MechChem Africa November-December 2024

MCA meets Dustin Schiller and Simon Hausknecht, CEOs of SHG Conveyor Control of Germany, the developers of the AI-based Rip Prevent+ conveyor monitoring system now available in South Africa from Tru-Trac Rollers. Tru-Trac adopts Rip Prevent+ conveyor monitoring

“We have developed the Rip Prevent+ sys tem based mostly on analysing the electri cal and performance characteristics of the drive pulley motor: the speed, the voltage, the amps, and the ambient temperature, amongst many others. All of this can be done by adding a small and inexpensive data monitoring unit near the motor controller,” he adds. SHG has developed software that en ables the system to be ‘tuned’ to the specific parameters of each conveyor system: its drum diameter, maximum power, torque and belt length, for example. “We are con tinuously monitoring a lot of data, mostly the electrical parameters. The mix of the data starting from the voltage, amps, Total Harmonic Distortion (THD), Cos Phi and other collected data gives us a specific signature. Other data such as associated the K-factor, the real, apparent and reac tive power, and so on and so forth is also collected and analysed,” continues Simon

Simon Hausknecht (left) and Dustin Schiller (right), developers of the AI-based Rip Prevent+ conveyor monitoring system.

S HG Conveyor Control was started about 18 months ago by Dustin Schiller, a mechanical engineer with experience working in the convey ing industry at Continental, and Simon Hausknecht, an electrical engineer. “We came up with an idea to find a way to moni tor the health of a conveyor without having to use sensitive sensor elements inside belts or remotely mounted across the length of a conveyor,” begins Schiller. They began by looking into rip detection events: “We contacted conveyor belt us ers and asked them to share the historical data they had collected from their systems, particularly for periods that included rip events. We then set out to use AI to analyse this data, collected from traditional condi tion monitoring systems, to see if we could see any early prediction signals for the rip events that had occurred,” he tells MCA . That was our starting point for the devel opment – and it was very successful. Traditional conveyor belt monitoring involves embedding wire loops into the belt that create eddy currents. It is only when the loops get broken that the rip is detected, which often means that the rip has already happened. So it is not usually a prevention system, it can only detect the rip after the event. “Our system is completely new, be cause it does not include any external sensor elements on the belt and we set out to find the root cause of the rip and to stop the belt before the rip occurs,” he continues. “So we did some reverse engineering on a rip event in the loading area of a belt and we quickly found a ‘signature’ pattern in the

data that was uniquely linked to the start of that event,” Dustin Schiller tells MCA , add ing that this signal was not linked to any of the installed sensors on the system.

By predicting the onset of rip events, the innovative Rip Prevent+ system will have a major impact on the reliability and uptime on any operation that uses conveyor belts.

6 ¦ MechChem Africa • November-December 2024