Science topic
Controller Design - Science topic
Explore the latest questions and answers in Controller Design, and find Controller Design experts.
Questions related to Controller Design
Hello Everyone
Does Nonlinear ARX model is suitable for controller design ? For example, can PID controller tuned on Nonlinear ARX Model ? or It is suitable for only prediction purpose ?
thanks for your interest.
I have a system with 3 input and single output. In system Identification toolbox of MATLAB, how i put the 3 input data sets for estimating the system models simultaneously?
controller design? energy management?
In my study there are 2 intervention groups and 1 control group. In the ANCOVA I take trait scores as covariates and state scores as AV (so I compare the differences in the groups after the intervention).
Is there a statistical way to check if there was a change in the control group? I am replicating a study, they assumed there was no effect in the control group - but what if I am not sure (because the control design could have an effect). Within the ANCOVA, I only know that the results differ, but not whether the changes differ, right?
Can anyone suggest good books/tutorials/websites for controller design of dc-dc dc-ac converters for PV and Wind applications?
Dear all,
Can we say, robust iterative learning control design, in other form is Data driven control design
In his name is the judge
Hi
There is a fuzzy logic control system in python. The system contain 2 inouts and 18 outputs.
inference of system is mamdani and shape function used to be guassian.
Then in term of refine performance of the controller I need to optimize specifications belong to shape functions of both input and output. In order to that I need to use multi objective optimization.
We have 2 input and 1 output in case of this problem. I have developed 3 shape functions for each entrance and 3 for output and the shape function is gaussian so we have 18 parameters totally.
I defined my problem as a function in python. But notice this I there is not any clear relationship between input and output of function. It’s just a function which is so complicated with 2 inputs and 18 outputs.
I made my decision to use NSGAII algorithm and I really don't want to change the algorithm.
So I try every way to optimize my function but I didn’t find any success. By searching about python library which can do multiobjective optimization I find Pymoo as the best solution but I really failed to optimize my function which is complicated custom function, with it.
So It’s really my pleasure if you can introduce me a library in python or suggest me a way that I can use Pymoo in order to this aim.
wish you best
Take refuge in the right.
In his name is the judge
Hi
I'm trying to make a fuzzy controller in order to optimize my absorber performance in opensees(in python).
I use adaptive neuro-fuzzy inference system (ANFIS) toolbox in matlab to make fuzzy system as controller.
input data for fuzzy logic system are acceleration and velocity of absorber and the output data is force wich controller send it to absorber for performance improvement.
in fact i want controller learn ,based on velocity and acceleration of a point of structure, how much force need for turn structure into it's balance position.
note that quantity of force determine by fuzzy controller system and applying force part of absorbr job.
logically we have to assign come load to the point of structure wich absorber locate there and then get acceleration and velocity of absorber as input training data of fuzzy logic system.
but i realy don't know how can i do this.
note that i want to give force to structure in balance position, in the otherwise i think make train data is possible when using dynamic loading so i entirely confused here
if you have any suggestion i realy eager to hear it.
wish you best
Take refuge in the right.
In his name is the judge
Hi
for using matlab facilities especially matlab controller like fuzzy have to sending and receiving data between matlab and opensees, in fact i want send two data as fuzzy controller form opensees to matlab after each time history step in opensees and then matlab send back one output to opensees.
to achieve this i have to connect matlab and opensees with openfresco it means i do hybrid simulation.
first is this possible to do this?
second is there any one do work like this, do hybrid simulation generally or specifically do hybrid simulation between matlab and opensees. if yes i am very thankful to share work with me or for some aim.
have to say i prefer not to use openseespy and i almost read all example and guide in openfresco site and even on each site but i cant do this right.
Any help is greatly appreciated.
Take refuge in the right.
In his name is the judge
Hi
I want design cotroller in to control active force of damper.
On this purpose, i analyzed structure with damper, wich is tlcdg, and get data whic use for generate controller. data are accleration and velocity of damper as input and dampers force as output ( i mean force wich made by own damper under earthquake excitation, not active force).
here is anfis properties :
number of inputs are 2
number of outputs is 1
generate fis method is grid partition
number of membership function are 4 for each input
input membership function type is guass2mf
output membership function type is linear
opt method is "hybrid".
(have to say i tried different epochs membership function , .....)
Unfortunately anfis toolbox in matlab refuse to train and build Suitable fuzzy controller wich means error is too much in training, so this answer is not acceptable.
i have some idea for make it true but i'm not sure.
here is my ideas :
* first i think train controller with more or less data ( training data are about 2000 wich is under 100s earthquake excitation but when i reduce earthquake excitation to 10 or 15 seconds the error is acceptable however i think this solution is not good.)
* second maybe i must try one damper for training.
* the last idea is to assign force on the damper location and get acceleration and velacity for generate trianing data.
here is my data and shot frome my try.
Any help is greatly appreciated.
Take refuge in the right.
In his name is the judge
Hi
I have to use tsk fuzzy system in python for my research.
Please recommend me a library for tsk ( not mamdani ) fuzzy system in python.
Also if this library exist please Introduce me a source for learning it in python.
wish you best
Take refuge in the right.
In his name is the judge
Hi
I want to learn fuzzy in python for my research.
Please recommend a good source for learning fuzzy logic system in python.
wish you best
Take refuge in the right.
In his name is the judge
Hi
can you say wich set of data is best to train and test ANFIS in matlab.
i mean how generate data to get less error, like nuber of data or Dissipation of them
Wish you best.
Take refuge in the right.
dear experts
i have estimated state space model form input output data using system identification, now plz guide how to design LQR controller
Hello,
I'm doing a meta-analysis with CMA software and my studies usually have a pre-post control design. CMA offers the possibility of getting the effect size in unmatched groups with pre-post by standardized change score SD or post score SD. With the change score SD I need the pre-post correlation, but that information is not given in many studies.
Another option I have is to calculate the effect size only with the post means and scores of both groups.
What should I do?
Thanks in advance,
Sonia
Hi everyone!
I would like to ask which control design method is more effective for controlling the steer-by-wire system? I am planning to use the MATLAB/Simulink.
Thank you in advance for your answers.
Dear Researchers, our team is interested in collaboration to conduct research on advanced control design for permanent magnet synchronous machines. In our lab we have an experimental setup to validate simulation results. If you are working on motor control and interested in collaboration, and do not have opportunity to test your simulations, we are glad to invite you to work with our research team. For more information, please go to https://pcmc.kz/
Hi
I intend to define a frequency domain objective function to design an optimal controller for Load-Frequency Control (LFC) of a power system. My purpose is to optimize this objective function (finding the proper location of zeros and poles) using meta-heuristic methods.
I will be happy if you share your valuable relevant and informative experiences, references and articles in this field including how to define and how to code.
Thanks
In controller design for mixed H2 - Hinf performance, ||W1(s)S(s)||inf<1 and ||W2(s)T(s)||inf<1 are to be satisfied simultaneously. The question is how to obtain the weight W1(s)?
Is there any generalized way to obtain W1(s) for any given system?
Hi All
I am doing a nonlinear optimal control design project, typically a practical engineering problem.
It is easy to solve that with direct method. However, I am wondering is there any way to prove that the optimal solution exists ? Further, since the direct method is employed, is there anyway to prove that the numerical solution will converge to that optimal solution?
Could anyone offer me some reference paper about that ?
thanks
Jie
Will be it a valid research design or else there is a need to control group? I have heard about 3-armed randomised controlled design. Is it possible to have 4-armed randomised controlled design.
I request all senior researcher to clarify it.
I am trying to design a PR controller to control the output voltage of a single-phase H bridge inverter having unipolar SPWM (fundamental frequency=50Hz and switching frequency=20kHz).
Can anyone please suggest to me a simple way to find Kp Kr values for the PR controller?
Your answers would be very helpful and I appreciate your help.
Kind regards, Silpa
Hello !
I have a PSIM circuit model where for the PI Controller Gain and Time constant values are mentioned. I would like to use the same values with Matlab PI controller where input is Proportional Gain and Integral Gain( Kp & Ki)
Any method to find Kp and Ki from Gain and Time constant values mentioned ?
Thanks in Advance
Sreeraj Arole
Optimal control and nonlinear control system.
I am trying to design a PI controller for a zeta converter to get a constant dc voltage at the output, but I am getting a somewhat sinusoidal ( not really a sine wave, but it looks similar) output... How can I design a PI controller in this case..?
Hello dear colleagues and experts!
When developing a model of a wind generator, I faced the problem of adjusting the PI controller for the pitch angle. I read a lot of articles, but the PI regulator failed ...
Please help with setting it up and getting characteristics
Thank you very much in advance!
Dear all,
I have a set of I/O data.
I estimated a transfer function based ( command: tfest) on this data in MATLAB. (I removed the trend from the data before it: command: detrend).
Can I design the PID-controller based on this (command: pidtune) and use it in the process (As the data is without trend)?
If no, how to adjust the PID controller settings for the real process?
Thanks
Hello all,
By modifying the PWM of the CP signal, I can change the charging current of Electric vehicles in AC charging.
But how the current is controlled/reduced accordingly? Which side does this current control?
Does charging station (EVSE) supply the corresponding charge current? or EV controller in the vehicle side reduces the incoming current?
Any literature/references/docs/links will be useful. I tried searching, but couldn't find how the current is controlled/reduced
Regards,
Praveen
Are there any advanced control strategies which could directly handle the non linear system for controller design, without doing any sort of linearization? Please give specific suggestions in this regard. Also please suggest some good sources to learn about the advanced nonlinear controller design for power electronic converters.
Hello all,
In all the literature and texts available for full-state feedback control design, it has been given that in the absence of reference input, all the states represented by state vector x(t) tends to 0 as time t tends to infinity. However, the reason is not given for this statement. Can anyone justify this statement? Looking forward to the answers. Thank you.
Regards,
Ankur Gajjar
Hope this helps. I am assuming a very simple realization of state-feedback.
Ankur Gajjar, my answer would be outdated for you. Still, I shall make an attempt. Regarding application of this in chemical process/power plant, 'A 'is the system model which describes the entire chemical process/power plant. If while modeling A, you considered all the process parameters(states) like temperature, reaction agents, stirring frequency, etc., and assure that process doesn't blow up (unstable or 'A<0' PSD-positive semi-definite), then it can guarantee that all the states settle down to finite value or zero in-case of absence of input.
I am new to both Simulink and USRP hardware. Basically, what I am trying to do is just to send and receive the data using only one USRP n210. I have created a transmitter and receiver blocks but I do not get any result from scopes. The scope graph showed nothing. There is not much examples about using Simulink to work with USRP hardware. Is there any simulink model to work fine? Thanks for any help and answer.
Please i need recommnedation on texts or literature that can improve my knowledge and skills on tuning of control systems ranging from sliding mode, LQR/LQG and others. I alwys have problem at this stage after rigor of modeling.
Most of control design problem involves tuning heuristically. In my opinion, this is randomness that doesnt have strategies. Even PID control with popular Ziegler Nichols still involve randomness!
there should be a way to know the range of tuning.
Dear friends,
Please see the attached Figure. The conventional PI-speed controller block diagram is here.
I had some technical queries about the controller design.
The questions are followed as :
- How to calculate the PI-speed controller Bandwidth?
- Also How to find the efficient calculation of K_p, K_i gain values? Not consider transfer functions.
- How to find the exact “β” angle for the MTPA?
- Is speed controller bandwidth dependent on current controller bandwidth?
I really appreciate to your beloved answers. definitely your answer helps me to understand and design an efficient PI-speed controller for IPMSM drives.
Thanks in advance.
Lets say for example that I need to control the z axis position of a single rotor uav. I knw that setting different throtle commands will vary the engines rpm ans thus change the altitude as well as creating a body torque in the oposite (of the motor rotation) direction. Considering that I can control yaw with cannards (like a rocket tail). In this scenario what is the main way of decoupling the dynamics?
I some papers, the mean square error is considered. In some other, the mse is normalized by dividing the error by the (total sampling instants x the total length of the reference trajectory).
Which solution does represent the factual error?
I am facing difficulty in:
1.designing the rule base for the fuzzy logic controller.
2.taking the number of membership function and there argument as a variable.
While designing the controller for a system in bond graph domain, I have gone through the terms and bond graph models titled "Virtual Bond graph" and "target bond graph". How are these models assisting me in the controller design in bond graph domain?
Hello,
Currently in my project, I am making a hardware controller which can modify the incoming PWM signal and vary the EV A.C charging. To start of with, does anyone know which exact chapters of IEC 61851 standards to buy ? I find that there are many chapters. Basically this hardware will modify the incoming charging current from EVSE to EV according to our inputs. Any literature/material regarding this will be useful. Thanks
Regards,
Praveen
How to calculate the settling time for the PV MPPT controller?
I mean for example if the operating conditions are changed: solar radiation, the duty cycle, etc.
How much time the does system need to reach the steady-state condition?
Can I make the sampling time changeable during the simulation?
For example, I want the sample time to be 0.0005 in the beginning and then change it to 0.00001 after 3 seconds.
I am designing a PI controller for a heating coil. Air flowing around the heating coil is heated before being supplied to a heated space. Hot water flows through the heating coil. In the suggested control system, the controlled variable is the temperature of air and the manipulated variable is water flow rate through the coil. The equation describing the dynamics of the system is shown below:
Cah*DTair/dt=Mw(Twi-Two)+ Ma(Ta-Tair)
Cah= heating capacity ,which is contant
Tair= Temperature of air supplied to heating space, varies of time
Mw= Mass of water flowing through the coil, varies with time since its the manipulated variable
Twi= Temperature of water supplied to the coil, varies with time since it is part of another process
Two=Temperature of water exiting the coil, constant
Tao=Temperatue of air before approaching the coil, constant
In the above equation, Mw and Twi are both variables (function of time), and therefore the equation is nonlinear due to Mw*Twi; To design a controller, I should linearize around an operating point, which will restric the validity of the controller to this specific point (or around that point). Can anyone guide me regarding a better control design approach? should I check nonlinear control theory? multivariable control? etc..
What analysis tool is most suitable for transient-state analysis and designing control to mitigate those transients in power distribution system? I am planning to carry out transient analysis caused by sudden increase in load on distribution feeder and want to control those transient by charging/discharging the battery storage system. Our load has a high ramp rate and high magnitude which could cause these transients in the distribution grid voltage profile.
Regards
What is the maximum number of population in quasi experimental research design using pre/post and control in a study?
Other than resource sharing and optimization, is there any direct affect of event based triggering in controlling chattering in sliding mode based control design?
Dear all,
As you know, one method of maximum power point tracking in wind turbines is to select the reference active power which complies with the maximum power that is achievable (power at MPP).
I wonder what happens if one increases the reference power beyond what is imposed by the value corresponding to MPP? Will it lead to power reduction? (from the control engineering point of view, we can design the power control loop so that the output power will approach the reference power; so, why not increase the reference power beyond MPP?!)
I'm trying to control temperature with PID(Kp,Ki, Kd). During the rise time the system use one PID and when target is reached the system switch to another PID with other parameter in order to eliminate the oscillation. I know that I can remove oscillations increasing Kd but in this case it does not work.
Results of real time system
Control temperature with constant PID: see figure onePID.png
Control temperature using PID1 during rise time (blue color) and when temperature reach the target, the controller is changed to PID2 (Steady state error: black color): see figure TWOPID.png
- void setTarget(...) { mPid = pid1; } void steadyStateError() { float e = fabs( temperature - target); if(e<=0) mPID = pid2; }
Hi Ruben Hernando ,
You can make a gain schedule so that the transition from Kp = 7.2 to Kp = 5.0 is smoother. In other words, the control action doesn't make a sudden 'jump' when error ≤ 0. You can design the gain schedule, Kp = f(e), as a function of error.
For example, if the error is far away from the origin, Kp = 7.2. If the error is close to the origin, then Kp approaches 5.0.
Hi all,
Is it possible to tune PID controllers for a Simscape model just by purely using Sim control design toolbox? It is a car model created in Simscape environment. The model contains Engine, CVT, wheels, etc. And the controller is a PI to ensure car follows the input drive cycle.
The screenshot of the model is attached.
Regards,
Sajad
wish to develop beam width control in an antenna not reconfigure beam???
Hi,
can anyone help me with the time delay which the controller will experience during the excitation ?
i wanna consider time delay in my controller while the controller is changing the coefficients of the damper.
To sum up, how can i consider time delay in time domain in my controller?
any suggestion ?
thanks in advance.
I have created a Simulink model of PID controlled Bouc Wen MR damper semi active suspension. The feedback signal to PID is taken as sprung mass displacement and reference value is set to '0' (zero). The output of PID control is to be voltage signal for MR damper.
While trying to auto tune the PID controller, I am getting the error:
Linearization aborted because the linear plant model seen by the PID block is effectively '0'.
Ensure the PID loop is physically closed and none of the block in the PID loop returns '0' when linearized at time t = 0.
I have used saturation block after PID with upper limit 5 and lower 0. (as MR damper max input voltage is 5 volts and min is 0).
Error image is attached.
Hi Pankaj,
In my previous post, I have mentioned that the Saturation block is a Discontinuous function block. Since your Reference Value is 0, the Absolute value function block |u|, is continuous at 0, but is not differentiable at 0. It may also cause an issue in the linearization process.
I ran the simulation and the Diagnostic Viewer (see attachment) showed that it contain an algebraic loop. The fatal error occurred when a suspected singularity was detected in the Integrator block inside the Bouc Wen MR Damper.
Please solve these issues before attempting to tune the PID controller. Also try to tidy up your Simulink model because certain parts look like "spaghetti". It will be helpful to identify I/O signals when you want to troubleshoot the system.
Now, let's see what Dr.
Zeashan Khan
will advise.
The problem is of PID controller for a ball beam system, intially some angle is given to the beam. A controller is to be designed to balance the ball in the beam.
if (*rtu_BallPositionValid)
{
real_T error=(0-*rtu_BallPosition);
double derivative=(error-lasterror)/0.001;
integral+=error*0.001;
*rty_RequestedBeamAngle=*rtu_ActualBeamAngle+ (Kp*error+Kd*derivative+Ki*integral);
lasterror=error;
} else {
*rty_RequestedBeamAngle = localDW->Delay_DSTATE;
}
I am testing for the power of art therapy to improve level of motivation in substance abusers in drug treatment.
Using a matched pairs experimental/control design. total sample=34, 17 per group.
Would testing for confounding demographic variables: age, education, and employment- which are identified as indicators of higher level of motivation - be useful in addition to standard t-test comparison? Is the sample big enough? What test would you use for confounding variables? Thanks- Mollie
why bounded condition (assumption) is needed for exogenous input disturbance in controller design problem?
the coefficient of the plant should satisfy what kind of requirement ? and can you give some exampled?
It is generally difficult to stabilize unstable high-order non-minimum phase plants with the 'ideal' PID controller. Take Lakshmanaprabu's model as an example, which has a real pole and a real zero in the right-half plane:
Gp = (5⋅s − 1)/(16000⋅s4 + 41200⋅s3 + 2940⋅s2 − 152⋅s − 5).
It is possible to stabilize the plant with a cascading compensator
Gc = k1 − (k4⋅s2 + k3⋅s + k2)/(s3 + N3⋅s2 + N2⋅s + N1)
where
k1 = −4.4690e+04
k2 = 4.9923e+04
k3 = −2.6980e+05
k4 = 1.1096e+05
N1 = −1.1172
N2 = 6.0334
N3 = −2.4079
The closed-loop transfer function (Gc*Gp)/(1 + Gc*Gp) is a 7th-order system:
Gcl = (−2.235e+05⋅s4 + 2.793e+04⋅s3 + 4190⋅s2 − 139.7⋅s − 5.586) / (1.6e+04⋅s7 + 2674⋅s6 + 270.4⋅s5 + 17.88⋅s4 + 0.8151⋅s3 + 0.02545⋅s2 + 0.0004759⋅s + 4.249e-06),
which has two real zeros in the left-half plane and another two real zeros in the right-half plane. Cancelling the two real zeros in the left-half plane with a pre-filter
Gf = 4.249e-06/(s2 + 0.125⋅s + 0.0025)
applied to the input command outside of the feedback loop yields Gclf = Gf*Gcl
Gclf = (−5.934e-05⋅s2 + 1.483e-05⋅s − 5.934e-07) / (s7 + 0.1671⋅s6 + 0.0169⋅s5 + 0.001118⋅s4 + 5.094e-05⋅s3 + 1.59e-06⋅s2 + 2.974e-08⋅s + 2.656e-10)
Computing the DC gain of Gclf (−2234.5) and placing a reference scaling factor "1/dcgain(Gclf)" ensures the unity gain when a unit step input is applied.
Hi,
Selecting the H∞ controller weighting parameters is one the most important part of designing of such a good controller, however, there is no specific method in the most references to do that and most of the works have been done by the trial and error methods.
Do you have any suggestion in order to tune the H∞ controller?
Thanks.
Kindly help me with the parameter selection of the value of gains of the PR controller.
Mathematically, the commonly used PR controller is not equivalent to PI controller. It can be deduced using the theory presented in the appendix of [1] that the controller in stationary frame shown in the attached figure is equivalent to a PI controller in synchronous frame.
[1] D. N. Zmood; D. G. Holmes. "Stationary frame current regulation of PWM inverters with zero steady-state error," IEEE Transactions on Power Electronics, 2003, 18(3): 814 - 822.
You can find in the attached figure that there exist coupling terms between the alpha axis-controller and the beta-axis controller. The commonly used PR controller in stationary frame does't include such coupling terms. The coupling terms come from the controller itself, not from the plant. If completely decoupled controller is required, the coupling terms generated by the plant model (e.g. the typical R-L model) should also be included.
I have done some test in matlab/simulink some time ago that such a controller performs exactly the same with the PI controller in stationary frame, if the parameters Kp and Ki have the same values.
However, in practice, the performance difference between the commonly used PR controller and PI controller may not be so obvious.
In addition, I would like to recommend the an advanced resonant controller, i.e. the vector-proportional-integral (VPI) controller. Its expression is s(Kps+Ki)/(s^2+w^2). Its numerator could be used to cancel the pole of the R-L plant. VPI controller could eliminate undesired peaks in the bode plot. According to my experience, it performs really better than the common PR controller. More information about the VPI controller could be found in [2] and other literature.
[2] Alejandro G. Yepes; Francisco D. Freijedo; Jesús Doval-Gandoy; Óscar López; Jano Malvar; Pablo Fernandez-Comesaña. " Effects of Discretization Methods on the Performance of ResonantControllers," IEEE Transactions on Power Electronics, 2010, 25(7): 1692 - 1712.
I want fuzzy logic controller design theoretically and in MATLAB simulink and programming any one have that material?especially for designing uavs and quadrotors
Hi
Consider the following closed loop generalized plant. P is the generalized plant, K is the controller to be designed and $\Delta$ matrix is the uncertainty matrix with the norm of smaller than one for all combination of the parameters.
I try to get the controller with H_inf design but the H_inf norm of the resulting closed loop from w_u to z_u is larger than 1. What does it mean?
I have tried to assume very power full actuation ( limitless). It never get smaller than 1. From theory means the controller is not robust. but why it is not possible to make it robust.
P.S: the real plant has two poles on origin.
I'm estimating sample size for a future study using the primary outcome (mean and SD) from a current study. The future study will use a randomized control design with two independent 'arms' (or groups): Breathing Training or Balance Training.
The current study uses Breathing Training only, for which I have primary outcome data. However, no study using the Balance Training has used the same primary outcome (and therefore no mean and SD data).
Can I use a secondary outcome to estimate sample size for this Balance Training group?
Given a Fractional Order PID controller designed for trajectory control of mobile robot. The question is How to implement it using one of the arduino platforms knowing that when approximating the fractional integrator and derivative using discrete transform it gives infinite orders of terms in the Z-domain.
I have a plant with transfer function, I design a PID Controller and the output response of the plant along with controller can be obtained with Ziegler method. What is meant by control signal, how to get response for it?
I have designed a controller using Integrator backstepping method and that controller is based on Lyapunov Theory. I have a question How I can check stability of resulting controller?
I want to know if there is any system that needs to operate on different variants (combinations of the P, I, D constants) of PID controller; like it operates on P-only controller first; then switches to PI and then may be to PD or PID alternatively or simultaneously or may be cascaded? Is there any need to have such a system? If yes where and if not why not?
I actually need active and reactive power controller design
Its a general question.
For example, I want to design a controller for a nonlinear system to cope with certain periodic disturbance. In order to get a linearized state space model, I linearize the system about certain operating region and state space linear model has no information of this periodic disturbance. how to remove or minimize it. One solution is to use model predictive control where a real system posed to this periodic disturbance is tracking a reference system model (this reference model doesn't included this disturbance into account). I would appreciate if you can suggest a solution other than this.
(Edited)
We found that there is a flow lag below the set point. Controller is slow to react to a set point change. Though we set the standard temperature and pressure conditions to 25oC and 14.696 psia, it increases to 34oC and 78 psia respectively. The control valve also heats up.
What could be the possible problems? How can we tune the PID values? we set the values as shown in the manual. But still, the problem occurs. Kindly give us the guidelines to rectify it.
If I find out the gain margin and phase margin of a transfer function, then whether I can relate it with the operating frequency.
1. Isn't worth checking the controllability of a nonlinear system before any regularisation methods (controller design, optimal control solution etc)?
2. Are they controller design techniques that do not require a controllability check?
I'm doing a project in a company who wants a design of a rudder control system for a ship. I finish the design and I did it on LabVIEW with the Control Design VI in a Simulation Loop. That was the easy part of the project but the company told me that they want a project where they can use it in the reality, so I have to simulate in real-time the ship and I can't use the simulation loop, beside I must to simulate two sensor with CompactRIo (compass and Synchro) and send it in any communication protocols to a computer. I know there is a VI where I can use to simulate a lag-leag controller.
The problems I have are: How to implement a plant equation without the help of Transfers Function? How to implement a integrator (1/s) in real-life? ( I had seen the possibility on use deadband). The last question, how to simulate the two sensor that I mentioned previously with CompactRIO?
What I know about the definition of ''Robustness Property'' for a controller designed for a particular set of parameters of the system it can be said to be robust if it works well under high-gain feedback to reject the disturbance gained by the system and eliminate the effect the system parameter uncertainty. So, is there any relation between robustness of the controller and faster response time such as raise time maximum overshoot settling time etc..?
Truly Yours
I want analyse the stability of chaotic systems and design a controller for such systems.
I need the value of those parameters that depends of the lenght of the ship
for my pid control designing my present value and the previous value being calculated as error. for this it is mandatory to define a kth value and time instant for it , can any one help for me to do this in linux c++ coding .
In linear system, to improve bandwidth, we can use 2 DOF control design with a reference filter. The design is performed in frequency domain. Is there an equivalent method in control of nonlinear system? Let say if I am using backstepping and Lyapunov theory to develop the control.
Non- Linear Control
1. Predictive control
2. Hysteresis Control
3. Sliding mode Control
4. Iterative learning control
5. Artificial Intelligence
6. Adaptive Control
I need to know all these above mentioned control techniques for Inverters In Microgrid in detail. Plz suggest me Helping materials thanks.
i plan to do hardware project on self tuning PID model reference adaptive controller for bldc motor. i simulated with help of matlab ,but how to implement hardware for MRAC controller ?
I want to know about the state of the art in multivariable control. Also I want to know about all different multivariable controllers that have been used so far upto recent times. It would be best if I can get hold of a review paper on multivariable controllers. Thank You.
I have to design a controller using Convex Optimization and Sum of Squares technique, please suggest some papers on this. What are the advantages of using this techniques from the classical controllers or other robust controllers.
Dear all,
In traditional sliding mode control, the approach is so clear. we design the sliding surface firstly and then find a control law that satisfy the condition of sliding mode existance. But in high order sliding mode, I think it should follow some principle that is similar to 1-sliding mode control.
From my point of view, in high order sliding mode control, we design a sliding function which is the expression of s and its r-1 time derivatives equal to zero. while in 1-sliding mode control, just s=0 . So in 1-sliding mode control, just 1-order accuracy while in high order sliding mode control, cos s=dot(s)=dot(dot(s))....... r-1 time derivatives..=0, then it will have a higher order accuracy. What's more, high relative degree makes the real control law to the system a smooth function,Then we base on some principles to design control law U.
Here is my problem, what principles we should take in high order sliding mode control to design control law? In 1-sliding mode control, the principle is s*dot(s)＜0, or dot(Lyfun) is negative defined. In some papers of Prof.Levant and Prof.Slotine, They gave some principle for high order SMC. But I not so clear, I want to communicate with people and see how you think of it.
Thank you very much!
I am working on high nonlinear system and I would like to design an efficient controller.
For a passivity-based control design the modification of dissipation energy function can be obtained by adding damping injection which is actually a virtual impedance matrix. which kind of this impedance, fictitous impedance or practical impedance?
Passivity-based control is the kind of nonlinear controller to achieve stability of simple electrical circuit, how can I expand this method to check stability for a complicated electrical circuits?
I will be appreciated your help…
In order to design this controller I need some books or references.
Hi Walid,
A Fractional Order Controller (FOC) typically uses the fractional-order derivative as “enhancement” in part of the mainstream control systems, such as fractional order sliding mode control, fractional order backstepping control, fractional order fuzzy control, and fractional order PID control. The following example shows how to design a simple fractional order sliding mode controller.
Consider a nonlinear control-affine system given by
x' = f(x) + Bu + d
where x is the state vector, f(x) is a smooth nonlinear function of its argument, B is the input matrix, u is the control vector, and d is the disturbance. A desired transformation
ς = λx
is introduced, where ς ∈ Rm×n and it is assumed that λB ≠ 0. Taking the time derivative of ς yield
ς' = λx'
and it easily follows that
ς' = λf(x) + λBu + λd.
The sliding manifold is chosen as
s = ς + Dας
where Dας is the fractional derivative term and 0 < α < 1. Taking the time derivative of s yield
s' = ς' + Dα+1ς.
Next, the Lyapunov function candidate is chosen as
V = (1/2) sTs
and the time derivative of V is
V' = sTs'
The sliding manifold will reach s = 0 in finite time if the control law u is chosen as
u = (λB)−1 [− k1s − k2 sgn(s) − Dα+1ς − λf(x)].
where k1, k2, and η are strictly positive with k2 = η + λ ||d||max .
Hence, according to Lyapunov stability theorem, V' becomes
V' = sT[− k1s − k2 sgn(s) + λd]
V' = − k1sTs − k2 ||s|| + (λd)Ts
V' ≤ − k1sTs − η ||s|| ≤ 0.
When s = 0 is reached, then
Dας = − ς
is asymptotically stable if the following condition is satisfied:
|arg[λi(-1)]| > (1/2) απ.
On top of the valuable literature suggested by Prof. Lafifi and Prof. Khan, you are also advised to refer to Prof. YangQuan Chen's works in Fractional-Order Control Systems. Here is just one of the books.
I need help on my problem. I managed to tune my PID controller using Ziegler-Nichols continuous cycling method and able to determine my Kp, Ti (integral time) and Td (derivative time). But the problem here is, I do not know how to use it in SIMULINK PID block. Let say my Kc is 500, Ti is 0.02 and Td is 0.01.
In SIMULINK PID controller (function block parameter), the parameter used is P (proportional gain), I (integral gain), D (derivative gain) and N (filter coefficient). May I know how to convert Ti to I and Td to D? What I know is Kc= P, and I used parallel form. Thank you..
Hi Hany,
In any undergrad control textbooks, you will find that the PID control signal, c(t) in time domain consists of a proportional term, an integral term, and a derivative term, which act on the error signal, e(t) as shown in Fig. 1.
When converted to transfer function via Laplace transform in s-domain, the PID control signal, C(s) is illustrated in Fig. 2. You can clearly see the relationships between Ki and Ti, as well as Kd and Td after factoring out Kp.
In practical engineering, unwanted high-frequency electrical disturbances known as noise may be generated in the electronic components of the measuring instruments (sensors). Because the noise is blended in the error signal (setpoint - measured process variable), the derivative term can greatly amplify the control signal that may lead to control actuator saturation.
To alleviate this problem, a first-order low-pass filter is included in the derivative term in modern PID controller, as shown in Fig. 3 as well as in the Simulink PID controller block. If noise is not considered in your Simulink model, you may set N = inf in the PID controller block at the preliminary design. Other things to consider include:
- disturbance rejection capability,
- gain and phase margins of the PID control system,
- making the controller less noise sensitive.
If the mathematical model for the biomedical system is developed, you may want to compare the performance between Ziegler-Nichols-tuned PID controller and the auto-tuned PID controller (by using PID Controller Tuning feature in the Simulink block).
I am working on high bandwidth controller design for high speed scan mirror mechanism for satellite. The structural frequency (or vibrational frequency) for this system is coming within the desired bandwidth for attaining the high speed motion.
Passivity based-stability criterion is a useful criteria for multi-parallel source and multi-parallel load converters for DC-distributed system, and it is a linear in nature (attached is the paper of the method) , I am thinking if there is some links between this method and sliding mode control.
I am investigating the feasibility of the nnMPC block for a MISO system and would like to know if anyone used the block successfully with more than one input.
Actually, I have a big doubt regarding the inputs of fuzzy logic controller (FLC), since I am new in this area. Most of the example that I found used error and derivative of error as the inputs for FLC. But, it is possible if I use error and integral of error as inputs of my FLC? Actually, I have a highly non-linear process to control.
Thanks..
I have a nonlinear system with order 5. I have designed linear or nonlinear control algorithms for the systems. The input of the system have to be a positive signal and should have some upper bounds to it. LQG , PID, State feedback (by linarization) as well as nonlinear control like Sliding mode control and super twisting algorithm is giving a negative peak in its control signal. For this reason I have to reject the negative peak in the control signal before giving it to the plant.
How can I overcome this problem?
I am trying to reduce(flattened) the flutuations in a signal in "realtime applications" using Simulink. For this , I am using a bandstop filter then a lowpass filter. But the delay is very high.(1-2 sec delay is permissible.). However the delay introduced is more than 20 sec and increases as the filter order increases. Is there any other technique to reduce the flutuations without delay?
I tried using filtfilt function but it of no use for real time systems.
Please help in this regard.
A non-minimum phase system is difficult to control because of RHP zeros. How to deal with this type of system?
I am using direct synthesis to design a controller for an open loop unstable processes. G(s) = k/(s-a) and I think I need to stabilize this by adding a feedback loop before applying direct synthesis. But I am not sure what kind of feedback loop I can add. Please help. Thanks.
I have a continuous-time state-space model and I designed in matlab a controller based to this model so yet the whole process is in continuous-time. In Simulink I simulate the model so that there is a sample time Ts (runge kutta). Of course the controller will be digital in the end, so 1. Am I simulating the process in discrete because of sample time? 2. Do zero-hold block enough to simulate desecrating continuous-time state-space model of controller? What I want to do is to convert my controller to desecrate (digital) and simulate it for continues time model…..
A focus on state-space design and MATLAB/Simulink applications is also desired.
I am searching for the best values for gain values for a PI controller, my supervisor suggested that I may choose the best values according to the error signal. he suggested that I would choose according to the integration of the area under the curve of the error signal, However, this led to incorrect results. I changed my metric to overshoot, settling time, and steady state error. However, I still think they are old metrics. Does anyone suggest a better metric for the error signal?
someone wants to implement fuzzy logic in simulink, what steps should he follow? Thanks in advance!
Hi
I have a problem I need to solve. The system consists of a hot plate type heater that is either on or off. The bag of fluid lays on top of it. There is a sensor that the bag of fluid rests on and checks it temperature. There is also a sensor that checks the hot plate temperature. The bag of fluid needs to be heated to and kept at 40 C. It cannot go above 41C. Fluid may start as low as 1C with the heater element off. The rise time from start (can be any temp) to set point (40 C) can be no longer than 1/2 hr. I was researching a pid controller for the heater, but am concerned about any overshoot. Some articles I've read about the MPC (Model Predictive Controller) makes it sound like it may be better suited to this application. If someone has worked on a problem like this, I would like to hear about your experiences.
Thanks...
Instead of PI or PD if we replace the system only by I or D, what will happen?Actually I need to know what will be the effect of system response. Also in fuzzy logic controller, if we don't connect P controller with (Z-1)/Z block in simulink, there will be any deviation occur in system response.
Okay, so I am doing a study on Iron Deficiency Anemia. In the first phase, I will determine the prevalence and then I need to go for Case control and an RCT. I need to ask if I have to calculate sample size for Case Control and RCT or can I have some standard minimum number of my own choice?
I mean, I have seen a lot of papers where the authors have just taken a certain number of cases and two or three controls per case. So can I take (for example) 50 cases and 100 controls, depending on my prevalence results or is there a specific formula for calculation?
Similar is the case with my Randomized Control Design, that is, can I take a certain number of people to start with or (again) is there any formula?
Thanks.
![[y8dgetye]_fig_2.png](profile/Yew-Chung-Chak/post/How_to_convert_Ti_integral_time_and_Td_derivative_time_to_I_and_D_parameter_in_function_block_parameter_for_PID_controller_SIMULINK/attachment/59d64fab79197b80779a8c08/AS%3A499005840531457%401495983671330/download/%5By8dgetye%5D_fig_2.png){kind=link}
![[y7jdxsws]_fig_1.png](profile/Yew-Chung-Chak/post/How_to_convert_Ti_integral_time_and_Td_derivative_time_to_I_and_D_parameter_in_function_block_parameter_for_PID_controller_SIMULINK/attachment/59d64fab79197b80779a8c07/AS%3A499005840531456%401495983671267/download/%5By7jdxsws%5D_fig_1.png){kind=link}
![[y7uwxp3u]_fig_3.png](profile/Yew-Chung-Chak/post/How_to_convert_Ti_integral_time_and_Td_derivative_time_to_I_and_D_parameter_in_function_block_parameter_for_PID_controller_SIMULINK/attachment/59d64fab79197b80779a8c09/AS%3A499005839941632%401495983671358/download/%5By7uwxp3u%5D_fig_3.png){kind=link}