About STAC

Project Information Center

Hydrogen Education

Previous Solicitations

News

Presentations

Contact

Home

STAC

Project Information Center

STATEMENT OF WORK

Project Title:

The Use of Real Time Measurement and Artificial Intelligence to Improve Efficiency and Reduce Emissions at Coal-Fired Power Plants

Contractor:

Energy Research Company
2571-A Arthur Kill Road
Staten Island, NY 10309
(718) 608-8788

Program Area:

Sensors and Control Sciences

Partners:

Brayton Point Generating Station - U.S. Gen New England, Inc.
Lehigh University (LU-ERC)

Project Description:

The objective of the ERCo's project is to develop a technique to measure coal properties in real time, and to process the data such that coal-fired electric utility operators can adjust their operation to avoid slagging and fouling.

Management Plan

The unique nature of STAC requires that projects be supported by multiple State entities, and to the extent necessary any other entity. As indicated in the STAC Agreement, it is the Contractor’s responsibility to coordinate the execution of work under the Contract, incorporated by reference hereto. Contractor, in conjunction with the other State entities, and to the extent necessary any other entity, shall conduct the project in accordance with the Management Plan –  described below.

Administration

ERCo will assure program objectives, costs, and schedule are maintained.  ERCo's program manager will use a computerized system of Pert charts, schedules and cost summaries to keep track of program status.  ERCo will conduct a kick-off meeting and any interim meetings as needed.  Quarterly progress reports will be provided.  A final report, detailing the project's procedures, results, and recommendations will be provided.

Task 1: Laboratory Coal Measurements

ERCo will measure coal samples to determine elemental concentrations that can be used to determine slagging and fouling indices for use to control boiler operations.  LU-ERC will provide the coal samples that represent different slagging/fouling behaviors.  A dozen coal samples will be collected from power plants that experience a range of problems typical of the power industry nowadays in relation to slagging/fouling.  The samples will be selected to include problems associated to high temperature slagging due to pyrite, low ash fusion temperature due to iron and silicate compounds, and low-temperature fouling due to alkali compounds.  Eastern and Western U.S. coals will be included, as well as foreign coals more in use by the U.S. power industry (Colombian, Indonesian and Canadian coals).

Coal samples are placed in the test chamber and the laser fired onto them.  The laser, a Big Sky Laser (Bozeman, MT) model CFR-400, emits coincident UV, visible, and near infrared laser pulses from right to left in the photograph.  The visible and near infrared pulses will be directed down into the chamber by the 1064nm, 532nm laser mirror.  The f/4 lens focused the pulses onto the surface of the pressed coal to create the LIBS spark.

The light emitted from the resultant spark will be viewed by lenses on the ends of the two fiber optic cables.  One fiber optic cable is connected to the spectrometer shown in the lower right hand corner.  This spectrometer is a conventional Czerny-Turner scanning spectrometer system by Roper Scientific (Princeton, NJ).  The other fiber optic cable is connected to ERCo's broadband Echelle spectrometer by LLA (Germany).  The Echelle spectrometer is capable of viewing the spectrum from 200-780nm at once.  The Czerny-Turner spectrometer has better resolution at longer wavelengths.  In this task, the elements of interest to the indices used in Task 2 will be measured. LU-ERC will provide the coal samples and ERCo will conduct this task.

Task 2: Laboratory Slagging/Fouling Index Calculations

In this task, data obtained from the experiments in Task 1 will be analyzed.  A literature and technologic review of coal/ash-related slagging and fouling indicators will be performed.  These indicators or indices will be applied to the laboratory LIBS data to determine the appropriateness of the indices and the sensitivity of the detection method to distinguish coals with different slagging and fouling potentials.  The results of Task 2 will be used to select the indicators that will be used in full-scale tests.  Some of these indices were discussed in Section 2.4. LU-ERC will conduct this task.

Task 3: Power Plant Coal Measurements

At the conclusions of the above tasks, ERCo will bring its portable LIBS system, shown in Figure 8, to Brayton Point and will be installed  to collect coal samples which will be analyzed.  The plant will arrange for burn-tests of three coals with distinct slagging/fouling characteristics.  The coal will be processed through the LIBS system, and from the elemental concentration analysis the indices determined above will be calculated.   ERCo and Brayton Point will conduct this task.

Task 4: Power Plant Parametric Testing

Parametric tests coupled to the measurements made in Task 3 will be performed in this task.  Field tests at Brayton Point Power Plant will investigate the impact of selected coals on boiler operation and emissions, in relation to their differences in coal quality.  In this task, based on the LOFA results, the following will be adjusted: changing fuel and secondary air requirements for full load generation, changing pulverizers loading and primary air flow requirements and temperature, flue gas temperatures and draft from the furnace exit to the stack, NOx and SO2 emissions, and qualitative observations of slagging and fouling patterns.  Additional test will be performed in this task to investigate the parametric effects of boiler control settings on slagging/fouling.  These tests will cover two aspects.  The first aspect will determine the effect of boiler controllable on slagging/fouling  These parameters will be selected based on their influence on slagging/fouling and will include air, burner tilt, mill configuration and biases, and air preheating.  The second aspect will determine the impact of sootblowing on slagging/fouling.  Tests will be performed to characterize the impact of the sootblowers on the deposits and boiler performance.  The plant DCS and CEM systems will be used for monitoring during testing and for data acquisition.  This task will be done at Brayton Point and will be conducted by LU-ERC and Brayton Point.

Task 5: Power Plant Data Analysis, Models and Expert System Development

The data acquired in Tasks 3 and 4 will be analyzed in terms of filtering data, determining trends and preparing  the data for AI modeling and developing.  An artificial intelligent engine approach will be used to develop neural network-based mathematical models that relate coal quality indexes with boiler operation and slagging/fouling, an optimization algorithm, and an expert system advisor.  Different supervised and unsupervised artificial neural network architectures and training algorithms will be evaluated, including hybrid models that combine limited training information.  Input to the model will include the slagging/fouling indices chosen from Task 2.  The multi-dimensional data-driven models will be based on the parametric field tests performed in Task 4.  The models will represent functional relationships that characterize the effect of coal composition on slagging/fouling and on boiler operation and performance, as well as the impact of boiler operating conditions on the slagging/fouling situation.  An optimization algorithm that determines boiler control settings that mitigate the particular objective of reduced slagging/fouling will be developed, subject to operational and environmental constraints.  Different optimization schemes that make sure of the neural network mathematical models will be tried.  A knowledge-based expert system will be developed and tested off-line, which contains a range of slagging and fouling index calculations and a set of calculations to estimate heat rate penalties of NOx and SO2 emissions. The expert system will also contain a set of rules that will be used to make advisory recommendations on modifications to the boiler control settings to respond to variations in fuel quality that have an impact on slagging/fouling and emissions.  Finally, the knowledge-based expert system will be tested, using off-line data and an evaluation will be made of the benefit of the LIBS/AI-based approach for optimizing boiler operation for changes in fuel quality.

Task 6: Information Dissemination

NYSERDA will disseminate the results of this project to NY based coal-fired power plants.  In addition, ERCo and LU-ERC will disseminate the project's results to other coal fired plants.  LU-ERC provides project information through various programs to over 2000 engineers at more than 100 U.S. companies with coal-fired electricity generating stations.  These include:

• Lehigh Energy Update (LEU) - The LEU is published three times a year and includes articles about current research projects at Lehigh.  The articles include a summary of the project results with key findings from the research.
• Workshops - The LU-ERC hold a number of topic specific workshops annually.  The workshops are used to provide engineers with the results from current research projects and proposals for transferring the technology to operating power plants.
• Energy Liaison Program (ELP) - The ELP is a membership program for electricity generating companies and equipment suppliers to the energy industry.  LU-ERC holds an annual meeting at Lehigh to discuss current operating problems and to provide the results from current research that is ready for technology transfer.  The meeting is open to ELP member companies and companies that may be interested in the program.  The annual meeting attracts about 16 generating companies and about 30 engineers from the largest coal-fired electricity generating stations in the U.S. and Canada.

Task 7: Reporting and Program Management

ERCo will assure program objectives, costs, and schedule are maintained.  ERCo's program manager will use a computerized system of Pert charts, schedules and cost summaries to keep track of program status.  ERCo will conduct a kick-off meeting and any interim meetings as needed.  Quarterly progress reports will be provided.  A final report, detailing the project's procedures, results, and recommendations will be provided.

Project Tasks, Status, and Deliverables

Last Updated: 10/24/06