Optimization And Its Risk Assessment Of Power System Generation Scheduling For Large-scale Integration Of Renewable Energy Resources

Renewable energy such as wind power and solar energy is inexhaustible and clean, which can improve the economic, energy-saving and emission-reduction benefit of the power system. However, renewable energy is affected by many natural conditions, resulting in the intermittence, fluctuation and randomness of its power output, and the difficulty to accurately forecast. At present, the common application form of renewable energy in China is to connect grid in large scale and high concentration, but the randomness and volatility greatly increase the risk of power system operation. In the optimization of power system generation scheduling, conventional generations are often not sufficient to stabilize the fluctuation of large-scale renewable energy power generation output due to the technical performance, which leads to the waste of renewable energy power generation output and load shedding. Therefore, it is necessary to evaluate the economic risk reasonably with large-scale renewable energy in the model of power system generation scheduling, and further optimize the generation scheduling of power system considering risk factors.This paper firstly introduces the traditional optimization model of power system generation scheduling, and put forward an improved particle swarm optimization algorithm(IPSO algorithm) for the model. The algorithm introduces feasibility criterion of unit commitment to judge the feasibility of particles, and deal with the constraints by time-interval and optimize the unit state, which make the algorithm efficient.Secondly, this paper applies conditional value at risk measure theory to measure the risk of the waste of renewable energy power generation output and load shedding after the access of large-scale renewable energy. The risk measurement indexes and their simulation method are proposed for power system dispatching and economic operation with large-scale renewable energy. Through the example the sensitivity and application value of the risk measure indexes proposed.Finally, on the research above, the optimization model and algorithm of power system generation scheduling for large-scale integration of renewable energy sources are further researched. Based on CVaR, we establish the optimization model of power system generation scheduling with renewable energy. Variable neighborhood descent algorithm is embedded in the IPSO algorithm to improve the local search ability of particle swarm algorithm. The algorithm proposed is an improved particle swarm optimization algorithm based on variable neighborhood search(VND-IPSO algorithm), which is suitable for solving the power system generation scheduling with renewable energy. Effect of the proposed algorithm is tested through the example, and the influence of risk measure indexes to the optimization model of generation scheduling is analyzed.