Abstract: Aiming at the high energy consumption caused by disregarding the changes of heating furnace's temperature in hot-rolling planning model,according to the productive technology features of hot-rolling,the hot-rolling batch planning was regarded as a Vehicle Route Problem (VRP) with uncertain rolling planning number.On this basis,the changing regularity of heating furnace temperature and the jump constraint of temperature between batches were considered,and a new mathematical model of rolling planning to minimize temperature jump penalty value as the main objective function was proposed.A fish seeking particle swarm algorithm was designed to solve the model.According to the practical hot-rolling batch planning problem of a steel enterprise,the proposed model and algorithm were verified,and the result showed that the rolling batch planning which considered the temperature curve could not only guide the production with the predetermined trend of temperature,but also be helpful for reducing energy consumption and extending the life of heating furnaces.Therefore,the proposed model and algorithm were effective and feasible.

Key words: temperature curve, hot-rolling, batch plan, vehicle route problem, fish seeking particle swarm algorithm

摘要： 针对目前热轧计划模型中未考虑加热炉温度变化而造成的能耗高的问题,根据钢铁企业热轧生产工艺的特点,将热轧批次生产计划归结为轧制计划数不确定的车辆路径问题。重点考虑了加热炉温度的变化规律和批次间温度的跳跃约束,以极小化温度跳跃惩罚值为目标建立了轧制计划数学模型,并设计出一种鱼群寻觅粒子群算法对模型进行求解。根据国内某钢铁企业热轧生产实际问题对模型和算法进行了验证,实验结果表明：考虑加热温度曲线的热轧批次计划不但能够按照预定的温度变化趋势来指导生产,而且有利于降低加热炉能耗和延长其寿命,因此所提出的模型和算法切实可行。

关键词: 温度曲线, 热轧, 批次计划, 车辆路径问题, 鱼群寻觅粒子群算法