A model-based temperature estimator for improving sensor dynamics in vehicle exhaust systems

With the advent of fast processors, real-time process control is becoming an important component of modern automobile industrial products. Normally, heat transfer processes possess large time constants and, as a consequence, temperature sensors are relatively slow. In order to protect against the hostile environment, temperature sensors used in vehicle exhaust systems need additional casing. Due to the casing, the heat transfer from the surface of the casing to the sensor is decreased even further. In order to design a feedback control for the vehicle, accurate information about the temperature inside the exhaust system is required which needs to be estimated from the measured temperature inside the exhaust system. In this paper, we develop a model of the sensor and use a Kalman filter for the prediction of the ambient temperature from the measurements. Further, we validate our results with suitable tests and compare the test results with the simulation results.