1949-3053 (c) 2019 IEEE. Personal use is permitted, but republication/redistribution requires IEEE permission. See http://www.ieee.org/publications_standards/publications/rights/index.html for more information. This article has been accepted for publication in a future issue of this journal, but has not been fully edited. Content may change prior to final publication. Citation information: DOI 10.1109/TSG.2020.2969256, IEEE Transactions on Smart Grid 1 Abstract—Aging generating units can gradually lose their efficiency, reliability, and economic viability, and violate certified levels of carbon emission in power system operations. Retirement is traditionally considered a viable option for managing aging units within the generation expansion planning (GEP) studies; while, the rehabilitation of such units can also be economically beneficial to power system and even outperform other options considering certain planning and operation criteria. Rehabilitation procedure depends on many physical factors including generating unit types and locations, as well as financial, regulatory, and transmission network constraints. This paper presents an optimization model for GEP considering retirement and rehabilitation options for generating unit planning and operation. The proposed GEP problem uses the Conditional Value-at-Risk (CVaR) as a risk index for considering rehabilitation cost and load forecast uncertainties. In order to evaluate performance of the proposed method, the IEEE 24-bus and 118-bus systems are examined in various operating conditions and the results are analyzed. Index Terms—Generation Expansion Planning (GEP), Retirement, Rehabilitation, Uncertainty, Risk Management. NOMENCLATURE Indices and symbols b Index for load blocks c Index for candidate units C Superscript for candidate units df Index for deterministic framework e Index for existing units E Superscript for existing units EO Superscript for old existing units h Index for buses k Index for existing lines m, t Index for buses Sc Index for scenarios sf Index for stochastic framework y Index for years Sets dh I Set of loads connected to bus h gh I Set of generators connected to bus h kh I Set of lines connected to bus h M. Farhoumandi and F. Aminifar are with the School of Electrical and Computer Engineering, College of Engineering, University of Tehran, Tehran 14395-515 Iran (e-mails: matin.farhoumandi@ut.ac.ir; faminifar@ut.ac.ir). M. Shahidehpour is with the Electrical and Computer Engineering Department, Illinois Institute of Technology, Chicago, IL 60616 USA (e-mail: ms@iit.edu). He is also a Research Professor with the Center of Research Excellence in Renewable Energy and Power Systems, King Abdulaziz University, Jeddah, 21589, Saudi Arabia. Parameters CLT Construction lead time (y) d Discount rate DT Duration of load block (h) IC Candidate unit investment cost ($M) MC Unit maintenance cost ($M) NB Number of load blocks NCU Number of candidate units NEL Number of lines NEU Number of existing units NH Number of buses NSc Number of scenarios NY Number of years OC Unit operation cost ($M/MWh) PD Load demand (MW) RC Existing unit rehabilitation cost ($M) RLT Rehabilitation lead time (y) SV Existing unit salvage value ($M) TEB Total expansion budget ($M)  Higher costs probability indicator [0,1]  Risk coefficient degree [0,1]  Growth rate for maintenance cost  Growth rate for operation cost  Declining rate of power generation  Inflation rate for the costs  Line susceptance (p.u.)  Present value conversion coefficient Variables s Construction termination indicator (0/1) u Power generation decision variable (0/1) v Rehabilitation termination indicator (0/1) x Investment decision variable (0/1) z Rehabilitation decision variable (0/1)  Retirement decision variable (0/1)  Voltage angle of bus (Radian)  Specific TRC for risk management PG Output power of generating units (MW) PL Power flow of transmission line (MW) TC Total cost ($M) TIC Total investment cost ($M) TOMC Total O&M cost ($M) TRC Total rehabilitation cost ($M) TSV Total salvage value ($M) I. INTRODUCTION OWER system loads continue to increase as populations around the world, particularly in developing countries, Generation Expansion Planning Considering the Rehabilitation of Aging Generating Units Matin Farhoumandi, Student Member, IEEE, Farrokh Aminifar, Senior Member, IEEE, and Mohammad Shahidehpour, Fellow, IEEE P