Induced draft fan fault identification method based on CNN-SVDD

Abstract

The invention relates to an induced draft fan fault recognition method based on CNN-SVDD, and the method comprises the steps: 1, collecting sufficient training data, and carrying out the data preprocessing; 2, constructing a CNN-SVDD model by using the preprocessed training data, performing dimension reduction on the time series data by using a CNN algorithm, and performing data single classification by using an SVDD algorithm; 3, collecting enough real-time data, and recognizing induced draft fan faults corresponding to the real-time data based on the constructed CNN-SVDD model. The method has the beneficial effects that feature extraction is firstly carried out by using deep learning, and single classification analysis is carried out by using SVDD after the feature dimension is greatly reduced, so that the time and memory overhead of SVDD can be saved. In addition, due to the fact that the induced draft fan data has the time sequence characteristics, the local correlation on the timedimension of the time sequence data can be fully utilized, learning parameters are reduced through the one-dimensional convolutional network, and calculation expenditure is further reduced.

Description

Technical field [0001] The invention relates to the technical field of reliability maintenance engineering, and in particular includes a method for identifying a fault of an induced draft fan based on CNN-SVDD. Background technique [0002] The induced draft fan is an important auxiliary equipment of a thermal power generating unit. The operating characteristics of the induced draft fan will affect the boiler combustion efficiency of the unit, the operating stability of the unit and the output of the unit. Therefore, timely detection of various signs of failure of the induced draft fan during the operation process can effectively grasp the operational status of the induced draft fan and optimize its operation status, avoid the further expansion of the failure trend and the occurrence of unplanned shutdown events, and improve the overall performance and efficiency of the unit. , It is of great significance to reduce equipment maintenance costs and increase equipment available time...

AI Technical Summary

Problems solved by technology

In the past, fault diagnosis methods based on machine learning modeling were mostly aimed at a single power plant, using all historical fault data of specific equipment in the power plant for fault pattern recognition model training, but fault pattern recognition modeling based on machine learning requires a large number of fault sample data to Ensure high diagnostic accuracy

In fact, the sample data of the same fault case of similar equipment in a single power plant is limited, and large diagnostic errors will occur in the actual application process

Images