Dynamical nonlinear particulate matter estimation based on laser induced incandescence measurements

Measurements of transient emissions become more important due to the increasing contribution of transient operation to the total tail pipe emissions. While for many quantities measurement devices with response time in the range of few milliseconds exist, the same is not true for particulate matter(PM). Pulsed Laser Induced Incandescence (LII) is widely used in experimental setups and may offer a viable approach also for production engines, but the specific nature of LII raises doubts on the quantitative precision achievable by the method, especially in transient operation. Indeed, there are two main problems in particular for dynamic measurements. On one side, the emitted laser power must be high enough to excite a sufficiently large number of particles within the observed area, but not as high to destroy them, and varying engine operating conditions imply changes in the number and size distribution of the particles as well. On the other side, during transients, flow conditions will vary, the particle flow may become less homogenous and changes in the gas velocity can make the time synchronization more difficult. Against this background, a LII measurement system has been setup at the JKU in Linz. The LII measurement system has been setup with a high power laser system. Measurements of the particulate emissions have been recorded for a transient cycle using this setup, and in parallel the AVL Micro Soot Sensor and the AVL Opacimeter. As the paper shows, strong nonlinear relations prevent the use of simple correlation methods to parameterize the LII output to reconstruct the particulate flow, but by using a robustified polynomial approach a law could be determined which handles this task. The reconstructed results not only achieve the same dynamical performance and exactitude of the AVL Micro Soot Sensor, but also retain many dynamical effects which are qualitatively confirmed by opacity measurements.