Artificial neural network for HCCI engine
Conventional spark and compression ignition engines suffer from low performance and high emissions, respectively. Homogeneous Charge Compression Ignition (HCCI) combustion is a promising combustion mode to solve inherent problems of internal combustion engines. HCCI engine has the potential to impro...
Saved in:
| Main Authors: | , , , , , , |
|---|---|
| Format: | Conference or Workshop Item |
| Language: | en |
| Published: |
2023
|
| Online Access: | http://eprints.utem.edu.my/id/eprint/27909/1/Artificial%20neural%20network%20for%20HCCI%20engine.pdf http://eprints.utem.edu.my/id/eprint/27909/ https://pubs.aip.org/aip/acp/article-abstract/2727/1/030026/2895121/Artificial-neural-network-for-HCCI-engine?redirectedFrom=fulltext |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Summary: | Conventional spark and compression ignition engines suffer from low performance and high emissions, respectively. Homogeneous Charge Compression Ignition (HCCI) combustion is a promising combustion mode to solve inherent problems of internal combustion engines. HCCI engine has the potential to improve spark-ignition engine fuel economy while at the same time solving the trade-off of NOx-soot emissions found in compression ignition engines. With the assistance of machine learning such as artificial neural networks (ANN), the potential of the HCCI engine can be maximized. In general, the HCCI engine model is often grouped into three major categories. The first category, the empirical model, requires a substantial amount of data from the experiment, while the second category, the thermo-kinetic model, needs computational resources that are often not accessible for instantaneous engine control. The third category, the artificial neural network model, provides a number of benefits over the first and second categories, offering a compromise between computational resources, accuracy and experimental data. |
|---|