Studies On Theory And Algorithm Of Robust Dispatch Considering Probability Distribution Information In Power Systems

With large-scale wind power integrated into power systems,the variability and uncertainty of wind power escalate the uncertainty in power system operation and pose the challenges of designing an effective dispatch method for power systems.To this end,a series of efficient dispatch methods considering the uncertainty have been proposed.Among them,the robust dispatch method has been widely used due to its advantages of high computational efficiency and less required information.However,in the robust dispatch methods,the uncertainty set is usually pre-determined and the probability information of uncertain variables is not well regarded in the optimization,which lead to over-conservative decisions.It should be noted that with large-scale wind power integrated into power systems,the traditional robust dispatch methods may not be applicable.Therefore,to improve the conservativeness,it is necessary to introduce the probability information into the robust dispatch methods.However,due to the insufficient historical data and modeling errors,the probability distritbuion of uncertainty estimated based on the historical data may be not the true one,which will affect the effectiveness of the dispatch methods that make decisions based on the accurate probability information.Therefore,seeking a solution to the above-mentioned issues and developing new robust dispatch methods considering the distributional uncertainty have significant theoretical and practical meanings.Therefore,this thesis aiming at copting with the uncertainty,making a tradeoff between the operational cost and risk,using optimization skill,does research on the robust dispatch approaches considering the probability distribution information.The main contributions of this thesis are summarized as follows:(1)Considering the issues that the decisions of robust dispatch approaches are usually over-conservative and the traditional robust dispatch approaches cannot handle the optimization dispatch problems of power systems with limited flexibility,a novel robust economic dispatch approach considering the precise probability distritbution information is proposed.By incorporating the wind power accommodation risk into the optimization model,the precise probability distribution information of wind power is introduced into the robust optimization model,which will improve the conservativeness of decisions.Then the wind power accommodation risk is regareded as the operational risk to develop the optimization model,where a tradeoff between the operational cost and risk is made.Meanwhile,to ensure the enough power system flexibility,both of reserve capacity and its response rate are considered in the reserve optimization.In order to improve the computational tractability,an efficient algorithm is also developed to transform the original optimization model into a mixed integer linear programming problem,which improves the computational efficiency(2)Considering the distributional uncertainty of uncertainty in the power system operation,a new robust economic dispatch approach considering the distributional uncertainty is proposed.Firstly,based on the characteristic of the distributional uncertainty of uncertainty in the power system operation,a new distributional uncertainty set construction method is developed,which can take full advantage of the historical data.Then the operational risk considering the distributional uncertainty is defined to model the wind power accommodation risk.Finally,by striking a balance between the operational cost and risk,the robust economic dispatch optimization model is developed.With respect to the developed model,an efficient algorithm is developed based on the characteristic of wind power distributional uncertainty set and the operational experience of power systems to improve the computational efficiency.(3)Considering the impact of the voltage regulation ability on the economic dispatch considering distributional uncertainty,a new distributionally robust real-time dispatch approach is proposed based on the distributionally robust economic dispatch approach.In the proposed approach,a decoupled linearized power flow model is adopted to develop linear power disturbance response model.Meanwhile,to take full advantage of the voltage regulation ability of power systems,the linear decision rule is adopted to control the reactive power output of wind farms.By combining the linear power disturbance response model and the linear decision rule,the distributionally robust real-time dispatch approach considering voltage security is proposed,where active and reactive power are co-optimized.With respect to the proposed optimization model,an efficient algorithm is developed to transform the original optimization model into a mixed integer linear programming problem,which improves the computational efficiency.(4)To address the coordination dispatch problem of the transmission and active distribution system,a new distributionally robust real-time dispatch approach for integrated transmission and distribution systems is proposed.In the proposed approach,a new coordination strategy for integrated transmission and distribution systems is developed,where the transmission and active distribtuions can be coordinated under the expected and disturbance scenarios.And then combining the ATC technique and the distritbutionally robust economic dispatch method,distributionally robust dispatch optimization model for integrated transmission and distribution systems is developed.To avoid the bilinear constraints that will prevent the convergence and degrade the computational efficiency,a new affinely strategy is adopted in the proposed optimization model to contribute to a linear programming problem.Meanwhile,in order to achieve the parallel solving of the ransmission and distribution optimization sub-problems,the diagonal quadratic approximation is applied so that the computational efficiency can be further improved.