R. Lakner

Multiagent realization of prediction-based diagnosis and loss prevention

A multiagent diagnostic system implemented in a Protege-JADE-JESS environment interfaced with a dynamic simulator and database services is described in this paper. The proposed system architecture enables the use of a combination of diagnostic methods from heterogeneous knowledge sources. The process ontology and the process agents are designed based on the structure of the process system, while the diagnostic agents implement the applied diagnostic methods. A specific completeness coordinator agent is implemented to coordinate the diagnostic agents based on different methods. The system is demonstrated on a case study for diagnosis of faults in a granulation process based on HAZOP and FMEA analysis.

An assumption-driven case-specific model editor

A case sensitive intelligent model editor has been developed for constructing consistent lumped dynamic process models and for simplifying them using modelling assumptions. The approach is based on a systematic assumption-driven modelling procedure and on the syntax and semantics of process,models and the simplifying assumptions.

Fault diagnosis based on hazard identification results

An intelligent diagnostic methodology is proposed in this paper that combines piping and instrumentation information with hazard analysis results. The process model is represented as a hierarchically ordered set of blended HAZOP and FMEA structures.

A Procedure Ontology for Advanced Diagnosis of Process Systems

An ontology for representing operation, safety and control procedures is proposed in this paper that supports diagnosis based on following these procedures and combining observed malfunctions with Failure Mode and Effects Analysis (FMEA) information. The procedure ontology is defined within interconnected components of the process plant, diagnostic analysis (where the FMEA is described) and procedures. The proposed method is illustrated on a simple operating procedure.

A procedure ontology for advanced diagnosis of process systems

An ontology for representing operating, safety and control procedures is proposed in this paper that descibes the procedure steps and their connections, and enables to represent exceptions during the normal flow of execution in terms of time-out and failed conditions. The procedure ontology is defined within interconnected components of the process plant and the diagnostic analysis based on risk assessment information, and has been implemented as an integrated ontology in a diagnostic framework using the the Protege ontology editor. A novel diagnostic method based on following these procedures and combining observed malfunctions with Failure Mode and Effects Analysis (FMEA) information is also proposed. The proposed ontology and the diagnostic method are illustrated on a simple operating procedure.

Diagnostic Goal-Driven Reduction of Multiscale Process Models

Fault detection and diagnosis in large-scale process systems is of great practical importance and present various challenging research problems at the same time. One of them is the computational complexity of the algorithms that causes an exponential growth of the computational resources (time and memory) with increasing system sizes [21]. One remedy of this problem is to decompose the system model and effectively focus on its relevant sub-model when doing the fault detection, isolation and loss prevention.

Agent-based diagnosis for granulation processes

A multiagent diagnostic system implemented in a Protege-JADE-JESS environment is described in this paper. It enables the use of a combination of diagnostic methods from heterogeneous knowledge sources. The system is demonstrated on a case study for diagnosis of faults in a granulation process.

Prediction-based diagnosis and loss prevention using qualitative multi-scale models

A prototype prediction based intelligent diagnostic system that is capable of integrating qualitative and quantitative process models and operation experience in the form of HAZOP result tables is proposed in this paper. The knowledge base of the system is organized in a hierarchical way following the hierarchy levels of the multi-scale model of the process system. This supports the focusing of the fault detection and loss prevention and thus decomposes the otherwise computationally hard problem. The system is illustrated on the example of a commercial fertilizer granulator drum.

Intelligent Assumption Retrieval from Process Models by Model-Based Reasoning

Process models can be seen as structured knowledge base elements with syntax and semantics dictated by the underlying physical and chemical laws. The effect of model simplification assumptions is then determined by forward reasoning in order to take into account all of their implications. A bidirectional reasoning method for the retrieval of modelling assumptions from two related process models is proposed. An intelligent model editor is constructed to perform and test the model-based reasoning methods related to the assumption retrieval.

Assumption retrieval from process models

Publisher Summary This chapter discusses the process models of lumped systems in their “canonical” form, where the equations and variables are classified and the “natural” set of design variables and “natural” assignment are selected. An efficient intelligent algorithm is proposed to generate the assumption sequences leading from one model to another in an automated way. Two simple assumption retrieval examples are also presented and discussed for analyzing and comparison purposes. The automated generation of process models from the engineering system description and modeling assumptions is one of the most important and challenging tasks in computer aided modeling (CAM). Efficient algorithms solving this task are essential for constructing automated modeling tools, which can be regarded as intelligent front-ends for dynamic simulators. Process models of lumped systems are differential algebraic equations (DAEs) with a well-defined structure dictated by the underlying physics and chemistry of the system. Modeling assumptions can be regarded as the representations of the engineering activity and decisions during the whole modeling process in constructing, simplifying, and analyzing process models. Assumption-driven modeling works directly with modeling assumptions, thus, enabling the definition and handling of process models as structured text with defined syntax and semantics. Algebraic manipulations are described as equivalence transformations and model simplification and enrichment assumptions as general modeling transformations acting on process models.