Power System Can Interrupt Load Management And Unit Commitment Problem

The implementation of power market reform in our country is in the transition phase. The kernel of power market is to activate all participants of power industry through the stimulation of interests and meet the demand of the economic optimality. On the other hand, the power grid has been under the situation of power shortage for a long time, thus it is very important to solve this problem through the Demand Side Management (DSM), while interruptible load management (ILM) is an important part of DSM in power market. In addition, unit commitment has long been an important issue in the power industry, which can improve the economy while keeping the system stable. The economy will become more prominent in power market, so it is extremely necessary to carry out the study of unit commitment.The importance of interruptible load management and unit commitment is introduced from the point of power market in this dissertation, and the current study about interruptible load management is elaborated. The common algorithms of unit commitment and the latest research progress at home and abroad are also summarized. Considering the successful implementation of interruptible load management abroad, this dissertation firstly gives the cost-benefit analysis for participants who actualize the interruptible load measure. Discrete customer types are utilized to embody the compensation cost for the users. Then this dissertation presents the optimal model of interruptible load contract. Taking the benefit of both Power Company and interruptible users into account, this model interprets the minimum of the compensation cost as the objective function. Case studies on a real power system demonstrate effectiveness of the proposed model.This dissertation introduces the traditional application of Lagrangian relaxation to unit commitment, and analyzes the performance of this algorithm. Then this work proposes a formulation and solution algorithm that retains the separation structure of classic Lagrangian relaxation while allowing the use of nonlinear AC network constraints. In order to solve the OPF subproblems faster and exploit the separation structure, the OPF problems are figured out using the Interior Point Method (IPM) that is rapid and reliable, and this dissertation makes use of parallel processing methods to speed up the computation. All of the measures above make the method presented in this dissertation valuable in large-scale unit commitment.