基于PBIL算法的炼钢生产低碳调度问题

Low-carbon scheduling problem in steelmaking production based on PBIL algorithm

为了将碳排放引入生产调度优化,针对其等待时间受限的生产调度问题,建立最小化最大完工时间与碳排放的双目标优化模型,利用加权效用函数与标准化方法将其转换为单目标,并采用种群增量学习算法对问题进行求解。仿真实验表明,作业等待时间因受上限约束对碳排放影响较小,设备空转是影响碳排放的主要因素,提高设备利用率可有效减少碳排放;最大完工时间与碳排放呈反相关关系,即为尽快响应客户要求企业需付出较大的碳排放代价,而在碳权交易市场下企业为降低碳排放成本需适当放宽以往的经济指标。

To introduce carbon emission into production scheduling optimization,aiming at the production scheduling problem with limited waiting time,a dual-objective optimization model for minimizing makespan and carbon emissions was constructed,and it was rebuilt as a single-objective model by using weighting utility function and standardized method.In addition,an Improved population-based increased learning(IPBIL)was utilized to solve the problem.Extensive simulation was conducted and the results showed that the operation waiting time had little impact on carbon emissions and lots of carbon were generated with equipment idling,thus the carbon emission could be reduced effectively by improving the utilization rate of the equipment.Moreover,makespan was inversely related to the carbon emission that enterprises needed to pay a greater cost of carbon emissions to quick respond to customer requirements,but they should relax economic indicators to reduce the cost of carbon emissions in carbon rights trading.