Search Results

A Machine Learning-Genetic Algorithm (ML-GA) approach was developed to virtually discover optimum designs using training data generated from multi-dimensional simulations. Machine learning (ML) presents a pathway to transform complex physical processes that occur in a combustion engine into compact informational processes. In the present work, a total of over 2000 sector-mesh computational fluid dynamics (CFD) simulations of a heavy-duty engine were performed. These were run concurrently on a supercomputer to reduce overall turnaround time. The engine being optimized was run on a low-octane (RON70) gasoline fuel under partially premixed compression ignition (PPCI) mode. A total of nine input parameters were varied, and the CFD simulation cases were generated by randomly sampling points from this nine-dimensional input space. These input parameters included fuel injection strategy, injector design, and various in-cylinder flow and thermodynamic conditions at intake valve closure (IVC).

Many recent studies have shown that the Reactivity Controlled Compression Ignition (RCCI) combustion strategy can achieve high efficiency with low emissions. However, it has also been revealed that RCCI combustion is difficult at high loads due to its premixed nature. To operate at moderate to high loads with gasoline/diesel dual fuel, high amounts of EGR or an ultra low compression ratio have shown to be required. Considering that both of these approaches inherently lower thermodynamic efficiency, in this study natural gas was utilized as a replacement for gasoline as the low-reactivity fuel. Due to the lower reactivity (i.e., higher octane number) of natural gas compared to gasoline, it was hypothesized to be a better fuel for RCCI combustion, in which a large reactivity gradient between the two fuels is beneficial in controlling the maximum pressure rise rate.

It is estimated that operating continuously on a B20 fuel containing the current allowable ASTM specification limits for metal impurities in biodiesel could result in a doubling of ash exposure relative to lube-oil-derived ash. The purpose of this study was to determine if a fuel containing metals at the ASTM limits could cause adverse impacts on the performance and durability of diesel emission control systems. An accelerated durability test method was developed to determine the potential impact of these biodiesel impurities. The test program included engine testing with multiple DPF substrate types as well as DOC and SCR catalysts. The results showed no significant degradation in the thermo-mechanical properties of cordierite, aluminum titanate, or silicon carbide DPFs after exposure to 150,000 mile equivalent biodiesel ash and thermal aging. However, exposure of a cordierite DPF to 435,000 mile equivalent aging resulted in a 69% decrease in the thermal shock resistance parameter.

Technical groups worldwide have been actively developing specifications and requirements for biodegradable hydraulic fluids for mobile applications. These groups have recognized that an industry-wide specification is necessary due to the increase in environmental awareness in the agriculture, construction, forestry, and mining industries, and to the increasing number of local regulations primarily throughout Europe. Caterpillar has responded to this need by publishing a requirement, Caterpillar BF-1, that may be used by Caterpillar dealers, customers, and industry to help select high-performance biodegradable hydraulic fluids. This requirement was written with the input of several organizations that are known to be involved with the development of similar types of specifications and requirements.

A global optimization method has been developed for an engine simulation code and utilized in the search of optimal fuel injection strategies. This method uses a Lagrange interpolation function which interpolates engine output data generated at the vertices and the intermediate points of the input parameters. This interpolation function is then used to find a global minimum over the entire parameter set, which in turn becomes the starting point of a CFD-based optimization. The CFD optimization is based on a steepest descent method with an adaptive cost function, where the line searches are performed with a fast-converging backtracking algorithm. The adaptive cost function is based on the penalty method, where the penalty coefficient is increased after every line search. The parameter space is normalized and, thus, the optimization occurs over the unit cube in higher-dimensional space.

Optimal fuel injection strategies are obtained with a micro-genetic algorithm and an adaptive gradient method for a nonroad, medium-speed DI diesel engine equipped with a multi-orifice, asynchronous fuel injection system. The gradient optimization utilizes a fast-converging backtracking algorithm and an adaptive cost function which is based on the penalty method, where the penalty coefficient is increased after every line search. The micro-genetic algorithm uses parameter combinations of the best two individuals in each generation until a local convergence is achieved, and then generates a random population to continue the global search. The optimizations have been performed for a two pulse fuel injection strategy where the optimization parameters are the injection timings and the nozzle orifice diameters.

Faced with competitive environments, pressure to lower development costs and aggressive timelines engineers are not only increasingly adopting numerical simulation techniques but are also embracing design optimization schemes to augment their efforts. These techniques not only provide more understanding of the trade-offs but are also capable of proactively guiding the decision making process. However, design optimization and exploration tools have struggled to find complete acceptance and are typically underutilized in many applications; especially in situations where the algorithms have to compete with existing swift decision making processes. In this paper we demonstrate how the type of setup and algorithmic choice can have an influence and make optimization more lucrative in a new product development atmosphere. We also present some results from a design exploration activity, involving linkage and structural development, of an earth moving machine application.

Neural networks are useful tools for optimization studies since they are very fast, so that while capturing the accuracy of multi-dimensional CFD calculations or experimental data, they can be run numerous times as required by many optimization techniques. This paper describes how a set of neural networks trained on a multi-dimensional CFD code to predict pressure, temperature, heat flux, torque and emissions, have been used by a genetic algorithm in combination with a hill-climbing type algorithm to optimize operating parameters of a diesel engine over the entire speed-torque map of the engine. The optimized parameters are mass of fuel injected per cycle, shape of the injection profile for dual split injection, start of injection, EGR level and boost pressure. These have been optimized for minimum emissions. Another set of neural networks have been trained to predict the optimized parameters, based on the speed-torque point of the engine.

Neural networks have been used for engine computations in the recent past. One reason for using neural networks is to capture the accuracy of multi-dimensional CFD calculations or experimental data while saving computational time, so that system simulations can be performed within a reasonable time frame. This paper describes three methods to improve upon neural network predictions. Improvement is demonstrated for in-cylinder pressure predictions in particular. The first method incorporates a physical combustion model within the transfer function of the neural network, so that the network predictions incorporate physical relationships as well as mathematical models to fit the data. The second method shows how partitioning the data into different regimes based on different physical processes, and training different networks for different regimes, improves the accuracy of predictions.

The University of Wisconsin - Madison hybrid vehicle team has designed and constructed a four-wheel drive, charge sustaining, parallel hybrid-electric sport utility vehicle for entry into the FutureTruck 2003 competition. This is a multi-year project utilizing a 2002 4.0 liter Ford Explorer as the base vehicle. Wisconsin's FutureTruck, nicknamed the ‘Moolander’, weighs 2000 kg and includes a prototype aluminum frame. The Moolander uses a high efficiency, 1.8 liter, common rail, turbo-charged, compression ignition direct injection (CIDI) engine supplying 85 kW of peak power and an AC induction motor that provides an additional 60 kW of peak power. The 145 kW hybrid drivetrain will out-accelerate the stock V6 powertrain while producing similar emissions and drastically reducing fuel consumption. The PNGV Systems Analysis Toolkit (PSAT) model predicts a Federal Testing Procedure (FTP) combined driving cycle fuel economy of 16.05 km/L (37.8 mpg).

The University of Wisconsin - Madison FutureTruck Team has designed and built a four-wheel drive, charge sustaining, parallel hybrid-electric sport utility vehicle for entry into the FutureTruck 2001 competition. The base vehicle is a 2000 Chevrolet Suburban. Our FutureTruck is nicknamed the “Moollennium” and weighs approximately 2427 kg. The vehicle uses a high efficiency, 2.5 liter, turbo-charged, compression ignition common rail, direct-injection engine supplying approximately 104 kW of peak power and a three phase AC induction motor that provides an additional 68.5 kW of peak power. This hybrid drivetrain is an attractive alternative to the large displacement V8 drivetrain, as it provides comparable performance with lower emissions and fuel consumption. The PNGV Systems Analysis Toolkit (PSAT) model predicts a Federal Testing Procedure (FTP) urban driving cycle fuel economy of 11.24 km/L (26.43 mpg) with California Ultra Low Emission Vehicle (ULEV) emissions levels.

We have developed a methodology for predicting combustion and emissions in a Homogeneous Charge Compression Ignition (HCCI) Engine. The methodology judiciously uses a fluid mechanics code followed by a chemical kinetics code to achieve great reduction in the computational requirements; to a level that can be handled with current computers. In previous papers, our sequential, multi-zone methodology has been applied to HCCI combustion of short-chain hydrocarbons (natural gas and propane). Applying the same procedure to long-chain hydrocarbons (iso-octane) results in unacceptably long computational time. In this paper, we show how the computational time can be made acceptable by developing a segregated solver. This reduces the run time of a ten-zone problem by an order of magnitude and thus makes it much more practical to make combustion studies of long-chain hydrocarbons.

Natural gas quality, in terms of the volume fraction of higher hydrocarbons, strongly affects the auto-ignition characteristics of the air-fuel mixture, the engine performance and its controllability. The influence of natural gas composition on engine operation has been investigated both experimentally and through chemical kinetic based cycle simulation. A range of two component gas mixtures has been tested with methane as the base fuel. The equivalence ratio (0.3), the compression ratio (19.8), and the engine speed (1000 rpm) were held constant in order to isolate the impact of fuel autoignition chemistry. For each fuel mixture, the start of combustion was phased near top dead center (TDC) and then the inlet mixture temperature was reduced. These experimental results have been utilized as a source of data for the validation of a chemical kinetic based full-cycle simulation.

This paper discusses the compression ratio influence on maximum load of a Natural Gas HCCI engine. A modified Volvo TD100 truck engine is controlled in a closed-loop fashion by enriching the Natural Gas mixture with Hydrogen. The first section of the paper illustrates and discusses the potential of using hydrogen enrichment of natural gas to control combustion timing. Cylinder pressure is used as the feedback and the 50 percent burn angle is the controlled parameter. Full-cycle simulation is compared to some of the experimental data and then used to enhance some of the experimental observations dealing with ignition timing, thermal boundary conditions, emissions and how they affect engine stability and performance. High load issues common to HCCI are discussed in light of the inherent performance and emissions tradeoff and the disappearance of feasible operating space at high engine loads.

Homogeneous Charge Compression Ignition (HCCI) has been proposed for natural gas engines in heavy duty stationary power generation applications. A number of researchers have demonstrated, through simulation and experiment, the feasibility of obtaining high gross indicated thermal efficiencies and very low NOx emissions at reasonable load levels. With a goal of eventual commercialization of these engines, this paper sets forth some of the primary challenges in obtaining high brake thermal efficiency from production feasible engines. Experimental results, in conjunction with simulation and analysis, are used to compare HCCI operation with traditional lean burn spark ignition performance. Current HCCI technology is characterized by low power density, very dilute mixtures, and low combustion efficiency. The quantitative adverse effect of each of these traits is demonstrated with respect to the brake thermal efficiency that can be expected in real world applications.

Previous research has shown that the homogeneous charge compression ignition (HCCI) combustion concept holds promise for reducing pollutants (i.e. NOx, soot) while maintaining high thermal efficiency. However, it can be difficult to control the operation of the HCCI engines even under steady state running conditions. Power density may also be limited if high inlet air temperatures are used for achieving ignition. A methodology using a small pilot quantity of diesel fuel injected during the compression stroke to improve the power density and operation control is considered in this paper. Multidimensional computations were carried out for an HCCI engine based on a CAT3401 engine. The computations show that the required initial temperature for ignition is reduced by about 70 K for the cases of the diesel pilot charge and a 25∼35% percent increase in power density was found for those cases without adversely impacting the NOx emissions.

The University of Wisconsin - Madison FutureTruck Team has designed and built a four-wheel drive, charge sustaining, parallel hybrid-electric sport utility vehicle for entry into the FutureTruck 2002 competition. This is a two-year project with tiered goals; the base vehicle for both years is a 2002 Ford Explorer. Wisconsin's FutureTruck, nicknamed the ‘Moolander’, weighs approximately 2050 kg. The vehicle uses a high efficiency, 2.5 liter, turbo-charged, compression ignition common rail, direct-injection engine supplying approximately 100 kW of peak power and a AC induction motor that provides an additional 33 kW of peak power. This hybrid drivetrain is an attractive alternative to the large displacement V6 drivetrain, as it provides comparable performance with similar emissions and drastically reduced fuel consumption.

Interest in operational cost reduction is driving engine manufacturers to consider low-cost fuel substitution in heavy-duty diesel engines. These dual-fuel (DF) engines could be operated either in diesel-only mode or operated with premixed natural gas (NG) ignited by a pilot flame of compression-ignited direct-injected diesel fuel. Under certain conditions, dual-fuel operation can result in increased cycle-to-cycle variability (CCV) during combustion. CFD can greatly help in understanding and identifying critical parameters influencing CCV. Innovative modelling techniques and large computing resources are needed to investigate the factors affecting CCV in dual-fuel engines. This paper discusses the use of the High Performance Computing resource Titan, at Oak Ridge National Laboratory, to investigate CCV of a dual-fuel engine.

The standard capability of engine experimental studies is that ensemble averaged quantities like in-cylinder pressure from multiple cycles and emissions are reported and the cycle to cycle variation (CCV) of indicated mean effective pressure (IMEP) is captured from many consecutive combustion cycles for each test condition. However, obtaining 3D spatial distribution of all the relevant quantities such as fuel-air mixing, temperature, turbulence levels and emissions from such experiments is a challenging task. Computational Fluid Dynamics (CFD) simulations of engine flow and combustion can be used effectively to visualize such 3D spatial distributions. A dual fuel engine is considered in the current study, with manifold injected natural gas (NG) and direct injected diesel pilot for ignition.

This paper presents on-the-fly chemical mechanism reduction (termed as dynamic mechanism reduction) for speeding up the chemistry solution for practical internal combustion engine simulations. Small mechanisms are built at each time step which are valid under the local conditions of each cell. The Directed Relation Graph with Error Propagation (DRGEP) algorithm is used for generating local skeletal mechanisms. Dynamic mechanism reduction is combined with adaptive zoning (termed as multi-zone) to achieve good computational speed-up for engine simulations. The accuracy and efficiency of dynamic mechanism reduction is evaluated for a wide range of scenarios including (a) Diesel combustion, (b) Homogeneous Charge Compression Ignition (HCCI) combustion, and (c) Dual fuel combustion.