The goal is to help to enhance the operational flexibility of the nation's natural gas storage system. The objectives of this project are to test ultrasonic meters for storage applications and to develop and test a prototype for a low-cost, inferential, energy flow rate meter for use within the natural gas storage infrastructure.

Southwest Research Institute – project management and research product

Location:
San Antonio, Texas 78238

Natural gas is priced and sold based on the energy delivered to the customer. The value of natural gas depends on the accurate determination of energy flow rate. Historically, the U.S. natural gas industry has determined energy flow rate using independent measurements of flow rate (rate of delivery) and heating value (combustible energy). In order to obtain the heating value of a gas, gas chromatography has traditionally been used to generate a composition assay from samples of the gas. This technology (installation and operational costs) and the sample-taking process required to support gas chromatographs, is often too cost-prohibitive to be used on a wide scale.

Energy flow rate measurement is critical to many industry aspects that rely on energy content (whether high or low) as a quality determination factor. Some suppliers (such as those with low-pressure Appalachian gas) provide richer gas than the rest of the country. If energy content is measured correctly, the richer portion can be stripped of its heavier hydrocarbons and sold separately without impacting the quality of the normal pipeline gas, thus providing a product with higher profitability for the suppliers. Most large end-users utilize gas for fuel and care about energy because a few percent change in heating value can have a large economic impact. Lower heating values mean more gas volume must be purchased so it is essential that energy rates be measured correctly. Finally, variations greater than 50 Btu/scf can adversely affect burner efficiencies in furnaces and engines, resulting in reduced operational efficiency.

In 1998, Southwest Research Institute® (SwRI®) performed an assessment of natural gas technology and a feasibility evaluation for the U.S. Department of Energy (DOE). The study aimed to compare traditional and alternative technologies for energy flow rate measurement in terms of their accuracy, capital investment, and “operational and maintenance” costs. The existing technology used by the industry is technically sound and fulfills accuracy requirements, but its inherent costs are difficult to justify in most cases. As an alternative to gas chromatographs, SwRI® investigated use of an inferential approach to energy measurement developed by Behring et al. [1998]. Behring et al. found that flow and energy measurement properties may be determined with just a few inferential measurements that characterize the natural gas composition without a full composition analysis.

The SwRI® study of 1998 determined that this alternative inferential approach to energy measurement was feasible. Heating values and densities may be calculated by sensing the speed of sound and the N2 and CO2 concentrations at a known temperature and pressure in a sample and then applying a gross inferential correlation equation. The correlation is based on a database of 102 different natural gas compositions (987-1150 Btu/scf and 83.4-98.3 mole percent methane). This database represents essentially a full practical range of natural gas mixtures under gas quality tariff authority. The inferential approach uses a cubic-spline fit to adjust the pressure and temperature of the reference database to sample conditions and a regression equation to predict molecular weight and heating value (based on the AGA-8 Gross Characterization Method (American Gas Association [1994]).

Energy measurement and metering errors in the United States, although very small (less than 0.5 percent), are very costly because of the huge quantities of natural gas distributed. Estimates of the cost of incorrect energy flow measurements, which gets passed through to consumers in tariffs, are around $300 million annually. Currently, natural gas flow rate is measured by orifice, turbine, and ultrasonic meters, whereas energy content is determined by a gas chromatograph. These two measurements are combined to provide energy flow. A gas chromatograph is very expensive to install and maintain. The new retrofit module developed by SwRI® is projected to sell at about one-fifth the cost of the gas chromatograph, but will have the same accuracy. Anticipated market segments include natural gas transmission and distribution, production and gathering, electric power generation, and large industrial users. The new retrofit module would make energy measurement more affordable at locations within the natural gas system where it is currently being carried out and justifiable at locations where economics currently can only justify spot (monthly) sampling. Estimates of the savings to consumers, assuming a 10 percent market penetration, would be around $30 million annually. The retrofit module can also determine other gas properties needed by the gas industry, opening up other avenues for commercialization vendors.

• Completed ultrasonic meter testing and evaluation,
• Completed assessment of energy meter requirements, and
• Successfully designed and tested a low-cost, inferential, natural gas energy flow rate meter prototype.

The first phase of this project included extensive ultrasonic meter testing and evaluation. SwRI® obtained both single-path and multi-path eight-inch ultrasonic meters from manufacturers and conducted a number of tests to determine baseline accuracy, rangeability, and repeatability. The compilation of laboratory test results suggests that ultrasonic flow meters can be a viable alternative to conventional measurement methods. When properly installed, operated, and maintained, ultrasonic meters can measure bi-directional flow rate to a level of accuracy comparable to traditional measurement methods, such as orifice or turbine flow meters. The ultrasonic meter has the potential to significantly reduce gas measurement costs at storage facilities since: (1) the number of meters and valves is reduced (as compared to orifice or turbine meter installations) because of its bi-directional capability and rangeability, and (2) maintenance costs should be substantially lower because there are fewer moving parts. The research included testing at the Meter Research Facility at Southwest Research Institute, and data was also provided by several gas transmission pipeline companies in the U.S. that performed field evaluations of ultrasonic meters for the report.

The second phase was to illustrate the technological feasibility of the inference approach through development of data correlations that could relate energy measurement properties (molecular weight, mass-based heating value, standard density, molar ideal gross heating value, standard volumetric heating value, density, and volume-based heating value) to three inferential properties: standard sound speed, carbon dioxide concentration, and nitrogen concentration (temperature and pressure are also required for the last two). The key advantage of this approach is that inexpensive on-line sensors may be used to measure the inferential variables, which can then be applied (through data correlations) to retrofit existing flow meters (ultrasonic, orifice, turbine, rotary, Coriolis, diaphragm, etc.) for on-line energy flow rate measurement. The practical issues for field development were evaluated using two transducers extracted from a $100 ultrasonic domestic gas meter and a $400 infrared sensor.

Following the natural gas energy meter assessment, development of the low cost inferential natural gas energy flow rate prototype retrofit module was carried out. This development included: (1) extension of the range of inferential property correlations, (2) evaluation of alternative sensors, (3) refinement of sensor performance, (4) design of the working prototype, and (5) construction of the working prototype. When these steps were fully completed, the natural gas energy flow rate prototype was tested in New Braunfels, Texas. The module performed well, with sensors and associated electronics functioning properly when exposed to field conditions. The trends in heating value and carbon dioxide from the module and the reference gas chromatograph were qualitatively similar, suggesting the module tracked the true heating value of the gas.

All work is complete and a final report produced. A Joint Industry Project (JIP) of eight companies, including several vendors interested in potential commercialization of the energy meter, has been established.

$1,085,473

$570,000

NETL – Anthony Zammerilli (304-285-4641 or Anthony.zammerilli@netl.doe.gov) 
SwRI – Eric Kelner (ekelner@swri.org or 210-522-3309)

Final Report - May 2005: Development of a Low Cost Inferential Natural Gas Energy Flow Rate Prototype Retrofit Module [PDF-4094KB]