Intelligent optimization system for peanut oil production control process

Abstract

The invention relates to the technical field of peanut oil production control, in particular to an intelligent optimization system for a peanut oil production control process, which comprises a process parameter modeling module for collecting peanut oil production parameters, performing linear combination on the peanut oil production parameters, analyzing the correlation of the peanut oil production parameters to quality indexes, and calculating the quality indexes of the peanut oil production parameters; and obtaining a process parameter mapping matrix. According to the method, through deep correlation analysis of multiple parameters in the production process, the mapping relation between the process parameters and the oil quality indexes is established, and the precision and reliability of quality control are improved. In the process control, the specific influence of each production link on the quality index is accurately reflected through standardized and linear combinatorial analysis on the parameters such as squeezing temperature, pressure, time, stirring speed and the like. By predicting the quality index under the current production condition, the response speed to the quality change is improved, so that the quality index deviation can be quickly analyzed and evaluated in real-time data monitoring.

Description

Technical Field

[0001] The invention relates to the technical field of peanut oil production control, and in particular to an intelligent optimization system for a peanut oil production control process. Background Art

[0002] Peanut oil production control technology involves the control and optimization of each production link from raw material processing, pressing and extraction, refining to packaging and storage. By controlling key production parameters such as temperature, pressure, humidity and time, the purity and nutritional content of the oil can be ensured, and impurities and harmful substances in the processing can be reduced. By using data collection and real-time monitoring, the effect of precise control and reduction of resource waste can be achieved, thus realizing an efficient and environmentally friendly production process.

[0003] Among them, the intelligent optimization system of the peanut oil production control process uses intelligent algorithms and data analysis methods to conduct real-time monitoring and automatic adjustment of each link in the peanut oil production, helping production companies to ensure the quality of oil products.

[0004] The existing technology is difficult to achieve in-depth correlation analysis of different process parameters in peanut oil production control, resulting in oil quality control relying on qualitative judgments of a small number of parameters, which cannot fully reflect the overall impact of each production link on oil quality, affecting the comprehensiveness of quality control. In the quality control process, when faced with real-time production data, it is difficult to quickly predict and analyze quality indicators, resulting in a delayed response to abnormal situations, which can easily lead to product quality fluctuations or substandard products. In terms of abnormal parameter processing, if certain parameters deviate from the standard value, the production process is difficult to adjust adaptively, which can easily lead to problems such as abnormal load increase of equipment, increased energy consumption, and equipment failure. At the same time, the existing technology fails to effectively regulate the dynamic balance of multiple parameters under load conditions, resulting in minor factors interfering with the optimization of key parameters under complex production conditions, making it difficult for the overall production efficiency to reach the expected level. Summary of the invention

[0005] The purpose of the present invention is to solve the shortcomings in the prior art and to propose an intelligent optimization system for the peanut oil production control process.

[0006] In order to achieve the above object, the present invention adopts the following technical scheme: an intelligent optimization system for peanut oil production control process comprises:

[0007] The process parameter modeling module collects peanut oil production parameters, analyzes the correlation between peanut oil production parameters and quality indicators by linearly combining the peanut oil production parameters, and obtains the process parameter mapping matrix;

[0008] The real-time data acquisition and prediction module maps the data of the process parameter mapping matrix acquired in real time, obtains the predicted value of the oil quality index under the current production conditions by solving the mapping relationship, and compares it with the expected quality standard to obtain the deviation evaluation result;

[0009] The dynamic parameter smoothing module compares the current value of the peanut oil production parameter with the preset threshold range, smoothes the parameters that are out of range, and obtains the smoothing parameter setting value. If the parameter is within the normal range, combined with the deviation evaluation result, the stable parameter setting value is obtained for the parameter within the normal range. Combined with the current peanut oil production status, it is determined whether load balancing is required. If necessary, load balancing calculation is performed on the key parameters to generate load balancing parameters.

[0010] The process optimization and adjustment module refines and adjusts the peanut oil production parameters based on the smoothing parameter setting value or the stable parameter setting value, obtains the optimized process parameter setting value, integrates the optimized process parameter setting value and the load balancing parameter, and obtains the peanut oil production control global parameter data processing result.

[0011] As a further solution of the present invention, the step of obtaining the linear combination of peanut oil production parameters is specifically as follows:

[0012] Based on the collected peanut oil production parameters, including pressing temperature, pressing pressure, pressing time and stirring speed, the parameters are standardized and given inconsistent weights for linear combination to generate process parameter combination values;

[0013] Based on the process parameter combination values, the formula is used:

[0014]

[0015] Calculate the correlation coefficient r to obtain the correlation coefficient between process parameters and oil quality;

[0016] in, is the average value of Q, which represents the average process parameter combination value in multiple batches of production, and Z is the oil quality index value. It is the average value of Z, which represents the average oil quality index value in multiple batches of production.

[0017] As a further solution of the present invention, the step of obtaining the process parameter mapping matrix is ​​specifically as follows:

[0018] According to the correlation coefficient between the process parameters and the oil quality, the linear relationship between the process parameter combination and the oil quality index is judged, and the influence of the parameter adjustment on the quality index is judged by analyzing the value of the correlation coefficient, so as to obtain the analysis result of the relationship between the process parameter combination and the oil quality index;

[0019] According to the analysis results of the relationship between the process parameter combination and the oil quality index, the process parameters are set as row items and the quality indexes are set as column items, and the correlation coefficient of each quality index is filled into the corresponding cell of the matrix to obtain the process parameter mapping matrix.

[0020] As a further solution of the present invention, the step of obtaining the predicted value of the oil quality index under the current production conditions is specifically as follows:

[0021] Based on the data in the process parameter mapping matrix, the standardized value of each parameter is associated with the corresponding correlation coefficient, and a mapping process is performed to obtain the mapping value of the peanut oil production parameter;

[0022] According to the mapping values ​​of the peanut oil production parameters, the formula is used:

[0023] Y=b0+b1·X1+b2·X2+b3·X3+b4·X4

[0024] Calculate the predicted value Y of the oil quality index under the current production conditions to obtain the predicted result of the oil quality index;

[0025] Among them, b0 is the constant term in the regression equation, which is the benchmark level of the predicted value; b1, b2, b3, and b4 are regression coefficients used to measure the influence of each process parameter on the quality index; X1, X2, X3, and X4 are the mapping values ​​of the standardized process parameters.

[0026] As a further solution of the present invention, the step of obtaining the deviation evaluation result is specifically as follows:

[0027] Based on the predicted results of the oil quality indicators, the predicted values ​​of the oil quality indicators under the current production conditions are compared with the expected quality standards, and the deviations between the quality indicators including acid value, peroxide value and purity are analyzed to obtain the deviation value of each indicator;

[0028] Based on the deviation value of each indicator, the deviation value is compared with the specified standard range to determine whether the process parameters need to be adjusted, analyze the direction that needs to be optimized, and generate a deviation evaluation result.

[0029] As a further solution of the present invention, the steps of obtaining the smoothing parameter setting value and the stability parameter setting value are specifically as follows:

[0030] Based on the real-time collected peanut oil production parameter values, the current values ​​of the parameters are compared with the preset threshold ranges one by one to determine whether the parameters exceed the maximum or minimum allowable values, and obtain a list of parameters that exceed the range;

[0031] Based on the out-of-range parameter list, the formula is used:

[0032]

[0033] Calculate the smoothed process parameter values Get the smoothing parameter setting value;

[0034] Among them, X′ is the process parameter value currently collected in real time, λ is the smoothing factor, n is the number of data points collected in the past, and X′ i is the i-th process parameter value in the previously collected data set;

[0035] Based on the parameters that meet the threshold range, according to the deviation evaluation results, the deviation of the parameters is analyzed and its stability is judged. If the deviation of the parameter is within the allowable range, its current value is directly used as the stable set value to obtain the stable parameter set value.

[0036] As a further solution of the present invention, the step of obtaining the load balancing parameters is specifically as follows:

[0037] Based on the current peanut oil production status, check the key production parameters collected in real time, compare the current value of the parameter with the preset threshold range, and determine whether load balancing is required. If the parameter exceeds the preset range, it is marked as a load state, and the determination result of the load balancing demand is obtained;

[0038] According to the determination result of the load balancing requirement, the formula is adopted:

[0039]

[0040] Calculate the load adjustment value h after balancing balance,j , generate load balancing parameters;

[0041] Among them, h j is the value of the current production parameter j collected, w j is the weight coefficient of the key parameter j, h ref,j is the reference value of parameter j, α j is the adjustment coefficient of parameter j.

[0042] As a further solution of the present invention, the steps for obtaining the peanut oil production control global parameter data processing results are specifically as follows:

[0043] Based on the smoothing parameter setting value or the stabilization parameter setting value, combined with the predicted value of the oil quality index under the current production conditions, the influence of each production parameter on the oil quality is evaluated, and iterative optimization is performed to obtain the optimized process parameter setting value;

[0044] According to the optimized process parameter setting values, the adaptability of the load balancing parameters and the process parameter setting values ​​under load conditions is compared, and adaptive adjustments are made to obtain the global parameter data processing results of peanut oil production control.

[0045] Compared with the prior art, the advantages and positive effects of the present invention are:

[0046] In the present invention, by deeply analyzing the correlation of multiple parameters in the production process, a mapping relationship between process parameters and oil quality indicators is established, thereby improving the accuracy and reliability of quality control. In process control, by standardizing and linearly combining parameters such as pressing temperature, pressure, time and stirring speed, the specific impact of each production link on quality indicators is accurately reflected. By predicting the quality indicators under the current production conditions, the reaction speed to quality changes is improved, so that in real-time data monitoring, the deviation of quality indicators can be quickly analyzed and evaluated, and the risk of product quality fluctuations caused by delayed reactions is reduced. In terms of abnormal parameter processing, parameters that exceed the preset threshold are smoothed to ensure the stability of production parameters, avoid the problems of increased equipment load and increased energy consumption caused by parameter deviations from standard values, and help to extend the stable operation cycle of the equipment. In addition, under load conditions, the load balancing operation of key parameters can effectively isolate the interference of secondary factors on key parameters under complex production conditions, thereby improving production efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0047] Figure 1 is a system flow chart of the present invention;

[0048] Figure 2 The flowchart of the present invention for linearly combining peanut oil production parameters;

[0049] Figure 3 A flow chart for obtaining a process parameter mapping matrix for the present invention;

[0050] Figure 4 A flow chart for obtaining the predicted value of the oil quality index under the current production conditions of the present invention;

[0051] Figure 5 A flow chart for obtaining deviation assessment results for the present invention;

[0052] Figure 6 A flow chart for obtaining smoothing parameter setting values ​​and stabilization parameter setting values ​​for the present invention;

[0053] Figure 7 A flow chart for obtaining load balancing parameters for the present invention;

[0054] Figure 8 The present invention is a flow chart for obtaining the processing results of global parameter data for peanut oil production control. DETAILED DESCRIPTION

[0055] In order to make the purpose, technical solution and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0056] In the description of the present invention, it should be understood that the terms "length", "width", "up", "down", "front", "back", "left", "right", "vertical", "horizontal", "top", "bottom", "inside", "outside" and the like indicate positions or positional relationships based on the positions or positional relationships shown in the drawings, and are only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore cannot be understood as limiting the present invention. In addition, in the description of the present invention, "multiple" means two or more, unless otherwise clearly and specifically defined.

[0057] See also Figure 1 , a peanut oil production control process intelligent optimization system comprises:

[0058] The process parameter modeling module collects the peanut oil production parameters of pressing temperature, pressing pressure, pressing time and stirring speed, and performs standardization to make the parameters have the same scale. The partial least squares regression algorithm is used to linearly combine the peanut oil production parameters, analyze the correlation between the peanut oil production parameters and the acid value, peroxide value, and purity oil quality indicators, and obtain the process parameter mapping matrix;

[0059] The real-time data acquisition and prediction module maps the data of the process parameter mapping matrix collected in real time through multi-dimensional vector operations, and obtains the predicted value of the oil quality index under the current production conditions by constructing a regression equation and solving the mapping relationship. It compares it with the expected quality standard, analyzes the deviation of the quality index, and obtains the deviation evaluation result;

[0060] The dynamic parameter smoothing module compares the current value of the peanut oil production parameter with the preset threshold range to determine whether the current peanut oil production parameter exceeds the preset threshold. The preset threshold is the maximum and minimum allowable values ​​of the pressing temperature, pressing pressure, pressing time and stirring speed. The Laplace smoothing algorithm is used to smooth the parameters that exceed the range to obtain the smoothing parameter setting value. If the parameter is within the normal range, combined with the deviation evaluation result, the stable parameter setting value is obtained for the parameter within the normal range. Combined with the current peanut oil production status, it is determined whether load balancing is required. If necessary, a multiple interpolation algorithm is used to perform load balancing calculations on key parameters to generate load balancing parameters.

[0061] The process optimization and adjustment module makes detailed adjustments to the peanut oil production parameters based on the smoothing parameter setting value or the stable parameter setting value, combined with the predicted value of the oil quality index under the current production conditions, and evaluates the influence of the parameters on the oil quality by adding or removing independent variables, retains the key parameters affecting the oil quality, obtains the optimized process parameter setting value, integrates the optimized process parameter setting value and the load balancing parameter, and obtains the global parameter data processing result of the peanut oil production control;

[0062] The process parameter mapping matrix includes the linear combination parameters of pressing temperature, pressing pressure, pressing time and stirring speed, as well as the weight distribution of their influence on acid value, peroxide value and purity. The deviation evaluation results include acid value deviation value, peroxide value deviation value and purity deviation value, and are refined into the deviation range and trend analysis of each indicator. The load balancing parameters include load distribution adjustment values ​​of pressing temperature, pressing pressure, pressing time and stirring speed, as well as the balance coefficient applicable to the load state. The global parameter data processing results of peanut oil production control include the optimized pressing temperature setting value, pressing pressure setting value, pressing time setting value and stirring speed setting value, in addition to the parameter selection that has a significant impact on key quality indicators.

[0063] See also Figure 2 , the specific steps for obtaining the linear combination of peanut oil production parameters are:

[0064] Based on the collected peanut oil production parameters, including pressing temperature, pressing pressure, pressing time and stirring speed, the parameters are standardized and given inconsistent weights for linear combination to generate process parameter combination values;

[0065] The collected peanut oil production parameters such as pressing temperature, pressing pressure, pressing time and stirring speed are read and transmitted to the system in turn through the collection equipment. In order to make the data have a unified dimension, the collected parameter data are unified by a standardization method to remove the differential effects of the parameters due to different dimensions. In the standardization process, for example, the pressing temperature is standardized to the [0,1] interval, that is, the original temperature value T min and T max Corresponding to 0 and 1 respectively, through the formula After conversion to standardized values, the pressing pressure is also divided into the [0,1] interval using a similar formula, while the pressing time is standardized to the [0,10] range, and the stirring speed is standardized to the [1,1] interval. After standardization, the parameters are integrated through linear combination. For example, the pressing temperature (T), pressing pressure (P), pressing time (t), and stirring speed (S) are assigned weight coefficients a, b, c, and d for linear weighted combination. The weight coefficient is adjusted by analyzing the impact of production data on oil quality. For example, by setting multiple groups of different weights through experiments, observing the impact of different weight combinations on acid value, in order to determine the appropriate weight configuration, the combination formula is obtained: combination value Q = a×T+b×P+c×t+d×S. The combination process is to take the standardized values ​​of each parameter in turn, multiply them by the corresponding weight coefficients and add them together, and finally obtain the combined process parameter value Q.

[0066] Based on the process parameter combination value, the formula is used:

[0067]

[0068] Calculate the correlation coefficient r to obtain the correlation coefficient between process parameters and oil quality;

[0069] Among them, r represents the linear correlation strength between the process parameter combination value and the oil quality index, which is used to judge the overall combination effect of the process parameters. is the average value of Q, which represents the average process parameter combination value in multiple batches of production. Z is the oil quality index value, which is obtained through actual testing, including parameters such as acid value, peroxide value and purity. Oil quality testing can be carried out by standard experimental methods (such as titration and chromatography). For example, the acid value is determined by standard titration method, and the acid value values ​​of multiple batches of oil products are recorded. It is the average value of Z, which indicates the average oil quality index value in multiple batches of production, and is used to compare the deviation between a single production and the overall quality level. It is the cumulative sum of the product of Q and Z deviations, and is used to measure the degree of linear deviation between process parameter values ​​and quality indicators. is the sum of squares of the deviations of the process parameter combination values, indicating the fluctuation range of the process parameters. It is the sum of square deviations of oil quality indicators and is used to assess the fluctuation range of production quality.

[0070] If the process parameter combination value Q is 54.75, the average of the combination values ​​of multiple batches is The actual measured oil quality index value Z is 55, and the average value of multiple batches of quality indicators is is 52.

[0071] Calculate the correlation coefficient:

[0072]

[0073] This result indicates that the correlation coefficient r is 1.

[0074] See also Figure 3 , the specific steps for obtaining the process parameter mapping matrix are:

[0075] According to the correlation coefficient between process parameters and oil quality, the linear relationship between process parameter combination and oil quality index is judged. By analyzing the value of the correlation coefficient, the influence of parameter adjustment on quality index is judged, and the analysis result of the relationship between process parameter combination and oil quality index is obtained.

[0076] A correlation coefficient of 1 indicates that there is a completely linear relationship between the process parameter combination value and the oil quality index. When r = 1, it means that there is a completely positive linear relationship between the variables, that is, the increase of one variable will lead to the increase of another variable at a constant ratio. This relationship means that the data points fall completely on a straight line with a positive slope. When r = -1, it means that there is a completely negative linear relationship, that is, the increase of one variable will lead to the decrease of another variable at a constant ratio, and the data points fall on a straight line with a negative slope. When r = 0, it means that there is no linear relationship, and there is no constant positive or negative linear trend between the two variables. Therefore, a correlation coefficient of 1 means that the linear fitting effect of the variables is optimal, and all data points completely match a straight line with a positive slope without any deviation. This value shows that under the current process combination, the relationship between the production parameters of peanut oil (such as pressing temperature, pressure, time, etc.) and the oil quality indicators (such as acid value, peroxide value, purity) is completely linear, that is, the adjustment of parameters directly leads to proportional changes in quality indicators without nonlinear fluctuations or randomness. Therefore, the result of a correlation coefficient of 1 indicates that the current combination of process parameters is stable and completely predictable, which helps to accurately control the quality indicators of oil products in production.

[0077] According to the analysis results of the relationship between the process parameter combination and the oil quality index, the process parameters are set as row items and the quality index is set as column items. The correlation coefficient of each quality index is filled into the corresponding cell of the matrix to obtain the process parameter mapping matrix.

[0078] In peanut oil production, process parameters may include pressing temperature, pressing pressure, pressing time and stirring speed, while quality indicators usually include acid value, peroxide value and purity. The purpose of the mapping matrix is ​​to show in detail the contribution ratio or correlation coefficient of each process parameter to each quality indicator. After determining the process parameters and quality indicators, the process parameters are set as the row items of the matrix, such as pressing temperature, pressing pressure, pressing time and stirring speed, etc., and the quality indicators acid value, peroxide value and purity are used as column items. In this way, each cell of the matrix represents the correlation or contribution value between a process parameter and a quality indicator. When filling the matrix, the correlation coefficient of each process parameter to each quality indicator needs to be calculated one by one. For example, the correlation coefficient between pressing temperature and acid value is calculated to be 0.85, which indicates that pressing temperature has a high influence on acid value. Therefore, 0.85 is filled in the intersection of pressing temperature and acid value in the matrix as a specific value. Similarly, the correlation coefficient of pressing pressure to acid value is 0.75, which is also filled in the corresponding position of pressing pressure and acid value in the matrix. Following this process, continue to fill in the combination relationship of other process parameters and quality indicators. For example, the correlation coefficient between pressing temperature and peroxide value is 0.6, and the correlation coefficient between stirring speed and purity is 0.9. These values ​​are filled in the corresponding matrix cells one by one. After completing the correlation filling of all parameters, the mapping matrix is ​​a structural diagram that comprehensively shows the impact of process parameters on quality indicators. This matrix not only provides an analysis of the contribution of each quality indicator (such as acid value, peroxide value and purity), but also provides a basis for production adjustments, so that producers can clearly identify the key parameters that affect different quality indicators and accurately adjust the parameters in the subsequent production process. The process parameter mapping matrix constructed in this way can support production control, so that the quality of oil products can reach stable standards.

[0079] See also Figure 4 , the steps for obtaining the predicted values ​​of oil quality indicators under current production conditions are as follows:

[0080] Based on the data in the process parameter mapping matrix, the standardized value of each parameter is associated with the corresponding correlation coefficient, and a mapping process is performed to obtain the mapping value of the peanut oil production parameter;

[0081] For the process parameter mapping matrix data collected in real time, including the standardized production parameters of pressing temperature, pressing pressure, pressing time and stirring speed, and the correlation coefficients of these parameters to oil quality indicators (such as acid value, peroxide value and purity), mapping processing is performed. The specific mapping process is to multiply each standardized parameter value collected in real time with the corresponding correlation coefficient to obtain the mapping value of the parameter to the target quality indicator. For example, assuming that the standardized pressing temperature value is 0.8 and the correlation coefficient is 0.85, its mapping value is 0.8×0.85=0.68. Similarly, assuming that the standardized pressing pressure is 0.9 and the correlation coefficient is 0.75, the mapping value is 0.9×0.75=0.675. According to this method, the standardized values ​​of all process parameters are multiplied one by one with their respective correlation coefficients to generate the mapping value of each parameter. Finally, these mapping values ​​will be used as input variables in the regression equation for the next step of predicting oil quality indicators.

[0082] According to the mapping values ​​of peanut oil production parameters, the formula is used:

[0083] Y=b0+b1·X1+b2·X2+b3·X3+b4·X4

[0084] Calculate the predicted value Y of the oil quality index under the current production conditions to obtain the predicted result of the oil quality index;

[0085] Among them, Y is used to estimate the quality indicators under the current process conditions. A regression model is constructed based on the previously collected production data and its corresponding quality inspection results. The process parameters are combined into a predicted value Y for quality control. b0 is the constant term in the regression equation, which is the benchmark level of the predicted value. It reflects the basic quality indicators when there are no other process parameters. It is obtained through the process of regression analysis. It usually depends on multiple batches of previously collected data and is determined based on multiple experiments or actual production data. For example, through 10 batches of production data, when all other parameter standardized values ​​are 0, the constant term is obtained as the basic predicted value. b1, b2, b3, and b4 are regression coefficients. In order to measure the influence of each process parameter on the quality index, each regression coefficient is obtained based on the previously collected data fitting of each process parameter and quality index, reflecting the weight of each parameter. The process of determining these coefficients is usually carried out through partial least squares regression (PLS) and other methods. For example, multiple batches of data are used to analyze the linear relationship between pressing temperature and acid value to obtain the value of b1. Other regression coefficients are calculated in a similar way. X1, X2, X3, and X4 are the mapping values ​​of the standardized process parameters, corresponding to pressing temperature, pressing pressure, pressing time, and stirring speed, respectively. They have been standardized according to the actual collected data and the mapping values ​​have been obtained through correlation analysis.

[0086] For example, if the mapping process obtains the following standardized mapping values: X1 = 0.68 (pressing temperature mapping value), X2 = 0.675 (pressing pressure mapping value), X3 = 0.72 (pressing time mapping value), X4 = 0.8 (stirring speed mapping value), the regression coefficients are assumed to be b0 = 0.5, b1 = 1.2, b2 = 1.1, b3 = 0.9, b4 = 1.0, and are substituted into the formula:

[0087] Y=0.5+(1.2·0.68)+(1.1·0.675)+(0.9·0.72)+(1.0·0.8)

[0088] Y=0.5+0.816+0.7425+0.648+0.8=3.5065

[0089] The results show that the predicted acid value under the current process conditions is 3.5065.

[0090] See also Figure 5 , the specific steps for obtaining the deviation assessment results are:

[0091] Based on the prediction results of oil quality indicators, the predicted values ​​of oil quality indicators under current production conditions are compared with the expected quality standards, and the deviations between quality indicators including acid value, peroxide value and purity are analyzed to obtain the deviation value of each indicator;

[0092] Under the current production conditions, the predicted value of the oil quality index Y = 3.5065 calculated by the regression equation is compared with the expected quality standard to evaluate the effectiveness of the current process settings. Assuming that the quality standard of the expected acid value is 3.5, this standard value is a benchmark determined based on previously collected data or industry requirements to ensure the quality consistency and safety of the product. Comparing the predicted value Y = 3.5065 with the target value of 3.5, the deviation value is calculated to be 3.5065-3.5 = 0.0065. The deviation value is 0.0065, indicating that the production parameters under the current process are close to the target value and are within a reasonable range with the standard deviation. In order to further verify the rationality of the process settings, the predicted values ​​of other quality indicators can be referred to. For example, if the predicted peroxide value is 1.02 and the expected standard is 1.0, through similar deviation calculation 1.02-1.0 = 0.02, it can be concluded that the deviation of the peroxide value is 0.02. Similarly, assuming that the predicted value of purity is 99.8% and the target standard is 99.9%, the deviation of purity is 99.8-99.9=-0.1%. Through this comparison, the deviation of process parameters from each quality indicator can be evaluated item by item, so as to understand the overall quality performance under the current production conditions. The calculation results of the deviation value provide a quantitative basis for the subsequent deviation evaluation.

[0093] Based on the deviation value of each indicator, compare the deviation value with the specified standard range, determine whether the process parameters need to be adjusted, analyze the direction that needs to be optimized, and generate deviation evaluation results.

[0094] Based on the calculated deviation results, the deviation size of each quality indicator is analyzed to obtain the final deviation evaluation result. The process of deviation evaluation is to compare the deviation value of each indicator with the specified allowable range to determine whether the current process needs to be adjusted. For example, assuming that the allowable deviation range is ±0.01, for the acid value deviation of 0.0065, the deviation value is less than 0.01, indicating that the predicted value of the acid value is close enough to the standard value and no adjustment is required. Similarly, the deviation of the peroxide value is 0.02, which exceeds the allowable range of 0.01, which means that the predicted value of the indicator deviates greatly from the target value, and further adjustment of related process parameters is required, such as reducing the pressing temperature or adjusting the stirring speed to improve the peroxide value. In addition, if the deviation of purity is 0.1%, which also exceeds the allowable range, it should be pointed out in the evaluation results that the process parameters need to be optimized, such as extending the pressing time or controlling the stirring speed to improve the purity of the product. Through this item-by-item analysis of the deviation size, the specific impact of each process parameter on the quality indicator can be identified, and the direction that needs to be optimized can be obtained. The deviation evaluation results provide a basis for adjusting the production process to ensure that the quality of the oil products is stable and meets the set quality standards. The final evaluation process identifies priorities and refines measures for process adjustments, helping to continuously optimize production conditions to meet quality control requirements.

[0095] See also Figure 6 , the steps for obtaining the smoothing parameter setting value and the stability parameter setting value are as follows:

[0096] Based on the real-time collected peanut oil production parameter values, the current values ​​of the parameters are compared with the preset threshold ranges one by one to determine whether the parameters exceed the maximum or minimum allowable values, and obtain a list of parameters that exceed the range;

[0097] If the peanut oil currently being produced is in the process of pressing, the following real-time process parameters are collected: Pressing temperature: the current value is 145°C (the preset threshold range is 120-140°C) Pressing pressure: the current value is 85MPa (the preset threshold range is 80-90MPa) Pressing time: the current value is 35 minutes (the preset threshold range is 30-40 minutes) Stirring speed: the current value is 75rpm (the preset threshold range is 50-80rpm), by comparing each parameter value collected in real time with the preset threshold range one by one, it is determined whether each parameter is within the allowable range: For pressing temperature: the current value is 145°C, which exceeds the preset upper limit of 140°C, so the parameter exceeds the threshold range and needs to be smoothed. For pressing pressure: the current value is 85MPa, which is within the threshold range (80-90MPa), indicating that the parameter is within the normal range and no further processing is required. For pressing time: the current value is 35 minutes, which is also within the range of 30-40 minutes, and no adjustment is required. For the stirring speed: the current value is 75 rpm, which is also within the threshold range of 50-80 rpm and does not require further processing. Through this process, it is determined which parameters are within the allowable range and which parameters (i.e., pressing temperature) need to be smoothed out.

[0098] Based on the out-of-range parameter list, the formula is used:

[0099]

[0100] Calculate the smoothed process parameter values Get the smoothing parameter setting value;

[0101] in, It is the result of smoothing the original parameter value. It is used to replace the parameter value that initially deviates from the threshold value to ensure that the parameter value returns to the allowable range. X′ is the process parameter value currently collected in real time, which is obtained in real time through the sensor. λ is the smoothing factor, which controls the strength of smoothing. It is usually obtained through analysis of previously collected data. For example, if deviations exceeding the threshold value occur multiple times during the production process, it is found through analysis that a smoothing factor of 0.5 can effectively adjust the excess value to the allowable range. Then 0.5 is used as the set value. It is generally set based on the stability of fluctuations in previously collected data. n is the number of previously collected data points, which indicates the number of previously collected data used for smoothing calculations. The appropriate amount of previously collected batch data can be selected through system settings to ensure that the smoothing calculation is in line with the current production situation. X′ i is the i-th process parameter value in the previously collected data set, indicating the corresponding parameter value collected in the previously collected batches.

[0102] For example, if the real-time collected pressing temperature X' is 145°C, it exceeds the threshold range, so it needs to be smoothed. Take the smoothing factor λ = 0.5 and use the past five collected temperature values ​​(140°C, 138°C, 136°C, 135°C and 137°C) as X' i , enter the formula to calculate:

[0103] Calculate the sum of differences of previously collected data:

[0104]

[0105] Find the mean difference:

[0106]

[0107] Compute the smoothed value:

[0108]

[0109] The results show that the temperature value obtained after smoothing is 141.1℃, which has been adjusted back to the threshold range compared to 145℃ which exceeded the threshold.

[0110] Based on the parameters that meet the threshold range, according to the deviation evaluation results, the deviation of the parameters is analyzed and its stability is judged. If the deviation of the parameter is within the allowable range, its current value is directly used as the stable set value to obtain the stable parameter set value;

[0111] For other parameters that meet the threshold range (pressing pressure, pressing time, and stirring speed), smoothing is not required, but the set values ​​are further optimized in combination with the deviation evaluation results. For example, for pressing pressure: the deviation evaluation results show that the deviation of the pressure is less than 0.01, indicating that the parameter matches the quality standard. 85MPa can be directly set as a stable production set value. For pressing time: the deviation evaluation shows that the acid value fluctuation at the current pressing time is small, and the acid value is within the standard range. Therefore, the current 35 minutes can be used as a stable set value for pressing time. For stirring speed: the deviation evaluation results show that the deviation is 0.02. Although there are slight fluctuations, it is still within the allowable range, so a stirring speed of 75rpm is suitable as a stable production set value. Through this process, the following stable process parameter settings are obtained: Pressing temperature: After smoothing, it is adjusted to 141.1℃, which meets the threshold range. Pressing pressure: 85MPa (stable set value) Pressing time: 35 minutes (stable set value) Stirring speed: 75rpm (stable set value). The results show that through real-time monitoring, smoothing and deviation evaluation, each parameter can be kept within a reasonable range to ensure the stability of the production process and the quality of the oil.

[0112] See also Figure 7, the specific steps for obtaining load balancing parameters are:

[0113] Based on the current peanut oil production status, check the key production parameters collected in real time, compare the current value of the parameter with the preset threshold range, and determine whether load balancing is required. If the parameter exceeds the preset range, it is marked as a load state, and the determination result of the load balancing demand is obtained;

[0114] Combined with the current peanut oil production status, check each key production parameter (such as pressing temperature, pressing pressure, pressing time and stirring speed) collected in real time one by one, compare its current value with the preset threshold range, and judge whether to trigger the load state balancing process according to the load state conditions. The specific judgment process is as follows: if the current value of a parameter exceeds the upper limit of its preset threshold range, it is marked as "out of range"; if all parameters are within the normal range, the production is in a stable state and there is no need to trigger load balancing; if a key parameter (such as pressing temperature) exceeds the set threshold range, it is judged that the production state has reached the load state and subsequent load balancing operations need to be performed. For example, in one test, the collected pressing temperature is 150℃ (exceeding the set upper limit of 140℃), the pressing pressure is 85MPa (within the preset range of 80-90MPa), the pressing time is 32 minutes (within the preset range of 30-40 minutes), and the stirring speed is 77rpm (also in the preset range of 50-80rpm). Although the pressing pressure, time and stirring speed are within the standard range, the pressing temperature exceeds the preset upper limit, so it is determined that the production load has reached a peak and needs to enter the load balancing process. On the contrary, if all the collected parameters are within the preset threshold range, the current production state is maintained and no load balancing operation is required.

[0115] According to the load balancing requirement determination result, the formula is used:

[0116]

[0117] Calculate the load adjustment value h after balancing balance,j , generate load balancing parameters;

[0118] Among them, h balance,j is the target value of the key parameter j after balanced calculation under load, which is used to guide the actual adjustment of the parameter to ensure that the parameter is close to its reference value. j is the value of the current production parameter j, which represents the real-time data of the key parameter (such as pressing temperature, pressing pressure, pressing time or stirring speed) under load, acquired in real time through sensors or monitoring equipment. jis the weight coefficient of key parameter j, reflecting the influence of this parameter on the overall load balance. It is usually set by analyzing the production data collected in the past. If a parameter has a greater impact on the equipment operating load, a higher weight will be set to ensure that the parameter is more significantly balanced. ref,j It is the reference value of parameter j, that is, the set value under ideal or standard load conditions, reflecting the target value of the parameter in production. It is usually determined by previously collected data or equipment recommended values. For example, if previously collected data shows that the pressing temperature is stable at 140°C, 140°C can be set as the reference value of the parameter h. ref,j , α j It is the adjustment coefficient of parameter j, which is used to control the amplitude of balance adjustment. This coefficient reflects the degree of smoothness of parameter j during adjustment, and avoids drastic adjustment of parameters with large deviations. The adjustment coefficient is usually set according to the volatility of the parameter. For example, when the pressing temperature deviates greatly, a higher α can be set. j value to reduce its adjustment range and avoid load instability caused by rapid changes.

[0119] For example, if the current load state production parameters are: pressing temperature 150℃, pressing pressure 88MPa, pressing time 32 minutes, stirring speed 77rpm. The baseline value is set as: pressing temperature 140℃, pressing pressure 85MPa, pressing time 35 minutes, stirring speed 75rpm. The weights and adjustment coefficients of each parameter are as follows: temperature weight is 0.4, adjustment coefficient is 0.5; pressure weight is 0.3, adjustment coefficient is 0.3; time weight is 0.2, adjustment coefficient is 0.2; stirring speed weight is 0.1, adjustment coefficient is 0.1. Substitute the formula for calculation:

[0120] Load balancing value of temperature:

[0121]

[0122] Load balance value of pressure:

[0123]

[0124] Load balancing value of time:

[0125]

[0126] Load balance value of stirring speed:

[0127]

[0128] The calculated load balancing adjustment value h balance,jThe target adjustment values ​​of each key parameter are as follows, for example, the temperature is adjusted to 147.33℃, the pressure is adjusted to 87.31MPa, the time is adjusted to 32.5 minutes, and the stirring speed is adjusted to 76.82rpm. Through this balanced adjustment, each key parameter gradually approaches its reference value, thereby reducing the fluctuation under load state, so that the production system can maintain stability under load state conditions and ensure smooth operation of the equipment.

[0129] See also Figure 8 , the specific steps for obtaining the processing results of global parameter data for peanut oil production control are as follows:

[0130] Based on the smoothing parameter setting value or the stable parameter setting value, combined with the predicted value of the oil quality index under the current production conditions, the influence of each production parameter on the oil quality is evaluated, and iterative optimization is performed to obtain the optimized process parameter setting value;

[0131] First, evaluate the degree of influence of each production parameter on oil quality. During the evaluation process, each parameter is analyzed step by step to examine the sensitivity of its fluctuation to quality indicators such as acid value and peroxide value, and the deviation evaluation is used to determine whether the impact is significant. Specifically, when the fluctuation of a certain parameter causes the change in acid value or peroxide value to exceed the allowable range (such as when the acid value changes by more than 0.1 or the peroxide value changes by more than 0.5), it can be considered that the parameter has a significant impact on oil quality; if the change in a certain parameter improves the acid value by less than 0.05 or the peroxide value by less than 0.2, and the quality indicators have met the standard requirements, it can be considered that the parameter has little effect on improving oil quality, and it can be considered to be eliminated or the optimization priority can be reduced. During the optimization process, dynamic adjustments are made in combination with the deviation between the predicted value and the actual result. First, the difference between the actual collected value of the parameter and the model predicted value during real-time monitoring of the production process is used. For example, when the acid value is predicted to be 3.50 at a pressing temperature of 140°C, but the actual production results show that the acid value is 3.65, and the deviation reaches 0.15, it is necessary to further refine the control range (such as 135-145°C) by dynamically adjusting the temperature optimization direction to reduce the deviation. At the same time, the stability of parameters with smaller deviations is evaluated. For example, if the deviation of the actual stirring speed is within the allowable range (±2rpm), and the change in the acid value or peroxide value is not enough to significantly improve the quality of the oil, its current value can be directly set as a stable value to reduce further intervention in the optimization process. By continuously analyzing the deviation between the predicted value and the actual result and its changing trend, the optimization direction and parameter combination are dynamically adjusted to ensure that the optimization of key parameters can significantly improve the quality indicators, and finally determine a set of optimized process parameter setting values ​​to achieve the stability of oil quality and the efficiency of process control.

[0132] According to the optimized process parameter setting values, the adaptability of the load balancing parameters and the process parameter setting values ​​under load conditions is compared, and adaptive adjustments are made to obtain the global parameter data processing results of peanut oil production control;

[0133] After completing the optimization and adjustment of each process parameter, the optimized process parameter setting value is integrated with the load balancing parameter according to the process requirements of the equipment under different load conditions. First, the load balancing parameters are compared with the optimized process parameter setting values ​​to evaluate their adaptability under load conditions. This process mainly includes the following steps: Based on the smoothing and optimized process parameters (such as pressing temperature, pressing pressure, stirring speed, etc.), the ideal setting value of each parameter is clarified. For example, under normal production conditions, the reference value of the pressing temperature is 125°C, the pressure is 85MPa, and the stirring speed is 75rpm. These reference values ​​will be used as the reference standard for the adaptation evaluation under load conditions. Under the production load state, the real-time data of each parameter is collected, such as the current value of the temperature rises to 130°C, the pressure rises to 88MPa, etc. Based on these load state data, the balance value of each parameter is calculated by the load balancing formula to measure the actual load level of each key parameter. Compare the load balancing parameters with the set values ​​of each process parameter to check whether each parameter is within the allowable deviation range. For example, if the equilibrium value of the pressing temperature is 128°C, which is 3°C different from the set value of 125°C, and is within the allowable deviation range of the equipment (±5°C), it is considered that the temperature parameter is well adapted under load; if the equilibrium value of the pressing pressure is 90MPa, which exceeds the allowable upper limit, it is necessary to further adjust the set value or add pressure reduction measures in the equipment. By comparing the deviations of various process parameters and load balancing parameters, a compatibility assessment report is generated to reflect the overall adaptation under load. If a key parameter deviates too much and is not within the allowable range, it is recorded as unsuitable and the parameter is marked for further optimization. For unsuitable parameters, further adjust their set values ​​and incorporate the adjusted new set values ​​into the global control scheme. For example, lower the pressing pressure to 87MPa to adapt to the load state, keep the temperature at 128°C, etc.

[0134] The above are only preferred embodiments of the present invention and are not intended to limit the present invention in other forms. Any technician familiar with the profession may use the technical contents disclosed above to change or modify them into equivalent embodiments with equivalent changes and apply them to other fields. However, any simple modification, equivalent change and modification made to the above embodiments based on the technical essence of the present invention without departing from the technical solution of the present invention still falls within the protection scope of the technical solution of the present invention.

Claims

1. An intelligent optimization system for peanut oil production control process, characterized in that: The system comprises: The process parameter modeling module collects peanut oil production parameters, analyzes the correlation between peanut oil production parameters and quality indicators by linearly combining the peanut oil production parameters, and obtains the process parameter mapping matrix; The real-time data acquisition and prediction module maps the data of the process parameter mapping matrix acquired in real time, obtains the predicted value of the oil quality index under the current production conditions by solving the mapping relationship, and compares it with the expected quality standard to obtain the deviation evaluation result; The dynamic parameter smoothing module compares the current value of the peanut oil production parameter with the preset threshold range, smoothes the parameters that are out of range, and obtains the smoothing parameter setting value. If the parameter is within the normal range, combined with the deviation evaluation result, the stable parameter setting value is obtained for the parameter within the normal range. Combined with the current peanut oil production status, it is determined whether load balancing is required. If necessary, load balancing calculation is performed on the key parameters to generate load balancing parameters. The process optimization and adjustment module refines and adjusts the peanut oil production parameters based on the smoothing parameter setting value or the stable parameter setting value, obtains the optimized process parameter setting value, integrates the optimized process parameter setting value and the load balancing parameter, and obtains the peanut oil production control global parameter data processing result.

2. The peanut oil production control process intelligent optimization system according to claim 1, characterized in that: The steps of obtaining the linear combination of peanut oil production parameters are specifically as follows: Based on the collected peanut oil production parameters, including pressing temperature, pressing pressure, pressing time and stirring speed, the parameters are standardized and given inconsistent weights for linear combination to generate process parameter combination values; Based on the process parameter combination values, the formula is adopted: Calculate the correlation coefficient r to obtain the correlation coefficient between process parameters and oil quality; in, is the average value of Q, which represents the average process parameter combination value in multiple batches of production, and Z is the oil quality index value. It is the average value of Z, which represents the average oil quality index value in multiple batches of production.

3. The peanut oil production control process intelligent optimization system according to claim 2, characterized in that: The steps for obtaining the process parameter mapping matrix are specifically as follows: According to the correlation coefficient between the process parameters and the oil quality, the linear relationship between the process parameter combination and the oil quality index is judged, and the influence of the parameter adjustment on the quality index is judged by analyzing the value of the correlation coefficient, so as to obtain the analysis result of the relationship between the process parameter combination and the oil quality index; According to the analysis results of the relationship between the process parameter combination and the oil quality index, the process parameters are set as row items and the quality indexes are set as column items, and the correlation coefficient of each quality index is filled into the corresponding cell of the matrix to obtain the process parameter mapping matrix.

4. The peanut oil production control process intelligent optimization system according to claim 3, characterized in that: The steps for obtaining the predicted value of the oil quality index under the current production conditions are specifically as follows: Based on the data in the process parameter mapping matrix, the standardized value of each parameter is associated with the corresponding correlation coefficient, and a mapping process is performed to obtain the mapping value of the peanut oil production parameter; According to the mapping values ​​of the peanut oil production parameters, the formula is used: Y=b0+b1·X1+b2·X2+b3·X3+b4·X4 Calculate the predicted value Y of the oil quality index under the current production conditions to obtain the predicted result of the oil quality index; Among them, b0 is the constant term in the regression equation, which is the benchmark level of the predicted value; b1, b2, b3, and b4 are regression coefficients used to measure the influence of each process parameter on the quality index; X1, X2, X3, and X4 are the mapping values ​​of the standardized process parameters.

5. The peanut oil production control process intelligent optimization system according to claim 4, characterized in that: The steps for obtaining the deviation assessment result are specifically as follows: Based on the predicted results of the oil quality indicators, the predicted values ​​of the oil quality indicators under the current production conditions are compared with the expected quality standards, and the deviations between the quality indicators including acid value, peroxide value and purity are analyzed to obtain the deviation value of each indicator; Based on the deviation value of each indicator, the deviation value is compared with the specified standard range to determine whether the process parameters need to be adjusted, analyze the direction that needs to be optimized, and generate a deviation evaluation result.

6. The peanut oil production control process intelligent optimization system according to claim 5, characterized in that: The steps for obtaining the smoothing parameter setting value and the stability parameter setting value are specifically as follows: Based on the real-time collected peanut oil production parameter values, the current values ​​of the parameters are compared with the preset threshold ranges one by one to determine whether the parameters exceed the maximum or minimum allowable values, and obtain a list of parameters that exceed the range; Based on the out-of-range parameter list, the formula is used: Calculate the smoothed process parameter values Get the smoothing parameter setting value; Among them, X′ is the process parameter value currently collected in real time, λ is the smoothing factor, n is the number of data points collected in the past, and X′ i is the i-th process parameter value in the previously collected data set; Based on the parameters that meet the threshold range, according to the deviation evaluation results, the deviation of the parameters is analyzed and its stability is judged. If the deviation of the parameter is within the allowable range, its current value is directly used as the stable set value to obtain the stable parameter set value.

7. The peanut oil production control process intelligent optimization system according to claim 1, characterized in that: The steps for obtaining the load balancing parameters are specifically as follows: Based on the current peanut oil production status, check the key production parameters collected in real time, compare the current value of the parameter with the preset threshold range, and determine whether load balancing is required. If the parameter exceeds the preset range, it is marked as a load state, and the determination result of the load balancing demand is obtained; According to the determination result of the load balancing requirement, the formula is adopted: Calculate the load adjustment value h after balancing balance,j , generate load balancing parameters; Among them, h j is the value of the current production parameter j collected, w j is the weight coefficient of the key parameter j, h ref,j is the reference value of parameter j, α j is the adjustment coefficient of parameter j.

8. The peanut oil production control process intelligent optimization system according to claim 7, characterized in that: The steps for obtaining the peanut oil production control global parameter data processing results are specifically as follows: Based on the smoothing parameter setting value or the stabilization parameter setting value, combined with the predicted value of the oil quality index under the current production conditions, the influence of each production parameter on the oil quality is evaluated, and iterative optimization is performed to obtain the optimized process parameter setting value; According to the optimized process parameter setting values, the adaptability of the load balancing parameters and the process parameter setting values ​​under load conditions is compared, and adaptive adjustments are made to obtain the global parameter data processing results of peanut oil production control.

Images