Integrated Diagnostic System for the Equipment of Power Plants: Part II — The Expert System

2002 ◽  
Author(s):  
J. Kubiak S. ◽  
G. Urquiza B. ◽  
A. Garci´a-Gutierrez
Keyword(s):  
Expert System ◽  
Expert Systems ◽  
Vibration Analysis ◽  
Power Plants ◽  
Diagnostic System ◽  
Heat Rate ◽  
Auxiliary Equipment ◽  
On Line ◽  
Knowledge Levels ◽  
Generating Equipment

This paper describes the development of an Expert System for identification of generating equipment faults caused by wearing out of their components, which decrease the efficiency and thus the heat rate of a generating plant. In a sister paper [1], the formulation was presented and the algorithms for the principal equipment were developed. The Expert Systems are based on the above algorithms. Also, in some case a vibration analysis is used jointly with thermodynamic analysis to locate precisely a fault, for example in a case of rubbing which damaged the seals of the turbine and/or compressors. The system is used off-line, however it can be installed on-line with a monitoring system. The Expert Systems identify the faults of the gas turbine, the compressor and the steam turbine. Auxiliary equipment faults are presented in the form of tables also, listing the symptoms and their causes [1]. The knowledge levels and the separate bases are built into the systems.

Monitoring and Improving Coal-Fired Power Plants Using the Input/Loss Method: Part IV

2004 ◽  
Author(s):  
Fred D. Lang
Keyword(s):  
Error Analysis ◽  
Power Plants ◽  
Heat Rate ◽  
Calorific Value ◽  
Mineral Matter ◽  
Fuel Flow ◽  
Direct Measurements ◽  
On Line ◽  
Plant Data ◽  
Loss Method

The Input/Loss Method is a unique process which allows for complete thermal understanding of a power plant through explicit determinations of fuel chemistry including fuel water and mineral matter, fuel heating (calorific) value, As-Fired fuel flow, effluent flow, boiler efficiency and system heat rate. Input consists of routine plant data and any parameter which effects system stoichiometrics, including: Stack CO2, Boiler or Stack O2, and, generally, Stack H2O. It is intended for on-line monitoring of coal-fired systems; effluent flow is not measured, plant indicated fuel flow is typically used only for comparison to the computed. The base technology of the Input/Loss Method was documented in companion ASME papers: Parts I, II and III (IJPGC 1998-Pwr-33, IJPGC 1999-Pwr-34 and IJPGC 2000-15079/CD). The Input/Loss Method is protected by US and foreign patents (1994–2004). This Part IV presents details of the Method’s ability to correct any data which effects system stoichiometrics, data obtained either by direct measurements or by assumptions, using multi-dimensional minimization techniques. This is termed the Error Analysis feature of the Input/Loss Method. Addressing errors in combustion effluent measurements is of critical importance for any practical on-line monitoring of a coal-fired unit in which fuel chemistry is being computed. It is based, in part, on an “L Factor” which has been proven to be remarkably constant for a given source of coal; and, indeed, even constant for entire Ranks. The Error Analysis feature assures that every computed fuel chemistry is the most applicable for a given set of system stoichiometrics and effluents. In addition, this paper presents comparisons of computed heating values to grab samples obtained from train deliveries. Such comparisons would not be possible without the Error Analysis.

Development of an on-line fuzzy expert system for integrated alarm processing in nuclear power plants

1995 ◽  
Vol 42 (4) ◽  
pp. 1406-1418 ◽  
Author(s):  
Seong Soo Choi ◽  
Ki Sig Kang ◽  
Han Gon Kim ◽  
Soon Heung Chang
Keyword(s):  
Expert System ◽  
Nuclear Power ◽  
Power Plants ◽  
Nuclear Power Plants ◽  
Fuzzy Expert System ◽  
On Line ◽  
Alarm Processing

The Application of On-Line Expert Systems to Supervisory Process Control

1988 ◽  
Vol 21 (6) ◽  
pp. 169-171 ◽  
Author(s):  
Howard M Foster ◽  
Simon R Meadowcroft
Keyword(s):  
Expert System ◽  
Expert Systems ◽  
Supervisory Control ◽  
System Structure ◽  
Algorithmic Problem ◽  
Potential Benefits ◽  
On Line ◽  
Supervisory Process ◽  
Improved Accuracy ◽  
Conventional Computer

The use of computers in the process industry is set to increase. However, the implementation of conventional computer technology is not adequate for executing the complex supervisory tasks. This paper speculates upon the use of on-line expert systems which have been developed for working in non-linear, non-algorithmic problem domains. Expert systems for supervisory control are explored, and the potential benefits for optimisation and improved accuracy of control are outlined. The complexity and multi-tasking nature of real-time supervisory control points to an expert system structure which is distributed. A modular expert system design is proposed and the potential benefits for the approach are evaluated.

Integrated Diagnostic System for the Equipment of Power Plants: Part I — Formulation and Algorithms

2002 ◽  
Author(s):  
J. Kubiak ◽  
A. Garci´a-Gutie´rrez ◽  
G. Urquiza ◽  
G. Gonza´lez
Keyword(s):  
Expert System ◽  
Expert Systems ◽  
Gas Turbines ◽  
Power Plants ◽  
Cooling System ◽  
Combined Cycle ◽  
Steam Turbines ◽  
Rotor Bearing ◽  
Bearing System ◽  
Plant Efficiency

The output capacity of combined cycle power plants is reduced in many cases, and sometimes forced to outages, when its main components are affected by faults, i.e., when the rotating equipment such as turbines, generators, compressors, pumps and fans suffer a failure. Normally, the overall reduction of the efficiency, and sometimes the component efficiencies, is monitored but it is difficult to identify the primary causes of the fault of the specific equipment that causes the reduction of plant efficiency. Therefore, to reduce the time of faulty operation, a precise diagnostic tool is needed. One such tool is an expert system approach, which is presented in this work. It consists of several expert systems for the identification of the faults caused by deterioration of the inner parts of the equipment, Fig. 1. Such faults not only reduce the plant efficiency but in many cases also increase the vibrations of the rotor-bearing system. Based on knowledge, the various expert systems have been constructed and their algorithms (efficiency reduction) developed for the following equipment: steam turbines, gas turbines and compressors, condenser, pumps and water cooling system. An expert system for detecting faults that increase the vibration of the rotor–bearing system is also presented. As far as the turbo compressor expert system is concerned the fault hybrid patterns previously developed were implemented and described elsewhere [1].

Application of On-Line Optimization Technology to Plant Operation

2000 ◽  
Author(s):  
Rodney R. Gay
Keyword(s):  
Power Generation ◽  
Power Plant ◽  
Gas Turbine ◽  
Power Plants ◽  
Heat Rate ◽  
Combined Cycle ◽  
Plant Performance ◽  
Plant Operation ◽  
Budget Analysis ◽  
On Line

Traditionally optimization has been thought of as a technology to set power plant controllable parameters (i.e. gas turbine power levels, duct burner fuel flows, auxiliary boiler fuel flows or bypass/letdown flows) so as to maximize plant operations. However, there are additional applications of optimizer technology that may be even more beneficial than simply finding the best control settings for current operation. Most smaller, simpler power plants (such as a single gas turbine in combined cycle operation) perceive little need for on-line optimization, but in fact could benefit significantly from the application of optimizer technology. An optimizer must contain a mathematical model of the power plant performance and of the economic revenue and cost streams associated with the plant. This model can be exercised in the “what-if” mode to supply valuable on-line information to the plant operators. The following quantities can be calculated: Target Heat Rate Correction of Current Plant Operation to Guarantee Conditions Current Power Generation Capacity (Availability) Average Cost of a Megawatt Produced Cost of Last Megawatt Cost of Process Steam Produced Cost of Last Pound of Process Steam Heat Rate Increment Due to Load Change Prediction of Future Power Generation Capability (24 Hour Prediction) Prediction of Future Fuel Consumption (24 Hour Prediction) Impact of Equipment Operational Constraints Impact of Maintenance Actions Plant Budget Analysis Comparison of Various Operational Strategies Over Time Evaluation of Plant Upgrades The paper describes examples of optimizer applications other than the on-line computation of control setting that have provided benefit to plant operators. Actual plant data will be used to illustrate the examples.

On-Line Heat Rate Monitoring as a Basis for an On-Line Turbine Diagnostic System

2005 ◽  
Author(s):  
Y. Bercovich ◽  
S. Glickman ◽  
L. Levin ◽  
A. Gordinsky ◽  
V. Belfor ◽  
...  
Keyword(s):  
High Precision ◽  
Measurement Errors ◽  
Diagnostic System ◽  
Heat Rate ◽  
High Sensitivity ◽  
Estimation Methods ◽  
Statistical Techniques ◽  
On Line ◽  
Rate Calculation ◽  
Cycle Efficiency

In this paper, a turbine on-line performance calculation system is presented. The system was implemented on a 575 MW unit of the Israel Electric Corporation and has been in operation for one year. The system was developed jointly by IEC and Berman Engineering Ltd. The main feature of the described system is the precision of the turbine heat rate calculation. This increased precision of the turbine heat rate calculation was accomplished by utilizing sophisticated statistical techniques, such as parametric and nonparametric regression, robust estimation, special filtration methods, autocorrelation methods, and uncertainty estimation methods. This high precision allows using the calculated heat rate as the main input to the turbine diagnostic system. The selection of turbine heat rate as the main diagnostic input is due to its high sensitivity to efficiency deviations of each turbine subsystem (turbine internal efficiency, condenser cleanliness, regenerative heaters’ cleanliness, etc.). However, despite this high sensitivity, the turbine heat rate cannot be used directly without implementing the sophisticated statistical techniques mentioned above because: • relatively small variation of the calculated heat rate over the entire turbine load range (only about 3%); • the presence of systematic and random measurement errors; • low signal/noise ratio as a result of the above items. In order to develop the techniques mentioned above, a detailed study of the error characteristics and error propagation was carried out. This study defined the problems which had to be solved in order to achieve an acceptably high precision of the calculation results. The current results allow using turbine heat rate as a tool for the following purposes: • turbine cycle efficiency estimation for all modes of operation and for turbine cycle scheme variations; • turbine internal condition estimation; • reliability control of measuring instrumentation which is used for turbine heat rate calculations; • determination of heat rate deviation which is above a preset acceptable value (heat rate “out of range”). The structure of the developed system is presented as well as examples of results which show the calculation precision. Also, examples are presented to illustrate how the heat rate can be using for identification of various abnormal situations which may impact the turbine cycle efficiency.

Sign in / Sign up

Export Citation Format

Share Document