Pulp batch cooking parameter optimization method

Abstract

The invention discloses a pulp batch cooking parameter optimization method, which comprises the following steps of: establishing a comprehensive multi-parameter multi-objective optimization model, namely taking six cooking parameters, namely the concentration and using amount of white liquor, the using amount of black liquor, degree of vulcanization, cooking time and cooking temperature as optimizing variables and considering the quality objective, yield objective, steam energy consumption objective and white liquor using amount objective of pulp; and optimizing the multi-parameter multi-objective optimization model by using an improved NSGA-II, namely introducing a novel fitness evaluation mechanism, integrating an adaptive crossover mutation operator, increasing an allochthonous population immigration mechanism and adopting a simple constraint processing method on the basis of the NSGA-II. The method solves the problem of batch cooking multi-parameter multi-objective optimization by comprehensively considering the influence of the cooking parameters, improves the convergence property and overall optimality of the algorithm in the aspect of solving the problem of high-dimensional optimization and can fulfill the aims of saving energy and reducing energy consumption.

Description

technical field [0001] The invention relates to multi-objective optimization technology, in particular to a parameter optimization method for pulp batch cooking. Background technique [0002] The description of the multi-objective optimization problem is as follows: Given a decision vector X=(x 1 , x 2 ,...,x n ), it satisfies the following constraints: [0003] g i (X)≥0 (i=1, 2, ..., k) [0004] h i (X) = 0 (i = 1, 2, ..., l) [0005] There are r optimization objectives, and these r objectives are in conflict with each other, the optimization objective can be expressed as [0006] f(X)=(f 1 (X), f 2 (X),...,f r (X)) [0007] seek make f(X * ) to reach the optimum at the same time under the condition of satisfying the constraints. [0008] Many optimization problems in scientific research or engineering applications are multi-objective optimization problems. Several objectives are often in conflict with each other, and there is no absolute or unique best solu...

AI Technical Summary

Problems solved by technology

However, there is no literature that studies all the main variable parameters together. Generally, some of the parameters are fixed, and then the other parameters are optimized. The optimal solution obtained in this way may not be the real optimal solution.

For example, Xiao Lan et al. published "Cleaner Production Strategy Based on Process Modeling and Optimization Technology" in the Journal of Zhejiang University (Natural Science Edition) in 1998. According to the model after regression, the method of nonlinear programming was used to analyze the cooking time and effective alkali. Concentration optimization, but did not consider other parameters of cooking; Yan Liexiang et al. published "Neural Network Dimensionality Reduction Analysis Method for Optimization of Pulping and Cooking Process Conditions" published in the Chinese Journal of Papermaking in 2000, using neural network dimensionality reduction analysis to generate pulp. The isoline of pulp rate and hardness can be used to determine the operating area where the pulp hardness is controlled within a certain range and the pulp yield is high, but this method is relatively rough; Jin Fujiang published in Huaqiao University Journal (Natural Science Edition) in 2001 "Multi-objective Optimal Control of Cooking Process in Pulping Production" published on the Internet established regression models for both continuous digesters and batch digesters, and optimized them with methods such as fuzzy logic decision-making, but did not consider effective alkali concentration, degree of sulfidation, etc. and made a lot of simplifications to the model

Images