Kohonen self-organizing map for multiple-outlet sluice gates control system

In multiple-outlet sluice gates control systems, the open-close state of each sluice gate is usually decided by personal experience based on upstream water height and downstream water height. In this paper, a new method is proposed to determine the state of multiple-outlet sluice gates. Using Kohonen self-organizing map neural network, the proposed method maps the expectation-running water heights into the sluice gates state. From the experimental results, the method can get effective multiple-outlet sluice gates features by mapping a continuous input vector into a discrete output vector.