Combining artificial neural network and multi-objective optimization to reduce a heavy-duty diesel engine emissions and fuel consumption

Nondominated sorting genetic algorithm Ⅱ(NSGA-Ⅱ) is well known for engine optimization problem. Artificial neural networks(ANNs) followed by multi-objective optimization including a NSGA-Ⅱ and strength pareto evolutionary algorithm(SPEA2) were used to optimize the operating parameters of a compression ignition(CI) heavy-duty diesel engine. First, a multi-layer perception(MLP) network was used for the ANN modeling and the back propagation algorithm was utilized as training algorithm. Then, two different multi-objective evolutionary algorithms were implemented to determine the optimal engine parameters. The objective of the present study is to decide which algorithm is preferable in terms of performance in engine emission and fuel consumption optimization problem.

Convergence of shape optimization calculations of mechanical components using adaptive biological growth and iterative finite element methods

Shape optimization of mechanical components is one of the issues that have been considered in recent years. Different methods were presented such as adaptive biological for reducing costs and increasing accuracy. The effects of step factor, the number of control points and the definition way of control points coordinates in convergence rate were studied. A code was written using ANSYS Parametric Design Language (APDL) which receives the studied parameters as input and obtains the optimum shape for the components. The results show that for achieving successful optimization, step factor should be in a specific range. It is found that the use of any coordinate system in defining control points coordinates and selection of any direction for stimulus vector of algorithm will also result in optimum shape. Furthermore, by increasing the number of control points, some non-uniformities are created in the studied boundary. Achieving acceptable accuracy seems impossible due to the creation of saw form at the studied boundary which is called "saw position".

A comparative study of hybrid electric vehicle fuel consumption over diverse driving cycles

Environmental pollution and declining resources of fossil fuels in recent years,have increased demand for better fuel economy and less pollution for ground transportation.Among the alternative solutions provided by researchers in recent decades,hybrid electric vehicles consisted of an internal combustion engine and an electric motor have been considered as a promising solution in the short-term.In the present study,fuel economy characteristics of a parallel hybrid electric vehicle are investigated by using numerical simulation.The simulation methodology is based on a fast forward facing simulation model of a parallel hybrid and an internal combustion engine powertrains.The objective of this study is to present the main parameters which result in an optimum combination of hybrid powertrain components in order to obtain a better fuel economy of hybrid powertrains regarding different driven cycles and hybridization factors.Then,the fuel consumption of the parallel hybrid electric vehicles are compared considering various driven cycles and hybridization factors.The results showed that the better fuel economy of hybrid powertrains increases by decreasing average load of the test cycle and the point of the best fuel economy for a particular average load of the cycle moves towards higher hybridization factors when the average load of the test cycle is reduced.

Design and energy absorption enhancement of vehicle hull under high dynamic loads

V-shape hulls are widely used in peacekeeping efforts such as demining vehicles in order to deflect the blast energy and reduce the effects of mine blast. Blast resistant design and energy absorption enhancement of V-shape plates were carried out using finite element analysis package ABAQUS. Various geometries of V-shape plates with and without interlayer of materials like Al-foams and honeycomb were employed to analyze their effects on the deformation of the plate and applied stresses and strains. The results obtained show that application of metallic foams leads to better response of the plate and consequently results in more energy dissipation, less dame to vehicle and enhances crew survivability.

Applying machine learning techniques to predict laminar burning velocity for ammonia/hydrogen/air mixtures

Ammonia utilisation in internal combustion engines has attracted wide interest due to the current trend toward decarbonisation,as ammonia is a zero-carbon fuel with different combustion properties to hydrocarbons.The laminar burning velocity(LBV)is a fundamental property of fuels with a significant effect on the combustion processes and accurate calculations and measurements of the LBV over a wide range of fuel blends,pressures and flow conditions is a time-consuming,complicated procedure.The main goal of the current study is to predict the LBV of NH_(3)/H_(2)/air mixtures using a hybrid machine learning(ML)approach based on a training dataset consisting of both the experimental LBV values and additional data obtained from numerical simulations with a detailed kinetic model.Initial ML model training data is collected from existing experimental LBV in the literature for NH_(3)/H_(2)/air mixtures.Then,synthetic data is generated using one-dimensional(1D)simulations to reduce data inhomogeneity and increase accuracy of the ML model.In total,24 different ML algorithms are tested to find the best model both for the experimental and the hybrid dataset.The results suggest that both Gaussian Process Regression(GPR)and Neural Networks(NNs)can be utilised to predict LBV of NH_(3)/H_(2)/air mixtures with reasonable accuracy.The hybrid ML model achieved a coefficient of determination of R^(2)=0.998.Finally,hybrid ML model hyperparameters are optimised to achieve a coefficient of determination of R^(2)=0.999.It was also found that ML can speed up LBV computation from 9500 to 27000 times compared to 1D simulations with a reduced mechanism.