Biomedical Engineering Reference
In-Depth Information
during the production process should be solved automatically and that the operator
must solve only a few problems, those that are too complex or unusual to be solved
automatically. Since any production process is affected by various disturbances, the
control system should be an adaptive one. Moreover, it should be self-learning
because it is impossible to foresee all kinds of disturbances in advance. AI that is
able to construct the self-learning algorithms and to minimize the participation of
the operator appears especially useful for this task. AI includes different methods
for creating autonomous control systems. The neural classifiers will be particularly
useful at the lowest level of the control system. They could be used to select
treatment modes, check cutting tool conditions, control assembly processes, etc.
They allow for more flexibility in the control system. The system will automatically
compensate for small deviations of production conditions, such as a change in the
cutting tool's shape or external environment parameters, variations in the structure
of workpiece materials, etc. AI will permit the design of self-learning classifiers and
should provide the opportunity to exclude the participation of a human operator at
this level of control.
At the second control level, the AI system should detect all deviations from the
normal production process and make decisions about how to modify the process to
compensate for the deviation. The compensation should be made by tuning the
parameters of the lower-level control systems. Examples of such deviations are
deviations from the production schedule, failures in some devices, and off-standard
production. At this level, the AI system should contain the structures in which the
interrelations of production process constituents are represented. As in the previous
case, it is desirable to have the algorithms working without the supervisor.
The third control level is connected basically with the change of nomenclature or
volume of the production manufactured by the factory. It is convenient to develop
such a system so that the set-up costs for a new production or the costs to change the
production volume are minimal. The self-learning AI structures formed at the lowest
level could provide the basis for such changes of set-up by selection of the process
parameters, the choice of equipment configuration for machining and assembly, etc.
At the third control level, the AI structures should detect the similarity of new
products with the products that were manufactured in the past. On the basis of this
similarity, the proposals about the manufacturing schedule, process modes, routing,
etc. will be automatically formed and then checked by the usual computational
methods of computer aided manufacturing (CAM). The results of the check, as well
as the subsequent information about the efficiency of decisions made at this level,
may be used for improving the AI system.
The most complicated AI structures should be applied at the top control level.
This AI system level must have the ability to reveal the recent unusual features in
the production process, to evaluate the possible influence of these new features on
the production process, and to make decisions about changing the control system
parameters at the various hierarchical levels or for calling for the operator's help. At
this level, the control system should contain the intelligence knowledge base, which
can be created using the results of the operation of the lower-level control systems
