Multi-information Substation-area Protection And State Estimation Application For Smart Distribution Networks

With the integration of power system and information system combining intelligent technologies,the power network is developed to a safe,self-healing,green,strong and reliable smart grid.The distribution network is facing diversified,complex and intelligent situations.The development trend and technical route of modern power system are bringing huge change to distribution system.And,all this promotes the applications of intelligent techonologies and theories to distribution network protection.The dissertation focus on the study of substation-area protection and state estimation application for smart distribution networks.On the basis of smart distribution network and cyber energy system analysis,supported by smart substation,communication technology and intelligent analysis methods,substation-area coordinated protection system and state estimation information application are studied.The main contents contain:(1)Research of substation-area protection system based on multi-information.The distributed protection device is designed based on modularization thinking.Applying information coordination and data mining theories,several protections could realize information support and function coordination.And,the reliability and adaption of distribution network protection are improved.On the other hand,the influence of substation-area information system is analyzed in three aspects:reliability,real-time performance and security risk.The corresponding assessment methods are proposed and analysed.(2)Research of fast state estimation algorithm for distribution network.According to the optimized power decoupling analysis,measurements are formulated in a unified decoupling expression.Applying retaining non-linearity principle in power and voltage measurement transformation,the measurement equation is linear and can be solved in a fast iterative algorithm.By different measurements’ decoupling transformation,this state estimation algorithm has constant measurement Jacobian matrix and has advantages in calculation speed and convergence reliability.This proposed state estimation algorithm has good adaption to different r/x ratios and can also meet the requirement for the real-time performance.(3)Research of new state estimation method for complex three-phase distribution network.Measurements were transformed in a mixed mode to improve the precision of the traditional current measurement transformation.Asymmetrical three-phase state estimation was solved based on current decoupling compensation model and Tellegen’s Theorem.This method could fit the measurement situation of distribution network.It was also suitable to network considering cable capacitance.It could satisfy the real-time requirement of distribution network automation and protection.(4)Research of a novel protection criterion based on dynamic trend analysis.The iterative expression of linear regression model is derived,and compound iteration helps to solve problems of initial value dependence and calculation precision.Change-point discriminant is based on squared R test rule,and the iterative expression of residual error improves the calculation speed.The protection criterion can judge fault rapidly and effectively,and the protection criterion is immune to bad data and satisfies reliability requirement of industrial application.(5)Research of an integrated protection algorithm based on two-step search within the system.A modified active power calculation,in conjunction with a tabu search method,is used to establish the faulted region of the network through the nodes with voltage measurements.And the second search precisely detects the fault section downstream of the faulted region by comparing current phase angle changes.This proposed algorithm can be applied to distribution systems with DGs and has no need of installation of additional voltage transformer equipment...