Gas Path Fault Diagnosis of Aeroengine Based on Soft Square Pinball Loss ELM
The ELM constructed based on the least squares loss function and ±1 label has poor generalization in the classification of data containing noise. The introduction of square pinball loss function can improve the robustness of ELM. However, the algorithm based on the squared loss function a...
Main Authors: | , , , |
---|---|
Format: | Article |
Language: | English |
Published: |
IEEE
2020-01-01
|
Series: | IEEE Access |
Subjects: | |
Online Access: | https://ieeexplore.ieee.org/document/9143161/ |
id |
doaj-cf43260229fa4d248b2f69a036c76405 |
---|---|
record_format |
Article |
spelling |
doaj-cf43260229fa4d248b2f69a036c764052021-03-30T03:35:34ZengIEEEIEEE Access2169-35362020-01-01813103213104610.1109/ACCESS.2020.30100969143161Gas Path Fault Diagnosis of Aeroengine Based on Soft Square Pinball Loss ELMYuyuan Cao0https://orcid.org/0000-0001-8081-4810Bowen Zhang1https://orcid.org/0000-0001-8598-2792Huawei Wang2https://orcid.org/0000-0003-3258-850XYu Bai3https://orcid.org/0000-0003-4111-4875College of Civil Aviation, Nanjing University of Aeronautics and Astronautics, Nanjing, ChinaCollege of Civil Aviation, Nanjing University of Aeronautics and Astronautics, Nanjing, ChinaCollege of Civil Aviation, Nanjing University of Aeronautics and Astronautics, Nanjing, ChinaCollege of Civil Aviation, Nanjing University of Aeronautics and Astronautics, Nanjing, ChinaThe ELM constructed based on the least squares loss function and ±1 label has poor generalization in the classification of data containing noise. The introduction of square pinball loss function can improve the robustness of ELM. However, the algorithm based on the squared loss function and the ±1 label imposes a margin of 1 for all training samples. At the same time, due to the unbounded nature of the loss function, the generalization of the algorithm in the classification problem is reduced. This paper proposes a soft threshold square pinball loss (SSP-Loss) function. This function can set more flexible thresholds for training samples while maintaining the robustness of the square pinball loss function. The soft-threshold square pinball loss function can approximate the bounded loss function in stages to further improve the classification performance of the algorithm. The performance of ELM based on the soft pinball loss function on several benchmark data sets proves the effectiveness of our proposed algorithm. More importantly, the excellent robustness and classification performance of the algorithm is very suitable for aeroengine gas path fault diagnosis, and is expected to become its candidate technology.https://ieeexplore.ieee.org/document/9143161/Extreme learning machinefault diagnosisaircraft enginemachine learning |
collection |
DOAJ |
language |
English |
format |
Article |
sources |
DOAJ |
author |
Yuyuan Cao Bowen Zhang Huawei Wang Yu Bai |
spellingShingle |
Yuyuan Cao Bowen Zhang Huawei Wang Yu Bai Gas Path Fault Diagnosis of Aeroengine Based on Soft Square Pinball Loss ELM IEEE Access Extreme learning machine fault diagnosis aircraft engine machine learning |
author_facet |
Yuyuan Cao Bowen Zhang Huawei Wang Yu Bai |
author_sort |
Yuyuan Cao |
title |
Gas Path Fault Diagnosis of Aeroengine Based on Soft Square Pinball Loss ELM |
title_short |
Gas Path Fault Diagnosis of Aeroengine Based on Soft Square Pinball Loss ELM |
title_full |
Gas Path Fault Diagnosis of Aeroengine Based on Soft Square Pinball Loss ELM |
title_fullStr |
Gas Path Fault Diagnosis of Aeroengine Based on Soft Square Pinball Loss ELM |
title_full_unstemmed |
Gas Path Fault Diagnosis of Aeroengine Based on Soft Square Pinball Loss ELM |
title_sort |
gas path fault diagnosis of aeroengine based on soft square pinball loss elm |
publisher |
IEEE |
series |
IEEE Access |
issn |
2169-3536 |
publishDate |
2020-01-01 |
description |
The ELM constructed based on the least squares loss function and ±1 label has poor generalization in the classification of data containing noise. The introduction of square pinball loss function can improve the robustness of ELM. However, the algorithm based on the squared loss function and the ±1 label imposes a margin of 1 for all training samples. At the same time, due to the unbounded nature of the loss function, the generalization of the algorithm in the classification problem is reduced. This paper proposes a soft threshold square pinball loss (SSP-Loss) function. This function can set more flexible thresholds for training samples while maintaining the robustness of the square pinball loss function. The soft-threshold square pinball loss function can approximate the bounded loss function in stages to further improve the classification performance of the algorithm. The performance of ELM based on the soft pinball loss function on several benchmark data sets proves the effectiveness of our proposed algorithm. More importantly, the excellent robustness and classification performance of the algorithm is very suitable for aeroengine gas path fault diagnosis, and is expected to become its candidate technology. |
topic |
Extreme learning machine fault diagnosis aircraft engine machine learning |
url |
https://ieeexplore.ieee.org/document/9143161/ |
work_keys_str_mv |
AT yuyuancao gaspathfaultdiagnosisofaeroenginebasedonsoftsquarepinballlosselm AT bowenzhang gaspathfaultdiagnosisofaeroenginebasedonsoftsquarepinballlosselm AT huaweiwang gaspathfaultdiagnosisofaeroenginebasedonsoftsquarepinballlosselm AT yubai gaspathfaultdiagnosisofaeroenginebasedonsoftsquarepinballlosselm |
_version_ |
1724183231771705344 |