An AI-based production data deep mining method and system

Abstract

The application discloses an AI-based production data deep mining method and system, relates to the technical field of production data deep mining, and comprises yield estimation analysis, quality fluctuation analysis and equipment fault analysis.The application firstly collects historical yield data through yield estimation analysis, obtains the correlation between production data and yield through analysis, screens effective data, combines current effective data and preset yield to regulate process parameters, obtains preset parameters, secondly mines historical quality data through quality fluctuation analysis, clearly defines the correlation between process parameters and quality, generates a re-regulation scheme, optimizes parameters to stabilize quality, and finally analyzes historical fault data through equipment fault analysis, identifies associated process parameters and fault intervals, monitors equipment parameters to mark potential fault equipment, and adjusts parameters to realize safe regulation.The application is suitable for manufacturing industry, can improve efficiency, stabilize quality and reduce fault risk.

Description

Technical Field

[0001] This invention relates to the field of deep mining technology for production data, and specifically to an AI-based method and system for deep mining of production data. Background Technology

[0002] With the intelligent development of manufacturing, there is an urgent need for in-depth utilization of production data. In order to meet the requirements of efficient and accurate production management, an AI-based method and system for in-depth mining of production data is needed to achieve intelligent optimization of the entire production process.

[0003] Current technologies for production data processing and management primarily rely on human experience and basic statistical analysis tools. In the field of output management, output forecasting often relies on historical averages or simple linear regression models. These methods can only capture the superficial relationship between production data and output, failing to quantify the comprehensive impact of multi-dimensional data such as raw material input and equipment operating power on output. This results in a high error rate in forecasting results, leading to frequent adjustments to production plans and causing problems such as raw material inventory backlog or idle production capacity.

[0004] In terms of quality control, traditional methods rely on quality inspectors to conduct sampling inspections and, based on experience, screen process parameters that may affect quality. However, due to the lack of systematic analysis of historical quality data, it is difficult to establish an accurate correlation model between process parameters and quality fluctuations. When quality anomalies occur, it is often necessary to blindly adjust multiple parameters, resulting in an excessively long average resolution cycle and a high recurrence rate, which increases monitoring costs.

[0005] In the field of equipment maintenance, existing technologies mainly rely on periodic inspections and post-failure repairs. Periodic inspections can lead to over-maintenance, increasing maintenance costs and waste; while post-failure repairs, due to the lack of a correlation between process parameters and fault warnings, result in excessively high production losses from a single downtime and cannot prevent the recurrence of similar faults in advance, further increasing production costs. Summary of the Invention

[0006] To address the aforementioned technical shortcomings, the present invention aims to provide an AI-based method and system for in-depth mining of production data.

[0007] To solve the above technical problems, the present invention adopts the following technical solution: The present invention provides an AI-based method for deep mining of production data, including the following steps: Step 1, output prediction analysis: Collect historical output data, analyze the historical output data to obtain the output correlation of various types of production data, and then obtain various types of effective production data. Collect the values ​​of the current types of effective production data, analyze the values ​​of the current types of effective production data based on the output correlation of various types of production data, obtain the current estimated output, and adjust the production process according to the preset output in the database, thereby obtaining the preset parameters of various production process parameters.

[0008] Step 2, Quality Fluctuation Analysis: Collect historical quality data, analyze the historical quality data to obtain the quality correlation of various production process parameters, generate re-regulation schemes for various production process parameters based on the preset parameters of various production process parameters and the quality correlation of various production process parameters, and carry out re-regulation of production process parameters.

[0009] Step 3: Equipment Fault Analysis: Collect historical fault data, analyze the historical fault data to obtain the fault correlation of various production process parameters, and then obtain the equipment with preset faults. Generate a safety production process control plan for each equipment with preset faults and carry out safety production control.

[0010] Preferably, the production process control is carried out as follows: the current estimated output is divided by the preset output in the database to obtain the current output increase rate, and the correction rate of each basic production data corresponding to each output increase rate is obtained from the database to obtain the current basic production data correction rate.

[0011] Historical data of various production process parameters and various effective historical production data are obtained from the database. Pearson correlation coefficients are calculated to obtain the production correlation between various production process parameters and various effective production data. If the current output growth rate is less than 1, the effective production data with the largest production correlation is selected as the associated production data for various production process parameters. This yields various associated production data for various production process parameters. After summarizing, the various production process parameters corresponding to various effective production data are obtained. If the current output growth rate is greater than 1, the effective production data with the smallest production correlation is selected as the associated production data for various production process parameters. This yields various associated production data for various production process parameters. After summarizing, the various production process parameters corresponding to various effective production data are obtained.

[0012] Multiply the current basic production data correction rate by the output correlation of various types of effective production data to obtain the correction rate of various types of effective production data. Obtain the production correction factor corresponding to each production correlation from the database to obtain the production correction factor of various types of production process parameters corresponding to various types of effective production data. Multiply the correction rate of various types of effective production data by the production correction factor of various types of production process parameters to obtain the production correction rate of various types of production process parameters.

[0013] Collect current production process parameters, multiply the current production process parameters by the corresponding production correction rate to obtain the preset parameters for each production process parameter, and set the values ​​of each production process parameter as the preset parameters.

[0014] On the other hand, the present invention provides an AI-based production data deep mining system, including the following modules: a production forecasting and analysis module, used to collect historical production data, analyze the historical production data, obtain the production correlation of various types of production data, and then obtain various types of effective production data; collect the values ​​of current types of effective production data; analyze the values ​​of current types of effective production data based on the production correlation of various types of production data to obtain the current estimated production; and adjust the production process according to the preset production in the database to obtain the preset parameters of various production process parameters.

[0015] The quality fluctuation analysis module is used to collect historical quality data, analyze the historical quality data, obtain the quality correlation of various production process parameters, generate re-regulation schemes for various production process parameters based on the preset parameters and the quality correlation of various production process parameters, and carry out re-regulation of production process parameters.

[0016] The equipment fault analysis module is used to collect historical fault data, analyze the historical fault data, obtain the fault correlation of various production process parameters, and then obtain the equipment with preset faults. It generates a safety production process control plan for each equipment with preset faults and carries out safety production control.

[0017] The beneficial effects of this invention are as follows: 1. This invention first collects historical production data through production forecasting analysis, analyzes the correlation between production data and output, filters effective data, and combines current effective data with preset production control process parameters to obtain preset parameters. Secondly, it mines historical quality data through quality fluctuation analysis, clarifies the correlation between process parameters and quality, generates a re-control scheme, optimizes parameters to stabilize quality, and finally analyzes equipment failure data through equipment failure analysis, identifies related process parameters and failure ranges, monitors equipment parameters to mark potentially faulty equipment, and adjusts parameters to achieve safe control. This invention is applicable to the manufacturing industry and can improve efficiency, stabilize quality, and reduce failure risks.

[0018] 2. This invention can monitor production status in real time, quickly identify abnormal situations, and, by establishing accurate predictive models, anticipate fluctuations in output and quality, as well as equipment failures. Compared with traditional methods, this invention has higher accuracy and timeliness, and can automatically adapt to changes in the production environment, continuously optimizing prediction results. In practical applications, it can effectively help enterprises rationally arrange production plans, reduce raw material waste, lower equipment downtime risks, and improve production efficiency and product quality. In various manufacturing industries, such as manufacturing and chemical industries, this invention can achieve intelligent production management, enhance their market competitiveness, and reduce operating costs while improving production efficiency. Attached Figure Description

[0019] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0020] Figure 1 This is a schematic diagram of the implementation steps of the method of the present invention.

[0021] Figure 2 This is a schematic diagram of the system structure connection of the present invention. Detailed Implementation

[0022] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0023] according to Figure 1 As shown, this invention provides an AI-based method for deep mining of production data, including the following steps: Step 1, Production Prediction Analysis: Collect historical production data, analyze the historical production data to obtain the production correlation of various types of production data, and then obtain various types of effective production data. Collect the values ​​of the current types of effective production data, analyze the values ​​of the current types of effective production data based on the production correlation of various types of production data, obtain the current estimated production, and adjust the production process according to the preset production in the database to obtain the preset parameters of various production process parameters.

[0024] In one specific embodiment, the historical production data is collected through the following process: the historical production data is collected through the production logs in the database. The historical production data includes the output of each production run and various types of production data, including but not limited to the amount of raw materials input, equipment operating power, production duration, and operator skill level.

[0025] In one specific embodiment, the analysis of historical production data is carried out as follows: the correlation coefficient between various types of production data and output is calculated using the Pearson correlation coefficient to obtain the output correlation of various types of production data; the effective interval of output correlation is obtained from the database; and various types of production data whose output correlation belongs to the effective interval of output correlation are recorded as various types of effective production data.

[0026] In one specific embodiment, the values ​​of the current various types of effective production data are analyzed. The specific analysis process is as follows: the output of each production and the values ​​of various types of effective production data are obtained from the historical output data. The average output and the average value of various types of effective production data are calculated. The maximum value is obtained from the output of each production and the values ​​of various types of effective production data, and recorded as the output threshold and the threshold of various types of effective production data.

[0027] Divide the current values ​​of various types of effective production data by the mean and threshold of each type of effective production data to obtain the current mean rate and current threshold rate of each type of effective production data. The current mean rate and current threshold rate of each type of effective production data are weighted according to the corresponding output correlation and weighted to obtain the current mean rate and current threshold rate of output. Multiply the mean output by the mean rate of current output to obtain the first type of output preset value. Multiply the output threshold by the current threshold rate of output to obtain the second type of output preset value. According to the weight factors of the first type of output preset value and the weight factors of the second type of output preset value preset by the staff, a weighted calculation is performed to obtain the current estimated output.

[0028] In one specific embodiment, the production process control is carried out as follows: the current estimated output is divided by the preset output in the database to obtain the current output increase rate; the correction rate of each basic production data corresponding to each output increase rate is obtained from the database to obtain the current basic production data correction rate.

[0029] Historical data of various production process parameters and various effective historical production data are obtained from the database. Pearson correlation coefficients are calculated to obtain the production correlation between various production process parameters and various effective production data. If the current output growth rate is less than 1, the effective production data with the largest production correlation is selected as the associated production data for various production process parameters. This yields various associated production data for various production process parameters. After summarizing, the various production process parameters corresponding to various effective production data are obtained. If the current output growth rate is greater than 1, the effective production data with the smallest production correlation is selected as the associated production data for various production process parameters. This yields various associated production data for various production process parameters. After summarizing, the various production process parameters corresponding to various effective production data are obtained.

[0030] Multiply the current basic production data correction rate by the output correlation of various types of effective production data to obtain the correction rate of various types of effective production data. Obtain the production correction factor corresponding to each production correlation from the database to obtain the production correction factor of various types of production process parameters corresponding to various types of effective production data. Multiply the correction rate of various types of effective production data by the production correction factor of various types of production process parameters to obtain the production correction rate of various types of production process parameters.

[0031] Collect current production process parameters, multiply the current production process parameters by the corresponding production correction rate to obtain the preset parameters for each production process parameter, and set the values ​​of each production process parameter as the preset parameters.

[0032] Step 2, Quality Fluctuation Analysis: Collect historical quality data, analyze the historical quality data to obtain the quality correlation of various production process parameters, generate re-regulation schemes for various production process parameters based on the preset parameters of various production process parameters and the quality correlation of various production process parameters, and carry out re-regulation of production process parameters.

[0033] In one specific embodiment, the historical quality data is collected through the following process: the data is collected through the detection logs, and the historical quality data includes the low quality rate of each historical production and the values ​​of various production process parameters.

[0034] In one specific embodiment, the analysis of historical quality data is carried out as follows: the correlation coefficient between various production process parameters and the low quality rate is calculated using the Pearson correlation coefficient to obtain the low quality correlation of various production process parameters; a quality influence factor model is constructed using a decision tree algorithm to obtain the quality fluctuation value of various production process parameters on the quality dimension; the total quality fluctuation value is statistically obtained; and the quality fluctuation value of various production process parameters on the quality dimension is divided by the total quality fluctuation value to obtain the contribution rate of various production process parameters on the quality dimension.

[0035] In one specific embodiment, the process of readjusting the production process parameters is as follows: historical production with a low quality rate greater than a preset low quality rate is recorded as risky production, and the values ​​of various production process parameters for each risky production are obtained. The average value of each production process parameter is calculated, and the maximum value of each production process parameter for each risky production is selected as the threshold value to obtain the threshold values ​​of various production process parameters.

[0036] By dividing the preset parameters of various production process parameters by the mean and threshold values ​​of each parameter, the risk threshold rate and the risk mean rate of each parameter are obtained. The preset correction rate of each parameter is then calculated by weighting the risk threshold rate and the risk mean rate. The risk correction factor corresponding to each contribution rate is retrieved from the database, and the risk correction factor corresponding to the contribution rate of each production process parameter to the quality dimension is obtained. The preset correction rate of each parameter is multiplied by the corresponding risk correction factor to obtain the correction rate of each parameter. Finally, the correction rate of each parameter is multiplied by the preset parameter to obtain the re-regulation value of each parameter. The values ​​of each parameter are then set as the re-regulation values.

[0037] Step 3: Equipment Fault Analysis: Collect historical fault data, analyze the historical fault data to obtain the fault correlation of various production process parameters, and then obtain the equipment with preset faults. Generate a safety production process control plan for each equipment with preset faults and carry out safety production control.

[0038] In one specific embodiment, the historical fault data collection process is as follows: the historical fault data is obtained from the fault maintenance log in the database, including the fault type corresponding to each fault and the values ​​of various production process parameters at the time of the fault.

[0039] In one specific embodiment, the analysis of historical fault data is described, and the specific analysis process is as follows: the support and confidence of various production process parameters and fault types are calculated using an association rule mining algorithm. Various production process parameters with preset support and confidence greater than the preset confidence are identified as various fault association parameters.

[0040] The fault values ​​corresponding to various fault-related parameters are summarized to obtain the historical fault ranges of various process parameters. The real-time values ​​of various process parameters of each equipment are monitored. When the parameter value of a certain type of fault-related parameter of a certain equipment belongs to the corresponding historical fault range and the duration is longer than the preset duration, the equipment is marked as a preset fault equipment. The fault-related parameter of this type is recorded as a risk-related parameter, thereby obtaining each preset fault equipment.

[0041] In one specific embodiment, the safety production control is specifically generated as follows: obtain the values ​​of various risk-related parameters of each preset fault equipment for each safety production from the database, calculate the average value to obtain the standard parameter values ​​of various risk-related parameters of each preset fault equipment, and set the values ​​of various risk-related parameters of each preset fault equipment to the corresponding standard parameter values.

[0042] according to Figure 2 As shown, the present invention provides an AI-based deep mining system for production data, including the following modules: a production forecasting and analysis module, a quality fluctuation analysis module, an equipment failure analysis module, and a database.

[0043] The quality fluctuation analysis module is connected to the production forecast analysis module and the equipment failure analysis module, respectively. The production forecast analysis module, the quality fluctuation analysis module, and the equipment failure analysis module are all connected to the database.

[0044] The production forecasting and analysis module is used to collect historical production data, analyze the historical production data, obtain the production correlation of various production data, and then obtain various effective production data. It collects the values ​​of the current effective production data, analyzes the values ​​of the current effective production data based on the production correlation of various production data, obtains the current estimated production, and adjusts the production process according to the preset production in the database, thereby obtaining the preset parameters of various production process parameters. The quality fluctuation analysis module is used to collect historical quality data, analyze the historical quality data, obtain the quality correlation of various production process parameters, generate re-regulation schemes for various production process parameters based on the preset parameters and the quality correlation of various production process parameters, and carry out re-regulation of production process parameters.

[0045] The equipment fault analysis module is used to collect historical fault data, analyze the historical fault data, obtain the fault correlation of various production process parameters, and then obtain the equipment with preset faults. It generates a safety production process control plan for each equipment with preset faults and carries out safety production control.

[0046] The database is used to store preset output, production logs, effective range of output correlation, correction rates of basic production data corresponding to each output increase multiple, historical data of various production process parameters, various effective historical production data, production correction factors corresponding to each production correlation, risk correction factors corresponding to each contribution rate, fault maintenance logs, and values ​​of various risk correlation parameters for each safe production of each preset fault equipment.

[0047] The above description is merely an example and illustration of the concept of the present invention. Those skilled in the art can make various modifications or additions to the specific embodiments described or use similar methods to replace them, as long as they do not deviate from the concept of the invention or exceed the scope defined in this specification, they should all fall within the protection scope of the present invention.

Claims

1. A method for deep mining of production data based on AI, characterized in that, Includes the following steps: Step 1: Production Forecast Analysis: Collect historical production data, analyze the historical production data to obtain the production correlation of various types of production data, and then obtain various types of effective production data. Collect the values ​​of the current types of effective production data, analyze the values ​​of the current types of effective production data based on the production correlation of various types of production data, obtain the current estimated production, and adjust the production process according to the preset production in the database to obtain the preset parameters of various production process parameters. Step 2, Quality Fluctuation Analysis: Collect historical quality data, analyze the historical quality data to obtain the quality correlation of various production process parameters, generate re-regulation schemes for various production process parameters based on the preset parameters of various production process parameters and the quality correlation of various production process parameters, and carry out re-regulation of production process parameters. Step 3: Equipment Fault Analysis: Collect historical fault data, analyze the historical fault data to obtain the fault correlation of various production process parameters, and then obtain the equipment with preset faults. Generate a safety production process control plan for each equipment with preset faults and carry out safety production control. The analysis of historical production data is carried out as follows: Historical production data includes the output of each production and various types of production data. The correlation coefficient between various types of production data and output is calculated using the Pearson correlation coefficient to obtain the output correlation of various types of production data. The effective interval of output correlation is obtained from the database, and various types of production data whose output correlation belongs to the effective interval of output correlation are recorded as various types of effective production data. The various types of production data include raw material input, equipment operating power, production time, and operator skill level; The analysis of historical quality data is as follows: Historical quality data includes the low quality rate of each historical production and the values ​​of various production process parameters. The correlation coefficient between various production process parameters and the low quality rate is calculated using the Pearson correlation coefficient to obtain the low quality correlation of various production process parameters. A quality impact factor model is constructed using a decision tree algorithm to obtain the quality fluctuation value of various production process parameters on the quality dimension. The total quality fluctuation value is statistically obtained. The contribution rate of various production process parameters to the quality dimension is obtained by dividing the quality fluctuation value of various production process parameters on the quality dimension by the total quality fluctuation value. The analysis of historical fault data is as follows: Historical fault data includes the fault type corresponding to each fault and the values ​​of various production process parameters when the fault occurred. The support and confidence of various production process parameters with fault type are calculated by using association rule mining algorithm. Various production process parameters with preset support and confidence greater than preset confidence are identified as various fault association parameters. The fault values ​​corresponding to various fault-related parameters are summarized to obtain the historical fault ranges of various process parameters. The real-time values ​​of various process parameters of each equipment are monitored. When the parameter value of a certain type of fault-related parameter of a certain equipment belongs to the corresponding historical fault range and the duration is longer than the preset duration, the equipment is marked as a preset fault equipment. The fault-related parameter of this type is recorded as a risk-related parameter, thereby obtaining each preset fault equipment.

2. The AI-based deep mining method for production data according to claim 1, characterized in that, The values ​​of various current valid production data are analyzed, and the specific analysis process is as follows: The production output and various types of effective production data are obtained from historical production data. The average output and the average of various types of effective production data are calculated. The maximum value of the production output and various types of effective production data are obtained from the production output and various types of effective production data, and recorded as the production threshold and various types of effective production data threshold. Divide the current values ​​of various types of effective production data by the mean and threshold of each type of effective production data to obtain the current mean rate and current threshold rate of each type of effective production data. The current mean rate and current threshold rate of each type of effective production data are weighted according to the corresponding output correlation and weighted to obtain the current mean rate and current threshold rate of output. Multiply the mean output by the mean rate of current output to obtain the first type of output preset value. Multiply the output threshold by the current threshold rate of output to obtain the second type of output preset value. According to the weight factors of the first type of output preset value and the weight factors of the second type of output preset value preset by the staff, a weighted calculation is performed to obtain the current estimated output.

3. The AI-based deep mining method for production data according to claim 2, characterized in that, The specific process for adjusting the production process is as follows: Divide the current estimated output by the preset output in the database to obtain the current output growth rate. Obtain the basic production data correction rate corresponding to each output growth rate from the database to obtain the current basic production data correction rate. Historical data of various production process parameters and various effective historical production data are obtained from the database. Pearson correlation coefficient is calculated to obtain the production correlation between various production process parameters and various effective production data. If the current output growth rate is less than 1, the various effective production data with the largest production correlation are selected as the various associated production data of various production process parameters. In this way, the various associated production data of various production process parameters are obtained. After summarizing, the various production process parameters corresponding to various effective production data are obtained. If the current output growth rate is greater than 1, the various effective production data with the smallest production correlation are selected as the various associated production data of various production process parameters. In this way, the various associated production data of various production process parameters are obtained. After summarizing, the various production process parameters corresponding to various effective production data are obtained. Multiply the current basic production data correction rate by the output correlation of various types of effective production data to obtain the correction rate of various types of effective production data. Obtain the production correction factor corresponding to each production correlation from the database to obtain the production correction factor of various types of production process parameters corresponding to various types of effective production data. Multiply the correction rate of various types of effective production data by the production correction factor of various types of production process parameters to obtain the production correction rate of various types of production process parameters. Collect current production process parameters, multiply the current production process parameters by the corresponding production correction rate to obtain the preset parameters for each production process parameter, and set the values ​​of each production process parameter as the preset parameters.

4. The AI-based deep mining method for production data according to claim 1, characterized in that, The specific adjustment process for re-adjusting production process parameters is as follows: Historical production with a low quality rate greater than the preset low quality rate is recorded as risky production. The values ​​of various production process parameters for each risky production are obtained. The average value of each production process parameter is calculated. The maximum value of each production process parameter for each risky production is selected as the threshold value to obtain the threshold values ​​of each production process parameter. By dividing the preset parameters of various production process parameters by the mean and threshold values ​​of each parameter, the risk threshold rate and the risk mean rate of each parameter are obtained. The preset correction rate of each parameter is then calculated by weighting the risk threshold rate and the risk mean rate. The risk correction factor corresponding to each contribution rate is retrieved from the database, and the risk correction factor corresponding to the contribution rate of each production process parameter to the quality dimension is obtained. The preset correction rate of each parameter is multiplied by the corresponding risk correction factor to obtain the correction rate of each parameter. Finally, the correction rate of each parameter is multiplied by the preset parameter to obtain the re-regulation value of each parameter. The values ​​of each parameter are then set as the re-regulation values.

5. The AI-based deep mining method for production data according to claim 4, characterized in that, The analysis of historical fault data is carried out in the following specific process: Historical fault data includes the fault type corresponding to each fault and the values ​​of various production process parameters at the time of the fault. The association rule mining algorithm is used to calculate the support and confidence of various production process parameters and fault types. Various production process parameters with support greater than the preset support and confidence greater than the preset confidence are identified as fault association parameters. The fault values ​​corresponding to various fault-related parameters are summarized to obtain the historical fault ranges of various process parameters. The real-time values ​​of various process parameters of each equipment are monitored. When the parameter value of a certain type of fault-related parameter of a certain equipment belongs to the corresponding historical fault range and the duration is longer than the preset duration, the equipment is marked as a preset fault equipment. The fault-related parameter of this type is recorded as a risk-related parameter, thereby obtaining each preset fault equipment.

6. The AI-based deep mining method for production data according to claim 5, characterized in that, The aforementioned safety production control measures are specifically generated as follows: The database is used to obtain the values ​​of various risk-related parameters for each preset fault device in each safe production operation. After the average value is calculated, the standard parameter values ​​of various risk-related parameters for each preset fault device are obtained. The values ​​of various risk-related parameters for each preset fault device are then set as the corresponding standard parameter values.

7. A data mining system applying the AI-based deep mining method for production data as described in any one of claims 1-6, characterized in that, Includes the following modules: The production forecasting and analysis module is used to collect historical production data, analyze the historical production data, obtain the production correlation of various production data, and then obtain various effective production data. It collects the values ​​of the current effective production data, analyzes the values ​​of the current effective production data based on the production correlation of various production data, obtains the current estimated production, and adjusts the production process according to the preset production in the database, thereby obtaining the preset parameters of various production process parameters. The quality fluctuation analysis module is used to collect historical quality data, analyze the historical quality data, obtain the quality correlation of various production process parameters, generate re-regulation schemes for various production process parameters based on the preset parameters and the quality correlation of various production process parameters, and carry out re-regulation of production process parameters. The equipment fault analysis module is used to collect historical fault data, analyze the historical fault data, obtain the fault correlation of various production process parameters, and then obtain the equipment with preset faults. It generates a safety production process control plan for each equipment with preset faults and carries out safety production control.

Images