Modeling and Optimization of a Zero VoltageSwitching Inverter for High Efficiency and Miniaturization
Abstract
In a Zero-Voltage Switching (ZVS) inverter, high conversion efficiency and miniaturization are expected since switching loss can be dramatically reduced with proper design. In order to realize ZVS condition, auxiliary components such as inductors, capacitors and switches are embedded to the inverter to implement the function. Since the design of auxiliary components is critical to the ZVS inverter, it is impossible to realize maximum efficiency or minimum size by following the conventional design procedure. This paper introduces an optimized design methodology for a three-phase ZVS inverter with objectives of both high efficiency and miniaturization. Based on the loss models of different commercial IGBT modules under different ZVSconditions, as well as the loss models of auxiliary components and filter inductors, the issue of pursuing highest efficiency is transformed into solving a constrained nonlinear multivariable problem. According to the proposed design methodology, all parameters that influence the efficiency and physical dimensions are considered simultaneously. Thus the optimized selection of height module, the parameters of the auxiliary components and the filter inductors would be obtained. A 30 kW three-phase ZVSinverter prototype was built to verify the proposed design method. With proposed design method, the improved prototype has achieved both smaller passive components volume and higher efficiency compared to the former prototype.
EXISTING SYSTEM:
One of the main drawbacks of soft-switching inverters is the voltage stress on the switches. The resonant dc link inverter has a very simple auxiliary circuit, but the voltage stress across the switches is high. Besides the output power quality may be an issue since discrete pulse modulation is utilized. Pulse width modulation and active clamping are utilized to improve the output power quality and voltage stress in . However with proposed modulation scheme, the soft switching can be only realized in low dc voltage utilization condition. The inverter demonstrated in utilizes a simple auxiliaries. the voltage stress a cross the main switches of this inverter is reduced. However, the voltage across the auxiliary switch is still much higher than the dc source. The negative bus auxiliary resonant circuit inverter. Achieved low voltage stress for all switches. However proper ZVS-on cannot be achieved in this inverter, which means the turn-on switching loss cannot be eliminated. In addition, inverters with complicated auxiliary circuits were reported in achieve low voltage stress across all switches. A all of the soft-switching inverters, the zero-voltage switching (ZVS) inverter with ZVS-SVM shown in Fig.1presents several advantages : All switches including auxiliary switch realize ZVS-on and ZVS-off; Reverse recovery losses of all freewheeling diodes are eliminated;
PROPOSED SYSTEM:
This paper proposes a design methodology to optimize both efficiency and size of the ZVS inverter. The basic operatingprinciples of the ZVS inverter are introduced in Section II toobtain the key waveforms of the switching modules andresonant components.Soft-switching techniques require additional componentssuch as capacitors, inductors and auxiliary switches. In order tosimplify the design procedures, the conventional methodologyis employed to fix a subset of design variables and to tuneseveral of the remaining variables. Following observation,another subset of design variables is set by designers’experience or assumptions and the abovementioned proceduresare repeated until the results are acceptable. In view of theconsiderable dependency of the design variables in the ZVSinverter, it may take a lot of effort to iterate the design with aconventional design methodology, and furthermore the finaldesign may not be the optimum one. Mathematical optimization techniques may provide a moreeffective way to allow all design variables and constrainedconditions to be considered simultaneously. With the aid ofmodern computers’ high processing speed, the optimum designresults will be computed in short time compared to theconventional methodology, and particularly, without the subjective influence of designers. The early attempt in applying mathematical optimization techniques to solve analog circuitdesign problems was published in the mid-1970s. Earlyresearch work introduced the sequential unconstrainedminimization technique (SUMT) to power electronics circuitdesign in the late 1970s. In , SUMT and the augmentedLagrangian (ALAG) penalty function technique werecompared. With the ALAG penalty function technique, theconstrained optimization problem can be transferred into anon-constrained optimization problem by constructing anaugment penalty function.
CONCLUSION
Conventional design methodologies for a ZVS inverter involve a subset of design variables and tuning several of theremaining design variables and observing the results. This stepis repeated, which requires extensive efforts to obtainacceptable results. This paper utilizes an optimizationalgorithm to optimize the efficiency and dimensions of the ZVSinverter. In order to evaluate the turn-off loss of the switchingmodules, measured data of turn-off loss under differentconducting currents and buffering capacitances are utilized tobuild the loss models. The dimensions of the main passive components in the circuits such as the resonant inductors,clamping capacitors and filter inductors are all modeled andconsidered in the design procedure. The optimum efficienciesunder different switching frequencies and volumes of thepassive components for five different IGBT modules wereobtained with the proposed design method. With the optimumselection of the switching module, switching frequency andvolume limitation, a 30 kW prototype was built and tested.According to the optimized design, the efficiency of theprototype has been increased by 0.2~0.5% compared to theprototype without optimization. Besides, the volume of theresonant inductor, clamping capacitor and the filter inductorhave been reduced by 61%, 76% and 73%, respectivelycompared to the former prototype.
REFERENCES
[1] B. K. Bose. “Energy, environment, and advances in power electronics,”IEEE Trans. Power Electron., vol. 15, no. 4, pp. 688–701, 2000.
[2] S. Milunovich and J. Rasco, “The sixth revolution: The coming ofcleantech,” Merrill Lynch, New York, 2008.
[3] Frankfurt School-UNEP Collaborating Centre for Climate & SustainableEnergy Finance, the United Nations Environment Programme (UNEP)and Bloomberg New Energy Finance, Global trends in renewable energyinvestment, 2014.
[4] European Photovoltaic Industry Association, Global market outlook forphotovoltaics 2014- 2018, [Online], available: http://www.epia.org,2014.
[5] F. Blaabjerg, C. Zhe, S. B. Kjaer, “Power electronics as efficient interfacein dispersed power generation systems,” IEEE Trans. Power Electron.,vol. 19, no. 5, pp. 1184-1194, 2004.
[6] M. D. Bellar, T. Wu, A. Tchamdjou, J. Mahdavi, and M. Ehsani, “Areview of soft-switched DC-AC converters,” IEEE Trans. Ind. Appl., vol.34, no. 4, pp. 847-860, 1998.
[7] D. M. Divan, “The resonant DC link converter-a new concept in staticpower conversion,” IEEE Trans. Ind. Appl., vol. 25, pp. 317-325, 1989.
[8] G. Venkataramanan, D.M. Divan, “Pulse width modulation with resonantDC link converters,” IEEE Trans. Ind. Appl., vol.29, no.1, pp. 113-120,1993.
[9] R. W. De Doncker, J. P. Lyons, “The auxiliary resonant commutated poleconverter,” in IEEE IAS annual meeting, pp. 1128-1235, 1990.