小区集中供热系统节费运行优化方法研究

Research on Optimization Methods for Cost-saving Operation of Self-heating and Heating Systems

随着“煤改电”工作的推进,电采暖供热系统得到广泛应用,但由于部分小区的供热系统缺乏合理的运行调控方式,电采暖供热系统运行费用高的问题日益凸显,为降低供热系统的运行费用,提出了一种基于室温预测的供热系统运行节费优化方法,以每日24h的逐时热源设备出力程度的组合作为一种运行模式,利用GA-BP神经网络预测每种运行模式的室温,从而判断其是否满足供暖标准,经验证,所建立GA-BP室温预测模型最大误差为0.7℃,平均误差为0.07℃,说明本模型可用于实际供热系统。最后利用遗传算法从满足供暖达标的运行模式中求解出最低运行费用的运行模式。并以山西省某小区为例,按照两种不同优化模式分别计算运行费用,并与原运行方式运行费用进行对比,结果表明,在不调整现有室温的情况下,节费效果为8.21%,在保证室温不低于18℃的情况下,可实现节费16.74%。

With the advancement of the"coal-to-electricity"initiative,electric heating systems have been widely used.However,due to the lack of reasonable operational control methods in the heating systems of some neighborhoods,the issue of high operational costs for electric heating systems has become increasingly prominent.To reduce the operational costs of heating systems,this paper proposes an optimization method for heating system operation cost reduction based on indoor temperature prediction.It treats the hourly output combinations of heat source equipment over a 24-hour period as an operational mode.The GA-BP neural network is used to predict the indoor temperature of each operational mode,thereby determining whether it meets the heating standard.Verification has shown that the maximum error of the GA-BP indoor temperature prediction model established in this paper is 0.7°C,with an average error of 0.07°C,indicating that this model can be used in actual heating systems.Finally,genetic algorithms are used to find the operational mode with the lowest operational cost from among the operational modes that meet the heating standard.Taking a neighborhood in Shanxi Province as an example,the operational costs were calculated according to two different optimization modes and compared with the original operational method.The results showed that,without adjusting the existing indoor temperature,the cost reduction effect was 8.21%.When ensuring that the indoor temperature is not lower than 18°C,a cost reduction of 16.74%can be achieved.