Adaptive and Improved Multi-population Based Nature-inspired Optimization Algorithms for Water Pump Station Scheduling

A common problem that the world faces is the waste of energy. In water pump stations, the situation is not different. Employees still use the traditional, manual, and empirical operation of the water pumps. This process gradually generates unwanted losses of energy and money. To avoid such profligacy, this paper presents two Adaptive and one Improved Multi-population based nature-inspired optimization algorithms for water pump station scheduling. The main goal here is to obtain the optimal operational scheduling of each group of pumps, wasting the minimum amount of energy. Therefore, since the objective function relies on the shaft power consumption of all the pumps running together, our aim becomes feasible. We implemented and tested the algorithms in the main water pump station of Shanghai, in China. Based on traditional multi-population based nature-inspired optimization algorithms, such as Genetic Algorithm (GA), Ant Colony Optimization (ACO), and Particle Swarm Optimization (PSO), this work adapts and improves the models to fit the complex constraints and characteristics of the system. It also compares and analyses the performance of each method used in this case study, considering the obtained results. The method which demonstrated outperformance was chosen as the best solution for the present problem.