ABSTRACT
State-space–based control, also known as modern control, presents a series of advantages for the intensive usage of computers. The entire design and analysis can be performed offline on computers, minimizing tune-up times. All control gains can be calculated offline with conventional library functions. The control code is also standardized and can automatically be generated for further implementation on microcontrollers. State-space control can be seen as two components: a PI equivalent module and a sum-of-products linear compensation term. The overall system can easily be optimized with an estimation of the inductor current, which saves a current sensor in the overall hardware. The state-space control method provides an arrangement able to keep the same software structure independently of the offline optimization problem (pole selection for the control law or estimation of certain state variables). The software structure can thus be validated and the code implemented while having several parameters as key development parameters. The code can be tested and certified for fault-free operation without uncertainty, independently of the actual dynamic performance of the power supply. Offline setting of parameters can be done under the same real-time code.
