Naresh Sundaram Iyer

Management of Complex Dynamic Systems based on Model-Predictive Multi-objective Optimization

Over the past two decades, model predictive control and decision-making strategies have established themselves as powerful methods for optimally managing the behavior of complex dynamic industrial systems and processes. This paper presents a novel model-based multi-objective optimization and decision-making approach to model-predictive decision-making. The approach integrates predictive modeling based on neural networks, optimization based on multi-objective evolutionary algorithms, and decision-making based on Pareto frontier techniques. The predictive models are adaptive, and continually update themselves to reflect with high fidelity the gradually changing underlying system dynamics. The integrated approach, embedded in a real-time plant optimization and control software environment has been deployed to dynamically optimize emissions and efficiency while simultaneously meeting load demands and other operational constraints in a complex real-world power plant. While this approach is described in the context of power plants, the method is adaptable to a wide variety of industrial process control and management applications

Framework for post-prognostic decision support

This paper describes a decision support system (DSS) for use in operational decision making with PHM-specific data. Challenges arise from the large amount of different information pieces upon which a decision maker has to act. Conflicting information from on-board and off-board PHM modules, seemingly contradictory and changing requirements from operations as well as maintenance for a multitude of different systems within strict time constraints make operational decision-making a difficult undertaking. The DSS enables the user to make optimal decisions based on his expression of rigorous trade-offs between different prognostic and external information sources. This is accomplished through guided evaluation of different optimal decision alternatives under operational boundary conditions using user-specific and interactive collaboration. We present some preliminary results of the use of such a DSS for post-prognostics decision-making

Soft computing applications to prognostics and health management (PHM): leveraging field data and domain knowledge

Assets Prognostics and Health Management (PHM) is a promising application area for Soft Computing (SC). To better understand PHM requirements, we introduce a decision-making framework in which we analyze PHM decisional tasks. This framework is the cross product of the decisions time horizon and the domain knowledge used by SC models. Within such a framework, we analyze the progression from simple to annotated lexicon, morphology, syntax, semantics, and pragmatics. We use this metaphor to monitor the leverage of domain knowledge in SC to perform anomaly detection, anomaly identification, failure mode analysis (diagnostics), estimation of remaining useful life (prognostics), on-board control, and off board logistics actions. We illustrate a case study in anomaly detection, which is solved by the construction and fusion of an ensemble of diverse detectors, each of which is based on different SC technologies.

Multivariate anomaly detection in real-world industrial systems

Anomaly detection is a critical capability enabling condition-based maintenance (CBM) in complex real-world industrial systems. It involves monitoring changes to system state to detect ldquoanomalousrdquo behavior. Timely and reliable detection of anomalies that indicate faulty conditions can help in early fault diagnostics. This will allow for timely maintenance actions to be taken before the fault progresses and causes secondary damage to the system leading to downtime. When an anomaly is identified, it is important to isolate the source of the fault so that appropriate maintenance actions can be taken. In this paper, we introduce effective multivariate anomaly detection techniques and methods that allow fault isolation. We present experimental results from the application of these techniques to a high-bypass commercial aircraft engine.

A Review of Two Industrial Deployments of Multi-criteria Decision-making Systems at General Electric

Two industrial deployments of multi-criteria decision-making systems at General Electric are reviewed from the perspective of their multi-criteria decision-making component similarities and differences. The motivation is to present a framework for multi-criteria decision-making system development and deployment. The first deployment is a financial portfolio management system that integrates hybrid multi-objective optimization and interactive Pareto frontier decision-making techniques to optimally allocate financial assets while considering multiple measures of return and risk, and numerous regulatory constraints. The second deployment is a power plant management system that integrates predictive modeling based on neural networks, optimization based on multi-objective evolutionary algorithms, and automated decision-making based on Pareto frontier techniques. The integrated approach, embedded in a real-time plant optimization and control software environment dynamically optimizes emissions and efficiency while simultaneously meeting load demands and other operational constraints in a complex real-world power plant

Evaluation of filtering techniques for aircraft engine condition monitoring and diagnostics

Engine condition monitoring and diagnostics are critical for safe, efficient and profitable aircraft operation. Snapshots of sensor measurements are typically used for engine health monitoring. Deviation of those measurements from a reference condition is a key feature for engine fault detection. Measurement noise and sensor failure, however, often contaminate those signals and affect performance of the fault detection. Filtering or signal smoothing is therefore a helpful technique to improve fault isolation and accuracy and robustness of the detection. Filtering is a very active area of research in the signal processing field; in the past several decades, a large variety of techniques have been developed. Each technique has its advantages and disadvantages, depending on the type of signal. The goal of this paper to review and evaluate various filtering techniques for aircraft engine condition monitoring applications, and recommend the most suitable filtering algorithms based on the unique characteristics of aircraft engine signals.

Automated Risk Classification and Outlier Detection

Risk assessment is a common task present in a variety of problem domains, ranging from the assignment of premium classes to insurance applications, to the evaluation of disease treatments in medical diagnostics, situation assessments in battlefield management, state evaluations in planning activities, etc. Risk assessment involves scoring alternatives based on their likelihood to produce better or worse than expected returns in their application domain. Often, it is sufficient to evaluate the risk associated with an alternative by using a predefined granularity derived from an ordered set of risk-classes. Therefore, the process of risk assessment becomes one of classification. Traditionally, risk classifications are made by human experts using their domain knowledge to perform such assignments. These assignments will drive further decisions related to the alternatives. We address the automation of the risk classification process by exploiting risk structures present in sets of historical cases classified by human experts. We use such structures to pre-compile risk signatures that are compact and can be used to classify new alternatives. Specifically, we use dominance relationships, exploiting the partial ordering induced by the monotonic relationship between the individual features and the risk associated with a candidate alternative, to extract such signatures. Due to its underlying logical basis, this classifier produces highly accurate and defensible risk assignments. However, due to its strict applicability constraints, it covers only a small percentage of new cases. In response, we present a weaker version of the classifier, which incrementally improves its coverage without any substantial drop in accuracy. Although these approaches could be used as risk classifiers on their own, we found their primary strengths to be in validating the overall logical consistency of the risk assignments made by human experts and automated systems. We refer to potentially inconsistent risk assignments as outliers and present results obtained from implementing our technique in the problem of insurance underwriting

Anomaly detection using data clustering and neural networks

Anomaly detection provides an early warning of unusual behavior in units in a fleet operating in a dynamic environment by learning system characteristics from normal operational data and flagging any unanticipated or unseen patterns. For a complex system such as an aircraft engine, normal operation might consist of multiple modes in a high dimensional space. Therefore, anomaly detection approaches based on single cluster data structure will not work. This paper investigates data clustering and neural network based approaches for anomaly detection, specifically addressing the situation which normal operation might exhibit multiple hidden modes. Results show detection accuracy can be improved by data clustering or learning the data structure using neural networks.

MONACO - Multi-Objective National Airspace Collaborative Optimization

The U.S. national Air Traffic Management (ATM) system is today operating at the edge of its capabilities, handling the real-time planning and coordination of over 50,000 flights per day. This situation will only worsen in the years to come, as it is expected that U.S. air traffic will nearly double by the year 2025. There is a pressing need therefore for increasing capacity to meet future demand, improving safety, enhancing efficiency, providing additional flexibility to airline operators, and equitable consideration of multiple stakeholder needs in this complex dynamic system. In this paper, we present a scalable enterprise framework for multi-stakeholder, multi-objective model-based planning and optimization of air traffic in the national airspace system (NAS). The approach is based on an intelligent evaluation and optimization at the strategic and flight route levels. At the strategic level, we focus on separations between flights to improve airspace system performance. At the flight route level, we focus on identifying an optimal portfolio of flight paths within a planning horizon that trades-off a reduction in miles flown and a reduction in congestion. This framework not only considers system-level objectives, but also regards the impact of decisions on the principal stakeholders within the NAS. It is expected that this system will serve as a key decision-support tool to address future NAS scalability and reliability needs.

Anomaly Detection Using Non-Parametric Information

Diagnostics and prognostics is particularly challenging in systems with a restricted suite of sensors; e.g., in aircraft engines where harsh operating conditions, weight considerations, and regulatory concerns limit the number of sensors. In this paper, we investigate anomaly detection techniques subject to these constraints. Specifically, we use as input to these techniques only controller-generated, log-data for the system. While such log-data is not designed to carry predictive information related to system health, we show that it is possible to extract early warning signals related to the failure of the system by looking for the presence or onset of anomalous or novel patterns in the log-data. We present preliminary results obtained by the application of this approach to some complex systems. We also provide a roadmap for extending this approach by the incorporation of minimal amount of system-specific knowledge of the kind that is typically available for complex systems. This extension is expected to strengthen the applicability of the approach to diagnostic and prognostic analysis at the level of the system components, as well as to the estimation of the root cause of a detected system anomaly.Copyright