CN-122018610-A - Rotary furnace parameter optimization recommendation method based on data analysis

CN122018610ACN 122018610 ACN122018610 ACN 122018610ACN-122018610-A

Abstract

The application relates to the technical field of parameter control, in particular to a rotary furnace parameter optimization recommendation method based on data analysis, which comprises the steps of obtaining the maximum allowable variation of each control parameter of a rotary furnace; and executing the iterative process of the differential evolution algorithm for multiple iterations until reaching the preset termination condition, and outputting the optimal recommended parameters. According to the technical scheme, key parameters which have obvious influence on the quality of the finished product can be focused in the iteration process, the accuracy of optimal technological parameter recommendation in a complex industrial baking scene is ensured, and the intelligent control of the rotary furnace is realized.

Inventors

WEI MIN
ZHOU DEWEN
LI CHANGXI
TANG HONGZHAO

Assignees

广州市赛思达机械设备有限公司

Dates

Publication Date: 20260512
Application Date: 20260408

Claims (10)

1. The rotary furnace parameter optimization recommendation method based on data analysis is characterized by comprising the steps of obtaining the maximum allowable variation of each control parameter of the rotary furnace; generating an initial population comprising a plurality of parameter vectors; The iterative process of the differential evolution algorithm is implemented, wherein the iterative process comprises the steps of selecting a parent vector from a population for any target vector to calculate a differential vector, limiting the scaling degree of the differential vector according to the maximum allowable variation to obtain an effective variation scaling factor of each control parameter and generate a variation vector, acquiring the accumulated sensitivity of each control parameter, wherein the accumulated sensitivity characterizes the influence degree of the control parameter on the quality score of a finished product, calculating the self-adaptive crossover probability of each control parameter according to the accumulated sensitivity, carrying out crossover operation on the variation vector and the target vector to generate a test vector, inputting the test vector into a pre-constructed finished product score prediction model to obtain a finished product score, updating the accumulated sensitivity of each control parameter according to the lifting amplitude of the finished product score relative to the target vector, and updating the population according to the finished product score; and iterating for a plurality of times until reaching a preset termination condition, and outputting optimal recommended parameters to control the rotary furnace.
2. The rotary furnace parameter optimization recommendation method based on data analysis according to claim 1, wherein the construction process of the finished product score prediction model comprises the following steps: Preprocessing the historical process parameters as input features, taking the quality scores as output tags, training a finished product score prediction model, and establishing a mapping relation between the input features and the output tags to obtain the finished product score prediction model.
3. The rotary furnace parameter optimization recommendation method based on data analysis according to claim 2, wherein the finished product score prediction model is a BP neural network or a support vector regression model.
4. The rotary furnace parameter optimization recommendation method based on data analysis of claim 1, wherein said generating an initial population comprising a plurality of parameter vectors comprises: Acquiring an environment correction coefficient of the current environment; The method comprises the steps of obtaining a reference parameter vector and a range vector of a rotary furnace, wherein the reference parameter vector comprises reference values of control parameters of each baking stage, the range vector comprises an allowable adjustment range of each control parameter of each baking stage, calculating the product of random gain and environment correction coefficient, multiplying the minimum value of the product sum of 1 by the range vector to obtain an offset vector, taking the sum of the offset vector and the reference parameter vector as the parameter vector, and obtaining parameter vectors corresponding to a plurality of random gains to obtain an initial population.
5. The rotary furnace parameter optimization recommendation method based on data analysis according to claim 4, wherein the calculating process of the environment correction coefficient is as follows: Collecting real-time temperature and input voltage of the environment where the rotary furnace is located; Calculating the temperature difference between the real-time room temperature and a preset reference room temperature, and determining a temperature correction amount according to the temperature difference, wherein the temperature correction amount is positively correlated with the temperature difference; calculating the pressure difference between the input voltage and a preset reference voltage, and determining a voltage correction amount according to the pressure difference, wherein the voltage correction amount is positively correlated with the pressure difference; And determining an environment correction coefficient according to the preset reference coefficient, the temperature correction amount and the voltage correction amount.
6. The rotary furnace parameter optimization recommendation method based on data analysis according to claim 1 is characterized in that obtaining effective variation scaling factors of all control parameters comprises calculating products of basic scaling factors and variation amplitudes of the control parameters in differential vectors for any control parameters to obtain theoretical variation amounts, calculating cutoff coefficients of the control parameters, wherein the cutoff coefficients are ratios of maximum allowable variation amounts to the theoretical variation amounts, taking the minimum value of 1 and the cutoff coefficients as correction coefficients, and taking products of the basic scaling factors and the correction coefficients as the effective variation scaling factors of the control parameters.
7. The rotary furnace parameter optimization recommendation method based on data analysis according to claim 1, wherein calculating an adaptive crossover probability of each control parameter according to the accumulated sensitivity comprises: Obtaining the maximum value and the minimum value of the accumulated sensitivity of each control parameter in the current population, and taking the difference value between the maximum value and the minimum value as the sensitivity extremely poor; taking the difference value between the cumulative sensitivity of any control parameter and the minimum value as the relative sensitivity; and carrying out linear interpolation on the preset minimum crossover probability and maximum crossover probability according to the ratio of the relative sensitivity to the sensitivity range, so as to obtain the self-adaptive crossover probability of the control parameter.
8. The rotary furnace parameter optimization recommendation method based on data analysis according to claim 1, wherein updating the cumulative sensitivity of each control parameter according to the elevation magnitude of the product score relative to the target vector comprises: Screening test vectors with the product scores higher than the target vectors to be used as winning sets; Calculating the parameter variation amplitude and the score lifting quantity of each test vector relative to the target vector according to any control parameter in response to the non-null winning set, calculating the weight coefficient of each test vector, and carrying out weighted summation on the score lifting quantity according to the weight coefficient, wherein the weight coefficient is the ratio of the parameter variation amplitude to the corresponding control parameter value range; and carrying out weighted summation on the accumulated sensitivity of the previous generation and the sensitivity contribution value by using a forgetting factor to obtain updated accumulated sensitivity.
9. The rotary furnace parameter optimization recommendation method based on data analysis of claim 8 wherein updating the cumulative sensitivity of each control parameter further comprises, in response to the winning set being empty, taking the last generation cumulative sensitivity as the updated cumulative sensitivity.
10. The rotary furnace parameter optimization recommendation method based on data analysis according to claim 1, wherein said control parameters include temperature, humidity and wind speed of the rotary furnace.

Description

Rotary furnace parameter optimization recommendation method based on data analysis Technical Field The application relates to the technical field of parameter control, in particular to a rotary furnace parameter optimization recommendation method based on data analysis. Background The baking process of the food is a key link of the modern food manufacturing industry, and the rotary furnace can finish the baking process of the pastry product through the cooperative control of a plurality of control parameters such as temperature, humidity, wind speed and the like. Along with the continuous improvement of the food quality requirements of the consumer market, how to ensure the color and the taste of each batch of finished products and improve the production efficiency at the same time has become a key index for measuring the performance of a baking control system. The existing rotary furnace control technology mainly adopts a parameter control mode based on a fixed formula. In actual operation, an operator typically manually sets target temperature, humidity and wind speed values for each baking stage according to a preset standard baking recipe. After the control system receives the instruction, the equipment such as a heating component, a humidifier, a fan and the like is driven to operate according to the target temperature, humidity and wind speed values corresponding to the baking stage, so that the environment in the furnace gradually approaches the set value and is maintained until the baking is finished. However, the fixed formula control lacks sensing and self-adaptive compensation capability to environmental temperature and humidity, and cannot adapt to interference of external environment, in addition, sensitivity contribution of each control parameter to quality of a finished product is different, adjustment amplitude of each control parameter directly influences improvement effect of quality of the finished product, different influences of each control parameter on quality of the finished product are ignored, and under a complex industrial baking scene, optimal parameters of a rotary furnace cannot be determined, so that control precision of the rotary furnace is not high, and therefore, in order to ensure quality of the finished product, key parameters which have obvious influence on quality of the finished product need to be focused in a rotary furnace control process. Disclosure of Invention In order to solve the technical problem of low control precision of the rotary furnace, the application provides a rotary furnace parameter optimization recommendation method based on data analysis, which can focus on key parameters with obvious influence on the quality of finished products in an iteration process and ensure the accuracy of optimal process parameter recommendation in a complex industrial baking scene. The application provides a rotary furnace parameter optimization recommendation method based on data analysis, which comprises the steps of obtaining the maximum allowable variation of each control parameter of a rotary furnace, generating an initial population comprising a plurality of parameter vectors, executing an iterative process of a differential evolution algorithm, selecting a parent vector from the population for any target vector to calculate a differential vector, limiting the scaling degree of the differential vector according to the maximum allowable variation to obtain an effective variation scaling factor of each control parameter, generating a variation vector, obtaining the cumulative sensitivity of each control parameter, representing the influence degree of the control parameter on a quality score of a finished product, calculating the self-adaptive crossover probability of each control parameter according to the cumulative sensitivity, performing crossover operation on the variation vector and the target vector, generating a test vector, inputting the test vector into a pre-constructed finished product score prediction model, obtaining a finished product score, updating the cumulative sensitivity of each control parameter according to the lifting amplitude of the finished product score relative to the target vector, updating the population according to the finished product score, iterating for a plurality of times until a preset termination condition is reached, and outputting the optimal recommendation parameter to control the rotary furnace. The maximum allowable variation of each control parameter of the rotary furnace is obtained to limit the scaling degree of the differential vector, the adaptive crossover probability is calculated according to the accumulated sensitivity, limited searching calculation force is concentrated on key parameters while invalid instructions exceeding the equipment capacity are prevented from being generated, the precision and the convergence speed of parameter optimization are improved, and the value of each control parameter in each baking