James W. Fonda

Assessment of an instrumented reinforced- concrete bridge with fiber-reinforced- polymer strengthening

Field instrumentation is investigated on an in-service highway bridge over a 2-year period. Extrinsic Fabry-Perot interferometric EFPI strain sensors provide a permanent health-monitoring capability. The bridge is a reinforced-concrete RC structure that was repaired and strengthened using fiber-reinforced-polymer FRP wraps. A sensor net- work monitors the load-induced strain in the FRP reinforcement and the steel rebar. Colocated electrical resistance strain gauges and a finite element analysis are used for comparison. Both dynamic and static load characteristics are analyzed for a near-capacity truck. The fiber optic measurements are generally consistent with the comparison measure- ments and the analytical results; and they show no failure or degradation as opposed to the electrical resistance gauges. We demonstrate the implementation and the performance of in situ EFPI sensors in a long- term field environment. Embedded fiber optic sensors can provide the required information for the intelligent management of a transportation infrastructure.

Embeddable modular hardware for multi-functional sensor networks

A multi-layer node is described for multi-functional sensor networks. The generation-4 smart sensor node (G4SSN) is light weight, has a small footprint, and is low power to support dedicated, embedded applications. It has core layers for data sensing, data processing and wireless networking. The modular physical layout is built around a flexible, multi-channel bus architecture and routing protocols are easily tailored. Additional stackable layers and devices can be easily configured and programmed to meet specific application requirements, especially for prototyping and research investigations. The feasibility for high-resolution sensor data acquisition and wireless transmission is demonstrated using the dynamic strain behavior of an instrumented cantilever beam. The G4SSN is adaptable with different hardware components such as different sensor types and radio layer capabilities.

Embeddable Sensor Mote for Structural Monitoring

An embeddable sensor mote for structural monitoring is described. The Missouri University of Science and Technology (MST) F1 mote is designed to provide a general platform for sensing, networking, and data processing. The platform consists of an 8051 variant processor, two 802.15.4 variant radio platform options, micro Smart Digital (SDTM) flash storage, USB connectivity, RS-232 connectivity, and various sensor capabilities. Sensor capabilities include, but are not limited to, strain gauges, a three-axis multi-range accelerometer, thermocouples, and interface options for other digital and analog sensors via a screw terminal block. In its default configuration the strain conditioning channel is appropriate for structural monitoring, but through reconfiguration it can be used with other resistive bridge transducers for pressure, force, displacement, etc. The F1 mote provides capabilities for strain, temperature, and vibration sensing in a small package. The mote is used at MST for networked monitoring of structures and networked robotic vehicles. In this paper an overview of the F1 mote will be given that emphasizes its operating architecture and potential applications. Applications include infrastructure monitoring for structures such as bridges, levees, and buildings as well as robotics, machine monitoring, and sensor networks. The described platform provides novelty in that it has the ability to be a dedicated structural monitoring system, however can be also used in development of other systems. The F1 platform was designed to combine features of available dedicated platforms and available development kits. The F1 provides a novel combination of sensing, processing, and application possibilities for the targeted application areas.

An online model-based fault diagnosis scheme for HVAC systems

In this paper, a model based fault detection and isolation (FDI) scheme with online fault learning capabilities is proposed for HVAC systems. An observer comprising of an online approximator in discrete-time (OLAD) and a robust term is used for detection. A fault is detected if the generated detection residual, which is defined as the error between the observer outputs and HVAC system states, exceeds an apriori chosen threshold. The OLAD term in the FD observer learns the fault dynamics online while the robust term guarantees asymptotic estimation of the system states. Subsequent to detection, a fault isolation observer, which comprises of the model of fault functions and another robust term, is initiated to identify the root cause. A fault is identified if the isolation residual converges to zero, where the residual is obtained by comparing outputs of the isolation observer and the system. Additionally, we consider different fault scenarios in the system such as single or simultaneous multiple faults. Analytical results for the FDI scheme guarantee the robustness and stability of the proposed scheme. Finally, a simulation example is used to demonstrate the proposed FDI scheme.

Development of autonomous triggering instrumentation

Triggering instrumentation for autonomous monitoring of load-induced strain is described for economical, fast bridge inspection. The development addresses one aspect for the management of transportation infrastructure - bridge monitoring and inspection. The objectives are to provide quantitative performance information from a load test, to minimize the setup time at the bridge, and to minimize the closure time to traffic. Multiple or networked measurements can be made for a prescribed loading sequence. The proposed smart system consists of in-situ strain sensors, an embedded data acquisition module, and a measurement triggering system. A companion control unit is mounted on the truck serving as the load. As the truck moves to the proper position, the desired measurement is automatically relayed back to the control unit. In this work, the testing protocol is developed and the performance parameters for the triggering and data acquisition are measured. The test system uses a dedicated wireless sensor mote and an infrared positioning system. The electronic procedure offers improvements in available information and economics.

Robust neural network RISE observer based fault diagnostics and prediction

A novel fault diagnostics and prediction scheme in continuous-time is introduced for a class of nonlinear systems. The proposed method uses a novel neural network (NN) based robust integral sign of the error (RISE) observer, or estimator, allowing for semi-global asymptotic stability in the presence of NN approximation errors, disturbances and unmodeled dynamics. This is in comparison to typical results presented in the literature that show only boundedness in the presence of uncertainties. The output of the observer/estimator is compared with that of the nonlinear system and a residual is used for declaring the presence of a fault when the residual exceeds a user defined threshold. The NN weights are tuned online with no offline tuning phase. The output of the RISE observer is utilized for diagnostics. Additionally, a method for time-to-failure (TTF) prediction, a first step in prognostics, is developed by projecting the developed parameter-update law under the assumption that the nonlinear system satisfies a linear-in-the-parameters (LIP) assumption. The TTF method uses known critical values of a system to predict when an estimated parameter will reach a known failure threshold. The performance of the NN/RISE observer system is evaluated on a nonlinear system and a simply supported beam finite element analysis (FEA) simulation based on laboratory experiments. Results show that the proposed method provides as much as 25% increased accuracy while the TTF scheme renders a more accurate prediction.

Adaptive Distributed Fair Scheduling for Multiple Channels in Wireless Sensor Networks

A novel adaptive and distributed fair scheduling (ADFS) scheme for wireless sensor networks (WSN) in the presence of multiple channels (MC-ADFS) is developed. The proposed MC-ADFS increases available network capacity and focuses on quality-of-service (QoS) issues. When nodes access a shared channel, the proposed MC-ADFS allocates the channel bandwidth proportionally to the packets weight which indicates the priority of the packets flow. The packets are dynamically assigned to channels based on the packet weight and current channel utilization. The dynamic assignment of channels is facilitated by use of receiver-based allocation and alternative routes. Moreover, MC-ADFS allows the dynamic allocation of network resources with little added overhead. Packet weights are initially assigned using user specified QoS criteria, and subsequently updated as a function of delay and queued packets. The back-off interval is also altered using the weight adaptation. The weight update and back-off interval selection ensure global fairness is attained even with variable service rates.

Optimized energy-delay sub-network routing protocol development and implementation for wireless sensor networks

The development and the implementation issues of a reactive optimized energy-delay sub-network routing (OEDSR) protocol for wireless sensor networks (WSN) are introduced and its performance is contrasted with the popular ad hoc on-demand distance vector (AODV) routing protocol. Analytical results illustrate the performance of the proposed OEDSR protocol, while experimental results utilizing a hardware testbed under various scenarios demonstrate improvements in energy efficiency of the OEDSR protocol. A hardware platform constructed at the University of Missouri-Rolla (UMR), now the Missouri University of Science and Technology (MST), based on the Generation 4 Smart Sensor Node (G4-SSN) prototyping platform is also described. Performance improvements are shown in terms of end-to-end (E2E) delay, throughput, route-set-up time and drop rates and energy usage is given for three topologies, including a mobile topology. Additionally, results from the hardware testbed provide valuable lessons for network deployments. Under testing OEDSR provides a factor of ten improvement in the energy used in the routing session and extends network lifetime compared to AODV. Depletion experiments show that the time until the first node failure is extended by a factor of three with the network depleting and network lifetime is extended by 6.7%.

Vibration isolation/suppression: research experience for undergraduates in mechatronics and smart structures

This paper provides an account of a student research project conducted under the sponsoring of the National Science Foundation (NSF) program on Research Experience for Undergraduates (REU) in Mechatronics and Smart Strictures in the summer of 2000. The objective of the research is to design and test a stand-alone controller for a vibration isolation/suppression system. The design specification for the control system is to suppress the vibrations induced by the external disturbances by at least fiver times and hence to achieve vibration isolation. Piezo-electric sensors and actuators are utilized for suppression of unwanted vibrations. Various steps such as modeling of the system, controller design, simulation, closed-loop testing using d- Space rapid prototyping system, and analog control implementation are discussed in the paper. Procedures for data collection, the trade-offs carried out in the design, and analog controller implementation issues are also presented in the paper. The performances of various controllers are compared. The experiences of an undergraduate student are summarized in the conclusion of the paper.

Structural identification using a low-cost search method

An easily implementable and trainable damage detection method is proposed and implemented for a simple truss structure. The approach uses the iterative search identification method and is compatible with low-cost and low-power microcontroller hardware. This method employs pattern matching for a data set from a strain sensor array and predicts location (truss member) and severity (member cross sectional area) of damage. As a health monitoring approach, the method is not as robust or rigorous as more complex methods. However, it has modest processing requirements and can handle noisy signals. The work presents an algorithm applied to a truss structure, the simulation performance from a finite-element-analysis, and a discussion of capabilities. The simulation demonstrates differing damage locations, damage severity, and signal noise. Its suitability for low-cost and low-power field processors is discussed.