Application of Fixed Length Gene Genetic Programming (FLGGP) in Hydropower Reservoir Operation

This paper develops a new method for real-time operation of reservoir systems. Genetic programming (GP) and a developed fixed length gene GP (FLGGP) are applied and compared in two approaches of static and dynamic operation rules with the aim of hydroelectric supply of Karun3 reservoir in Iran. Results are compared with those of genetic algorithm (GA) and nonlinear programming (NLP) method, indicating that GP and FLGGP have a higher efficiency (on average, 5 %) than GA and NLP operation methods. In addition, results showed that the FLGGP method is a powerful and efficient tool without the limitations of GP and can be used as a suitable replacement to GP. Comparison of two approaches of static and dynamic operation rules demonstrated the superiority of dynamic operation rules and this approach has an average superiority of 10 % to static operation rules in all methods.