Wendy Foslien

Use of SCADA Data for Failure Detection in Wind Turbines

This paper discusses the use of existing wind turbine SCADA data for development of fault detection and diagnostic techniques for wind turbines.

Hybrid Neural Network/Algorithmic Approaches to System Identification

Abstract Linear discrete-time models predominate in process system identification, but suffer from some drawbacks. An appealing alternative is to identify continuous-time linear models, expressed as differential equations or Laplace transforms. Two problems must be solved in a computationally feasible way for continuous-time transfer function identification: the model structure must be determined and, subsequently, values for parameters associated with the structural description must be estimated. Our approach to the identification of continuous-time models integrates two technologies. We use the nonlinear classification capabilities of neural networks for structure determination. For parameter estimation we use a nonlinear identification algorithm. A second neural network is employed to provide the algorithm initial conditions, thereby improving its convergence properties. Network training is greatly facilitated by the dynamic generation of examples—overparametrization and overfitting concerns are alleviated entirely. A system identifier targeted for petrochemical applications has been developed. Some details of the implementation are described and experimental results presented. These demonstrate that both accurate and efficient identification of continuous-time models is feasible.

Balancing Trust and Automation Needs for Effective Home Energy Management

With the increasing shortage of energy resources and the adverse impact of non-renewable fuels on the environment, there is a shift in the consumer’s mindset to emphasize managing and utilizing energy efficiently, reducing green house emissions and contributing to a clean environment. This is especially true to the residential markets where a trusted Home Energy Manager (HEM) device can aid in automating and delivering effective energy management strategies in homes. Home users are often passive in their interaction and have to be engaged and reassured that a HEM device contributes positively to the goal of home energy management. The objective is to boost their trust and confidence in HEM by making information (such as energy costs, usage patterns, etc.) accessible and enabling them to act and conserve energy effectively based on the same. In this research, we explore and understand the potential factors that influence how users would engage and interact with HEM device. Some basic functions of the HEM device include: a trusted advisor that provides dynamic recommendations based on user’s interaction and behavior in the home; ability to sense occupancy within the home and automatically adjust schedules without the need for explicit human intervention; deduce energy usage patterns; and adapt energy management strategies based on the user profiles derived from their behaviors and interaction with the thermostat. Using a HEM device with the proper balance of automation and user engagement can have a positive impact on reducing the global energy consumption and the sustenance of our environment.

Optimization with neural memory for process parameter estimation

The speed and accuracy of convergence of iterative optimization algorithms often depend critically upon the choice of a starting point. With a near optimum starting point, both speed and accuracy can be improved. A two step approach to optimization has been developed which utilizes the feedforward predictive capability of a neural network in conjunction with the feedback capability of an iterative optimization algorithm. This approach is taken in order to improve the speed of the iterative optimization algorithm, and also enhance the iterative algorithms ability to locate a global optimum. This technique has been applied to the problem of system identification for continuous time transfer function models. The neural network is used to select an initial set of process parameters for a given model structure using unit step response data. We present results on the accuracy of the predictive capability of the neural network, and results showing the improved performance of the iterative nonlinear system identification algorithm.

Incremental supervised learning: localized updates in nonlocalized networks

We present a novel yet simple approach to incremental learning in neural networks: the problem of updating a mapping based on limited new data. The approach consists of forming a training set by appending to the new data additional training examples generated by exercising the network. This strategy enables the mapping to be updated in the neighborhood of the new data without causing distortions elsewhere in the input space. The approach can be used with any neural network model; it is particularly useful for the popular multilayer sigmoidal networks in which small parameter changes can have nonlocal consequences. Demonstrations and parametric explorations on a toy problem are described.

Comparing Active vs. Passive Participation in a Virtual Tour An Initial Look at User Attitudes and Preferences

In corporate virtual world (VW) demonstrations, it can oftentimes be difficult to gain the active participation (i.e., first hand interaction) of all users of the demonstration. Due to the general willingness or ability to register with VWs (e.g., Second Life®) and self-efficacy associated with controlling an avatar, many users may be more apt to participate passively (e.g., watch someone else interact). Therefore, in the present work, we investigated the differences and similarities in the attitudes between visitors to a virtual tour, who either actively or passively participated. The results of the study indicated that large group active participation led to more confusion and distraction when compared to large group passive participation. However, passive participants indicated less confidence in their ability to interact with the tour on their own. This paper concludes with lessons learned and recommendations for this virtual tour.