Prediction of combustion state through a semi-supervised learning model and flame imaging

• Published in: Fuel (Guildford) 2021-04, Vol.289, p.119745, Article 119745
• Main Authors: Han, Zhezhe, Li, Jian, Zhang, Biao, Hossain, Md. Moinul, Xu, Chuanlong
• Format: Article
• Language: English
• Subjects: Accuracy; Artificial neural networks; Boiler furnaces; Classifiers; Combustion; Combustion state prediction; Comparative studies; Denoising autoencoder; Feature extraction; Gaussian process; Gaussian process classifier; Generative adversarial network; Learning; Machine learning; Model accuracy; Neural networks; Noise reduction; Novel loss function; Prediction models; Retraining; Semi-supervised learning; Support vector machines
• ISSN: 0016-2361; 1873-7153
• DOI: 10.1016/j.fuel.2020.119745

•A novel semi-supervised learning model is established for combustion state prediction.•An innovative loss function is proposed by using adversarial learning mechanism and structural similarity metric.•The model is evaluated through seen and unseen combustion states of heavy oil-fired boiler furnace.•Generalization and robustness of the model are also verified.•Suggestions are provided for selecting optimal feature learning networks and classifiers. Accurate prediction of combustion state is crucial for an in-depth understanding of furnace performance and optimize operation conditions. Traditional data-driven approaches such as artificial neural networks and support vector machine incorporate distinct features which require prior knowledge for feature extraction and suffers poor generalization for unseen combustion states. Therefore, it is necessary to develop an advanced and accurate prediction model to resolve these limitations. This study presents a novel semi-supervised learning model integrating denoising autoencoder (DAE), generative adversarial network (GAN) and Gaussian process classifier (GPC). The DAE network is established to extract representative features of flame images and the network trained through the adversarial learning mechanism of the GAN. Structural similarity (SSIM) metric is introduced as a novel loss function to improve the feature learning ability of the DAE network. The extracted features are then fed into the GPC to predict the seen and unseen combustion states. The effectiveness of the proposed semi-supervised learning model, i.e., DAE-GAN-GPC was evaluated through 4.2 MW heavy oil-fired boiler furnace flame images captured under different combustion states. The averaged prediction accuracy of 99.83% was achieved for the seen combustion states. The new states (unseen) were predicted accurately through the proposed model by fine-tuning of GPC without retraining the DAE-GAN and averaged prediction accuracy of 98.36% was achieved for the unseen states. A comparative study was also carried out with other deep neural networks and classifiers. Results suggested that the proposed model provides better prediction accuracy and robustness capability compared to other traditional prediction models.