Sharif Digital Repository

This article investigates optimization of long-term operation of hydrothermal power systems consisting of cascaded reservoirs. Due to stochasticity of reservoir inflows, demand for energy, and unit forced outages, the uncertainty of this system is so significant that reliability of demand satisfaction becomes an indispensable component of the modelling process. On the other hand, existence of stochastic parameters, especially in the case of cascaded reservoirs, makes the problem very difficult to solve by applying existing optimization techniques. A hybrid genetic algorithm with dynamic tuning of its control parameters is developed that incorporates real number encoding and an analytical... 

A special hybrid genetic algorithm (GA) is designed to solve the long-term coordination of hydro-thermal power systems with cascaded reservoirs and stochastic inflows. Since decision variables are continuous, in the proposed GA we employ real number rather than binary encoding. To create superior children we introduce dynamic tuning of the weights of operators. An exponential normalization is also developed such that better chromosomes have more chance to reproduce. To test the proposed method, 16 GAs are investigated which differ based on real or binary encoding, dynamic tuning or fixed weights for operators, inverse or exponential normalization and mixed or pure random initial populations.... 

In this paper, an algebraic tuning rule is presented for fractional PI controllers to control first order plus dead-time processes. By using the performance map (PM) method, this tuning rule is derived in order to set the gain margin of the control system close to 3 and the phase margin close to 60 degrees. The robustness and performance of this tuning rule are compared with some well-known PI tuning rules. Simulation results are brought to demonstrate the effectiveness and robustness of this tuning formula against process dynamic uncertainties in comparison with the other tuning methods