A5049290253- Catalytic Processes in Materials Science
- Vehicle emissions and performance
- Catalysis and Oxidation Reactions
- Air Quality Monitoring and Forecasting
- Advanced Combustion Engine Technologies
- Advanced Control Systems Optimization
- Advanced Memory and Neural Computing
- Process Optimization and Integration
- Catalysis and Hydrodesulfurization Studies
- Analytical Chemistry and Sensors
- Neural Networks and Applications
- Industrial Gas Emission Control
- Field-Flow Fractionation Techniques
- Extraction and Separation Processes
- Electrocatalysts for Energy Conversion
- CCD and CMOS Imaging Sensors
- Water Quality Monitoring and Analysis
Delft University of Technology
1999-2002
University of Twente
1995-1998
For an adequate control of the reductant flow in selective catalytic reduction NOx diesel exhaust, a tool has to be available accurately and quickly predict engine's emission. these purposes, elaborate computer models expensive analyzers are not feasible. The application neural network is proposed instead. Measurements were performed on transient operating engine. One part data was used train for emission prediction, other test. average absolute deviation between predicted measured 6.7 %....
Abstract For an adequate control of the reductant flow in selective catalytic reduction NO x diesel exhaust, a tool has to be available predict accurately and fast engine's emission. In this article application neural network is proposed. Measurements were performed on transient engine. The average absolute deviation between measured emission predicted by 6.7%. high accuracy predictions, combined with short computation times (0.2 ms/data point), makes very promising automotive control.
To adequately control the reductant flow for catalytic reduction of NOx in diesel exhaust, a tool is required that capable accurately and quickly predicting emissions from engine's operating variables. In this paper three algorithms nonlinear modeling are evaluated: neural networks, split fit algorithm Bakker et al., polynomial NARX model, which linear its parameters. Measurements were carried out on transient automotive engine. Each was able to make excellent predictions, combined with...
Abstract This review article describes and discusses the literature on reducing agent control systems for NO x emission reduction in exhaust gas of full lean‐burn engines. The can be classified as feedback, feedforward, feedforward‐feedback, feedforward/feedforward‐feedback periodical dosing systems, which further classical, override, co‐ordinated, constrained split or a combination those. As long reproducible, fast cheap sensors are not commercially available, system should an inferential...