采用改进温湿度变量策略的夏季短期负荷预测方法

为了充分考虑温度和湿度变量对夏季电力负荷的综合影响,提出一种改进的基于温湿度多形式变量的夏季短期负荷预测方法。首先通过分析夏季气象因素对负荷变化的影响,构造了三种不同形式的温湿度变量作为模型输入变量。然后根据周特性变化对负荷进行分层,对各层负荷建立基于LASSO回归的预测模型,并通过枚举搜索求解算法对输入变量进行选择,优化预测模型。最后通过计算剩余变量对应的系数从而进一步估计出各时段负荷的分布。算例结果表明该方法能有效提高模型的预测精度及鲁棒性。

Multi-objective optimization strategy fordistribution network considering V2Genabledelectric vehicles in buildingintegrated energy system

Based on the large-scale penetration of electric vehicles (EV) into the building cluster, a multi-objective optimal strategy considering the coordinated dispatch of EV is proposed, for improving the safe and economical operation problems of distribution network. The system power loss and node voltage excursion can be effectively reduced, by taking measures of time-of-use (TOU) price mechanism bonded with the reactive compensation of energy storage devices. Firstly, the coordinate charging/discharging load model for EV has been established, to obtain a narrowed gap between load peak and valley. Next, a multi-objective optimization model of the distribution grid is also defined, and the active power loss and node voltage fluctuation are chosen to be the objective function. For improving the efficiency of optimization process, an advanced genetic algorithm associated with elite preservation policy is used. Finally, reactive compensation capacity supplied by capacitor banks is dynamically determined according to the varying building loads. The proposed strategy is demonstrated on the IEEE 33-node test case, and the simulation results show that the power supply pressure can be obviously relieved by introducing the coordinated charging/discharging behavior of EV; in the meantime, via reasonable planning of the compensation capacitor, the remarkably lower active power loss and voltage excursion can be realized, ensuring the safe and economical operation of the distribution system.