This study presents a robust vision-based monitoring approach for detecting flame combustion states in solid oxide fuel cell(SOFC)afterburners.Industrial SOFC systems face significant challenges in flame state detecti...This study presents a robust vision-based monitoring approach for detecting flame combustion states in solid oxide fuel cell(SOFC)afterburners.Industrial SOFC systems face significant challenges in flame state detection due to the inherent instability of combustion processes and fluctuating gas flow rates.To address these issues and enhance monitoring reliability,we introduce FlameNet-MSF,an innovative deep learning framework that integrates multi-scale feature fusion.The architecture adopts a dual-branch design:a coarse-grained branch to extract global flame characteristics and a fine-grained branch to capture local pattern details.These complementary features are adaptively fused through a dedicated fusion module,enabling accurate flame state monitoring under complex operating conditions.Extensive experiments conducted on industrial SOFC datasets demonstrate the superior performance of FlameNet-MSF,achieving an overall classification accuracy of 99.21%,with recognition accuracies exceeding 96.9%across all three flame states.Furthermore,the framework supports real-time processing with a latency of just 29.3 ms per frame.Cross-dataset validation and ablation studies further validate the robustness and generalization capabilities of the proposed method.By providing a reliable and practical solution for automated flame monitoring,the FlameNet-MSF framework contributes to improved combustion efficiency and operational safety in industrial SOFC applications.展开更多
This paper presents a study on a novel instrumentation system for the measurement of temperature distribution of combustion flames. This system operates upon the three-color principle combining advanced optical sensin...This paper presents a study on a novel instrumentation system for the measurement of temperature distribution of combustion flames. This system operates upon the three-color principle combining advanced optical sensing and digital image processing techniques. It comprises an endoscope, a light splitter assembly, a CCD camera, a frame-grabber and associated software. This system was calibrated using a blackbody furnace as standard temperature source. The relationship between flame temperatures and grey-level of the images was established through image processing and function correlation. Experimental results obtained on a gas-fired combustion rig provide flame images and temperature distributions on three different wavelengths. Based on the flame temperature distribution the combustion conditions can be analyzed. Experimental results also reveal that this system is capable of online measurement of temperature distribution in a combustion zone. This system can potentially be applied to many areas such as power generation, metallurgy, chemical engineering. It is also a powerful tool for improving the control of combustion process. Keywords three-color method - flame monitor - temperature - imaging processing CLC number TK311展开更多
基金supported by the National Key Research and Development Program of China(Grant No.2022YFB4003605)the Foundation for Outstanding Research Groups of Hubei Province of China(Grant No.2025AFA012)+3 种基金Jiangsu Provincial Key Research and Development Program(Grant No.BE2023092-3)Hubei Provincial Natural Science Foundation(Grant No.2024AFB226)Belt and Road Joint Laboratory on Measurement and Control Technology Fund(Grant No.MCT2024003)the Beijing Natural Science Foundation(Grant No.IS23050)。
文摘This study presents a robust vision-based monitoring approach for detecting flame combustion states in solid oxide fuel cell(SOFC)afterburners.Industrial SOFC systems face significant challenges in flame state detection due to the inherent instability of combustion processes and fluctuating gas flow rates.To address these issues and enhance monitoring reliability,we introduce FlameNet-MSF,an innovative deep learning framework that integrates multi-scale feature fusion.The architecture adopts a dual-branch design:a coarse-grained branch to extract global flame characteristics and a fine-grained branch to capture local pattern details.These complementary features are adaptively fused through a dedicated fusion module,enabling accurate flame state monitoring under complex operating conditions.Extensive experiments conducted on industrial SOFC datasets demonstrate the superior performance of FlameNet-MSF,achieving an overall classification accuracy of 99.21%,with recognition accuracies exceeding 96.9%across all three flame states.Furthermore,the framework supports real-time processing with a latency of just 29.3 ms per frame.Cross-dataset validation and ablation studies further validate the robustness and generalization capabilities of the proposed method.By providing a reliable and practical solution for automated flame monitoring,the FlameNet-MSF framework contributes to improved combustion efficiency and operational safety in industrial SOFC applications.
文摘This paper presents a study on a novel instrumentation system for the measurement of temperature distribution of combustion flames. This system operates upon the three-color principle combining advanced optical sensing and digital image processing techniques. It comprises an endoscope, a light splitter assembly, a CCD camera, a frame-grabber and associated software. This system was calibrated using a blackbody furnace as standard temperature source. The relationship between flame temperatures and grey-level of the images was established through image processing and function correlation. Experimental results obtained on a gas-fired combustion rig provide flame images and temperature distributions on three different wavelengths. Based on the flame temperature distribution the combustion conditions can be analyzed. Experimental results also reveal that this system is capable of online measurement of temperature distribution in a combustion zone. This system can potentially be applied to many areas such as power generation, metallurgy, chemical engineering. It is also a powerful tool for improving the control of combustion process. Keywords three-color method - flame monitor - temperature - imaging processing CLC number TK311
