Experimental Identification, Modeling, and Control Analysis of a Steam Power Plant

Increasing demands for electricity, and the need for more and safer power generation has motivated investigation of exploiting new control methods. Such control methods should increase the power generation capacity, and also reduce the maintenance costs. To design proper controllers, adequate information about the system is required, which leads to an accurate dynamics model. In this project, Shazand Power Plant (10 km away from Arak, a city in central Iran) was considered for data collection. A set of data were obtained and used to prepare computational models using various methods; i.e. Least Square Error (LSE), Neural Networks, and Adaptive Neural Fuzzy Inference System (ANFIS). Next, based on the various approaches, new controllers were designed to control the system. In order to simulate the performance of the proposed controllers, developed models were used and the results were compared to the present control strategies. The obtained results reveal an acceptable and reasonable performance of the power plant, with higher efficiencies, to be used in other power plants.