A mathematical approach: RMU professors built predictive modeling tools to help place natural gas stations

It’s a moment of panic for drivers: You suddenly notice the warning light on your fuel gauge, but you don’t know when it came on or how close you are to the nearest gas station.

Now, imagine you drive a compressed or liquefied natural gas-powered vehicle. The Pittsburgh region, for example, has only five natural gas fueling stations. That scarcity is a barrier to more people adopting cleaner, more fuel-efficient natural gas vehicles that could also reduce dependence on foreign oil.

Three professors at Robert Morris University have developed a mathematical model that determines the optimal locations for natural gas fueling stations in Pittsburgh, based on existing traffic flow and traffic density. The paper, authored by Tony Kerzmann, assistant professor of mechanical engineering; Gavin Buxton, associate professor of physics; and Jonathan Preisser, assistant professor of mathematics, predicts the optimal locations for up to 128 fueling stations in Pittsburgh. The paper is being published in the journal Sustainable Energy Technology and Assessments.

A step toward energy independence

According to the U.S. Department of Energy’s Alternative Fuels Data Center, 14.8 million natural gas vehicles operate worldwide but only 112,000, including buses and trucks, operate in the United States.

Yet as the RMU professors note in their paper, while the U.S. imports 45 percent of its total petroleum consumption, the nation produces nearly 90 percent of the natural gas it consumes.

“A transportation sector dominated by natural gas vehicles would provide a huge step toward energy independence, but this is not the only advantage of natural gas vehicles,” the authors write. “Natural gas vehicles significantly lower carbon monoxide, nitrogen oxide, non-methane hydrocarbon, particulate matter and greenhouse gas emissions.”

Taking a look at the model

As a first pass, the authors considered the total vehicle miles traveled through each location as the numerical variable with which to optimize the distribution of natural gas fueling stations. The fueling stations are treated in the model as wandering around, trying to find the regions with the higher numerical variable.

The computer model finds the optimum locations for a large number of natural gas fueling stations simultaneously, while penalizing the overlapping of natural fueling station locations — fueling stations that are close enough to one another that potential customers will be divided, and use both.

However, the numerical variable used to optimize the distribution of natural gas fueling stations is at present too crude. The authors, therefore, are looking at the socioeconomic factors that might make a given location a sound investment for placing a natural gas fueling station.

“Say you are BP and you wanted to change some of your existing gas stations to supply natural gas. Our computer program could tell you where to distribute the natural gas stations that would make the most sense, that would increase the likelihood of customers switching over to natural gas vehicles,” Buxton says.

Socioeconomic factors

The typical customer of a natural gas-powered vehicle might prefer to see natural gas fueling stations near his or her neighborhood, where he or she is likely to be refueling his or her vehicle, than in areas of high vehicle traffic. Therefore, while vehicle miles traveled through a given location may be an important variable, it is not the only variable to consider.

The number of residents in a given neighborhood might influence the decision of where to place a natural gas fueling station, along with the average household income — with customers from more affluent neighborhoods being more likely to purchase a new vehicle.

Furthermore, the typical political persuasion within a neighborhood might also influence the customers’ likelihood to purchase a natural gas-powered vehicle. For example, might it be possible to consider the type of customers that are more likely to buy a more environmentally-friendly vehicle, or a vehicle that reduces our dependence on foreign fuel?

The presence of existing gasoline fueling stations that could be converted to also provide natural gas refueling capabilities would be another influencing factor. For example, currently no gas stations are located in downtown Pittsburgh, making the Golden Triangle an unlikely destination for natural gas fueling stations.

However, it’s important to remember that ultimately a robust network of natural gas fueling stations could ease the transition to even cleaner hydrogen fuels. The infrastructure that we invest in now to transition to natural gas-powered vehicles is the same infrastructure required to store and transport hydrogen for hydrogen-fueled vehicles.

 

Natural gas basics

Natural gas is an odorless, nontoxic, gaseous mixture of hydrocarbons — predominantly methane. Because of the gaseous nature of this fuel, when stored onboard a vehicle, it must be in either a compressed gaseous (CNG) or liquefied (LNG) state. Both CNG and LNG are clean burning, domestically produced, relatively low priced and widely available for fuel vehicles.

  There are three types of natural gas vehicles:

  • Dedicated vehicles are designed to run only on natural gas.
  • Bi-fuel vehicles have two separate fueling systems that enable them to run on either natural gas or gasoline.
  • Dual-fuel vehicles, traditionally limited to heavy-duty applications, have fuel systems that run on natural gas and use diesel fuel for ignition assistance

  Natural gas accounts for about a quarter of the energy used in the United States.

  About one-third goes to residential and commercial uses, such as heating and cooking; one-third to industrial uses; and one-third to electric power production.

  Only about one-tenth of 1 percent is used for transportation fuel.

Source: Alternative Fuels Data Center, part of the U.S. Department of Energy’s Clean Cities program

How to reach: Robert Morris University, Jonathan Potts, senior director, public relations, [email protected] or (412) 397-5291. For more information, visit www.rmu.edu.