Research On Several Issues Of Centralized State Estimation For Integrated Energy System

To improve the overall energy utilization and the renewable energy accommodation,the concept of Integrated Energy System(IES)has been proposed and regarded as the main form of energy for human society.Therefore,it is required to propose a proper SE method for IES to provide accurate and reliable data for the Energy Management System(EMS)to realize precise predictions,decision-making,and control for IES.In this paper,the static state estimation(SSE)and dynamic state estimation(DSE)of the integrated energy system are studied.The main work is as follows:(1)Aiming at the existing nonlinear and non-convex SSE model,a bilinear robust state estimation(BRSE)method for the Integrated Electricity-Heat System(IEHS)is proposed.The original nonlinear measurement equations are transformed into linear equations by introducing auxiliary state variables and measurements.Then a robust SE model based on Weighted Least Absolute Values(WLAV)for IEHS is established.In the case study,by comparing with the SE method based on Weighted Least Squares(WLS),it is verified that the method has higher computational efficiency and better robustness.(2)A robust state estimation model based on second-order cone programming(SOCP)is proposed for IEHS.Based on BRSE,to compensate for the loss of measurement redundancy,the quadratic equality constraints satisfied by the auxiliary state variables in the electricity system are relaxed to the second-order cone constraints and added to the existing linear WLAV model.The linear penalty function term is added to the objective function to tighten the SOC relaxation,and then a robust state estimation method based on SOCP is proposed.Compared with WLS and BRSE,it is proved that this method has better estimation accuracy than WLS and BRSE on the premise of ensuring the computational efficiency and obtaining the global optimal solution.Also,it has good robustness to the multiple interacting and conforming bad data that appears in the measurements.(3)Because the electricity and gas systems have different time scales and sampling periods,the DSE method based on the Kalman filter that considers the dynamic characteristics of gas pipelines is proposed.Firstly,by applying the finite difference method,the partial differential equations of the gas system are transformed into the standardized state transition equations.Then,the DSE model for IEGS is formed based on the Kalman filter.Also,the electricity system and the gas system’s measurements with different sampling intervals are joined to ensure the observability of DSE by using the interpolation method.In the case study,compared with the SSE method,it is proved that this method can reflect the system’s real operation state more accurately and has higher computational efficiency.