Improved Response Speed of NDIR SF₆ Gas Sensors With Regression Kalman Data Fusion Algorithm

The leakage of sulfur hexafluoride (SF6) gas in high-pressure equipment will cause great risk, so it is of great significance to the development of SF6 gas concentration detection system. Nondispersive infrared (NDIR) technology offers several advantages, including miniaturization, low-power consumption, simple structure, affordability, and nondestructive testing, making it well suited for establishing large-scale sensor networks. However, while efforts have primarily focused on enhancing the detection limit and accuracy of NDIR sensors, shortening their response time has remained a persistent challenge. This is because sensor stability and response time are inherently linked to the length of the gas cell, presenting a dilemma where optimizing one aspect often compromises the other. Short gas cells offer faster response times but sacrifice stability, whereas long gas cells prioritize stability but elongate response times. To address this issue, this article introduces a novel solution: a dual gas cell NDIR sensor coupled with the regression Kalman data fusion algorithm. The results demonstrate that the fused data exhibit a response time equivalent to that of a short gas cell, surpassing that of a long gas cell by 20%. Moreover, the stability achieved is on par with that of a long gas cell, with the variance limited to less than 0.2. This innovative approach represents a significant advancement in SF6 leak detection technology, reconciling the conventional tradeoff between response time and stability.