Dispatch of generations and flexible loads in power systems considering renewable energy

Abstract

The development of renewable energy sources (RESs) is one of the most effective and feasible solutions to the sustainability of human society. However, it is challenging for power systems to admit a larger percentage of RESs, especially wind and solar power which are less controllable. The paradigm of traditional power systems has to be changed in order to accommodate more RESs. More flexibility, possibly from the load side, may be necessary to overcome the uncertainties from RESs. New resources that can provide flexibility, as well as new strategies to dispatch these resources, are desired in both transmission level and distribution level of power systems.

In this thesis, methodologies for the dispatch of generations and distributed resources in power systems are studied considering the impact of RESs. First of all, a new algorithm is proposed for the real-time economic dispatch of power systems. The algorithm is able to adjust the generation schedule of the system in a most economical way if the real-time conditions of the system deviate from the original plan. It can be used to handle the forecast errors and fast fluctuations from both loads and RESs. The advantages of the proposed algorithm comparing to existing methods are illustrated and discussed.

Electrical vehicles (EVs) is one of the most potential flexible resources for future power systems. Proper charging control strategies are crucial to eliminate the adverse influence and realize the benefits of EVs to power systems. Two kinds of charging control strategies for EVs are proposed aiming at better feasibility and effectiveness. The strategies require more reasonable input information, and have less computational complexity to handle large amount of EVs. They can also achieve outstanding scheduling results in comparison with existing algorithms and strategies.

Besides EVs, operation strategies for other types of potential storage options in the load side are investigated. A comprehensive operation scheme is built for air-conditioning systems with real thermal storage devices, which can utilize the advantages of thermal storage devices more effectively. The operation scheme aims at load shifting and the cooperation between the storage system and wind power generation. The features of wind power forecast errors are studied to assist the design of the scheme. The storage system dispatched under the proposed scheme can achieve valuable benefits to power systems.

Finally, a load management strategy is proposed for electric water heaters with storage tanks in residential distribution systems. A new model to schedule multiple water heaters in a distribution system is proposed with enhanced feasibility. The strategy utilize the flexibility of water heating load to improve the load profile of the system. The influence of high PV generations to short-term load profile and load balance is considered. The results confirm that water heaters also have considerable potential to increase the controllability in the load side.

In summary, this thesis has developed new methodologies for real-time economic dispatch of generations and scheduling of flexible resources in power systems in response to the uncertainties from RESs.