TruckStop

Abstract

A multibody system (MBS) model of a scaled experimental attenuator is being developed, the aim being to design a new kind of truck-mounted attenuator (TMA) made of recycled materials. In this work, we implement the elasto-plastic behavior of the recycled material modules as a constitutive law of a MBS model which also includes the impacting vehicle, in order to simulate the collision.
The simulation results aim at predicting the potential risk for the occupants, referring to the Acceleration Severity Index (ASI), and for the road workers by considering the truck displacement. The envisaged collisions consider a vehicle weighting from 1 ton (at 100 km/h) up to 13 tons (at 70 km/h).

Motivation

According to the National census of fatal occupational injuries in 2008 (U.S Departmentof Labor) there were 345 workers struck by a vehicle for the year 2007, which corresponds to more than 2.5% of the global number of fatal occupational injuries. For the workers, especially when they are working on highways, a truck-mounted attenuator (TMA) is systematically used to prevent from the fatal accident of inattentive drivers approaching the roadwork at high speed.

Background

One of the key points when designing a TMA is its shape and the dynamic characteristics of the absorbing material (i.e. to absorb the energy of the collision). In this context, theArGEnCo Laboratory of the Université de Liège has developed a novel concept based on recycled materials: their TMA modules are composed of metallic cans issuing from “kitchen waste" for which a particular attention is paid to their pre-compression state.
Preliminary tests on a scaled test bench, consisting of a 600 kg mass mounted on wheel sets impacting a fixed attenuator module (shape: cubic, side: 40 cm), provide encouraging results.
In order to extrapolate those to the actual TMA, a dynamic model of the test-bench and of the real TMA is thus required. On the basis of the experimental characteristics of the attenuator modules, a multibody model has been developed by the Center for Research in Mechatronics (CEREM) of the Université catholique de Louvain.

Methodology

  • The first step of this project focused on the implementation of the recycled material properties, measured by the ArGEnCo Laboratory, by considering the elastic loading force, the plastic loading force, the elastic unloading force, the dissipated energy and the elastic-to-plastic transition.
  • Using a multibody,  one can question about the number of bodies to consider per TMA module for an axial collision. By comparing models involving 1, 3 and 10 bodies respectively, there are discrepancies during the collision (oscillations) but not in terms of maximum deceleration of the impacting mass (600 kg): a good correspondence between the three models is observed and also with the experimental results. Also, comparing those results in terms of the Acceleration Severity Index (ASI), which is the reference of our calculations, shows differences that are lower than 0.4% between the different models, and 7.4% with experiments which is rather acceptable.
  • The above point shows that results are not really affected by the number of bodies per attenuator module: this led us to restrict it simply to one. However, since there are actual difficulties to couple serial nonlinear stiffness in mechanical models , all the more in the present case with the discontinuity caused by the loading/unloading phenomenon, we propose the following model formulation. In the figure, a three-module MBS is considered, each of them (dotted rectangle) having its own stiffness and mass. Tocircumvent the above- mentioned serial stiffness problem, intermediate masses (20% of the module masses) are inserted and connected to the others as coupling masses via forces elements according to the law de
    picted in Fig. 1. To check the pertinence of this modeling technique, it was measured a low discrepancy, less than 0.13%, between results obtained with 20% and 40% transition masses and 11.3% with experiments, which is rather encouraging.
  • This scaled MBS model has been generalized to simulate a longitudinal impact between a car and a real TMA carried by a simplified truck, i.e. a rigid block having one degree of freedom in the longitudinal direction and subject to a longitudinal friction force (brake and/or rolling resistance). The attenuator model is inspired from the one described in point 3 and the car, being more rigid than the attenuator modules, is, presently, modeled as a rigid MBS, with articulated suspension and wheels.
  • The performance of the attenuator was evaluated by considering the following criteria: the safety of the driver, computed via the ASI Criteria and the safety of road workers, estimated via the total displacement of the truck. To optimize the TMA, the module pre-compression has been considered as a parameter, being the ASI the cost function to minimize.

Why multibody?

With respect to a FEM approach, which would be rather unfeasible here because the cans deformation can only be measured globally at the level of the modules, the MBS approach, while being a more “lumped” technique, gives satisfactory results; moreover, without requiring lengthy simulations. According to  a car-TMA collision simulation of 250ms takes approximately 8 days to run. Using Robotran which generates symbolic MBS equations, it is possible to simulate the collision in less than 80 sec. To optimize the TMA, it was required several simulations (more than 100), which shows the importance of high speed simulations.

 

Collaborations

IMMC/CEREM Center for Research in Mechatronics, Université catholique de Louvain (UCL), Louvain-la-Neuve, Belgium

ArGEnCo Institut de Mécanique et Génie Civil, Université de Liège, Liège, Belgium

 

copyright © 2011 CEREM - www.cerem.be - This e-mail address is being protected from spambots. You need JavaScript enabled to view it.