Mica material production optimization method and system based on big data analysis

Abstract

The invention discloses a mica material production optimization method and system based on big data analysis, and relates to the field of mica production technologies, and the method comprises the steps: obtaining a mica processing product type and a historical mica processing product type; acquiring initial equipment parameters of all mica processing equipment; acquiring mica processing data and mica historical processing data according to multiple types of sensors preset in mica processing equipment; screening out core process parameters in the initial equipment parameters; calculating a core process standard value corresponding to the core process parameter; calculating the loss rate of the mica raw material in the mica processing process; screening and outputting local correction factors and global correction factors; and correcting the initial equipment parameters to obtain optimal equipment parameters. The mica processing technology is optimized, and the utilization rate of mica processing raw materials and the quality of mica processing products are improved.

Description

Technical Field

[0001] The embodiments of the present application relate to the field of mica production technology, and in particular, to a mica material production optimization method and system based on big data analysis. Background Art

[0002] Mica has insulating and heat-insulating properties, good chemical stability, and resistance to strong acids, strong alkalis and pressure. Therefore, it is an important raw material for manufacturing electrical equipment. With the advancement of social science and technology, mica materials are also widely used in other fields, such as the building materials industry, fire protection industry, etc. Since mica materials play an indispensable role in many industries, once the produced mica materials do not meet the relevant industry quality standards, it will have a significant economic impact on the relevant industries. Therefore, it is necessary to optimize the mica material production process and improve and strictly control the quality of mica materials.

[0003] The existing technology for optimizing the mica material production process mainly involves updating and optimizing the mica processing equipment. Although it can improve the production quality of mica materials to a certain extent, equipment upgrades require a lot of money, and the mica material processing process is complicated, and there are many factors that affect the quality of mica materials. Therefore, a method is needed to optimize the mica material production process from multiple dimensions to improve the mica production quality. Summary of the invention

[0004] The embodiments of the present application provide a mica material production optimization method and system based on big data analysis, which are used to solve the problem of insufficient quality of mica materials produced by existing mica production processes.

[0005] To achieve the above objectives, the embodiments of the present application adopt the following technical solutions:

[0006] In a first aspect, a mica material production optimization method based on big data analysis is provided, the method comprising:

[0007] Obtain the mica processing product type of the current material order through the mica material order system, and select the samples to be processed of the same type as the mica processing product type;

[0008] Retrieving historical processing data and initial equipment parameters of the same type as the mica processing product type from a preset mica processing database, wherein the initial equipment parameters are equipment parameters of all mica processing equipment in the mica processing technology corresponding to the mica processing product type;

[0009] Based on the initial equipment parameters, the sample to be processed is processed by the mica processing equipment, and the mica processing data of the sample to be processed during the processing is collected by using multiple types of sensors preset in all the mica processing equipment, and the sample quality parameters of the sample to be processed after processing are detected;

[0010] Classifying and analyzing the timing characteristics of the mica processing data, and screening out the core process parameters in the initial equipment parameters according to the analysis results of the timing characteristics analysis;

[0011] Calculate the average value of the mica historical processing data based on the core process parameters, and use the average value as the core process standard value corresponding to the core process parameters;

[0012] Combining the core process standard value with the mica processing data, calculating the mica raw material loss rate during the mica processing process;

[0013] The core process parameters are selected by calculating the correlation coefficient between the mica raw material loss rate and the core process parameters, and the selected core process parameters are output as local correction factors;

[0014] Obtaining mica product quality parameters corresponding to the mica processing product type through a preset database, calculating an influence degree data set between the initial equipment parameters and the mica product quality parameters, prioritizing the initial equipment parameters based on the influence degree data set, and outputting the initial equipment parameters that have completed the priority grading as a global correction factor;

[0015] The initial equipment parameters are corrected in combination with the local correction factor and the global correction factor to obtain optimal equipment parameters.

[0016] Optionally, the step of classifying and analyzing the timing characteristics of the mica processing data, and screening out the core process parameters in the initial equipment parameters according to the analysis results of the timing characteristics analysis comprises the following steps:

[0017] Feature classification of the mica processing data is performed according to the sensor types of the multi-type sensors to obtain mica weight change data, mica temperature change data, and mica pH change data;

[0018] An upper limit of the total number of inflection points is set according to the total number of processing steps included in the mica processing process, the upper limit of the total number of inflection points is used as an algorithm constraint, and an inflection point detection algorithm is used to perform inflection point detection on the mica weight change data to obtain weight change inflection points distributed in the mica weight change data;

[0019] Dividing the mica processing technology into process nodes according to the weight change inflection point to obtain a plurality of mica processing sub-processes within the mica processing technology;

[0020] Extracting the maximum temperature and the minimum temperature of the mica temperature change data in each mica processing sub-process, calculating the difference between the maximum temperature and the minimum temperature, and obtaining the temperature difference;

[0021] Extracting the maximum pH value and the minimum pH value of the mica pH change data in each mica processing sub-process, calculating the difference between the maximum pH value and the minimum pH value, and obtaining the pH difference;

[0022] If the temperature difference corresponding to the mica processing sub-process is greater than a preset temperature difference threshold, analyzing whether the pH difference corresponding to the mica processing sub-process is greater than a preset pH difference threshold;

[0023] If the pH difference corresponding to the mica processing sub-process is greater than the pH difference threshold, it is determined that the initial equipment parameters corresponding to the mica processing sub-process are mica purification parameters;

[0024] If the pH difference is less than or equal to the pH difference threshold, determining that the initial equipment parameter corresponding to the mica processing sub-process is a mica grinding parameter;

[0025] The mica purification parameters and the mica grinding parameters are integrated to obtain core process parameters.

[0026] Optionally, calculating the average value of the mica historical processing data according to the core process parameters, and using the average value as the core process standard value corresponding to the core process parameters comprises the following steps:

[0027] Extracting historical purification weight change data from the historical processing data corresponding to the mica purification parameters to obtain historical purification weight change data, calculating the purification weight average value of the historical purification weight change data, and using the purification weight average value as the purification process standard value;

[0028] Extracting historical grinding weight change data from the historical processing data corresponding to the mica grinding parameters to obtain historical grinding weight change data, calculating the grinding weight average value of the historical grinding weight change data, and using the grinding weight average value as a grinding process standard value;

[0029] The standard values ​​of the purification process and the standard values ​​of the grinding process are integrated to obtain the standard values ​​of the core process parameters.

[0030] Optionally, calculating the correlation coefficient between the mica raw material loss rate and the core process parameters, screening the core process parameters according to the correlation coefficient, and outputting the screened core process parameters as local correction factors comprises the following steps:

[0031] Calculating a core parameter correlation coefficient set between the mica core process parameter and the mica raw material loss rate, screening out a core parameter correlation coefficient greater than a preset core parameter correlation coefficient threshold value in the core parameter correlation coefficient set, and outputting the core process parameter corresponding to the core parameter correlation coefficient as a local correction factor;

[0032] According to multiple types of sensors preset in the mica processing plant, mica processing environment data is obtained, wherein the mica processing environment data includes mica processing environment temperature data and mica processing environment humidity data;

[0033] Establishing a multiple linear regression equation with the mica raw material loss rate as the dependent variable, inputting the mica processing environment temperature data and the mica processing environment humidity data as independent variables into the multiple linear regression equation, and calculating the environment temperature correlation coefficient and the environment humidity correlation coefficient;

[0034] If the environmental temperature correlation coefficient is greater than a preset environmental correlation coefficient threshold, the mica processing environmental temperature data is output as a local correction factor;

[0035] If the environmental humidity correlation coefficient is greater than the environmental correlation coefficient threshold, the mica processing environmental humidity data is output as a local correction factor.

[0036] Optionally, calculating a core parameter correlation coefficient set between the mica core process parameter and the mica raw material loss rate, screening out core parameter correlation coefficients in the core parameter correlation coefficient set that are greater than a preset core parameter correlation coefficient threshold, and outputting the core process parameter corresponding to the core parameter correlation coefficient as a local correction factor comprises the following steps:

[0037] A correlation coefficient model is established with the mica raw material loss rate as an independent variable;

[0038] The vibration frequency, grinding time and grinding method included in the mica grinding parameters are respectively input into the correlation coefficient model as independent variables to obtain a grinding correlation coefficient set;

[0039] Screening out the grinding correlation coefficients in the grinding correlation coefficient set that are greater than a preset correlation coefficient threshold, and outputting the mica grinding parameters corresponding to the grinding correlation coefficients as local correction factors;

[0040] The solution ratio, purification time and purification mesh parameters included in the mica purification parameters are respectively input into the correlation coefficient model as independent variables to obtain a purification correlation coefficient set;

[0041] The purification correlation coefficients in the purification correlation coefficient set that are greater than a preset correlation coefficient threshold are screened out, and the mica purification parameters corresponding to the purification correlation coefficients are output as local correction factors.

[0042] Optionally, obtaining mica product quality parameters through a preset database, calculating an influence degree data set between the initial equipment parameters and the mica product quality parameters, prioritizing the initial equipment parameters based on the influence degree data set, and outputting the initial equipment parameters that have completed the priority grading as a global correction factor comprises the following steps:

[0043] Obtaining mica product quality parameters corresponding to the mica processed product type through a preset database;

[0044] Establishing a mica regression model with the mica product quality parameter as the dependent variable and the initial equipment parameter as the independent variable;

[0045] The mica equipment parameters, mica raw material data, and mica ingredient ratio included in the initial equipment parameters are respectively input as independent variables into the mica regression model to obtain influence degree data, and the influence degree data are integrated to obtain an influence degree data set;

[0046] Outputting the initial device parameter corresponding to the largest impact degree data in the impact degree data set as a global correction factor of the first priority;

[0047] Outputting the initial device parameter corresponding to the second largest impact degree data in the impact degree data set as a global correction factor of the second priority;

[0048] The initial device parameters corresponding to the third largest impact data in the impact data set are output as global correction factors of the third priority, the first priority is greater than the second priority, and the second priority is greater than the third priority.

[0049] Optionally, the initial device parameters are corrected in combination with the local correction factor and the global correction factor to obtain the optimal device parameters, including the following steps:

[0050] After analyzing the local correction factors, locally correct the core process parameters and mica processing environment data included in the primary equipment parameters;

[0051] The initial equipment parameters that have been partially corrected are globally corrected according to the global correction factors to obtain optimal equipment parameters.

[0052] Optionally, performing global correction on the initial device parameters that have undergone local correction according to the global correction factor to obtain the optimal device parameters comprises the following steps:

[0053] Acquire mica product quality standard parameters corresponding to the mica processed product type according to the preset database;

[0054] Calculating the standard deviation between the mica product quality parameter and the mica product quality standard parameter;

[0055] Optimizing the initial device parameters corresponding to the global correction factor of the first priority to obtain first device parameters;

[0056] Calculating a first difference between a mica product quality parameter obtained by processing according to the first equipment parameter and the mica production standard quality parameter, and if the first difference is less than a preset difference threshold, outputting the first equipment parameter as an optimal equipment parameter;

[0057] If the first difference is greater than or equal to the difference threshold, optimizing the initial device parameters corresponding to the global correction factor of the second priority to obtain second device parameters;

[0058] Calculating a second difference between a mica product quality parameter obtained by processing according to the second equipment parameter and the mica production standard quality parameter, and if the second difference is less than the difference threshold, outputting the second mica equipment parameter as an optimal equipment parameter;

[0059] If the second difference is greater than or equal to the difference threshold, the initial device parameters corresponding to the global correction factors of the third priority are optimized to obtain third device parameters, and the third device parameters are output as optimal device parameters.

[0060] In the second aspect, a mica material production optimization system based on big data analysis includes:

[0061] a memory configured to store instructions; and

[0062] A processor is configured to call the instructions from the memory and to implement the mica material production optimization method based on big data analysis described in any one of the first aspects when executing the instructions.

[0063] In a third aspect, a computer-readable storage medium is provided, on which instructions are stored, and when the instructions are executed by a processor, the processor is configured to execute the mica material production optimization method based on big data analysis according to any one of the first aspects.

[0064] Through the above technical scheme, the mica material order system is used to obtain the mica processing product type and the historical mica processing product type, the initial equipment parameters are obtained according to the mica processing product type, and the mica processing data and the mica processing historical data in the mica processing process are collected by multi-type sensors. The mica processing data is classified after time series feature extraction, and the inflection point detection algorithm is used to perform inflection point detection on the mica weight change data obtained after classification. The mica processing process is divided into process nodes using the obtained weight change inflection point, and the standard deviation is calculated according to the division result. The core process parameters in the initial equipment parameters are screened out according to the standard deviation, and the core process standard values ​​are calculated according to the core process parameters. The mica raw material loss rate is obtained according to the core process standard values ​​and the mica processing data. The local correction factors are screened out according to the correlation coefficient between the mica raw material loss rate and the core process standard value. The global correction factors are screened out according to the influence degree data between the initial equipment parameters and the mica product quality parameters, and the equipment parameters are optimized in combination with the local correction factors and the global correction factors to obtain the optimal equipment parameters.

[0065] As described above, the present invention obtains factors that affect the raw material loss rate and the quality of mica processed products from a multi-dimensional perspective, performs local and global optimization on the primary equipment parameters, and realizes comprehensive optimization of the primary equipment parameters. In summary, the present application comprehensively considers a variety of factors that affect the raw material loss rate and the quality of mica processed products, and also considers the influence of temperature and humidity in the environment, which greatly improves the utilization rate of mica raw materials and the quality of mica processed products, and reduces the production cost of mica.

[0066] Other features and advantages of the embodiments of the present application will be described in detail in the subsequent specific implementation section. BRIEF DESCRIPTION OF THE DRAWINGS

[0067] Figure 1 A schematic diagram of a process for optimizing mica material production based on big data analysis provided in an embodiment of the present application;

[0068] Figure 2 A schematic diagram of a process for screening out core process parameters in initial equipment parameters provided in an embodiment of the present application;

[0069] Figure 3 A flowchart of a method for globally correcting initial device parameters provided in an embodiment of the present application. DETAILED DESCRIPTION

[0070] In order to make the purpose, technical scheme and advantages of the embodiments of the present application clearer, the technical scheme in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. It should be understood that the specific implementation methods described herein are only used to illustrate and explain the embodiments of the present application, and are not used to limit the embodiments of the present application. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of this application.

[0071] It should be noted that if the embodiments of the present application involve directional indications (such as up, down, left, right, front, back...), the directional indications are only used to explain the relative position relationship, movement status, etc. between the components under a certain specific posture (as shown in the accompanying drawings). If the specific posture changes, the directional indication will also change accordingly.

[0072] In addition, if there are descriptions involving "first", "second", etc. in the embodiments of the present application, the descriptions of "first", "second", etc. are only used for descriptive purposes and cannot be understood as indicating or suggesting their relative importance or implicitly indicating the number of technical features indicated. Therefore, the features defined as "first" and "second" may explicitly or implicitly include at least one of the features. In addition, the technical solutions between the various embodiments can be combined with each other, but they must be based on the ability of ordinary technicians in the field to implement them. When the combination of technical solutions is contradictory or cannot be implemented, it should be deemed that such combination of technical solutions does not exist and is not within the scope of protection required by this application.

[0073] Figure 1 A flow chart of a mica material production optimization method based on big data analysis according to an embodiment of the present application is schematically shown. Figure 2 A schematic diagram of a process flow of a method for screening out core process parameters in initial equipment parameters according to an embodiment of the present application is shown schematically. Figure 3 A schematic diagram of a process of a method for globally correcting initial device parameters according to an embodiment of the present application is schematically shown.

[0074] like Figure 1 As shown, the embodiment of the present application provides a method for optimizing mica material production based on big data analysis, which may include the following steps:

[0075] S101. Obtain the mica processing product type of the current material order through the mica material order system, and select samples to be processed of the same type as the mica processing product type.

[0076] In this embodiment, the mica material order system can store and update the orders of customers cooperating with the mica factory in real time. The orders can display the types of mica processed products required by the customers, such as mica sheets, mica powder, mica plates, and mica special parts. The mica material order system can query the types of mica processed products of the current material order, and select the same type of processing samples according to the mica processed product types. Processing the samples first can effectively reduce the waste of mica raw materials and save costs. In addition, the mica material order system can also query the historical mica processed product types of the historical material orders of the mica processing factory in the past.

[0077] S102. Retrieve historical processing data and initial equipment parameters of the same type as the mica processing product type from a preset mica processing database. The initial equipment parameters are equipment parameters of all mica processing equipment in the mica processing technology corresponding to the mica processing product type.

[0078] In this embodiment, the initial equipment parameters include core process parameters, mica equipment parameters, mica raw material data and mica proportion. The core process parameters refer to mica purification parameters and mica grinding parameters. The mica equipment parameters refer to the processing equipment parameters used in the mica processing process, such as the stroke and frequency of the jig for coarse screening of mica. The mica raw material data include the type of mica ore (such as biotite, phlogopite, muscovite, etc.), the density of mica (for example, the density of muscovite is 2.7 g / cm 3 -2.9g / cm 3 ) etc. The mica batching ratio refers to the ratio of mica ore and other raw materials required for processing mica materials. For example, in addition to mica, the raw materials required for processing mica tapes also require binders, solvents, silane coupling agents, catalysts, stabilizers and other raw materials. The addition of these raw materials requires accurate proportions. Insufficient addition of a certain raw material will lead to reduced electrical performance of the mica sheets produced, which will not meet the mica material order delivery standards and cause financial losses to the mica processing plant. First, retrieve the historical processing data and initial equipment parameters of the same type as the mica processing product type from the preset mica processing database to prepare for the subsequent correction of the initial equipment parameters.

[0079] S103, based on the initial equipment parameters, the samples to be processed are processed by mica processing equipment, and the mica processing data during the processing of the samples to be processed are collected by using the multi-type sensors preset in all mica processing equipment, and the sample quality parameters after the processing of the samples to be processed are detected.

[0080] In this embodiment, the sample to be processed is first processed by the mica processing equipment using the initial equipment parameters. While processing the sample, the mica processing data of the sample to be processed during the processing is collected using the multi-type sensors preset in all the mica processing equipment, and the sample quality parameters after the processing of the sample to be processed are detected. The main function of this step is to detect the processing step parameters that need to be modified later. Among them, the multi-type sensors are set inside all the mica processing equipment, and the multi-type sensors include temperature sensors, weight sensors and pH detection sensors, which are used to detect the temperature changes, weight changes and pH changes that occur during the mica processing process. The mica processing data includes temperature change data, weight change data and pH change data. In addition, all sensors will upload the detected data to the database in real time during the detection process, and the database will store it.

[0081] S104, classifying and analyzing the timing characteristics of mica processing data, and screening out core process parameters in the initial equipment parameters according to the analysis results of the timing characteristics analysis.

[0082] In this embodiment, the mica processing data refers to the temperature data, weight data and pH data in the mica processing process. These data are first arranged according to the time of collection, converted into data with time series characteristics, and then the time series characteristic data are classified to obtain mica weight change data, mica temperature change data and mica pH change data. Since each step in the mica processing process will have a weight change, the mica processing technology can be node-divided according to the mica weight change data, and then the core process parameters in the initial equipment parameters, i.e., mica purification parameters and mica grinding parameters, are screened out according to the mica temperature change data and mica pH change data. The mica purification parameters and mica grinding parameters are the core steps for subsequent correction. Since the temperature and pH changes will occur in the mica purification parameters and mica grinding parameters, they can be screened out according to the temperature and pH change amplitudes. The role of this step is to screen out the parameters corresponding to the steps in the mica processing process that are prone to waste of mica raw materials, so as to facilitate the subsequent reduction of the mica raw material loss rate.

[0083] S105. Calculate the average value of historical mica processing data based on the core process parameters, and use the average value as the core process standard value corresponding to the core process parameters.

[0084] In this embodiment, the corresponding historical mica processing data is screened out according to the core process parameters, the mica weight change data involving the two steps of mica purification and mica grinding in the historical mica processing data are extracted, and the purification weight average value of the historical purification weight change data is calculated, and the calculated purification weight average value is used as the purification process standard value. In the same way, the grinding weight average value of the historical grinding weight change data is calculated, and the calculated grinding weight average value is used as the grinding process standard value. The core process standard value is obtained by integrating the purification process standard value and the grinding process standard value.

[0085] S106. Calculate the mica raw material loss rate in the mica processing process by combining the core process standard value with the mica processing data.

[0086] In this embodiment, the difference between the standard value of the purification process and the historical purification weight change data in the mica processing data is calculated, and then the difference is divided by the historical purification weight change data to obtain the mica raw material loss rate in the mica purification process. In the same way, the difference between the standard value of the grinding process and the historical grinding weight change data in the mica processing data is calculated, and then the difference is divided by the historical grinding weight change data to obtain the mica raw material loss rate in the mica grinding process.

[0087] S107, screening the core process parameters by calculating the correlation coefficient between the mica raw material loss rate and the core process parameters, and outputting the screened core process parameters as local correction factors.

[0088] In this embodiment, the correlation coefficients between the mica purification parameters and the mica grinding parameters and the mica raw material loss rate are calculated respectively, and the parameters with the largest mica raw material loss rate are screened out according to the correlation coefficients, and then the screened parameters are output as local correction factors, so as to facilitate the subsequent correction of the core process parameters in the initial equipment parameters. Since the mica processing environment data (ambient temperature data and ambient humidity data) will have a certain impact on the mica raw material loss rate, it is necessary to calculate the correlation coefficient between the mica processing environment data and, and determine whether the mica processing environment data needs to be output as a local correction factor according to the obtained correlation coefficient.

[0089] S108. Obtain mica product quality parameters corresponding to the mica processing product type through a preset database, calculate the influence degree data set between the initial equipment parameters and the mica product quality parameters, prioritize the initial equipment parameters based on the influence degree data set, and output the prioritized initial equipment parameters as global correction factors.

[0090] In this embodiment, mica product quality parameters are obtained through a preset database, and the obtained mica product quality parameters are parameters used to measure the quality of mica processed products, such as sand content, loss on ignition, loose density, water content, whiteness, etc. The quality parameters of mica powder are calculated as the influence degree data set between the mica equipment parameters, mica raw material data, mica ingredient ratio and mica product quality parameters contained in the initial equipment parameters, and the initial equipment parameters are prioritized according to the influence degree data set. The initial equipment parameters with high priority are corrected first. If the mica product quality standard parameters are reached, there is no need to continue to correct them, which can save time and reduce unnecessary waste of resources. After completing the priority classification, the initial equipment parameters are output as global correction factors.

[0091] S109, correcting the initial equipment parameters by combining the local correction factor and the global correction factor to obtain the optimal equipment parameters.

[0092] In this embodiment, the core process parameters and mica processing environment data are first corrected according to local correction factors. After the local correction is completed, the initial equipment parameters are corrected according to the global correction factors and the priority corresponding to the global correction factors. If the mica product quality standard parameters are reached or three corrections are completed, the final correction results are output as the optimal equipment parameters.

[0093] In one embodiment, referring to Figure 2 , classifying and analyzing the timing characteristics of mica processing data, and screening out the core process parameters in the initial equipment parameters according to the analysis results of the timing characteristics analysis include the following steps:

[0094] S201, feature-classifying mica processing data according to sensor types of the multi-type sensors to obtain mica weight change data, mica temperature change data, and mica pH change data;

[0095] S202, setting an upper limit of the total number of inflection points according to the total number of processing steps included in the mica processing technology, taking the upper limit of the total number of inflection points as an algorithm constraint, and performing inflection point detection on the mica weight change data using an inflection point detection algorithm to obtain weight change inflection points distributed in the mica weight change data;

[0096] S203, dividing the mica processing technology into process nodes according to the weight change inflection point, and obtaining multiple mica processing sub-processes within the mica processing technology;

[0097] S204, extracting the maximum temperature and the minimum temperature of the mica temperature change data in each mica processing sub-process, calculating the difference between the maximum temperature and the minimum temperature, and obtaining the temperature difference;

[0098] S205, extracting the maximum pH value and the minimum pH value of the mica pH change data in each mica processing sub-process, calculating the difference between the maximum pH value and the minimum pH value, and obtaining the pH difference;

[0099] S206, if the temperature difference is greater than a preset temperature difference threshold, analyzing whether the pH difference corresponding to the mica processing sub-process is greater than a preset pH difference threshold;

[0100] S207, if the pH difference corresponding to the mica processing sub-process is greater than the pH difference threshold, determining that the initial equipment parameters corresponding to the mica processing sub-process are mica purification parameters;

[0101] S208, if the pH difference is less than or equal to the pH difference threshold, determining that the initial equipment parameters corresponding to the mica processing sub-process are mica grinding parameters;

[0102] S209, integrating mica purification parameters and mica grinding parameters to obtain core process parameters.

[0103] In this embodiment, mica processing data refers to temperature data, weight data and pH data in the mica processing process. These data are first arranged according to the time of collection and converted into data with time series characteristics. Then, the time series characteristic data is classified according to the sensor type of the multi-type sensors, and mica weight change data, mica temperature change data and mica pH change data can be obtained. Since weight changes will occur in each step of the mica processing process, the mica processing technology can be divided into nodes according to the mica weight change data.

[0104] The inflection point detection algorithm is used to detect inflection points of mica weight change data, and a control function is constructed according to the mica processing data. The process of detecting inflection points is the process of minimizing the value of the control function. Since the loss function value of the control function is lower in a uniform subsequence and higher in an uneven subsequence, the total number of inflection points of the mica weight change data and the distribution position of the inflection points of the mica weight change data can be calculated according to this principle. In order to prevent the total number of inflection points from being too large, it is necessary to set a total inflection point threshold based on the total number of processing steps included in the mica processing technology, and constrain the total number of inflection points generated. For example, the total inflection point threshold is set to three times the total number of processing steps. Finally, the weight change inflection points distributed in the mica weight change data can be obtained.

[0105] The mica processing technology is divided into process nodes according to the obtained weight change inflection point, and multiple mica processing sub-processes within the mica processing technology are obtained. The maximum temperature and the minimum temperature of the mica temperature change data corresponding to each mica processing sub-process are extracted, and then the difference between the maximum temperature and the minimum temperature is calculated to obtain the temperature difference. The maximum pH value and the minimum pH value of the mica pH change data corresponding to each mica processing sub-process are extracted, and the difference between the maximum pH value and the minimum pH value is calculated to obtain the pH difference. The mica purification parameters and mica grinding parameters included in the initial equipment parameters are screened out according to the temperature difference and the pH difference.

[0106] If the temperature difference is greater than the preset temperature difference threshold, the mica processing sub-process with a pH difference greater than the preset pH difference threshold is screened out, and the initial equipment parameters corresponding to the mica processing sub-process are determined to be mica purification parameters. This is because acid and alkali solution or organic solvent will be added during the mica purification process, and chemical reactions will occur in this process, resulting in temperature changes and pH changes. According to this change feature, the initial equipment parameters corresponding to the mica processing sub-process are determined to be mica purification parameters. The mica processing sub-process with a pH difference less than or equal to the preset pH difference threshold is screened out, and the initial equipment parameters corresponding to the mica processing sub-process are determined to be mica grinding parameters. This is because temperature changes will occur due to continuous friction during the mica grinding process. According to this change feature, the initial equipment parameters corresponding to the mica processing process are determined to be mica grinding parameters. Among them, the mica grinding parameters refer to the vibration frequency, grinding time and grinding method involved in the mica grinding process, and the mica purification parameters refer to the solution ratio, purification time and purification mesh parameters involved in the mica purification process. Finally, the mica purification parameters and mica grinding parameters are integrated to obtain the core process parameters.

[0107] In one embodiment, calculating the average value of historical mica processing data according to the core process parameters, and using the average value as the core process standard value corresponding to the core process parameters includes the following steps:

[0108] Extracting historical purification weight change data from historical processing data corresponding to the mica purification parameters, obtaining historical purification weight change data, calculating the purification weight average of the historical purification weight change data, and using the purification weight average as the purification process standard value;

[0109] Extracting historical grinding weight change data from historical processing data corresponding to mica grinding parameters, obtaining historical grinding weight change data, calculating the grinding weight average value of the historical grinding weight change data, and using the grinding weight average value as a grinding process standard value;

[0110] The standard values ​​of the purification process and the grinding process are integrated to obtain the standard values ​​of the core process parameters.

[0111] In this embodiment, the corresponding historical processing data of mica is screened out according to the core process parameters, and the historical processing data is first arranged in chronological order to extract the timing features. The mica weight change data involving the two steps of mica purification and mica grinding in the historical processing data of mica after the timing features are extracted are screened out, and the purification weight average value of the historical purification weight change data is calculated, and the calculated purification weight average value is used as the purification process standard value. In the same way, the grinding weight average value of the historical grinding weight change data is calculated, and the calculated grinding weight average value is used as the grinding process standard value. The core process standard value is obtained by integrating the purification process standard value and the grinding process standard value.

[0112] In one embodiment, calculating the correlation coefficient between the mica raw material loss rate and the core process parameters, screening the core process parameters according to the correlation coefficient, and outputting the screened core process parameters as local correction factors includes the following steps:

[0113] Calculate the core parameter correlation coefficient set between the mica core process parameters and the mica raw material loss rate, screen out the core parameter correlation coefficient set that is greater than a preset core parameter correlation coefficient threshold, and output the mica core process parameters corresponding to the core parameter correlation coefficient as local correction factors;

[0114] According to the multi-type sensors preset in the mica processing plant, the mica processing environment data is obtained, and the mica processing environment data includes the mica processing environment temperature data and the mica processing environment humidity data;

[0115] A multiple linear regression equation was established with the mica raw material loss rate as the dependent variable, and the mica processing environment temperature data and the mica processing environment humidity data were input into the multiple linear regression equation as independent variables, and the environment temperature correlation coefficient and the environment humidity correlation coefficient were calculated.

[0116] If the environmental temperature correlation coefficient is greater than a preset environmental correlation coefficient threshold, the mica processing environmental temperature data is output as a local correction factor;

[0117] If the environmental humidity correlation coefficient is greater than the environmental correlation coefficient threshold, the mica processing environmental humidity data is output as a local correction factor.

[0118] In this embodiment, first, the core parameter correlation coefficient set between the mica core process parameters and the mica raw material loss rate is calculated. This step is intended to quantify the degree of influence of the core process parameters on the raw material loss rate. In specific implementation, the Pearson correlation coefficient method can be used. For each core process parameter, such as grinding time, vibration frequency, solution ratio, etc., the correlation coefficient is calculated with the raw material loss rate. The calculation formula is:

[0119]

[0120] Where X represents the core process parameters, Y represents the raw material loss rate, and X and are the average values ​​of the corresponding variables respectively. For example, suppose that the correlation coefficient between grinding time and raw material loss rate is 0.85, and the correlation coefficient between vibration frequency is 0.72. After the calculation is completed, a set of correlation coefficients of all core process parameters will be obtained. Next, set a suitable correlation coefficient threshold, such as 0.7, to screen out the correlation coefficients greater than the threshold and their corresponding core process parameters. In this example, both grinding time and vibration frequency will be selected. These selected parameters will be output as local correction factors for subsequent process optimization. The effect of implementing this step is to identify the core process parameters that have the most significant impact on the raw material loss rate, provide a clear direction for subsequent parameter optimization, and help improve the efficiency of mica processing and reduce raw material loss.

[0121] Then, the mica processing environment data is obtained according to the multi-type sensors preset in the mica processing plant. These data mainly include the temperature data of the mica processing environment and the humidity data of the mica processing environment. During implementation, temperature sensors and humidity sensors can be installed at key locations in the processing workshop, such as the raw material storage area, grinding area, and purification area. These sensors need to have high precision and stability, and be able to record environmental parameters in real time and continuously. For example, a PT100 platinum resistance temperature sensor can be used to measure temperature with an accuracy of ±0.1°C; a capacitive humidity sensor can be used to measure relative humidity with an accuracy of ±2%RH. The data acquisition frequency of the sensor can be set to once per minute to capture subtle changes in environmental parameters. The data acquisition system pre-processes these raw data, including removing outliers and smoothing, and then stores them in the database. The final mica processing environment data set contains the following information: timestamp, location identifier, temperature value (°C), and relative humidity value (%RH). The implementation effect of this step is to provide high-quality and comprehensive environmental parameter data for subsequent analysis, which reflects the actual environmental conditions during mica processing.

[0122] Then, a multiple linear regression equation with mica raw material loss rate as the dependent variable is established. The purpose of this step is to quantify the impact of ambient temperature and humidity on raw material loss rate. The general form of the multiple linear regression equation is: Y = β0 + β1X1 + β2X2 + ε, where Y represents the mica raw material loss rate, X1 represents the ambient temperature, X2 represents the ambient humidity, β0 is the intercept, β1 and β2 are regression coefficients, and ε is the error term. When implementing, it is necessary to first prepare enough historical data, including temperature and humidity records under different environmental conditions and the corresponding raw material loss rate. Then the regression coefficient is estimated using the least squares method. Statistical software such as R or Python's statsmodels library can be used for calculation. For example, assume that the regression equation obtained is: loss rate (%) = 2.5 + 0.15 * temperature (℃) + 0.08 * humidity (%). This means that for every 1℃ increase in temperature, the loss rate increases by an average of 0.15%; for every 1% increase in relative humidity, the loss rate increases by an average of 0.08%. Next, the mica processing environment temperature data and the mica processing environment humidity data are input into the regression equation as independent variables. By calculating the t statistic of the partial regression coefficient and its corresponding p value, the environment temperature correlation coefficient and the environment humidity correlation coefficient can be obtained.

[0123] Then determine whether the ambient temperature has a significant effect on the raw material loss rate. When implementing, first set a suitable environmental correlation coefficient threshold, the selection of this threshold is usually based on industry experience or statistical standards, for example, it can be set to 0.3. Next, compare the ambient temperature correlation coefficient calculated in the previous step with this threshold. If the ambient temperature correlation coefficient is greater than the threshold, for example, the correlation coefficient obtained is 0.42, which is greater than the set threshold of 0.3, then it is considered that the ambient temperature has a significant effect on the raw material loss rate. In this case, the mica processing ambient temperature data is output as a local correction factor. This means that in the subsequent process optimization, special attention needs to be paid to and control of the temperature of the processing environment. The specific output includes information such as the temperature measurement range, fluctuation range, and the most suitable temperature range. For example, it is concluded that the processing environment temperature should be controlled between 20-25°C, and the temperature fluctuation should not exceed ±2°C.

[0124] Finally, it is determined whether the environmental humidity correlation coefficient is greater than the environmental correlation coefficient threshold. The purpose and principle of this step are similar to those of determining the temperature correlation coefficient, which aims to determine whether the environmental humidity has a significant effect on the raw material loss rate. During implementation, the same environmental correlation coefficient threshold (such as 0.3) is used to compare with the calculated environmental humidity correlation coefficient. If the environmental humidity correlation coefficient is greater than the threshold, for example, the obtained correlation coefficient is 0.38, which is greater than the set threshold of 0.3, it is considered that the environmental humidity has a significant effect on the raw material loss rate. In this case, the mica processing environmental humidity data is output as a local correction factor. The output content includes the appropriate range of humidity, control accuracy requirements, etc. For example, it is concluded that the relative humidity of the processing environment should be controlled between 45-55%, and the humidity fluctuation should not exceed ±5%RH. The implementation effect of this step is to determine environmental humidity as another potential key influencing factor, which provides a scientific basis for the subsequent formulation of humidity control strategies. Combining the two factors of temperature and humidity, the environmental conditions of mica processing can be more comprehensively optimized, thereby effectively reducing the raw material loss rate and improving product quality and production efficiency.

[0125] In one embodiment, calculating a core parameter correlation coefficient set between mica core process parameters and mica raw material loss rate, screening out core parameter correlation coefficients greater than a preset core parameter correlation coefficient threshold value in the core parameter correlation coefficient set, and outputting the core process parameters corresponding to the core parameter correlation coefficients as local correction factors includes the following steps:

[0126] A correlation coefficient model is established with the mica raw material loss rate as an independent variable;

[0127] The vibration frequency, grinding time and grinding method included in the mica grinding parameters are input into the correlation coefficient model as independent variables to obtain the grinding correlation coefficient set;

[0128] Screening out the grinding correlation coefficients in the grinding correlation coefficient set that are greater than a preset correlation coefficient threshold, and outputting the mica grinding parameters corresponding to the grinding correlation coefficients as local correction factors;

[0129] The solution ratio, purification time and purification mesh parameters included in the mica purification parameters are respectively input into the correlation coefficient model as independent variables to obtain a purification correlation coefficient set;

[0130] The purification correlation coefficients in the purification correlation coefficient set that are greater than a preset correlation coefficient threshold are screened out, and the mica purification parameters corresponding to the purification correlation coefficients are output as local correction factors.

[0131] In this embodiment, it is first necessary to clarify that the mica grinding parameters include vibration frequency, grinding time and grinding method. The vibration frequency is usually measured in Hertz (Hz), and the possible range is 20-100 Hz; the grinding time is measured in minutes, and the possible range is 30-180 minutes; the grinding method may include dry grinding, wet grinding or mixed grinding. The correlation coefficient model usually adopts the Pearson correlation coefficient method. During implementation, it is first necessary to collect a large amount of historical data, including mica processing results under different vibration frequencies, grinding times and grinding methods. For example (50Hz, 60min, dry grinding, 95% purity), (75Hz, 90min, wet grinding, 97% purity), etc. Then, the correlation coefficient between each grinding parameter and the processing result is calculated separately. Assume that the calculation results show that the vibration frequency correlation coefficient is 0.75, the grinding time correlation coefficient is 0.82, and the grinding method (after conversion to numerical value) correlation coefficient is 0.68. These correlation coefficients constitute the grinding correlation coefficient set. The implementation effect of this step is to quantify the degree of influence of each grinding parameter on the mica processing result. The resulting set of correlation coefficients can intuitively show which parameters have a greater impact on the processing results. For example, in the above hypothetical data, the correlation coefficient of grinding time is the highest, which means that it may be the most critical grinding parameter. This provides a clear direction for subsequent parameter optimization, which helps to improve the efficiency of the mica grinding process and product quality.

[0132] Then set a suitable correlation coefficient threshold to distinguish between parameters with significant influence and those with non-significant influence. When implemented, the correlation coefficient of each grinding parameter obtained in the previous step is compared with the set threshold. Assuming that the threshold is set to 0.7, then according to the assumed data in the previous step, the correlation coefficients of vibration frequency (0.75) and grinding time (0.82) both exceed the threshold, while the grinding method (0.68) does not exceed the threshold. Therefore, vibration frequency and grinding time will be selected as significant influencing parameters and output as local correction factors. The output local correction factors may include the following information: vibration frequency: correlation coefficient 0.75, recommended adjustment range 50-80Hz; grinding time: correlation coefficient 0.82, recommended adjustment range 60-120 minutes. These outputs will be used for subsequent process parameter optimization. For example, the grinding time can be adjusted first because it has the highest correlation coefficient and has the greatest impact on the processing results. Then consider adjusting the vibration frequency. The grinding method does not need to be adjusted temporarily because of its low correlation coefficient. The implementation effect of this step is to select the most critical parameters from many grinding parameters, simplifying the subsequent optimization process. By focusing on these key parameters, the mica grinding process can be improved more effectively, product quality can be improved, unnecessary parameter adjustments can be reduced, and time and resources can be saved. In addition, this data-based parameter screening method provides a scientific basis for process optimization and reduces the errors that may be caused by subjective judgment.

[0133] Next, the mica purification parameters are input as independent variables into the correlation coefficient model to obtain the set of purification correlation coefficients. The mica purification parameters mainly include solution ratio, purification time and purification mesh parameters. Solution ratio usually refers to the concentration of the pickling solution, and the possible range is 5%-30%; the purification time is measured in hours, and the possible range is 2-12 hours; the purification mesh parameter refers to the mesh size of the mesh, and the possible range is 100-400 mesh. The correlation coefficient model also uses the Pearson correlation coefficient method. When implementing it, it is first necessary to collect a large amount of historical data, including the mica purification results under different solution ratios, purification times and purification mesh parameters. For example (15%, 6h, 200 mesh, 98% purity), (20%, 8h, 300 mesh, 99% purity), etc. Then, the correlation coefficient between each purification parameter and the purification result is calculated separately. Assume that the calculation results show that the solution ratio correlation coefficient is 0.78, the purification time correlation coefficient is 0.85, and the purification mesh parameter correlation coefficient is 0.72. These correlation coefficients constitute the purification correlation coefficient set. Some special cases need to be paid attention to during the calculation process. For example, the purification mesh parameter is a discrete mesh value, which needs to be converted into a continuous pore size (in microns) before the correlation calculation. In addition, if there is an interaction between certain parameters, it is also necessary to consider introducing interaction terms for analysis. The effect of this step is to quantify the degree of influence of each purification parameter on the mica purification result. Through the obtained correlation coefficient set, it can be intuitively seen which parameters have a greater impact on the purification results. For example, in the above hypothetical data, the purification time has the highest correlation coefficient, which means that it may be the most critical purification parameter. This provides a clear direction for subsequent parameter optimization, which helps to improve the efficiency of the mica purification process and product quality.

[0134] Next, the correlation coefficient of each purification parameter is compared with the set threshold. Assuming that the threshold is set to 0.75, then according to the assumed data in S3, the correlation coefficients of the solution ratio (0.78) and the purification time (0.85) both exceed the threshold, while the purification mesh parameter (0.72) does not exceed the threshold. Therefore, the solution ratio and purification time will be selected as significant influencing parameters and output as local correction factors. The output local correction factors may include the following information: solution ratio: correlation coefficient 0.78, recommended adjustment range 15%-25%; purification time: correlation coefficient 0.85, recommended adjustment range 6-10 hours. These outputs will be used for subsequent process parameter optimization. For example, priority can be given to adjusting the purification time because it has the highest correlation coefficient and has the greatest impact on the purification results. Secondly, consider adjusting the solution ratio. The purification mesh parameter does not need to be adjusted for the time being due to its low correlation coefficient. In practical applications, the feasibility and economy of parameter adjustment also need to be considered. For example, although increasing the purification time may increase purity, it will also increase energy consumption and production costs. Therefore, when determining the final parameter adjustment plan, it is necessary to find a balance between product quality and production efficiency.

[0135] In one embodiment, the mica product quality parameters are obtained through a preset database, the influence degree data set between the initial equipment parameters and the mica product quality parameters is calculated, the initial equipment parameters are prioritized based on the influence degree data set, and the initial equipment parameters with priority classification are output as global correction factors, including the following steps:

[0136] Obtain mica product quality parameters corresponding to mica processing product types through a preset database;

[0137] A mica regression model was established with mica product quality parameters as dependent variables and initial equipment parameters as independent variables.

[0138] The mica equipment parameters, mica raw material data, and mica ingredient ratio included in the initial equipment parameters are respectively input into the mica regression model as independent variables to obtain the influence degree data, and the influence degree data are integrated to obtain the influence degree data set;

[0139] Outputting the initial equipment parameters corresponding to the data with the greatest impact degree in the impact degree data set as the global correction factor with the first priority;

[0140] Outputting the initial equipment parameters corresponding to the second largest impact data in the impact data set as the global correction factor of the second priority;

[0141] The initial device parameters corresponding to the three largest impact data in the impact data set are output as global correction factors of the third priority, and the first priority is greater than the second priority, and the second priority is greater than the third priority.

[0142] In this embodiment, the preset database is usually a structured information storage system, using a relational database (such as MySQL, Oracle) or a non-relational database (such as MongoDB). The database stores detailed information on various mica processing products, including but not limited to product type, quality parameters, production process, etc. Mica product quality parameters include multiple aspects, such as purity, sheet diameter, thickness, heat resistance, insulation, etc. For example, for mica powder products, quality parameters include: purity (≥98%), particle size distribution (D50≤45μm), whiteness (≥85%), water content (≤0.5%), etc. For mica sheet products, including: purity (≥99%), sheet diameter (10-100mm), thickness (0.02-0.1mm), heat resistance temperature (≥800℃), etc. During implementation, it is first necessary to determine the type of mica product currently being processed, such as mica powder, mica sheet, mica paper, etc. Then, the quality parameters corresponding to the product type are retrieved from the preset database through SQL query statements or other database query methods. These acquired quality parameters will be used as dependent variables for establishing regression models in subsequent steps to analyze the impact of initial equipment parameters on product quality. This not only provides reliable basic data for subsequent regression analysis, but also helps to ensure the practicality and reliability of the final optimization results. At the same time, this database-based parameter acquisition method is also convenient for future data updates and maintenance, ensuring that the analysis process always uses the latest and most accurate quality standards.

[0143] A mica regression model is established with mica product quality parameters as dependent variables and initial equipment parameters as independent variables. The choice of regression model depends on the complexity of the relationship between variables. Multiple linear regression, polynomial regression, or more complex nonlinear regression models are used. For complex industrial processes such as mica processing, multiple linear regression models are usually preferred because they are simple to understand and can provide a good enough approximation in many cases. When implementing, it is first necessary to collect enough historical data, including product quality data obtained under different equipment parameter settings. For example, there are the following data sets:

[0144] (Temperature: 200°C, Pressure: 5MPa, Speed: 1000rpm) → Purity: 98.5%

[0145] (Temperature: 220°C, Pressure: 6MPa, Speed: 1200rpm) → Purity: 99.0%

[0146] (Temperature: 180°C, Pressure: 4MPa, Speed: 800rpm) → Purity: 97.8%

[0147] Then, the regression coefficients are estimated using the least squares method or other regression analysis methods. Statistical software such as R, Python's statsmodels library, or SPSS can be used for calculations. Assume that the regression equation obtained is:

[0148] Purity (%) = 90 + 0.03 * temperature (°C) + 0.5 * pressure (MPa) + 0.001 * speed (rpm)

[0149] This equation shows that for every 1°C increase in temperature, the purity increases by an average of 0.03%; for every 1MPa increase in pressure, the purity increases by an average of 0.5%; and for every 1rpm increase in speed, the purity increases by an average of 0.001%.

[0150] After the model is established, model diagnosis is required, including checking the normality, homoscedasticity, and multicollinearity of the residuals. The coefficient of determination R is also required to be calculated. 2 To assess the goodness of fit of the model.

[0151] Next, the mica equipment parameters, mica raw material data, and mica ingredient ratios included in the initial equipment parameters are input as independent variables into the mica regression model to obtain the influence degree data, and the influence degree data are integrated to obtain the influence degree data set. The core of this step is to use the regression model established in the above steps to quantify the influence degree of each initial equipment parameter on the quality of mica products. The initial equipment parameters can be divided into three categories:

[0152] Mica equipment parameters: such as grinding machine speed, pressure, temperature, etc.

[0153] Mica raw material data: such as raw material purity, water content, particle size distribution, etc.

[0154] Mica ingredient ratio: such as the mixing ratio of different grades of mica, the ratio of additives, etc.

[0155] When implementing, it is necessary to first determine the specific indicators for each type of parameter. For example:

[0156] Mica equipment parameters: grinding machine speed (500-2000rpm), pressure (1-10MPa), temperature (100-300℃).

[0157] Mica raw material data: raw material purity (95-99%), water content (0.1-1%), average particle size (10-100μm).

[0158] Mica ingredient ratio: high-purity mica ratio (60-90%), additive ratio (0.5-2%).

[0159] The following results can be obtained through the calculation of the mica regression model:

[0160] Grinding machine speed: β = 0.002 (every increase of 1rpm, the purity increases by 0.002%)

[0161] Pressure: β = 0.5 (every increase of 1MPa, the purity increases by 0.5%)

[0162] Temperature: β = 0.03 (purity increases by 0.03% for every 1°C increase)

[0163] Raw material purity: β = 0.8 (for every 1% increase, the product purity increases by 0.8%)

[0164] Water content: β = -0.3 (for every 1% increase, the product purity decreases by 0.3%)

[0165] Average particle size: β = -0.01 (product purity decreases by 0.01% for every 1 μm increase)

[0166] High purity mica ratio: β = 0.1 (each increase of 1% will increase the product purity by 0.1%)

[0167] Additive ratio: β = -0.2 (for every 1% increase, product purity decreases by 0.2%)

[0168] These beta values ​​constitute the impact dataset.

[0169] The effect of this step is to obtain a quantitative impact dataset, which clearly shows the magnitude and direction of the impact of each initial equipment parameter on product quality. This provides direct data support for subsequent parameter optimization, helps to identify the parameters that have the most significant impact on product quality, and thus formulate targeted optimization strategies.

[0170] The initial equipment parameters corresponding to the data with the largest impact in the impact data set are output as the global correction factors with the first priority. The core of this step is to identify the equipment parameters that have the most significant impact on the quality of mica products and use them as the top priority factors in the optimization process. When implementing, it is first necessary to sort the impact data sets obtained in the above steps. The basis for sorting is the absolute value of the impact of each parameter, because no matter whether it is a positive impact or a negative impact, as long as the impact is large, it means that the parameter has an important impact on product quality.

[0171] Based on the calculated data in the above steps, the impact datasets are ranked as follows:

[0172] 1. Raw material purity: β = 0.8

[0173] 2. Pressure: β = 0.5

[0174] 3. Water content: β = -0.3

[0175] 4. Temperature: β = 0.03

[0176] 5. Additive ratio: β = -0.2

[0177] 6. High purity mica ratio: β = 0.1

[0178] 7. Average particle size: β = -0.01

[0179] 8. Grinding machine speed: β = 0.002

[0180] According to the sorting results, the raw material purity has the greatest impact (β=0.8), so the raw material purity is output as the global correction factor with the first priority.

[0181] The output information includes:

[0182] Parameter name: Raw material purity

[0183] Impact: 0.8

[0184] Impact direction: Positive (increasing the purity of raw materials will increase the purity of products)

[0185] Recommended adjustment range: 95-99%

[0186] Optimization suggestion: Under the premise of ensuring economic feasibility, improve the purity of raw materials as much as possible

[0187] The effect of implementing this step is to clarify the equipment parameters that have the most significant impact on product quality, providing a clear direction for subsequent process optimization. By prioritizing the adjustment of this parameter, product quality can be improved most quickly and effectively. This data-based priority setting method not only improves the efficiency of the optimization process, but also reduces the errors caused by subjective judgment. At the same time, using the most important parameter as a global correction factor means that this parameter needs to be focused on and controlled throughout the production process, which helps to maintain the stability and consistency of product quality.

[0188] Next, the initial equipment parameters corresponding to the second largest impact data in the impact data set are output as the second priority global correction factor. The core of this step is to identify the equipment parameters that have the second most significant impact on the quality of mica products, and use them as the second priority factor in the optimization process. Continuing to use the sorting results in the above steps, the parameter corresponding to the second largest impact data is pressure (β=0.5). Therefore, pressure is output as the second priority global correction factor. Taking pressure as the second priority global correction factor means that in the entire production process, in addition to paying attention to the purity of the raw materials, it is also necessary to focus on controlling and monitoring the pressure parameters.

[0189] Next, the initial equipment parameters corresponding to the third largest impact data in the impact data set are output as the global correction factors of the third priority, while ensuring that the first priority is greater than the second priority, and the second priority is greater than the third priority. The core of this step is to identify the equipment parameters that have the third most significant impact on the quality of mica products, and include them in the optimization considerations, while establishing a clear priority structure. Continuing to use the previous sorting results, the parameter corresponding to the third largest impact data is water content (β = -0.3). Therefore, water content is output as the global correction factor of the third priority. During implementation, it should be noted that the adjustment of water content will affect other parameters, such as the fluidity of the material or the processing temperature. Therefore, when adjusting the water content, it is necessary to comprehensively consider its impact on the entire process. Establishing a clear priority structure (first priority > second priority > third priority) helps to make reasonable decisions when resources are limited.

[0190] In one embodiment, the initial device parameters are corrected by combining the local correction factor and the global correction factor to obtain the optimal device parameters, including the following steps:

[0191] After analyzing the local correction factors, the core process parameters and mica processing environment data included in the primary equipment parameters are partially corrected;

[0192] The initial equipment parameters that have completed local corrections are globally corrected according to the global correction factors to obtain the optimal equipment parameters.

[0193] In this embodiment, it is first necessary to analyze the previously determined local correction factors in detail. These local correction factors include core process parameters such as grinding time, vibration frequency, solution ratio, purification time, purification screen parameters, and environmental data such as temperature and humidity. The analysis process involves an in-depth analysis of the influencing mechanism, adjustment range and interaction of each factor. For example, for the core process parameter of grinding time, the analysis includes: (1) Influencing mechanism: Grinding time directly affects the fineness and uniformity of mica particles. Too short will lead to insufficient grinding, and too long will cause excessive grinding and energy waste. (2) Adjustment range: Based on historical data and equipment capabilities, the adjustment range is 30-180 minutes. (3) Interaction: Grinding time interacts with parameters such as vibration frequency and grinding media, and needs to be considered in a coordinated manner. For the processing environment data of ambient temperature, the analysis includes: (1) Influencing mechanism: Temperature affects the brittleness of mica and the performance of processing equipment. Too high temperature causes mica to become overly dry and brittle, and too low temperature affects the normal operation of the equipment. (2) Adjustment range: According to the physical properties of mica and equipment requirements, the adjustment range is 18-28°C. (3) Interaction: Temperature and humidity are closely related and need to be considered simultaneously.

[0194] After the analysis is completed, local correction is performed. This process adopts the fine adjustment method, that is, small and gradual adjustments are made on the basis of the original parameters. The specific implementation method can adopt the orthogonal test method or the response surface method. Taking the grinding time as an example, a 3-level orthogonal test can be designed: L9 (3^4), in which the three levels of grinding time can be set to 60 minutes, 90 minutes and 120 minutes. At the same time, consider other 3 related factors, such as vibration frequency, grinding medium size and grinding medium material. Through these 9 groups of experiments, the optimal level of grinding time can be found. For ambient temperature, the response surface method can be used. Design a central composite design with a temperature range of 18-28°C, while considering humidity factors (such as 40-60% RH). By fitting the quadratic response surface model, the optimal temperature and humidity combination can be obtained. The effect of the correction is evaluated by comparing the product quality parameters before and after the correction. For example, the purity, particle size distribution, whiteness and other indicators of mica before and after the correction can be compared. If the corrected parameters significantly improve these indicators, the local correction is considered to be effective.

[0195] According to the global correction factors, the initial equipment parameters that have completed the local correction are globally corrected to obtain the optimal equipment parameters. This step combines the results of the previous local correction with the global correction factors, comprehensively considers the influence of each parameter, and finally determines the optimal combination of equipment parameters. Global correction factors usually include several key parameters that have the most significant impact on product quality, such as raw material purity, pressure, temperature, etc. The priority of these factors has been determined in the previous steps. The purpose of global correction is to optimize all parameters as a whole while meeting the requirements of these key factors. An effective way to implement global correction is to use a multi-objective optimization algorithm, such as a genetic algorithm or a particle swarm optimization algorithm. Taking the genetic algorithm as an example, its implementation steps are as follows:

[0196] 1. Encoding: Encode all parameters that need to be optimized into chromosomes. For example, parameters such as raw material purity, pressure, temperature, grinding time, etc. can be encoded into a chromosome.

[0197] 2. Initialize the population: Generate a set of initial solutions (usually 100-200) based on the locally modified parameters.

[0198] 3. Fitness evaluation: Define the fitness function, which usually includes product quality indicators (such as purity, particle size) and production efficiency indicators (such as output, energy consumption). For example:

[0199] Fitness = w1*purity+w2*(1 / particle size)+w3*yield-w4*energy consumption

[0200] Among them, w1, w2, w3, and w4 are weight coefficients, which need to be determined according to actual production needs.

[0201] 4. Selection: Select excellent individuals based on fitness. Roulette selection method can be used.

[0202] 5. Crossover and mutation: Generate new parameter combinations through crossover operations, and introduce new properties through mutation operations.

[0203] 6. Repeat steps 3-5 until the termination condition is reached (such as the number of iterations reaches the preset value or there is no significant improvement in the optimal solution for multiple consecutive generations).

[0204] Finally, the effect of this set of optimal parameters needs to be verified in actual production. A small-scale trial production can be carried out first. If the effect is as expected, it can be gradually expanded to full production. At the same time, a continuous monitoring mechanism is established to promptly discover and solve problems. The effect of this step is to obtain a set of globally optimal equipment parameters. These parameters not only take into account local fine adjustments, but also take into account global key factors, which can optimize production efficiency and costs while ensuring product quality.

[0205] In one embodiment, referring to Figure 3 , according to the global correction factor, the initial device parameters that have completed the local correction are globally corrected, and the optimal device parameters are obtained, including the following steps:

[0206] S301, obtaining mica product quality standard parameters corresponding to the mica processing product type according to a preset database;

[0207] S302, calculating the standard deviation between the mica product quality parameter and the mica product quality standard parameter;

[0208] S303, optimizing the initial device parameters corresponding to the global correction factor of the first priority to obtain first device parameters;

[0209] S304, calculating a first difference between a mica product quality parameter obtained by processing according to the first equipment parameter and a mica production standard quality parameter, and if the first difference is less than a preset difference threshold, outputting the first equipment parameter as an optimal equipment parameter;

[0210] S305: If the first difference is greater than or equal to the difference threshold, optimizing the initial device parameters corresponding to the global correction factor of the second priority to obtain second device parameters;

[0211] S306, calculating a second difference between the mica product quality parameter obtained by processing according to the second equipment parameter and the mica production standard quality parameter, and if the second difference is less than the difference threshold, outputting the second mica equipment parameter as the optimal equipment parameter;

[0212] S307: If the second difference is greater than or equal to the difference threshold, the initial device parameters corresponding to the global correction factors of the third priority are optimized to obtain third device parameters, and the third device parameters are output as optimal device parameters.

[0213] In this embodiment, the preset database is usually a structured information storage system, using a relational database (such as MySQL, Oracle) or a non-relational database (such as MongoDB). The database stores detailed quality standards for various mica processing products, which are derived from industry specifications, customer requirements or internal quality control standards of the company. During implementation, it is first necessary to determine the type of mica product currently being processed, such as mica powder, mica sheets, mica paper, etc. Then, the quality standard parameters corresponding to the product type are retrieved from the preset database through SQL query statements or other database query methods. These obtained quality standard parameters will serve as a benchmark for evaluating product quality in subsequent steps.

[0214] Next, calculate the standard deviation between the mica product quality parameters and the mica product quality standard parameters. The purpose of this step is to quantify the gap between the current product quality and the standard, and provide a basis for subsequent parameter optimization. Standard deviation is a statistical concept used to measure the degree of dispersion of a set of data. Here, it reflects the degree of difference between the actual product quality and the standard quality. The formula for calculating the standard deviation is as follows:

[0215]

[0216] Where s is the standard deviation, xi is each sample value, is the sample mean, and n is the sample size.

[0217] In actual operation, it is necessary to calculate the standard deviation for multiple quality parameters. For example, suppose there is the following data:

[0218] Actual purity: 98.5%, standard purity: ≥99%; actual particle size distribution (D50): 48μm, standard particle size distribution (D50): ≤45μm; actual whiteness: 88%, standard whiteness: ≥90%; actual moisture content: 0.6%, standard moisture content: ≤0.5%.

[0219] For each parameter, calculate the difference from the standard value, and then calculate the standard deviation of these differences. It should be noted that for standards of "≥" or "≤" type, the standard value can be used as a reference point to calculate the deviation of the actual value from the reference point. The calculated standard deviations are as follows: purity standard deviation: 0.5%; particle size distribution standard deviation: 3μm; whiteness standard deviation: 2% moisture content standard deviation: 0.1%.

[0220] Then optimize the initial equipment parameters corresponding to the global correction factors of the first priority to obtain the first equipment parameters. The core of this step is to optimize the parameters with the greatest impact in order to quickly improve product quality. The first priority global correction factors are usually the parameters that have the most significant impact on product quality determined in the previous analysis, such as raw material purity or key process parameters. The optimization process can adopt a variety of methods, such as gradient descent, Newton's method or more complex optimization algorithms. Taking the gradient descent method as an example, the basic idea is to iterate along the gradient direction of the objective function until the local optimal solution is found. The specific implementation steps are as follows:

[0221] 1. Define the objective function: usually the difference between the product quality parameters and the standard parameters, such as the mean square error.

[0222] 2. Calculate the gradient of the objective function with respect to the optimization parameters.

[0223] 3. Update parameters along the gradient direction: Where α is the learning rate, is the gradient.

[0224] 4. Repeat steps 2-3 until the termination condition is reached (such as the number of iterations reaches a preset value or the gradient changes very little).

[0225] For example, suppose the first priority global correction factor is raw material purity, and the initial value is 98%. Through optimization, the following results are obtained: Iteration 1: 98.2% (objective function value: 0.8); Iteration 2: 98.5% (objective function value: 0.5); Iteration 3: 98.8% (objective function value: 0.2); Iteration 4: 99.0% (objective function value: 0.1). In the first equipment parameter finally obtained, the raw material purity is optimized to 99.0%. The implementation effect of this step is to optimize the most critical parameters. Since the focus is on the parameters with the greatest impact, the optimization effect is often more obvious.

[0226] Next, the first difference between the mica product quality parameters obtained by processing the first equipment parameters and the mica production standard quality parameters is calculated. If the first difference is less than the preset difference threshold, the first equipment parameters are output as the optimal equipment parameters. The core of this step is to evaluate the effect of the optimization of the first priority parameters and decide whether further optimization is needed. First, it is necessary to use the optimized first equipment parameters for trial production or simulation calculations to obtain new mica product quality parameters. Then, these parameters are compared with the standard quality parameters to calculate the difference. The difference can be calculated using a variety of methods, such as Euclidean distance or weighted average difference. For example, assume that the following results are obtained after optimization:

[0227] Actual purity: 99.1%, standard purity: ≥99%; Actual particle size distribution (D50): 46μm, standard particle size distribution (D50): ≤45μm; Actual whiteness: 89%, standard whiteness: ≥90%; Actual moisture content: 0.5%, standard moisture content: ≤0.5%

[0228] The relative difference of each parameter can be calculated and then averaged:

[0229] Purity difference: (99.1%-99%) / 99%=0.1%; particle size difference: (46μm-45μm) / 45μm=2.2%; whiteness difference: (90%-89%) / 90%=1.1%; moisture content difference: (0.5%-0.5%) / 0.5%=0%; average difference=(0.1%+2.2%+1.1%+0%) / 4=0.85%.

[0230] This average difference is then compared with the preset difference threshold. Assuming that the preset difference threshold is 1%, then 0.85% < 1%, so the first equipment parameter can be output as the optimal equipment parameter. If the first difference is greater than or equal to the difference threshold, the initial equipment parameter corresponding to the global correction factor of the second priority is optimized to obtain the second equipment parameter. The global correction factor of the second priority is usually the parameter with the second greatest impact on product quality determined in the previous analysis, such as pressure, temperature or a key process parameter. The optimization process can adopt a method similar to step S303, such as gradient descent method, Newton method or other optimization algorithms. For example, assuming that the global correction factor of the second priority is pressure, the initial value is 5MPa. Through optimization, the following results are obtained: Iteration 1: 5.5MPa (objective function value: 0.7); Iteration 2: 6.0MPa (objective function value: 0.4); Iteration 3: 6.3MPa (objective function value: 0.2); Iteration 4: 6.5MPa (objective function value: 0.1). In the final second equipment parameter, the pressure was optimized to 6.5MPa. The effect of this step is to further improve product quality by optimizing the less important parameters when the optimization effect of the first priority parameters is insufficient.

[0231] Calculate the second difference between the mica product quality parameter obtained by processing according to the second equipment parameter and the mica production standard quality parameter. If the second difference is less than the difference threshold, the second mica equipment parameter is output as the optimal equipment parameter. Specifically, firstly, it is necessary to use the optimized second equipment parameter for trial production or simulation calculation to obtain new mica product quality parameters. Then, compare these parameters with the standard quality parameters to calculate the difference. The method for calculating the difference can follow the method in step S304. For example, assume that the following results are obtained after optimization:

[0232] Actual purity: 99.2%, standard purity: ≥99%; Actual particle size distribution (D50): 44μm, standard particle size distribution (D50): ≤45μm; Actual whiteness: 90%, standard whiteness: ≥90%; Actual moisture content: 0.4%, standard moisture content: ≤0.5%

[0233] The relative difference of each parameter can be calculated and then averaged:

[0234] Purity difference: (99.2%-99%) / 99%=0.2%; particle size difference: (44μm-45μm) / 45μm=-2.2% (negative value indicates better than the standard); whiteness difference: (90%-90%) / 90%=0%; moisture content difference: (0.4%-0.5%) / 0.5%=-20% (negative value indicates better than the standard); average difference=(0.2%+(-2.2%)+0%+(-20%)) / 4=-5.5%.

[0235] Then the absolute value of this average difference is compared with the preset difference threshold. Assuming that the preset difference threshold is still 1%, then |-5.5%|>1%, but it should be noted here that a negative value actually indicates that the product quality exceeds the standard requirements. Therefore, in this case, although the absolute value of the difference is greater than the threshold, it is a positive excess, so the second device parameter can still be output as the optimal device parameter.

[0236] The effect of implementing this step is to comprehensively evaluate the product quality after the second priority parameters are optimized, and determine the optimal equipment parameters when the effect meets or exceeds expectations. This method not only ensures that the product quality meets the requirements, but also identifies situations where the product quality exceeds the standard. In actual production, this "exceeding the standard" means higher product value or greater process adjustment space. At the same time, this step also reflects the flexibility and adaptability of the optimization process. By gradually optimizing parameters of different priorities and evaluating the effect at each step, the target quality can be achieved within the minimum necessary optimization steps, thereby improving the optimization efficiency. This method can not only ensure product quality, but also avoid the waste of time and resources caused by over-optimization.

[0237] If the second difference is greater than or equal to the difference threshold, the initial equipment parameters corresponding to the global correction factors of the third priority are optimized to obtain the third equipment parameters, and the third equipment parameters are output as the optimal equipment parameters. The global correction factors of the third priority are usually parameters that have a relatively small but still important impact on product quality determined in the previous analysis, such as certain auxiliary process parameters or environmental factors. The optimization process can use a similar method to the first two steps, but considering that this is the last step of optimization, a more complex or comprehensive optimization algorithm will be used, such as a genetic algorithm or a particle swarm optimization algorithm. Taking the genetic algorithm as an example, the specific implementation steps are as follows:

[0238] 1. Encoding: Encode the third priority parameter into chromosomes.

[0239] 2. Initialize the population: Generate a set of initial solutions (usually 50-100) based on the current parameter values.

[0240] 3. Fitness evaluation: Define the fitness function, which is usually the difference between the product quality parameters and the standard.

[0241] 4. Selection: Select excellent individuals based on fitness.

[0242] 5. Crossover and mutation: Generate new parameter combinations through crossover operations, and introduce new properties through mutation operations.

[0243] 6. Repeat steps 3-5 until the termination condition is reached (such as the number of iterations reaches the preset value or there is no significant improvement in the optimal solution for multiple consecutive generations).

[0244] For example, suppose the global correction factor of the third priority is the grinding time, and the initial value is 60 minutes. Through optimization, the following results are obtained: Optimal for the first generation: 65 minutes (fitness: 0.85); Optimal for the fifth generation: 72 minutes (fitness: 0.92); Optimal for the tenth generation: 75 minutes (fitness: 0.97); Optimal for the fifteenth generation: 78 minutes (fitness: 0.99). In the final third equipment parameter, the grinding time is optimized to 78 minutes. The effect of this step is to make the final fine-tuning and improvement of product quality by optimizing the third priority parameters. Although the impact of this parameter is relatively small, it is a key factor in meeting or exceeding the quality standards based on the optimization of the first two steps. At the same time, this step also reflects the comprehensiveness and depth of the optimization process, ensuring that all important parameters are fully optimized. Outputting the third equipment parameters as the optimal equipment parameters means that the optimization process ends here. This final output parameter combination should be the best solution after multi-level and multi-angle optimization, which can ensure product quality while taking into account production efficiency and cost control.

[0245] The present application also discloses a mica material production optimization system based on big data analysis, which is characterized by comprising:

[0246] a memory configured to store instructions; and

[0247] The processor is configured to call instructions from the memory and implement the above-mentioned mica material production optimization method based on big data analysis when executing the instructions.

[0248] Among them, the processor can adopt a central processing unit (CPU). Of course, according to actual usage, other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. can also be adopted. The general-purpose processor can adopt a microprocessor or any conventional processor, etc., and this application does not impose any restrictions on this.

[0249] Among them, the memory can be an internal storage unit of a computer device, such as a hard disk or memory of a computer device, or an external storage device of a computer device, such as a plug-in hard disk, a smart memory card (SMC), a secure digital card (SD) or a flash memory card (FC) equipped on the computer device, etc., and the memory can also be a combination of an internal storage unit and an external storage device of a computer device. The memory is used to store computer programs and other programs and data required by the computer device. The memory can also be used to temporarily store data that has been output or is to be output, and this application does not impose any restrictions on this.

[0250] An embodiment of the present application also provides a machine-readable storage medium having instructions stored thereon, the instructions being used to enable a machine to execute the above-mentioned method for optimizing mica material production based on big data analysis.

[0251] Those skilled in the art will appreciate that the embodiments of the present application may be provided as methods, systems or computer program products. Therefore, the present application may adopt the form of a complete hardware embodiment, a complete software embodiment or an embodiment in combination with software and hardware. Moreover, the present application may adopt the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) that contain computer-usable program code.

[0252] The present application is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems), and computer program products according to the embodiments of the present application. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of the processes and / or boxes in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0253] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate 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 A function specified in one or more boxes.

[0254] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process in the computer or other programmable device. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0255] In a typical configuration, a computing device includes one or more processors (CPU), input / output interfaces, network interfaces, and memory.

[0256] The memory may include non-permanent memory in a computer-readable medium, random access memory (RAM) and / or non-volatile memory in the form of read-only memory (ROM) or flash RAM. The memory is an example of a computer-readable medium.

[0257] Computer readable media include permanent and non-permanent, removable and non-removable media that can be implemented by any method or technology to store information. Information can be computer readable instructions, data structures, program modules or other data. Examples of computer storage media include, but are not limited to, phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technology, compact disk read-only memory (CD-ROM), digital versatile disk (DVD) or other optical storage, magnetic cassettes, magnetic tape disk storage or other magnetic storage devices or any other non-transmission media that can be used to store information that can be accessed by a computing device. As defined herein, computer readable media does not include temporary computer readable media (transitory media), such as modulated data signals and carrier waves.

[0258] It should also be noted that the terms "include", "comprises" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, method, commodity or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, commodity or device. In the absence of more restrictions, the elements defined by the sentence "comprises a ..." do not exclude the existence of other identical elements in the process, method, commodity or device including the elements.

[0259] The above are only embodiments of the present application and are not intended to limit the present application. For those skilled in the art, the present application may have various changes and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application should be included within the scope of the claims of the present application.

Claims

1. A mica material production optimization method based on big data analysis, characterized in that: The method comprises the following steps: Obtain the mica processing product type of the current material order through the mica material order system, and select the samples to be processed of the same type as the mica processing product type; Retrieving historical processing data and initial equipment parameters of the same type as the mica processing product type from a preset mica processing database, wherein the initial equipment parameters are equipment parameters of all mica processing equipment in the mica processing technology corresponding to the mica processing product type; Based on the initial equipment parameters, the sample to be processed is processed by the mica processing equipment, and the mica processing data of the sample to be processed during the processing is collected by using multiple types of sensors preset in all the mica processing equipment, and the sample quality parameters of the sample to be processed after processing are detected; Classifying and analyzing the timing characteristics of the mica processing data, and screening out the core process parameters in the initial equipment parameters according to the analysis results of the timing characteristics analysis; Calculate the average value of the mica historical processing data based on the core process parameters, and use the average value as the core process standard value corresponding to the core process parameters; Combining the core process standard value with the mica processing data, calculating the mica raw material loss rate during the mica processing process; The core process parameters are selected by calculating the correlation coefficient between the mica raw material loss rate and the core process parameters, and the selected core process parameters are output as local correction factors; Obtaining mica product quality parameters corresponding to the mica processing product type through a preset database, calculating an influence degree data set between the initial equipment parameters and the mica product quality parameters, prioritizing the initial equipment parameters based on the influence degree data set, and outputting the initial equipment parameters that have completed the priority grading as a global correction factor; The initial equipment parameters are corrected in combination with the local correction factor and the global correction factor to obtain optimal equipment parameters.

2. The method according to claim 1, characterized in that The step of classifying and analyzing the timing characteristics of the mica processing data, and screening out the core process parameters in the initial equipment parameters according to the analysis results of the timing characteristics analysis comprises the following steps: Feature classification of the mica processing data is performed according to the sensor types of the multi-type sensors to obtain mica weight change data, mica temperature change data, and mica pH change data; An upper limit of the total number of inflection points is set according to the total number of processing steps included in the mica processing process, the upper limit of the total number of inflection points is used as an algorithm constraint, and an inflection point detection algorithm is used to perform inflection point detection on the mica weight change data to obtain weight change inflection points distributed in the mica weight change data; Dividing the mica processing technology into process nodes according to the weight change inflection point to obtain a plurality of mica processing sub-processes within the mica processing technology; Extracting the maximum temperature and the minimum temperature of the mica temperature change data in each mica processing sub-process, calculating the difference between the maximum temperature and the minimum temperature, and obtaining the temperature difference; Extracting the maximum pH value and the minimum pH value of the mica pH change data in each mica processing sub-process, calculating the difference between the maximum pH value and the minimum pH value, and obtaining the pH difference; If the temperature difference corresponding to the mica processing sub-process is greater than a preset temperature difference threshold, analyzing whether the pH difference corresponding to the mica processing sub-process is greater than a preset pH difference threshold; If the pH difference corresponding to the mica processing sub-process is greater than the pH difference threshold, it is determined that the initial equipment parameters corresponding to the mica processing sub-process are mica purification parameters; If the pH difference is less than or equal to the pH difference threshold, determining that the initial equipment parameter corresponding to the mica processing sub-process is a mica grinding parameter; The mica purification parameters and the mica grinding parameters are integrated to obtain core process parameters.

3. The method according to claim 2, characterized in that The step of calculating the average value of the mica historical processing data based on the core process parameters and using the average value as the core process standard value corresponding to the core process parameters comprises the following steps: Extracting historical purification weight change data from the historical processing data corresponding to the mica purification parameters to obtain historical purification weight change data, calculating the purification weight average value of the historical purification weight change data, and using the purification weight average value as the purification process standard value; Extracting historical grinding weight change data from the historical processing data corresponding to the mica grinding parameters to obtain historical grinding weight change data, calculating the grinding weight average value of the historical grinding weight change data, and using the grinding weight average value as a grinding process standard value; The standard values ​​of the purification process and the standard values ​​of the grinding process are integrated to obtain the standard values ​​of the core process parameters.

4. The method according to claim 3, characterized in that: The steps of selecting the core process parameters by calculating the correlation coefficient between the mica raw material loss rate and the core process parameters, and outputting the selected core process parameters as local correction factors are as follows: Calculating a core parameter correlation coefficient set between the mica core process parameter and the mica raw material loss rate, screening out a core parameter correlation coefficient greater than a preset core parameter correlation coefficient threshold value in the core parameter correlation coefficient set, and outputting the core process parameter corresponding to the core parameter correlation coefficient as a local correction factor; According to multiple types of sensors preset in the mica processing plant, mica processing environment data is obtained, wherein the mica processing environment data includes mica processing environment temperature data and mica processing environment humidity data; Establishing a multiple linear regression equation with the mica raw material loss rate as the dependent variable, inputting the mica processing environment temperature data and the mica processing environment humidity data as independent variables into the multiple linear regression equation, and calculating the environment temperature correlation coefficient and the environment humidity correlation coefficient; If the environmental temperature correlation coefficient is greater than a preset environmental correlation coefficient threshold, the mica processing environmental temperature data is output as a local correction factor; If the environmental humidity correlation coefficient is greater than the environmental correlation coefficient threshold, the mica processing environmental humidity data is output as a local correction factor.

5. The method according to claim 4, characterized in that The steps of calculating the core parameter correlation coefficient set between the mica core process parameter and the mica raw material loss rate, screening out the core parameter correlation coefficient set that is greater than a preset core parameter correlation coefficient threshold, and outputting the core process parameter corresponding to the core parameter correlation coefficient as a local correction factor include the following steps: A correlation coefficient model is established with the mica raw material loss rate as an independent variable; The vibration frequency, grinding time and grinding method included in the mica grinding parameters are respectively input into the correlation coefficient model as independent variables to obtain a grinding correlation coefficient set; Screening out the grinding correlation coefficients in the grinding correlation coefficient set that are greater than a preset correlation coefficient threshold, and outputting the mica grinding parameters corresponding to the grinding correlation coefficients as local correction factors; The solution ratio, purification time and purification mesh parameters included in the mica purification parameters are respectively input into the correlation coefficient model as independent variables to obtain a purification correlation coefficient set; The purification correlation coefficients in the purification correlation coefficient set that are greater than a preset correlation coefficient threshold are screened out, and the mica purification parameters corresponding to the purification correlation coefficients are output as local correction factors.

6. The method according to claim 1, characterized in that The method of obtaining mica product quality parameters corresponding to the mica processed product type through a preset database, calculating an influence degree data set between the initial equipment parameters and the mica product quality parameters, prioritizing the initial equipment parameters based on the influence degree data set, and outputting the initial equipment parameters after priority grading as global correction factors comprises the following steps: Obtaining mica product quality parameters corresponding to the mica processed product type through a preset database; Establishing a mica regression model with the mica product quality parameter as the dependent variable and the initial equipment parameter as the independent variable; The mica equipment parameters, mica raw material data, and mica ingredient ratio included in the initial equipment parameters are respectively input as independent variables into the mica regression model to obtain influence degree data, and the influence degree data are integrated to obtain an influence degree data set; Outputting the initial device parameter corresponding to the largest impact data in the impact data set as a global correction factor of the first priority; Outputting the initial device parameter corresponding to the second largest impact degree data in the impact degree data set as a global correction factor of the second priority; The initial device parameters corresponding to the third largest impact data in the impact data set are output as global correction factors of the third priority, the first priority is greater than the second priority, and the second priority is greater than the third priority.

7. The method according to claim 6, characterized in that The correcting the initial equipment parameters by combining the local correction factor and the global correction factor to obtain the optimal equipment parameters comprises the following steps: After analyzing the local correction factors, locally correct the core process parameters and mica processing environment data included in the primary equipment parameters; The initial device parameters that have been partially corrected are globally corrected according to the global correction factors to obtain optimal device parameters.

8. The method according to claim 7, characterized in that The step of globally correcting the initial device parameters after local correction according to the global correction factor to obtain the optimal device parameters comprises the following steps: Acquire mica product quality standard parameters corresponding to the mica processed product type according to the preset database; Calculating the standard deviation between the mica product quality parameter and the mica product quality standard parameter; Optimizing the initial device parameters corresponding to the global correction factor of the first priority to obtain first device parameters; Calculating a first difference between a mica product quality parameter obtained by processing according to the first equipment parameter and the mica production standard quality parameter, and if the first difference is less than a preset difference threshold, outputting the first equipment parameter as an optimal equipment parameter; If the first difference is greater than or equal to the difference threshold, optimizing the initial device parameters corresponding to the global correction factor of the second priority to obtain second device parameters; Calculating a second difference between a mica product quality parameter obtained by processing according to the second equipment parameter and the mica production standard quality parameter, and if the second difference is less than the difference threshold, outputting the second mica equipment parameter as an optimal equipment parameter; If the second difference is greater than or equal to the second difference threshold, the initial device parameters corresponding to the global correction factor of the third priority are optimized to obtain third device parameters, and the third device parameters are output as optimal device parameters.

9. A mica material production optimization system based on big data analysis, characterized in that: include: a memory configured to store instructions; as well as A processor is configured to call the instructions from the memory and to implement the mica material production optimization method based on big data analysis according to any one of claims 1 to 8 when executing the instructions.

10. A computer-readable storage medium having instructions stored thereon, characterized in that: When the instruction is executed by a processor, the processor is configured to execute the mica material production optimization method based on big data analysis according to any one of claims 1 to 8.

Images