Research On Real-time Storage-consumptive Scheduling Of Renewable Energy Grid

Wind power and solar power have the advantages of being clean and renewable, but they also have the disadvantages of generation power uncertain. Uncertain renewable energy has bad effect on power grid, and solving the uncertainty o f generation power of great significance to power grid. Renewable energy grid is the grid where renewable energy is the main source, and the uncertainty of generation power lead to the fact that it can’t become the reliable main source. The method using high-capacity energy storage facilities to solve the uncertainty of power generation is called reserving method, and the pumped storage plant applying high-capacity energy storage facilities is named storage-consumptive pumped storage plant. Based on day-ahead schedule, renewable energy gird real-time storage-consumptive scheduling coordinates renewable energy, pumped storage and thermal power, eliminating the uncertainty of renewable energy and making the renewable energy grid safe, reliable and economic.The paper comes up with the concept of storage-consumptive pumped storage plant, building the model of the plant. Compared with the traditional pumped storage plant, the former works longer, transforms pumped or generation states more frequently and contacts with real-time scheduling more closely. It is unreasonable to describe it with traditional model. Based on energy topological graph, this mathematical model describes storage process of storage-consumptive pumped storage plant, providing the theory basis to construct real-time storage-consumptive scheduling model.Construct real-time storage-consumptive scheduling model in renewable energy grid without considering network constrains. Average power represents energy during real-time scheduling period, and unbalanced power indicates the relationship of renewable power and load power. Compared unbalanced power constrains and energy constrains of pumped storage plant, it concludes different operation modes of renewable energy gird. Take highest efficiency of energy conversion as optimal object, and then use lagrangian multiplier method to solve the model.Construct real-time storage-consumptive scheduling model considering power flow equations, node voltage constrains and branch power constrains. Consider 3 factors of maximum renewable energy connecting to gird, least adjustment amount of thermal power and highest energy conversion as optimal objectives at the same time and convert multiobjective optimization problem to one objective.Improve the prime-dual interior point method, and use improved interior point method to solve real-time storage-consumptive scheduling model. By adjusting the order of iteration variables, using relevance decoupling reduced method, it reduces calculation scale and improves computational efficiency.Using example in sufficient and insufficient renewable energy conditions to verify the statement, it reveals that the real-time storage-consumptive model is rational and that the prime-dual interior point method is effective.