MVI56-LTQ

The data collected by the MVI56-LTQ is used to train a neural network so that any deviation from “good” machine behaviour and performance can be detected and predicted; automatically generating PC programs suitable for subsequent online performance monitoring and diagnostic purposes. As part of the software development process, NUM beta-tested the technology on a CNC milling machine equipped with three axes and a spindle, which required a neural network consisting of 36 neurons and three hidden layers. In this particular instance, the values of 396 parameters needed to be accurately defined; this required the acquisition of more than 2 million “known good” data points and 300 iterations of the teaching phase, which took approximately four hours per axis.

The MVI56-LTQ also supports the following features:
Freeze images to stop refreshing and allow detailed analysis of the last capture
Zooming and moving the image in the visual controls (by touch gestures, mouse input or direct input of specific values) to view image details more precisely
Display of toolbars with directly usable control elements (e.g. for selecting an image, zooming, creating shapes, freezing the image refresh and downloading the displayed image)
Display an information bar showing current details and values such as image size, pixel coordinates, colour values and shape data
Draw shapes (points, lines, rectangles, ellipses, and polygons) with modifiable positions and sizes for determining size, area, and coordinates, setting areas of interest, etc.
Display of shapes (crosses, rectangles and circles) or image overlays for setting up and positioning cameras and artefacts