtle changes are not obvious, and relying solely on collecting and analyzing conventional parameters is insufficient to comprehensively capture early signs of equipment failure, easily leading to missed or misdiagnosed faults, which can then cause equipment downtime or even more serious production accidents. Therefore, there is an urgent need for a fault prediction method for production equipment to achieve early prediction and prevention of faults. Summary of the Invention

[0004] The purpose of this invention is to provide a production equipment management system and method based on multi-source data analysis to solve the problems raised in the prior art.

[0005] To achieve the above objectives, the present invention provides the following technical solution: a production equipment management method based on multi-source data analysis, the management method comprising: Step S100: Acquire historical operating data records of the production equipment during the production process; the collection of operating data in each historical operating data record ends when a fault occurs; wherein, the historical operating data includes vibration frequency data, vibration intensity data, high-temperature areas of the production equipment during operation, and temperature data corresponding to the high-temperature areas; classify the historical operating data records according to preset fault type labels; calculate the target time period; Step S200: Obtain the historical operating data of the production equipment within the target time period from each classified historical operating data record; calculate the temperature threshold, temperature growth rate threshold, vibration frequency threshold, and vibration intensity threshold for each target area; Step S300: Extract temperature data, vibration frequency data, and vibration intensity data collected for each target area within the target time period; calculate the fault assessment index threshold for each target area; Step S400: Monitor the operating data of the production equipment in real time, and judge the operating data of the production equipment based on the temperature threshold, temperature growth rate threshold, vibration frequency threshold, vibration intensity threshold, and fault assessment index threshold, and issue a fault warning.

[0006] Furthermore, step S100 includes: Step S101: Acquire acoustic data of the production equipment at regular intervals using an acoustic sensor; wherein, the acoustic data includes vibration frequency data and vibration intensity data of the production equipment during operation; acquire infrared imaging data of the production equipment at regular intervals using an infrared imager; wherein, the infrared imaging data includes high-temperature areas of the production equipment during operation and temperature data corresponding to the high-temperature areas; the acoustic data and the infrared imaging data are acquired synchronously. Step S102: Label the historical operation data records with preset fault type tags according to fault type; group the historical data records carrying the same fault type tag into a group; identify the high temperature area in each historical operation data record in the same group, and set the high temperature area as the target area; classify the historical data records corresponding to the same target area in the same group into one category; Step S103: Extract vibration intensity data and temperature data of the target area from the same type of historical data records; sort the vibration intensity data and temperature data of the target area according to the order of collection time to generate a feature map; Step S104: Extract the segment of vibration intensity rise curve and temperature rise curve closest to the end time of the acquisition from the feature map; extract the start time t1 and end time t1' of the vibration intensity rise curve, and the start time t2 and end time t2' of the temperature rise curve; Step S105: Select the largest start time from the start times t1 and t2 as the target start time t; select the smallest end time from the end times t1' and t2' as the target end time t'; the target time period is [t, t'], and calculate the target duration Δt = t' - t for each target time period in the same type of historical running data records; The above steps involve simultaneously acquiring multi-source data using acoustic fingerprint sensors and infrared imagers to comprehensively obtain equipment operation information, overcoming the shortcomings of traditional single-data acquisition and providing rich evidence for fault diagnosis. Historical operating data is categorized by fault type, focusing on target areas and further categorizing them to make subsequent analysis more targeted and facilitate the discovery of data features related to specific faults and areas. Vibration intensity and temperature data are extracted to generate feature maps, presenting the data in an intuitive form for easy observation of data trends. Key rising curves near the end of the acquisition process are extracted from the feature maps, and time points are obtained, as these curves contain important information about when production equipment faults are imminent. The target time period and duration are determined, providing a unified and crucial time standard for subsequent threshold calculations and fault risk assessment, thereby improving the accuracy and timeliness of fault diagnosis and effectively predicting equipment faults.

[0007] Furthermore, step S200 includes: Step S201: Extract temperature data of the target area collected within each target time period from the same type of historical operation data records; perform a weighted average of the temperature data of the target area collected within each target time period to obtain the average temperature; set the average temperature as the temperature threshold of the target area; calculate the temperature growth rate of the target area within each target time period. Among them, V C C represents the rate of temperature increase in the target area within each target time period. t This represents the temperature value of the target area at the start time of each target time period, expressed in C. t The temperature value of the target region at the end of each target time period is represented; the temperature growth rate threshold is obtained by weighted averaging the temperature growth rates corresponding to each target time period in the target region. Step S202: Extract vibration frequency data of the target area collected in each target time period from the same type of historical operation data records; perform discrete Fourier transform on the extracted vibration frequency data to obtain the frequency distribution spectrum; select the vibration frequency range with the highest energy proportion in the frequency distribution spectrum; take the center value of the vibration frequency range, and set the center value as the vibration frequency threshold of the target area; Step S203: Divide each target time period in the same type of historical operation data records into several time windows according to a preset window size; fit the vibration intensity data included in each time window using the least squares method to obtain the slope α. i ; where a i Represent the slope corresponding to the vibration intensity data included in the i-th time window; calculate the rate of change of slope between two adjacent slopes. Where r represents the rate of change of the slope between two adjacent slopes, and a i+1The slope of the vibration intensity data included in the (i+1)th time window is represented by ΔT; the time interval between two adjacent time windows is represented by ΔT. The slope change rate calculated in each target time period is integrated. If the slope change rate in the target time period is greater than a preset slope change rate threshold, the time window corresponding to the minimum slope change rate that is greater than the preset slope change rate threshold is selected. The vibration intensity data in the time window is extracted, and the minimum vibration intensity in the vibration intensity data is selected. The minimum vibration intensity corresponding to each target time period in the target region is weighted and averaged to obtain the vibration intensity threshold of the target region. Step S204: Integrate the temperature threshold, temperature growth rate threshold, vibration frequency threshold, and vibration intensity threshold corresponding to each target area in the same fault type label. The above steps, by extracting temperature data from the target area and performing a weighted average to determine the temperature threshold, and simultaneously calculating the temperature growth rate and its threshold, reflect the stable state and changing trend of the temperature in the target area, providing a standard for judging whether the temperature of production equipment is abnormal. Performing a discrete Fourier transform on the vibration frequency data and selecting the center value of the frequency range with the highest energy proportion as the threshold can capture the main frequency characteristics of the production equipment vibration, more accurately judging whether the vibration is normal. By dividing the time window and fitting the vibration intensity data to obtain the slope, and then calculating the slope change rate, the minimum vibration intensity within the time window is selected based on the change rate to determine the vibration intensity threshold. This method can keenly capture abnormal fluctuations in vibration intensity. Integrating various thresholds and associating different thresholds for different target areas under the same fault type label facilitates subsequent comprehensive judgment of equipment operating status, providing a comprehensive and accurate reference for equipment fault early warning, and improving the reliability of fault diagnosis.

[0008] Furthermore, step S300 includes: Step S301: Substitute the temperature data, vibration frequency data, and vibration intensity data of a target area within the same fault type label into the normalization formula for normalization processing, obtaining normalized data sets for temperature, vibration frequency, and vibration intensity; the normalization formula is as follows: Where F represents the normalized value of the input data, and X represents the input data; X min X represents the minimum value in the input data; max This represents the maximum value in the input data; Step S302: Calculate the average values ​​of the data in the temperature normalized data set, vibration frequency normalized data set, and vibration intensity normalized data set, respectively. Let the average values ​​be the temperature normalized value F1, vibration frequency normalized value F2, and vibration intensity normalized value F3, respectively. Take the weighted average of the temperature normalized value F1, vibration frequency normalized value F2, and vibration intensity normalized value F3 of a target area in a certain fault type label to obtain the fault assessment index threshold of the target area in the fault type label. The above steps normalize the temperature, vibration frequency, and vibration intensity data of a target area under the same fault type label because these data may have different dimensions and value ranges. Normalization can eliminate the influence of dimensions, making different types of data comparable and transforming them to a unified scale for comprehensive analysis. The average value of each normalized data set is calculated and weighted to obtain the fault assessment index threshold. In this way, data from multiple different dimensions can be combined to form a quantitative indicator, which can comprehensively assess the fault risk level of the target area under this fault type. This provides an intuitive and effective basis for subsequent judgment on whether there are potential faults in the equipment based on this threshold, thereby improving the scientificity and accuracy of fault assessment.

[0009] Furthermore, step S400 includes: The system monitors the temperature, temperature rise rate, vibration frequency, and vibration intensity of the production equipment in real time. If the temperature of a target area of ​​the production equipment is found to be higher than a temperature threshold, a fault warning is issued based on the fault type label corresponding to the target area, and the operating parameters of the production equipment are adjusted within a target duration Δt to lower the temperature below the temperature threshold. If the temperature of a target area of ​​the production equipment is found to be lower than the temperature threshold, and any one of the temperature rise rate, vibration frequency, or vibration intensity is found to be higher than the corresponding threshold, a fault warning is issued based on the fault type label corresponding to the target area, and the operating parameters of the production equipment are adjusted within a target duration Δt to bring the data above the threshold below the threshold. If the temperature, temperature rise rate, vibration frequency, and vibration intensity of a target area of ​​the production equipment are all found to be lower than the corresponding threshold, a fault assessment index for the target area is calculated. If the fault assessment index for the target area is higher than the corresponding fault assessment index threshold, a fault warning is issued based on the fault type label corresponding to the target area, and the operating parameters of the production equipment are adjusted within a target duration Δt to bring the fault assessment index below the fault assessment index threshold.

[0010] Furthermore, to better implement the above method, a production equipment management system based on multi-source data analysis is also provided. The management system includes: a data acquisition and preprocessing module, a threshold calculation module, a fault assessment index calculation module, and an early warning module. The data acquisition and preprocessing module is responsible for collecting historical operating data records of the production equipment during operation, and preprocessing and extracting features from the historical operating data records. The threshold calculation module calculates various thresholds for judging production equipment failures based on preprocessed historical operating data. The fault assessment index calculation module obtains the fault assessment index threshold by normalizing the preprocessed historical operating data. The monitoring and early warning module is used to monitor the operating data of production equipment in real time and to make judgments and issue early warnings based on the operating data.

[0011] Furthermore, the data acquisition and preprocessing module includes a data acquisition unit, a data classification unit, and a feature extraction unit; The data acquisition unit periodically collects voiceprint data of the production equipment using a voiceprint sensor and periodically collects infrared imaging data of the production equipment using an infrared imager; the voiceprint data and the infrared imaging data are collected synchronously. The data classification unit labels historical operation data records with preset fault type tags according to fault type; groups historical data records carrying the same fault type tag into a group; identifies high-temperature areas in each historical operation data record in the same group, and sets the high-temperature area as the target area; and classifies historical data records corresponding to the same target area in the same group into one category. The feature extraction unit extracts vibration intensity data and temperature data of the target area from the same type of historical data records; sorts the vibration intensity data and the temperature data of the target area according to the order of collection time to generate a feature map; and obtains the target time period corresponding to each target area based on the feature map and calculates the target duration.

[0012] Furthermore, the threshold calculation module includes a temperature threshold calculation unit, a vibration frequency threshold calculation unit, and a vibration intensity threshold calculation unit; The temperature threshold calculation unit extracts temperature data of the target area collected within each target time period from historical operation data records of the same type; performs a weighted average of the temperature data of the target area collected within each target time period to obtain an average temperature; sets the average temperature as the temperature threshold of the target area; calculates the temperature growth rate of the target area within each target time period; and performs a weighted average of the temperature growth rates corresponding to each target time period of the target area to obtain a temperature growth rate threshold. The vibration frequency threshold calculation unit extracts vibration frequency data of the target area collected within each target time period from the same type of historical operation data records; performs discrete Fourier transform on the extracted vibration frequency data to obtain a frequency distribution spectrum; selects the vibration frequency range with the highest energy proportion in the frequency distribution spectrum; takes the center value of the vibration frequency range, and sets the center value as the vibration frequency threshold of the target area. The vibration intensity threshold calculation unit divides each target time into several time windows with a preset window size; it fits the vibration intensity data included in each time window using the least squares method to obtain the slope; it calculates the slope change between two adjacent slopes; it integrates the slope change rates calculated within each target time period, and if the slope change rate within the target time period is greater than a preset slope change rate threshold, it selects the time window corresponding to the smallest slope change rate greater than the preset slope change rate threshold; it extracts the vibration intensity data within the time window and filters out the minimum vibration intensity among the vibration intensity data; it then performs a weighted average of the minimum vibration intensity corresponding to each target time period in the target region to obtain the vibration intensity threshold of the target region.

[0013] Furthermore, the fault assessment index calculation module includes a data normalization unit and a fault assessment index threshold calculation unit. The data normalization unit substitutes the temperature data, vibration frequency data, and vibration intensity data of a target area in the same fault type label into the normalization formula for normalization processing, and obtains the temperature normalized data set, vibration frequency normalized data set, and vibration intensity normalized data set. The fault assessment index threshold calculation unit calculates the average values ​​of the data in the temperature normalized data set, the vibration frequency normalized data set, and the vibration intensity normalized data set, respectively. Let the average values ​​be the temperature normalized value F1, the vibration frequency normalized value F2, and the vibration intensity normalized value F3, respectively. The unit then performs a weighted average of the temperature normalized value F1, the vibration frequency normalized value F2, and the vibration intensity normalized value F3 of a target area in a certain fault type label to obtain the fault assessment index threshold of the target area in the fault type label.

[0014] Furthermore, the early warning module includes a fault judgment unit and an early warning prompt unit; The fault diagnosis unit monitors the temperature, temperature rise rate, vibration frequency, and vibration intensity of the production equipment in real time; and makes judgments on the real-time monitored operating data of the production equipment based on temperature threshold, temperature rise rate threshold, vibration frequency threshold, vibration intensity threshold, and fault assessment index threshold. The early warning unit issues a fault warning based on the judgment result.

[0015] Compared with the prior art, the beneficial effects of the present invention are: 1. Improved Fault Prediction Accuracy: Traditional methods rely solely on conventional parameter collection and analysis, making it difficult to capture early, subtle signs of equipment failure. This invention uses a soundprint sensor and an infrared imager to simultaneously collect multi-source data such as vibration frequency, intensity, high-temperature areas, and temperature, reflecting the equipment's operating status from multiple dimensions. This significantly improves the accuracy of fault prediction and reduces missed detections and misjudgments.

[0016] 2. Take proactive measures: Calculate the thresholds for key data and fault assessment index thresholds, and monitor equipment operation data in real time. Once an anomaly is detected, issue an early warning immediately and adjust the operating parameters within the target time to bring the equipment status back to the normal range, prevent the fault from worsening, reduce equipment downtime and production accidents, and ensure the continuity and stability of production.

[0017] 3. Optimize equipment management and maintenance: By analyzing historical data and calculating thresholds, we can gain a deeper understanding of the operating characteristics of equipment under different fault types, helping managers to develop more reasonable maintenance plans, improve equipment maintenance efficiency, and reduce maintenance costs. Attached Figure Description

[0018] Figure 1 This is a schematic diagram of the process flow of a production equipment management system and method based on multi-source data analysis according to the present invention; Figure 2 This is a schematic diagram of the system structure of a production equipment management system and method based on multi-source data analysis according to the present invention. Detailed Implementation

[0019] Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0020] Example 1: As Figure 1 As shown, the present invention provides a technical solution, a production equipment management method based on multi-source data analysis, the management method comprising: Step S100: Acquire historical operating data records of the production equipment during the production process; the collection of operating data in each historical operating data record ends when a fault occurs; wherein, the historical operating data includes vibration frequency data, vibration intensity data, high-temperature areas of the production equipment during operation, and temperature data corresponding to the high-temperature areas; classify the historical operating data records according to preset fault type labels; calculate the target time period; Step S100 includes: Step S101: Acquire acoustic data of the production equipment at regular intervals using an acoustic sensor; wherein, the acoustic data includes vibration frequency data and vibration intensity data of the production equipment during operation; acquire infrared imaging data of the production equipment at regular intervals using an infrared imager; wherein, the infrared imaging data includes high-temperature areas of the production equipment during operation and temperature data corresponding to the high-temperature areas; the acoustic data and the infrared imaging data are acquired synchronously. Step S102: Label the historical operation data records with preset fault type tags according to fault type; group the historical data records carrying the same fault type tag into a group; identify the high temperature area in each historical operation data record in the same group, and set the high temperature area as the target area; classify the historical data records corresponding to the same target area in the same group into one category; Step S103: Extract vibration intensity data and temperature data of the target area from the same type of historical data records; sort the vibration intensity data and temperature data of the target area according to the order of collection time to generate a feature map; Step S104: Extract the segment of vibration intensity rise curve and temperature rise curve closest to the end time of the acquisition from the feature map; extract the start time t1 and end time t1' of the vibration intensity rise curve, and the start time t2 and end time t2' of the temperature rise curve; Step S105: Select the largest start time from the start times t1 and t2 as the target start time t; select the smallest end time from the end times t1' and t2' as the target end time t'; the target time period is [t, t'], and calculate the target duration Δt = t' - t for each target time period in the same type of historical running data records; Step S200: Obtain the historical operating data of the production equipment within the target time period from each classified historical operating data record; calculate the temperature threshold, temperature growth rate threshold, vibration frequency threshold, and vibration intensity threshold for each target area; Step S200 includes: Step S201: Extract temperature data of the target area collected within each target time period from the same type of historical operation data records; perform a weighted average of the temperature data of the target area collected within each target time period to obtain the average temperature; set the average temperature as the temperature threshold of the target area; calculate the temperature growth rate of the target area within each target time period. Among them, V C C represents the rate of temperature increase in the target area within each target time period. t This represents the temperature value of the target area at the start time of each target time period, expressed in C. tThe temperature value of the target region at the end of each target time period is represented; the temperature growth rate threshold is obtained by weighted averaging the temperature growth rates corresponding to each target time period in the target region. Step S202: Extract vibration frequency data of the target area collected in each target time period from the same type of historical operation data records; perform discrete Fourier transform on the extracted vibration frequency data to obtain the frequency distribution spectrum; select the vibration frequency range with the highest energy proportion in the frequency distribution spectrum; take the center value of the vibration frequency range, and set the center value as the vibration frequency threshold of the target area; Step S203: Divide each target time period in the same type of historical operation data records into several time windows according to a preset window size; fit the vibration intensity data included in each time window using the least squares method to obtain the slope α. i ; where a i Represent the slope corresponding to the vibration intensity data included in the i-th time window; calculate the rate of change of slope between two adjacent slopes. Where r represents the rate of change of the slope between two adjacent slopes, and a i+1 The slope of the vibration intensity data included in the (i+1)th time window is represented by ∆T; the time interval between two adjacent time windows is represented by ∆T. The slope change rate calculated in each target time period is integrated. If the slope change rate in the target time period is greater than a preset slope change rate threshold, the time window corresponding to the minimum slope change rate that is greater than the preset slope change rate threshold is selected. The vibration intensity data in the time window is extracted, and the minimum vibration intensity in the vibration intensity data is selected. The minimum vibration intensity corresponding to each target time period in the target region is weighted and averaged to obtain the vibration intensity threshold of the target region. Step S204: Integrate the temperature threshold, temperature growth rate threshold, vibration frequency threshold, and vibration intensity threshold corresponding to each target area in the same fault type label. Step S300: Extract temperature data, vibration frequency data, and vibration intensity data collected for each target area within the target time period; calculate the fault assessment index threshold for each target area; Step S300 includes: Step S301: Substitute the temperature data, vibration frequency data, and vibration intensity data of a target area within the same fault type label into the normalization formula for normalization processing, obtaining normalized data sets for temperature, vibration frequency, and vibration intensity; the normalization formula is as follows: Where F represents the normalized value of the input data, and X represents the input data; X min X represents the minimum value in the input data;max This represents the maximum value in the input data; Step S302: Calculate the average values ​​of the data in the temperature normalized data set, vibration frequency normalized data set, and vibration intensity normalized data set, respectively. Let the average values ​​be the temperature normalized value F1, vibration frequency normalized value F2, and vibration intensity normalized value F3, respectively. Take the weighted average of the temperature normalized value F1, vibration frequency normalized value F2, and vibration intensity normalized value F3 of a target area in a certain fault type label to obtain the fault assessment index threshold of the target area in the fault type label. Step S400: Monitor the operating data of the production equipment in real time, and judge the operating data of the production equipment based on the temperature threshold, temperature growth rate threshold, vibration frequency threshold, vibration intensity threshold, and fault assessment index threshold, and issue a fault warning prompt. Step S400 includes: The system monitors the temperature, temperature rise rate, vibration frequency, and vibration intensity of the production equipment in real time. If the temperature of a target area of ​​the production equipment is found to be higher than a temperature threshold, a fault warning is issued based on the fault type label corresponding to the target area, and the operating parameters of the production equipment are adjusted within a target duration ∆t to lower the temperature below the temperature threshold. If the temperature of a target area of ​​the production equipment is found to be lower than the temperature threshold, and any one of the temperature rise rate, vibration frequency, or vibration intensity is found to be higher than the corresponding threshold, a fault warning is issued based on the fault type label corresponding to the target area, and the operating parameters of the production equipment are adjusted within a target duration ∆t to bring the data above the threshold below the threshold. If the temperature, temperature rise rate, vibration frequency, and vibration intensity of a target area of ​​the production equipment are all found to be lower than the corresponding threshold, a fault assessment index for the target area is calculated. If the fault assessment index for the target area is higher than the corresponding fault assessment index threshold, a fault warning is issued based on the fault type label corresponding to the target area, and the operating parameters of the production equipment are adjusted within a target duration ∆t to bring the fault assessment index below the fault assessment index threshold. In an embodiment of the invention, an acoustic fingerprint sensor and an infrared imager are installed on a milling machine to synchronously collect data at 10-minute intervals. The collected historical operating data is recorded from the machine startup until the tool wear failure occurs. This data is classified according to the "tool wear" failure type label, and the high-temperature area near the tool is identified as the target area. The historical data of the same target area are then grouped together. Vibration intensity and temperature data are extracted from this type of data, sorted by collection time to generate a feature map, and the target time period is determined to be 2-4 hours before tool wear, with a target duration of 2 hours. For the temperature data of the target area within the target time period, a weighted average is used to obtain a temperature threshold of 50°C and a temperature increase rate threshold of 2°C per hour. Discrete Fourier transform is performed on the vibration frequency data to find the frequency range with the highest energy proportion, which is 50-60Hz. The vibration frequency threshold is set to 55Hz. The target time period is divided into 30-minute windows. Vibration intensity data is fitted and slope change rate is calculated. The minimum vibration intensity is selected and weighted averaged to obtain a vibration intensity threshold of 3 m / s². The temperature, vibration frequency, and vibration intensity data of the target area are normalized to calculate the normalized values ​​F1 (temperature), F2 (vibration frequency), and F3 (vibration intensity). The weighted averaged values ​​yield a fault assessment index threshold of 0.6. The milling machine operation is monitored in real time. When the temperature in the area near the tool reaches 52°C, which is higher than the temperature threshold, the system issues an early warning based on the "tool wear" fault type label and adjusts the milling machine speed within 2 hours to reduce the temperature to below 50°C. Example 2: Figure 2 As shown, in order to better implement the above method, a production equipment management system based on multi-source data analysis is also provided. The management system includes: a data acquisition and preprocessing module, a threshold calculation module, a fault assessment index calculation module, and an early warning module. The data acquisition and preprocessing module is responsible for collecting historical operating data records of the production equipment during operation, and preprocessing and extracting features from the historical operating data records. The threshold calculation module calculates various thresholds for judging production equipment failures based on preprocessed historical operating data. The fault assessment index calculation module obtains the fault assessment index threshold by normalizing the preprocessed historical operating data. The monitoring and early warning module is used to monitor the operating data of production equipment in real time and to make judgments and issue early warnings based on the operating data. The data acquisition and preprocessing module includes a data acquisition unit, a data classification unit, and a feature extraction unit. The data acquisition unit periodically collects voiceprint data of the production equipment using a voiceprint sensor and periodically collects infrared imaging data of the production equipment using an infrared imager; the voiceprint data and the infrared imaging data are collected synchronously. The data classification unit labels historical operation data records with preset fault type tags according to fault type; groups historical data records carrying the same fault type tag into a group; identifies high-temperature areas in each historical operation data record in the same group, and sets the high-temperature area as the target area; and classifies historical data records corresponding to the same target area in the same group into one category. The feature extraction unit extracts vibration intensity data and temperature data of the target area from historical data records of the same type; sorts the vibration intensity data and the temperature data of the target area according to the order of collection time to generate a feature map; and obtains the target time period corresponding to each target area based on the feature map and calculates the target duration. The threshold calculation module includes a temperature threshold calculation unit, a vibration frequency threshold calculation unit, and a vibration intensity threshold calculation unit. The temperature threshold calculation unit extracts temperature data of the target area collected within each target time period from historical operation data records of the same type; performs a weighted average of the temperature data of the target area collected within each target time period to obtain an average temperature; sets the average temperature as the temperature threshold of the target area; calculates the temperature growth rate of the target area within each target time period; and performs a weighted average of the temperature growth rates corresponding to each target time period of the target area to obtain a temperature growth rate threshold. The vibration frequency threshold calculation unit extracts vibration frequency data of the target area collected within each target time period from the same type of historical operation data records; performs discrete Fourier transform on the extracted vibration frequency data to obtain a frequency distribution spectrum; selects the vibration frequency range with the highest energy proportion in the frequency distribution spectrum; takes the center value of the vibration frequency range, and sets the center value as the vibration frequency threshold of the target area. The vibration intensity threshold calculation unit divides each target time into several time windows with a preset window size; it fits the vibration intensity data included in each time window using the least squares method to obtain the slope; it calculates the slope change between two adjacent slopes; it integrates the slope change rates calculated within each target time period, and if the slope change rate within the target time period is greater than a preset slope change rate threshold, it selects the time window corresponding to the smallest slope change rate greater than the preset slope change rate threshold; it extracts the vibration intensity data within the time window and filters out the minimum vibration intensity among the vibration intensity data; it then performs a weighted average of the minimum vibration intensity corresponding to each target time period in the target region to obtain the vibration intensity threshold of the target region. The fault assessment index calculation module includes a data normalization unit and a fault assessment index threshold calculation unit. The data normalization unit substitutes the temperature data, vibration frequency data, and vibration intensity data of a target area in the same fault type label into the normalization formula for normalization processing, and obtains the temperature normalized data set, vibration frequency normalized data set, and vibration intensity normalized data set. The fault assessment index threshold calculation unit calculates the average values ​​of the data in the temperature normalized data set, the vibration frequency normalized data set, and the vibration intensity normalized data set, respectively, and assuming that the average values ​​are the temperature normalized value F1, the vibration frequency normalized value F2, and the vibration intensity normalized value F3, respectively; and performs a weighted average of the temperature normalized value F1, the vibration frequency normalized value F2, and the vibration intensity normalized value F3 of a target area in a certain fault type label to obtain the fault assessment index threshold of the target area in the fault type label; The early warning module includes a fault judgment unit and an early warning prompt unit; The fault diagnosis unit monitors the temperature, temperature rise rate, vibration frequency, and vibration intensity of the production equipment in real time; and makes judgments on the real-time monitored operating data of the production equipment based on temperature threshold, temperature rise rate threshold, vibration frequency threshold, vibration intensity threshold, and fault assessment index threshold. The early warning unit issues a fault warning based on the judgment result.

[0021] Finally, it should be noted that the above descriptions are merely preferred embodiments of the present invention and are not intended to limit the present invention. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art can still modify the technical solutions described in the foregoing embodiments or make equivalent substitutions for some of the technical features. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A production equipment management method based on multi-source data analysis, characterized in that, The management method includes: Step S100: Acquire historical operating data records of the production equipment during the production process; the collection of operating data in each historical operating data record ends when a fault occurs; wherein, the historical operating data includes vibration frequency data, vibration intensity data, high-temperature areas of the production equipment during operation, and temperature data corresponding to the high-temperature areas; classify the historical operating data records according to preset fault type labels; calculate the target time period; Step S200: Obtain the historical operating data of the production equipment within the target time period from each classified historical operating data record; calculate the temperature threshold, temperature growth rate threshold, vibration frequency threshold, and vibration intensity threshold for each target area; Step S300: Extract temperature data, vibration frequency data, and vibration intensity data collected for each target area within the target time period; calculate the fault assessment index threshold for each target area; Step S400: Monitor the operating data of the production equipment in real time, and judge the operating data of the production equipment based on the temperature threshold, temperature growth rate threshold, vibration frequency threshold, vibration intensity threshold, and fault assessment index threshold, and issue a fault warning.

2. The production equipment management method based on multi-source data analysis according to claim 1, characterized in that... Step S100 includes: Step S101: Acquire acoustic data of the production equipment at regular intervals using an acoustic sensor; wherein, the acoustic data includes vibration frequency data and vibration intensity data of the production equipment during operation; acquire infrared imaging data of the production equipment at regular intervals using an infrared imager; wherein, the infrared imaging data includes high-temperature areas of the production equipment during operation and temperature data corresponding to the high-temperature areas; the acoustic data and the infrared imaging data are acquired synchronously. Step S102: Label the historical operation data records with preset fault type tags according to fault type; group the historical data records carrying the same fault type tag into a group; identify the high temperature area in each historical operation data record in the same group, and set the high temperature area as the target area; classify the historical data records corresponding to the same target area in the same group into one category; Step S103: Extract vibration intensity data and temperature data of the target area from the same type of historical data records; sort the vibration intensity data and temperature data of the target area according to the order of collection time to generate a feature map; Step S104: Extract the segment of vibration intensity rise curve and temperature rise curve closest to the end time of the acquisition from the feature map; extract the start time t1 and end time t1' of the vibration intensity rise curve, and the start time t2 and end time t2' of the temperature rise curve; Step S105: Select the largest start time from the start times t1 and t2 as the target start time t; select the smallest end time from the end times t1' and t2' as the target end time t'; the target time period is [t, t'], and calculate the target duration ∆t = t' - t for each target time period in the same type of historical running data records.

3. The production equipment management method based on multi-source data analysis according to claim 1, characterized in that, Step S200 includes: Step S201: Extract temperature data of the target area collected within each target time period from the same type of historical operation data records; perform a weighted average of the temperature data of the target area collected within each target time period to obtain the average temperature; set the average temperature as the temperature threshold of the target area; calculate the temperature growth rate of the target area within each target time period. Among them, V C C represents the rate of temperature increase in the target area within each target time period. t This represents the temperature value of the target area at the start time of each target time period, expressed in C. t The temperature value of the target region at the end of each target time period is represented; the temperature growth rate threshold is obtained by weighted averaging the temperature growth rates corresponding to each target time period in the target region. Step S202: Extract vibration frequency data of the target area collected in each target time period from the same type of historical operation data records; perform discrete Fourier transform on the extracted vibration frequency data to obtain the frequency distribution spectrum; select the vibration frequency range with the highest energy proportion in the frequency distribution spectrum; take the center value of the vibration frequency range, and set the center value as the vibration frequency threshold of the target area; Step S203: Divide each target time period in the same type of historical operation data records into several time windows according to a preset window size; fit the vibration intensity data included in each time window using the least squares method to obtain the slope α. i ; where a i Represent the slope corresponding to the vibration intensity data included in the i-th time window; calculate the rate of change of slope between two adjacent slopes. Where r represents the rate of change of the slope between two adjacent slopes, and a i+1 The slope of the vibration intensity data included in the (i+1)th time window is represented by ∆T; the time interval between two adjacent time windows is represented by ∆T. The slope change rate calculated in each target time period is integrated. If the slope change rate in the target time period is greater than a preset slope change rate threshold, the time window corresponding to the minimum slope change rate that is greater than the preset slope change rate threshold is selected. The vibration intensity data in the time window is extracted, and the minimum vibration intensity in the vibration intensity data is selected. The minimum vibration intensity corresponding to each target time period in the target region is weighted and averaged to obtain the vibration intensity threshold of the target region. Step S204: Integrate the temperature threshold, temperature growth rate threshold, vibration frequency threshold, and vibration intensity threshold corresponding to each target area in the same fault type label.

4. The production equipment management method based on multi-source data analysis according to claim 1, characterized in that, Step S300 includes: Step S301: Substitute the temperature data, vibration frequency data, and vibration intensity data of a target area within the same fault type label into the normalization formula for normalization processing, obtaining normalized data sets for temperature, vibration frequency, and vibration intensity; the normalization formula is as follows: Where F represents the normalized value of the input data, and X represents the input data; X min X represents the minimum value in the input data; max This represents the maximum value in the input data; Step S302: Calculate the average values ​​of the data in the temperature normalized data set, vibration frequency normalized data set, and vibration intensity normalized data set, respectively. Let the average values ​​be the temperature normalized value F1, vibration frequency normalized value F2, and vibration intensity normalized value F3, respectively. Take a weighted average of the temperature normalized value F1, vibration frequency normalized value F2, and vibration intensity normalized value F3 of a target area in a certain fault type label to obtain the fault assessment index threshold of the target area in the fault type label.

5. The production equipment management method based on multi-source data analysis according to claim 1, characterized in that, Step S400 includes: The system monitors the temperature, temperature rise rate, vibration frequency, and vibration intensity of the production equipment in real time. If the temperature of a target area of ​​the production equipment is found to be higher than a temperature threshold, a fault warning is issued based on the fault type label corresponding to the target area, and the operating parameters of the production equipment are adjusted within a target duration ∆t to lower the temperature below the temperature threshold. If the temperature of a target area of ​​the production equipment is found to be lower than the temperature threshold, and any one of the temperature rise rate, vibration frequency, or vibration intensity is found to be higher than the corresponding threshold, a fault warning is issued based on the fault type label corresponding to the target area, and the operating parameters of the production equipment are adjusted within a target duration ∆t to bring the data above the threshold below the threshold. If the temperature, temperature rise rate, vibration frequency, and vibration intensity of a target area of ​​the production equipment are all found to be lower than the corresponding threshold, a fault assessment index for the target area is calculated. If the fault assessment index for the target area is higher than the corresponding fault assessment index threshold, a fault warning is issued based on the fault type label corresponding to the target area, and the operating parameters of the production equipment are adjusted within a target duration ∆t to bring the fault assessment index below the fault assessment index threshold.

6. A production equipment management system based on multi-source data analysis, used to execute the production equipment management method based on multi-source data analysis according to any one of claims 1-5, characterized in that, The management system includes: a data acquisition and preprocessing module, a threshold calculation module, a fault assessment index calculation module, and an early warning module; The data acquisition and preprocessing module is responsible for collecting historical operating data records of the production equipment during operation, and preprocessing and extracting features from the historical operating data records. The threshold calculation module calculates various thresholds for judging production equipment failures based on preprocessed historical operating data. The fault assessment index calculation module obtains the fault assessment index threshold by normalizing the preprocessed historical operating data. The monitoring and early warning module is used to monitor the operating data of production equipment in real time and to make judgments and issue early warnings based on the operating data.

7. A production equipment management system based on multi-source data analysis according to claim 6, characterized in that: The data acquisition and preprocessing module includes a data acquisition unit, a data classification unit, and a feature extraction unit; The data acquisition unit periodically collects voiceprint data of the production equipment using a voiceprint sensor and periodically collects infrared imaging data of the production equipment using an infrared imager; the voiceprint data and the infrared imaging data are collected synchronously. The data classification unit labels historical operation data records with preset fault type tags according to fault type; Historical data records carrying the same fault type label are grouped together; high-temperature areas are identified in each historical operation data record in the same group, and these high-temperature areas are designated as target areas; historical data records corresponding to the same target area in the same group are classified into one category. The feature extraction unit extracts vibration intensity data and temperature data of the target area from the same type of historical data records; sorts the vibration intensity data and the temperature data of the target area according to the order of collection time to generate a feature map; and obtains the target time period corresponding to each target area based on the feature map and calculates the target duration.

8. A production equipment management system based on multi-source data analysis according to claim 6, characterized in that: The threshold calculation module includes a temperature threshold calculation unit, a vibration frequency threshold calculation unit, and a vibration intensity threshold calculation unit; The temperature threshold calculation unit extracts temperature data of the target area collected within each target time period from the same type of historical operation data records; performs a weighted average of the temperature data of the target area collected within each target time period to obtain the average temperature; and sets the average temperature as the temperature threshold of the target area. Calculate the temperature growth rate of the target region within each target time period; calculate the temperature growth rate threshold by weighted averaging the temperature growth rates of the target region for each target time period. The vibration frequency threshold calculation unit extracts vibration frequency data of the target area collected within each target time period from the same type of historical operation data records; performs discrete Fourier transform on the extracted vibration frequency data to obtain a frequency distribution spectrum; selects the vibration frequency range with the highest energy proportion in the frequency distribution spectrum; takes the center value of the vibration frequency range, and sets the center value as the vibration frequency threshold of the target area. The vibration intensity threshold calculation unit divides each target time into several time windows with a preset window size; it fits the vibration intensity data included in each time window using the least squares method to obtain the slope; it calculates the slope change between two adjacent slopes; it integrates the slope change rates calculated within each target time period, and if the slope change rate within the target time period is greater than a preset slope change rate threshold, it selects the time window corresponding to the smallest slope change rate greater than the preset slope change rate threshold; it extracts the vibration intensity data within the time window and filters out the minimum vibration intensity in the vibration intensity data; it then performs a weighted average of the minimum vibration intensity corresponding to each target time period in the target region to obtain the vibration intensity threshold of the target region.

9. A production equipment management system based on multi-source data analysis according to claim 6, characterized in that: The fault assessment index calculation module includes a data normalization unit and a fault assessment index threshold calculation unit. The data normalization unit substitutes the temperature data, vibration frequency data, and vibration intensity data of a target area in the same fault type label into the normalization formula for normalization processing, and obtains the temperature normalized data set, vibration frequency normalized data set, and vibration intensity normalized data set. The fault assessment index threshold calculation unit calculates the average values ​​of the data in the temperature normalized data set, the vibration frequency normalized data set, and the vibration intensity normalized data set, respectively. Let the average values ​​be the temperature normalized value F1, the vibration frequency normalized value F2, and the vibration intensity normalized value F3, respectively. The temperature normalized value F1, the vibration frequency normalized value F2, and the vibration intensity normalized value F3 of a target area in a certain fault type label are weighted and averaged to obtain the fault assessment index threshold of the target area in the fault type label.

10. A production equipment management system based on multi-source data analysis according to claim 6, characterized in that: The early warning module includes a fault diagnosis unit and an early warning notification unit; The fault judgment unit monitors the temperature, temperature growth rate, vibration frequency, and vibration intensity of the production equipment in real time; and judges the real-time monitored operating data of the production equipment based on temperature threshold, temperature growth rate threshold, vibration frequency threshold, vibration intensity threshold, and fault assessment index threshold. The early warning unit issues a fault warning based on the judgment result.

Images