Lars Vabbersgaard Andersen

A new energy optimizing control strategy for switched reluctance motors

This paper describes a new and machine-independent method to minimize the energy consumption of a speed controlled switched reluctance motor (SRM). The control strategy is to vary the duty cycle of the applied DC voltage in order to obtain the desired speed quickly and when operating in steady-state vary the turn-on angle (a/sub on/) of the phase voltage to minimize the energy consumption. The power flow is measured in the DC-link and used to control the turn-on angle. Simulations carried out on a three-phase 6/4 pole SRM justify the algorithm and the physical implementation in a Siemens SAB 80C517A microcontroller is described. Measurements on two different load systems show it is possible to minimize the energy consumption on-line in a speed controlled switched reluctance motor without losing the dynamic performance. A comparison with an ordinary mode-shift controlled SRM shows more than an 8% increase in overall efficiency for some operation points. The algorithm is fully applicable to other switched reluctance motors at other power levels or with other pole configurations. >

Boundary Element Analysis of the Steady-state Response of an Elastic Half-Space to a Moving Force on its Surface

The paper deals with the boundary element method formulation of the steady-state wave propagation through elastic media due to a source moving with constant velocity. The Greens function for the three-dimensional full-space is formulated in a local frame of reference following the source. This is appropriate for the analysis of, for example, ground borne noise from railway or road traffic. The frequency-domain fundamental solution is derived from the corresponding time-domain solution by means of Fourier transformation. To obtain a closed form solution, a part of the time-domain kernel functions is approximated, but the error which is introduced in this way is insignificant. Numerical examples are given for a moving rectangular load on an elastic half-space. The results from a boundary element code based on the derived Greens function are compared with a semi-analytic solution.

Vibrations of a Track Caused by Variation of the Foundation Stiffness

Abstract The paper deals with the stochastic analysis of a single-degree-of-freedom vehicle moving at constant velocity along a simple track structure with randomly varying support stiffness. The track is modelled as an infinite Bernoulli–Euler beam resting on a Kelvin foundation, which has been modified by the introduction of a shear layer. The vertical spring stiffness in the support is assumed to be a stochastic homogeneous field consisting of a small random variation around a deterministic mean value. First, the equations of motion for the vehicle and beam are formulated in a moving frame of reference following the vehicle. Next, a first-order perturbation method is proposed to establish the relationship between the variation of the spring stiffness and the responses of the vehicle mass and the beam. Numerical examples are given for various parameters of the track. The response spectra obtained from the perturbation analysis are compared with the numerical solution, in which finite elements with transparent boundary conditions are used. The circumstances, under which the first-order perturbation approach provides satisfactory results, are discussed.

Vehicle Moving Along an Infinite Beam With Random Surface Irregularities on a Kelvin Foundation

The paper deals with the stochastic analysis of a single-degree-of-freedom vehicle moving at a constant velocity along an infinite Bernoulli-Euler beam with surface irregularities supported by a Kelvin foundation. Both the Bernoulli-Euler beam and the Kelvin foundation are assumed to be constant and deterministic. This also applies to the mass, spring stiffness, and damping coefficient of the vehicle. At first the equations of motion for the vehicle and beam are formulated in a coordinate system following the vehicle. The frequency response functions for the displacement of the vehicle and beam are determined for harmonically varying surface irregularities. Next, the surface irregularities are modeled as a random process. The variance response of the mass of the vehicle as well as the displacement variance of the beam under the oscillator are determined in terms of the autospectrum of the surface irregularities. ©2002 ASME

Numerical and Experimental Investigation of Stop-Bands in Finite and Infinite Periodic One-Dimensional Structures

Adding periodicity to structures leads to wavemode interaction, which generates pass- and stop-bands. The frequencies at which stop-bands occur are related to the periodic nature of the structure. Thus structural periodicity can be shaped in order to design vibro-acoustic filters for reducing vibration and noise transmission. The aim of this paper is to investigate, numerically and experimentally, stop-bands in periodic one-dimensional structures. Two methods for predicting stop-bands are described: the first method applies to infinite periodic structures using a wave approach; the second method deals with the evaluation of a vibration level difference (VLD) in a finite periodic structure embedded within an infinite one-dimensional waveguide. This VLD is defined to predict the performance in terms of noise and vibration insulation of periodic cells embedded in an otherwise uniform structure. Numerical examples are presented, and results are discussed and validated experimentally. Very good agreement between the numerical and experimental models in terms of stop-bands is shown. In particular, the results show that the stop-bands obtained using a wave approach (applied to a single cell of the structure) predict those obtained from the VLD of the corresponding finite periodic structure.

Probabilistic finite element stiffness of a laterally loaded monopile based on an improved asymptotic sampling method

AbstractThe mechanical responses of an offshore monopile foundation mounted in over-consolidated clay are calculated by employing a stochastic approach where a nonlinear p–y curve is incorporated with a finite element scheme. The random field theory is applied to represent a spatial variation for undrained shear strength of clay. Normal and Sobol sampling are employed to provide the asymptotic sampling method to generate the probability distribution of the foundation stiffnesses. Monte Carlo simulation is used as a benchmark. Asymptotic sampling accompanied with Sobol quasi random sampling demonstrates an efficient method for estimating the probability distribution of stiffnesses for the offshore monopile foundation.

Flooring-Systems and Their Interaction with Usage of the Floor

Some flooring-system designs might be sensitive to their vibrational performance, as there might be the risk that serviceability-limit-state problems may be encountered. For evaluating the vibrational performance of the flooring-system at the design stage, decisions need to be made by the engineer in charge of computations. On a flooring-system often passive humans and/or furniture are present. Often these masses and their way of interacting with the floor mass are ignored in predictions of vibrational behavior of the flooring-system. The paper explores and quantifies how these masses can influence central parameters describing the dynamic behavior of the flooring-system.

Robust and efficient handling of yield surface discontinuities in elasto-plastic finite element calculations

Purpose – The purpose of this paper is to present several methods on how to deal with yield surface discontinuities. The explicit formulations, first presented by Koiter (1953), result in multisingular constitutive matrices which can cause numerical problems in elasto-plastic finite element calculations. These problems, however, are not documented in previous literature. In this paper an amendment to the Koiter formulation of the constitutive matrices for stress points located on discontinuities is proposed. Design/methodology/approach – First, a review of existing methods of handling yield surface discontinuities is given. Examples of the numerical problems of the methods are presented. Next, an augmentation of the existing methods is proposed and its robustness is demonstrated through footing bearing capacity calculations that are usually considered “hard”. Findings – Previous studies documented in the literature all present “easy” calculation examples, e.g. low friction angles and few elements. The ame...

Effect of Spatial Correlation Length on the Interpretation of Normalized CPT Data Using a Kriging Approach

AbstractIn geotechnical engineering analysis and design, the frequency and spacing of borehole information is of great interest, especially when field data are limited. This paper uses random field models to deal with uncertainty in soil properties owing to spatial variability, by analyzing in-situ cone penetration test (CPT) data from a sandy site in northern Denmark. To provide a best estimate of properties between observation points in the random field, a Kriging interpolation approach has been applied. As expected, for small correlation lengths, the estimated field quantities at intermediate locations between data points are close to the mean value of the measured results, and a high uncertainty is associated with the estimate. A longer correlation length reduces the error and implies more variation in the estimated values between the data points.

Non-structural Masses and Their Influence on Floor Natural Frequencies

Excessive floor vibrations are problematic and may potentially render a floor unfit for its intended use. A design-stage check of vibrational performance of a floor design would encompass design-stage estimates of floor dynamic characteristics such as floor natural frequencies. Non-structural masses such as furniture might be present on the in-service floor. For a prediction of floor dynamic characteristics it is not common to account for the fact that non-structural masses elevated above the floor plane may contribute with inertial energy as a result of their horizontal motion occurring during vertical floor vibration. The paper addresses this subject by setting up a finite element model for the floor, which also accounts for an elevation of the non-structural masses. It is shown how different configurations of non-structural masses influence floor natural frequencies. For the investigations, the elevations and weights of the masses are modelled as random variables and Monte Carlo simulations are used for setting up the random configurations of non-structural masses across the floor area.