Study On The Theory And Method Of Optimizing Distributed Power System Containing Intermittent Renewable Energy

Renewable energy sources, such as wind and solar, are widely accepted to be the key to a sustainable energy supply infrastructure, because of their zero carbon emission and their inexhaustibility. Since the wind and solar more dependent on climate, the wind/PV power system is characterized by intermittent and randomness. Considering the stochastic characteristic of renewable energy, the optimal configuration of power system with intermittent energy integration is studied in this dissertation.The power availability of the renewable power system containing energy storage is studied in this dissertation. Based on the analysis of intermittent characteristic of wind power and solar poaer, the discrete-time Markov chain is used to model the residual energy of battery storage system in charging and discharging process. Then the power availability can be obtained by the stationary distribution of the Markov chain. At last, the Monte Carlo simulation is introduced to verify the results of this method.In order to utilize the renewable energy cost effectively, many researchers have studied the algorithm to calculate the capacity of applicable generator units that can constitute a reliable power system with low cost. Firstly, a simple optimization algorithm to determine the optimum component size for a Wind/PV/Battery power system employing an iterative scheme is introduced. Based on the observation that the state of charge (SOC) of battery should be periodically invariant, the needed number of generator units can be calculated directly. Then in order to get the optimum system configuration that simultaneously minimizes the life cycle cost of system and the loss of power supply probability, the multi-objective design model is presented, and an improved Pareto archive Multi-objective Particle Swarm Optimization algorithm is adopted to get its solutions.In this study, a methodology has been proposed to optimally allocate wind-based generator units in the distribution system for minimizing average energy loss. The methodology is based on a probabilistic model of wind power considering the wind speed Weibull probability distribution function and wind turbine force outage rate. Due to the intermittent nature of wind power, there are possible changes in the distribution system infrastructure and operation by deploying wind turbines in distribution system. So the optimal allocation of DG is critical to make them work better in Distribution system. And the stochastic model of energy loss can be analyzed by using the probabilistic power flow based on Cornish-Fisher expansion. High computational Monte Carlo simulation is avoided, so that the proposed algorithm can greatly improve the executing efficient.We propose a branch flow model for the analysis and optimization of radial distribution network. The model leads to a new approach to solving optimal power flow that consists of two relaxation steps, which are angle relaxation and convex relaxation. Optimal power flow is generally nonconvex and NP-hard. After the two relaxation steps, the nonlinear optimal power flow can be transformed to second order cone programming problem. This dissertation introduces the branch flow model, explains the two relaxation steps, and proves the conditions for exact relaxation.