In the manufactering of electronically commutated electric motors for the automotive
industry during Industry 4.0, 100-% traceability of products and their manufacturing
parameters is required. It is necessary to ensure distrubuted quality control during the
macufacturing process to eliminate defective parts as soon as possible, thus reducing the cost
of further processing the parts or even complaits from customers in the event of defective
parts not being eliminated during the final inspection because of manufacturing errors not
being detected. The diploma thesis describes the current state of the quality control location
for the phase contacts on the stator assembly line. Several errors are selected from the error
catalog, which is updated by the company's quality control department. This types of errors
currently occur, but are not covered by the quality control at this location. A suitable camera
is selected according to the application, and a new adjustable camera mount is made. Two
different types of stators are being manufactured on the production line, so adjustable
position of the cameras is necessary when changing the type of stator being produced.
Outside the production line, a test site has been prepared, where we installed the cameras
and the structural configuration accoring to the requirements and the conditions inside the
production line were imitated. Two types of programs have been created and set up. After
testing, a program with machine learning is selected, and the results are presented in more
detail. At the end of this diploma thesis, the machine vision system is ready for installation
on the production line
|
