Asko Ellman

Fluid Transmission Line Modeling Using a Variational Method

A variational method is used to derive numerical models for transient flow simulation in fluid transmission lines. These are generalizations of models derived using the more traditional modal method. Three different transient compressible laminar pipe flow models are considered (inviscous, one-dimensional linear viscous, and two-dimensional dissipative viscous flow), and a model for transient turbulent pipe flow is given. The (model) equations in the laminar case are given in the form of a set of constant coefficient ordinary differential equations, and for the turbulent case (model) in the form of a set of nonlinear ordinary differential equations. Explicit equations are given for various end conditions. Attenuation factors, similar to the window functions used in spectral analysis, are used to attenuate Gibbs phenomenon oscillations.

A Two Regime Orifice Flow Formula for Numerical Simulation

Because the conventional formula for turbulent orifice flow rate has an infinite derivative when the pressure difference is zero, ODE solvers may fail during numerical simulation of fluid power circuits. To remedy this, a two-regime orifice flow formula is proposed in which an empirical polynomial laminar flow function is used for small pressure differences. The proposed formula has a smooth transition between laminar and turbulent regimes, and its derivative does not have any singularities.

Sliding mode control of a pneumatic muscle actuator system with a PWM strategy

In this paper, a sliding mode control (SMC) strategy is applied to a pulse width modulation (PWM)-driven pneumatic muscle actuator system using high speed on/off solenoid valves. Servo-pneumatic systems with PWM-driven on/off valves can be used instead of expensive servo valves to decrease complexity, weight, and cost of servo-pneumatic systems. Due to the highly nonlinear nature of pneumatics, the system is difficult to model accurately which leads to un-modelled dynamics and uncertainties. In this paper, a robust and nonlinear SMC approach is implemented in order to control the system with sufficient accuracy. A nonlinear model is developed in a single-input single-output form by studying the flow, pressure, and force dynamics of the system. The SMC strategy is applied to three different system configurations: single on/off valve, two on/off valves, and a servo valve. The performance and effectiveness of these configurations are investigated under sinusoidal tracking at different frequencies. The robustness of the controllers is studied by varying the inertia of the system and by applying external disturbances to the system.

Numerical Integration of Fluid Power Circuit Models Using Two-Stage Semi-Implicit Runge-Kutta Methods

Fluid power circuits that contain fluid volumes of different orders of magnitude are difficult to simulate because the system of ordinary differential equations is numerically stiff. Even algorithms specially designed for stiff systems require excessively small time steps to avoid numerical oscillation in simulations of some circuits. In this paper the accuracy and numerical stability of several two-stage semi-implicit Runge-Kutta methods that have been proposed in circuit simulation literature are analysed and compared. It is shown that, for integration of very stiff circuits, the best method in this class is an L-stable method. A simple numerical example is used to verify the theoretical results. The example includes a novel way of modelling orifice flow that is especially suitable for numerical simulations.

Combination of Virtual and Physical Objects in User-Centered Design of a Mobile Work Machine Cabin

User-centred design methods improve the understanding of user work practices and enable construction of customized and user-friendly products. Applying these methods is, however, challenging since the users must to be able to test prototypes which is too time consuming and expensive with real prototypes. This is particularly true in the case of a mobile work machine cabin because the cabin forms an integral part of the machine so that elements need to be prototyped. The main properties of the cabin are the drivers’ visibility, functionality, ergonomics and safety. Virtual environment (VE) offers an effective way to realize prototyping and provides a means to study the drivers’ visual field from the cabin. Today’s design work is already performed using 3D CAD software. Introducing such models in VE is, however, not without its obstacles, since no native CAD format is supported in VE. Employing a general-purpose 3D graphics format usually destroys the model structure and also visualization parameters such as textures and lighting. When the aim is to have users test the functionality of the cabin, the VE model is unsatisfactory because certain physical parts are also required. First the bench is needed to ensure natural posture of the test driver. Second, the steering wheel and pedals are the objects with which the driver most typically interacts. Third, a set of control panels, including gauges and switches, are also often interacted by the driver. This study presents a setup for virtual testing of a mobile work machine cabin as a resource for user-centred design. The study focuses on the importance of physical objects in making the test situation realistic for hands-on professionals. The prototypes are tested by cabin design professionals experienced the use of CAD tools and real prototypes. The aim is to obtain designers’ evaluations and interpretations of different combinations of virtual and physical objects in prototypes. To achieve this a procedure for user-centred design of mobile work machine cabins is presented. More generally, the study discusses the participation of users in the design process employing VE as a design tool.© 2007 ASME

Techniques for Studying a Mobile Hydraulic Crane in Virtual Reality

Abstract Mobile hydraulic applications are exposed to changing environmental conditions and working processes. Furthermore, the fact that those systems consist of mechanical, fluid power and electronic control parts, make the design phase of the product to become complex. In a product development process, system configurations, components selections and parameter optimization must be accomplished in an evaluation-iteration method until fulfilling the performance specification. Replacing real physical prototypes by mathematical models and virtual prototyping in the design process is a major benefit in terms of reducing costs and time in the design phase. This paper introduces a modular method that generates dynamic models for a mobile hydraulic crane and a 3D graphical interface for visualizing of the simulation results in real-time. From the visual feedback provided by the interface, the user interacts with the course of the simulation by driving the crane model with joystick controllers. Such a tool is ideal to be utilised in virtual prototyping, since user can virtually drive and test the prototype and evaluate the system behaviour in real-time. The simulator also allows the user to instantly modify parameters and components of the model. A two degree of freedom hydraulically-driven crane is studied as an example.

Wearable Haptic Device for an IPT System Based on Pneumatic Muscles

The human haptic system has an important role to play in human interaction with Virtual Environments (VEs). Unlike the visual and auditory systems, the haptic sense is capable of both sensing and acting on the environment and is an indispensable part of many human activities. In order to provide the realism needed for effective and compelling applications, VEs need to provide inputs to, and mirror the outputs of, the haptic system. These characteristics are the most important issues in the design of confined spaces and mechanical constructions using 6 DOF input devices in Immersive Projection Technology (IPT) or Cave-like systems. Inputs to the haptic system are in the form of haptic displays and outputs are actuator action commands, where the primary input/output variables are displacements and forces. The idea of total free movement in the IPT system was one of the corner stones of the present study. Therefore the force feedback system should also be wearable. The observer is able to reach objects placed above and below, grip them, and move freely in virtual space. This paper discusses the hardware and software structure of the haptic force feedback system for an IPT system, the theoretical functionality of McKibben actuators and the measured performance of a glove. Force control is based on fast solenoid valves, PID controller and a developed pneumatic muscle model. Muscle actuators are attached to the forearm and control the electronics and valves to a wearable backpack. A developed sensing and force-reflecting exoskeleton applies force to all four fingers and also the thumb. The device has five active DOFs, one for each finger. An ascension motion tracking device is used to track the position and orientation of the forearm. A Fifth Dimensional Technologies fiber optics data glove is used to measure the position of the fingers and provides better information on finger movements. Virtools 4.0 software and VRPN interface was used to connect the data glove, feedback device and PC-cluster.Copyright

Applicability of Pump Models for Varying Operational Conditions

It is commonly known that the characteristics of a fluid power pump depend on pump type, pressure, rotational speed and displacement. But in addition to these, also all the other parameters or factors associated with the operating conditions may have a significant effect on the characteristics. The most important of these are the pump construction and size, operating point temperature and the characteristics of the oil, which also depend on temperature and pressure. The aim of this study is to show the effects that the varying operational conditions have on the characteristics of a axial piston pump, to compare the measured characteristics with other published characteristics of axial piston pumps and to study the capability of pump models to represent these characteristics. The results include information of the effects of fluid temperature, type of fluid and the setting value of the displacement on the pump characteristics along with the effects of pressure and rotational speed. The sensitivity of the pump to each of the parameters is discussed. The effect of limited information of pump characteristics on the reliability of simulation results is studied using the Schlosser models.© 2003 ASME

Effect of the Immersion Level of a Virtual Loader Simulator on the Sense of Presence

In mobile machine industry, work machine simulators can be used to save time and costs in product development and training. However, investing into expensive or difficult-to-use technology should be avoided if it does not affect measurable user performance or feeling of presence. At Tampere University of Technology, a virtual loader simulator has been constructed in an three-wall walk-in virtual environment. The goal of this research was to investigate how the immersivity of the simulator affects testers’ feeling of presence. A total of 25 test users performed two test runs each with different simulator setups, and the presence level of each user was evaluated by a questionnaire after both runs. The results indicate that switching from 2D view to 3D stereoscopic display did not significantly affect the sense of presence. However, with the motion platform, the test users reported clearly higher presence ratings than without it. In addition, we examined some effects of the users’ backgrounds: Test users who had experience from driving large work machines reported lower presence ratings than those without such experience. The sense of presence for frequent computer players was higher than for non-players, but only in one simulator setup.Copyright

The Effect of Parameter Uncertainty on the Reliability of Pressure Accumulator Simulations

The aim of this study is to demonstrate how the various parameters and the uncertainty associated with them affect the simulation results of a pressure accumulator. The parameters to be studied are related mainly to the pre-charging procedure of a membrane accumulator and cover the constants for the pressure and temperature of the nitrogen gas and the efficient volume of accumulator. These parameters are included in the non-linear model which is suited especially for large amplitude and low frequency transients. The background of the work is related to the fact that simulation is already an important tool in product development work. To be able to design the control of the system and predict the performance of it an estimate of the accuracy of the calculations is needed. The usability of simulation is determined by the fact how reliable the information is. To gain full benefit of simulation more attention has to be paid to the validity of the models, the accuracy of the parameter values needed in the models and the sensitivity of these parameters. The most sensitive parameters have to be recognized and paid special attention to the accuracy of the values given to them. Also the changes of these values in time due to wear or other modification in the system have to be noticed. Numerous factors have an influence on the accumulator operation. The most important of these are the ratio of the operation pressure to the pre-charge pressure, the amplitude of the flow rate disturbances, the temperature of the hydraulic fluid and the gas, the viscosity of the hydraulic fluid, the thermodynamic process of the nitrogen gas in the accumulator and the changes in the speed of sound. Also the mounting of the accumulator, the fittings used and the connecting pipes may have a significant role in the dynamics of an accumulator. The methods used in the study include measurements in time domain, modeling, simulations, and analytical work. The accumulator dynamics may alter due to remarkable changes in operating points like pressure levels. These are possible because of large amplitudes of flow rates particularly at the low frequency area where the presented study is focused on. The results include model analysis and information of the importance of the most fundamental parameters of the models and suggestions for future research work.Copyright