Research On Coil Scheduling Problem In Continuous Annealing Line Considering Energy Consumption

Continuous annealing in iron and steel enterprises is the process of making steel performance better through changing the grain distribution inside the steel by annealing the steel coils after acid rolling.The production process in continuous annealing line is energy-consuming due to the high working temperature.In the traditional production process of continuous annealing,the production and energy management are independent.However,during the annealing process,the production schedule has great influence on the energy consumption.By coordinating the production and energy scheduling,it would reduce the total production and energy costs.This thesis studies the coil scheduling problem in continuous annealing line considering energy consumption from the perspective of production and energy coordination,which is to decide the coil schedule and heating temperature in order to reduce the total production and energy costs,under the premise of production process constraints and energy restrictions.For the deterministic case,a mixed-integer programming model is formulated,and a dynamic programming algorithm and a two-stage heuristic are proposed to solve it.For the case with uncertain parameters,the model is approximated by sampling,and a scenario decomposition based algorithm is proposed to solve it.The main contents of this thesis include:1)For the deterministic case,a mixed integer linear programming model is established.Based on the practical decision manner,a two-phase heuristic is proposed.Moreover,a dynamic programming algorithm is proposed based on the production and energy consumption coordination manner.Finally,the performance of production and energy consumption coordination is illustrated by the experiments.2)For the case with uncertain parameters,a stochastic optimization model with the objective of minimizing the expectation of energy and production costs is established.Based on sample average approximation,the model is approximated to a deterministic one through sampling.Then the model is solved by CPLEX.The experimental results illustrate that CPLEX cannot solve the model optimally in reasonable time.3)For the case with uncertain parameters,a sample average approximation algorithm is proposed to solve the problem.A heuristic algorithm is designed to determine the sample size.The approximated model is decomposed based on scenarios,and solved by Lagrangian relaxation based method.The performances of the algorithms are verified by computational experiments.4)Based on the above model and algorithms,the coil scheduling system for continuous annealing considering energy consumption is developed,which realizes the automatic production schedule and visualized modification,in order to provide alternative production schedule for the problem.