CNN-GRU-BINN-based intelligent BIT design method for heavy duty gas turbine control system

Abstract

The invention discloses a CNN-GRU-BINN-based intelligent BIT design method for a heavy duty gas turbine control system, and belongs to the field of intelligent BIT design of heavy duty gas turbines. Aiming at the problem of high false alarm rate of a conventional BIT of the heavy duty gas turbine control system, the problem that the conventional BIT cannot extract time sequence characteristics of detection signals of a hardware module is solved by adopting a structure of combining two neural networks of CNN and GRU, and the spatial characteristics of the detection signals of a controller module are extracted by utilizing a CNN convolutional neural network structure; and time characteristics of the detection signals of the controller module are extracted in combination with a GRU recurrent neural network structure, the spatial characteristics and the time characteristics are fused and then input into a biological excitation neural network to achieve the functions of correct state recognition and false alarm reduction of the controller module, the recognition precision of the intelligent BIT is improved, and the operation reliability of the heavy duty gas turbine control system is enhanced.

Description

technical field [0001] The technical field of the present invention is the intelligent BIT design field of heavy-duty gas turbine control system, which is a convolutional neural network (Convolutional Neural Networks, CNN), a gated recurrent unit neural network (Gated Recurrent Unit, GRU) and a biologically inspired neural network (Biologically Inspired Neural Network, BINN) integrated intelligent BIT design method. Background technique [0002] Under the background of my country's vigorous development of domestic heavy-duty gas turbines, how to ensure the long-term reliable operation of heavy-duty gas turbine control systems has become a research hotspot. However, with the improvement of system integration and complexity, the detection and maintenance of heavy-duty gas turbine control systems are also facing difficulties. For real-time troubleshooting and maintenance, especially for hardware module level maintenance, the difficulty of system self-inspection and verification ...

AI Technical Summary

Problems solved by technology

[0003] At present, the self-inspection methods of domestic heavy-duty gas turbine control systems in my country are still in the conventional BIT. However, due to the limitation of conventional BIT detection only through simple threshold judgment, it is difficult to realize the effective identification of sudden changes and intermittent faults caused by the internal environment of the cabinet. Therefore, The inability to accurately detect the state of the hardware module of the heavy-duty gas turbine not only increases the maintenance cost of the control system, but also makes the domestic heavy-duty gas turbine unable to meet the requirements of long-term reliable operation, resulting in great economic losses

Images