The energy performance of various types of voltage-source and current-source converters is examined. For fairness and completeness, efficiency is calculated for three major battleground scenarios. The first is a low dynamic nonregenerative group of applications such as pumps, fans, and compressors. This group represents 85% of high power (2 MW) industrial applications where energy savings are usually a primary consideration justifying investment. The second scenario considers applications requiring good dynamic response and regenerative braking. Finally, the third group considers very high power applications (over 20 MW). The evaluation presented takes into account semiconductor switching and conduction losses, losses in the medium voltage feeding transformer (determined per IEEE Standard C57.18.10-1998), and the losses in ac and dc filters. For purposes of analysis, computer simulations validated against measurements taken on a 1-MW voltage source inverter (VSI) and a 1.4-MW current source inverter (CSI) were used. The results of the first scenario show competitive efficiencies for VSI and CSI drives, whereas voltage source-based solutions are more energy efficient in the second scenario considered. For the last group, the current source load-commutated inverter exhibits the best performance.

On the Efficiency of Voltage Source and Current Source Inverters for High-Power Drives

This paper addresses the important issue of power quality management for smart grids and proposes a load management strategy based on transformer derating for minimizing harmonic distortion in distribution feeders and transformers. Ongoing development of smart grid technologies such as smart metering and smart appliances are creating new opportunities for improving distribution system performance. One area undergoing study is effective control of demand response through (semi)automated load management practices (e.g., smart appliances). Despite these developments, the impact on power quality has not been taken into consideration from a demand side management point of view. Smart grids provide an excellent opportunity to better manage power quality and reduce harmonic distortions present in power networks. In this paper, it is proposed that the impact of harmonics generated by nonlinear loads should be factored into overall load control strategies of smart appliances. This work focuses on the impact on residential distribution transformers which are adversely impacted by harmonic current distortions. A growing concern is the potentially high penetration of plug-in electric vehicles in smart grids. Load management of electric vehicles is studied for an IEEE 30-bus 23 kV distribution system to demonstrate the benefits of the proposed power quality and load management strategy. This paper proposes computing transformer K-Factor derating to control scheduling of smart appliances/loads to reduce harmonic stresses.

Load management in smart grids considering harmonic distortion and transformer derating

A new analysis into the steady-state operation and derating of three-phase transformers under nonsinusoidal and asymmetric operating conditions is proposed. The combined effects of transformer core and load asymmetry, nonlinearity, and harmonics, as well as nonsinusoidal input excitation are examined. A time-domain nonlinear model for three-phase three-leg transformers is implemented. Transformer derating is estimated by modeling additional power losses due to harmonics generated by the iron core, nonsinusoidal excitation, and nonlinear (rectifier and electric drive) loading. Laboratory tests are performed to verify simulated waveforms. The contribution of this paper is a nonlinear transformer modeling technique for steady-state operation under unbalanced, asymmetric, and nonsinusoidal operation, capable of computing derating factors.

/pdf/derating-of-asymmetric-three-phase-transformers-serving-36cuzz5hxp.pdf

Derating of Asymmetric Three-Phase Transformers Serving Unbalanced Nonlinear Loads

A power amplifier is used to supply sinusoidal currents of different frequencies for measuring eddy-current losses of a 25 kVA single-phase transformer under short-circuit condition. Measured data show that eddy-current loss is a linear function of frequency with power of 2 and the eddy-current loss coefficient P/sub EC-Rlinear/ is computed. New measurement techniques are applied to determine the derating of single-phase transformers with full-wave diode and thyristor rectifier loads. The derating of transformers has been defined such that for the (apparent, real) power transfer of a transformer the total losses are identical to the rated losses at rated temperature. A relation between apparent power, derating and K-factor is given taking into account iron-core and stray-power losses. Measured derating values are compared with computed results based on the eddy-current losses, iron-core losses, stray-power losses, and K-factors. The eddy-current loss coefficient P/sub EC-Rnonlinear/ is computed from harmonics caused by diode/thyristor bridge loads.

Measurement of eddy-current loss coefficient P/sub EC-R/, derating of single-phase transformers, and comparison with K-factor approach

No-load loss calculation of distribution transformers supplied by nonsinusoidal voltage using three-dimensional finite element analysis

The use of adjustable-speed drives requires transformers capable of withstanding high levels of harmonic currents under normal operating conditions. Experience has been that overheating problems are much more common with dry-type transformers than with liquid-filled transformers. Transformer insulation life is determined by the hot spot temperature but confirmation of hot spot temperature rise is one performance characteristic which is ignored in industry standards. This is especially important for transformers rated for nonsinusoidal load currents. The design of transformers for nonsinusoidal load currents should include an analysis of the eddy loss distribution in the windings and calculation of the hot spot temperature rise. Calculations and thermal tests giving only average winding temperature rises are not sufficient. Thermal tests with nonsinusoidal currents and measurements of hot spot temperature rises are extremely difficult on large transformers. The combination of testing and analysis may be the only economically practical approach. Analysis indicates that the dry type transformer hot spot temperature is very sensitive to the eddy loss magnitude and distribution. The Underwriters Laboratories Inc. (UL) K-factor rated dry type transformer and the recommended practices given in ANSI/IEEE C57.110 are reviewed. When purchasing transformers subject to nonsinusoidal load currents, considerations should be given to the manufacturer's development program and capability to calculate the eddy loss distribution and hot spot temperatures.

L.W. Pierce

Papers

Transformer design and application considerations for nonsinusoidal load currents