Research On Operation Optimization Of Production Processes In Process Industry Based On Improved Intelligent Algorithms

With increasing international competition, steel, petrochemical and other process industries put forward a higher demand for operation level of production processes. In process industry, optimization operation of production process lies between the production scheduling layer and the process control layer in the automation system and its task is to reasonably determine the parameter values in production process and set goals for the process control layer in order to achieve stable operation of the production process, improve equipment efficiency and product quality, and reduce energy consumption and production costs. Therefore the production process operation optimization in has become a hot research issue in international process control field and is of great significance for the process industry companies.Strip cold rolling process and ethylene pyrolysis process are the typical production processes of steel and petrochemical industries, respectively, which have multivariable, nonlinear, strong coupling and other features. This paper focuses on the modeling and algorithm development for the operation optimization of the two processes. A hybrid scatter search algorithm based on differential evolution and an adaptive genetic algorithm are proposed and then applied to the operation optimization problems of strip cold rolling process and ethylene pyrolysis process, respectively. Specific contents of this paper are as follows:(1) An improved hybrid scatter search algorithm is proposed, in which the differential evolution is incorporated to generate new solutions. In addition, a variety of mutation strategies and an adaptive mutation selection strategy are developed to improve the efficiency and robustness of the hybrid algorithm.(2) For the strip cold rolling production process, the fluctuations in the production of the actual production process are taken into account, and thus a robust operation optimization model of the process is established to minimize the total energy consumption. The hybrid scatter search algorithm is used to solve the problem, and the experimental results demonstrate the effectiveness of the model and algorithm.(3) An improved adaptive genetic algorithm using a variety of crossover is proposed. In this algorithm, a selection strategy which can adaptively select efficient crossover based on the analysis of the search results is designed. Computational results based on benchmark problems illustrate the effectiveness of the proposed algorithm.(4) In the production of ethylene pyrolysis process, the least squares support vector machine is used to establish the operation optimization model. The improved adaptive genetic algorithm is used to solve the problem and the computational results based on actual production data show that the algorithm can help petrochemical companies to improve yields of both ethylene and propylene.(5) Based on the operation optimization model and algorithms, an operation optimization system is developed for the ethylene pyrolysis process.