As part of the proving trials of the GRANIT system, a number of anchorages were installed at a field site near to AMEC plc's headquarters in Swynnerton, England.
The goal of these tests was to prove the GRANIT equipment over intensive field use, and for the successful diagnosis of load for a number of anchorages from the same trained neural network.
The anchorages were installed in hard concrete and possessed the following characteristics:
- Hole diameter - 32mm diamond drilled
- Bolt type - M24 by 2400mm by 150mm threaded
- Slow set resin used
- Standard bearing plate
- Free length from 110mm to 400mm
Following the success of the GRANIT tests on the first six installed anchorages, a further three were installed with similar characteristics at the same site. Further data was taken, and again success was found with the diagnosis of their load with the GRANIT system.
Further blind tests were taken from all eight anchorages (anchorage number 1 was inaccessible), and the GRANIT system again achieved a high level of accuracy in the diagnosis of their load for a number of different load ranges.
It was decided at this point to move to a different location, and install similar anchorages with the same dimensions as above. These anchorages were installed in soft sandstone at the Kibblestone site, and had the same characteristics as the Swynnerton site, although the free lengths here ranged from 945mm to 295mm.
In spite of the obvious differences between the two sites, and the minor differences between the anchorages installed, success was again achieved in the diagnosis of their load. This diagnosis was performed using the same neural network that had been trained on the Swynnerton site, and constitutes a major step forward for the GRANIT technique, showing its flexibility and high level of accuracy.