Development of a dynamic model for the reconstruction of tailpipe emissions from measurements on a constant volume sampling dilution system

Abstract

The need for better understanding and further improvement of the operation of modern low emission vehicles has drawn the attention to fast response time-resolved emission measurements. The constant volume sampling system can provide such data. The resolution of the recorded data though is limited due to design limitations of such systems. Several factors such as gas transport delays, mixing and dispersion phenomena, etc are responsible for smoothing and distortion of the original emission signal. The current paper describes a model developed to compensate for all these distortion sources, by taking into account the dynamic behaviour of the system. The problem is approached with purely mathematical tools derived from system control theory. The parameters of the model are estimated using measurements of the system response to known and controlled emission step inputs. The accuracy of the results is validated using ultra-fast response emission analysers. The final model is proven to be capable of increasing the quality of measurements in terms of time resolution and accuracy, at levels comparable to sophisticated ultra-fast response measurement techniques.