Industrial-grade data acquisition instrument data processing method and system

Abstract

The invention discloses a data processing method and system for an industrial-grade data acquisition instrument, and relates to the technical field of electric digital data processing. The system comprises an acquisition module, an analysis module and a management module, the acquisition module acquires operation parameters and production environment data of industrial equipment and transmits the operation parameters and the production environment data to the analysis module; performing single-domain analysis and association processing on the operation parameters and the production environment data to generate a cross-domain association analysis result, generating a threshold range through dynamic threshold adjustment, re-evaluating the operation parameters and the production environment data, generating an analysis result after threshold optimization, transmitting the analysis result to a management module, and managing the operation parameters and the production environment data on the basis of the analysis result after threshold optimization. And generating an optimal scheme and automatically synchronizing the optimal scheme to a workshop system, presetting edge side primary early warning and edge side multi-level early warning mechanisms, mining parameter causality through cross-domain association analysis, dynamically adjusting a threshold value in combination with working conditions and environmental factors, generating an optimal scheme through edge cloud cooperation and multiple algorithms, and enabling industrial intelligent fine management.

Description

Technical Field

[0001] The present invention relates to the technical field of electronic digital data processing, and in particular to a data processing method and system for an industrial-grade data acquisition instrument. Background Art

[0002] In recent years, intelligent management of industrial equipment and precise control of the production environment have become the key to improving the core competitiveness of enterprises. As a bridge connecting physical equipment and digital systems, the data processing capabilities of industrial data acquisition instruments directly affect the efficiency, quality and safety of industrial production.

[0003] Traditional systems often analyze operating parameters and production environment data in isolation, and are unable to explore the inherent connections between multi-source data, making fault prediction and root cause location difficult; static threshold warning mechanisms are difficult to adapt to changes in working conditions and are prone to false alarms and missed alarms; edge computing capabilities are weak, and reliance on cloud processing results in high latency and low reliability. At the same time, industrial scenarios have special requirements for high-frequency data collection, multi-objective optimization decision-making, and adaptability to complex environments, but existing technologies are insufficient in data correlation analysis, dynamic threshold adjustment, and warning response mechanisms, and cannot meet these requirements. Summary of the Invention

[0004] The technical problems solved by the present invention are: traditional systems often analyze operating parameters and production environment data in isolation, and are unable to explore the intrinsic connections between multi-source data, resulting in difficulties in fault prediction and root cause location; static threshold warning mechanisms are difficult to adapt to changes in working conditions and are prone to false alarms and missed alarms; edge computing capabilities are weak, and reliance on cloud processing causes high latency and low reliability. At the same time, industrial scenarios have special requirements for high-frequency data collection, multi-objective optimization decision-making, and adaptability to complex environments, but existing technologies are insufficient in data correlation analysis, dynamic threshold adjustment, and warning response mechanisms, and cannot meet the needs.

[0005] In order to solve the above technical problems, the present invention provides the following technical solutions: comprising a collection module, an analysis module and a management module;

[0006] The acquisition module collects the operating parameters and production environment data of the industrial equipment and transmits them to the analysis module;

[0007] The analysis module performs single-domain analysis and correlation processing on the operating parameters and production environment data, generates a cross-domain correlation analysis result, generates a threshold range based on the cross-domain correlation analysis result through dynamic threshold adjustment, re-evaluates the operating parameters and production environment data, generates an analysis result after threshold optimization, and transmits it to the management module;

[0008] The management module generates an optimal solution based on the analysis results after threshold optimization and automatically synchronizes it to the workshop system, presetting the edge-side primary warning and edge-side multi-level warning mechanisms.

[0009] As a preferred solution of the data processing system of an industrial data acquisition instrument described in the present invention, wherein: the operating parameters include rotation speed, temperature, pressure, current and vibration frequency;

[0010] The production environment data includes temperature, humidity, dust concentration, gas content and air pressure;

[0011] The acquisition module includes an edge computing node for collecting operating parameters and production environment data through a sensor network;

[0012] The edge computing node is deployed with a lightweight data processing engine, which performs high-frequency sampling of operating parameters and production environment data based on a time window algorithm, and the sampling frequency is greater than or equal to a first frequency threshold;

[0013] Extracting fluctuation characteristic values ​​from the operating parameters and production environment data by applying a sliding filter algorithm;

[0014] When it is detected that the change rate of operating parameters and production environment data exceeds the first grade warning threshold 1st gradient threshold, the edge side primary warning is triggered. The edge side primary warning only uploads the abnormal data packet to the cloud through an encrypted channel, and does not execute the sound and light alarm.

[0015] As a preferred solution of the data processing system for an industrial-grade data acquisition instrument of the present invention, wherein: the single-domain analysis result includes a first single-domain analysis result and a second single-domain analysis result;

[0016] The first single-domain analysis result is obtained by analyzing and processing the operating parameters by the operating parameter processing unit, including a time domain feature extraction result, a state quantification evaluation result and a fault matching result;

[0017] The time domain feature extraction results include calculating the mean, variance, rate of change and kurtosis coefficient of the operating parameters;

[0018] The state quantification evaluation result is to input the time domain characteristics into a preset evaluation function to generate a health index of the industrial equipment;

[0019] The fault matching result is obtained by comparing the time domain feature with a preset state feature library of industrial equipment to obtain a comparison result with the state feature library of industrial equipment;

[0020] The second single domain analysis result is obtained by analyzing and processing the production environment data through the production environment data processing unit, including spatial feature extraction results, impact quantification results and environmental suitability assessment results;

[0021] The spatial feature extraction result is obtained by calculating the regional mean, standard deviation, gradient and skewness coefficient of the production environment data;

[0022] The impact quantification result generates an association rule base based on the correlation analysis of historical data, and calculates the impact of environmental parameters on production indicators;

[0023] The environmental suitability assessment result is generated by constructing a multidimensional space of environmental parameters, calculating the Euclidean distance between the current environmental state point and the optimal production state point of historical data, and generating an environmental suitability index.

[0024] As a preferred solution of the data processing system for an industrial-grade data acquisition instrument according to the present invention, the root cause of the abnormal situation is located based on the first single-domain analysis result and the second single-domain analysis result, specifically including:

[0025] When it is detected that the operating parameters or production environment data exceed the dynamic threshold, based on the abnormality occurrence time point t, the multidimensional operating parameters and production environment data within the t±Δt time window are extracted;

[0026] The dynamic threshold is set based on historical data statistics;

[0027] Construct a temporal correlation network between operating parameters and production environment data, and calculate the causal strength value of each parameter node through the Bayesian network;

[0028] Sorting based on the causal strength values ​​to generate an ordered list of possible root causes;

[0029] A time series trend analysis module is built into the operating parameter processing unit and the production environment data processing unit to construct a sliding time window based on historical operating parameters and production environment data;

[0030] The window length is configured as a first time threshold range, and a difference prediction is performed on the future trends of the operating parameters and the production environment data;

[0031] When the deviation between the predicted value and the dynamic threshold exceeds the preset warning coefficient, a trend warning signal is generated, and the time-space distribution characteristics of the operating parameter fluctuations are visualized through a heat map. The heat map color mapping is based on the deviation and includes safety, warning and danger.

[0032] As a preferred solution of the data processing system for an industrial-grade data acquisition instrument described in the present invention, wherein: the first single-domain analysis result and the second single-domain analysis result are correlated by a cross-domain correlation analysis unit to generate a cross-domain correlation analysis result including key environmental factors;

[0033] The cross-domain correlation analysis unit uses a multivariate correlation analysis algorithm to quantify the transmission lag between operating parameters and production environment data and establish a multivariate causal relationship network map;

[0034] The key environmental factors are ranked based on causal strength, identifying key environmental factors that affect the status of industrial equipment and generating a factor importance ranking report.

[0035] As a preferred solution of the industrial-grade data acquisition instrument data processing system described in the present invention, dynamic threshold adjustment is performed based on the cross-domain correlation analysis results, specifically including:

[0036] Determine the current operating condition by combining the operating parameters, and establish an operating condition-threshold template mapping table;

[0037] When the working condition switches, the threshold template of the corresponding working condition is automatically called;

[0038] The confidence interval of the threshold is updated based on the mean ± 3σ of the statistical properties of historical data;

[0039] Based on the key environmental factors identified in the cross-domain correlation analysis results, independently adjust the threshold range of the corresponding operating parameters, and the adjustment range does not exceed the first proportional threshold range of the current threshold;

[0040] The operating parameters and production environment data are re-evaluated based on the threshold range to generate analysis results after the threshold is optimized.

[0041] As a preferred solution of the industrial-grade data acquisition instrument data processing system described in the present invention, wherein: the management module includes an SOP management unit;

[0042] The SOP management unit is a subunit of the management module for performing standard operating procedure optimization. Based on the analysis results and feedback data after the threshold optimization, the dynamic verification and optimization of the SOP are achieved through three-dimensional process deduction and multi-objective optimization algorithm;

[0043] Multiple sets of SOP optimization plans are generated through multi-objective optimization algorithms, and the plans are optimized based on production efficiency, energy consumption, industrial equipment and life indicators, and automatically synchronized to the workshop execution system.

[0044] As a preferred solution of the data processing system for an industrial-grade data collector described in the present invention, wherein: the edge computing node includes mixed-precision data storage;

[0045] High-frequency sampling data is stored with n-bit floating-point precision, and abnormal data packets are automatically upgraded to 1-bit floating-point precision;

[0046] and reducing transmission bandwidth occupancy through differential coding compression technology, with a compression ratio not lower than a first numerical threshold;

[0047] And a multi-level early warning mechanism is preset on the edge side.

[0048] As a preferred solution of the industrial-grade data acquisition instrument data processing system described in the present invention, the multi-level early warning mechanism includes:

[0049] According to the change rate of operating parameters or production environment data, it is divided into yellow warning and red warning;

[0050] When the rate of change of operating parameters or production environment data exceeds the PST gradient threshold, a yellow warning is triggered;

[0051] When the rate of change of operating parameters or production environment data exceeds the end gradient threshold, a red alert is triggered;

[0052] When the yellow warning is triggered, the local sound and light alarm is activated. When the red warning is triggered, the local sound and light alarm and remote SMS notification are triggered simultaneously.

[0053] A data processing method for an industrial data acquisition instrument, which is applied to a data processing system for an industrial data acquisition instrument, comprises:

[0054] Step S1, collecting operating parameters and production environment data of industrial equipment;

[0055] Step S2: performing single-domain analysis and correlation processing on the operating parameters and production environment data to generate a cross-domain correlation analysis result; generating a threshold range based on the cross-domain correlation analysis result through dynamic threshold adjustment; re-evaluating the operating parameters and production environment data and generating an analysis result after threshold optimization;

[0056] In step S3, based on the analysis results after threshold optimization, an optimal solution is generated and automatically synchronized to the workshop system, and the edge-side primary warning and edge-side multi-level warning mechanisms are preset.

[0057] The beneficial effects of the present invention are: mining parameter causal relationships through cross-domain data correlation analysis, realizing dynamic threshold adaptive adjustment based on working conditions and key environmental factors, utilizing edge computing and cloud collaboration to reduce latency, combining three-dimensional process deduction and multi-objective optimization algorithms to generate optimal solutions, effectively solving existing problems, and providing a comprehensive solution for industrial intelligent management and refined control. BRIEF DESCRIPTION OF THE DRAWINGS

[0058] Figure 1 A basic flow chart of a data processing system for an industrial-grade data acquisition instrument provided by one embodiment of the present invention.

[0059] Figure 2 A schematic flow chart of the steps of a data processing method for an industrial-grade data acquisition instrument provided by one embodiment of the present invention. DETAILED DESCRIPTION

[0060] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the specific implementation methods of the present invention are described in detail below in conjunction with the drawings. It is obvious that the described embodiments are only part of the embodiments of the present invention, but not all of the embodiments.

[0061] Example 1, with reference to Figure 1 , as an embodiment of the present invention, provides an industrial-grade data acquisition instrument data processing system, including an acquisition module, an analysis module and a management module;

[0062] The acquisition module collects the operating parameters and production environment data of industrial equipment and transmits them to the analysis module;

[0063] The analysis module performs single-domain analysis and correlation processing on operating parameters and production environment data, generates cross-domain correlation analysis results, generates threshold ranges based on the cross-domain correlation analysis results through dynamic threshold adjustment, re-evaluates operating parameters and production environment data, generates analysis results after threshold optimization, and transmits them to the management module;

[0064] The management module generates the optimal solution based on the analysis results after threshold optimization and automatically synchronizes it to the workshop system, presetting the edge-side primary warning and edge-side multi-level warning mechanisms.

[0065] In one embodiment, the system consists of an acquisition module, an analysis module and a management module, and the modules operate in coordination. The acquisition module collects the operating parameters and production environment data of industrial equipment in real time, and transmits the operating parameters and production environment data to the analysis module; the analysis module first performs single-domain analysis on the operating parameters and production environment data respectively to obtain single-domain analysis results, and then performs correlation processing on the single-domain analysis results to generate cross-domain correlation analysis results. Based on the cross-domain correlation analysis results, an adjusted threshold range is generated through dynamic threshold adjustment. The operating parameters and production environment data are re-evaluated according to the threshold range to generate analysis results after threshold optimization, and the analysis results are transmitted to the management module; based on the analysis results after threshold optimization, the management module generates an optimal solution and automatically synchronizes it to the workshop system, and at the same time presets the edge-side primary warning and edge-side multi-level warning mechanisms to ensure efficient and safe operation of industrial production.

[0066] Operating parameters include speed, temperature, pressure, current and vibration frequency;

[0067] Production environment data includes temperature, humidity, dust concentration, gas content and air pressure;

[0068] The acquisition module includes edge computing nodes, which are used to collect operating parameters and production environment data through sensor networks;

[0069] The edge computing node is deployed with a lightweight data processing engine, which performs high-frequency sampling of operating parameters and production environment data based on a time window algorithm, with the sampling frequency being greater than or equal to the first frequency threshold;

[0070] The operating parameters and production environment data are subjected to the sliding filter algorithm to extract the fluctuation characteristic values;

[0071] When it is detected that the change rate of operating parameters and production environment data exceeds the first-level warning threshold 1st gradient threshold, the edge-side primary warning is triggered. The edge-side primary warning only uploads the abnormal data packet to the cloud through an encrypted channel, and does not execute the sound and light alarm.

[0072] In one embodiment, the operating parameters include industrial equipment speed (unit: revolutions per minute, measurement accuracy ±0.5%), temperature (range -50°C to 200°C, resolution 0.1°C), pressure (0-10MPa, accuracy ±0.2%FS), current (0-100A, resolution 0.01A) and vibration frequency (1-10kHz, sampling bit 16bit); production environment data include temperature and humidity (temperature -20°C to 80°C, humidity 0-100%RH, accuracy ±0.3°C / ±3%RH), dust concentration (PM2.5 / PM10, measurement range 0-1000μg / m3, resolution 1μg / m3), gas content (including O2, CO and CO2, measurement range 0-100%VOL, accuracy ±1%) and air pressure (700-1100hPa, accuracy ±0.3hPa). The acquisition module is composed of distributed edge computing nodes (using ARM At the same time, the acquisition module is also responsible for the collection and storage of historical data. It automatically retains operating parameters and production environment data according to the set time period (such as daily and weekly), and the storage time is not less than 90 days, providing historical reference for subsequent data analysis. The storage of historical data adopts a hierarchical compression strategy. The real-time data retains the original accuracy. The historical data is aggregated by day / week and then lossy compression is adopted (such as retaining 3 decimal places). The compression ratio is not less than 5:1. The edge node and the cloud adopt an incremental synchronization mechanism, and the data of the previous 24 hours are automatically synchronized at dawn every day. The synchronization success rate is not less than 99.9%. The edge computing node is deployed with a lightweight data processing engine, which performs high-frequency sampling of operating parameters and production environment data based on the time window algorithm. The sampling frequency is The rate is greater than or equal to 100Hz (the first frequency threshold). After preprocessing with a sliding time window, the high-frequency sampling data is cached in real time in the ring buffer of the edge node (capacity ≥ 1 million records). When the data change rate is detected to exceed the 1st gradient threshold, the feature extraction module is triggered to batch process the most recent 1000 records in the buffer. Operating parameters and production environment data are collected through the industrial-grade sensor network (supporting the Modbus RTU protocol) and transmitted in JSON format, including timestamp, industrial equipment ID, parameter category (operating parameters / production environment data), and operating parameter and production environment data values ​​and units. The edge computing node and the analysis module transmit through the MQTT protocol, and the cloud and the management module interact through the HTTPS interface. Operating parameters and production environment data are collected through the industrial-grade sensor network (supporting the Modbus RTU protocol). Each edge computing node is deployed with a lightweight data processing engine (an embedded AI inference framework accelerated by TensorRT) that supports the following processing capabilities:

[0073] High-frequency sampling mechanism, based on a sliding time window algorithm with an adjustable window length of 100ms to 10s, synchronously samples operating parameters and production environment data at a frequency greater than or equal to 100Hz (the first frequency threshold), with a sampling interval error of less than or equal to ±1ms;

[0074] Feature extraction algorithm, applying adaptive sliding filter algorithms (Kalman filter, median filter and low-pass filter) to extract time domain features of operating parameters and production environment data, including mean, variance and kurtosis, and frequency domain features (FFT spectrum, wavelet transform coefficients);

[0075] When the change rate of operating parameters and production environment data exceeds the manually preset 1st (first warning classification threshold) gradient threshold, the edge computing node triggers a primary warning and only uploads the abnormal data packet to the cloud through an encrypted channel without performing an audible or visual alarm. This warning is designed to quickly capture potential anomalies and complete data transmission, providing a basis for subsequent analysis;

[0076] The single domain analysis results include a first single domain analysis result and a second single domain analysis result;

[0077] The first single domain analysis result is obtained by analyzing and processing the operating parameters through the operating parameter processing unit, including a time domain feature extraction result, a state quantification evaluation result and a fault matching result;

[0078] The time domain feature extraction results include the calculation of the mean, variance, rate of change and kurtosis coefficient of the operating parameters;

[0079] The result of state quantification evaluation is to input the time domain characteristics into the preset evaluation function to generate the health index of industrial equipment;

[0080] The fault matching result is obtained by comparing the time domain characteristics with the preset state feature library of industrial equipment;

[0081] The second single domain analysis result is obtained by analyzing and processing the production environment data through the production environment data processing unit, including spatial feature extraction results, impact quantification results and environmental suitability assessment results;

[0082] The spatial feature extraction results are obtained by calculating the regional mean, standard deviation, gradient and skewness coefficient of the production environment data;

[0083] The impact quantification results generate an association rule base based on the correlation analysis of historical data, and calculate the impact of environmental parameters on production indicators;

[0084] The environmental suitability assessment results are generated by constructing a multidimensional space of environmental parameters, calculating the Euclidean distance between the current environmental state point and the optimal production state point of historical data, and generating an environmental suitability index.

[0085] In one embodiment, the first single-domain analysis result includes a time domain feature extraction result, a state quantification evaluation result, and a fault matching result obtained by analyzing and processing the operation parameters by the operation parameter processing unit;

[0086] The time domain feature extraction results specifically include the calculated mean, variance, rate of change, and kurtosis coefficient of the operating parameters, which can reflect the characteristics of the operating parameters from different angles. For example, the mean reflects the average level of the operating parameters, the variance measures the degree of dispersion of the operating parameters, the rate of change shows how the operating parameters change over time, and the kurtosis coefficient is used to describe the steepness of the distribution of the operating parameters.

[0087] The state quantitative assessment results input the extracted time domain features into a preset evaluation function to generate the health index of the industrial equipment. The health index is a quantitative assessment of the overall operating status of the industrial equipment and can intuitively reflect the current health status of the industrial equipment.

[0088] The fault matching result is obtained by comparing the time domain features with the preset state feature library of industrial equipment. The state feature library stores feature data under different operating states. Through comparison, it can be determined whether the current state of the industrial equipment is normal and the possible fault type;

[0089] The second single domain analysis result is obtained by analyzing and processing the production environment data through the production environment data processing unit, including spatial feature extraction results, impact quantification results and environmental suitability assessment results;

[0090] The spatial feature extraction results are obtained by calculating the regional mean, standard deviation, gradient, and skewness coefficient of the production environment data. The regional mean reflects the average level of production environment data within a certain area, the standard deviation reflects the degree of dispersion of production environment data, the gradient is used to describe the spatial variation trend of production environment data, and the skewness coefficient shows the degree of skewness of the production environment data distribution.

[0091] The impact quantification results are generated based on the correlation analysis of historical data to generate an association rule base, and then the impact of environmental data on production indicators is calculated. By analyzing the relationship between production environment data and production indicators in historical data, an association rule base is established to quantitatively evaluate the impact of production environment data on production indicators;

[0092] The results of the environmental suitability assessment are generated by constructing a multidimensional space of production environment data, calculating the Euclidean distance between the current environmental state point and the optimal production state point in historical data, and generating an environmental suitability index. This index can reflect the suitability of the current production environment for the operation of industrial equipment and production activities, and provide a basis for optimizing the production environment.

[0093] Locate the root cause of the abnormality based on the first and second single-domain analysis results, specifically including:

[0094] When it is detected that the operating parameters or production environment data exceed the dynamic threshold, based on the abnormality occurrence time point t, the multidimensional operating parameters and production environment data within the t±Δt time window are extracted;

[0095] Dynamic thresholds are set based on historical data statistics;

[0096] Construct a temporal correlation network between operating parameters and production environment data, and calculate the causal strength value of each parameter node through the Bayesian network;

[0097] Sort by causal strength value to generate an ordered list of possible root causes;

[0098] A time series trend analysis module is built into the operating parameter processing unit and the production environment data processing unit to build a sliding time window based on historical operating parameters and production environment data;

[0099] The window length is configured as the first time threshold range, and the future trend of the operating parameters and production environment data is predicted by difference;

[0100] When the deviation between the predicted value and the dynamic threshold exceeds the preset warning coefficient, a trend warning signal is generated, and the time-space distribution characteristics of the operating parameter fluctuations are visualized through a heat map. The heat map color mapping is based on the deviation and includes safety, warning and danger.

[0101] In one embodiment, multi-dimensional operating parameters or production environment data extraction: when it is detected that the operating parameters or production environment data exceed the threshold dynamically set based on the 95% confidence interval of historical data, the system automatically extracts multi-dimensional operating parameters and production environment data within ±5 minutes (configurable time window Δt) before and after the abnormality occurrence time point t, including industrial equipment speed, temperature and pressure operating parameters, as well as temperature, humidity and dust concentration production environment data. The dynamic threshold setting is automatically updated based on the statistical characteristics of the historical data (mean ± 3σ). The default value of Δt is 5 minutes and can be adjusted according to the response characteristics of the industrial equipment. The window length is dynamically configured according to the type of industrial equipment: the default value for fast-responding industrial equipment (such as motors) is 15 minutes and can be manually adjusted to 1-12 hours; the default value for slow-responding industrial equipment (such as kilns) is 4 hours and can be manually adjusted to 6-24 hours. Using the moving average method, the operating parameter points and production environment data points in the window are taken from the latest 200 sampling values ​​corresponding to a time period of 2 minutes to 4 hours, and are calculated at a sampling rate of 100 Hz.

[0102] Causal relationship analysis: A temporal correlation network between operating parameters and production environment data is constructed. The Pearson correlation coefficient (value range [-1, 1]) and Granger causality test (significance level p < 0.05) are calculated between all operating parameters involved in the causal relationship analysis and production environment data to quantify the strength of the causal relationship between all operating parameters and production environment data involved in the analysis. Based on the calculation results, an ordered list of possible root causes is generated. The sorting rule is: parameters with larger absolute values ​​of correlation coefficients and higher causal relationship significance are ranked higher.

[0103] Trend prediction and warning: A time series trend analysis module is built into the operating parameter processing unit and the production environment data processing unit. Based on the moving average method, the window length can be configured to 15 minutes to 24 hours (the first time threshold range) to predict future operating parameters and production environment data values. When the deviation between the predicted value and the dynamic threshold exceeds the preset warning coefficient (default 0.7, configurable range 0.1-1.0), the system automatically generates a trend warning signal and visualizes the time-space distribution characteristics of the fluctuations of operating parameters and production environment data through a heat map. The heat map uses a three-color gradient mapping, specifically including:

[0104] Green: safe area (deviation < 0.3);

[0105] Yellow: Warning area (0.3≤deviation<0.7);

[0106] Red: Dangerous area (deviation ≥ 0.7);

[0107] The preset warning coefficient for sensitive industrial equipment is recommended to be set to 0.5, and for ordinary industrial equipment to be set to 0.7. The refresh frequency of the heat map is: real-time update, and the historical data retention period is ≥30 days.

[0108] The first single-domain analysis result and the second single-domain analysis result are correlated by a cross-domain correlation analysis unit to generate a cross-domain correlation analysis result including key environmental factors;

[0109] The cross-domain correlation analysis unit uses a multivariate correlation analysis algorithm to quantify the transmission lag between operating parameters and production environment data and establish a multivariate causal relationship network map;

[0110] Key environmental factors are ranked based on causal strength, identifying key environmental factors that affect the status of industrial equipment and generating a factor importance ranking report.

[0111] In one embodiment, multivariate correlation calculation, the cross-domain correlation analysis unit adopts a multivariate correlation analysis algorithm to quantify the degree of correlation between the operating parameters and the production environment data by calculating the Pearson correlation coefficient, which ranges from -1 to 1. The closer the absolute value is to 1, the stronger the correlation is. For example, when the Pearson correlation coefficient of the industrial equipment temperature and the ambient humidity is 0.8, it indicates that the industrial equipment temperature and the ambient humidity are highly positively correlated. At the same time, cross-correlation analysis is used to calculate the conduction hysteresis between the operating parameters and the production environment data. It is determined how long it takes for a change in one variable to cause a change in another variable. For example, the analysis found that after the ambient temperature increases, there is a 2-hour delay in the decrease in the viscosity of the lubricating oil of the industrial equipment.

[0112] Causal network construction: Based on the above correlation and conduction hysteresis analysis results, a multivariate causal network map is constructed. In the map, nodes represent operating parameters or production environment data, and the lines between nodes represent the causal relationship between variables. The thickness of the line indicates the strength of the causal relationship. For example, if changes in environmental dust concentration significantly affect the degree of bearing wear in industrial equipment, there will be a thicker line between the two nodes in the map.

[0113] Identification of key environmental factors: Based on the causal strength of each connection in the causal network diagram, key environmental factors that have a significant impact on the status of industrial equipment are identified. For example, through sorting, it is found that ambient temperature, humidity and dust concentration are the top three key environmental factors affecting the failure rate of industrial equipment. Finally, a factor importance ranking report is generated to intuitively present the degree of influence of each environmental factor on the status of industrial equipment.

[0114] Dynamic threshold adjustment based on cross-domain correlation analysis results, including:

[0115] Determine the current working condition by combining the operating parameters and establish a working condition-threshold template mapping table;

[0116] When the working condition switches, the threshold template of the corresponding working condition is automatically called;

[0117] The confidence interval of the threshold is updated based on the mean ± 3σ of the statistical properties of historical data;

[0118] Based on the key environmental factors identified in the cross-domain correlation analysis results, independently adjust the threshold range of the corresponding operating parameters, and the adjustment range does not exceed the first proportional threshold range of the current threshold;

[0119] Re-evaluate operating parameters and production environment data based on threshold ranges to generate analysis results after threshold optimization.

[0120] In one embodiment, the operating condition-threshold template mapping determines the current operating condition by combining the speed, load current and pressure of the industrial equipment, such as industrial equipment startup, full-load operation and low-load standby, and establishes an operating condition-threshold template mapping table. For example, the "full-load operation" operating condition is associated with a threshold template including a temperature upper limit of 80°C and a vibration frequency upper limit of 100Hz, and different operating conditions correspond to independent threshold ranges; operating condition switching response: when the operating parameter combination triggers the operating condition switching condition, such as the speed increases from 500 rpm to 1500 rpm, the system automatically calls the threshold template of the corresponding operating condition to avoid false alarms under different operating conditions; confidence interval dynamic update: based on the statistical characteristics of the historical data mean ±3σ, the fluctuation range of the operating parameters and production environment data is calculated, and the confidence interval of the threshold is updated once an hour. For example, when the mean of the historical temperature data of the industrial equipment is 65°C and the standard deviation is 3°C, the dynamic threshold interval is 56-74°C (65±3×3);

[0121] Fine-tuning of key factor thresholds: Based on key environmental factors identified by cross-domain correlation analysis, such as dust concentration and ambient temperature, the threshold range of affected operating parameters is independently adjusted. The adjustment range is limited to ±20% (the first proportional threshold range) of the current threshold. For example, when the dust concentration exceeds the standard, the bearing wear threshold is dynamically adjusted from 50μm to 60μm (+20%).

[0122] Re-evaluate operating parameters and production environment data. Based on the updated threshold range, conduct real-time monitoring and re-evaluation of operating parameters such as temperature and pressure and production environment data such as humidity and gas content, and generate analysis results after threshold optimization for subsequent early warning decisions.

[0123] The management module includes an SOP management unit;

[0124] The SOP management unit is a subunit of the management module used to perform standard operating procedure optimization. Based on the analysis results and feedback data after threshold optimization, it realizes dynamic verification and optimization of SOPs through three-dimensional process deduction and multi-objective optimization algorithm.

[0125] Multiple sets of SOP optimization plans are generated through multi-objective optimization algorithms, and the plans are optimized based on production efficiency, energy consumption, industrial equipment and life indicators, and automatically synchronized to the workshop execution system.

[0126] In one embodiment, based on the analysis results after threshold optimization (such as industrial anomaly warning and production parameter fluctuation assessment) and workshop execution feedback data (such as actual operation time and resource consumption records), an optimization decision-making basis is constructed. For example, when the threshold adjustment is triggered by the abnormal temperature of industrial equipment, the SOP optimization process is started in combination with the production efficiency decline data; through the three-dimensional process deduction mechanism, the process execution is simulated from three dimensions: time axis, resource axis and risk axis; the time consumption of each link of the SOP is disassembled on the time axis, the bottleneck steps are accurately located, the occupancy of manpower, materials and industrial equipment is analyzed on the resource axis, and the risk axis assessment process may trigger risk points of abnormal thresholds of operating parameters and production environment data, such as a certain operation step that may cause the industrial equipment load to exceed the limit;

[0127] Based on the analysis results and feedback data after threshold optimization, dynamic verification and optimization of SOPs are achieved through three-dimensional process simulation and multi-objective optimization algorithms. Multiple SOPs are generated using a Pareto optimal algorithm. Normalization eliminates dimensionality effects. The system comprehensively optimizes production efficiency (output per unit time), energy consumption (power consumption per unit product), and industrial equipment life (remaining operating hours predicted based on historical failure data) to generate multiple SOP optimization plans. These plans are then quantitatively compared based on the core indicators of production efficiency, energy consumption, and industrial equipment life. For example, Plan A can improve efficiency by 15% by shortening the number of operating steps, but increase energy consumption by 10%. Plan B reduces the load on industrial equipment through phased operations, extending the life of industrial equipment by 20% while maintaining efficiency. The system prioritizes plans that simultaneously improve efficiency by ≥5% and reduce energy consumption by ≥3%. If both cannot be achieved, priorities are manually assigned by production stage (such as "capacity priority" or "industrial equipment maintenance priority") and automatically synchronized to the shop floor execution system, linking industrial equipment control parameters with operating instructions to ensure seamless implementation of the new process. The system also continuously monitors the execution results, forming a closed-loop management system of "analysis-optimization-execution-feedback."

[0128] Edge computing nodes include mixed-precision data storage;

[0129] High-frequency sampling data is stored with n-bit floating-point precision, and abnormal data packets are automatically upgraded to l-bit floating-point precision;

[0130] and reducing transmission bandwidth occupancy through differential coding compression technology, with a compression ratio not lower than a first numerical threshold;

[0131] And a multi-level early warning mechanism is preset on the edge side.

[0132] Multi-level early warning mechanism, including:

[0133] According to the change rate of operating parameters or production environment data, it is divided into yellow warning and red warning;

[0134] When the rate of change of operating parameters or production environment data exceeds the PST gradient threshold, a yellow warning is triggered;

[0135] When the rate of change of operating parameters or production environment data exceeds the end gradient threshold, a red alert is triggered.

[0136] When the yellow warning is triggered, the local sound and light alarm is activated. When the red warning is triggered, the local sound and light alarm and remote SMS notification are triggered simultaneously.

[0137] In one embodiment, the edge computing node integrates mixed-precision data storage and intelligent compression technology, optimizes data management through differentiated storage strategies, and uses 16(n)-bit floating-point precision to store high-frequency sampling data (sampling frequency ≥ 100Hz) of industrial equipment speed and temperature, which can not only meet routine monitoring needs but also reduce storage overhead. For example, "25.3℃" is encoded as 0xC359; when it is detected that the rate of change of operating parameters exceeds the threshold or triggers the risk mode, the abnormal data packet (including 5s original data + 30s historical data) is automatically stored. The accuracy of the floating point data is improved to 32(1) bits, ensuring that key abnormality details are fully recorded. For example, the abnormal vibration peak value of industrial equipment "123.456Hz" is stored with higher precision. At the same time, differential coding compression technology is used to calculate the difference between adjacent sampling points for encoding and transmission, achieving a compression ratio of no less than 8:1 (the first numerical threshold). For example, [10.0A, 10.2A, 10.4A] is compressed to [10.0A, 0.2A, 0.2A], significantly reducing bandwidth usage and ensuring that abnormal data is quickly uploaded to the cloud within 500ms.

[0138] Based on the primary warning on the edge side, the edge computing node presets a multi-level warning mechanism on the edge side, which accurately responds based on the change rate of operating parameters or production environment data. When the change rate of operating parameters or production environment data exceeds the 1st (pst) gradient threshold, for example, the temperature rises by 5°C within 10 minutes, a yellow warning is triggered, and a local sound and light alarm (sound pressure level ≥ 85dB, light intensity ≥ 1000cd) is activated to alert on-site personnel. When the change rate of operating parameters or production environment data exceeds the 2nd (end) gradient threshold, for example, the temperature rises by 15°C within 10 minutes, a red warning is triggered. In addition to the local sound and light alarm, a remote SMS notification is sent to the operation and maintenance personnel at the same time, such as "The temperature of industrial equipment in production line A exceeds the threshold, the current value is 95°C, and the threshold is 70°C", ensuring that key anomalies are handled in a timely manner, forming a closed-loop management of "monitoring-warning-response".

[0139] For example, the motor on a factory production line has a speed of 0-3000RPM and a temperature monitoring range of -20℃ to 120℃. Under normal operating conditions, the speed is 1500RPM, the temperature is 65℃, the ambient humidity is 50% RH, and there is no warning. Under abnormal operating conditions, the speed suddenly rises to 2800RPM (exceeding the threshold of 2500RPM), triggering a primary warning on the edge side, and the abnormal data is uploaded to the cloud. When the temperature rises by 12℃ within 10 minutes (exceeding the 1st gradient threshold of 5℃ / 10min), a yellow warning is further triggered and a local sound and light alarm is activated. If the temperature suddenly rises by 15℃ (exceeding the end gradient threshold), a red warning is triggered, and a remote SMS is sent to notify the operation and maintenance personnel.

[0140] Example 2, reference Figure 2 Another embodiment of the present invention is different from the first embodiment in that it provides an industrial-grade data acquisition instrument data processing method, including:

[0141] Step S1, collecting operating parameters and production environment data of industrial equipment;

[0142] Step S2: Perform single-domain analysis and correlation processing on the operating parameters and production environment data to generate cross-domain correlation analysis results. Based on the cross-domain correlation analysis results, generate a threshold range through dynamic threshold adjustment, re-evaluate the operating parameters and production environment data, and generate analysis results after threshold optimization.

[0143] In step S3, based on the analysis results after threshold optimization, an optimal solution is generated and automatically synchronized to the workshop system, and the edge-side primary warning and edge-side multi-level warning mechanisms are preset.

[0144] In one of the embodiments, the operating parameters and production environment data of industrial equipment are collected in real time, and single-domain analysis is performed on the operating parameters and production environment data respectively to obtain single-domain analysis results. The single-domain analysis results are then correlated to generate cross-domain correlation analysis results. Based on the cross-domain correlation analysis results, an adjusted threshold range is generated through dynamic threshold adjustment. The operating parameters and production environment data are re-evaluated based on the threshold range to generate analysis results after threshold optimization. Based on the analysis results after threshold optimization, an optimal solution is generated and automatically synchronized to the workshop system. The edge-side primary warning and edge-side multi-level warning mechanisms are preset to ensure efficient and safe operation of industrial production.

[0145] The present invention mines parameter causal relationships through cross-domain data correlation analysis, implements dynamic threshold adaptive adjustment based on working conditions and key environmental factors, uses edge computing and cloud collaboration to reduce latency, and combines three-dimensional process deduction and multi-objective optimization algorithms to generate optimal solutions, effectively solving existing problems and providing a comprehensive solution for industrial intelligent management and refined control.

[0146] Those skilled in the art will appreciate that embodiments of the present invention may be provided as methods, systems, or computer program products. Therefore, the present invention may take the form of a fully hardware embodiment, a fully software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media containing computer-usable program code. The storage medium may be implemented by any type of volatile or non-volatile storage device, or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic storage, flash memory, magnetic disk, or optical disk. These computer program instructions may also be stored in a computer readable memory that can direct a computer or other programmable data processing device to work in a specific manner, so that the instructions stored in the computer readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 a process or multiple processes and / or boxes Figure 1 The function specified in one or more boxes.

[0147] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit the present invention. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical solutions of the present invention, which should all be included in the scope of the claims of the present invention.

Claims

1. An industrial-grade data acquisition instrument data processing system, characterized in that: Including acquisition module, analysis module and management module; The acquisition module collects the operating parameters and production environment data of the industrial equipment and transmits them to the analysis module; The analysis module performs single-domain analysis and correlation processing on the operating parameters and production environment data, generates a cross-domain correlation analysis result, generates a threshold range based on the cross-domain correlation analysis result through dynamic threshold adjustment, re-evaluates the operating parameters and production environment data, generates an analysis result after threshold optimization, and transmits it to the management module; The management module generates an optimal solution based on the analysis results after threshold optimization and automatically synchronizes it to the workshop system, presetting the edge-side primary warning and edge-side multi-level warning mechanisms.

2. The industrial-grade data acquisition instrument data processing system according to claim 1, characterized in that: The operating parameters include rotation speed, temperature, pressure, current and vibration frequency; The production environment data includes temperature, humidity, dust concentration, gas content and air pressure; The acquisition module includes an edge computing node for collecting operating parameters and production environment data through a sensor network; The edge computing node is deployed with a lightweight data processing engine, which performs high-frequency sampling of operating parameters and production environment data based on a time window algorithm, and the sampling frequency is greater than or equal to a first frequency threshold; Extracting fluctuation characteristic values ​​from the operating parameters and production environment data by applying a sliding filter algorithm; When it is detected that the change rate of operating parameters and production environment data exceeds the first grade warning threshold 1st gradient threshold, the edge side primary warning is triggered. The edge side primary warning only uploads the abnormal data packet to the cloud through an encrypted channel, and does not execute the sound and light alarm.

3. The industrial-grade data acquisition instrument data processing system according to claim 1, wherein: The single domain analysis results include a first single domain analysis result and a second single domain analysis result; The first single-domain analysis result is obtained by analyzing and processing the operating parameters by the operating parameter processing unit, including a time domain feature extraction result, a state quantification evaluation result and a fault matching result; The time domain feature extraction results include calculating the mean, variance, rate of change and kurtosis coefficient of the operating parameters; The state quantification evaluation result is to input the time domain characteristics into a preset evaluation function to generate a health index of the industrial equipment; The fault matching result is obtained by comparing the time domain feature with a preset state feature library of industrial equipment to obtain a comparison result with the state feature library of industrial equipment; The second single domain analysis result is obtained by analyzing and processing the production environment data through the production environment data processing unit, including spatial feature extraction results, impact quantification results and environmental suitability assessment results; The spatial feature extraction result is obtained by calculating the regional mean, standard deviation, gradient and skewness coefficient of the production environment data; The impact quantification result generates an association rule base based on the correlation analysis of historical data, and calculates the impact of environmental parameters on production indicators; The environmental suitability assessment result is generated by constructing a multidimensional space of environmental parameters, calculating the Euclidean distance between the current environmental state point and the optimal production state point of historical data, and generating an environmental suitability index.

4. The industrial-grade data acquisition instrument data processing system according to claim 3, wherein: Locating the root cause of the abnormality based on the first single-domain analysis result and the second single-domain analysis result specifically includes: When it is detected that the operating parameters or production environment data exceed the dynamic threshold, based on the abnormality occurrence time point t, the multidimensional operating parameters and production environment data within the t±Δt time window are extracted; The dynamic threshold is set based on historical data statistics; Construct a temporal correlation network between operating parameters and production environment data, and calculate the causal strength value of each parameter node through the Bayesian network; Sorting based on the causal strength values ​​to generate an ordered list of possible root causes; A time series trend analysis module is built into the operating parameter processing unit and the production environment data processing unit to construct a sliding time window based on historical operating parameters and production environment data; The window length is configured as a first time threshold range, and a difference prediction is performed on the future trend of the operating parameter and the production environment data; When the deviation between the predicted value and the dynamic threshold exceeds the preset warning coefficient, a trend warning signal is generated, and the time-space distribution characteristics of the operating parameter fluctuations are visualized through a heat map. The heat map color mapping is based on the deviation and includes safety, warning and danger.

5. The industrial-grade data acquisition instrument data processing system according to claim 4, characterized in that: The first single-domain analysis result and the second single-domain analysis result are correlated by a cross-domain correlation analysis unit to generate a cross-domain correlation analysis result including key environmental factors; The cross-domain correlation analysis unit uses a multivariate correlation analysis algorithm to quantify the transmission lag between operating parameters and production environment data and establish a multivariate causal relationship network map; The key environmental factors are ranked based on causal strength, identifying key environmental factors that affect the status of industrial equipment and generating a factor importance ranking report.

6. The industrial-grade data acquisition instrument data processing system according to claim 5, characterized in that: Dynamic threshold adjustment is performed based on the cross-domain correlation analysis results, specifically including: Determine the current operating condition by combining the operating parameters, and establish an operating condition-threshold template mapping table; When the working condition switches, the threshold template of the corresponding working condition is automatically called; The confidence interval of the threshold is updated based on the mean ± 3σ of the statistical properties of historical data; Based on the key environmental factors identified in the cross-domain correlation analysis results, independently adjust the threshold range of the corresponding operating parameters, and the adjustment range does not exceed the first proportional threshold range of the current threshold; The operating parameters and production environment data are re-evaluated based on the threshold range to generate analysis results after the threshold is optimized.

7. The industrial-grade data acquisition instrument data processing system according to claim 1, wherein: The management module includes an SOP management unit; The SOP management unit is a subunit of the management module for performing standard operating procedure optimization. Based on the analysis results and feedback data after the threshold optimization, the dynamic verification and optimization of the SOP are achieved through three-dimensional process deduction and multi-objective optimization algorithm; Multiple sets of SOP optimization plans are generated through multi-objective optimization algorithms, and the plans are optimized based on production efficiency, energy consumption, industrial equipment and life indicators, and automatically synchronized to the workshop execution system.

8. The industrial-grade data acquisition instrument data processing system according to claim 2, wherein: The edge computing node includes mixed precision data storage; High-frequency sampling data is stored with n-bit floating-point precision, and abnormal data packets are automatically upgraded to 1-bit floating-point precision; and reducing transmission bandwidth occupancy through differential coding compression technology, with a compression ratio not lower than a first numerical threshold; And a multi-level early warning mechanism is preset on the edge side.

9. The industrial-grade data acquisition instrument data processing system according to claim 1, characterized in that: The multi-level early warning mechanism includes: According to the change rate of operating parameters or production environment data, it is divided into yellow warning and red warning; When the rate of change of operating parameters or production environment data exceeds the PST gradient threshold, a yellow warning is triggered; When the rate of change of operating parameters or production environment data exceeds the end gradient threshold, a red alert is triggered; When the yellow warning is triggered, the local sound and light alarm is activated. When the red warning is triggered, the local sound and light alarm and remote SMS notification are triggered simultaneously.

10. A data processing method for an industrial data acquisition instrument, applied to an industrial data acquisition instrument data processing system according to any one of claims 1 to 9, characterized in that: Step S1, collecting operating parameters and production environment data of industrial equipment; Step S2: performing single-domain analysis and correlation processing on the operating parameters and production environment data to generate a cross-domain correlation analysis result; generating a threshold range based on the cross-domain correlation analysis result through dynamic threshold adjustment; re-evaluating the operating parameters and production environment data and generating an analysis result after threshold optimization; In step S3, based on the analysis results after threshold optimization, an optimal solution is generated and automatically synchronized to the workshop system, and the edge-side primary warning and edge-side multi-level warning mechanisms are preset.

Images