Micro-grid Economic Dispatch Considering Output Uncertainty Of Wind Generators

With the development of power technology and the consumption of fossil energy, energy crisis and environmental problems have become increasingly prominent. The renewable energy has the advantages of low loss, less pollution, flexible operation and economy, therefore, the future energy development direction of the world gives priority to use renewable energy. Micro-grid is composed of distributed power supply, energy storage unit and load, etc. It has flexible operation mode and dispatch performance. The randomness of wind generation has the shock and negative impact on the power system of operation, power quality, and stability. Micro-grid economic dispatch including wind resource encounters a serious problem. Therefore, correctly describing the randomness of the wind generation, building a reasonable economic dispatch model can reduce the effect of randomness on the safe operation and economic loss, considering the uncertainties of wind output in micro-grid economic dispatch in this paper have important theoretical and practical significance.Firstly, the output randomness of wind generation was studied in the paper. When random variable obeyed determined probability distribution, the wind generation unit's safety margin in micro-grid was analyzed. Based on the theory of CVaR(Conditional Value-at-Risk) having good characterization for wind generation unit safety margin of the tail probability, the wind loss risk constraint was proposed. Under the condition of wind energy condition constraint at risk and mirco-grid safe operation, power of the conventional generators, exchange power with power grid, power of wind generators as the decision variables, conserdering the effect of pollutants on the micro-grid system efficiency, for the target of minimizing the sum of cost of conventional generating units, pollutant emission, and exchange cost between the micro-grid and power grid, the micro-grid economic dispatch model considering the risk loss of wind was established. With Rockefeller and Uryasey theory to solve the CVaR constraints, which arising multidimensional integration problem after treatment, sample average approximation(SAP) is employed to slove it. Finally simulation analysis was done, quantitative analysis impact on micro-grid economic dispatch by random optimization model, the risk level of confidence as well as Monte Carlo sampling frequency.Secondly, micro-grid system unable to accurately estimate the wind power output of the distribution function, a set of uncertainty moment of wind power output was build up, by using mean and covariance matrix of wind power output' historical data. Based on the set of uncertainty moment of wind power output, with power of conventional generating units, interruptible load and the input power of distribution network as decision variables, for the goal of the minimum expected cost of the controllable unit cost, micro-grid loss cost, a one-way interaction of the distribution network cost, and interruptible load(IL Load) compensation cost, a distributional robust economic dispatch model of micro-grid based on the uncertainty of wind power output moment(mean and matrix) was established. This model overcame the problem that the robust linear optimization was too much conservative. The distributional robust optimization problem is transformed into a semi-definite programming(SDP) problem by Lagrange duality theory, meanwhile, solve SDP problem by using interior point method. Take a micro-grid system for example, the feasibility and effectiveness of the model are demonstrated.