Search

This work extends the wavelet modulation (WM) technique for operating three-phase ($3\phi$), five-level (5L) power electronic converters (PECs). The WM technique is extended by resolution segmentation of the synthesis scale-based linearly combined wavelet functions, which are used as switching signals. The resolution segmentation is created based on the dilation property of synthesis scale-based linearly combined wavelet functions. Each leg of the $3\phi$ VS, 5L dc–ac PEC is operated by two sets of resolution-segmented synthesis scale-based linearly combined wavelet basis functions to produce the voltage of one phase. The extended WM technique is implemented for performance testing, when operating $3\phi$ VS, 5L, diode-clamped and flying-capacitor dc–ac PECs. In this article, Part I, simulation test results are presented and compared with results obtained using the level-shifted pulsewidth modulation, phase-shifted PWM, and space vector modulation. Test results show significant reductions of output harmonics, along with improvements in fundamental components of output voltages and PEC efficiency. [ABSTRACT FROM AUTHOR]

This article presents the implementation and experimental testing of the wavelet-modulation technique for three-phase ($3\phi$) 5-level (5L) power electronic converters (PECs). This version of the wavelet modulation technique is structured as a multiresolution analysis that can support a redundant nonuniform recurrent sampling-reconstruction of three reference-modulating signals. The reconstruction of reference-modulating signals is achieved by sets of resolution-segmented scale-base linearly combined synthesis wavelet basis functions, which are used as switching signals to activate the switching elements of a $3\phi$ , 5L, VS dc–ac PEC. The presented wavelet-modulation technique is implemented in real time using a digital signal-processing board. Switching pulses generated by the extended wavelet-modulation technique are used to operate $3\phi$ , 5L, VS, diode-clamped and flying-capacitor dc–ac PECs. Experimental performance of the tested 5L dc–ac PECs is evaluated, when operated by the presented wavelet modulation, level-shifted pulsewidth modulation (PWM), phase-shifted PWM, and space vector modulation techniques. Experimental tests are conducted for linear, dynamic, and nonlinear loads fed by the tested 5L PECs. Test results show that magnitudes of output voltage fundamental components can be significantly increased, and harmonic distortions can be effectively reduced using the wavelet-modulation technique. These features are further demonstrated by performance comparisons with other techniques under similar operating conditions. [ABSTRACT FROM AUTHOR]

This paper presents the development and performance evaluation of a direct torque controller (DTC) for permanent-magnet synchronous motor (PMSM) drives fed by a 3\phi 6-pulse wavelet-modulated dc–ac power electronic converters. The developed DTC is designed to adjust the d-q-axis components of stator voltages (vd and vq) of a PMSM in response to changes in the load torque or drive speed. The desired adjustments in vd and vq are achieved by changing the angle $\vartheta$ of a frame created by \text{-hp} PMSM drive system. The performance of the frame-angle-based DTC is investigated for different changes in the command torque and drive speed. Simulation and experimental test results demonstrate stable, fast, and accurate responses that are complimented by negligible sensitivity to variations in the system parameters. [ABSTRACT FROM PUBLISHER]

This paper presents the implementation and performance evaluation of a coordinated antiislanding protection for systems with multiple distributed generation units (DGUs). The proposed coordinated antiislanding protection is structured to process the d-q-axis components of the instantaneous three phase apparent powers (s_{d} and s_{q}) determined at the point-of-common-coupling for each DGU. The processing of sd and sq, for each DGU, is carried out by the wavelet packet transform (WPT) in order to extract their low- and high-frequency subband contents. The contents of WPT low- and high-frequency subbands offer signature information that can facilitate detecting the islanding condition and identifying the islanded DGU(s). The coordinated antiislanding protection is implemented in real time for experimental testing on a laboratory collector system that has three different DGUs. Experimental results reveal fast and accurate responses to islanding events, accurate identification of islanded DGUs, and negligible sensitivity to the type or ratings of protected DGUs. In addition, test results show that the $d-q$ WPT-based coordinated antiislanding protection can accurately distinguish between islanding and nonislanding events, including faults, step changes in power delivery to the grid, unintentional loss of grid connection, low-voltage ride through, and sudden harmonic distortion on the grid side. [ABSTRACT FROM AUTHOR]

In this paper, the performance of a passive anti-islanding method is experimentally tested for three phase (3\phi) cogeneration systems. The tested method is based on determining the wavelet packet transform (WPT) high-frequency subbands present in the d– q-axis components of instantaneous 3\phi, when evaluated at the point of common coupling (PCC). This passive anti-islanding method is founded based on the nature of instantaneous 3\phi apparent powers that have components continuously exchanged between both sides of the PCC. An islanding condition can be considered as a transient disturbance that creates nonperiodic and nonstationary high frequency components in sd and sq. These frequency components can be parameterized by WPT high-frequency subbands, which can provide accurate detection of the islanding condition. The $d$– $q$ WPT-based anti-islanding method is tested for a $3\phi$ cogeneration system under various loading and power delivery conditions. Performance results reveal accurate, fast, and reliable detection and response to the islanding condition. [ABSTRACT FROM AUTHOR]

This paper proposes a new passive antiislanding method for three-phase (3φ) distributed generation units (DGUs). The proposed method is based on extracting signatures from the instantaneous 3φ apparent powers determined at the point of common coupling (PCC). This new method is found on the fact that the instantaneous 3φ apparent powers have components continuously exchanged between loads and sources. The islanding condition creates transient high-frequency components in the instantaneous 3φ apparent powers.These high-frequency components contain signature information capable of identifying the islanding condition. These transient high-frequency components can be extracted using the wavelet packet transform (WPT), when applied to the direct and quadrature (d-q-axis) components of the instantaneous 3φ apparent powers. The d-q WPT-based antiislanding method is implemented for testing on a 3φ permanent magnet generator-based wind energy conversion system. Test results demonstrate an accurate, fast, and reliable response to the islanding condition occurring when supplying different load types at different levels of power delivery to the host grid. [ABSTRACT FROM AUTHOR]

This paper presents the real-time testing of a digital protection for microgrids operated in utility-connection and stand-alone modes. The tested protection is structured with coordinated digital relays, which are featured with a fault identification based on the wavelet packet transform (WPT). The coordination of the employed digital relays is established based on the second-level high-frequency subbands extracted from the d-q-axis components of the currents flowing through each point of branching. The second-level high-frequency subbands are extracted and parametrized by two sets of WPT coefficients. The first set of WPT coefficients is utilized for identifying faults, while the second set is employed to locate identified faults. The d-q WPT-based digital multirelay protection is realized without time-based coordination, adaptive features, or relay-to-relay communication. Performances of the d-q WPT-based digital multirelay protection are investigated for a laboratory microgrid. Test results demonstrate accurate, fast, reliable, and coordinated responses, along with insensitivity to the mode of operating the microgrid. [ABSTRACT FROM AUTHOR]

This paper presents the implementation and performance evaluation of a new digital protection for microgrid systems, The proposed digital protection is based on utilizing the wavelet packet transform (WPT) for extracting the first-level high-frequency subband contents present in the $d$– $q$-axis current components. The utilized WPT is realized by a half-band digital high-pass filter, whose coefficients are determined by the Daubechies $db4$ wavelet basis functions. The $d$– $q$ WPT-based digital protection is implemented for offline and experimental performance evaluation. Offline tests are conducted using data collected from a simulation model of a microgrid located in Fermeuse, NL, Canada. The experimental tests are conducted on a laboratory microgrid system that is constructed as a scaled version of the Fermeuse microgrid. Several transient disturbances occurring in different parts of the test microgrid systems are investigated both offline and in experimentation. Performance results demonstrate significant abilities of the $d$– $q$ WPT-based digital protection for fast, accurate, and reliable detection, classification, and response to different transient disturbances. In all tested faults, the developed digital protection is able to generate a trip signal in less than a quarter cycle based on the 60-Hz system operating frequency. [ABSTRACT FROM PUBLISHER]

This paper presents the implementation and performance evaluation of an embedded digital protection for three-phase $(3\phi)$ interior permanent-magnet synchronous motor (IPMSM) drives. The presented digital protection is established through extracting the high-frequency subband present in the $d$– $q$-axis components of the currents drawn by the protected IPMSM. The desired high-frequency subband is extracted using the wavelet packet transform (WPT) from the $d$– $q$-axis current components, which are obtained from the controller of the IPMSM drive. The extracted high-frequency subband contains signature information that allows detecting and classifying transient disturbances into fault and nonfault events. The presented method is insensitive to the variations in the drive parameters, loading levels, nature of electrical faults (open circuit or short circuit), and operation of the drive controller. These features of the $d$– $q$ WPT-based digital protection simplify its implementation and facilitate its embedding within the controller of the IPMSM drive. The performance of the $d$– $q$ WPT-based digital protection is experimentally evaluated on a laboratory 5-hp IPMSM drive for various fault and nonfault events. Test results demonstrate fast, accurate, and reliable detection, classification, and response to fault and nonfault events. [ABSTRACT FROM PUBLISHER]

The switching pulses of the wavelet modulation technique are generated as sets of dilated and shifted scale-based linearly combined synthesis wavelet functions. The duration and location of each switching pulse are set by the scale (resolution level) at which its associated synthesis wavelet function is generated. In this paper, a novel controller is developed and tested for three phase (3\phi) voltage source (VS) six-pulse wavelet-modulated ac–dc converters. The proposed controller is structured to adjust the maximum value reached by the scale. This maximum value, J, is adjusted based on two angles. The first angle, \thetaI, maps changes in the 3\phi supply voltages to adjustments in J, while the second angle, \thetaO, maps changes in the dc output voltage to adjustments in J. The net adjustment in J is optimized based on the mapped changes by \thetaI and \thetaO. The optimal resolution-level controller is implemented in order to evaluate its performances for operating 3\phi VS six-pulse ac–dc converters. The performances of the proposed controller are tested when the converter is supplied with balanced, unbalanced, and distorted 3\phi voltages for feeding static and dynamic loads. Test results demonstrate fast and accurate responses, stable functioning, and good dynamics to ensure that the output dc voltage follows its command value for different supply and loading conditions. [ABSTRACT FROM PUBLISHER]

This paper presents the implementation and performance testing of the wavelet modulation technique for operating three phase, voltage source (VS), six-pulse ac–dc converters. The wavelet modulation technique is realized by a nondyadic-type multiresolution analysis (MRA), which is constructed using sets of dilated and translated scale-based linearly combined wavelet basis functions. A dc reference signal is processed using this MRA, where three sets of groups of nonuniform recurrent samples are created by the analysis stage. The synthesis stage of the nondyadic MRA reconstructs the dc reference signal using three sets of dilated and translated scale-based linearly combined synthesis wavelet basis functions, which are used to activate the switching elements of the ac–dc converter. Simulation and experimental performances of the 3\phi ac–dc converter, that is, operated by the wavelet modulation technique, are investigated for supplying static and dynamic load types. Performances of the 3\phi wavelet modulated ac–dc converter are also investigated for unbalanced input 3\phi voltages. Simulation and experimental results show that high magnitude of output dc components, and significant reductions of input and output harmonic components of the 3\phi VS ac–dc converter can be accomplished using the wavelet modulation technique. These improvements in the performances of 3\phi ac–dc converters are further demonstrated through comparisons with the pulse-width and space-vector modulation techniques under similar conditions of loading and 3\phi input voltages. [ABSTRACT FROM AUTHOR]

This paper presents the performances of a new method for detecting ice accumulation on wind turbines. The presented method is based on constructing a multiresolution analysis (MRA) to extract frequency components present in the electric currents flowing out of an electric generator driven by a wind turbine. The foundations of the proposed ice detection method are established based on the fact that ice accumulation leads to a slow increase in the wind turbine inertia, which triggers pulsations in the electromagnetic torque of the electric generator. Such torque pulsations create certain frequency components that can be extracted from the direct and quadrature components of the electric generator output currents. The Daubechies db6 basis functions are selected for constructing the desired MRA that will extract the frequency components from the electric generator output currents. The MRA-based ice detection method is implemented for simulation and experimental testing for two types of electric generators driven by wind turbines. Simulation and experimental performances demonstrate significant capabilities for fast, accurate, and consistent detection of ice accumulation on wind turbines. Furthermore, these performance results show that the responses of the proposed method are insensitive to turbine specifications, wind speed, and/or generator type. [ABSTRACT FROM PUBLISHER]

This paper presents the development, implementation, and performance testing of a permanent-magnet generator (PMG)-based wind energy conversion system (WECS) for grid-connected applications. The grid-connected operation is constructed using the new three-phase resolution-level-controlled wavelet-modulated inverter. The output of the wind generator is fed to a switch-mode ac–dc converter, where the dc voltage is controlled to ensure extracting power from the wind under variable wind speed conditions in order to meet the grid active and reactive power demands. The resolution-level control of the grid-side inverter is designed as a current controller for adjusting active and reactive powers delivered to the grid. The proposed WECS is implemented both in software and hardware for performance testing on a direct-drive 6-kW laboratory PMG operated with variable speed. Test results demonstrate that an accurate control of the dc voltage on the generator side ensures adjusting the generator speed to extract power at each wind speed in order to meet the demand active and reactive power delivery to the grid. Also, test results show significant abilities of the resolution-level controller to initiate fast and accurate adjustments in the active and reactive powers delivered to the grid in order to follow any changes in their demand values under variable wind speed. [ABSTRACT FROM AUTHOR]

This paper presents the real-time implementation and experimental performances of the wavelet-modulation technique for single-phase voltage-source (VS) inverters. The wavelet-modulation technique is realized through constructing a nondyadic-type multiresolution analysis, which supports sampling of a sinusoidal reference-modulating signal in a nonuniform recurrent manner, then reconstructing it using the inverter-switching actions. The required nonuniform recurrent sampling is carried out by using dilated and translated sets of wavelet basis functions, which are generated by the scale-base linearly combined scaling function. The reconstruction of the sampled signal is accomplished by using dilated and translated sets of wavelet basis functions, which are generated by the scale-base linearly combined synthesis scaling function. The dilated and translated sets of wavelet basis functions used in the reconstruction are employed as switching signals to activate the inverter-switching elements. The wavelet-modulation technique is implemented in real time by using a digital signal processing board to generate switching pulses for a single-phase VS H-bridge (four-pulse) inverter. Experimental performances of the single-phase inverter, which is operated by the wavelet-modulation technique are investigated while supplying linear, dynamic, and nonlinear loads with different frequencies. Experimental test results show that high magnitude of fundamental components and significantly reduced harmonic contents of the inverter outputs can be achieved using the wavelet-modulation technique. The efficacy of the developed modulation technique is further demonstrated through performance comparisons with the pulsewidth- and random-pulsewidth-modulation techniques for similar loading conditions. [ABSTRACT FROM AUTHOR]

This paper presents the analysis, development, and implementation of a wavelet modulation (WM) technique for three-phase voltage-source (VS) six-pulse inverters. The WM technique is based on sampling three continuous-time (CT) sinusoidal reference modulating signals in a nonuniform recurrent manner using three sets of scale-based linearly combined wavelet basis functions. These CT signals are recovered from their samples by a three-phase VS six-pulse inverter, which is activated by three sets of synthesis scale-based linearly combined wavelet basis functions. Each set of synthesis basis functions is generated for activating one leg of the three-phase inverter in order to recover one CT reference modulating signal. The WM technique is implemented for both simulation and experimental performance testing. The performances of the WM technique are compared with those obtained using the space vector modulation, random pulsewidth modulation, and hysteresis band current control techniques under the same loading conditions. Simulation and experimental test results show that the proposed WM technique is able to switch a three-phase VS six-pulse inverter to produce outputs with significantly improved fundamental components and low harmonic contents. [ABSTRACT FROM PUBLISHER]

This paper presents the development, implementation, and performances of a resolution-level vector controller (RLVC) for three-phase induction motor (IM) drive systems. The RLVC is structured to adjust the output voltages of a three-phase voltage-source wavelet-modulated inverter which is used to supply a three-phase IM. The objectives of the RLVC are established by controlling the quadrature axis component of the stator currents while maintaining a constant direct axis component of the stator currents. The complete IM drive system incorporating the RLVC is successfully implemented in real time for performance testing using a three-phase squirrel-cage IM. The performances of the proposed IM drive system are tested for sudden changes in load torque and command speed. Simulated and experimental performances demonstrate reliable, stable, fast, and accurate responses of the proposed controller under various operating conditions. [ABSTRACT FROM PUBLISHER]

This paper presents an extension of real-time tests of a wavelet-packet-transform-based technique for the differential protection of three-phase power transformers. The proposed technique is implemented using a DS1102 digital signal processor board and tested on two different three-phase power transformers with neutral resistance grounded. Different magnetizing inrush and internal fault currents are investigated with current transformer (CT) saturation for different loading conditions, including capacitive loads. The results show a complete independence from transformer parameters, load types, grounding method, or CT saturation. Furthermore, the proposed technique has high speed, good accuracy, small required memory, and reliable responses with reduced computational burden. In all cases of the investigated internal faults, the proposed algorithm is capable of identifying the fault and generating a trip signal in less than a quarter cycle based on a 60-Hz system. [ABSTRACT FROM AUTHOR]

This paper presents real-time implementation and testing of a wavelet-packet-transform-based algorithm for power transformer protection. The proposed algorithm is implemented using a DS1102 digital signal processor board and tested on two different three-phase power transformers. Different magnetizing inrush, normal (through-fault) and internal fault currents are investigated for different loading conditions. The results show a complete independence from both the transformer parameters and current transformer saturation. In addition, the proposed algorithm has reduced computational burden, high speed, high accuracy, small required memory, and high reliability. In all cases of the investigated internal faults, the proposed algorithm is capable of identifying the fault and generating a trip signal in less than a quarter cycle based on a 60-Hz system. [ABSTRACT FROM AUTHOR]