Multi-objective Operation Optimization Of Continuous Annealing Production Process With Safety Constraints

In the continuous annealing production process of iron and steel enterprises,the setting of operation variables have long been relied on the experience of on-site operators.However,the continuous annealing production process needs to consider several objectives such as product quality,energy consumption and production capacity,and involves many coupling variables and complex constraints,which makes it difficult for operators to make global optimization settings.In the meantime,in the practical production process of continuous annealing,there often occur safety problems such as strip buckling and strip deviation,seriously affecting the quality of products and the continuity and safety of production,thus causing unnecessary losses to the enterprises.Therefore,taking the continuous annealing production process of a large iron and steel enterprise as the background,this dissertation considers the factors that affect safety production when establishing the operation optimization model,and establishes the operation optimization model with the goals of product quality,energy consumption,unit capacity and process execution.Based on the characteristics of the problem and the model,a multi-objective evolutionary algorithm with multi-population,in which the classification and the coordination and competition are incorporated,is proposed.The detailed research contents are as follows:(1)For the continuous annealing production process,the operation optimization model is established.Through the research,the four optimization objectives of product quality,energy consumption,unit production capacity and process execution and the operation variables related to the objectives are determined.Through the analysis of the continuous annealing production process,the process constraints are determined.By analyzing the factors influencing the safety in actual production,the quality safety constraints and the production safety constraints are determined.Finally,the multi-objective operation optimization model of the continuous annealing production process considering the safety constraints is established.(2)For the optimization model,a multi-objective evolutionary algorithm(denoted as MOEA-CMCC)based on the classification and the coordination and competition of multi-population is developed.In the algorithm,the strategy of multi-population with different evolutionary strategies is adopted.Through the competition and coordination mechanism,the population can adaptively adjust the size of the population and share information among populations so as to improve the searching diversity and robustness of the algorithm and to avoid the algorithm from falling into the local optimum.The classification mechanism is introduce into the external archive and the partial search is conducted in the class to improve the dispersibility and convergence speed of the external archive.The experimental results based on the benchmark problems show the effectiveness of the proposed mechanism and algorithm.(3)For the practical production,the multi-objective operation optimization of continuous annealing process based on MOEA-CMCC is studied.In order to solve the problem efficiently,the problem characteristics are further considered in the algorithm.The method of determining unknown parameters in safety constraints is proposed,and a feasibility repair strategy based on heuristic method is designed for process constraints.Experimental results based on practical problems illustrate that the proposed algorithm is very effective and that its performance is superior to those of the other traditional multi-obj ective evolutionary algorithms.