Hello everyone,
I am an electrical and electronics engineer and responsible for hydraulic dynamic testing machines at Utest. We mostly design machines for fatigue test and i would like to ask help for the problem that limits me for new designs.
First of all, in fatigue tests the main concern is max and min values of displacement according to the controled force. Test may be stopped if the displacemend exceed 10% or any declared value of the initial value. RMC let me to reach status block and current angle. However even at 1Hz. i can not get the value at 90-270 degrees or anywhere between 85 to 95 or 265 to 275. Even if i get the aproximate value , the trend would be unlogical since the value at, lets say at1000 may be less even 1% than at cycle 1001 because one is catched at 85 while other one is catched at 90 degrees. So i want to have a parameter for sine wave values at 0 90 180 270.
I have already designed an algorithm for min max values but it makes the software complicated and slower. Also i would like to create a software for adaptive control. In fatigue testing, consider sample as a spring, the stiffness property is changing in time. In fact purpose is finding how much it changes. And this effects all control parameters during the test. Test cycles are usualy 2million and above and when you observe the trend in the end, the parameters setted are not sufficient to trace the target value. So if you give me a tip or set a parameter for that angles, i will change the open loop parameter according to the difference between target and actual and minimize the error . Of cource my estimation will be there wont be any phase difference between the actual and target.
At first i think, you may solve this problem by taking target value as reference. But this time if the actual values leading or lagging behind the actual, it wont be the actual max or min. If you can add a property that finds min max values , or triggering option at a exact value (For example, catch values when load is at 100kN or at 10mm), it would be perfect.
Also if i know these values, at some application i may use adaptive displacement control for force. Since F = k.X simply, any deformation value meant to be force on it. At some cases our displacement sensor resolution is higher and not noisy. So by changing target displacement value , i may also hold the force cycle proper.
I would appretiate any advice or information you have.
Regards,
