Event-triggered reactive power tracking optimization for second-level power fluctuations of renewables and stochastic loads

• Published in: International journal of electrical power & energy systems Vol. 155; p. 109581
• Main Authors: Zhang, Zhaoyi, Wang, Zijiang, Shang, Ben, Fan, Youping, Shu, Yinbiao
• Format: Journal Article
• Language: English
• Published: Elsevier Ltd 01.01.2024
• Subjects: Chance constraint programming; Event-triggered; Reactive power tracking optimization; Second-level power fluctuations; Chance constraint programming; Second-level power fluctuations; Event-triggered; Reactive power tracking optimization
• ISSN: 0142-0615; 1879-3517
• DOI: 10.1016/j.ijepes.2023.109581

•Reactive power tracking optimization to adapt to second-level power fluctuations.•Short-temporal-scale optimization to avoid voltage violation and objective increase.•Event-triggered algorithm to skip unnecessary MGO and LTO when small fluctuations.•Comprehensive index considering effects on voltage and objective for event trigger.•Multi-stage optimization based on chance constraint programming for optimal states. The power fluctuations of renewable energy sources and stochastic loads (RESL) makes it difficult for system operators to achieve optimal control. This paper proposes an event-triggered reactive power optimization strategy based on the event-triggered algorithm to rapidly track the optimal operating states for accommodating the second-level power fluctuations of RESL, with less computation and communication. Firstly, on a longer temporal scale, event-triggered multi-stage global optimization (MGO) is performed on a rolling basis to obtain the time-interval optimal operating states according to chance constraint programming, and the system operation security is improved by narrowing the voltage-feasible regions. Next, at each moment within the time interval, event-triggered local tracking optimization (LTO) is performed at each controller of RESL by rapidly regulating reactive power to adapt to its second-level active power fluctuations. The LTO is completed at the local level by using the sensitivity information from MGO, without running the optimal power flow program, to relieve the computation and communication burdens. Then, event-triggered LTOs performed at all controllers synchronously form the system-level real-time optimization. The unnecessary MGO and LTO are skipped by using the event-triggered algorithm when small power fluctuations are occurring. Finally, a case study on the IEEE-39 bus power system validates the feasibility of the proposed strategy.