基于负荷预测的大型集中供热管网管径改造规划方法
Planning method of pipe diameter renovation of a large-scale centralized heating network based on load prediction
摘要：
提出了一种基于全网热用户供热负荷预测模型及改进遗传算法的管径改造规划方法。全网热用户供热负荷预测模型通过选取典型住宅和公共建筑用户并将其供热负荷特性推广至全网,被证明可快速准确预测全网热用户供热负荷。改进遗传算法通过改造潜力指标筛选出可通过改造获利的管段作为优化变量,可显著减少遗传算法的优化变量数量。通过与传统遗传算法对比,发现其具有更高的计算效率和更好的寻优能力。将上述方法应用于西安市某大型集中供热管网的管径改造规划,发现可将优化变量从246个减少至13个,最终优化结果表明,仅通过改造3个管段就可获得173万元的年改造利润。
Abstract：
This paper presents a pipe diameter renovation planning method based on the heating load prediction model for all heat users in the whole network and the improved genetic algorithm. The heating load prediction model for all heat users in the whole network is proven to be able to quickly and accurately predict the heating load of all heat users in the whole network by selecting typical residential and public building users and extending their heating load characteristics to the whole network. The improved genetic algorithm screens out pipe sections that can profit from renovation as optimization variables through renovating potential index, which can significantly reduce the number of optimization variables of the genetic algorithm. By comparing with the traditional genetic algorithm, it is found that it has higher computational efficiency and better optimization ability. Applying the above method to the pipe diameter renovation planning of a large-scale centralized heating network in Xi’an, it is found that the optimization variables can be reduced from 246 to 13. Moreover, the final optimization result shows that an annual renovation profit of 1.73 million yuan can be obtained by only renovating three pipe sections.
Keywords:large-scale centralized heating network; pipe diameter renovation planning; renovation potential; genetic algorithm; load prediction