Abstract
A novel two-stage feature extraction scheme is proposed in this paper for eliciting discriminant information contained in the data from various nuclear power plant (NPP) sensors to facilitate event identification. Based on the idea of sensor typewise block projection, the primal features can be extracted without losing the intrinsic structure contained in the multisensor data. The features are then subject to further dimensionality reduction by selecting the sensors that are most relevant to the events under consideration. Results from detailed experiments with data generated from a simulator of Taiwan Maanshan NPP illustrate the efficacy of the proposed scheme.