On some technical and economic aspects of Volt-VAR optimization and distributed renewable energy

Abstract

Conventional power systems have significantly promoted economic and cultural prosperity in human society for centuries, while they are associated with environmental pollution, greenhouse gas emissions, rapid fossil fuel depletion, and significant power losses.With the prospect of a more sustainable and cleaner future, distributed renewable energy sources, mainly including photovoltaics and wind turbines, are growing enormously. In power distribution networks, energy-saving related schemes like Volt-VAR optimization (VVO) and conservation voltage regulation (CVR) are widely studied and implemented. However, with the increment of renewable penetration level, the performance of energy-saving related VVO and CVR schemes is adversely affected due to the uncertain nature of load demand and renewables. It is necessary to develop planning and operation strategies for the coordination of voltage control devices, renewables, and energy storages. Furthermore, the marketization of the power industry requires planning and operation strategies to be more economical and profitable. There are motivations to further integrate market theories in renewable integrated power distribution networks. II In this thesis, various work has been conducted on the aforementioned topics. First of all, a novel scenario formation method is proposed to capture the uncertain characteristics of photovoltaics, wind turbines, load demand, and electricity price for power distribution network planning. Instead of modeling the characteristics of uncertainty resources independently and combining them based on hypothesis, the proposed method employs the multivariate Gaussian mixture model to cluster the characteristics jointly, where the characteristics in each scenario approach a multivariate Gaussian distribution. Based on this method, a scenario-based optimal renewable distributed generation planning model is proposed considering CVR effects and uncertainties of existing renewables. The proposed model helps enhance the CVR effects on loss saving and load reduction at the planning stage, which is rarely investigated in existing studies. Next, the potential of improving the economic performance of VVO is investigated in both the planning stage and the operation stage. An optimal dispatchable and non-dispatchable energy source planning model is proposed. The objective of this work is to maximize economic benefit of energy-saving while maintaining the minimum operation cost. The locations and capacities of capacitor bank, dispatchable distributed generator, and distributed energy storage are optimally decided to achieve the goal. Besides, a market-based VVO model is proposed with a data-driven load model which reflects not only load-to-voltage dependence but also energy-to-voltage dependence. The market-based VVO model helps distribution companies to achieve the maximum cost saving in addition to energy saving, which is contributed from the electricity price differences between wholesale markets and retail markets. Finally, a cooperative game-based business model for rooftop PV investment and profit-sharing is proposed. This work aims at maximizing the total profit and sharing the profits fairly considering the contributions of multiple entities who participate in the planning and operation stages. In summary, this thesis provides novel insights on some technical and economic aspects of Volt-VAR optimization and distributed renewables, which helps the sustainable development of future power distribution networks.