To assure power quality and stability of the electricity network it is important to perform control and synchronization of the inverters which are used in the connection of distributed DC power generation units to the grid. It is proven that the model of the inverters, is a differentially flat one. By exploiting differential flatness properties it is shown that the multiple inverters model can be transformed into a set of local inverter models which are decoupled and linearized. For each local inverter the design of a state feedback controller becomes possible, e.g. using pole placement methods. Such a controller processes measurements not only coming from the individual inverter but also coming from other inverters which are connected to the grid. Moreover, to estimate the nonmeasurable state variables of each local inverter, the Derivativefree nonlinear Kalman Filter is used. This consists of the Kalman Filter recursion applied to the local linearized model of the inverter and of an inverse transformation that is based on differential flatness theory, which enables to compute estimates of the state variables of the initial nonlinear model of the inverter. Furthermore, by redesigning the aforementioned filter as a disturbance observer it becomes also possible to estimate and compensate for disturbance terms that affect each local inverter.
Nonlinear synchronizing control of parallel inverters connected to the electricity grid
SIANO, PIERLUIGI;
2015-01-01
Abstract
To assure power quality and stability of the electricity network it is important to perform control and synchronization of the inverters which are used in the connection of distributed DC power generation units to the grid. It is proven that the model of the inverters, is a differentially flat one. By exploiting differential flatness properties it is shown that the multiple inverters model can be transformed into a set of local inverter models which are decoupled and linearized. For each local inverter the design of a state feedback controller becomes possible, e.g. using pole placement methods. Such a controller processes measurements not only coming from the individual inverter but also coming from other inverters which are connected to the grid. Moreover, to estimate the nonmeasurable state variables of each local inverter, the Derivativefree nonlinear Kalman Filter is used. This consists of the Kalman Filter recursion applied to the local linearized model of the inverter and of an inverse transformation that is based on differential flatness theory, which enables to compute estimates of the state variables of the initial nonlinear model of the inverter. Furthermore, by redesigning the aforementioned filter as a disturbance observer it becomes also possible to estimate and compensate for disturbance terms that affect each local inverter.I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.