Adaptive Parameters

Under the deregulation of generation market in China, all distributed generators will particular in electric power bidding. Therefore power purchase cost optimization (PPCO) problem has been getting more attention of power grid Company. However, under the competition principle, they can purchase power from several of power plants, therefor, there exist continuous and integral variables in purchase cost model, which is difficult to solve by classical linear optimization method. An improved differential evolution algorithm is proposed and employed to solve the PPCO problem, which targets on minimum purchase cost, considering the supply and demand balance, generation and transfer capability as constraints. It yields the global optimum solution of the PPCO problem. The numerical results show that the proposed algorithm can solve the PPCO problem and saves the costs of power purchase. It has a widely practical value of application.

The paper concerns the application of Genetic Algorithms and Genetic Programming to complex tasks such as automated design of control systems, where the space of solutions is non-trivial and may contain discontinuities. An adaptive value-switching mechanism for mutation rate control is proposed. It is shown that the proposed mechanism is useful in preventing the search from getting trapped in local extremes of the fitness landscape.

The paper concerns the application of Genetic Algorithms and Genetic Programming to complex tasks such as automated design of control systems, where the space of solutions is non-trivial and may contain discontinuities. Several adaptive mechanisms for control of the search algorithm's parameters are proposed, investigated and compared to each other. It is shown that the proposed mechanisms are useful in preventing the search from getting trapped in local extremes of the fitness landscape.