Industrial data collection and management method based on Internet of Things

Abstract

The present invention discloses an industrial data collection and management method based on the Internet of Things, which relates to the field of data management technology and includes the following steps: performing demand analysis on industrial environments and production processes, and clarifying industrial equipment indicators to be monitored in the industrial environments and production processes, which are equipment performance indicators and production performance indicators, respectively. The industrial data collection and management method based on the Internet of Things significantly improves the reliability of equipment and reduces unexpected downtime by real-time monitoring of equipment performance indicators and production performance indicators, and can predict possible equipment failures and performance degradations through comprehensive analysis of equipment status evaluation indexes, fault analysis evaluation indexes, and production performance evaluation indexes, so as to take preventive maintenance measures in advance, which can not only reduce production interruptions caused by equipment failures, but also improve the efficiency and effectiveness of maintenance work by optimizing maintenance plans.

Description

Technical Field

[0001] The present invention relates to the technical field of data management, and in particular to an industrial data collection and management method based on the Internet of Things. Background Art

[0002] With the rapid development of science and technology, it is emphasized that the production process should be made intelligent, networked and service-oriented through the Internet of Things, big data, cloud computing, artificial intelligence and other advanced technologies, so as to improve production efficiency, reduce costs, and enhance product quality and customer satisfaction. By collecting and analyzing data, enterprises can more accurately understand the utilization of resources such as raw materials, equipment, and personnel, thereby optimizing resource allocation and reducing production costs. The rapid development of Internet of Things technology has provided strong support for industrial data collection.

[0003] In the existing technology, the requirements of industrial environment and production process are constantly changing, and as the amount of collected industrial data increases, the analysis of related equipment and production environment in the industrial environment will become more difficult, which in turn affects the judgment of the maintenance cycle of industrial equipment. Therefore, how to improve the management efficiency of industrial data and the evaluation accuracy of the maintenance cycle of industrial equipment and identify potential problems in the production process is the problem we need to solve. To this end, an industrial data collection and management method based on the Internet of Things is proposed. Summary of the invention

[0004] To achieve the above objectives, the present invention is implemented through the following technical solutions: an industrial data collection and management method based on the Internet of Things, comprising the following steps:

[0005] Step 1: Conduct demand analysis on the industrial environment and production process, identify the industrial equipment indicators to be monitored in the industrial environment and production process, namely equipment performance indicators and production performance indicators, and plan data collection points to deploy IoT devices for data collection;

[0006] Step 2: pre-process the collected data, perform data analysis on the equipment performance indicators, and extract sub-indicators of the equipment performance indicators, wherein the sub-indicators of the equipment performance indicators include equipment status monitoring indicators and fault analysis indicators;

[0007] Step 3: Extract features based on the sub-indicators of the equipment performance indicators and the production performance indicators, further analyze the associated data of the sub-indicator features of the equipment performance indicators and the production performance indicator features, and obtain the equipment performance feature set and the production performance feature set;

[0008] Step 4: Use machine learning technology combined with equipment performance feature sets to evaluate and analyze the sub-indicators of equipment performance indicators respectively, and use deep learning technology combined with production performance feature sets to evaluate and analyze production performance indicators;

[0009] Step 5: Based on the analysis results of equipment performance indicators and production performance indicators, a predictive maintenance model is constructed in combination with relevant characteristic data to predict the maintenance trend of industrial equipment, and the maintenance demand level of industrial equipment is preset to perform preventive maintenance.

[0010] Preferably, in step 1, the process of deploying IoT devices to collect data includes:

[0011] Conduct demand analysis on industrial environment and production process, analyze the performance indicators of industrial equipment that need to be monitored in industrial environment and production process, which are equipment performance indicators and production performance indicators. Among them, equipment performance indicators include temperature, vibration, operation time, fault frequency, fault type, and repair time. Production performance indicators include production speed, raw material consumption and energy efficiency. Temperature is used to monitor the temperature change of equipment and prevent equipment damage caused by overheating. Vibration is used to monitor the vibration of equipment and evaluate the operating status and potential failures of equipment. Operation time is used to record the operation time of equipment and analyze the frequency and life of equipment. Failure frequency is used to count the number of equipment failures and evaluate the reliability and stability of equipment. Failure type is used to record the specific type and cause of failure and provide a basis for fault handling. Repair time is used to record the time from fault discovery to repair and evaluate the efficiency of fault handling. Production speed is used to monitor the output speed of products on the production line and evaluate production efficiency. Raw material consumption is used to count the amount of raw materials used in the production process and evaluate production costs and resource utilization efficiency. Energy efficiency is used to monitor energy consumption in the production process and evaluate the energy efficiency level of the production line.

[0012] Clarify the monitoring purpose of each indicator, plan data collection points on relevant industrial equipment according to the determined industrial equipment performance indicators, and deploy IoT devices at the corresponding data collection points, such as temperature sensors, vibration sensors, pressure sensors, energy consumption monitors, etc.;

[0013] Deploy a data transmission network, connect IoT devices at each data collection point, and configure a data center to store, process and analyze data from IoT devices. Integrate IoT devices with the data transmission network and data center, and transmit the collected data to the data center via wireless communication devices for centralized data management and analysis.

[0014] Preferably, in step 2, the process of performing data analysis on the equipment performance indicators includes:

[0015] Preprocess the collected equipment performance indicators and production performance indicators, including data cleaning and data conversion, and aggregate the collected data at time intervals to reduce the amount of data and extract trend information. Check and remove abnormal data points caused by sensor failure, data transmission errors, etc. through data cleaning. For missing data, use interpolation methods (such as linear interpolation, mean interpolation, etc.) or deletion methods to process according to the characteristics and missing conditions of the data. For continuously changing data such as vibration and temperature, use smoothing algorithms (such as moving average, exponential smoothing, etc.) to remove noise and improve data quality. Perform data conversion to ensure that all data units are consistent to facilitate subsequent data analysis and comparison. Standardize or normalize the data as needed to eliminate the impact of different dimensions on data analysis results.

[0016] Perform data analysis on the pre-processed equipment performance indicators and further determine the sub-indicators of the equipment performance indicators, which are equipment status monitoring indicators and fault analysis indicators;

[0017] Based on the relevant data of equipment status monitoring indicators, analyze the temperature, vibration and operating time data of industrial equipment, monitor the operating status of industrial equipment under different working conditions, and analyze the fault frequency, fault type and repair time data of industrial equipment based on the relevant data of fault analysis indicators. Monitor the fault change trend of industrial equipment and mark the normal data and abnormal data in the equipment status monitoring indicators and fault analysis indicators. Among them, analyze the change trend of equipment temperature, identify temperature abnormalities, evaluate whether the equipment has overheating risks, identify characteristic frequencies and vibration modes in equipment vibration, evaluate the operating status and potential faults of the equipment, count the operating time of the equipment, analyze the frequency of use and life distribution of the equipment, provide a basis for equipment maintenance and replacement, count the number of equipment failures, analyze the change trend of fault frequency, identify the high-incidence period and vulnerable parts of the equipment, classify and count the faults, analyze the causes and influencing factors of various types of faults, provide a basis for fault prevention and handling, analyze the time from fault discovery to repair, evaluate the efficiency and response speed of fault handling, and propose improvement measures to shorten the repair time.

[0018] Preferably, in step 3, the process of acquiring the equipment performance feature set and the production performance feature set includes:

[0019] Extract features from the sub-indicators of equipment performance indicators. For equipment status monitoring indicators, extract temperature features, vibration signal features, and operating time features, extract the maximum and minimum values ​​of temperature data, calculate the temperature change rate to identify rapid temperature changes and abnormal temperature patterns, extract the characteristic frequency, amplitude, and phase of the vibration signal through spectrum analysis to identify the relationship between the vibration pattern and the equipment status, and count the total operating time and periodic operating time of industrial equipment, analyze the relationship between operating time and equipment maintenance cycle, and identify the correlation between operating time and fault occurrence. For fault analysis indicators, extract fault frequency features, fault type features, and fault repair features, count the fault frequency and the time distribution of fault occurrence, classify the fault types, and extract features of each type to analyze the distribution and influencing factors of fault repair time.

[0020] Extract features from production performance indicators, obtain production speed features, raw material consumption features, and energy efficiency features, calculate the average value, fluctuation range, and stability of production speed, analyze the changing trend and periodic pattern of production speed, calculate the raw material consumption rate and raw material consumption per unit product, analyze the relationship between raw material consumption and production speed and product quality, extract abnormal patterns of raw material consumption, and calculate the energy consumption per unit product and the overall energy efficiency ratio, so as to analyze the relationship between energy efficiency and production speed and equipment status, and identify energy efficiency anomalies;

[0021] The extracted equipment status monitoring indicators and fault analysis indicator features are integrated into an equipment performance feature set, and the extracted production speed, raw material consumption and energy efficiency features are integrated into a production performance feature set.

[0022] Preferably, in step 4, the process of evaluating and analyzing the equipment performance index sub-indicators and the production performance index includes:

[0023] Extract features associated with equipment status monitoring indicators from the equipment performance feature set, build an equipment monitoring model in combination with a neural network model, conduct a comprehensive analysis of the relevant data of temperature features, vibration signal features, and operating time features, and use the trained equipment monitoring model to evaluate the equipment status, obtain the equipment status evaluation index, and analyze the health status trend of the equipment;

[0024] Extract features associated with fault analysis indicators from the equipment performance feature set, train a fault classification model based on a decision tree, conduct a comprehensive analysis of relevant data on fault frequency features, fault type features, and fault repair features, and use the trained model to analyze fault data to obtain a fault analysis evaluation index to assess the equipment's fault risk.

[0025] Based on the characteristics of the production performance characteristic set that are associated with the production performance indicators, the production performance evaluation model is trained in combination with a convolutional neural network. A comprehensive analysis is conducted on the relevant data of production speed characteristics, raw material consumption characteristics, and energy efficiency characteristics. The trained production performance evaluation model is used to evaluate the production performance and obtain a production performance evaluation index to evaluate the efficiency and production status of the production process and identify production bottlenecks and optimization opportunities.

[0026] Preferably, the expression of the equipment status evaluation index is:

[0027] ;

[0028] in, is the equipment condition assessment index, For the The temperature reading at each time point, For the The characteristic value of the vibration signal at a time point, is the total operating time of the device, Indicates the maximum operating time of the device, which is the preset benchmark value. is the calibration parameter used to adjust the scale of the equipment condition assessment index, is the lowest temperature reading during the evaluation period, is the baseline temperature, which represents the typical temperature of the device during normal operation. and It is the adjustment parameter in the exponential function, which is used to control the influence of operating time and temperature on the equipment status evaluation index value. is the total number of observation time points, The larger the value, the better the device status;

[0029] The expression of the fault analysis evaluation index is:

[0030] ;

[0031] in, is the failure analysis evaluation index, For the The fault frequency of the fault type, For the The failure cost of this type of failure is the direct maintenance cost. For the The mean repair time for each type of failure is is the total number of fault types, is the weight factor used to adjust the failure cost and repair time The impact of is the maximum fault frequency among all fault types, is the reference value of the fault frequency, which is the preset fault frequency threshold during normal operation. is an index parameter used to measure the impact of the maximum fault frequency on the fault analysis evaluation index. The larger the value, the lower the impact of the fault.

[0032] Preferably, the expression of the production performance evaluation index is:

[0033] ;

[0034] in, is the production performance evaluation index, For the The production speed at a point in time, For the The raw material consumption per unit product at a time point, For the Energy consumption per unit product at a time point, is the total number of observation time points, is the benchmark energy consumption, the energy consumption per unit product under optimal production conditions, For the lowest energy consumption, is an index parameter used to adjust the energy efficiency The impact of value, The value range is from 0 to 1, where 1 represents the best state (i.e., fastest production speed, least raw material consumption, and highest energy efficiency), and 0 represents the worst state.

[0035] Preferably, in step 5, the process of predicting the maintenance trend of industrial equipment includes:

[0036] Traverse the relevant characteristic data of equipment performance indicators and production performance indicators, and integrate the equipment status evaluation index, fault analysis evaluation index and production performance evaluation index to obtain a comprehensive data set, ensure the quality and consistency of the data, and handle any missing or abnormal values;

[0037] Extract historical data from the comprehensive data set, divide the historical data into a training set and a test set, use the training set data to train the predictive maintenance model in combination with the time series prediction model, and use the test set data to test the constructed predictive maintenance model, where the historical data includes equipment performance data, production performance data, and known maintenance records;

[0038] The output of the predictive maintenance model is used to combine the equipment status evaluation index, the fault analysis evaluation index and the production performance evaluation index to obtain the predictive maintenance coefficient, comprehensively analyze the overall health status and production efficiency of industrial equipment, and determine the urgency of the maintenance required for industrial equipment in the future.

[0039] Combined with historical data and equipment maintenance requirements, different maintenance requirement levels for industrial equipment are preset, namely high maintenance requirement level, medium maintenance requirement level and low maintenance requirement level, and the corresponding evaluation threshold is matched to each maintenance requirement level with the predicted maintenance coefficient;

[0040] Develop preventive maintenance plans based on the maintenance demand level, including maintenance schedules, resource allocation, and expected results, and deploy predictive maintenance models to analyze the maintenance needs of industrial equipment, determine the current maintenance demand level of industrial equipment, and implement corresponding preventive maintenance plans.

[0041] Preferably, the expression of the predicted maintenance coefficient is:

[0042] ;

[0043] in, is the predicted maintenance coefficient, It is the equipment status evaluation index, which reflects the health status of the equipment. It is a failure analysis evaluation index that reflects the failure risk of the equipment. It is a production performance evaluation index that reflects the efficiency of the production process. , , is the weight coefficient, used to balance different index pairs Contribution To adjust the parameters, used to control the growth rate of the exponential function, It is the benchmark value of the production performance evaluation index, indicating the target production efficiency. The value range is from 0 to 1, where 1 represents the best state (i.e., healthy equipment, no risk of failure, and high production efficiency), and 0 represents the worst state.

[0044] Preferably, the plurality of maintenance requirement levels correspond to the plurality of evaluation thresholds, the evaluation thresholds include an upper threshold and a lower threshold, and the correspondence between the plurality of maintenance requirement levels and the plurality of evaluation thresholds is as follows:

[0045] High maintenance level ; Indicates that the equipment may have a greater maintenance demand and needs to be given priority for maintenance;

[0046] Medium maintenance requirement level ; Indicates that the equipment has certain maintenance requirements, but can be arranged in the short term;

[0047] Low maintenance level ; It indicates that the equipment is in good condition, the maintenance requirement is low, and the maintenance cycle can be extended;

[0048] in, is the predicted maintenance coefficient, is the lower threshold corresponding to the high maintenance demand level and the upper threshold corresponding to the medium maintenance demand level, is the lower threshold corresponding to the medium maintenance requirement level and the upper threshold corresponding to the low maintenance requirement level, , .

[0049] The present invention provides an industrial data collection and management method based on the Internet of Things. It has the following beneficial effects:

[0050] 1. The industrial data collection and management method based on the Internet of Things significantly improves the reliability of equipment and reduces unexpected downtime by real-time monitoring of equipment performance indicators and production performance indicators. It can also predict possible equipment failures and performance degradations through comprehensive analysis of the equipment status evaluation index, the fault analysis evaluation index, and the production performance evaluation index, so as to take preventive maintenance measures in advance. It can not only reduce production interruptions caused by equipment failures, but also improve the efficiency and effectiveness of maintenance work by optimizing maintenance plans. In addition, by analyzing the frequency and type of failures, the patterns and trends of failures can be identified, so as to take targeted improvement measures to reduce the probability of future failures.

[0051] 2. This industrial data collection and management method based on the Internet of Things realizes real-time monitoring of various indicators on the production site by deploying a large number of sensors and equipment. It can continuously collect equipment performance data and production performance indicators, ensuring the real-time and accuracy of the data. Compared with traditional manual inspections or intermittent data collection methods, the Internet of Things technology greatly reduces data delays and errors, and provides production managers with more accurate and comprehensive data support. Real-time and accurate data not only helps to promptly discover abnormal situations in the production process, but also provides a strong basis for production optimization and decision-making, thereby improving overall production efficiency and product quality. BRIEF DESCRIPTION OF THE DRAWINGS

[0052] Figure 1 This is an external method flow chart of an industrial data collection and management method based on the Internet of Things of the present invention;

[0053] Figure 2 A flow chart of the present invention for evaluating and analyzing equipment performance index sub-indicators and production performance indicators;

[0054] Figure 3 A flow chart for predicting the maintenance trend of industrial equipment according to the present invention. DETAILED DESCRIPTION

[0055] The present invention is further described in detail below in conjunction with the accompanying drawings and specific embodiments. The embodiments of the present invention are provided for the purpose of illustration and description, and are not intended to be exhaustive or to limit the present invention to the disclosed forms. Many modifications and variations will be apparent to those of ordinary skill in the art. The embodiments are selected and described in order to better illustrate the principles and practical applications of the present invention, and to enable those of ordinary skill in the art to understand the present invention and thereby design various embodiments with various modifications suitable for specific uses.

[0056] The first embodiment, as Figure 1 , Figure 2 As shown, the present invention provides a technical solution: an industrial data collection and management method based on the Internet of Things, comprising the following steps:

[0057] Step 1: Conduct a demand analysis on the industrial environment and production process, clarify the industrial equipment indicators to be monitored in the industrial environment and production process, which are equipment performance indicators and production performance indicators, and plan data collection points to deploy IoT devices for data collection. Conduct a demand analysis on the industrial environment and production process, and analyze the performance indicators of industrial equipment to be monitored in the industrial environment and production process, which are equipment performance indicators and production performance indicators. Among them, equipment performance indicators include temperature, vibration, operating time, failure frequency, failure type, and repair time. Production performance indicators include production speed, raw material consumption, and energy efficiency. Temperature is used to monitor the temperature changes of the equipment and prevent equipment damage caused by overheating. Vibration is used to monitor the vibration of the equipment and evaluate the operating status and potential failures of the equipment. Operating time is used to record the operating time of the equipment and is used to analyze the frequency of use and life of the equipment. Failure frequency is used to count the number of equipment failures and evaluate the reliability and stability of the equipment. Failure type is used to record the specific type of failure. The type and cause of the fault provide a basis for fault handling. The repair time is used to record the time from the discovery to the repair of the fault and evaluate the efficiency of fault handling. The production speed is used to monitor the output speed of products on the production line and evaluate the production efficiency. The raw material consumption is used to count the amount of raw materials used in the production process and evaluate the production cost and resource utilization efficiency. The energy efficiency is used to monitor the energy consumption in the production process and evaluate the energy efficiency level of the production line. The monitoring purpose of each indicator is clarified. According to the determined industrial equipment performance indicators, data collection points are planned on the relevant industrial equipment, and IoT devices such as temperature sensors, vibration sensors, pressure sensors, energy consumption monitors, etc. are deployed at the corresponding data collection points. A data transmission network is laid out to connect the IoT devices at each data collection point, and a data center is configured to store, process and analyze data from IoT devices. The IoT devices are integrated with the data transmission network and the data center, and the collected data is transmitted to the data center through wireless communication devices for centralized data management and analysis.

[0058] Step 2: Preprocess the collected data, analyze the equipment performance indicators, and extract sub-indicators of the equipment performance indicators. The sub-indicators of the equipment performance indicators include equipment status monitoring indicators and fault analysis indicators. Preprocess the collected equipment performance indicators and production performance indicators, including data cleaning and data conversion, and aggregate the collected data at time intervals to reduce the amount of data and extract trend information. Check and remove abnormal data points caused by sensor failure, data transmission errors, etc. through data cleaning. For missing data, interpolation methods (such as linear interpolation, mean interpolation, etc.) or deletion methods are used for processing according to the characteristics and missing conditions of the data. For continuously changing data such as vibration and temperature, smoothing algorithms (such as moving average, exponential smoothing, etc.) are used to remove noise and improve data quality. Data conversion is performed on the data to ensure that the units of all data are consistent to facilitate subsequent data analysis and comparison. The data is standardized or normalized as needed to eliminate the influence of different dimensions on the data analysis results. Data analysis is performed on the preprocessed equipment performance indicators to further determine the sub-indicators of the equipment performance indicators, which are respectively Equipment status monitoring indicators and fault analysis indicators: Based on the relevant data of equipment status monitoring indicators, analyze the temperature, vibration and operating time data of industrial equipment, monitor the operating status of industrial equipment under different working conditions, analyze the fault frequency, fault type and repair time data of industrial equipment based on the relevant data of fault analysis indicators, monitor the fault change trend of industrial equipment, and mark the normal data and abnormal data in the equipment status monitoring indicators and fault analysis indicators. Among them, analyze the change trend of equipment temperature, identify temperature abnormalities, evaluate whether the equipment has overheating risks, identify characteristic frequencies and vibration modes in equipment vibration, evaluate the operating status and potential faults of the equipment, count the operating time of the equipment, analyze the frequency of use and life distribution of the equipment, provide a basis for equipment maintenance and replacement, count the number of equipment failures, analyze the change trend of fault frequency, identify the high-incidence period of failure and vulnerable parts of the equipment, classify and count the failures, analyze the causes and influencing factors of various types of failures, provide a basis for fault prevention and handling, analyze the time from fault discovery to repair, evaluate the efficiency and response speed of fault handling, and propose improvement measures to shorten the repair time;

[0059] Step 3: Extract features based on the sub-indicators of the equipment performance indicators and the production performance indicators, further analyze the associated data of the sub-indicator features of the equipment performance indicators and the production performance indicator features, obtain the equipment performance feature set and the production performance feature set, extract features from the sub-indicators of the equipment performance indicators, extract temperature features, vibration signal features and operating time features for the equipment status monitoring indicators, extract the maximum and minimum values ​​of the temperature data, calculate the temperature change rate to identify rapid temperature changes and abnormal temperature patterns, extract the characteristic frequency, amplitude and phase of the vibration signal through spectrum analysis to identify the relationship between the vibration pattern and the equipment status, and count the total operating time and periodic operating time of the industrial equipment, analyze the relationship between the operating time and the equipment maintenance cycle, identify the correlation between the operating time and the occurrence of faults, and extract fault frequency features, fault type features and faults for the fault analysis indicators. Repair features, count fault frequencies and the time distribution of fault occurrence, classify fault types, extract features of each type to analyze the distribution and influencing factors of fault repair time, extract features of production performance indicators, obtain production speed features, raw material consumption features and energy efficiency features, calculate the average value, fluctuation range and stability of production speed, analyze the changing trend and periodic pattern of production speed, count the raw material consumption rate and raw material consumption per unit product, analyze the relationship between raw material consumption and production speed and product quality, extract abnormal patterns of raw material consumption, and calculate the energy consumption per unit product and the overall energy efficiency ratio to analyze the relationship between energy efficiency and production speed and equipment status, identify energy efficiency anomalies, integrate the extracted equipment status monitoring indicators and fault analysis indicator features into an equipment performance feature set, and integrate the extracted production speed, raw material consumption and energy efficiency features into a production performance feature set;

[0060] Step 4: Use machine learning technology in combination with equipment performance feature sets to evaluate and analyze the sub-indicators of equipment performance indicators respectively, and use deep learning technology in combination with production performance feature sets to evaluate and analyze production performance indicators. Extract features associated with equipment status monitoring indicators from the equipment performance feature set, build an equipment monitoring model in combination with a neural network model, conduct a comprehensive analysis of relevant data on temperature features, vibration signal features, and operating time features, and use the trained equipment monitoring model to evaluate the equipment status to obtain an equipment status evaluation index, analyze the health status trend of the equipment, extract features associated with fault analysis indicators from the equipment performance feature set, train a fault classification model in combination with a decision tree, conduct a comprehensive analysis of relevant data on fault frequency features, fault type features, and fault repair features, and use the trained model to analyze the fault data to obtain a fault analysis evaluation index, evaluate the failure risk of the equipment, and train a production performance evaluation model based on features associated with production performance indicators in the production performance feature set in combination with a convolutional neural network, conduct a comprehensive analysis of relevant data on production speed features, raw material consumption features, and energy efficiency features, and use the trained production performance evaluation model to evaluate production performance to obtain a production performance evaluation index, so as to evaluate the efficiency and production status of the production process and identify production bottlenecks and optimization opportunities;

[0061] Furthermore, the expression of the equipment status evaluation index is:

[0062] ;

[0063] in, is the equipment condition assessment index, For the The temperature reading at each time point, For the The characteristic value of the vibration signal at a time point, is the total operating time of the device, Indicates the maximum operating time of the device, which is the preset benchmark value. is the calibration parameter used to adjust the scale of the equipment condition assessment index, is the lowest temperature reading during the evaluation period, is the baseline temperature, which represents the typical temperature of the device during normal operation. and It is the adjustment parameter in the exponential function, which is used to control the influence of operating time and temperature on the equipment status evaluation index value. is the total number of observation time points, The larger the value, the better the device status. and When the value of increases, if it exceeds the normal operation threshold, will decrease, indicating that the device status is deteriorating. near hour, will decrease, because the longer the running time, the greater the possibility of equipment wear. near hour, The value of will decrease, resulting in reduce;

[0064] The expression of the failure analysis evaluation index is:

[0065] ;

[0066] in, is the failure analysis evaluation index, For the The fault frequency of the fault type, For the The failure cost of this type of failure is the direct maintenance cost. For the The mean repair time for each type of failure is is the total number of fault types, is the weight factor used to adjust the failure cost and repair time The impact of is the maximum fault frequency among all fault types, is the reference value of the fault frequency, which is the preset fault frequency threshold during normal operation. is an index parameter used to measure the impact of the maximum fault frequency on the fault analysis evaluation index. The larger the value of is, the lower the impact of the fault is. , or When the value of increases, will decrease, indicating an increased risk of failure. near hour, The value of will decrease, resulting in Decrease, indicating an increase in the frequency of faults;

[0067] Furthermore, the expression of the production performance evaluation index is:

[0068] ;

[0069] in, is the production performance evaluation index, For the The production speed at a point in time, For the The raw material consumption per unit product at a time point, For the Energy consumption per unit product at a time point, is the total number of observation time points, is the benchmark energy consumption, the energy consumption per unit product under optimal production conditions, For the lowest energy consumption, is an index parameter used to adjust the energy efficiency The impact of value, The value range is from 0 to 1, where 1 represents the best state (i.e., the fastest production speed, the least raw material consumption, and the highest energy efficiency), and 0 represents the worst state. When the average value of the production speed increases, Increase, when the fluctuation range of raw material consumption decreases, Increase, when the unit product energy consumption is close to the benchmark energy consumption, Increase;

[0070] Step 5: Based on the analysis results of equipment performance indicators and production performance indicators, a predictive maintenance model is constructed in combination with relevant characteristic data to predict the maintenance trend of industrial equipment, and the maintenance demand level of industrial equipment is preset to perform preventive maintenance.

[0071] The second embodiment is based on the first embodiment. Figure 3 As shown, in step 5, the process of predicting the maintenance trend of industrial equipment includes:

[0072] Traverse the relevant characteristic data of equipment performance indicators and production performance indicators, and integrate the equipment status evaluation index, fault analysis evaluation index and production performance evaluation index to obtain a comprehensive data set, ensure the quality and consistency of the data, handle any missing or outliers, extract historical data from the comprehensive data set, divide the historical data into training set and test set, use the training set data to train the predictive maintenance model in combination with the time series prediction model, and use the test set data to test the constructed predictive maintenance model, where the historical data includes equipment performance data, production performance data and known maintenance records, use the output of the predictive maintenance model, combine the equipment status evaluation index, fault analysis evaluation index and production performance The evaluation index is used to obtain the predictive maintenance coefficient, which comprehensively analyzes the overall health status and production efficiency of industrial equipment, determines the urgency of the maintenance required for industrial equipment in the future, and presets different maintenance demand levels for industrial equipment, namely high maintenance demand level, medium maintenance demand level and low maintenance demand level, based on historical data and equipment maintenance demand. The corresponding evaluation threshold is matched to each maintenance demand level with the predictive maintenance coefficient. A preventive maintenance plan is formulated according to the maintenance demand level, including the maintenance schedule, resource allocation and expected effect. The predictive maintenance model is deployed to analyze the maintenance demand of industrial equipment, determine the current maintenance demand level of industrial equipment, and implement the corresponding preventive maintenance plan.

[0073] Furthermore, the expression of the predicted maintenance coefficient is:

[0074] ;

[0075] in, is the predicted maintenance coefficient, It is the equipment status evaluation index, which reflects the health status of the equipment. It is a failure analysis evaluation index that reflects the failure risk of the equipment. It is a production performance evaluation index that reflects the efficiency of the production process. , , is the weight coefficient, used to balance different index pairs Contribution To adjust the parameters, used to control the growth rate of the exponential function, It is the benchmark value of the production performance evaluation index, indicating the target production efficiency. The value range is from 0 to 1, where 1 represents the best state (i.e., healthy equipment, no risk of failure, and high production efficiency), and 0 represents the worst state. When increasing, Increase, indicating that the device is in better condition. When it increases, due to the effect of radical and logarithmic function, The contribution of near hour, Increase, indicating higher production efficiency, The exponential function in the term is used to capture the risk of production efficiency being lower than ideal. Reduce, risk increase, through Apply a scaled logarithmic function and divide by The square root of the logarithm of Apply the natural logarithm function and compare it to the base value By comparison, we can evaluate the production efficiency and use the exponential function to simulate the risk when the production efficiency is lower than the ideal state. decrease, the exponential term increases rapidly, indicating that the decline in production efficiency has a great impact on The negative impact of

[0076] Furthermore, the multiple maintenance requirement levels correspond to multiple evaluation thresholds, the evaluation thresholds include an upper threshold and a lower threshold, and the corresponding relationship between the multiple maintenance requirement levels and the multiple evaluation thresholds is:

[0077] High maintenance level ; Indicates that the equipment may have a greater maintenance demand and needs to be given priority for maintenance;

[0078] Medium maintenance requirement level ; Indicates that the equipment has certain maintenance requirements, but can be arranged in the short term;

[0079] Low maintenance level ; It indicates that the equipment is in good condition, the maintenance requirement is low, and the maintenance cycle can be extended;

[0080] in, is the predicted maintenance coefficient, is the lower threshold corresponding to the high maintenance demand level and the upper threshold corresponding to the medium maintenance demand level, is the lower threshold corresponding to the medium maintenance requirement level and the upper threshold corresponding to the low maintenance requirement level, , .

[0081] Obviously, the described embodiments are only some embodiments of the present invention, not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field and related fields without creative work should fall within the scope of protection of the present invention. The structures, devices and operating methods not specifically described and explained in the present invention are implemented according to the conventional means in the field unless otherwise specified and limited.

Claims

1. An industrial data collection and management method based on the Internet of Things, characterized in that: The following steps are involved: Step 1: Conduct demand analysis on the industrial environment and production process, identify the industrial equipment indicators to be monitored in the industrial environment and production process, namely equipment performance indicators and production performance indicators, and plan data collection points to deploy IoT devices for data collection; Step 2: pre-process the collected data, perform data analysis on the equipment performance indicators, and extract sub-indicators of the equipment performance indicators, wherein the sub-indicators of the equipment performance indicators include equipment status monitoring indicators and fault analysis indicators; Step 3: Extract features based on the sub-indicators of the equipment performance indicators and the production performance indicators, further analyze the associated data of the sub-indicator features of the equipment performance indicators and the production performance indicator features, and obtain the equipment performance feature set and the production performance feature set; Step 4: Use machine learning technology in combination with the equipment performance feature set to evaluate and analyze the sub-indicators of the equipment performance indicator respectively, and use deep learning technology in combination with the production performance feature set to evaluate and analyze the production performance indicator. In step 4, the process of evaluating and analyzing the sub-indicators of the equipment performance indicator and the production performance indicator includes: Extract features associated with equipment status monitoring indicators from the equipment performance feature set, build an equipment monitoring model in combination with a neural network model, conduct a comprehensive analysis of the relevant data of temperature features, vibration signal features, and operating time features, and use the trained equipment monitoring model to evaluate the equipment status, obtain the equipment status evaluation index, and analyze the health status trend of the equipment; Extract features associated with fault analysis indicators from the equipment performance feature set, train a fault classification model based on a decision tree, conduct a comprehensive analysis of relevant data on fault frequency features, fault type features, and fault repair features, and use the trained model to analyze fault data to obtain a fault analysis evaluation index to assess the equipment's fault risk. Based on the characteristics of the production performance characteristic set that are associated with the production performance indicators, the production performance evaluation model is trained with a convolutional neural network to conduct a comprehensive analysis of the relevant data of the production speed characteristics, raw material consumption characteristics, and energy efficiency characteristics. The trained production performance evaluation model is used to evaluate the production performance and obtain a production performance evaluation index to evaluate the efficiency and production status of the production process and identify production bottlenecks and optimization opportunities. The expression of the fault analysis evaluation index is: Among them, FAI is the failure analysis assessment index, F j is the fault frequency of the jth fault type, C j is the failure cost of the jth failure type, is the direct maintenance cost, R j is the average repair time of the jth fault type, m is the total number of fault types, λ is the weight factor used to adjust the impact of fault cost and repair time on FAI, max(F) is the maximum fault frequency among all fault types, F base is the reference value of the fault frequency, which is the preset fault frequency threshold during normal operation; γ is an index parameter used to measure the impact of the maximum fault frequency on the fault analysis evaluation index; Step 5: Based on the analysis results of equipment performance indicators and production performance indicators, a predictive maintenance model is constructed in combination with relevant characteristic data to predict the maintenance trend of industrial equipment, and the maintenance demand level of industrial equipment is preset to perform preventive maintenance.

2. The method for industrial data collection and management based on the Internet of Things according to claim 1 is characterized in that: In step 1, the process of deploying IoT devices for data collection includes: Conduct demand analysis on industrial environment and production process, and analyze the performance indicators of industrial equipment that need to be monitored in the industrial environment and production process, namely equipment performance indicators and production performance indicators. Equipment performance indicators include temperature, vibration, operating time, fault frequency, fault type, and repair time. Production performance indicators include production speed, raw material consumption, and energy efficiency. Clarify the monitoring purpose of each indicator, plan data collection points on relevant industrial equipment according to the determined industrial equipment performance indicators, and deploy IoT devices at the corresponding data collection points; Deploy a data transmission network, connect IoT devices at each data collection point, and configure a data center to store, process and analyze data from IoT devices. Integrate IoT devices with the data transmission network and data center, and transmit the collected data to the data center via wireless communication devices for centralized data management and analysis.

3. The method for industrial data collection and management based on the Internet of Things according to claim 2 is characterized in that: In the step 2, the process of performing data analysis on the equipment performance indicators includes: Pre-process the collected equipment performance indicators and production performance indicators, including data cleaning and data conversion, and aggregate the collected data at time intervals; Perform data analysis on the pre-processed equipment performance indicators and further determine the sub-indicators of the equipment performance indicators, which are equipment status monitoring indicators and fault analysis indicators; Based on the relevant data of equipment status monitoring indicators, analyze the temperature, vibration and operating time data of industrial equipment, monitor the operating status of industrial equipment under different working conditions; based on the relevant data of fault analysis indicators, analyze the fault frequency, fault type and repair time data of industrial equipment, monitor the fault change trend of industrial equipment, and mark the normal data and abnormal data in the equipment status monitoring indicators and fault analysis indicators.

4. The industrial data collection and management method based on the Internet of Things according to claim 3 is characterized in that: In step 3, the process of obtaining the equipment performance characteristic set and the production performance characteristic set includes: Extract features from the sub-indicators of equipment performance indicators. For equipment status monitoring indicators, extract temperature features, vibration signal features, and operating time features. For fault analysis indicators, extract fault frequency features, fault type features, and fault repair features. Count the fault frequency and the time distribution of fault occurrence. Extract features from production performance indicators, obtain production speed features, raw material consumption features, and energy efficiency features to analyze the relationship between energy efficiency and production speed, equipment status, and identify energy efficiency anomalies; The extracted equipment status monitoring indicators and fault analysis indicator features are integrated into an equipment performance feature set, and the extracted production speed, raw material consumption and energy efficiency features are integrated into a production performance feature set.

5. The method for industrial data collection and management based on the Internet of Things according to claim 4 is characterized in that: The expression of the equipment status evaluation index is: Among them, ESI is the equipment status evaluation index, T i is the temperature reading at the ith time point, V i is the vibration signal characteristic value at the i-th time point, R is the total operating time of the equipment, and R max represents the maximum operating time of the equipment, k is a calibration parameter used to adjust the scale of the equipment condition assessment index, T min is the lowest temperature reading during the evaluation period, T base is the baseline temperature, which indicates the typical temperature of the equipment during normal operation. α and β are adjustment parameters in the exponential function, which are used to control the influence of operating time and temperature on the equipment status assessment index value. n is the total number of observation time points.

6. The method for industrial data collection and management based on the Internet of Things according to claim 5 is characterized in that: The expression of the production performance evaluation index is: Among them, PPI is the production performance evaluation index, S i is the production speed at the i-th time point, M i is the raw material consumption per unit product at the i-th time point, E i is the energy consumption per unit product at the i-th time point, n is the total number of observation time points, E base is the benchmark energy consumption, the energy consumption per unit product under optimal production conditions, E min is the minimum energy consumption, and δ is an exponential parameter used to adjust the impact of energy efficiency on the PPI value.

7. The method for industrial data collection and management based on the Internet of Things according to claim 6 is characterized in that: In step 5, the process of predicting the maintenance trend of industrial equipment includes: Traverse the relevant characteristic data of equipment performance indicators and production performance indicators, and integrate the equipment status evaluation index, the fault analysis evaluation index and the production performance evaluation index to obtain a comprehensive data set; Extract historical data from the comprehensive data set, divide the historical data into a training set and a test set, use the training set data to train the predictive maintenance model in combination with the time series prediction model, and use the test set data to test the constructed predictive maintenance model, where the historical data includes equipment performance data, production performance data, and known maintenance records; The output of the predictive maintenance model is used to combine the equipment status evaluation index, the fault analysis evaluation index and the production performance evaluation index to obtain the predictive maintenance coefficient, and comprehensively analyze the overall health status and production efficiency of industrial equipment; Combined with historical data and equipment maintenance requirements, different maintenance requirement levels for industrial equipment are preset, namely high maintenance requirement level, medium maintenance requirement level and low maintenance requirement level, and the corresponding evaluation threshold is matched to each maintenance requirement level with the predicted maintenance coefficient; Develop preventive maintenance plans based on the maintenance demand level, including maintenance schedules, resource allocation, and expected results, and deploy predictive maintenance models to analyze the maintenance needs of industrial equipment, determine the current maintenance demand level of industrial equipment, and implement corresponding preventive maintenance plans.

8. The method for industrial data collection and management based on the Internet of Things according to claim 7 is characterized in that: The expression of the predicted maintenance coefficient is: Among them, PMC is the predictive maintenance coefficient, ESI is the equipment status evaluation index, FAI is the fault analysis evaluation index, PPI is the production performance evaluation index, w1, w2, w3 are weight coefficients used to balance the contribution of different indexes to PMC, ∈ is an adjustment parameter used to control the growth rate of the exponential function, PPI base It is the benchmark value of the production performance evaluation index and represents the target production efficiency.

9. The method for industrial data collection and management based on the Internet of Things according to claim 8, characterized in that: The plurality of maintenance requirement levels correspond to the plurality of evaluation thresholds, the evaluation thresholds include an upper threshold and a lower threshold, and the correspondence between the plurality of maintenance requirement levels and the plurality of evaluation thresholds is as follows: High maintenance levelθ H ≤PMC<1; Medium maintenance requirement levelθ M ≤PMC<θ H ; Low maintenance level 0 <PMC<θ M ; Where PMC is the predicted maintenance coefficient, θ H is the lower threshold corresponding to the high maintenance demand level and the upper threshold corresponding to the medium maintenance demand level, θ M is the lower threshold corresponding to the medium maintenance demand level and the upper threshold corresponding to the low maintenance demand level, θ H =0.8,θ M =0.5.

Images