Doctoral thesis deals with optimal sun-tracking of a photovoltaic system considering the electric drive losses. The Sun tracking PV system assures that the highest possible share of the available solar radiation reaches the surface of the PV modules. The electric drive which enables tracking is considered as the loss of the energy produced in the PV system. The maximum of the energy produced in the PV system is achieved by the continuous tracking of the PV system. Since the electric drives are determined by constant speed and time, and angle quantization the maximum of the energy produced can only approximate. The doctoral thesis presents a new method for determining such trajectories of the PV modules that change the position of the PV modules in such a way that the production of the electric energy in the given time interval of the observation reaches its maximum. The goal is to determine the maximum efficiency of the PV tracking system considering the tracking system energy consumption. To achieve this, exact values of the available solar energy are needed for a given moment. To do this, a new method for predicting direct and diffuse solar radiation on the Earthćs surface, in the form of the time dependent function, is developed. The method is confirmed by the comparison of the measured and the predicted solar radiation for clear days. The developed method for predicting the solar radiation in the form of the time dependent function and energy consumption of the tracking system, given as the functions of the azimuth and tilt angle change, are applied together to determine those trajectories of the PV module, where the PV system energy production, gives the maximum. To find a solution of the nonlinear and bounded optimization problem, a stochastic search algorithm called DifferentialEvolution is applied. The approach for determining the maximum of the energy produced in the PV system has not been researched yet. The explicitly defined objective function, which is minimized in the optimization procedure considering the optimization bounds, is used. Thus it is assured the maximum of the possible energy produced in the PV system, considering the applied model of the PV system, tracking system consumption, predicted solar radiation, and the properties of the applied optimization method. The use of the different and more advanced models of the PV system, the sun tracking system consumption or prediction of the solar radiation, can lead to the different optimal trajectories of the sun tracking system. However, this cannot reduce the importance of the proposed method. The proposed method gives the maximum of the possible energy produced in the discussed PV system, considering the applied models and data.
D.09 Tutoring for postgraduate students
COBISS.SI-ID: 14925078In control realizations, the parameters of electric machines realizations are often treated as constant, which is not truth especially in the case of machines with an iron core. Variation of machine parameters due to the magnetically nonlinear behavior of the iron core can be considered by changing static and dynamic inductance. The use of static inductance can lead to stability problems especially in the case of sensorless control realizations. Such problems are unknown when dynamic inductance is used. A proper use of dynamic inductances can substantially improve performances current controllers in existing control realizations.
F.07 Improvements to an existing product
COBISS.SI-ID: 14884630Three-phase voltage sources are normally required to determine parameters of three-phase permanent magnet machines. The results of investigation show the also single phase voltage sources in the form of inverters or linear amplifiers can be successfully applied to determine parameters of three-phase permanent magnet machines.
F.02 Acquisition of new scientific knowledge
COBISS.SI-ID: 14883606