Evaluation And Identification Methods Of Power System Transmission Congestion Considering Component Failures

Transmission congestion(TC) occurs when the power flow of a transmission line in a power system(PS) is larger than its thermal limit due to the limitations of thermal capacity of transmission lines and the stability of power system. Generally, the occurrence of TC can be directly recognized by power flow analysis using the PS topology, electrical parameters, operation parameters, and load level of the PS. Traditional TC methods are, in essence, deterministic, and can only be used to answer whether a TC phenomenon occurs or not, but are unable to quantify the severity of the TC.Numerous uncertainties exist in power system operations and electricity markets, such as in load forecasting, electricity price, generation plan and component(generating units, transmission lines, and transformers) failures, which result in the uncertainties of TC. Therefore, power system operators must have the information on the likelihood of TC occurrence and the severity of the TC in a given period, such as TC probability, TC frequency, and the adequacy of transmission lines. In addition, effective measures can be taken to alleviate or eliminate the TC in a power system if the key components causing the TC can be identified. Therefore, studies on TC considering uncertainties and identification of key TC components can be used to provide sufficient basis for the decisions in power system operations, repair and maintenance, which is very important in the theory and practical applications of PSs.This thesis mainly focuses on a TC evaluation method considering the uncertainties, TC tracing technique of power systems, and the identification method of weak parts. Then the proposed method was extend to evaluate the TC of power systems containing wind farms. The thesis mainly includes the following discussion points:(1)To overcome the shortage of most methods that only analyze TC from a single-period power flow point of view, a TC probability evaluation model considering multi-period load curves was built based on the Monte Carlo simulation and load classification using a cluster analysis method.The probability characteristics of TC phenomenons are analyzed based on the load sequential curve. Case studies on the IEEE Reliability Test System(RTS) indicate that a single TC probability index can be evaluated by using the proposed model, and the proposed method is effective and feasible.(2)An index system was presented to describe the probability TC severity of transmission components and the system level. Indices for TC severity of transmission components include TC probability, TC frequency, TC capacity and TC energy, and indices for system TC severity include system TC probability, TC frequency, TC capacity, and TC energy. The index system is a complete index system from two levels, i.e. transmission component and system, and three aspects, i.e. probability, frequency and capacity. Hence, the proposed index system can be used to evaluate the overall TC severity of system and components.(3)A TC evaluation model considering random failures of PS components, such as generating units, transmission lines, and transformers, was proposed based on the non-sequential Monte Carlo simualtion. System states of a PS can be formed by the randomly sampled component states using the Monte Carlo method. The occurrence of TC can be determined by the power flow analysis for failure system state sthrough the measures of re-dispatching generating units and shedding load. If TC of transmission components occurs, then TC indices are updated. Random sampling process is repeated until all samplings are finished.Finally, TC probability, frequency, capacity, and energy of the PS and transmission components can be calculated. Case studies on the IEEE-RTS indicate that the proposed method is effective and feasible. Based on the results, the following conclusions can be drawn that the electrical and reliability parameters of components and load-shedding policy have significant effects on the TC severity.(4)Based on a similar idea of reliability tracing, TC tracing principles i.e., Failed Components Sharing Principle(FCSP) and Proportional Sharing Principle(PSP), and TC tracing models and algorithm based on Monte Carlo simulation were presented, which can be used to recognize the source of TC. For a sampling state, if TC of the system occurs, then the indices of TC probability, frequency, and energy can be distributed to each failed component based on the PSP. TC tracing indices of each component can be accumulated by the shared indices of each sampling system state, and then the weak parts from a TC point of view can be identified. Case studies on the IEEE-RTS indicate that the TC indices can be completely distributed among all components using the proposed congestion tracing principles, models and algorithms.(5) Output power of a wind turbine generator(WTG) has the features of randomicity and intermittence, which has significant effect on the dispatching, operations of a power system containing wind energies. To characterize the effects on TC of a PS containing wind energies, a TC contribution index system of wind power was proposed, which can be used to quantify the “contribution” of a wind farm to the change of TC in a PS. The TC evaluation model for a PS considering the correlation between the series of wind speeds and the series of hourly loads was proposed. In addition, capacity credit(CC) of a wind farm was defined from a TC point of view. The CC model considering component failures was built based on Monte Carlo simulation. The algorithm for the CC model is proposed using a bisection method. The impacts of locations and capacities of wind farm on TC were analyzed using the case studies on a modified IEEE-RTS.