Thermal power generating unit state prediction method based on multi-source data and deep learning

• Main Authors: ZHENG WEIDONG, LIN JINGRU, LI HANHUA, LI JIAQIN, HE WENLONG, GE HENG, PAN BO, NIU YONGZHE, CHEN JINDAN
• Format: Patent
• Language: Chinese; English
• Published: 09.04.2024
• Subjects: CONTROL OR REGULATING SYSTEMS IN GENERAL; CONTROLLING; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS ORELEMENTS; PHYSICS; REGULATING

The invention provides a thermal power generating unit state prediction method based on multi-source data and deep learning, and the method comprises the steps: controlling a thermal power generating unit state monitoring system to be powered on for operation, and controlling the thermal power generating unit state monitoring system to collect multi-type data of a thermal power generating unit, the multi-type data comprises image data, vibration data, sound data and operation parameters; preprocessing the multi-type data, and performing hierarchical feature extraction on the preprocessed multi-type data based on a pre-trained convolutional neural network to obtain a feature quantity; constructing a CNN-GRU model for the multi-type data, and determining an optimal model parameter of the CNN-GRU model suitable for the current prediction task based on an NAS algorithm; and inputting the characteristic quantity into the trained CNN-GRU model, and obtaining a thermal power generating unit state prediction result i