Eureka AIR delivers breakthrough ideas for toughest innovation challenges, trusted by R&D personnel around the world.

LSTM-based gas turbine online detection method and apparatus

A gas turbine and detection method technology, which is applied in gas turbine engine testing, jet engine testing, neural learning methods, etc., can solve problems such as large prediction errors and control effects, and achieve high accuracy, large data volume, and small analysis errors. Effect

Active Publication Date: 2019-11-12
杭州华电江东热电有限公司
View PDF8 Cites 13 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

Its disadvantage is that the prediction error is large, which has a great impact on subsequent control

Method used

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • LSTM-based gas turbine online detection method and apparatus
  • LSTM-based gas turbine online detection method and apparatus
  • LSTM-based gas turbine online detection method and apparatus

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0046] Such as Figure 1-3As shown, the proposed LSTM-based gas turbine anomaly online detection method and device include the following steps:

[0047] Step 1: Data collection. A gas turbine is mainly composed of three major components: a compressor, a combustor and a turbine. Select the main measuring point data from the monitoring software of the gas turbine, and analyze it as the input of the LSTM neural network. Measuring point data include: GT IGV position (angle), GT IGV position, air humidity, gas turbine compressor intake temperature, gas turbine intake filter differential pressure measurement point (1) and measurement point (2), gas turbine intake filter Internal pressure, gas turbine inlet static pressure, GT IGV position-1 measuring point (1) and measuring point (2), gas turbine compressor inlet temperature measuring point (1), measuring point (2) and measuring point (3), Gas turbine compressor outlet air temperature, gas turbine combustor shell pressure measuri...

Embodiment 2

[0077] An LSTM-based gas turbine online detection device with the above method installed, including a memory, a processor, an I / O device and an alarm device that are electrically connected and stored with a program for realizing the above method, and the I / O device is connected to a computer with monitoring software installed on the gas turbine and / or network to access and obtain real-time measurement point data.

[0078] The processor is connected to the hand-held user terminal through wireless transmission. Remote monitoring and early warning through handheld devices.

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

PUM

No PUM Login to View More

Abstract

The present invention belongs to the field of power plant security control systems, and particularly relates to an LSTM-based gas turbine online detection method and apparatus. The method is characterized in comprising the following steps: data collection; normalization processing; feature extraction; LSTM abnormality detection model training; and online abnormality detection: inputting to-be-detected data to a trained detection model, to obtain a model prediction value, calculating a difference between the prediction value and a sensor measured value to obtain an absolute value, and if the absolute value exceeds a given threshold, determining that abnormality occurs. The method is suitable for processing and predicting important events with a relative long interval and delay in a time series, and is suitable for analysis and fitting of the time series. A deep learning concept and technology is fully utilized to automatically select and extract hidden characteristics in data information detected by a device sensor, so as to implement online abnormality detection based on gas turbine real-time measuring point data. The LSTM-based gas turbine online detection method is large in collected data quantity, small in analysis error, and high in early warning result accuracy rate.

Description

technical field [0001] The invention belongs to the field of power plant safety control systems, and in particular relates to an LSTM-based gas turbine online detection method and device. Background technique [0002] Gas turbine is one of the most important power machines at present, and has a wide range of applications in aviation, electric power, ships and other fields. Gas turbines have complex structures and harsh working environments, which are prone to various failures. If they cannot be found and repaired in time, their safety and reliability will be seriously affected. As the application of gas turbines increases, people pay more and more attention to their working conditions. Once a gas turbine fails and shuts down, it will affect the stability of the power system, cause huge economic losses, and even affect the stable development of the national economy. For gas turbines The study of faults has also been put on the agenda. [0003] In the prior art, methods for ...

Claims

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

Application Information

Patent Timeline
no application Login to View More
IPC IPC(8): G01M15/14G06N3/08
CPCG01M15/14G06N3/08
Inventor 王新保方继辉李勇辉
Owner 杭州华电江东热电有限公司
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Eureka Blog
Learn More
PatSnap group products