考虑异质性的定频空调负荷聚合建模及功率跟踪策略

Constant Frequency Air Conditioning Load Aggregated Modeling and Its Power Tracking Strategy Considering Heterogeneity

为便于建立空调负荷聚合模型,目前研究往往假设空调个体参数一致,忽略了空调受控群内部参数差异较大的实际情况,亦未计及空调最小停运时间与整数设定温度的影响,因此该文考虑空调负荷的异质性,提出一种改进型聚合建模方法。该方法根据空调额定制冷量分组,并通过模糊C-均值聚类算法提高受控群电功率的模拟精度。引入最小停运时间以及设定温度的整数约束,使仿真结果更接近实际定频空调运行工况。同时,在实际运算中采用坐标格式存储稀疏矩阵,提高运算速度和模型的实用性。基于该改进型聚合模型,进一步提出一种功率跟踪策略,该策略计算各聚类群每一时刻的需求响应能力,通过求解0—1整数规划问题实现调控对象最优组合。仿真算例验证所提策略对目标功率跟踪的良好效果。

To simplify the aggregated model for air conditioning loads,uniform parameters are usually assumed in recent research. How ever,the actual situation that the parameters of controlled air-conditioning group are significantly various,and the influence of the minimum outage time and integer setting temperature are neglected. This paper proposes an improved aggregated modeling method considering the heterogeneity of air conditioning loads,w hich improves the pow er simulation precision of the controlled group through grouping air conditioners by rated refrigeration capacity and using fuzzy C-means clustering algorithm. By introducing the minimum outage time and setting temperature adjustment w ith integer constraints,the simulation scenarios became closer to the actual operation condition of constant frequency air conditioning. Besides,in the actual calculation, the sparse matrix w as stored in the coordinate format to improve the calculation speed and practicability of the model. Based on the improved aggregated model,a pow er tracking strategy is proposed consequently,w hich calculates the demand response capacity of each cluster at each time step,and optimizes the regulating objects by solving the 0-1 integer programming problem. The simulation results verify the effectiveness of the proposed strategy in pow er tracking.