Fast learning in networks of locally-tuned processing units
Neural Computation
Advances in artificial immune systems
IEEE Computational Intelligence Magazine
Learning and optimization using the clonal selection principle
IEEE Transactions on Evolutionary Computation
Hi-index | 0.00 |
Establishing the condition prediction model of characteristic parameters is one of the key parts in the implementation of condition based maintenance (CBM) of the hydroelectric generating unit (HGU). The performance of radial basis function neural network (RBFNN) in prediction mainly depends on the determination of the number and locations of data centers at the hidden layer. A novel approach inspired from the immune optimization principles is proposed in this paper and used to determine and optimize the structure at the hidden layer. The immune optimized RBFNN has been applied to the vibration condition prediction of the hydroturbine guiding bearing. The prediction results are compared with those by some other intelligent algorithms and the actual values, which shows the effectiveness and the preciseness of the proposed immune optimized RBFNN.