@article {
author = {Torabi, Seyed Ali and Shayegan, Mohsen and Mohammadi, Mohammad},
title = {A hybrid Mathematical Programming-Fuzzy Expert Systems Approach for Improving Combined Cycle Power Plants’ Operations},
journal = {Advances in Industrial Engineering},
volume = {49},
number = {2},
pages = {165-176},
year = {2015},
publisher = {University of Tehran},
issn = {2783-1744},
eissn = {2783-1744},
doi = {10.22059/jieng.2015.57056},
abstract = {By developing the level of life in societies, the need for new resources of energy becomes more important. In almost all cases, the aim is to convert one form of energy to the electrical form. With advances of technology and use of new electric and electronic devices and gadgets, the need for electrical energy is increasing and providing stable and continuous electric power becomes very important and urgent. The most common place to convert other forms of energy to electric energy is power plant. To maintain the stability and continuity of delivering the electric energy to costumers, the operation and control of the plant is great important. Advances in power plants’ technology have resulted in more reliance on human (i.e. expert-based) decision making. In this way, a mathematical programming-based decision support system could provide a suitable support for human decision making in optimizing power plant operations. In this study, a new approach is presented for optimizing the load distribution among units in a combined cycle (Gas and Steam Turbine) power plant. In the normal operations status, two parameters, i.e. the efficiency and risk are the most important factors for load distribution. Power is demanded by the control center (national electric grid control) and this needed power is often equally distributed by the control center, power market and operators between units. However, this distribution scheme is not economic in terms of efficiency and risk. In the suggested method, online data from units after fuzzification are fed into a fuzzy expert system and according to the unit conditions, two defuzzified scales, i.e. the efficiency scale and risk scale, are calculated for each unit. These scales are then used as coefficients in the proposed bi-objective mathematical programming model by which the best possible load distribution scheme is obtained according to the demanded power and the efficiency and risk of current units which is then displayed to the operator to set in the load control system. Finally, the efficiency and effectiveness of the proposed method is shown by a real case study.},
keywords = {Augmented ε-constraint method,Decision support systems,Fuzzy expert systems,Mathematical Programming,Power plant operations,Rule-based systems},
url = {https://aie.ut.ac.ir/article_57056.html},
eprint = {https://aie.ut.ac.ir/article_57056_0db585869a71e77c204283b9da8d3728.pdf}
}