Design and Simulation of optimized Anti-surge Adaptive control system for Centrifugal Compressor

Please, I'm trying to design and simulate An anti-surge adaptive control system for Centrifugal compressor, please can anyone point me to the right direction using Matlab

[MADUKA CHIKA]

Ok, thanks Sam chak

[MADUKA CHIKA]

Please can you direct me on how to accurately model an anti-surge controller using Fuzzy logic on simulink?

[MADUKA CHIKA]

I've taken simulink course and yet can't run a simulation on centrifugal compressors

[Sam Chak]

Hi @MADUKA CHIKA

Could you please provide the mathematical model of a typical centrifugal compressor? This can be in the form of a transfer function , a linear state-space representation , or the more general form of a state-space model .

Having a model is essential as it enables the design of a controller and allows for analysis and simulation of the control system under different conditions and reasonable uncertainties before implementing it in real-time on the actual equipment.

Once you have the model, please post it in your recent thread:

https://www.mathworks.com/matlabcentral/answers/2074101-how-can-i-design-and-simulate-an-anti-surge-controller-for-centrifugal-compressor-using-fuzzy-logic?s_tid=srchtitle

Answer 1

[Sam Chak]

Hi @MADUKA CHIKA

To simulate the dynamics and control of the Centrifugal compressor system in MATLAB, you must adhere to a systematic approach. The first step involves formulating the governing equations based on fluid dynamics and mechanics, subsequently designing an adaptive control system, and finally implementing anti-surge measures.

Accurately characterizing the anti-surge behavior requires some mathematical equations that only you know. Specifically, advanced control theory is essential for formulating a stabilizing adaptive control system, while the Compressor system dynamics can be expressed through differential equations, grounded in the principles of fluid dynamics and mechanics.

In MATLAB, the implementation unfolds through the creation of three distinct function files: one for the compressor (compressor()), another for the controller (myController()), and a third for anti-surge measures (antiSurge()). Lastly, a Main Script utilizing the appropriate ODE solver is required to solve the compressor system problem by integrating these three function files.

[MADUKA CHIKA]

Thank Sam, please have you done anything similar on Matlab or a Tutorial to guide me please, I'll appreciate

[Sam Chak]

Hi @MADUKA CHIKA

I don't have experience simulating a Compressor, but the general approach should be similar. I can provide you with basic code to simulate the control of a Double Integrator system in MATLAB. To model the Compressor, you should consult popular textbooks or academic papers from reputable journals to derive the mathematical model. I've also come across some math here: https://www.mathworks.com/help/hydro/ref/compressorg.html.

However, I'm unsure if it's related to your application.

%% Main Script
tspan   = [0 10];   % simulation time
x0      = [1, 0];   % initial values, x1 starts from 1, x2 starts from 0
[t, x]  = ode45(@mySystem, tspan, x0);

% plot the result
plot(t, x, 'linewidth', 1.5), grid on
xlabel('Time'), ylabel('Amplitude')
legend('x_{1}(t)', 'x_{2}(t)')

%% 'System' function file (Double Integrator)
function dxdt = mySystem(t, x)
    dxdt    = zeros(2, 1);  % 2nd-order system described in column vector form
    
    % call myControl() function
    u       = myControl(x);
    
    % Ordinary Differential Equations in first-order derivatives
    dxdt(1) = x(2);     % x1' = x2
    dxdt(2) = u;        % x2' = u   <-- equivalent to x1" = u
end

%% 'Control' function file
function u = myControl(x)
    x1      = x(1);
    x2      = x(2);
    Kp      = 1;        % proportional gain
    Kd      = 2;        % derivative gain
    u       = - Kp*x1 - Kd*x2;
end