PUTTING A DARWINIAN SPIN ON THE DIESEL ENGINE

September 22, 2000

SCIENCE & ENGINEERING NEWS

New York, N.Y. — Bruce Schechter reports for the New York Times that to engineers, the silent machinery of a living cell is a humbling reminder of the crudeness of their own designs. Every cell is a tiny, elegant engine that converts chemical fuel to energy while emitting innocuous byproducts. By contrast, a diesel engine, the most efficient type of internal combustion engine, is a monument to waste.

So engineers have begun to imitate nature by letting a computerized version of Darwinian natural selection guide their design processes. In this approach, known as genetic algorithms, a computer simulates the performance of a group of machines, each with a slightly different design. These machines compete against one another and, just as in evolution, the best performing, or fittest, survives, to serve as the basis for another generation of designs. This process is repeated until an evolutionary winner, whose performance is maximized, emerges.

Scientists at the University of Wisconsin have recently applied this approach to the design of a diesel engine that, while far from the biological ideal, is more efficient and produces less waste than others in its class.

Recently genetic algorithms have been used in a wide variety of fields as diverse as creating artistic masterpieces, playing expert checkers and designing robots. But, Dr. Peter Senecal, a postdoctoral student at the university, said, “This is the first application of genetic algorithms to engine design.”

Despite numerous refinements, the modern diesel engine is remarkably similar to the 1893 prototype by Rudolf Diesel. A French-born engineer, Diesel made his engine, using the newly discovered principles of thermodynamics, as a replacement of the gasoline engine, then in its infancy and extremely inefficient.

In a gasoline engine, a mixture of air and gas is injected into the engine’s cylinder and then compressed by the piston. The mixture, ignited by a spark from the spark plug, explodes, and this explosive energy drives the piston and eventually makes the wheels turn round.

The efficiency of a gas engine is limited by (among other things) the compression ratio, the amount that the fuel-air mixture is compressed by the cylinder: the more compression the more efficient the engine. But when a gas is compressed it grows hotter and this heating can prematurely ignite the fuel-air mixture, causing a phenomenon known as knocking.

But in a diesel engine, this heating is put to good use. The cylinder of a diesel engine is filled with only air, which is compressed to as much as 500 pounds a square inch, causing its temperature to rise to 1,000 degrees. Then, a tiny bit of fuel is injected into the combustion chamber, where the compressed, heated air causes it to explode without a spark plug. The result is an engine more efficient than one that runs on gasoline, but also one more complex because it requires machinery for compressing and injecting the fuel.

Refining the diesel engine has been a major project of engineers, arousing the interest of vast research laboratories and garage tinkerers alike. Their efforts have allowed diesel engines to evolve from hulking behemoths appropriate only in mines, factories and power plants to engines suitable for automobiles. Still, said Dr. Rolf D. Reiz, a professor of mechanical engineering at the University of Wisconsin, “For the first 80 years of diesel history, design has all been done by trial and error.”

Over the past two decades, with the advent of high-speed computing, this has begun to change, but slowly, because the problem is so difficult. To understand how a diesel engine works in detail requires more than a knowledge of piston and valve; the physics of fuel injection, exactly how fuel breaks into a mist and swirls around a cylinder, must be considered; every detail of physics and chemistry of the exploding, turbulent fireball of fuel must be tracked.

It’s the kind of problem that chokes even the most powerful supercomputers. But over the last few years, computers running software developed by Dr. Reiz and his colleagues at government laboratories, universities and in industry have begun to make progress, though the progress is slow.

“A typical simulation will run for several days on a supercomputer,” Dr. Reiz said. That simulation is of one engine cycle, which actually takes place in less than a tenth of a second. “We’re much slower than real-time. It’s kind of like the early days of weather prediction where you’d be predicting the weather that occurred three weeks ago. We’re in that realm, but nevertheless we can use the tools to do things that cannot be done in the laboratory.”

The computer simulation can be used to show the effects of varying any of the parameters of the engine’s design: the timing of the fuel injection, how much fuel is injected, at what pressure and so on. There can be dozens of parameters to adjust, each of which affects the others. Finding an optimal combination by trial and error on a real-world engine could take practically forever. But, with simulations taking two days apiece, trying all the combinations of variables with a computer does not seem to work much faster.

After searching the mathematical literature, Dr. Senecal, a student of Dr. Reiz, found a better way to speed up the evolution of designs by borrowing genetic principles.

The problem of how to optimize a process based on many parameters can be likened to locating the peak of a mountain range. One approach is to start somewhere and keep walking up. Eventually the walker will reach a point where he can walk up no further, and this is a peak. But is this the highest peak?

Evolution confronts this problem in its search for an optimal combination of genes for survival. Mutating genes can be thought of as tweaking parameters. Somehow evolution manages to find a way of conquering mountains while not planting its flag on insignificant peaks. It does so by sending out not just one explorer but a veritable army.

By randomly mutating genes, evolution scatters a group of explorers across the so-called fitness landscape. The ones that find the highest ground, and are thus the fittest, survive and share their genes. Their offspring will then explore an area of the mountain closer to the place their parents landed, and perhaps discover even higher ground corresponding to even greater fitness.

The basic idea of genetic algorithms is that entire classes of designs, strategies or artworks can be written down as depending upon a set of parameters. In the case of engine design, Dr. Senecal chose to make these parameters correspond to elements like injection timing, pressure and other operating variables. These parameters are, in effect, the genes of the engine and the computer starts by generating a random set of these genes.

The engines bearing these “genes” are simulated and the results are compared. Dr. Senecal rated the fitness of his engines on their fuel efficiency and the amount of soot and nitrate wastes they generated. The best of these designs are mated together by swapping genes in a way inspired by nature, and the process is repeated.

Ordinarily, genetic algorithms require hundreds of “organisms” to be evaluated each generation, but given how time-consuming it is to simulate a diesel engine cycle, Dr. Senecal needed a better technique. He discovered in the literature an approach called microgenetic algorithms, a refinement that allowed him to consider generations of just five organisms. Still, using a Silicon Graphics Origin 2000 supercomputer with 32 processors it took more than two weeks of continuous operation to find an optimal set of parameters.

The effort was worthwhile; Dr. Senecal’s test engine consumed 15 percent less fuel than a standard engine while producing one-third as much nitric oxide and half the soot.

These results go beyond theoretical. To make sure their simulation corresponds to reality, the Wisconsin scientists have simulated a Caterpillar truck engine used to power real- world machinery. By tweaking the parameters of this real-world engine they confirmed their computer prediction, and that confirmation is of great interest to engine designers.

“What we can now do,” Dr. Reiz said, “is indicate to engine designers those variables that are most important or ones that might have been overlooked had it not been for the computer identifying it.” In particular, these studies have highlighted the importance of injecting the fuel into the cylinder in tiny bursts instead of in a single pulse. Doing so increases the surface area of the fuel, which leads to cleaner and more efficient burning.

So far the Wisconsin engineers have focused only on tweaking parameters while keeping the overall engine design constant. The next stage will involve having the computer vary the engine shape, particularly in the curve of the cylinder head.

“If you look at the shape of the piston in the truck engines you see it really hasn’t changed much in the last 40 years,” Dr. Reiz said. Meanwhile, the fuel injection system has changed radically.

“In the old days injectors used to operate at 2,000 pounds per square inch,” he added. “Now we’re talking about 20,000 pounds per square inch. So it stands to reason that an engine design that might have been useful for 2,000 pounds per square inch may not be useful for 20,000 pounds per square inch.”

Nobody has tinkered with the design because tinkering would involve creating a design, having a new piston made, and running it in the laboratory without much guidance from theory, and doing it over and over again. Rather than confront this expensive and frustrating process, designers have left the piston alone.

Now, Dr. Reiz said, “We can do all this on the computer without even having to cut metal.” The result of this application of genetic algorithms could mean, in a few years, a revolution in the diesel engine.

============================================================

Subscribe to HPCwire's Weekly Update!

Be the most informed person in the room! Stay ahead of the tech trends with industy updates delivered to you every week!

Intel Debuts New GPU – Ponte Vecchio – and Outlines Aspirations for oneAPI

November 17, 2019

Intel today revealed a few more details about its forthcoming Xe line of GPUs – the top SKU is named Ponte Vecchio and will be used in Aurora, the first planned U.S. exascale computer. Intel also provided a glimpse of Read more…

By John Russell

SC19: Welcome to Denver

November 17, 2019

A significant swath of the HPC community has come to Denver for SC19, which began today (Sunday) with a rich technical program. As is customary, the ribbon cutting for the Expo Hall opening is Monday at 6:45pm, with the Read more…

By Tiffany Trader

SC19’s HPC Impact Showcase Chair: AI + HPC a ‘Speed Train’

November 16, 2019

This year’s chair of the HPC Impact Showcase at the SC19 conference in Denver is Lori Diachin, who has spent her career at the spearhead of HPC. Currently deputy director for the U.S. Department of Energy’s (DOE) Read more…

By Doug Black

Microsoft Azure Adds Graphcore’s IPU

November 15, 2019

Graphcore, the U.K. AI chip developer, is expanding collaboration with Microsoft to offer its intelligent processing units on the Azure cloud, making Microsoft the first large public cloud vendor to offer the IPU designe Read more…

By George Leopold

At SC19: What Is UrgentHPC and Why Is It Needed?

November 14, 2019

The UrgentHPC workshop, taking place Sunday (Nov. 17) at SC19, is focused on using HPC and real-time data for urgent decision making in response to disasters such as wildfires, flooding, health emergencies, and accidents. We chat with organizer Nick Brown, research fellow at EPCC, University of Edinburgh, to learn more. Read more…

By Tiffany Trader

AWS Solution Channel

Making High Performance Computing Affordable and Accessible for Small and Medium Businesses with HPC on AWS

High performance computing (HPC) brings a powerful set of tools to a broad range of industries, helping to drive innovation and boost revenue in finance, genomics, oil and gas extraction, and other fields. Read more…

IBM Accelerated Insights

Data Management – The Key to a Successful AI Project

 

Five characteristics of an awesome AI data infrastructure

[Attend the IBM LSF & HPC User Group Meeting at SC19 in Denver on November 19!]

AI is powered by data

While neural networks seem to get all the glory, data is the unsung hero of AI projects – data lies at the heart of everything from model training to tuning to selection to validation. Read more…

China’s Tencent Server Design Will Use AMD Rome

November 13, 2019

Tencent, the Chinese cloud giant, said it would use AMD’s newest Epyc processor in its internally-designed server. The design win adds further momentum to AMD’s bid to erode rival Intel Corp.’s dominance of the glo Read more…

By George Leopold

Intel Debuts New GPU – Ponte Vecchio – and Outlines Aspirations for oneAPI

November 17, 2019

Intel today revealed a few more details about its forthcoming Xe line of GPUs – the top SKU is named Ponte Vecchio and will be used in Aurora, the first plann Read more…

By John Russell

SC19: Welcome to Denver

November 17, 2019

A significant swath of the HPC community has come to Denver for SC19, which began today (Sunday) with a rich technical program. As is customary, the ribbon cutt Read more…

By Tiffany Trader

SC19’s HPC Impact Showcase Chair: AI + HPC a ‘Speed Train’

November 16, 2019

This year’s chair of the HPC Impact Showcase at the SC19 conference in Denver is Lori Diachin, who has spent her career at the spearhead of HPC. Currently Read more…

By Doug Black

Cray, Fujitsu Both Bringing Fujitsu A64FX-based Supercomputers to Market in 2020

November 12, 2019

The number of top-tier HPC systems makers has shrunk due to a steady march of M&A activity, but there is increased diversity and choice of processing compon Read more…

By Tiffany Trader

Intel AI Summit: New ‘Keem Bay’ Edge VPU, AI Product Roadmap

November 12, 2019

At its AI Summit today in San Francisco, Intel touted a raft of AI training and inference hardware for deployments ranging from cloud to edge and designed to support organizations at various points of their AI journeys. The company revealed its Movidius Myriad Vision Processing Unit (VPU)... Read more…

By Doug Black

IBM Adds Support for Ion Trap Quantum Technology to Qiskit

November 11, 2019

After years of percolating in the shadow of quantum computing research based on superconducting semiconductors – think IBM, Rigetti, Google, and D-Wave (quant Read more…

By John Russell

Tackling HPC’s Memory and I/O Bottlenecks with On-Node, Non-Volatile RAM

November 8, 2019

On-node, non-volatile memory (NVRAM) is a game-changing technology that can remove many I/O and memory bottlenecks and provide a key enabler for exascale. That’s the conclusion drawn by the scientists and researchers of Europe’s NEXTGenIO project, an initiative funded by the European Commission’s Horizon 2020 program to explore this new... Read more…

By Jan Rowell

MLPerf Releases First Inference Benchmark Results; Nvidia Touts its Showing

November 6, 2019

MLPerf.org, the young AI-benchmarking consortium, today issued the first round of results for its inference test suite. Among organizations with submissions wer Read more…

By John Russell

Supercomputer-Powered AI Tackles a Key Fusion Energy Challenge

August 7, 2019

Fusion energy is the Holy Grail of the energy world: low-radioactivity, low-waste, zero-carbon, high-output nuclear power that can run on hydrogen or lithium. T Read more…

By Oliver Peckham

Using AI to Solve One of the Most Prevailing Problems in CFD

October 17, 2019

How can artificial intelligence (AI) and high-performance computing (HPC) solve mesh generation, one of the most commonly referenced problems in computational engineering? A new study has set out to answer this question and create an industry-first AI-mesh application... Read more…

By James Sharpe

Cray Wins NNSA-Livermore ‘El Capitan’ Exascale Contract

August 13, 2019

Cray has won the bid to build the first exascale supercomputer for the National Nuclear Security Administration (NNSA) and Lawrence Livermore National Laborator Read more…

By Tiffany Trader

DARPA Looks to Propel Parallelism

September 4, 2019

As Moore’s law runs out of steam, new programming approaches are being pursued with the goal of greater hardware performance with less coding. The Defense Advanced Projects Research Agency is launching a new programming effort aimed at leveraging the benefits of massive distributed parallelism with less sweat. Read more…

By George Leopold

AMD Launches Epyc Rome, First 7nm CPU

August 8, 2019

From a gala event at the Palace of Fine Arts in San Francisco yesterday (Aug. 7), AMD launched its second-generation Epyc Rome x86 chips, based on its 7nm proce Read more…

By Tiffany Trader

D-Wave’s Path to 5000 Qubits; Google’s Quantum Supremacy Claim

September 24, 2019

On the heels of IBM’s quantum news last week come two more quantum items. D-Wave Systems today announced the name of its forthcoming 5000-qubit system, Advantage (yes the name choice isn’t serendipity), at its user conference being held this week in Newport, RI. Read more…

By John Russell

Ayar Labs to Demo Photonics Chiplet in FPGA Package at Hot Chips

August 19, 2019

Silicon startup Ayar Labs continues to gain momentum with its DARPA-backed optical chiplet technology that puts advanced electronics and optics on the same chip Read more…

By Tiffany Trader

Crystal Ball Gazing: IBM’s Vision for the Future of Computing

October 14, 2019

Dario Gil, IBM’s relatively new director of research, painted a intriguing portrait of the future of computing along with a rough idea of how IBM thinks we’ Read more…

By John Russell

Leading Solution Providers

ISC 2019 Virtual Booth Video Tour

CRAY
CRAY
DDN
DDN
DELL EMC
DELL EMC
GOOGLE
GOOGLE
ONE STOP SYSTEMS
ONE STOP SYSTEMS
PANASAS
PANASAS
VERNE GLOBAL
VERNE GLOBAL

Intel Confirms Retreat on Omni-Path

August 1, 2019

Intel Corp.’s plans to make a big splash in the network fabric market for linking HPC and other workloads has apparently belly-flopped. The chipmaker confirmed to us the outlines of an earlier report by the website CRN that it has jettisoned plans for a second-generation version of its Omni-Path interconnect... Read more…

By Staff report

Kubernetes, Containers and HPC

September 19, 2019

Software containers and Kubernetes are important tools for building, deploying, running and managing modern enterprise applications at scale and delivering enterprise software faster and more reliably to the end user — while using resources more efficiently and reducing costs. Read more…

By Daniel Gruber, Burak Yenier and Wolfgang Gentzsch, UberCloud

Dell Ramps Up HPC Testing of AMD Rome Processors

October 21, 2019

Dell Technologies is wading deeper into the AMD-based systems market with a growing evaluation program for the latest Epyc (Rome) microprocessors from AMD. In a Read more…

By John Russell

Rise of NIH’s Biowulf Mirrors the Rise of Computational Biology

July 29, 2019

The story of NIH’s supercomputer Biowulf is fascinating, important, and in many ways representative of the transformation of life sciences and biomedical res Read more…

By John Russell

Xilinx vs. Intel: FPGA Market Leaders Launch Server Accelerator Cards

August 6, 2019

The two FPGA market leaders, Intel and Xilinx, both announced new accelerator cards this week designed to handle specialized, compute-intensive workloads and un Read more…

By Doug Black

When Dense Matrix Representations Beat Sparse

September 9, 2019

In our world filled with unintended consequences, it turns out that saving memory space to help deal with GPU limitations, knowing it introduces performance pen Read more…

By James Reinders

With the Help of HPC, Astronomers Prepare to Deflect a Real Asteroid

September 26, 2019

For years, NASA has been running simulations of asteroid impacts to understand the risks (and likelihoods) of asteroids colliding with Earth. Now, NASA and the European Space Agency (ESA) are preparing for the next, crucial step in planetary defense against asteroid impacts: physically deflecting a real asteroid. Read more…

By Oliver Peckham

Cray, Fujitsu Both Bringing Fujitsu A64FX-based Supercomputers to Market in 2020

November 12, 2019

The number of top-tier HPC systems makers has shrunk due to a steady march of M&A activity, but there is increased diversity and choice of processing compon Read more…

By Tiffany Trader

  • arrow
  • Click Here for More Headlines
  • arrow
Do NOT follow this link or you will be banned from the site!
Share This