Reid and Gurgenci – under the ACARP, CRCMining-funded longwall shearer cutting force estimation project – have put together a methodology for estimating the cutting forces acting on an operation shearer.

“The ability to estimate cutting forces turns the shearer into a sensor capable of monitoring the local cutting environment, making it possible to detect changes in the seam [for example, track features of the seam – important for the problem of shearer horizon control], in addition to keeping tabs on the operation and workload of the shearer itself,” the researchers said.

“A key benefit to knowing the cutting forces, rather than just the torques of the cutterhead motors, is that they provide information regarding force direction, and not just the magnitude of the force/torque required to excavate coal.”

To measure cutting force a Kalman filter (where information is blended from measurements with predictions from a dynamic model of the shearer) was adapted using the method of state augmentation to estimate six external forces acting on the longwall shearer.

The estimated cutting forces were employed to track the vertical motion of the shearer relative to the seam by comparing a force profile generated over complete rotations of the drum to a baseline for the shearer on its desired horizon.

“As an application for the shearer cutting forces, we introduced the problem of tracking a feature of the seam as the shearer advanced across the face. This is an important problem for the longwall shearer because it enables the shearer to locate itself vertically in the seam, and opens up the possibility of autonomous horizon control,” Reid and Gurgenci said.

A high-fidelity model of the shearer representing the plant and the coal/cutterhead interaction was used to generate data to demonstrate the concept. Force estimates were generated from four simulated cutting trials of both plain coal and coal with an included stone band.

“It was shown that the cutting force estimates in the absence of the stone band were unbiased and had standard deviations of less than 5 percent of the mean force level,” they said.

“When the leading cutter drum began to cut through the modelled 5cm stone band, a bias developed in the cutting force estimates associated with it. The force estimate standard deviations also increased.

“The force estimates were then applied to the problem of tracking the embedded stone band as it moved relative to the drum. The stone band shift was clearly evident for two distinct vertical shifts of the stone band.”

CRCMining plans to seek further funding to prove the application on an operating face, and to investigate the possibility of using this approach to provide real-time data on cutting drum pick condition and the face stress state.