Neural Computing and Applications
In this paper, two optimal operation plans of a cascaded reservoir system with four hydropower plants are studied for reaching the highest power energy of all the power plants. In the first plan, called local optimization plan, optimal water discharge of each upstream hydropower plant is determined first and the optimal discharge is then used as input data for determining optimal water discharge of other downstream hydropower plants. In the second plan, called global optimization plan, optimal water discharge of all hydropower plants is simultaneously determined. For reaching the highest total power energy of the system, nine methods consisting of moth swarm algorithm, Harris hawks optimization, cuckoo search algorithm (CSA), snap drift cuckoo search algorithm (SD-CSA), stochastic fractal search algorithm, parasitism predation algorithm, marine predator algorithm, tunicate swarm algorithm, and the proposed improved cuckoo search algorithm (ICSA) are implemented. Among the methods, the proposed ICSA is first developed by improving the exploitation phase of CSA. The comparisons of power energy from each hydropower plant for the two plans indicate that local optimization plan is more useful for upstream plants but for downstream plants, whereas global optimization plan is more useful for downstream plants. For the purpose of reaching the highest total power energy for the whole system, the global optimization plan is more effective because it reaches higher power energy than local optimization plan. For investigating the improvement of the proposed method over CSA and SD-CSA, the three methods are run for seven benchmark functions and for the global optimization plan with different population sizes and different iteration numbers. As a result, it is recommended that global optimization plan should be applied for all hydropower plants in cascaded systems and downstream plants with higher power energy can share benefit to upstream ones. In addition, one of the most effective methods that should be recommended for implementing global optimization plan is the proposed ICSA.
