I am designing an electric spring for mitigating power quality issues. I would like to know which is the most new and advanced control strategy (like PI controller) for controlling half bridge inverter.
The electric spring I am going to design is realized using a single phase half bridge inverter which when connected in series with a non-critical load forms a smart load. This smart load would act to provide Demand Side Management for DC micro grids. I would like to find the best control strategy for this inverter.
Half-bridge suggests a topology which has two push-pull transistors. Between them is connected one lead to your step-up transformer primary.
The other lead is connected between two stacked capacitors, acting in a 'springy' give-and-take. This arrangement mimics an H-bridge. As a result, AC square waves are applied to the transformer primary.
It is easy to alternately switch the transistors, at 50% duty cycle. However you wish to regulate the voltage to the load. This is the difficult part of the design job, namely to switch the transistors at the proper duty cycle and frequency.
before jumping to exotic control you should first determine if its actually needed. the beauty of PI control is that it is easy to understand, easy to implement, and the best part easy to prove stability over the applied range of use.
The electric spring I am going to design is realized using a single phase half bridge inverter which when connected in series with a non-critical load forms a smart load. This smart load would act to provide Demand Side Management for DC micro grids. I would like to find the best control strategy for this inverter.
SMart Loads suggest a non-linear diode like VI characteristic, which is as we (EE's) all know, is very inefficient, low PF and high THD.
"Electric Spring" suggest a linear response like F=kX with force constant k.
THe only effective linear front end is an active PFC circuit with makes the load look linear and the step-up design will have a finite ESR source impedance and thus with impedance transformed by n^2 will drop with load impedance, which is non-linear unless with a PFC on the secondary. ( and primary)
So your spec is extremely vague and has zero measurable terms in it.
It is neither linear nor spring like, although any proportional feedback may be considered linear, it does not mean the current (pulses) are linear unless designed with suitable specs.