Page 340 - Decision Making Applications in Modern Power Systems
P. 340

302  Decision Making Applications in Modern Power Systems


            is even more necessary when the weather conditions are adverse to the diffu-
            sion of effluents. The authors present a dynamic dispatch procedure, which
            is able to hold the integral nature of the restrictions of issue. So, the environ-
            mental economic dispatch (EED) in thermal plants is a very important task
            to ensure the power demand, which is to make a distribution to all the mill
            engines, ensuring that the cost is minimal.
               In this chapter a model and a mathematical method for EED tools using
            evolutionary algorithms (EAs) [nondominated sorting genetic algorithm
            (GA) II (NSGA-II)] to reduce both the cost of energy production in thermal
            power plants (TPPs) and the environmental impact are applied. The identifi-
            cation of different ways of evaluating the emissions produced by power
            plants suggests mathematical models and computational tools to be used for
            the assessment of the economic (cost generation and fuel consumption) and
            environmental (emissions) variables, considering the pollution generated as
            well as the permissibility of each pollutant in the atmosphere to allow the
            construction of different simulation scenarios. It also formulates the optimi-
            zation of bi-objective EED problem, using a computational tool (NSGA-II-
            EA) analysis for the selection of the configuration of independent and
            dependent variables of the mathematical model, considering the demanded
            power and the environmental impacts.


            12.2 Materials and methods

            12.2.1 Heuristic optimization techniques
            The use of heuristic methods increases to quickly get tools to give solutions
            to actual problems. It is important to note that these methods do not guaran-
            tee the best optimization solution found, although the purpose is to find the
            solution next to the optimal solution in a reasonable time. Fig. 12.1 shows
            the classification of global optimization methods [4,5].




















            FIGURE 12.1 Global optimization methods.
   335   336   337   338   339   340   341   342   343   344   345