Weigh-in-motion (WIM) systems are devices which measure the loads imposed by the axles of vehicles travelling at normal highway speeds. The present generation of WIM equipment has limited accuracy when used to estimate static axle weights. The main limiting factor is the bouncing of the vehicles on their tyres and suspensions as they travel at speed. To reduce these errors, multiple-sensor arrays can be designed to 'filter out' the dynamic component of the axle's load profile. A genetic algorithm technique has been used to design optimal arrays of sensors. These were designed to cope with a range of suspensions and vehicle speeds. (It was assumed that the sensors were perfectly accurate when measuring the instantaneous vertical load imposed by the axle). The arrays were then tested against experimental data. The results showed that a good design is to use a symmetrical eight sensor array about 14 metres long and that there was little advantage in using more than eight sensors.

Want to know more about this project?