Research On Coordinated Optimal Operation Of Power System Considering New Energy Uncertainty And Demand Side Active Response

As one of the important topics in power system research area,power system optimal operation is always the research hotspot for scholars at home and abroad.With large-scale new energy(such as wind power)accessed to the grid,uncertainty of generation side in power system has been significantly enhanced,making new energy output analysis and modeling research attract great attentions.After introducing new energy uncertainty,deterministic power flow calculation,which need to be taken into consideration in the coordinated optimal operation of power system,becomes probabilistic power flow calculation.Thus,research on probabilistic power flow calculation is the foundation of solving the coordinated optimal operation problem of power system.As we know,energy saving and emission reduction are the fundamental purposes of accessing new energy to the grid,which can help relieve the worsening environmental problem.The carbon emission is one of the important indices of evaluating the environment.Consequently,coordinated optimal operation of power system considering carbon emission is the current focus of research.In the power system operation,a branch may be congested due to the system failure,etc.Research on the congestion management problem in coordinated optimal operation of power system can't be avoided.Moreover,with further deregulation of demand side and enhanced initiative of loads,research on coordinated optimal operation of power system considering demand response(DR)meets the current needs.In summary,this paper does the five aspects of coordinated optimal operation of power system as follows:1)In the new energy output characteristic analysis and modeling,this paper uses the power system containing multiple wind farms as the research object and has achieved several research results.First,a non-sequential single wind farm model is established.The effect of wake,wind turbine failure and failure of power cables inside a wind farm on the single wind farm output is analyzed.Then,a novel MC-Nataf-CD method based on MC sampling method,Nataf transformation and Cholesky decomposition is proposed and a multiple wind farms' output model considering correlated wind speeds is built.Finally,the effect of the above three factors on the multiple wind farms' output is reflected through the reliability indices.This part provides the theoretical basis for follow-up study in this paper.2)In the area of probabilistic power flow calculation,a new probabilistic power flow model and calculation method considering correlated input variables is presented.On the basis of MC-Nataf-CD method and point estimate method,an innovative probabilistic power flow calculation method considering correlated input variables based on 2m+1 point estimate method and Cornish-Fisher series is proposed,which can effectively improve the calculation accuracy.When processing the probabilistic power flow problem containing the non-gaussian random variables,such as wind power,approximation of Cornish-Fisher series is more accurate than that of Gram-Charlier series,which provides a reliable solution for the follow-up study on the probabilistic power flow chance constraints in the coordinated optimal operation of power system.3)In the area of coordinated optimal operation of power system considering carbon emission,a novel concept of regional carbon emission and an innovative multi-objective regional coordinated optimal dispatch model considering carbon emission are proposed.Different from traditional model,the proposed model adds a new objective regional carbon emission difference to control the balance of regional carbon emission.The proposed probabilistic power flow calculation method has been applied to ensure the accuracy of simulation results.The proposed model can be both applied to the test and actual power systems,which can solve the new energy integrated power system coordinated optimal operation problem considering regional carbon emission balance properly.It is an effective tool for managing urban regional carbon emission and estimate the rationality of the wind farms' location plan.4)In the area of congestion management research in coordinated optimal operation of power system,both uncertainty of new energy and initiative of active distribution network nodes(ADNNs)have been studied.In the research of congestion management considering new energy uncertainty,a single-level power system congestion management model based on the probabilistic power flow is proposed,which is analyzed and discussed under different modes,including the pool mode,nodal price mode and direct electricity purchase contract mode.The proposed model can effectively prevent the occurrence of secondary branch congestion and over-reliance on power system rescheduling,which can also guide signing direct electricity purchase contract rationally.In the research of congestion management of power system containing ADNNs,a new congestion management model based on bi-level optimization is proposed,which can be solved by there heuristic algorithms based on the KKT optimization condition.The propsed congestion management model based on bi-level optimization can help ISO dispatch the traditional and distributed generators reasonably in case of generator or transmission line faults.It can also effectively control the market power and ensure the stability of nodal price,especially the nodes not far away from the ADNNs.5)In the area of coordinated optimal operation of power system considering DR,the issue of coordinated optimal operation of power system considering price volatility and DR is studied.A novel coordinated optimal dispatch model considering price volatility and DR is proposed.This model combines the traditional robust optimization method with the coordinated optimal dispatch model considering DR.The effects of the initiative of demand side,probabilistic power flow and uncertainty of settlement electricity price on the system optimal operation are taken into consideration to guarantee the system robustness against the uncertainty.The objective function of the model mainly consistes of the sale of electricity and the generation cost.The sale of electricity contains electricity sales of the first and second class load.The electricity sale of the first class is determined by LMP while that of the second is determined by both their own quotes and LMP.The model has two advantages: the first is controlling of the LMP and congestion;the second is shifting the peak load to the periods with low load.