Reliability analysis methods for power systems with substantial penetration of renewable generating resources

Energy storage sizing;Expected unserved energy;Loss of load probability;Probabilistic production costing;Seasonal correlation;Reliability assessment;Variable generation

Alamri, Abdullah Saad N. Abdullah ; Meliopoulos, A. P. Electrical and Computer Engineering Saeedifard, Maryam Gebraeel, Nagi Grijalva, Santiago Molzahn, Daniel ; Meliopoulos, A. P.

University:Georgia Institute of Technology

Department:Electrical and Computer Engineering

关键词: Energy storage sizing;    Expected unserved energy;    Loss of load probability;    Probabilistic production costing;    Seasonal correlation;    Reliability assessment;    Variable generation;   

Others  :  https://smartech.gatech.edu/bitstream/1853/62773/1/ALAMRI-DISSERTATION-2020.pdf

美国|英语

来源: SMARTech Repository

【 摘 要 】

Today’s typical power systems have a mix of conventional generation (CG) and variable generation (VG). Examples of VG are wind turbine systems (WTSs) and solar cell generators (SCGs). VG is characterized by intermittency and its integration into power systems affects power systems operation, reliability and planning practices. The focus of this research is generation adequacy when adding VG. The objective of this thesis is to develop reliability assessment models of power systems with wind farms (WF) and/or solar farms (SF); each WF may have a number of WTSs and each SF may have a number of SCGs. These models involve finding WF/SF power output probability. Three different methods for computing the generated power probability distribution function (PDF) of a WF or a SF are proposed: (1) analytical method, (2) non-sequential Monte Carlo simulation, and (3) sequential Monte Carlo simulation. Historical wind speed/solar radiation data are utilized to perform the study. Further, force outage rates (FORs) of components are incorporated in the process of computing the WF/SF generated power PDF. All methods yield comparable results. The usefulness of the computed PDF is demonstrated by integrating them into a probabilistic production costing (PPC) model for assessing the reliability of a system comprising one or more WFs/SFs. A further development of these models includes a unit commitment economic dispatch model (UCED) to simulate real operation of power systems, specifically CG constraints in presence of VG at different penetration levels. Also, the UCED will be used to solve for optimal energy storage system (ESS) sizing and its expected charging and discharging power profile. This profile determines the expected ESS effect on reliability. The optimal ESS sizing problem is modeled as a mixed integer linear program (MILP) that takes into account: (a) VG units FORs, (b) reserve and demand requirements, and (c) CG operational constraints. (d) seasonal wind speed, solar radiation, and demand correlation.