No abstract

No abstract

The optimization of roadway safety design by experimental means is expensive and time consuming. Computer simulation of such complex engineering systems improves understanding of how and why the system behaves as it does, aids in decision making, and reduces development costs and time involved. The simulation presented is based on a computer model developed from a study of the results of full-scale experiments of impact on the Brifen wire-rope safety fence (WRSF). The code comprises a dynamic vehicle model and a quasi-static fence model interacting in time through the important collapse mechanisms of the system. The principles governing them are described and their inclusion is validated by demonstrating good correlation between the predictions of the simulation and the experimental test data. Sensitivity studies show that the performance of a WRSF is particularly sensitive to the impact conditions of vehicle speed and angle and the design parameters offence height, post spacing, post strength, and rope pre-tension. The sensitivity work is extended to show that for fences installed with a low rope pre-tension, performance may not be significantly impaired if rope pre-tension is not maintained. However, significant gains in fence performance may be made should a fence be installed and maintained with a high rope pre-tension. The use of the simulation in assessing cost-effectiveness of alternative designs in achieving a target performance is also demonstrated.

A computer simulation has been developed at The University of Sheffield that predicts the impact performance of roadway wire rope safety fences. Effective use of this simulation for design, or other purposes, requires it to be validated by showing that the predicted vehicle motion and fence response closely represent that of the real system. Two distinct validation studies are described in the current paper. The first compares the major results of eight impact tests undertaken at the UK Motor Industry Research Association and a further four impact tests undertaken at the Swedish Road and Transport Research Institute with the simulation predictions. The close prediction of the test results validates the use of the simulation as a design tool for a range of fence designs and impact conditions. The second validation study compares, in detail, the test results and simulation prediction for a single test. This suggests that, if required, results may be improved further by considering suspension effects in the vehicle model, and by a better understanding of frictional effects at the rope-post and the rope-vehicle contacts.