Detonation cell size prediction based on artificial neural networks with chemical kinetics and thermodynamic parameters
In this paper, we develop a series of Artificial Neural Networks (ANN) using different chemical kinetic and thermodynamic input parameters to predict detonation cell sizes. The feedforward neural networks are trained and validated using available experimental data from the Caltech detonation databas...
| Main Authors: | , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Elsevier
2023-03-01
|
| Series: | Fuel Communications |
| Subjects: | |
| Online Access: | http://www.sciencedirect.com/science/article/pii/S2666052022000346 |
| _version_ | 1797961113316360192 |
|---|---|
| author | Georgios Bakalis Maryam Valipour Jamal Bentahar Lyes Kadem Honghui Teng Hoi Dick Ng |
| author_facet | Georgios Bakalis Maryam Valipour Jamal Bentahar Lyes Kadem Honghui Teng Hoi Dick Ng |
| author_sort | Georgios Bakalis |
| collection | DOAJ |
| description | In this paper, we develop a series of Artificial Neural Networks (ANN) using different chemical kinetic and thermodynamic input parameters to predict detonation cell sizes. The feedforward neural networks are trained and validated using available experimental data from the Caltech detonation database covering a wide variety of gaseous combustible mixtures at different initial conditions. For each combination of input parameters, a multiple-stage process is followed, which is described in detail, to first determine the best hyperparameters of the ANN (hidden layers, nodes per layer, etc.) and secondly to establish through a fitting process the optimal parameters for each specific network. The performance of the artificial neural networks with different input features is assessed using data from the same source, but that is kept independent and separate from the training and validation process of the ANN. It is found that ANN with three features can provide an accurate estimation of detonation cell size, while increasing the number of features does not improve the accuracy of the ANN. It is also found that the input parameters with the best performance relate indirectly to the stability parameter χ. |
| first_indexed | 2024-04-11T00:54:20Z |
| format | Article |
| id | doaj.art-536152ce01754d97aaf5db6fe4938dec |
| institution | Directory Open Access Journal |
| issn | 2666-0520 |
| language | English |
| last_indexed | 2024-04-11T00:54:20Z |
| publishDate | 2023-03-01 |
| publisher | Elsevier |
| record_format | Article |
| series | Fuel Communications |
| spelling | doaj.art-536152ce01754d97aaf5db6fe4938dec2023-01-05T06:24:44ZengElsevierFuel Communications2666-05202023-03-0114100084Detonation cell size prediction based on artificial neural networks with chemical kinetics and thermodynamic parametersGeorgios Bakalis0Maryam Valipour1Jamal Bentahar2Lyes Kadem3Honghui Teng4Hoi Dick Ng5Department of Mechanical, Industrial and Aerospace Engineering, Concordia University, Montréal, Quebec H3G 1M8, Canada; Corresponding author at: Department of Mechanical, Industrial and Aerospace Engineering, Concordia University, Montréal, Quebec H3G 1M8, Canada.Concordia Institute for Information Systems Engineering, Concordia University, Montréal, Quebec H3G1M8, CanadaConcordia Institute for Information Systems Engineering, Concordia University, Montréal, Quebec H3G1M8, CanadaDepartment of Mechanical, Industrial and Aerospace Engineering, Concordia University, Montréal, Quebec H3G 1M8, CanadaSchool of Aerospace Engineering, Beijing Institute of Technology, Beijing 100081, ChinaDepartment of Mechanical, Industrial and Aerospace Engineering, Concordia University, Montréal, Quebec H3G 1M8, CanadaIn this paper, we develop a series of Artificial Neural Networks (ANN) using different chemical kinetic and thermodynamic input parameters to predict detonation cell sizes. The feedforward neural networks are trained and validated using available experimental data from the Caltech detonation database covering a wide variety of gaseous combustible mixtures at different initial conditions. For each combination of input parameters, a multiple-stage process is followed, which is described in detail, to first determine the best hyperparameters of the ANN (hidden layers, nodes per layer, etc.) and secondly to establish through a fitting process the optimal parameters for each specific network. The performance of the artificial neural networks with different input features is assessed using data from the same source, but that is kept independent and separate from the training and validation process of the ANN. It is found that ANN with three features can provide an accurate estimation of detonation cell size, while increasing the number of features does not improve the accuracy of the ANN. It is also found that the input parameters with the best performance relate indirectly to the stability parameter χ.http://www.sciencedirect.com/science/article/pii/S2666052022000346Machine learningArtificial neural networkGaseous detonationCell sizeChemical kineticsStability parameter |
| spellingShingle | Georgios Bakalis Maryam Valipour Jamal Bentahar Lyes Kadem Honghui Teng Hoi Dick Ng Detonation cell size prediction based on artificial neural networks with chemical kinetics and thermodynamic parameters Fuel Communications Machine learning Artificial neural network Gaseous detonation Cell size Chemical kinetics Stability parameter |
| title | Detonation cell size prediction based on artificial neural networks with chemical kinetics and thermodynamic parameters |
| title_full | Detonation cell size prediction based on artificial neural networks with chemical kinetics and thermodynamic parameters |
| title_fullStr | Detonation cell size prediction based on artificial neural networks with chemical kinetics and thermodynamic parameters |
| title_full_unstemmed | Detonation cell size prediction based on artificial neural networks with chemical kinetics and thermodynamic parameters |
| title_short | Detonation cell size prediction based on artificial neural networks with chemical kinetics and thermodynamic parameters |
| title_sort | detonation cell size prediction based on artificial neural networks with chemical kinetics and thermodynamic parameters |
| topic | Machine learning Artificial neural network Gaseous detonation Cell size Chemical kinetics Stability parameter |
| url | http://www.sciencedirect.com/science/article/pii/S2666052022000346 |
| work_keys_str_mv | AT georgiosbakalis detonationcellsizepredictionbasedonartificialneuralnetworkswithchemicalkineticsandthermodynamicparameters AT maryamvalipour detonationcellsizepredictionbasedonartificialneuralnetworkswithchemicalkineticsandthermodynamicparameters AT jamalbentahar detonationcellsizepredictionbasedonartificialneuralnetworkswithchemicalkineticsandthermodynamicparameters AT lyeskadem detonationcellsizepredictionbasedonartificialneuralnetworkswithchemicalkineticsandthermodynamicparameters AT honghuiteng detonationcellsizepredictionbasedonartificialneuralnetworkswithchemicalkineticsandthermodynamicparameters AT hoidickng detonationcellsizepredictionbasedonartificialneuralnetworkswithchemicalkineticsandthermodynamicparameters |