By Douglas J. Smith IEng.,
As the electric utility industry transitions from a regulated to a non-regulated competitive market, improving and maintaining the efficiencies of electric generating plants is critical. A unit’s efficiency, or heat rate, is a function of the plant’s design, its size, the type of fuel used and the quality of its operations and maintenance. In 1998, EL&P, a sister publication of Power Engineering, reported that the average heat rate for the 50 best performing coal-fired plants ranged from 9,382 Btu/kWh to 10,146 Btu/kWh. Although heat rates have been declining in recent years, there is still room for improvement.
Over the years power plants have installed a variety of continuous monitoring, diagnostic and performance monitoring systems on a variety of power plant equipment. Today, many electric utilities have integrated data from these systems into real-time heat rate monitoring systems.
Operators and engineers can use this information to optimize a plant’s efficiency and heat rate. However, continuous on-line heat rate monitoring systems cannot make up for the loss in heat rate due to equipment design, poor quality fuel and/or poor maintenance.
While there are many tools available to help power plants optimize heat rate and fuel consumption, it is essential that an assessment of the plant’s performance be first carried out. Assessing a plant’s performance provides a comprehensive picture of actions that can be taken to optimize heat rate. It also identifies the strengths and weaknesses of plant operations.
Control room heat rate interface terminal. Photo: Courtesy of NeuCo, Inc.
Plant performance assessments offered by EPRI provide a cost effective approach for determining current plant performance. EPRI uses this information to establish practical performance goals for optimizing a plant’s heat rate.
The first priority before initiating a performance assessment is to have a meeting with representatives of all departments that are involved in the efficient operation of the plant. Attendees should include plant management, operations, maintenance and performance engineers. According to EPRI, it is essential that the performance assessment team work closely with the plant’s staff to ensure their complete support of any actions recommended. For the program to be a success it is important that everyone at the plant level be involved.
Although there are different facets in assessing the performance and heat rate of a plant, the following must first be determined:
- What is the current mechanical condition of the plant’s equipment?
- Is heat rate a criterion in making operating decisions?
- What performance monitoring tools and methods are utilized to monitor and improve heat rate?
- What is the level of heat rate awareness among the plant’s staff and is any specific heat rate training provided?
- How can all of the information currently available be utilized to improve heat rate?
- What is the optimum heat rate (benchmarking) for the particular plant?
According to Duane Hill, manager performance administration, Dairyland Cooperative, their operating staff has been involved in the selection and application of the heat rate optimization programs at all four of their power plants. Because the operators were involved from the beginning they have accepted the heat rate optimizing programs without reservations. At Dairyland, the operators are able to take immediate action to rectify any operating problems that directly affect heat rate. If the problem is equipment related, they work with the engineering and maintenance staff to rectify the problem.
Heat Rate Improved 1.5 Percent
EPRI’s Plant Performance Assessment program has successfully been applied by Hoosier Energy at their 1,000 MW coal-fired Merom generating station. The plant is located in Sullivan County, Ind. The Merom assessment project was a plant-wide analysis of the work culture and practices that existed at the plant. Before proceeding with the full assessment, EPRI documented and verified the existing plant’s heat rate, established a benchmark for the best achievable heat rate and performed an assessment of the plant’s instruments.
EPRI recommended a number of improvements to improve operations and heat rate at Merom. They also recommended heat rate awareness training, formation of a heat rate improvement task force, instrument upgrade and the installation of an on-line performance monitoring system. One of the first actions by the plant was the organization of eight heat rate awareness seminars. As a result of the seminars, more work requests for heat rate improvements were generated than in the past.
At Merom, data was available from a variety of monitoring systems. Unfortunately, because the reliability of data depended upon the accuracy of the instruments already installed, the data was not always reliable. As a result, EPRI recommended that a number of key instruments be upgraded including feed-water flowmeters and various temperature and pressure sensors.
With improved monitoring, management and awareness training, Merom has reduced in one year the plant’s heat rate to its lowest level since the plant went into operation in the early 1980s. According to EPRI, the plant’s heat rate has been improved by 1.5 percent (Figure 1) and the net capacity factor by 5.2 percent. Cost has also been reduced by 2.3 percent.
Santee Cooper Heat Rate Program
In 1993 Santee Cooper began a program to produce weekly unit heat rate curves from on-line performance monitoring systems at seven coal-fired power plants. The goal of the program was to use the information to dispatch the units. Approximately every seven seconds real-time data is received from different monitoring systems within the plants. Using this data the heat rate and controllable losses are calculated and displayed on computer terminals at the company’s corporate office.
A separate program, developed to save data from the performance monitoring system every five minutes, calculates the heat rate for the previous week. However, in developing the system the engineers had to design it to eliminate artificial values. The artificial values were required to keep the existing performance monitoring system running in the event of a field instrument failure. This required the establishment of acceptable limits for the normal operating range of each instrument. If any of the instruments used in calculating heat rate indicate a reading outside of the normal range, or differ from its duplicate by more than a specified amount, the program skips to the next record.
Prior to using data from the on-line performance monitoring system, three performance engineers were needed to calculate and produce a heat rate curve. Initially this was done eight times per year but was eventually improved to 14 curves per year. With the new operating heat rate program, two engineers require one week to produce seven heat rate curves. Today, after further improvements to the system, one engineer is able to produce seven heat rate curves in two hours.
Curves for Dispatch
At the end of each quarter Santee Cooper averages the weekly heat rates and develops a regression equation that is used to produce a smooth input/output curve. The SCADA system uses this information to dispatch the units. Regular dispatch updates ensure that any equipment problems and improvements in a unit’s performance are taken into account.
As a check on the validity of the curves produced, the curves are compared with the station’s monthly generation reports and fuel use. Using all of the available data for the month, including the coal’s heating value, Santee Cooper calculates the heat rate. Using the average hourly net generation, and the calculated heat rate, the input value for the boiler is calculated for each hour. These values are then totaled and divided by the total net generation. The calculated value is compared against the station’s heat rate that has been determined by the amount of coal used and the recorded net generation.
According to Santee Cooper, the biggest benefit is the calculation of heat rate on a weekly basis. This now forces the plant’s management to look at unit performance weekly. Prior to the seven units being fitted with the heat rate performance monitoring systems the units were dispatched based on data collected during heat rate tests done in the late 1980s. According to the Santee Cooper Energy Control Center, dispatching the units using the new curves saves $1,400/day during a typical summer load.
VAX to NT
Dairyland Power Cooperative is located in Western Wisconsin. Four of their coal-fired units, 60 MW and 90 MW units at the Alma Station, a 360 MW at John P. Madgett Station and a 360 MW unit at the Genoa Station all have Scientech PMAX performance monitoring systems. Originally these systems were VAX based but were converted to NT based platforms in the late 1990s.
Data collected at the plants can be accessed at all four stations and also at the corporate office. All four plants are load following units and are within a 50-mile radius of the corporate office. In order to develop accurate heat rate curves the units are tested at several unit loads. At the John P. Madgett and Genoa plants there is a total of eight test loads for each unit and six for each of the Alma units.
The PMAX systems collect data whenever the units are at stable conditions. A stable condition is achieved when the unit has reached its desired load and thermodynamic equilibrium.
Two sequences per unit are required to conduct the automatic testing of the data. The first sequence determines the stability of the conditions and calculates the test values and the second sequence stores the test values in the archive file. Raw data is averaged every five minutes. The PMAX system then compares the actual MW output with the test values. Every hour the test data is moved to a second archive file.
Analysis of the data to determine unit performance is carried out by Scientech’s PEPSE program. This program performs an integrated heat balance of the plant. Spreadsheet templates developed by Dairyland Power Cooperative evaluate the test data prior to submitting it to the PEPSE program for analysis.
Because of the need to determine the reliability of the data supplied by the sensors, an Advanced Calibration Monitoring (ACM) system was installed in 1998 at the John P. Madgett Station. The ACM system compares how one piece of data is performing compared to all of the other inputs that are modeled in the system. Although the ACM system was initially intended to reduce the number of periodic instrument calibrations it is also used to check the reliability of the data prior to calculating the heat rate.
Using the real-time monitoring systems Dairyland Energy Cooperative is now able to determine performance degradation of their units more quickly. Using this information the utility is able to optimize a plant’s heat rate. Figures 2 and 3 are typical screens used by Dairyland’s operators at the John P. Madgett Station. Over the last 10 years the heat rate at the John P. Madgett station has improved by 6-7 percent and at the other plants by 4 percent, says Duane Hill, manager, performance administration, Dairyland Power Cooperative.
TVA On-line Heat Rate Study
Tennessee Valley Authority (TVA) has recently completed a study of on-line performance monitoring/heat rate software packages. The study included ABB, Babcock & Wilcox, Bailey, Black & Veatch/EPRI, Encoteck, Exergetic Systems, EPRI, Foxboro, GE Enter, General Physics, Honeywell, Power Technologies, Scientech, SE-ACE Innovations and Westinghouse. The intent of the study was to assess the use of existing on-line monitoring systems for possible application to TVA’s fossil fired power plants.
Any performance monitoring system chosen by TVA must interface with the utility’s existing data acquisition (DAS) or distributed control systems (DCS). The company supplying the monitoring systems must also supply the interface. Although many suppliers have interfaces available, some companies must use other vendor interfaces.
Some of TVA’s DCS systems have access to a historian for archiving the plant’s performance data. As a result they have no need for a separate data archiving system. However, units fitted with DAS would require a historian to archive the data. According to TVA, performance monitoring systems supplied by ABB, Black & Veatch, GE Enter, Scientech and General Physics can provide data archiving systems without the need for special stand-alone historians. TVA’s study determined that the cost of a stand-alone historian would increase the cost of a performance monitoring system significantly.
A less costly stand-alone option is available from Honeywell. The Honeywell system can collect data for up to four units. Scientech’s PMAX and EPRI’s PMW monitoring systems have their own embedded historians and are the most cost-effective for the existing DCS and DAS.
For ease of operation TVA states that the performance monitor server platform should be a Windows NT operating system. Since Windows operating systems are familiar to the engineers and operators they can easily make configuration changes. A simple graphic interface is also important as it allows engineering and operations personnel to quickly make changes to the system. Good graphics that show the status of the unit with the cost related to off target performance is also a requirement.
An integral part of any performance monitoring package is the archiving of data. After all of the data has been collected and analyzed it is essential that it be stored for later access and trending. TVA surveyed three methods for storing data:
- A stand-alone data archiving system-the-most expensive.
- OSI’s PI embedded system-less expensive than a stand-alone system.
- Performance monitoring systems with their own embedded archiving system-the best and least expensive method.
In general, configuration of the system can be done by the company supplying the software. Many companies also offer training for a utility’s personnel so that they can configure the systems in-house. Since configuration is the largest cost, and TVA has the engineers and technical staff who can do the work, the company will perform system configuration themselves, says Robert Frank, Director of EPRI’s on site I&C Center.