Optimization of the critical plane direction
A very important target in the work with multiaxial fatigue criteria is to get to the goal as fast as possible. Thus the way of thorough mapping of fatigue damage parameter on a dense predefined system of planes, which is used in PragTic by integral methods is not desirable here. If a more sparse scanning is used, the important plane directions can be missed. This could be very unpleasant by critical plane methods, since only the plane with a maximum load or damage defines the exact damage of the whole construction.
This is the reason, why the Optimize option should be every time switched on in the Calculation Methods menu, if a critical plane method is examined. Switching it off leads to a non-conservative prediction or sets very high demands on number of scanned planes.
In the case of maximum damage methods, the optimization is run after the casual scanning within the globe analogy method, Bannantine-Socie method or random concept ended. The optimization itself is a gradient method. It seeks for the highest gradient of the damage parameter, until it approaches close enough to the final maximum.
Notification: Only the plane with the maximum damage parameter from the casual scanning is passed on to the optimization procedure within the MD methods. Such solution necessitates the casual scanning enough dense (I usually use Number of scanned planes = 45-72 in the Calculation Methods dialogue), so that the correct maximum was found. The solution implemented also expects that no singularities or very abrupt changes will happen near the maximum. In my opinion, this is a correct expectation in this case, but I have no proof now.
More:
© PragTic, 2007
This help file has been generated by the freeware version of HelpNDoc