Navigation of autonomous surface vessels, less than 20 m in length, in large sea states is difficult and often precludes successful completion of the assigned mission or, in the worst case, survivability. Operation in high seas requires sensing of the local wave environment and determining a vessel trajectory that maximizes survivability based on knowledge of the vessel response functions and prediction of the incident wave field forward in time. To achieve this objective, new technologies are being developed and tested in full scale at the Marine Autonomy Research Site (MARS), located in central Lake Superior and operated by Michigan Technological University. In this initial set of experiments, a skilled human operator was used as a surrogate for an envisioned wave-adaptive autonomous control system. The test vehicle is a fully instrumented personal watercraft, operated by a U.S. Coast Guard-trained surf-boat operator in moderate sea states with Froude number (Fr) = 1.0 through a course consisting of upwave, cross-wave, and down-wave legs. Results dramatically document that the wave-dodging maneuvers employed are designed to minimize vessel pitch (preserve propulsor and rudder control) while allowing increased vessel roll. Comparisons of the straight line with wave-dodging circuits during constant sea state conditions show that vehicle roll is at times twice greater in wave-dodging runs while vehicle pitch averages half to one third of that in the straight-line course. These data suggest that optimum paths do exist through steep, evolving incident wave fields and these optimum paths can produce significant improvements in vessel survivability.

Energy storage requirements and its management are important considerations for dc microgrid designs that have a high penetration of stochastic distributed sources and loads. Modern control methods, such as Hamiltonian Surface Shaping and Power Flow Control (HSSPFC), often rely on a reduced order model of the microgrid for controller design. This paper explores (1) the reduced order boost converter model for use in development of advanced control schemes via a detailed, switching mode model implemented on a Typhoon HIL 602 with a controller-in-the-loop (CIL) and (2) a design methodology that may be used for determining converter distributed storage requirements for the closed loop controls.

Changing emissions regulations, fuel price fluctuations and development of new energy-intensive mission systems are driving both component technological innovation and require more sophisticated controls on-board modern ships. Historically, numerous components and systems aboard ships perform energy conversions from one form to another. Conversion from chemical to thermal to kinetic to electrical and back to thermal energy are common. Today, subsystems are designed separately without opportunity for optimizing overall system-of-systems performance. By modeling multiple system domains simultaneously and applying the Second Law of Thermodynamics, this progresses towards overall ship system optimization. Exergy, the available energy for performing useful work, and exergy destruction was calculated in each energy conversion process. Knowledge of the exergy flows leads to a system-of-systems optimization to minimize overall exergy destruction translating into lower emissions and fuel costs. This can eventually result in more efficient, smaller and lighter shipboard systems. A model of notional shipboard power and cooling system is presented that features a pulsed load (an electromagnetic railgun) and have implemented both traditional and exergy-based control schemes. This paper will briefly review the modeling, which has been previously published, and present results using exergy destruction for optimization of the ship system controls.

This paper proposes a novel concept of optimizing the load profile for a given capacitive energy storage system using a Lyapunov based closed loop controller where the objective is to balance energy and power usage of a load. The energy storage source is modeled as a capacitor for simplicity to explore the theoretical and practical implications of the developed technique. After illustrating the control law experimentally, simulation and experimental results are presented to extrapolate its use to other conditions. Experimentally, the load is implemented with a controllable load instrument. Weights are used to emphasize different features of the optimal control law’s objective. When the objective is focused towards minimizing load voltage, the control law results in a simple, exponential discharge of the storage energy. In contrast, when load power is the objective, the majority of the storage energy is extracted in an initial pulse with a higher loss of energy through the system’s parasitic resistance. These two behaviors allow for a shaping of the storage power through closed loop load scheduling. Results show that the optimal discharge shape approaches that of a pulse load, and is applicable to pulse load applications such as rail guns and radar on electric naval ships.

The more electric aircraft (MEA) concept aims to reduce emissions, fuel costs, and enable incorporation of electric weapon systems and advanced sensor platforms. These systems will further burden the electrical system due to the pulse like loading and require advanced control strategies and distributed energy storage systems to ensure stability. Furthermore, multi-physical coupling of thermal electrical systems are often compartmentalize and analyzed separately, forgoing congruency that could occur if analyzed together. Here, we study how exergy, the amount of useful energy throughout a system, can guide control design and system operation. A multi-physics networked microgrid model was developed of an aircraft with two generation sources, interconnecting power converters, a lumped thermal mass model and pulsed loading. The Hamiltonian Surface Shaping Power Flow Control (HSSPFC) strategy is applied to the electrical grid via idealized and distributed storage elements. The HSSPFC was first developed to solve a general, scalable, form a networked microgrid architecture and then applied to the specific aircraft model. Implementation of the HSSPFC requires an outer loop to balance installed generation and to manage storage. This was accomplished through an exergy optimal set point generation scheme that minimized exergy destruction in the power converters. Bus regulation of within 3% of the desired set point was achieved while servicing a 100 kW pulsed load. A tradeoff between optimization update rate and storage regulation was found to be limited by the algorithm execution speed. Increased optimization update rates were linked to reduced storage use and fewer transients in bus voltage. The thermal model was electrically coupled through pumping loads and by cooling the power electronics. Exergy optimal coolant pump operation was also studied. The minimal exergy and pump energy consumption were obtained by operating the coolant system near the upper temperature limit of the coolant, which minimized cooling electrical loads.

This paper presents two control strategies: (i) An optimal exergy destruction (OXD) controller and (ii) a decentralized power apportionment (DPA) controller. The OXD controller is an analytical, closed-loop optimal feedforward controller developed utilizing exergy analysis to minimize exergy destruction in an AC inverter microgrid. The OXD controller requires a star or fully connected topology, whereas the DPA operates with no communication among the inverters. The DPA presents a viable alternative to conventional P − ω / Q − V droop control, and does not suffer from fluctuations in bus frequency or steady-state voltage while taking advantage of distributed storage assets necessary for the high penetration of renewable sources. The performances of OXD-, DPA-, and P − ω / Q − V droop-controlled microgrids are compared by simulation.

The US Navy is moving towards more electric ships with ongoing development of high-powered radars and electronic warfare systems, directed energy and electromagnetic weapons. One of the key enablers for these systems is improving the ability to control the power and energy flows aboard existing and future warships. This includes electrical, chemical and thermal energy flows. Most advanced control efforts focus on a single shipboard system (i.e. electrical or thermal). This paper is investigating the ability to leverage the Second Law of Thermodynamics by developing controls which minimize Exergy destruction across domains. An Exergy-based approach is developed to allow improved control design. The development and validation of reduced order dynamic models for shipboard electrical, thermal and missions systems is also described. These models have been developed with additional exergy terms so that they can be utilized to conduct optimization with exergy destruction based objective functions. Some of the components do not yet exist in hardware, therefore the reduced order models are validated against well-known detailed dynamic models.

A vision based tracking algorithm was developed for active load damping on a 1/9 th scale extending boom crane using the Microsoft Xbox 360 Kinect RGB-D camera. The damping performance was compared to the IMU approach based on the Microstrain Inertia Link sensor. With the IMU solution, the radial damping ratio was improved by a factor of 2.2 and an order of magnitude in the tangential direction. The vision solution achieved a damping ratio improvement by a factor 2.0 and 4.7 in the radial and tangential directions respectively. The settling times for both sensors were with a few seconds at a total duration of 25 and 15 seconds for the radial and tangential directions when active damping was applied. The cranes response was a limiting factor in the system's damping performance as it would be with a full scale crane.

Microgrids with distributed generation and storage assets often form an underdetermined system of power flow equations for balancing loads and generation. This feature is compounded for networked microgrid topologies. Advanced control strategies offer a solution to this power flow problem and often require a feedforward reference. This provides an opportunity to compute energy optimal references but with the limitation of solution computation time. Therefore, four optimal reference command generators were developed focusing on solution time and scalability to both asset quantity and number of microgrids networked. Strategies explored were pure numerical, closed form, and numerical hybrids, and a Lagrange multiplier method. A tradeoff existed between smaller solution time of the Lagrange multiplier method and guaranteeing a feasible solution intrinsic to the hybrid approaches. Timing trials were performed where generation assets per microgrid ranged from 1 to 130000 and networked microgrid quantity ranged from 1 to 250. These trials quantified the bounds where subsecond updates rates could be achieved for two types of topologies.

A sensor isolation system was developed to reduce vibrational and noise effects on MEMS IMU sensors. A single degree of freedom model of an isolator was developed and simulated. Then a prototype was constructed for use with a Microstrain 3DM-GX3-25 IMU sensor and experimentally tested on a six DOF motion platform. An order of magnitude noise reduction was observed on the z accelerometer up to seven Hz. The isolator was then deployed on a naval ship along with a DMS TSS-25 IMU used as a truth measurement and a rigid mounted 3DM sensor was used for comparison. Signal quality improvements of the IMU were characterized and engine noise at 20 Hz was reduced by tenfold on x, y, and z accelerometers. A heave estimation algorithm was implemented and several types of filters were evaluated. Lab testing with a six DOF motion platform with pure sinusoidal motion, a fixed frequency four pole bandpass filter provided the least heave error at 12.5% of full scale or 0.008m error. When the experimental sea data was analyzed a fixed three pole highpass filter yielded the most accurate results of the filters tested. A heave period estimator was developed to adjust the filter cutoff frequencies for varying sea conditions. Since the ship motions were small, the errors w.r.t. full scale were rather large at 78% RMS as a worst case and 44% for a best case. In absolute terms when the variable filters and isolator were implemented, the best case peak and RMS errors were 0.015m and 0.050m respectively. The isolator improves the heave accuracy by 200% to 570% when compared with a rigidly mounted 3DM sensor.

An actively controlled airbag system used as a three axis motion platform is characterized by a unique application of the multiple input H1 Frequency Response Function (FRF) estimator. The goal of the testing is to understand the lateral (in-plane) stiffness properties of the inflated airbag at different inflation pressures and heights. The characterization is required to explore the controllability of the platform. The dynamic lateral magnitude increased 11.3 %, from 0.62 to 0.69 G/N, on average while damping decreased by 11.5 %, from 0.164 to 0.147, as height increased. Furthermore, the natural frequency shifted from 8.6 to 4.2 Hz as pressure and height varied. The static lateral stiffness increased on average 3.4 fold, from 6.5 to 22.2 N/cm from the lowest to highest pressure.

Inexpensive, commercial available off-the-shelf (COTS) Global Positioning Receivers (GPS) have typical accuracy of ±3 meters when augmented by the Wide Areas Augmentation System (WAAS). There exist applications that require position measurements between two moving targets. The focus of this work is to explore the viability of using clusters of COTS GPS receivers for relative position measurements to improve their accuracy. An experimental study was performed using two clusters, each with five GPS receivers, with a fixed distance of 4.5 m between the clusters. Although the relative position was fixed, the entire system of ten GPS receivers was on a mobile platform. An expensive RTK GPS system, accurate to the centimeter level, was also fixed to the mobile platform and used for comparison. Data was recorded while moving the system over a rectangular track with a perimeter distance of 7564 m. The cluster data was post processed and yielded approximately one meter accuracy for the relative position vector between the two clusters. Furthermore, the 2σ and RMS values were decreased by approximately 13 fold from a worst case scenario.

The US military is moving toward the electrification of many weapon systems and platforms. Advanced weapon systems such as high energy radar, electro-magnetic kinetic weapons and directed energy pose significant integration challenges due to their pulsed power electrical load profile. Additionally, the weapons platforms, including ships, aircraft, and vehicles can be studied as a mobile microgrids with multiple generation sources, loads, and energy storage. There is also a desire to extend the mission profile and capabilities of these systems. Common goals are to increase fuel efficiency, maintaining system stability, and reduce energy storage size as typically required to enable pulsed load devices. To achieve these goals, there is an opportunity to optimize system performance by considering system wide exergy, a measure of the useful energy within the system. By studying exergy, systems with multi-physical coupling, as with electrically pulsed devices that require cooling, the system can be optimized holistically. While numerous optimization approaches exist, many focus on the long term, hours to days, energy management problem. Furthermore, advance control strategies such as the Hamiltonian Surface Shaping Power Flow Controller (HSSPFC) require feedforward operating points about which storage is actuated to maintain stability. Storage size can be reduced by combining the HSSPFC with exergy based optimization strategies designed for sub-second update rates. In this dissertation, several numerical and closed form/numerical hybrid optimization strategies were developed where speed of solution was explored vs. microgrid asset size, in non-realtime simulation. Then an exergy based optimization strategy was combined with the HSSPFC on a three bus networked microgrid model using average switch mode models, pulse loading, and a thermal system. The three bus model was then extended to a co-simulation on Hardware-in-the-loop (HIL) using an Opal- RT OP5700 and Typhoon HIL 600 realtime simulators, where the optimization was executed asynchronously through UDP Ethernet communication. The storage utilization was reduced by orders of magnitude when comparing two cases of optimized vs. non-optimized generation settings. Bus voltage regulation was within 5 % where there was a trade off between optimization update rate, transient regulation, and storage utilization. Contributions of this work are summarized as follows. An exergy based, asynchronous, optimization strategy was developed to work in concert with the HSSPFC strategy where sub-second update rates were achieved. A co-simulation test bench was developed to allow the study of advanced control strategies through the use of multiple realtime HIL simulators. The methodology for integrating the HIL simulators is given including wiring, calibration, signal scaling, and implementation specific details. An on-line optimization strategy was also developed to interact with the HIL system and used for determining power converter duty cycle biases.