James H. Garrett

Engineering applications of neural networks

This paper describes several prototypical applications of neural network technology to engineering problems. The applications were developed by the authors as part of a graduate-level course taught at the University of Illinois at Urbana-Champaign by the first author (now at Carnegie Mellon University). The applications are: an adaptive controller for building thermal mass storage; an adaptive controller for a combine harvester; an interpretation system for non-destructive evaluation of masonry walls; a machining feature recognition system for use in process planning; an image classification system for classifying land coverage from satellite or high-altitude images; and a system for designing the pumping strategy for contaminated groundwater remediation. These applications are representative of many of the engineering problems for which neural networks are applicable: adaptive control, feature recognition, and design.

Knowledge based standards processing

Abstract This paper presents the representational model of a standard, background research in standards processing, and a current research project which uses a knowledge-based system approach to standards processing. The representational model of a standard, based on decision tables, was originally developed in 1966 and has been continually modified and improved since that time. Concurrent with the development of this model was the development of its uses in computer-aided standards processing, such as automated conformance checking and design. As an extension of this standards processing research, a knowledge-based standards processor is being built that will act as an interface between CAD programs and design standards. The knowledge-based standards processor uses a blackboard architecture similar to that of Hearsay-II, a speech understanding program. The processors knowledge is divided up into about ten knowledge sources, most of which are standard-independent. The knowledge sources are responsible for the development of a design script that can be used to perform the CAD programs member design tasks, while satisfying the governing design standard.

Analysis of Three Indoor Localization Technologies for Supporting Operations and Maintenance Field Tasks

AbstractLocating building components that need to be worked on during maintenance tasks is critical for timely repair of the component and mitigation of the damage. The process of locating a component or a person in a facility is called indoor localization. The objective of this research study is to analyze the feasibility of three indoor localization technologies for supporting operations and maintenance (OM namely, wireless local area network (WLAN), radio frequency identification (RFID) tags, and inertial measurement units (IMU). These technologies have been selected on the basis of the requirements of the localization needed for supporting O&M field activities. A previous work has been extended, which tested RFID-based locations in an indoor environment, by testing the three selected technologies in the same test bed and using the same hypothesis and fingerprinting approach developed in the previous work. The two main motivations behind using the same test bed and same approach are to h...

Toward Data-Driven Structural Health Monitoring: Application of Machine Learning and Signal Processing to Damage Detection

AbstractA multilayer data-driven framework for robust structural health monitoring based on a comprehensive application of machine learning and signal processing techniques is introduced. This paper focuses on demonstrating the effectiveness of the framework for damage detection in a steel pipe under environmental and operational variations. The pipe was instrumented with piezoelectric wafers that can generate and sense ultrasonic waves. Damage was simulated physically by a mass scatterer grease-coupled to the surface of the pipe. Benign variations included variable internal air pressure and ambient temperature over time. Ultrasonic measurements were taken on three different days with the scatterer placed at different locations on the pipe. The wave patterns are complex and difficult to interpret, and it is even more difficult to differentiate the changes produced by the scatterer from the changes produced by benign variations. The sensed data were characterized by 365 features extracted from a variety of...

A density-based spatial clustering approach for defining local indicators of drinking water distribution pipe breakage

Abstract The physical condition of American infrastructure systems has raised concerns that have been addressed, in part, by studies addressing their condition assessment. Condition assessment aims at describing current condition and estimating remaining service lives of infrastructure network components. This is a predominantly time-based analysis, which can be complemented by the spatial analysis of physical condition data of infrastructure components. More specifically, exploratory spatial data analysis might identify areas with high failure rates and generate local indicators of condition for subsets of pipe segments within the physical network. Such local indicators can be used in cost/benefit analysis for planning capital investments with the advantage of allowing the identification of critical customers within critical regions and therefore better accounting for social costs. This paper aims to provide an approach to spatial data clustering of networked infrastructure failure data, which is presented and demonstrated by applying it to a drinking water pipe breakage dataset. Clusters are, in this paper, the set of break points that occurred in regions that present high breakage density per mile of pipe compared to the pipes in their vicinity. The proposed approach can be framed as density-based data clustering approach, and its output consists of a hierarchical clustering of breaks. The root node of the cluster hierarchy, which contains the set of all break points, is subsequently partitioned into smaller clusters. In this hierarchy, clusters are subdivided to reflect different breakage densities along the network space. Therefore, clusters in the lower level of the hierarchy present more homogeneous breakage rates. The results of the proposed approach are assessed according to their sensitivity to choice of parameters and according to a clustering quality measure. The chosen parameters are shown to provide results that are superior compared to a range of parameter choices, in terms of clustering quality, and also compared to the less important though relevant criteria of the number and size of clusters generated.

A neural network-based machine learning approach for supporting synthesis

The goal of machine learning for artifact synthesis is the acquisition of the relationships among form, function, and behavior properties that can be used to determine more directly form attributes that satisfy design requirements. The proposed approach to synthesis knowledge acquisition and use (SKAU) described in this paper, called NETSYN, creates a function to estimate the probability of each possible value of each design property being used in a given design context. NETSYN uses a connectionist learning approach to acquire and represent this probability estimation function and exhibits good performance when tested on an artificial design problem. This paper presents the NETSYN approach for SKAU, a preliminary test of its capability, and a discussion of issues that need to be addressed in future work.

MIA: a wearable computer for bridge inspectors

MIA (Mobile Inspection Assistance) is a wearable computer system that helps bridge inspectors collecting multimedia information in the field and producing the inspection report. MIA allows an inspector to fill out the inspection form, access previous inspection reports, make sketch(s) of the bridge element(s), take photograph(s), and produce the inspection report via a voice or pen interface. The first prototype of the system was developed by students in a Wearable Computer Project Design Course at Carnegie Mellon University during Spring 1998 semester. The hardware and software aspects of the system is discussed, as well as the application of the system for inspection.

Sensor data driven proactive management of infrastructure systems

In a paper presented at the ASCE International Conference on Computing in Civil Engineering in Cancun, Mexico, a vision was laid out for sensor data-driven, proactive management of infrastructure systems in which information and communication technology is used to more efficiently and effectively construct infrastructure systems, monitor their performance, and enable an intelligent operation of these systems. Since that time, a research center at Carnegie Mellon, the Center for Sensed Critical Infrastructure Research (CenSCIR), has been established with a mission to perform research towards this vision. The objectives for this paper are: 1) to discuss the motivation for such sensor-data driven proactive infrastructure management; 2) to identify and discuss the major research questions that need to be addressed by CenSCIR to achieve this vision; and 3) to present several CenSCIR projects that address some of these research questions.

A Knowledge Discovery Framework for Civil Infrastructure: A Case Study of the Intelligent Workplace

Large databases are becoming increasingly common in civil infrastructure applications. Although it is relatively simple to specifically query these databases at a low level, more abstract questions like ‘How does the environment affect pavement cracking?’ are difficult to answer with traditional methods. Data mining techniques can provide a solution for learning abstract knowledge from civil infrastruc-ture databases. However, data mining needs to be performed within a systematic process to ensure correct and reproducible results. Many decisions must be made during this process, making it difficult for novice analysts to apply data mining techniques thoroughly. This paper presents an application of a knowledge discovery process to data collected for an ‘intelligent’ building. The knowledge discovery process is illustrated and explained through this case study. Additionally, we discuss the importance of this case study in the context of a research effort to develop an interactive guide for the knowledge discovery process.

An object-centered approach for modelling engineering design products: Combining description logic and object-oriented modelling

Class-centered data models, such as the object-oriented data model, are inadequate for supporting engineering design product models because of their lack of support for object evolution, schema evolution, and semantic and user-defined relationships. Description logic overcomes these limitations by providing constructs for intentional description of classes, relationships, and objects. By combining description logic with object-oriented modelling concepts, design product schemas and data can be uniformly represented and modified throughout the design process.