Active Network Management Considering Wind and Load Forecasting Error

Active network management, such as economically shifting flexible electric vehicle (EV) load over time, can help relieve network congestion caused by excessive wind output with the least wind power curtailment. This paper proposes an enhanced EV load shifting strategy by considering the uncertainty associated with wind and load forecasting. Due to wind and load forecasting error, there will be risks in expected benefits from EV shifting over differing time horizon. The longer the time horizon, the higher will be the risk. Such tradeoffs between benefits and risks in EV shifting for mitigating network congestions are not considered in existing literature. Besides, the general method to analyze uncertainty is based on Monte Carlo simulation, which is time-consuming. This paper addresses the challenge by adopting risk adjusted return on capital concept, which is widely used in the financial sector for assessing returns under differing risk levels. The proposed strategy converts the operational benefits, generated from EV shifting under different uncertainty levels into an equivalent benefit value under per unit uncertainty level, i.e., 'mitigating' the impact of uncertainties in the optimization of EV load shifting. As demonstrated in this paper, the proposed strategy assesses the impacts of wind and load forecasting error on the expected operational benefit in an analytical and scalable way, thus extending the traditional deterministic network operation to stochastic network operation.