关于控制方法的选择
请教一下大家,当控制一个油缸-质量系统时,在运动过程中油缸的等效质量会发生较大的变化,这意味着该控制对象是一个时变系统,请问在建立控制模型时等效质量应该选择最大的等效质量吗?利用何种控制方法可以实现系统的平稳,准确运行呢?开环,位置反馈,或是反馈加前馈的方法?我使用平均质量的估计值。 如果我知道质量在运动过程中会如何变化,我可以计算随质量变化而变化的闭环增益。
Delta RMC可以每毫秒改变增益,以适应不断变化的负载。
下面是一部短片,展示了在移动过程中发生变化的负荷。 质量不会改变,但惯性随着角度的变化而变化。 RMC可以根据角度改变控制器增益。
该液压系统旨在培养学生如何在运动过程中改变控制器增益。
https://deltamotion.com/peter/Videos/Non-Linear-Lab_Medium.mp4
I use an estimate for the average mass.If I know how the mass will change during movement, I can calculate closed loop gains that change as the mass changes.
The Delta RMC can change gains every millisecond to adapt to changing loads.
Below is a short movie showing a load that changes during movement.The mass does not change but the inertia changes as the angle changes. The RMC can change the controller gains as a function of the angle.
This hydraulic system is designed to train students on how to change controller gains during motion.
https://deltamotion.com/peter/Videos/Non-Linear-Lab_Medium.mp4
我们还没有完成。
闭环传递函数的符号公式是:
Ga(s)=K*ω^2/(s*(s^2+2*ζ*ω*s+ ω^2 ))
Gc(s)=Kp+Kd*s+K2*s^2
Kp,Kd和K2的符号公式是什么?
(s+ λ)^3=(s^3+( K*ω^2*K2+2*ζ*ω)*s^2+( K*ω^2*Kd+ ω^2)*s+ K*ω^2*Kp)
现在你可以回答mayseven的问题了。 您可以在质量变化时计算固有频率,然后使用固有频率的新值来计算Kp,Kd和K2的新值。
您曾经问过如何补偿从1吨到100吨的负荷。 我会在活塞的两端放置压力传感器来测量负载引起的力。 然后,我可以使用VCCM方程计算开环增益,新的固有频率,然后是控制器增益的新值。 这可以在负载稳定后的几微秒内完成。
We are not done yet.
The symbolic formula for the closed loop transfer function is:
Ga(s)=K*ω^2/(s*(s^2+2*ζ*ω*s+ ω^2 ))
Gc(s)=Kp+Kd*s+K2*s^2
What are the symbolic formulas for Kp, Kd and K2?
(s+ λ)^3=(s^3+( K*ω^2*K2+2*ζ*ω)*s^2+( K*ω^2*Kd+ ω^2)*s+ K*ω^2*Kp)
Now you can answer the question for mayseven.You can calculate the natural frequency as the mass changes then use the new value for the natural frequency to calculate new values for Kp, Kd and K2.
You once asked how to compensate for a load changing from 1 ton to 100 tons.I would put pressure sensors on both ends of the piston to measure force induced by the load.I could then calculate the open loop gain using the VCCM equation, the new natural frequency and then new values for the controller gains. This can all be done in microseconds after the load settles.
PEN 发表于 2019-6-23 03:47 static/image/common/back.gif
我们还没有完成。
闭环传递函数的符号公式是:
Ga(s)=K*ω^2/(s*(s^2+2*ζ*ω*s+ ω^2 ))
非常感谢您的回复!那在质量的变化过程中,根据在PID控制下闭环反馈系统的特征方程,可以通过修改三个pid 参数来对闭环极点进行配置?
PEN 发表于 2019-6-23 03:47
我们还没有完成。
闭环传递函数的符号公式是:
Ga(s)=K*ω^2/(s*(s^2+2*ζ*ω*s+ ω^2 ))
You once asked how to compensate for a load changing from 1 ton to 100 tons.I would put pressure sensors on both ends of the piston to measure force induced by the load.
我记得,我知道,我知道用压力传感器检测压力,然后输入到控制器A2模拟量模块,对于变化的负载,我知道要用”变化的增益曲线”的方法,(图片曲线),我知道这种方法可以解决变化负载问题,但是具体在的DELTA中如何操作,具体使用我还不会。
I remember, I know, I know, I know to use pressure sensor to detect pressure, and then input it into the analog module of controller A2. For variable load, I know to use the method of "variable gain curve" (picture curve). I know this method can solve the problem of variable load, but I don't know how to operate it in DELTA.
PEN 发表于 2019-6-23 03:47
我们还没有完成。
闭环传递函数的符号公式是:
Ga(s)=K*ω^2/(s*(s^2+2*ζ*ω*s+ ω^2 ))
You once asked how to compensate for a load changing from 1 ton to 100 tons.I would put pressure sensors on both ends of the piston to measure force induced by the load.
我记得,我知道,我知道用压力传感器检测压力,然后输入到控制器A2模拟量模块,对于变化的负载,我知道要用”变化的增益曲线”的方法,(图片曲线),我知道这种方法可以解决变化负载问题,但是具体在的DELTA中如何操作,具体使用我还不会。
I remember, I know, I know, I know to use pressure sensor to detect pressure, and then input it into the analog module of controller A2. For variable load, I know to use the method of "variable gain curve" (picture curve). I know this method can solve the problem of variable load, but I don't know how to operate it in DELTA. PEN 发表于 2019-6-23 03:47
我们还没有完成。
闭环传递函数的符号公式是:
Ga(s)=K*ω^2/(s*(s^2+2*ζ*ω*s+ ω^2 ))
We are not done yet.
Yes, I'm waiting for you to go on with the explanation, as well as the poles of the integral, the more the better. Let me try this formula.
对,我还等着您继续讲解,还有积分极点等,越多越好。我尝试一下这个公式。 有一个错误。
Ga(s)是开环传递函数。
PEN 发表于 2019-6-24 01:18 static/image/common/back.gif
有一个错误。
Ga(s)是开环传递函数。
这种情况下PI控制会有效果吗?还有您介绍过p-only,PD,PD2的增益计算方法,但PI控制却没有具体讲过,能展开讨论一下吗?
mayseven 发表于 2019-6-24 18:12
这种情况下PI控制会有效果吗?还有您介绍过p-only,PD,PD2的增益计算方法,但PI控制却没有具体讲过,能展 ...
PI将起作用,但必须使用前馈。 如果未使用前馈,积分器将增加并导致过冲。 如果使用并优化前馈,则积分器的作用非常小。
如果未使用前馈,则可以关闭积分器增益直到接近命令位置,但这会导致响应缓慢。 最好将积分器增益和前馈一起使用。
PI will work but feed forwards must be used. If feedforwards are not used, the integrator will increase up and cause overshoot.If feedforwards are used and optimized, the integrator does very little.
If feedforwards are not used, it is possible to turn off the integrator gain until close to the command position, but this results in a slow response.It is best to use the integrator gain and feed forwards together.
页:
[1]
2