An Unsupervised Classification Method for Flame Image of Pulverized Coal Combustion Based on Convolutional Auto-Encoder and Hidden Markov Model
Combustion condition monitoring is a fundamental and critical issue that needs to be addressed in the wide-load operation of coal-fired boilers. In this paper, an unsupervised classification framework based on the convolutional auto-encoder (CAE), the principal component analysis (PCA), and the hidd...
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2019-07-01
|
| Series: | Energies |
| Subjects: | |
| Online Access: | https://www.mdpi.com/1996-1073/12/13/2585 |
| _version_ | 1811306830835482624 |
|---|---|
| author | Tian Qiu Minjian Liu Guiping Zhou Li Wang Kai Gao |
| author_facet | Tian Qiu Minjian Liu Guiping Zhou Li Wang Kai Gao |
| author_sort | Tian Qiu |
| collection | DOAJ |
| description | Combustion condition monitoring is a fundamental and critical issue that needs to be addressed in the wide-load operation of coal-fired boilers. In this paper, an unsupervised classification framework based on the convolutional auto-encoder (CAE), the principal component analysis (PCA), and the hidden Markov model (HMM) is proposed to monitor the combustion condition with the uniformly spaced flame images, which are collected from the furnace combustion monitoring system. First, CAE is adopted to extract the features from the flame images, which obtain the sparse representations in the images. Then, PCA is applied to project the feature vectors into the orthogonal space for robustness and computation efficiency. Finally, a HMM is built to calculate the corresponding optimal states by learning the temporal behaviors in the compressed representations. A coal combustion adjustment experiment was conducted in a 660 MW opposed-firing boiler, and the sequential 14,400 flame images with three different combustion states were obtained to evaluate the effectiveness of the proposed approach. We tested six different compression dimensions of the latent variable <i>z</i> in the CAE model and ensured that the appropriate compress parameter was 1024. The proposed framework is compared with five other methods: the CAE + Gaussian mixture model (GMM), CAE + Kmean, the CAE + fuzzy c-mean method, CAE + HMM, and the traditional handcraft feature extraction method (TH) + HMM. The results show that the proposed framework has the highest classification accuracy (95.25% for the training samples and 97.36% for the testing samples) and has the best performance in recognizing the semi-stable state (85.67% for the training samples and 77.60% for the testing samples), indicating that the proposed framework is capable of identifying the combustion condition, changing when the combustion deteriorates as the coal feed rate falls. |
| first_indexed | 2024-04-13T08:53:35Z |
| format | Article |
| id | doaj.art-c2ab93d20eb041f7b9c282b48f138ca3 |
| institution | Directory Open Access Journal |
| issn | 1996-1073 |
| language | English |
| last_indexed | 2024-04-13T08:53:35Z |
| publishDate | 2019-07-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Energies |
| spelling | doaj.art-c2ab93d20eb041f7b9c282b48f138ca32022-12-22T02:53:25ZengMDPI AGEnergies1996-10732019-07-011213258510.3390/en12132585en12132585An Unsupervised Classification Method for Flame Image of Pulverized Coal Combustion Based on Convolutional Auto-Encoder and Hidden Markov ModelTian Qiu0Minjian Liu1Guiping Zhou2Li Wang3Kai Gao4Beijing Key Laboratory of New Technology and System on Measuring and Control for Industrial Process, North China Electric Power University, Beijing 102206, ChinaBeijing Key Laboratory of New Technology and System on Measuring and Control for Industrial Process, North China Electric Power University, Beijing 102206, ChinaState Grid Liaoning Electric Power Supply Co, Ltd., Shenyang 110004, Liaoning Province, ChinaState Grid Liaoning Electric Power Supply Co, Ltd., Shenyang 110004, Liaoning Province, ChinaState Grid Liaoning Electric Power Supply Co, Ltd., Shenyang 110004, Liaoning Province, ChinaCombustion condition monitoring is a fundamental and critical issue that needs to be addressed in the wide-load operation of coal-fired boilers. In this paper, an unsupervised classification framework based on the convolutional auto-encoder (CAE), the principal component analysis (PCA), and the hidden Markov model (HMM) is proposed to monitor the combustion condition with the uniformly spaced flame images, which are collected from the furnace combustion monitoring system. First, CAE is adopted to extract the features from the flame images, which obtain the sparse representations in the images. Then, PCA is applied to project the feature vectors into the orthogonal space for robustness and computation efficiency. Finally, a HMM is built to calculate the corresponding optimal states by learning the temporal behaviors in the compressed representations. A coal combustion adjustment experiment was conducted in a 660 MW opposed-firing boiler, and the sequential 14,400 flame images with three different combustion states were obtained to evaluate the effectiveness of the proposed approach. We tested six different compression dimensions of the latent variable <i>z</i> in the CAE model and ensured that the appropriate compress parameter was 1024. The proposed framework is compared with five other methods: the CAE + Gaussian mixture model (GMM), CAE + Kmean, the CAE + fuzzy c-mean method, CAE + HMM, and the traditional handcraft feature extraction method (TH) + HMM. The results show that the proposed framework has the highest classification accuracy (95.25% for the training samples and 97.36% for the testing samples) and has the best performance in recognizing the semi-stable state (85.67% for the training samples and 77.60% for the testing samples), indicating that the proposed framework is capable of identifying the combustion condition, changing when the combustion deteriorates as the coal feed rate falls.https://www.mdpi.com/1996-1073/12/13/2585flame imagesconvolutional auto-encoderhidden Markov modelunsupervised classification |
| spellingShingle | Tian Qiu Minjian Liu Guiping Zhou Li Wang Kai Gao An Unsupervised Classification Method for Flame Image of Pulverized Coal Combustion Based on Convolutional Auto-Encoder and Hidden Markov Model Energies flame images convolutional auto-encoder hidden Markov model unsupervised classification |
| title | An Unsupervised Classification Method for Flame Image of Pulverized Coal Combustion Based on Convolutional Auto-Encoder and Hidden Markov Model |
| title_full | An Unsupervised Classification Method for Flame Image of Pulverized Coal Combustion Based on Convolutional Auto-Encoder and Hidden Markov Model |
| title_fullStr | An Unsupervised Classification Method for Flame Image of Pulverized Coal Combustion Based on Convolutional Auto-Encoder and Hidden Markov Model |
| title_full_unstemmed | An Unsupervised Classification Method for Flame Image of Pulverized Coal Combustion Based on Convolutional Auto-Encoder and Hidden Markov Model |
| title_short | An Unsupervised Classification Method for Flame Image of Pulverized Coal Combustion Based on Convolutional Auto-Encoder and Hidden Markov Model |
| title_sort | unsupervised classification method for flame image of pulverized coal combustion based on convolutional auto encoder and hidden markov model |
| topic | flame images convolutional auto-encoder hidden Markov model unsupervised classification |
| url | https://www.mdpi.com/1996-1073/12/13/2585 |
| work_keys_str_mv | AT tianqiu anunsupervisedclassificationmethodforflameimageofpulverizedcoalcombustionbasedonconvolutionalautoencoderandhiddenmarkovmodel AT minjianliu anunsupervisedclassificationmethodforflameimageofpulverizedcoalcombustionbasedonconvolutionalautoencoderandhiddenmarkovmodel AT guipingzhou anunsupervisedclassificationmethodforflameimageofpulverizedcoalcombustionbasedonconvolutionalautoencoderandhiddenmarkovmodel AT liwang anunsupervisedclassificationmethodforflameimageofpulverizedcoalcombustionbasedonconvolutionalautoencoderandhiddenmarkovmodel AT kaigao anunsupervisedclassificationmethodforflameimageofpulverizedcoalcombustionbasedonconvolutionalautoencoderandhiddenmarkovmodel AT tianqiu unsupervisedclassificationmethodforflameimageofpulverizedcoalcombustionbasedonconvolutionalautoencoderandhiddenmarkovmodel AT minjianliu unsupervisedclassificationmethodforflameimageofpulverizedcoalcombustionbasedonconvolutionalautoencoderandhiddenmarkovmodel AT guipingzhou unsupervisedclassificationmethodforflameimageofpulverizedcoalcombustionbasedonconvolutionalautoencoderandhiddenmarkovmodel AT liwang unsupervisedclassificationmethodforflameimageofpulverizedcoalcombustionbasedonconvolutionalautoencoderandhiddenmarkovmodel AT kaigao unsupervisedclassificationmethodforflameimageofpulverizedcoalcombustionbasedonconvolutionalautoencoderandhiddenmarkovmodel |