Modelling and evaluation of combustion emission characteristics of COME biodiesel using RSM and ANN—a lead for pollution reduction
| Autor: | Santhosh Murugan, Ramachandran Thulasiram, Dharmalingam Ramasamy, Surendarnath Sundaramoorthy |
|---|---|
| Rok vydání: | 2021 |
| Předmět: |
Biodiesel
Variable compression ratio business.industry Health Toxicology and Mutagenesis Design matrix General Medicine 010501 environmental sciences Combustion 01 natural sciences Pollution Diesel fuel Lead Biofuel Biofuels Compression ratio Environmental Chemistry Environmental science Neural Networks Computer Response surface methodology Process engineering business Gasoline Vehicle Emissions 0105 earth and related environmental sciences |
| Zdroj: | Environmental Science and Pollution Research. 28:34730-34741 |
| ISSN: | 1614-7499 0944-1344 |
| Popis: | Nowadays, the emissions from the diesel engines are focused lot to minimise the environmental pollutions in accordance with the emission standards. In this regard, biodiesels are found to be efficient for the diesel engines due to their higher energy contents and low exhaust emissions. Use of biofuel in association with diesel will be an efficient way for the cost-effective performance of the diesel engines with reduced pollutions. The COME is an efficient combustible oil, in which the diesel is blended at different proportions to identify their suitability in the diesel engines. In this regard, the properties of the COME-Diesel blends are determined and analysed for their influence on the combustion characteristics. To understand the performance and emission characteristics of blends, experiments are carried out on the variable compression ratio (VCR) engine keeping the blend, compression ratio, load, and speed as variables. The response surface methodology (RSM) used as a tool for designing and conducting the experiments according to the predetermined variables. The experimental sets generated are performed to determine the NO and HC emissions (response functions). The adequacy of the models is verified through ANOVA and through the p and F tests. The experimental design matrix is also used in training the artificial neural network (ANN) to develop the empirical models. The models from RSM and ANN are experimented and the results obtained from both the models are compared for their accuracy levels. Once the hypothesis is developed for the biodiesel and engine setup, the emission models will be used for the optimising the engine operating parameters and blends to minimise the pollutions from engine for wide operating conditions. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|
Full text from SpringerLink