Reactive Power Optimization based on Data-driven Load Curve Segmentation

Since the operation of distribution system is highly impacted by fluctuating characteristics of loads, reactive power optimization over a certain time period is essential to provide effective strategies to maintain the security and economic operation of distribution system. In this paper, two methods are proposed to minimize network losses and reactive power compensation device adjustment times for a long time horizons, while satisfying the operational constrains such as satisfying voltage magnitude limits. One method conducts optimization for each time point independently. The other first segments measured load curve into several sections based on a filtered signal calculation method, and then optimizes reactive power dispatch for each load section. Through a case study for a modified IEEE 34 bus system, the optimization method with load curve segmentation is found to be able to achieve both low losses and adjustment times. Furthermore, its computational efficiency is also verified through experiments compared with other methods.