碳足迹
瞬态(计算机编程)
补偿(心理学)
汽车工程
碳纤维
工作(物理)
环境科学
原位
材料科学
工艺工程
降级(电信)
电化学
计算机科学
灵敏度(控制系统)
足迹
电极
瞬态响应
电化学气体传感器
瞬态分析
作者
Quanquan Tang,Zuorong Huang,Bin Wu,Xishuang Liang,Mingliang Fu,Geyu Lu
出处
期刊:ACS Sensors
[American Chemical Society]
日期:2026-07-23
卷期号:11 (8): 7264-7274
标识
DOI:10.1021/acssensors.6c01547
摘要
High-temperature CO2 detection in automotive exhaust remains a formidable challenge due to the harsh chemical environment and significant cross-sensitivity of solid electrolyte sensors. This study reports a robust sensing platform integrating a NASICON-based solid electrolyte sensor with a random forest (RF) machine learning (ML) architecture for high-fidelity carbon emission monitoring. To achieve superior electrochemical performance, the sensor utilizes surface-optimized catalysts and stabilizers, with NASICON doping to enhance ionic conductivity and structural integrity under thermal cycling. The hardware is coupled with an automotive-grade data acquisition system, enabling in situ signal capture directly from the engine tailpipe. To circumvent the analytical limitations of solid electrolyte sensors-specifically thermal drift and multi-gas cross-interference-a software-defined compensation strategy is proposed. Rather than relying on auxiliary gas sensors, the RF model leverages intrinsic vehicle operating parameters as surrogate multidimensional features to dynamically decouple interfering factors from the CO2 response. The sensor reliability was rigorously validated across multiple combustion platforms under standardized transient cycles, including the Worldwide Harmonized Light Vehicles Test Cycle (WLTC), the World Harmonized Transient Cycle (WHTC), and the Real Driving Emissions (RDE) protocols. The ML-enhanced sensor achieved a coefficient of determination (R2) exceeding 0.8 even in highly non-linear transient states. Crucially, the cumulative mass emission error was suppressed to less than 0.4%, demonstrating metrological-grade accuracy in total carbon quantification. By synergizing solid-state electrochemistry with intelligent data analytics, this work provides a scalable, cost-effective solution for real-time carbon footprint tracking, directly supporting global initiatives for transportation decarbonization and environmental compliance.
科研通智能强力驱动
Strongly Powered by AbleSci AI