Industrial personal computer operation monitoring system and method based on data analysis

Abstract

The invention discloses an industrial personal computer operation monitoring system and method based on data analysis, and relates to the technical field of operation monitoring, and the system comprises a data acquisition and preprocessing module, a real-time monitoring and fault identification module and a performance optimization module. Analyzing in real time and displaying the state through a graphical interface; based on deep analysis of a preset rule algorithm, fault hidden dangers are automatically identified, warnings are pushed, and faults are rapidly positioned by means of a three-dimensional visual interface in combination with a factory layout; analyzing historical data to evaluate a performance trend, and providing optimization suggestions; multi-module cooperation realizes comprehensive monitoring, fault identification and performance optimization, improves the operation reliability and stability of the industrial personal computer, and ensures high efficiency and safety of industrial production.

Description

Technical Field

[0001] The present invention relates to the technical field of operation monitoring, and in particular to an industrial computer operation monitoring system and method based on data analysis. Background Art

[0002] With the rapid development of industrial information technology, industrial computers are widely used in industrial scenarios, and their operation monitoring is extremely important. Traditional monitoring relies on manual inspections and simple threshold judgments, which is inefficient and difficult to cope with complex working conditions, and cannot accurately analyze multi-source data. Modern industry has increasingly higher requirements for the reliability and real-time operation of industrial computers, and urgently needs intelligent monitoring methods. In order to meet the needs of real-time monitoring, fault prediction and optimized operation, monitoring systems and methods based on data analysis have emerged. By mining the value of data to achieve accurate diagnosis and prediction, and realize real-time monitoring of the operating status of industrial computers, it promotes the advancement of monitoring towards intelligence and precision, improves industrial production efficiency and stability, and enhances industry competitiveness.

[0003] Existing technologies include an industrial computer operation monitoring system based on data analysis disclosed in the invention application patent with announcement number CN116679668A. This invention is aimed at the industrial computer operation monitoring system, and a rapid diagnosis unit diagnoses faults and generates logs; an automatic maintenance unit performs secondary detection after troubleshooting; a working condition monitoring unit generates a reminder signal, and a supervision unit generates an adjustment signal based on this, and an execution control unit executes and analyzes the cause of the fault; a data retention unit saves important data based on the decision; the system realizes a closed loop of fault diagnosis, maintenance, monitoring and data management, thereby improving the operational reliability of the industrial computer.

[0004] Combined with the above solutions, it can be found that the current industrial computer operation monitoring system and method have limitations. In the existing technology, fault detection is delayed, the automatic monitoring capability is weak, and anomalies cannot be captured in real time; the processing of complex working condition data is insufficient, and it is difficult to accurately diagnose potential faults; data is difficult to communicate, forming "data islands", and integration and analysis are difficult; safety measures are weak, and the visual interface is simple, making it difficult to intuitively present the operating status, affecting monitoring efficiency and decision-making accuracy; restricting the effectiveness and reliability of industrial computer operation monitoring, and unable to meet the needs of modern industry for efficient and stable monitoring. Summary of the Invention

[0005] The purpose of the present invention is to provide an industrial computer operation monitoring system and method based on data analysis, which solves the problems existing in the background technology.

[0006] To solve the above technical problems, the present invention adopts the following technical solutions: In a first aspect, the present invention provides an industrial computer operation monitoring system based on data analysis, comprising: a data acquisition and preprocessing module, which collects the operation data of all industrial computers in the factory within a set period through various sensors and data acquisition equipment, and processes the data; The real-time monitoring and fault identification module analyzes and processes the collected operating data of all IPCs in the factory within a set period in real time to determine whether any IPCs in the factory have experienced any faults. If a fault occurs, the module identifies the fault category set of the IPC. Based on the fault category set, the module determines the fault level of the IPC and sends a warning message to the IPC operation and maintenance supervisor. The module also dynamically displays the IPC and its fault category set through a 3D visual monitoring interface. The performance optimization module obtains the operating data of all industrial computers in the factory in each historical period, identifies the fault category set of all industrial computers in each historical period, and classifies the inspection levels of automatic inspection robots according to all fault categories in the factory.

[0007] A second aspect of the present invention provides a method for executing the data analysis-based industrial computer operation monitoring system of the present invention, comprising: S1. data acquisition and preprocessing, collecting the operation data of all industrial computers in the factory within a set period through various sensors and data acquisition equipment, and processing the data; S2. Real-time monitoring and fault identification: This system analyzes and processes the collected operational data of all IPCs within a set period in real time to determine whether any IPCs in the factory have experienced any faults. If a fault occurs, the system identifies the fault category set for that IPC. Based on the fault category set, the system determines the fault level of the IPC and sends a warning message to IPC operation and maintenance supervisors. The system also dynamically displays the IPC and its fault category set through a 3D visual monitoring interface. S3. Performance optimization: Obtain the operating data of all industrial computers in the factory during each historical period, identify the set of fault categories of all industrial computers in each historical period, and classify the inspection levels of automatic inspection robots according to all fault categories in the factory.

[0008] The beneficial effects of the present invention are: 1. The present invention analyzes data in real time in the real-time monitoring and fault identification module, collects and analyzes data in real time, captures anomalies in a timely manner, improves the automatic monitoring capability, and displays the operating status in a graphical interface, which is convenient for operators to grasp the status of the industrial control computer in a timely and intuitive manner.

[0009] 2. The present invention automatically identifies potential faults in the real-time monitoring and fault identification module, combines it with three-dimensional visualization to dynamically display the status and quickly locate faults, thereby improving processing efficiency and reducing downtime losses.

[0010] 3. The present invention analyzes and divides inspection frequencies in the performance optimization module, provides optimization suggestions for operation and maintenance supervisors, helps shift from passive response to active predictive maintenance, improves the performance and stability of industrial computers, and extends equipment life. BRIEF DESCRIPTION OF THE DRAWINGS

[0011] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0012] Figure 1 This is a schematic diagram of the system structure connection of the present invention.

[0013] Figure 2 This is a connection diagram for executing the method of the present invention. DETAILED DESCRIPTION

[0014] The technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than all the embodiments; based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present invention.

[0015] Reference Figure 1 As shown, the present invention provides an industrial computer operation monitoring system based on data analysis, comprising: The data acquisition and preprocessing module collects and processes the operating data of all industrial computers in the factory within a set period through various sensors and data acquisition equipment.

[0016] In a specific embodiment of the present invention, the collecting of the operating data of the industrial computer within a set period and processing of the collected data specifically includes: Multi-source data collection is performed to obtain the operating data of the industrial computer in a time series based on sensor data and industrial computer performance indicators; the operating data includes hardware data and software data, wherein the hardware data includes a three-dimensional image of the housing, fan speed, temperature of the industrial computer's heat dissipation area, number of connection interruptions, signal transmission delay duration, measured equivalent series resistance, actual operating temperature of the capacitor, and capacitor usage time; the software data includes data transmission error rate, voltage fluctuation amplitude, memory usage rate, number of memory errors, and page swap rate; The collected data is processed by data cleaning, missing values ​​processing, denoising, time series feature extraction, format conversion, and data normalization to eliminate dimensional differences.

[0017] It should be noted that data cleaning includes removing noise and outliers. Missing values ​​can be processed by interpolation and mean filling. Denoising can be done by sliding window averaging and Kalman filtering. Time series features are extracted including mean, variance, trend, and periodicity. Format conversion includes unified timestamp and unit standardization. Data normalization is also performed to eliminate dimensional differences. Since the dimensions of the data in the calculation are different, the data needs to be normalized to convert several numerical values ​​into values ​​between 0 and 1.

[0018] The real-time monitoring and fault identification module analyzes and processes the collected operating data of all industrial computers in the factory within a set period in real time to determine whether any industrial computers in the factory have faults. If an industrial computer has a fault, the fault category set of the industrial computer is identified; the fault level of the industrial computer is determined based on the fault category set that has occurred, and a warning message is pushed to the industrial computer operation and maintenance supervisor. At the same time, the industrial computer and its fault category set are dynamically displayed through a three-dimensional visual monitoring interface.

[0019] In a specific embodiment of the present invention, the method of determining whether all industrial computers in the factory have faults and identifying the fault category set of an industrial computer if a fault occurs is as follows: B1. Shell deformation failure: Compare the collected 3D shell images with the original 3D shell images stored in the database to obtain the overlapping volume of the current 3D shell image and the original 3D shell image, and substitute it into the tilt deformation judgment model , is the number of the industrial computer, , is any integer greater than 2, is the number of each measurement, , is any integer greater than 2, where is the overlapping volume of the shell image measured by the nth industrial computer for the i-th time and the original three-dimensional image, Measure the volume of the original 3D image of the shell for the i-th time for the n-th industrial computer; If the overlapping volume ratio of the shell of an industrial computer is , then the shell of the device is deformed, K is the minimum volume ratio of the shell image measured by the industrial computer and the original three-dimensional image; B2. Heat dissipation failure, ,in is the fan operation judgment value, is the actual fan speed measured by the nth industrial computer for the i-th time, is the rated speed of the fan. When the fan fails, The administrator sets it according to the hardware parameters; ,in is the actual temperature of the heat dissipation area measured for the i-th time by the n-th industrial computer, The normal operating temperature of the industrial computer heat dissipation area is set. is the critical offset of temperature anomaly, when When the temperature of the heat dissipation area of ​​the industrial computer is abnormal; when , the cooling system of the industrial computer fails, is a logical AND symbol; B3. Cable failure, calculate the connector failure value ,in The monitoring data transmission error rate measured for the i-th time by the n-th industrial computer, is the allowed data transmission error rate, The number of connection interruptions measured for the i-th time for the n-th industrial computer, To allow for connection interruptions, is the signal transmission delay time, To allow for signal transmission delay, is the voltage fluctuation abnormal value and , is the maximum threshold of voltage fluctuation, They are the weight coefficients for monitoring data transmission error rate, number of connection interruptions, signal transmission delay time, and voltage fluctuation amplitude. , the average value of multiple measurements , When the connector is abnormal, is the set fault threshold; B4. Memory failure, calculate memory failure score ,in Score the memory usage of the nth industrial computer measured for the i-th time , is the actual memory usage rate measured for the i-th time on the n-th industrial computer, is the maximum threshold of memory usage, is the memory error score of the nth industrial computer measured for the i-th time and , is the number of memory errors, is the maximum threshold for the number of memory errors, is the page swap abnormality judgment value measured by the nth industrial computer for the i-th time and , is the actual page swap rate, is the maximum threshold of the page swap rate, are the weight coefficients of memory usage, number of memory errors, and page swap rate respectively. , When the memory fails, The memory failure threshold is set; B5. Capacitor failure, scoring the aging degree of the motherboard capacitor ,in Score the aging degree of the capacitor measured for the i-th time on the n-th industrial computer. is the measured equivalent series resistance, is the initial equivalent series resistance, index 1.5: reflects the acceleration effect of ESR nonlinear growth, is the actual operating temperature of the capacitor measured by the nth industrial computer for the i-th time, Capacitor rated temperature, is the temperature acceleration factor, The usage time of the capacitor measured for the i-th time on the n-th industrial computer, Design life of the capacitor, is the time aging weight coefficient, when When the motherboard capacitors are healthy and there is no aging fault When the mainboard capacitor is slightly aged; when When the motherboard capacitors are completely aged, Specifically, it indicates the maximum and minimum values ​​of the motherboard capacitor aging score.

[0020] It should be noted that in the formula for scoring the aging degree of motherboard capacitors, Indicates the nonlinear degradation of the capacitor. The aging growth trend of the capacitor's equivalent series resistance (ESR) is not linear, but rather shows an accelerated degradation trend. The exponent 1.5 is based on the data fitting of the accelerated aging test (105°C / 2000 hours), which found that the slope of the ESR growth curve is approximately equal to 1.5. Indicates temperature stress. Capacitors will age faster in high temperature environments. Indicates the cumulative working time of the capacitor. Aging accumulates at a uniform rate within the design life, and the risk increases linearly after exceeding the design life.

[0021] In a specific embodiment of the present invention, determining the fault level of the industrial computer according to the set of fault categories that have occurred specifically includes: ,in Score the fault level of the nth industrial computer, The weight for each fault type can be adjusted as needed, and , For each fault type, ,in To allow the shell image measured by the industrial computer to coincide with the original 3D image, , , , , the fault level is ; Recommendations based on industry experience .

[0022] In a specific embodiment of the present invention, the warning information is pushed to the IPC operation and maintenance supervisor, and the IPC and its fault category set are dynamically displayed through a three-dimensional visual monitoring interface, which specifically includes: Notifications are triggered through the rule engine Drools, which are divided into sound alarms, pop-up alarms, and SMS or email alarms. When an industrial computer fails, a specific sound is played to alert the operation and maintenance supervisor. At the same time, the monitoring image is used to detect whether the operation and maintenance supervisor is in the monitoring room. If so, an alarm window will pop up on the graphical interface to display the alarm information, including the alarm time, alarm type, and fault description to the supervisor. If not, the alarm information can be sent to the relevant personnel via SMS or email. All alarm information is recorded and stored in the database, and a query function for alarm records is provided. Operation and maintenance supervisors can query according to time range, fault level, and fault type.

[0023] The anomaly recognition module conducts in-depth analysis of operating data based on preset rules and algorithms, automatically identifies potential fault hazards, and pushes warning information through the system; through the three-dimensional visual monitoring interface, combined with the factory layout diagram, it dynamically displays the status of each industrial computer and quickly locates the fault location.

[0024] In a specific embodiment of the present invention, the automatic identification of potential fault hazards specifically includes: B1. Shell deformation degree: Compare the collected shell 3D image with the original shell 3D image stored in the database to obtain the overlapping volume of the current shell 3D image and the original shell 3D image, and substitute it into the tilt deformation judgment model ,in is the overlapping volume of the shell image of the p-th device and the original three-dimensional image, is the volume of the original three-dimensional image of the p-th device shell; if the overlapping volume ratio of a device shell is , the housing of the device is deformed.

[0025] B2. Industrial computer cooling system, , is the number of measurements, , is any integer greater than 2, where The actual fan speed is is the rated speed of the fan. When the fan fails, The administrator sets it according to the hardware parameters; ,in is the measured temperature of the heat dissipation area of ​​the industrial computer, The normal operating temperature of the industrial computer heat dissipation area is set. is the critical offset of temperature anomaly, when When the temperature of the heat dissipation area of ​​the industrial computer is abnormal; when , the cooling system of the industrial computer is abnormal, is the logical AND symbol.

[0026] B3. Connector, calculate the connector fault value ,in To monitor the data transmission error rate, is the normal maximum value of the data transmission error rate, is the normal value of data transmission error rate, is the number of connection interruptions, The maximum number of connection interruptions. is the signal transmission delay time, is the maximum value of the signal transmission delay time, is the normal value of signal transmission delay time, is the voltage fluctuation amplitude and , is the actual voltage fluctuation amplitude value, is the maximum value of voltage fluctuation amplitude, They are the weight coefficients for monitoring data transmission error rate, number of connection interruptions, signal transmission delay time, and voltage fluctuation amplitude. , the average value of multiple measurements , When the connector is abnormal, is the set fault threshold.

[0027] B4. Memory failure, calculate memory failure score ,in is the actual memory usage, This is the maximum value of normal memory usage. is the memory usage warning value, ,and , is the number of memory errors and , is the maximum number of memory errors, is the page swap rate and , is the maximum value of the page swap rate, are the weight coefficients of memory usage, number of memory errors, and page swap rate respectively. .

[0028] B5. Motherboard capacitor aging, score the degree of motherboard capacitor aging , is the number of measurements, , is any integer greater than 2, where is the measured equivalent series resistance, is the initial equivalent series resistance, index 1.5: reflects the acceleration effect of ESR nonlinear growth, is the actual operating temperature of the capacitor, Capacitor rated temperature, is the temperature acceleration factor, is the time the capacitor has been used, Design life of the capacitor, is the time aging weight coefficient, when When the motherboard capacitors are healthy, maintenance costs are low and no intervention is required; when When the detection cycle needs to be shortened, heat dissipation or load needs to be optimized; when When it is damaged, it must be replaced immediately to avoid damaging other components. Specifically, it indicates the maximum and minimum values ​​of the motherboard capacitor aging score.

[0029] In a specific embodiment of the present invention, the push warning information specifically includes: Notifications are triggered through the rule engine Drools and are divided into sound alarms, pop-up alarms, and SMS or email alarms. When an alarm is required to alert the operator, a specific sound is played to remind the user. At the same time, the monitoring image is used to monitor whether there are any staff in the monitoring room. If there are, an alarm window will pop up on the graphical interface to display the alarm information, including alarm time, alarm type, fault description, etc. to prompt the monitoring personnel. If not, the alarm information can be sent to relevant personnel via SMS or email.

[0030] All alarm information is recorded and stored in the database, and an alarm record query function is provided. Users can query according to time range, alarm level, alarm type and other conditions to quickly locate and analyze historical alarm information. At the same time, statistical analysis of alarm records can be performed to understand the fault distribution and trend of industrial computers, providing a reference for system maintenance and optimization.

[0031] It should be noted that for 3D scene construction, data collection can obtain the 3D structure of the factory through laser scanning or BIM modeling tools, assign a unique ID to each industrial computer, and associate its physical coordinates, production line, person in charge and other information. Multi-dimensional positioning assistance and plane map positioning provide a factory floor plan view, mark the equipment status with different colors, support quick search, and automatic positioning by entering the equipment number; equipment list positioning lists all equipment in a table, and the abnormal equipment row is marked red and flashing. Clicking the row record can jump to the corresponding position in the 3D scene; dividing the graphical interface into different areas, each area displays different types of information; adding interactive functions to the graphical interface to facilitate users to view and analyze data; for some key indicators, such as CPU usage and memory usage, it may be necessary to update once per second or every minute to ensure that users can understand the operating status of the industrial computer in a timely manner.

[0032] The performance optimization module obtains the operating data of all industrial computers in the factory in each historical period, identifies the set of fault categories of all industrial computers in each historical period, and classifies the inspection frequency of automatic inspection robots for all fault categories in the factory.

[0033] In a specific embodiment of the present invention, the step of obtaining the operating data of all industrial computers in the factory in each historical period and identifying the set of fault categories of all industrial computers in each historical period specifically includes: Identification number of all industrial computers in the factory, fault category, number of occurrences of each fault category, number of cycles, and total number of inspections in each cycle.

[0034] In a specific embodiment of the present invention, the occurrence probability of all fault types in the factory is calculated by the following specific steps: The fault category set of all industrial computers in the factory in each historical period is obtained , f is the fault category number, ,in is the occurrence frequency of fault f, is the number of times fault f occurs in the historical cycle, and z is the number of cycles.

[0035] In a specific embodiment of the present invention, the classification to obtain the inspection level of the automatic inspection robot for all fault categories in the factory specifically includes: According to the number and frequency of each fault, the inspection level coefficient of each fault is obtained ,like , then the inspection level of the fault is recorded as level j, is the fault ratio coefficient interval corresponding to the j-th inspection level.

[0036] It should be noted that the fault ratio coefficient intervals corresponding to the inspection levels are specifically determined based on the industrial computer inspection personnel. For example, in order to prevent high-frequency fault categories from causing significant losses to the factory, the fault ratio coefficient intervals corresponding to the inspection levels can be divided into: It is the first level of fault inspection. It is the second level of fault inspection. This is the third level of fault inspection.

[0037] It should also be noted that, among the first level, second level and third level of fault inspection, the importance of the first level of fault inspection is less than that of the second level of fault inspection, and the importance of the second level of fault inspection is less than that of the third level of fault inspection. That is to say, the number and frequency of failures of each industrial computer in the first level of fault inspection in the historical period are low, which can reduce the frequency of daily inspections and reduce the number of automatic inspection robots for inspection and maintenance. The number and frequency of failures of each industrial computer in the second level of fault inspection in the historical period are average, which can maintain the frequency of historical daily inspections and the number of automatic inspection robots. The number and frequency of failures of each industrial computer in the third level of fault inspection in the historical period are high, which can increase the frequency of daily inspections and increase the number of automatic inspection robots. This provides solid data support for the subsequent inspection frequency and cost investment of industrial computers in the factory, which is conducive to reducing the maintenance and management costs of industrial computers, while ensuring the rationality of the inspection frequency and cost investment of industrial computers.

[0038] In a specific embodiment of the present invention, the fault ratio coefficient interval is specifically: The greater the fault ratio coefficient, the higher the inspection level of the fault, the higher the inspection frequency, and the greater the number of automatic inspection robots for inspection and maintenance.

[0039] It should be noted that the present invention also includes a database for storing reference original data, including the original three-dimensional image of the industrial computer, the maximum threshold of voltage fluctuation, the cable fault threshold, the maximum threshold of memory usage, the maximum threshold of the number of memory errors, the maximum threshold of the page swap rate, the memory fault threshold, the maximum and minimum values ​​of the motherboard capacitor aging score, and the total number of factory inspections.

[0040] Reference Figure 2 As shown, the second aspect of the present invention provides a method for an industrial computer operation monitoring system based on data analysis, including: S1. Data acquisition and preprocessing, through various sensors and data acquisition equipment, collecting the operation data of all industrial computers in the factory within a set period and processing it.

[0041] S2. Real-time monitoring and fault identification: The collected operating data of all IPCs in the factory within a set period is analyzed and processed in real time to determine whether any IPCs in the factory have any faults. If an IPC has a fault, the fault category set of the IPC is identified. The fault level of the IPC is determined based on the fault category set that has occurred, and a warning message is pushed to the IPC operation and maintenance supervisor. At the same time, the IPC and its fault category set are dynamically displayed through a 3D visual monitoring interface.

[0042] S3. Performance optimization: Obtain the operating data of all industrial computers in the factory during each historical period, identify the set of fault categories of all industrial computers in each historical period, and classify the inspection levels of automatic inspection robots according to all fault categories in the factory.

[0043] The above contents are merely examples and explanations of the concept of the present invention. Those skilled in the art may make various modifications or additions to the described specific embodiments or replace them in a similar manner. As long as they do not deviate from the concept of the invention or exceed the scope defined by the present invention, they should all fall within the scope of protection of the present invention.

Claims

1. An industrial computer operation monitoring system based on data analysis, characterized in that: include: The data acquisition and preprocessing module collects and processes the operating data of all industrial computers in the factory within a set period through various sensors and data acquisition equipment; The real-time monitoring and fault identification module analyzes and processes the collected operating data of all industrial computers in the factory within a set period in real time to determine whether all industrial computers in the factory have faults. If an industrial computer has a fault, the fault category set of the industrial computer is identified, including shell deformation fault, heat dissipation fault, cable fault, memory fault, and capacitor fault; the fault level of the industrial computer is determined based on the fault category set, including ,in Score the fault level of the nth industrial computer, is the weight of each fault type, and , For each fault type, ,in To allow the shell image measured by the industrial computer to coincide with the original 3D image, , , , The fault level is, ; It also pushes warning information to the IPC operation and maintenance supervisor, and dynamically displays the IPC and its fault category set through a 3D visual monitoring interface; The performance optimization module obtains the operating data of all industrial computers in the factory in each historical period, identifies the fault category set of all industrial computers in each historical period, and classifies the inspection levels of automatic inspection robots according to all fault categories in the factory.

2. The industrial computer operation monitoring system based on data analysis according to claim 1 is characterized in that: The collecting of the operating data of the industrial computer within a set period and processing of the collected data specifically includes: Multi-source data collection is performed to obtain the operating data of the industrial computer in a time series based on sensor data and industrial computer performance indicators; the operating data includes hardware data and software data, wherein the hardware data includes a three-dimensional image of the housing, fan speed, temperature of the industrial computer's heat dissipation area, number of connection interruptions, signal transmission delay duration, measured equivalent series resistance, actual operating temperature of the capacitor, and capacitor usage time; the software data includes data transmission error rate, voltage fluctuation amplitude, memory usage rate, number of memory errors, and page swap rate; The collected data is processed by data cleaning, missing values ​​processing, denoising, time series feature extraction, format conversion, and data normalization to eliminate dimensional differences.

3. The industrial computer operation monitoring system based on data analysis according to claim 2 is characterized in that: The method for determining whether all industrial computers in the factory have faults and identifying the fault category set of an industrial computer if a fault occurs is as follows: B1. Shell deformation failure: Compare the collected 3D shell images with the original 3D shell images stored in the database to obtain the overlapping volume of the current 3D shell image and the original 3D shell image, and substitute it into the tilt deformation judgment model , is the number of the industrial computer, , is any integer greater than 2, is the number of each measurement, , is any integer greater than 2, where is the overlapping volume of the shell image measured by the nth industrial computer for the i-th time and the original three-dimensional image, Measure the volume of the original 3D image of the shell for the i-th time for the n-th industrial computer; If the overlapping volume ratio of the shell of an industrial computer is , then the shell of the device is deformed, K is the minimum volume ratio of the shell image measured by the industrial computer and the original three-dimensional image; B2. Heat dissipation failure, ,in is the fan operation judgment value, is the actual fan speed measured by the nth industrial computer for the i-th time, is the rated speed of the fan. When the fan fails, The administrator sets it according to the hardware parameters; ,in is the actual temperature of the heat dissipation area measured for the i-th time by the n-th industrial computer, The normal operating temperature of the industrial computer heat dissipation area is set. is the critical offset of temperature anomaly, when When the temperature of the heat dissipation area of ​​the industrial computer is abnormal; when , the cooling system of the industrial computer fails, is a logical AND symbol; B3. Cable failure, calculate the cable failure value ,in The monitoring data transmission error rate measured for the i-th time by the n-th industrial computer, is the allowed data transmission error rate, The number of connection interruptions measured for the i-th time for the n-th industrial computer, To allow for connection interruptions, is the signal transmission delay time, To allow for signal transmission delay, is the voltage fluctuation abnormal value and , is the actual voltage fluctuation amplitude value, is the maximum threshold of voltage fluctuation, They are the weight coefficients for monitoring data transmission error rate, number of connection interruptions, signal transmission delay time, and voltage fluctuation amplitude. , the average value of multiple measurements , When the cable fails, is the set cable fault threshold; B4. Memory failure, calculate memory failure score ,in Score the memory usage of the nth industrial computer measured for the i-th time , is the actual memory usage rate measured for the i-th time on the n-th industrial computer, is the maximum threshold of memory usage, is the memory error score of the nth industrial computer measured for the i-th time and , is the number of memory errors, is the maximum threshold for the number of memory errors, is the page swap abnormality judgment value measured by the nth industrial computer for the i-th time and , is the actual page swap rate, is the maximum threshold of the page swap rate, are the weight coefficients of memory usage, number of memory errors, and page swap rate respectively. , When the memory fails, The memory failure threshold is set; B5. Capacitor failure, scoring the aging degree of the motherboard capacitor ,in Score the aging degree of the capacitor measured for the i-th time on the n-th industrial computer. is the measured equivalent series resistance, is the initial equivalent series resistance, index 1.5: reflects the acceleration effect of ESR nonlinear growth, is the actual operating temperature of the capacitor measured by the nth industrial computer for the i-th time, Capacitor rated temperature, is the temperature acceleration factor, The usage time of the capacitor measured for the i-th time on the n-th industrial computer, Design life of the capacitor, is the time aging weight coefficient, when When the motherboard capacitors are healthy and there is no aging fault When the mainboard capacitor is slightly aged; when When the motherboard capacitors are completely aged, Specifically, it is the maximum and minimum values ​​of the motherboard capacitor aging score; According to the above method, the fault categories of all industrial computers are determined, and a set of fault categories of all industrial computers is constructed.

4. The industrial computer operation monitoring system based on data analysis according to claim 1 is characterized in that: The warning information is pushed to the IPC operation and maintenance supervisor, and the IPC and its fault category set are dynamically displayed through a three-dimensional visual monitoring interface, which specifically includes: Notifications are triggered through the rule engine Drools, which are divided into sound alarms, pop-up alarms, and SMS or email alarms. When an industrial computer fails, a specific sound is played to alert the operation and maintenance supervisor. At the same time, the monitoring image is used to detect whether the operation and maintenance supervisor is in the monitoring room. If so, an alarm window will pop up on the graphical interface to display the alarm information, including the alarm time, alarm type, and fault description to the supervisor. If not, the alarm information can be sent to the relevant personnel via SMS or email. All alarm information is recorded and stored in the database, and a query function for alarm records is provided. Operation and maintenance supervisors can query according to time range, fault level, and fault type.

5. The industrial computer operation monitoring system based on data analysis according to claim 1 is characterized in that: The step of obtaining the operating data of all industrial computers in the factory in each historical period and identifying the set of fault categories of all industrial computers in each historical period specifically includes: Identification number of all industrial computers in the factory, fault category, number of occurrences of each fault category, number of cycles, and total number of inspections in each cycle.

6. The industrial computer operation monitoring system based on data analysis according to claim 5, characterized in that: The probability of occurrence of all fault types in the factory is calculated by the following steps: The fault category set of all industrial computers in the factory in each historical period is obtained , f is the fault category number, ,in is the occurrence frequency of fault f, is the number of times fault f occurs in the historical cycle, and z is the number of cycles.

7. The industrial computer operation monitoring system based on data analysis according to claim 6, characterized in that: The classification obtains the inspection level of the automatic inspection robot for all fault categories in the factory, which specifically includes: According to the number and frequency of each fault, the inspection level coefficient of each fault is obtained ,like , then the inspection level of the fault is recorded as level j, is the fault ratio coefficient interval corresponding to the j-th inspection level.

8. The industrial computer operation monitoring system and method based on data analysis according to claim 7 is characterized in that: The fault ratio coefficient interval is specifically: The greater the fault ratio coefficient, the higher the inspection level of the fault, the higher the inspection frequency, and the greater the number of automatic inspection robots for inspection and maintenance.

9. A method for implementing the data analysis-based industrial computer operation monitoring system according to any one of claims 1 to 8, comprising: S1. Data collection and preprocessing: Using various sensors and data acquisition equipment, we collect and process the operating data of all industrial computers in the factory within a set period. S2. Real-time monitoring and fault identification module, which analyzes and processes the collected operating data of all industrial computers in the factory within a set period in real time, determines whether all industrial computers in the factory have faults, and if an industrial computer has a fault, identifies the fault category set of the industrial computer, including shell deformation fault, heat dissipation fault, cable fault, memory fault, and capacitor fault; determines the fault level of the industrial computer based on the fault category set, including ,in Score the fault level of the nth industrial computer, is the weight of each fault type, and , For each fault type, ,in To allow the shell image measured by the industrial computer to coincide with the original 3D image, , , , , the fault level is ; and push warning information to the IPC operation and maintenance supervisor, while dynamically displaying the IPC and its fault category set through a 3D visual monitoring interface; S3. Performance optimization: Obtain the operating data of all industrial computers in the factory during each historical period, identify the set of fault categories of all industrial computers in each historical period, and classify the inspection levels of automatic inspection robots according to all fault categories in the factory.

Images