Report

Programme Gaspard Monge pour l'Optimisation et la Recherche Opérationnelle – 3 et 4 octobre 2013 Hybrid Approaches Combining Metaheuristics and Methods of Mathematical Analysis for Environmental Unit Commitment Problem Leaders : Saïd Hanafi and Nenad Mladenovic Mohammed BELLALIJ, LAMAV, University of Valenciennes Rachid BENMANSOUR, LAMIH, University of Valenciennes Igor CRÉVITS , LAMIH, University of Valenciennes Saïd HANAFI, LAMIH, University of Valenciennes Bassem JARBOUI, University of Sfax, Tunisia Raca TODOSIJEVIC, LAMIH, Valenciennes Rita MACEDO, LAMIH, University of Valenciennes Sandrine CHAROUSSET, EDF R&D Marko MLADENOVIC, University of Valenciennes Nenad MLADENOVIC, University of Brunel, United Kingdom Christophe WILBAUT, LAMIH, University of Valenciennes Hard Optimization Problems Hybrid Approaches They combine relaxation, heuristics, metaheuristics and exact NP-hard problems are often tackled in fields such as methods to provide efficient methods for hard problems . mathematics, statistics, computer science, physics, Hard Optimization Problem engineering, economics, and social sciences to solve realf* = Max{f(x) : x X E} world business problems appearing in Airline, Telecommunications, Manufacturing, Healthcare, Scheduling, Planning, Data mining , Transportation and Energy. Enviromental aspects Unit Commitment Problem Unit commitment Problem (UCP) is a well-known combinatorial optimization problem. It consists of determining an optimal production plan for a given set of power plants over a given time horizon so that the total production cost is minimized, while various constraints are satisfied. The constraints that must be respected are: 1) Unit power generation limits - upper and lower bounds of production for each unit 2) Load balance - the total production of all active plants must satisfy the required demand in each time period 3) Spinning reserve constraints 4) Minimum up/down time constraints – minimal number of consecutive time periods during which units must be turned on/off • Major polluters of the environment are power plants • Globally, more than 70% of power plants use fossil fuels such as coal, natural gas and oil • After combustion more CO2 is produced than fuel is used • Almost all countries enforce certain environmental penalties • In some countries CO2 taxes are more expensive than the fuel itself Environmental Unit Commitment Problem Some Goals • • Effective methods for bi-objective problem that makes balance between supply and demand. • Hybrid approaches combining exact methods, heuristics and relaxations for solving resource allocation and scheduling problems. • Variable neighborhood search for the problem of mobilization of production units, taking into account both economic and ecological aspects. Proposed Animations Optimization Seminars - Seminar Organization • • Debates between participants and speakers driven by the chairman Real time synthesis to precise developments, complements and needs - Structure of Half Day Seminars • • Presentation of the basics focused on the area and recent developments Discussion to define links between other research fields and synthesis to precise needs - Main Topics • Hybrid Approaches in Combinatorial Optimization Continuous Reflection on Education and Profession Methodological issues of optimization • Key points on mastering complexity for both methods and formulation • Optimization in concrete situations and link between the numerical and the real world • Improvement of the link by common vocabulary and concepts on optimization References • F. Glover, S. Hanafi (2010). Metaheuristic Search with Inequalities and Target Objectives for Mixed Binary Optimization. International Journal of Applied Metaheuristic Computing. • S. Hanafi, C. Wilbaut. (2011). Improved Convergent Heuristic for the 0-1 Multidimensional Knapsack Problem. Annals of Operations Research. • J. Lazić, S. Hanafi, N. Mladenović, D. Urošević (2010). Variable Neighbourhood Decomposition Search for 0-1 Mixed Integer Programs. Computers and Operations Research. • M. Vasquez, Y. Vimont (2005). Improved results on the 0-1 Multi Dimensional Knapsack problem. European Journal of Operational Research. • R. Todosijevic, M. Mladenovic, S. Hanafi, I. Crévits (2012). VNS based heuristic for solving the Unit Commitment problem. Electronic Notes in Discrete Mathematics. • Y.W. Jeong, J.B. Park, S.H. Jang, K. Lee (2010). A new quantum inspired binary pso: Application to unit commitment problems for power systems. IEEE Trans Power Systems. • P. Hansen, N. Mladenovic, J.A. Moreno-Perez (2010). Variable neighbourhood search: methods and applications. Annals of Operation Research.