This paper presents the development and evaluation of a multi-objective linear programming (LP) optimization tool for an operations planning platform (OPP). The LP optimization model is coded using a high level programming language called AMPL (A Mathematical Programming Language) and solved using a solver called CPLEX. This model was developed by closely working with operations planning engineers at BC Hydro, in Canada, with the support of a research team from the University of British Columbia. The optimization model incorporates two objectives to optimize the operation of a multi reservoir system: maximize revenue from power generation, and to minimize penalties resulting from deviations of reservoir elevations and spill releases from a preferred operating regime. Several multi-objective optimization techniques are being investigated including traditional methods such as the Weighting Method and the Constraint Method, as well as Goal Programming (GP) techniques. These methods would be tested for optimality and computational efficiency and would be generalized and used to study a number of multi-reservoir systems in British Columbia. Stochastic inflows would be considered in the model making use of chance-constraints and probability distribution functions based on historical inflow records. The model also incorporates features which simulate maintenance outages on hydropower plants. The goal is to determine the optimal maintenance schedules that minimize the cost of these outages. We present results of a case study to illustrate the capabilities of the model to provide decision makers with timely information on trade-off between different objectives. Maintenance schedules of generating units with and without optimization are also considered.
Archila, Daniel; Shawwash, Ziad K.; Jiyi, Zhou; Vassilev, Paul; Kong, Gillian; Plesa, Vladimir; and Abdalla, Alaa, "A Multi-Objective Optimization Model For Operations Planning Of Multi-Reservoir Systems" (2014). CUNY Academic Works.