skip to main content
OSTI.GOV title logo U.S. Department of Energy
Office of Scientific and Technical Information

Title: Diagnostics and Control of Natural Gas-Fired furnaces via Flame Image Analysis using Machine Vision & Artificial Intelligence Techniques

Abstract

A new approach for the detection of real-time properties of flames is used in this project to develop improved diagnostics and controls for natural gas fired furnaces. The system utilizes video images along with advanced image analysis and artificial intelligence techniques to provide virtual sensors in a stand-alone expert shell environment. One of the sensors is a flame sensor encompassing a flame detector and a flame analyzer to provide combustion status. The flame detector can identify any burner that has not fired in a multi-burner furnace. Another sensor is a 3-D temperature profiler. One important aspect of combustion control is product quality. The 3-D temperature profiler of this on-line system is intended to provide a tool for a better temperature control in a furnace to improve product quality. In summary, this on-line diagnostic and control system offers great potential for improving furnace thermal efficiency, lowering NOx and carbon monoxide emissions, and improving product quality. The system is applicable in natural gas-fired furnaces in the glass industry and reheating furnaces used in steel and forging industries.

Authors:
Publication Date:
Research Org.:
University of Missouri-Columbia
Sponsoring Org.:
USDOE Office of Energy Efficiency and Renewable Energy (EERE)
OSTI Identifier:
862201
Report Number(s):
DOE/CH/11032
TRN: US200712%%68
DOE Contract Number:
FC36-00CH11032
Resource Type:
Technical Report
Country of Publication:
United States
Language:
English
Subject:
03 NATURAL GAS; 36 MATERIALS SCIENCE; ARTIFICIAL INTELLIGENCE; BURNERS; CARBON MONOXIDE; COMBUSTION; COMBUSTION CONTROL; CONTROL SYSTEMS; DETECTION; FLAMES; FORGING; FURNACES; GLASS INDUSTRY; NATURAL GAS; ON-LINE SYSTEMS; STEELS; TEMPERATURE CONTROL; THERMAL EFFICIENCY

Citation Formats

Shahla Keyvan. Diagnostics and Control of Natural Gas-Fired furnaces via Flame Image Analysis using Machine Vision & Artificial Intelligence Techniques. United States: N. p., 2005. Web. doi:10.2172/862201.
Shahla Keyvan. Diagnostics and Control of Natural Gas-Fired furnaces via Flame Image Analysis using Machine Vision & Artificial Intelligence Techniques. United States. doi:10.2172/862201.
Shahla Keyvan. Thu . "Diagnostics and Control of Natural Gas-Fired furnaces via Flame Image Analysis using Machine Vision & Artificial Intelligence Techniques". United States. doi:10.2172/862201. https://www.osti.gov/servlets/purl/862201.
@article{osti_862201,
title = {Diagnostics and Control of Natural Gas-Fired furnaces via Flame Image Analysis using Machine Vision & Artificial Intelligence Techniques},
author = {Shahla Keyvan},
abstractNote = {A new approach for the detection of real-time properties of flames is used in this project to develop improved diagnostics and controls for natural gas fired furnaces. The system utilizes video images along with advanced image analysis and artificial intelligence techniques to provide virtual sensors in a stand-alone expert shell environment. One of the sensors is a flame sensor encompassing a flame detector and a flame analyzer to provide combustion status. The flame detector can identify any burner that has not fired in a multi-burner furnace. Another sensor is a 3-D temperature profiler. One important aspect of combustion control is product quality. The 3-D temperature profiler of this on-line system is intended to provide a tool for a better temperature control in a furnace to improve product quality. In summary, this on-line diagnostic and control system offers great potential for improving furnace thermal efficiency, lowering NOx and carbon monoxide emissions, and improving product quality. The system is applicable in natural gas-fired furnaces in the glass industry and reheating furnaces used in steel and forging industries.},
doi = {10.2172/862201},
journal = {},
number = ,
volume = ,
place = {United States},
year = {Thu Dec 01 00:00:00 EST 2005},
month = {Thu Dec 01 00:00:00 EST 2005}
}

Technical Report:

Save / Share:
  • The United States Department of Energy, Industrial Technologies Program has invested in emerging Process Design and Optimizations Technologies (PDOT) to encourage the development of new initiatives that might result in energy savings in industrial processes. Gas fired furnaces present a harsh environment, often making accurate determination of correct air/fuel ratios a challenge. Operation with the correct air/fuel ratio and especially with balanced burners in multi-burner combustion equipment can result in improved system efficiency, yielding lower operating costs and reduced emissions. Flame Image Analysis offers a way to improve individual burner performance by identifying and correcting fuel-rich burners. The anticipated benefitmore » of this technology is improved furnace thermal efficiency, and lower NOx emissions. Independent validation and verification (V&V) testing of the FIA technology was performed at Missouri Forge, Inc., in Doniphan, Missouri by Environ International Corporation (V&V contractor) and Enterprise Energy and Research (EE&R), the developer of the technology. The test site was selected by the technology developer and accepted by Environ after a meeting held at Missouri Forge. As stated in the solicitation for the V&V contractor, 'The objective of this activity is to provide independent verification and validation of the performance of this new technology when demonstrated in industrial applications. A primary goal for the V&V process will be to independently evaluate if this technology, when demonstrated in an industrial application, can be utilized to save a significant amount of the operating energy cost. The Seller will also independently evaluate the other benefits of the demonstrated technology that were previously identified by the developer, including those related to product quality, productivity, environmental impact, etc'. A test plan was provided by the technology developer and is included as an appendix to the summary report submitted by Environ (Appendix A). That plan required the V&V contractor to: (1) Establish the as-found furnace operating conditions; (2) Tune the furnace using currently available technology to establish baseline conditions; (3) Tune the furnace using the FIA technology; and (4) Document the improved performance that resulted from application of the FIA technology. It is important to note that the testing was not designed to be a competition or comparison between two different methodologies that could be used for furnace tuning. Rather, the intent was to quantify improvements in furnace performance that could not be achieved with existing technology. Therefore, the measure of success is improvement beyond the furnace efficiency obtainable using existing furnace optimization methods rather than improvement from the as found condition.« less
  • Investigation is made of various aspects of parallel computer vision, with the goal of building behaving, real-time systems that perform multi-sensory integration. A commitment is made to the idea that an intimate coupling of sensory and motor capabilities is a way to make progress on the vision problem. Behaving animals have such a coupling, and its benefits have been demonstrated analytically. A special-purpose parallel pipelined hardware was acquired for low-level vision, and have upgraded our 16-processor Butterfly Parallel Processor with faster CPUs and floating point hardware. The Butterfly is now connected with the rest of the vision hardware (the pipelinedmore » device and a fast Sun/3) via the VME bus. This year's work broadened its focus from the optimal selection of feature detectors in a Bayesian framework. The results were used to apply to a working system that applies a Markov Random Field formulation of the segmentation (objecthood recognition, figure-ground separation) problem. Further work moved in the direction of acquiring and commissioning real-time vision hardware and using the hardware in applications. Theoretical work continues to be important, and this year we made progress on the theory of principal views and convergence properties of two sorts of parallel networks: connectionist nets for recognition and Markov Random Field models of segmentation.« less
  • The Northeast Artificial Intelligence Consortium's purpose is to conduct pertinent research in artificial intelligence and to perform activities ancillary to this research. These volumes describe progress that has been made in the second year of the existence of the NAIC on the technical research tasks undertaken at the member universities. The topics covered in general are: versatile expert system for equipment maintenance, distributed AI for communications system control, automatic photo interpretation, time-oriented problem solving, speech understanding systems, knowledge base maintenance, hardware architectures for very large systems, knowledge-based reasoning and planning, and a knowledge acquisition, assistance, and explanation system. This partmore » addresses various aspects of parallel, structural, and optimal techniques in computer vision.« less
  • The Northeast Artificial Intelligence Consortium is purpose is to conduct pertinent research in artificial intelligence and to perform activities ancillary to this research. This report describes progress made in the second year of the existence of the NAIC on the technical research tasks undertaken at the member universities. The topics covered in general are: versatile expert system for equipment maintenance, distributed AI for communications-system control, automatic photointerpretation, time-oriented problem solving, speech understanding systems, knowledge-base maintenance, hardware architectures for very large systems, knowledge-based reasoning and planning, and a knowledge acquisition, assistance, and explanation system. The specific topics for this volume aremore » the use of expert systems for automated photointerpretation and other AI techniques to image segmentation and region identification.« less
  • The objective of this project was to develop and apply enabling tools and methods towards advanced combustion diagnostics and control of fired-equipment in large-scale petrochemical manufacturing. There are a number of technology gaps and opportunities for combustion optimization, including technologies involving advanced in-situ measurements, modeling, and thermal imaging. These technologies intersect most of manufacturing and energy systems within the chemical industry. This project leveraged the success of a previous DOE funded project led by Dow, where we co-developed an in-situ tunable diode laser (TDL) analyzer platform (with Analytical Specialties Inc, now owned by Yokogawa Electric Corp.). The TDL platform hasmore » been tested and proven in a number of combustion processes within Dow and outside of Dow. The primary focus of this project was on combustion diagnostics and control applied towards furnaces, fired heaters and boilers. Special emphasis was placed on the development and application of in-situ measurements for O2, CO and methane since these combustion gases are key variables in optimizing and controlling combustion processes safely. Current best practice in the industry relies on measurements that suffer from serious performance gaps such as limited sampling volume (point measurements), poor precision and accuracy, and poor reliability. Phase I of the project addressed these gaps by adding improved measurement capabilities such as CO and methane (ppm analysis at combustion zone temperatures) as well as improved optics to maintain alignment over path lengths up to 30 meters. Proof-of-concept was demonstrated on a modern olefins furnace located at Dow Chemical's facility in Freeport TX where the improved measurements were compared side-by-side to accepted best practice techniques (zirconium oxide and catalytic bead or thick film sensors). After developing and installing the improved combustion measurements (O2, CO, and methane), we also demonstrated the ability to improve control of an olefins furnace (via CO-trim) that resulted in significant energy savings and lower emissions such as NOx and other greenhouse gases. The cost to retrofit measurements on an existing olefins furnace was found to be very attractive, with an estimated payback achieved in 4 months or less.« less