David E. Hess

Impact of Flow Control Technologies on Naval Platforms (Invited)

This paper highlights flow control technologies and draws a connection between the technologies and US Navy applications, specifically submarines, where possible. This paper does not provide an exhaustive citation listing typical of review papers, but instead focuses on selected applications along with a brief history of submarines. The major theme of the paper is to provide some rationale for which technologies may work in real operational conditions, why technologies have successfully been demonstrated in the laboratory but fail in real operational scenarios, and provide some directions for future research on promising flow control technologies.

Improved Simulation of Ship Maneuvers Using Recursive Neural Networks

An improved Recursive Neural Network (RNN) maneuvering simulation tool for surface ships is described. Inputs to the simulation, cast in the form of forces and moments, are redefined and extended in a manner that more accurately captures the physics of ship motion; the new model is used to extend initial efforts toward RNN surface ship simulations. These extensions include improved formulations of propeller thrust, lift from deflected rudders, and the explicit inclusion of roll and pitch righting moments. Two maneuvers are simulated: tactical circles and horizontal overshoots. Simulation errors for the circles averaged over all maneuvers for such variables as speed, trajectory components and heading were 5% or less. The horizontal overshoot simulation errors were also 5% or less for the same variables with the exception of the transverse trajectory component. The explanation for the latter deficiency is believed to be the result of the exclusion of wind forces acting on the vehicle, which will be the subject of later work.

Measurement and Modeling of the Motions of a High-Speed Catamaran in Waves

Wetdeck slamming can be defined as a large vertical acceleration event that occurs when ship motions cause an impact between the cross deck and the ocean’s surface. The use of Computational Fluid Dynamics (CFD) and other simulation tools to accurately predict wetdeck slamming loads and ship motions has become the objective of a number of efforts (Hess, et al, 2007; Lin, et al, 2007; Faller et al, 2008; for example). The Sea Fighter, FSF-1, is a high-speed research vessel developed by the U.S. Office of Naval Research (ONR). Christened in 2005, she is an aluminum catamaran propelled by four steerable water jets capable of speeds up to 50 knots. In 2006, Sea Fighter underwent a series of rough water trials to assess its operational profile in high sea states (Fu, et. al., 2007). Along with this assessment, ONR sponsored an effort to obtain full-scale qualitative and quantitative wave slamming and ship motion data. One of these rough water trials took place 18–20 April 2006 as the ship transited from Esquimalt, British Columbia, Canada to San Diego, California, USA. During this trial, the significant wave height ranged from 1.5 to 2.7 m and the ship speed ranged from 20 to 40 knots. This paper describes the results of the effort to characterize the Sea Fighter’s motion in waves. To provide suitable full-scale validation data, the incoming ambient waves had to be characterized. A Light Detecting and Ranging, (LiDAR) system, an array of ultrasonic distance sensors, and several video cameras were used to characterize the incoming wave field. In addition, three fiber optic gyro motion units were deployed to record ship motions. Additionally, a GPS unit was utilized to measure ship speed, pitch, roll, and heading. Several slam and near slam events are discussed over the range of ship’s speed, heading, and sea states tested. Similarities and differences between these events are also noted. Additionally, this data was used to develop a simulation of the Sea Fighter’s motion in waves similar to previous work done utilizing model test data (Hess, et al, 2007; Faller et al, 2008).Copyright

Uncertainty Analysis Applied to Feedforward Neural Networks

Abstract Three problems associated with uncertainty in feedforward neural network predictions are discussed. First, the uncertainty present in the input vector propagates through a trained network into the output vector, and this uncer- tainty is determined using the matrix of partial derivatives defining the change in each output with respect to the inputs. Second, because the partial derivative information conveys the relative sensitivity of a given output to each of the inputs, it can be used as a tool to determine the relevance of each of the inputs to the out prediction. Finally, the influence of random choices for training and testing data sets is investigated. The variability in these solutions provides a measure of the fossilized bias error in the network with respect to development decisions. The approaches are illustrated using examples of four-quadrant propeller predictions.

Real-Time Simulation Based Design Part II: Changes in Hull Geometry

As a second step, and a critical milestone, in the development of a new simulation based design (SBD) tool based on recursive neural network (RNN) technologies the capability to change hull geometries and compute in real-time the new vehicle dynamics has been tested. Previously, this approach was successfully demonstrated for design changes on the ONR Body 1 submarine appendages (sternplane and rudder). The advantages of this RNN based tool are that simulation based design can be performed in a real-time nonlinear simulation (RNS) environment, and this approach enables the fusion of experimental data, when available, with steady Reynolds Averaged Navier-Stokes (RANS) solutions. Building upon the previous work done using RNNs to support submarine simulation based design, the focus of this paper is on the extension of these RNN based approaches, and in particular for the design and modification of hull shape and size, including the computation of non-symmetric hull shapes. As in the previous SBD work on the appendages, a parent training data set for a particular vehicle was used to train an RNN to model how input forces and moments lead to particular output motions. Design changes to the vehicle were then implemented by changing the input force and moment database. As previously shown, appendages can be changed directly by specifying a new geometry and/or lift coefficient. As described herein, the hull geometry can be modified either through the use of other empirical data and/or through the use of steady RANS solutions. The new force and moment database for the hull and the new geometry are used as the input into the RNS based design code to determine the design change impact on vehicle maneuvering. Since only the input force and moment database is changed, no re-training of the RNN is required. As such, the new design simulations can be made in real-time, and the design cycle can, in theory, be shortened significantly. With the results for the hull geometry changes the full utility of this approach can now be defined, and bounds placed on the use of RNN based SBD approaches. This approach resolves the main limitation in RNN technology, namely that it was difficult or impossible to design vehicles using this technology. Now, RNNs can be used not only for vehicle design, but also to determine the result of the design changes in real-time.

Simulation Based Design: A Real-Time Approach Using Recursive Neural Networks

A new simulation based design (SBD) tool has been developed based on recursive neural network (RNN) technologies. This approach permits simulation based design to be performed in a real-time nonlinear simulation (RNS) environment. Further, this approach enables the fusion of experimental data, when available, with steady Reynolds Averaged Navier-Stokes (RANS) solutions. Building upon the extensive work done using RNNs to support Navy submarine simulation and control, a second-generation RNN simulation code has been developed to support submarine simulation based design. Previously, the 1 st generation RNN maneuvering simulation tools were used for the prediction of blind submarine maneuvers in the ONR sponsored Maneuvering Simulation Challenge. A blind maneuver was one for which only the initial conditions and the controls directing the vehicle were provided to the participants. Inputs to the simulation were the controls acting on the vehicle such as propeller rotation speed, rudder and sternplane deflection time histories and the initial conditions. The outputs were time histories of the submarine state variables, the three linear and three angular velocity components. These output data were integrated to recover trajectory and attitude, and differentiated to determine the accelerations acting on the vehicle. Overall, the RNN simulations performed better than any of the other simulation tools including other empirical methods, potential codes and/or vortex tracking methods and unsteady RANS simulations. The focus of this paper is on the extension of these RNN based approaches for use as geometry-to-motion simulation tools. Specifically, a parent training data set for a particular vehicle is used to train an RNN to model how input forces and moments lead to particular output motions. Design changes to the vehicle can then be implemented by changing the input force and moment database. Appendages can be changed directly by specifying a new geometry and/or lift coefficient. The hull geometry can be modified either through the use of other empirical data and/or through the use of steady RANS solutions. The new force and moment database and the new geometry are then used as the input into the RNS based design code to determine the design change impact on vehicle maneuvering. Since only the input force and moment database is changed, no re-training of the RNN is required. As such, the new design simulation can be made in real-time, and the design cycle can, in theory, be shortened significantly. To date, this approach has been used to successfully demonstrate the maneuvering impact of design changes on the ONR Body 1 submarine appendages (sternplane and rudder). This approach resolves the main limitation in RNN technology, namely that it was difficult or impossible to design vehicles using this technology. Now, RNNs can not only be used for vehicle design, but the result of the design changes may be determined in real-time.

Analytic Redundancy for Automatic Control Systems: Recursive Neural Network Based Virtual Sensors

Advanced Automatic Control and Fault Detection systems are being developed for Navy Submarines and Surface Ships. Within this context a critical concern is the vehicle automatic control system response to the environment, damage that may significantly degrade the vehicle performance, as well as unexpected sensor, actuator, or control surface failures. Such an occurrence may well lead to an automatic control system response that is either inadequate or inappropriate given the current state of the vehicle. In order to maintain mission effectiveness, changes in the vehicle dynamics, as well as component failures must be rapidly detected and recovery actions must be promptly initiated within the automatic control loop. The overall system consists of combining three state-of-the-art techniques—Robust/Reconfigurable Control, Recursive Neural Networks (RNNs), and Fault Detection and Isolation (FDI) algorithms—into a real-time system that provides vehicle monitoring, fault detection, and automatic control of the vehicle from within the executive control loop. The combination of these technologies provides an innovative approach for the identification of both discrete and continuous failures, for differentiating between component failure and environmental influences, and for incorporating model-based fault protection into the autonomous control loop. The current paper describes the use of the RNNs as Virtual Sensors (VS) to provide real-time analytic redundancy of the sensor readings provided to the automatic control system. Since sensor failures will feed directly into the automatic control system, it is critical to check and verify the sensor readings prior to utilization in the automatic control loop since all subsequent commands are predicated on the assumption that the sensed information is correct. As such, a virtual sensor system has been developed which provides complete analytic redundancy of the sensor measurements utilized by the automatic control system. Each sensor reading is checked against real-time simulation predictions of the “true” sensor values and a decision is made as to the validity of the measurement. The sensed values are then either passed on as correct or flagged as being in error and a Virtual Sensor estimate of the “true” sensor reading values are provided to the automatic control system. The results show that the typical sensor failure modes, sensor drift, sensor lock-up, sensor drop-out, sensor data spikes, and sensor noise can be detected and corrected analytically using this approach with zero false positives. A key requirement in developing this system was the capability to avoid reporting false positives, sensor problems, when none actually existed. These real-time systems for Advanced Control and Monitoring can mean the difference between safe, continued operation and potentially catastrophic failures.

A Tool to Predict the Four-Quadrant Performance of the Wageningen B-Screw Series for Ship Performance Simulations

Abstract Feedforward neural network predictions of the four-quadrant thrust and torque behavior for the Wageningen B-Screw Series of propellers have been developed. The resulting prediction tool can be used to provide reasonable four-quadrant thrust and torque predictions for propellers, interpolating automatically between measured data.

Feedforward Neural Networks Applied to Problems in Ocean Engineering

The Maneuvering and Control Division of the Naval Surface Warfare Center, Carderock Div. (NSWCCD) along with Applied Simulation Technologies have been developing and applying feedforward neural networks (FFNN) to problems of naval interest in Ocean Engineering. A selection of these will be discussed. Together, they show the power of the nonlinear method as well as its utility in diverse applications. Experimental data describing a subset of the B-Screw series of propellers operating in all four quadrants have been reported by MARIN in the Netherlands. The data contain varying pitch to diameter ratios, expanded area ratios, number of blades and advance angle. These four variables were used to train a FFNN to predict the four-quadrant thrust and torque characteristics for the entire B-screw series over a range of beta from 0 to 360 deg. The results show excellent agreement with the existing data and provide a means for estimating 4-quadrant performance for the entire series. For submarine simulation and design, knowledge of the total forces and moments acting on the hull as a function of angle-of-attack, sideslip angle and dimensionless turning rate across a large parameter space is required. This data is acquired experimentally and/or numerically and can be used to train a FFNN to act as a Virtual Tow Tank or Virtual CFD Code. The network not only recovers the training data but also serves as a very fast, nonlinear six degree-of-freedom look-up table of the forces and moments acting on the hull throughout the parameter space described by the vehicle dynamics. Example solutions demonstrating this approach will be presented. Wave impact loads pose continuing problems for vessels in high sea states, with damage to hatches and appendages, suggesting that these loads may be greater than current design guidelines. Such forcing is complex and often difficult to estimate numerically. Experimental data were acquired at NSWC to measure the hydrodynamic loads of regular, nonbreaking waves on a plate and a cylinder while varying incident wave height, wavelength, wave steepness, plate angle and immersion level of the plate/cylinder. Predictions of wave impact forces from a FFNN trained on the experimental data will be presented.Copyright

Development of an advanced ship simulation & control system using neural networks

Initial efforts in a three-year program to develop an advanced simulation & control system for surface ships are described. The system employs a recursive neural network to simulate the motion of the vehicle in the presence of wind and waves. The faster-than-real-time response of the trained network will permit the use of advanced control techniques such as model-reference control or predictive control and the implementation of path planning for improved performance in the presence of adverse environmental conditions. Early results showing accurate simulation of a U.S. Navy ship conducting overshoot (zig-zag) maneuvers in the presence of wind are shown