How to develop a Closed loop current control for converter

I have developed a Buck-Boost converter to charge a SuperCap from a source, in Buck mode and supply a load in Boost mode, I have created a voltage control loop, to maintain the Voltage level within the limit 2.7V.

I also need to include a current controller for maintaing the current to the reference value, (ex.10A)

I need support/guidance with this part.

[Sam Chak]

Hi @Santhosh

Could you provide the differential equations for the Buck-Boost converter?

If it is common practice to linearize the Buck-Boost converter around a steady-state operating point for small-signal analysis, then the transfer function models of the Buck-Boost converter become valid tools for analyzing the frequency response and designing controllers to simultaneously regulate both output voltage and current in a dual-loop control architecture.

[Santhosh]

Dear @Sam Chak,

Please find the equations, the transfer function, Simulink Model and the behaviour of the simulink model below,

Thank you for your kind support

Equations:

1. a) Buck Mode: Q1– Closed, Q2 – Open

1. b) Buck Mode: Q1 – Open, Q2 – Open

Transfer Function

2. a) Boost Mode: Q2 – Open, Q1 – Open

2. b) Boost Mode: Q2– Closed, Q1 – Open

Inductor Voltage & Current

Capacitor Current

Transfer Function

Simulink Model

Behaviour of the Model

[Sam Chak]

Hi @Santhosh

Thank you for your update. Since the decoupled transfer functions are available and are of 2nd order, PID controllers can theoretically be designed to regulate the voltage and current outputs satisfactorily. In fact, the pidtune() tool can make the design process even easier when the performance requirements are properly specified. What is the issue with the controller design?

By the way, the voltage plot shows excessive oscillatory behavior upon reaching the target limit of 2.7V. However, it is very important to note that this behavior is not mathematically reflected in the differential equations and the transfer function models. Have you verified if the Simulink model behaves almost similarly to the transfer function model?