Prediction of the Hydrodynamic Parameters in the Coupled Heave and Pitch Motion Equations for Underwater Robotic Vehicles Using Measured Responses at Sea

2001 ◽  
Vol 123 (3) ◽  
pp. 93-102 ◽  
Author(s):  
Ayman B. Mahfouz ◽  
Mahmoud R. Haddara ◽  
Christopher D. Williams
Keyword(s):  
Neural Networks ◽  
Parametric Identification ◽  
Computer Time ◽  
Random Waves ◽  
Pitch Motion ◽  
Hydrodynamic Parameters ◽  
Identification Technique ◽  
Coupling Parameters ◽  
Robotic Vehicle ◽  
Identification Techniques

A method for the identification of the damping, restoring, and coupling parameters in the equations describing the coupled heave and pitch motions for an underwater robotic vehicle (URV) sailing near sea surface in random waves using only its measured responses at sea is presented. The random decrement equations are derived for the URV performing coupled heave and pitch motions in random waves. The hydrodynamic parameters in these equations are identified using a new identification technique called RDLRNNT, which uses a combination of a multiple linear regression algorithm and a neural networks technique. The combination of the classical parametric identification techniques and the neural networks technique provides robust results and does not require a large amount of computer time. The developed identification technique would be particularly useful in identifying the parameters for both moderately and lightly damped motions under the action of unknown excitations effected by a realistic sea. Numerically generated data for the coupled heave and pitch motion of a URV are used initially to test the accuracy of the technique. Experimental data are also used to validate the identification technique. It is shown that the developed technique is reliable in the identification of the parameters in the equations describing the coupled heave and pitch motions for an URV.

NON-PARAMETRIC IDENTIFICATION TECHNIQUES FOR INTELLIGENT PNEUMATIC ACTUATOR

2015 ◽  
Vol 77 (20) ◽  
Author(s):  
Abdulrahman A. A. Emhemed ◽  
Rosbi Mamat ◽  
Ahmad ‘Athif Mohd Faudzi
Keyword(s):  
System Identification ◽  
Parametric Identification ◽  
Pneumatic Actuator ◽  
Step Response ◽  
Model Parameters ◽  
Force Model ◽  
Force System ◽  
Identification Technique ◽  
Identification Techniques ◽  
Non Parametric

The aim of this paper is to present experimental, empirical and analytic identification techniques, known as non-parametric techniques. Poor dynamics and high nonlinearities are parts of the difficulties in the control of pneumatic actuator functions, which make the identification technique very challenging. Firstly, the step response experimental data is collected to obtain real-time force model of the intelligent pneumatic actuator (IPA). The IPA plant and Personal Computer (PC) communicate through Data Acquisition (DAQ) card over MATLAB software. The second method is approximating the process by curve reaction of a first-order plus delay process, and the third method uses the equivalent n order process with PTn model parameters. The obtained results have been compared with the previous study, achieved based on force system identification of IPA obtained by the (Auto-Regressive model with eXogenous) ARX model. The models developed using non-parameters identification techniques have good responses and their responses are close to the model identified using the ARX system identification model. The controller approved the success of the identification technique with good performance. This means the Non-Parametric techniques are strongly recommended, suitable, and feasible to use to analyze and design the force controller of IPA system. The techniques are thus very suitable to identify the real IPA plant and achieve widespread industrial acceptance.

Identification of ’Effective’ Linear Joints Using Coupling and Joint Identification Techniques

1998 ◽  
Vol 120 (2) ◽  
pp. 331-338 ◽  
Author(s):  
Y. Ren ◽  
C. F. Beards
Keyword(s):  
Dynamic Properties ◽  
Real Life ◽  
Model Parameters ◽  
Coupling Technique ◽  
Alternative Approach ◽  
Identification Technique ◽  
Joint Identification ◽  
Stiff Joint ◽  
Almost All ◽  
Identification Techniques

Almost all real-life structures are assembled from components connected by various types of joints. Unlike many other parts, the dynamic properties of a joint are difficult to model analytically. An alternative approach for establishing a theoretical model of a joint is to extract the model parameters from experimental data using joint identification techniques. The accuracy of the identification is significantly affected by the properties of the joints themselves. If a joint is stiff, its properties are often difficult to identify accurately. This is because the responses at both ends of the joint are linearly-dependent. To make things worse, the existence of a stiff joint can also affect the accuracy of identification of other effective joints (the term “effective joints” in this paper refers to those joints which otherwise can be identified accurately). This problem is tackled by coupling these stiff joints using a generalized coupling technique, and then the properties of the remaining joints are identified using a joint identification technique. The accuracy of the joint identification can usually be improved by using this approach. Both numerically simulated and experimental results are presented.

Prediction of Hydrodynamic Forces and Moments on Submarines Using Neural Networks

2002 ◽  
Author(s):  
Ibrahim Mohamed ◽  
Mahmoud Haddara ◽  
Christopher D. Williams ◽  
Michael Mackay
Keyword(s):  
Neural Network ◽  
Neural Networks ◽  
Hydrodynamic Model ◽  
Parametric Identification ◽  
Hydrodynamic Forces ◽  
Degree Of Freedom ◽  
Experimental Time ◽  
Time Histories ◽  
Trained Network ◽  
Identification Tool

This paper describes a parametric identification tool for predicting the hydrodynamic forces acting on a submarine model using its motion history. The tool uses a neural network to identify the hydrodynamic forces and moments; the network was trained with data obtained from multi-degree-of-freedom captive maneuvering tests. The characteristics of the trained network are demonstrated through reconstruction of the force and moment time histories. This technique has the potential to reduce experimental time and cost by enabling a full hydrodynamic model of the vehicle to be obtained from a relatively limited number of test maneuvers.

Simulation of the electroerosion treatment process using neural networks

2022 ◽  
Vol 14 (4) ◽  
pp. 5-12
Author(s):  
Ol'ga Ermilina ◽  
Elena Aksenova ◽  
Anatoliy Semenov
Keyword(s):  
Neural Network ◽  
Neural Networks ◽  
Equivalent Circuit ◽  
Electrical Discharge ◽  
Electrical Discharge Machining ◽  
Calculated Data ◽  
Parametric Identification ◽  
Identification Problem ◽  
Rlc Circuit ◽  
Matlab Package

The paper provides formalization and construction of a model of the process of electrical discharge machining. When describing the process, a T-shaped equivalent circuit containing an RLC circuit was used. Determine the transfer function of the proposed substitution scheme. Also, a task is formulated and an algorithm for neural network parametric identification of a T-shaped equivalent circuit is proposed. The problem is posed and an algorithm is developed for neural network parametric identification of the equivalent circuit with a computational experiment, the formation of training samples on its basis, and the subsequent training of dynamic and static neural networks used in the identification problem. The process was simulated in Simulink, Matlab package. Acceptable coincidence of the calculated data with the experimental ones showed that the proposed model of electrical discharge machining reflects real electromagnetic processes occurring in the interelectrode gap.

scholarly journals Identification of Nonlinear Vibrating Structures: Part I—Formulation

1987 ◽  
Vol 54 (4) ◽  
pp. 918-922 ◽  
Author(s):  
S. F. Masri ◽  
R. K. Miller ◽  
A. F. Saud ◽  
T. K. Caughey
Keyword(s):  
Time Domain ◽  
Parametric Identification ◽  
Companion Paper ◽  
Sufficient Accuracy ◽  
Recursive Least Squares ◽  
Estimation Methods ◽  
Free Oscillations ◽  
Vibrating Structures ◽  
Parameter Estimation Methods ◽  
Identification Techniques

A self-starting multistage, time-domain procedure is presented for the identification of nonlinear, multi-degree-of-freedom systems undergoing free oscillations or subjected to arbitrary direct force excitations and/or nonuniform support motions. Recursive least-squares parameter estimation methods combined with non-parametric identification techniques are used to represent, with sufficient accuracy, the identified system in a form that allows the convenient prediction of its transient response under excitations that differ from the test signals. The utility of this procedure is demonstrated in a companion paper.

Based on Walsh packet parametric identification technique for dynamic systems

PAMM ◽  
2005 ◽  
Vol 5 (1) ◽  
pp. 777-778
Author(s):  
Magdalena Napiorkowska-Alykow ◽  
Wojciech Glabisz
Keyword(s):  
Dynamic Systems ◽  
Parametric Identification ◽  
Identification Technique

scholarly journals Non-Linear Generalized Maxwell Model for Dynamic Characterization of Viscoelastic Components and Parametric Identification Techniques

2012 ◽  
Author(s):  
Hanen Jrad ◽  
Jean Luc Dion ◽  
Franck Renaud ◽  
Imad Tawfiq ◽  
Mohamed Haddar
Keyword(s):  
Dynamic Stiffness ◽  
Maxwell Model ◽  
Parametric Identification ◽  
Mechanical Analysis ◽  
Dynamical Behaviour ◽  
Static Displacement ◽  
Generalized Maxwell Model ◽  
Non Linear ◽  
Identification Techniques

Viscoelastic components are incorporated into automobile and aerospace structures system in order to damp mechanical vibrations. Viscoelastic components are a key element in designing desired dynamic behaviour of mechanical systems. Viscoelastic components dynamic characteristics are often very complex, due to the dependence of its response on several variables, such as frequency, amplitude, preload, and temperature. These dependencies can be critical in capturing the mechanical proprieties and so non linear dynamical behaviour may appear. Assuming that non linearities are due to non linear elasticity, the non linear Generalized Maxwell Model (GMM) is proposed to characterize dynamics of viscoelastic components. Parameters of GMM are identified from Dynamic Mechanical Analysis (DMA) tests for different excitation frequencies. A particular result from identification is that the non linear stiffness is dependent upon displacement amplitude and static displacement under static preload. The significance of this result is that the non linear dynamics of the viscoelastic component can be represented by a simple analytical model capable to produce accurate results. Comparison between measurements and simulations of dynamic stiffness of viscoelastic component has been carried on.

Sign in / Sign up

Export Citation Format

Share Document