Oil well fault diagnosis method based on improved capsule neural network

• Main Authors: WANG YING, LIU MIAOMIAO, ZHANG QIANG, ZHU LIUTAO, WANG MEI, WANG HUI
• Format: Patent
• Language: Chinese; English
• Published: 30.08.2022
• Subjects: CALCULATING; COMPUTING; COUNTING; ELECTRIC DIGITAL DATA PROCESSING; PHYSICS

The invention relates to an oil well fault diagnosis method based on an improved capsule neural network. The method comprises the following steps: 1, constructing a data set required for training and testing an oil well fault diagnosis model of the improved capsule neural network; 2, establishing an oil well fault diagnosis model based on the improved capsule neural network; 3, dividing the processed data set into a training set and a test set, inputting the training set into the constructed oil well fault diagnosis model based on the improved capsule neural network, carrying out learning training on the oil well fault diagnosis model based on the improved capsule neural network, and finding a model parameter combination with the best effect; and 4, performing fault diagnosis on the to-be-identified fault indicator diagram data by using the trained oil well fault diagnosis model based on the improved capsule neural network. According to the method, existing indicator diagram data of oilfield development is ut