Studies On Modern Heuristic Optimization Techuiques With Their Applications To The Power Unit Coordinate Contrcl

A fossil power plant is usually composed of a boiler-turbine system, coupled to an electric power generator as a prime mover. The boiler-turbine configuration is a multivariable, nonlinear, and time varying-system, which suffers from strong coupling between the pressure and power output loops, large settling time, nonlinearities, and uncertain disturbances. Coordinate control is the fittest strategy to control this system during the load following mode, considering an optimization process under economics and quality considerations to satisfy the momentary load demand within the minimum cost of operation. Consequently, the energy and water resources can be saved from being depleted. The successful solution for the multi-objective coordinated control problem can be achieved by developing the optimal mapping that is capable to operate the power unit within the permissible limits. The main work of this research is arranged as following.1. Theoretical analysis on multi-objective optimization problem, heuristic algorithms, and power plant control techniques are presented. The most popular modern heuristic algorithms are studied in detail, such as Genetic Algorithm (GA), Differential Evolution (DE), Particle Swarm Optimization (PSO), Simulated Annealing (SA), and Multi-objective Evolutionary Algorithms (MOEAs). These algorithms are then employed to solve several benchmark functions, to verify and validate their performances.2. A typical supervisory steady-state coordinate control system is introduced to provide the optimal process for the boiler-turbine system. Considering that, the coordinate controller is a high-level layer, while the pressure and power control loops are the low-level layers of the system. Then the multi-objective power unit coordinated control problem is optimized by generating the optimal input command signals to adjust the positions of valve actuators to ensure the optimal operating mode for each setting point.3. Comparative studies are taken to verify and validate the performances of the algorithms such as GA, DE, SA, and PSO in solving the multi-objective power unit coordinated control problem.4. A modified differential evolution (m-DE) algorithm is adaptively proposed to solve the power unit coordinated control problem. The proposed algorithm is uniformly initialized with a controlling zone factor (m) then, mutated by inserting a multiplier term to reduce the randomness and improve the performance of the algorithm. A numerical case study is employed to verify the performance of the proposed algorithm with respect to the basic DE, GA, and PSO algorithms. The experimental simulation results show that the proposed algorithm outperforms the other comparative algorithms, which demonstrate its efficiency in solving the problem.5. Multi-objective evolutionary algorithms are implemented to solve the problem, such as multi-objective genetic algorithm (MOGA), non-dominated sorting genetic algorithm (NSGA-?), and strength Pareto evolutionary algorithm (SPEA-?). These algorithms have the ability to produce a set of solutions according to Pareto concept in an independent pilot run rather than the conventional method. Furthermore, they can provide the decision maker an opportunity to choose a reasonable solution according to his experience and problem preferences.