Fuzzy Expert Systems for the Diagnosis of Component and Sensor Faults in Complex Energy Systems

Andrea Toffolo

doi:10.1115/1.4000175

Fuzzy Expert Systems for the Diagnosis of Component and Sensor Faults in Complex Energy Systems

Journal of Energy Resources Technology ◽

10.1115/1.4000175 ◽

2009 ◽

Vol 131 (4) ◽

Cited By ~ 2

Author(s):

Andrea Toffolo

Keyword(s):

Expert Systems ◽

Numerical Models ◽

Energy Systems ◽

Combined Cycle ◽

Sensor Data ◽

Complex Energy ◽

Fuzzy Expert Systems ◽

Mass Flow Rates ◽

Expected Values ◽

Fault Mode

Locating the causes of malfunctions in complex energy systems is an extremely difficult task, since more than one fault mode may produce similar and possibly undistinguishable patterns of effects. This paper shows how fuzzy expert systems can exploit the available measurements from the data acquisition system to identify different component and sensor fault modes. Real sensor data (mass flow rates, pressures, temperatures, and key operating parameters) are compared with the expected values of the same quantities that are calculated using numerical models of local subsystems. This comparison simply determines if the differences between measured and expected values are “negative,” “zero,” or “positive” in fuzzy logic terms. The final objective is to verify the existence of some patterns of these attributes that univocally identify the considered fault modes. These patterns are then implemented as the set of rules forming the knowledge base of a fuzzy expert system. The proposed diagnostic methodology is tested on the gas section of a real combined-cycle cogeneration plant, and the effect of measurement noise is also discussed.

Download Full-text

Fuzzy Expert Systems for the Diagnosis of Component and Sensor Faults in Complex Energy Systems

Volume 6: Energy Systems: Analysis, Thermodynamics and Sustainability ◽

10.1115/imece2007-41818 ◽

2007 ◽

Author(s):

Andrea Toffolo

Keyword(s):

Expert Systems ◽

Numerical Models ◽

Energy Systems ◽

Combined Cycle ◽

Sensor Data ◽

Complex Energy ◽

Fuzzy Expert Systems ◽

Mass Flow Rates ◽

Expected Values ◽

Fault Mode

Locating the causes of malfunctions in complex energy systems is an extremely difficult task, since more than one fault mode may produce similar and possibly undistinguishable patterns of effects. This paper shows how fuzzy expert systems can exploit the available measurements from the data acquisition system to identify different component and sensor fault modes. Real sensor data (mass flow rates, pressures, temperatures, and key operating parameters) are compared to the expected values of the same quantities that are calculated using numerical models of local subsystems. This comparison simply determines if the differences between measured and expected values are “negative”, “zero” or “positive” in fuzzy logic terms. The final objective is to verify the existence of some patterns of these attributes that univocally identify the considered fault modes. These patterns are then implemented as the set of rules forming the knowledge base of a fuzzy expert system. The proposed diagnostic methodology is tested on the gas section of a real combined-cycle cogeneration plant and the effect of measurement noise is also discussed.

Download Full-text

Fault Diagnosis with Fuzzy Expert System

Applied Mechanics and Materials ◽

10.4028/www.scientific.net/amm.48-49.519 ◽

2011 ◽

Vol 48-49 ◽

pp. 519-522 ◽

Cited By ~ 1

Author(s):

Yang Lan Ou

Keyword(s):

Expert System ◽

Numerical Models ◽

Energy Systems ◽

Sensor Data ◽

Fuzzy Expert System ◽

Sensor Fault ◽

Knowledge Based ◽

Fuzzy Expert Systems ◽

Mass Flow Rates ◽

Fault Mode

Locating the causes of malfunctions in complex energy systems is an extremely difficult task, more than one fault mode may produce similar and possibly undistinguishable patterns of effects. This paper shows how fuzzy expert systems can exploit the available measurements from the data acquisition system to identify different component and sensor fault modes. Real sensor data (mass flow rates, pressures, temperatures, and key operating parameters) are compared with the expected values of the same quantities that are calculated using numerical models of local subsystems. The final objective is to verify the existence of some patterns of these attributes that univocally identify the considered fault modes. These patterns are then implemented as the set of rules forming the knowledge based on fuzzy expert system.

Download Full-text

Thermoeconomic Diagnosis: Zooming Strategy Applied to Highly Complex Energy Systems: Part 1 — Detection and Localization of Anomalies

Advanced Energy Systems ◽

10.1115/imece2002-33183 ◽

2002 ◽

Cited By ~ 5

Author(s):

Vittorio Verda ◽

Luis Serra ◽

Antonio Valero

Keyword(s):

Inverse Problem ◽

Complex Systems ◽

Reference Conditions ◽

Energy Systems ◽

Combined Cycle ◽

Operating Conditions ◽

The Other ◽

Complex Energy ◽

Operational Diagnosis ◽

Detection And Localization

This paper presents a summary of our most recent advances in Thermoeconomic Diagnosis, developed during the last three years [1–3], and how they can be integrated in a zooming strategy oriented towards the operational diagnosis of complex systems. In fact, this paper can be considered a continuation of the work presented at the International Conference ECOS’99 [4–6] in which the concepts of malfunction (intrinsic and induced) and dysfunction [7] were analyzed in detail. These concepts greatly facilitate and simplify the analysis, the understanding and the quantification of how the presence of an anomaly, or malfunction, affects the behavior of the other plant devices and of the whole system. However, what remains unresolved is the so-called inverse problem of diagnosing [3], i.e. given two states of the plant (actual and reference operating conditions), find the causes of deviation of the actual conditions with respect to the reference conditions. The present paper tackles this problem and describes significant advances in addressing how to locate the actual causes of malfunctions, based on the application of procedures for filtering induced effects that hide the real causes of degradation. In this paper a progressive zooming thermoeconomic diagnosis procedure, which allows one to concentrate the analysis in an ever more specific zone is described and applied to a combined cycle. In an accompanying paper (part 2 [8]) the accuracy of the diagnosis results is discussed, depending on choice of the thermoeconomic model.

Download Full-text

Thermoeconomic Diagnosis: Zooming Strategy Applied to Highly Complex Energy Systems. Part 1: Detection and Localization of Anomalies*

Journal of Energy Resources Technology ◽

10.1115/1.1819315 ◽

2005 ◽

Vol 127 (1) ◽

pp. 42-49 ◽

Cited By ~ 14

Author(s):

Vittorio Verda ◽

Luis Serra ◽

Antonio Valero

Keyword(s):

Inverse Problem ◽

Complex Systems ◽

Reference Conditions ◽

Energy Systems ◽

Combined Cycle ◽

Operating Conditions ◽

The Other ◽

Complex Energy ◽

Operational Diagnosis ◽

Detection And Localization

This paper presents a summary of our most recent advances in Thermoeconomic Diagnosis, developed during the last three years, and how they can be integrated in a zooming strategy oriented toward the operational diagnosis of complex systems. In fact, this paper can be considered a continuation of the work presented at the International Conference ECOS’99 in which the concepts of malfunction (intrinsic and induced) and dysfunction were analyzed in detail. These concepts greatly facilitate and simplify the analysis, the understanding, and the quantification of how the presence of an anomaly, or malfunction, affects the behavior of the other plant devices and of the whole system. However, what remains unresolved is the so-called inverse problem of diagnosing, i.e., given two states of the plant (actual and reference operating conditions), find the causes of deviation of the actual conditions with respect to the reference conditions. The present paper tackles this problem and describes significant advances in addressing how to locate the actual causes of malfunctions, based on the application of procedures for filtering induced effects that hide the real causes of degradation. In this paper a progressive zooming thermoeconomic diagnosis procedure, which allows one to concentrate the analysis in an ever more specific zone is described and applied to a combined cycle. In an accompanying paper the accuracy of the diagnosis results is discussed, depending on choice of the thermoeconomic model.

Download Full-text

A Second-Law-Based Optimization: Part 1—Methodology

Journal of Engineering for Gas Turbines and Power ◽

10.1115/1.2816981 ◽

1996 ◽

Vol 118 (4) ◽

pp. 693-697 ◽

Cited By ~ 18

Author(s):

Y. M. El-Sayed

Keyword(s):

Objective Function ◽

Energy Systems ◽

Second Law Of Thermodynamics ◽

Energy System ◽

Combined Cycle ◽

Second Law ◽

Complex Energy ◽

System A ◽

Decision Variables ◽

Further Development

This paper deals with the optimization of complex energy systems given a cost objective function. The optimization uses a decomposition strategy based on the second law of thermodynamics and a concept for costing the components of a system. A large number of nonlinear decision variables can be optimized with enhanced convergence to an optimum. The paper is in two parts. In this part, the methodology is described. In Part 2, the methodology is applied to a simple energy system of 10 components and 19 manipulated decision parameters. The system is treated once as a single purpose combined cycle and once as a power-heat cogenerating system. The results of the application are summarized and evaluated. Further development is encouraged.

Download Full-text

An Application of Fuzzy Expert Systems to Strategic Investments: The Case of Florim S.P.A.

SSRN Electronic Journal ◽

10.2139/ssrn.1473008 ◽

2001 ◽

Cited By ~ 1

Author(s):

Gisella Facchinetti ◽

Carlo Alberto Magni ◽

Giovanni Mastroleo ◽

Marina Vignola

Keyword(s):

Expert Systems ◽

Strategic Investments ◽

Fuzzy Expert Systems

Download Full-text

Integrating multiple fuzzy expert systems under restricting requirements

Agricultural Economics (Zemědělská ekonomika) ◽

10.17221/5015-agricecon ◽

2012 ◽

Vol 52 (No. 4) ◽

pp. 187-196

Author(s):

S. Aly ◽

I. Vrana

Keyword(s):

Decision Making ◽

Expert Systems ◽

Business Processes ◽

New Product ◽

Decision Problems ◽

Food Chains ◽

Problem Context ◽

Multiple Factors ◽

Decision Consistency ◽

Fuzzy Expert Systems

The multiple, different and specific expertises are often needed in making YES-or-NO (YES/NO) decisions for treating a variety of business, economic, and agricultural decision problems. This is due to the nature of such problems in which decisions are influenced by multiple factors, and accordingly multiple corresponding expertises are required. Fuzzy expert systems (FESs) are widely used to model expertise due to its capability to model real world values which are not always exact, but frequently vague, or uncertain. In addition, they are able to incorporate qualitative factors. The problem of integrating multiple fuzzy expert systems involves several independent and autonomous fuzzy expert systems arranged synergistically to suit a varying problem context. Every expert system participates in judging the problem based on a predefined match between problem context and the required specific expertises. In this research, multiple FESs are integrated through combining their crisp numerical outputs, which reflect the degree of bias to the Yes/No subjective answers. The reasons for independency can be related to maintainability, decision responsibility, analyzability, knowledge cohesion and modularity, context flexibility, sensitivity of aggregate knowledge, decision consistency, etc. This article presents simple algorithms to integrate multiple parallel FES under specific requirements: preserving the extreme crisp output values, providing for null or non-participating expertises, and considering decision-related expert systems, which are true requirements of a currently held project. The presented results provides a theoretical framework, which can bring advantage to decision making is many disciplines, as e.g. new product launching decision, food quality tracking, monitoring of suspicious deviation of the business processes from the standard performance, tax and customs declaration issues, control and logistic of food chains/networks, etc. 

Download Full-text