EPE Association: EPE 2011 - DS3d: Topic 08: Measurements and Sensors

EPE 2011 - DS3d: Topic 08: Measurements and Sensors
You are here: EPE Documents > 01 - EPE & EPE ECCE Conference Proceedings > EPE 2011 ECCE Europe - Conference > EPE 2011 - Topic 08: Measurements and Sensors > EPE 2011 - DS3d: Topic 08: Measurements and Sensors

   [return to parent folder]

   A digital platform for real-time simulation of power converters with high switching frequency
By Kevin DE CUYPER, Michel OSEE, Frederic ROBERT, Pierre MATHYS [View]
[Download]

Abstract: In this paper, we propose a new implementation method for the rapid prototyping of power converters. AField-Programmable Gate Array emulates the temporal equations of the device while analog interfacesallow to use the platform to test control and protection systems in a nearly seamless way.A description of the converter based on an ideal on-off model for the power switches is used in orderto obtain multiple topologies for the same device. Each on these topologies is described by its ownequations and corresponds to a specific set of states of the switches. A topology selector based on thecontrol signals and the internal currents and voltages allows to find the correct equations in less than 10ns and a mid-point linear solver calculates the outputs in a few hundreds of nanoseconds. The solver also takes care of the passive switches (i.e. diodes) to ensure that they don’t go outside their current/voltage limit values.The whole process is described with a boost converter, and results show the great accuracy of our method even for high (> 100kHz) switching frequencies and sampling periods of less than 1μs.

   A fast estimation algorithm for low-frequency oscillations in power systems
By Mebtu BEZA, Massimo BONGIORNO [View]
[Download]

Abstract: This paper deals with the application of the Recursive Least Square (RLS) algorithm for estimation oflow-frequency oscillations. A novel resetting method to improve the speed of response of the estimationwill be described. The effectiveness of the proposed method will be proven through simulation as wellas experimental verification.

   A Fully Automated Measurement Set-up for the Determination of the Reverse Recovery Behaviour of Ultra-Fast Diodes
By Juergen STAHL, Daniel KUEBRICH, Thomas DUERBAUM, Janina PATZ, Anne-Christine LEICHT [View]
[Download]

Abstract: Often data sheets provide only poor information about the recovery behavior of ultra-fast diodes. Onthe other hand, existing diode models do not predict the real characteristic of all diodes. Nevertheless,due to its importance, this behavior needs to be known and therefore measured. For this purpose, afully automated measurement set-up for determining the reverse recovery characteristic of ultra-fastdiodes in an accurate manner was designed and is described here. All the data obtained is immediatelytransferred into MATLAB and therefore available for further calculation, model building and modelvalidation. Since the whole set-up is automated, a complete field of variations in the reverse voltage,the forward current, the temperature, and the rate of current drop di/dt can be easily applied to thetested diode. Hence, a lot of information can be obtained effortlessly. This uncomplicatedmethodology makes it readily available for circuit designers, allowing them to predict the contributionof the reverse recovery of rectifiers to the total losses more accurately. In addition, a real comparisonof different diodes at many operation points is possible.

   A new Digital Modulated Interface for a Precise and Robust Current Sensor
By Abdoulkarim BOUABANA, Constantinos SOURKOUNIS [View]
[Download]

Abstract: Compared to conventional current sensors, the proposed topology is advantageous because of its high resolution throughout its full measurement range, thus reducing offsets to a minimum. This aim is reached through active offset compensation using an analog/digital chopper/de-chopper. A further advantage is the excellent suppression of 1/f noise and a low noise density (35 nV/√Hz) down to the DC-threshold. Common current sensors may be replaced with this technology whose one output is likewise analog. The power supply requirements are also far lower than for standard topologies as only a single supply is needed. The F-PWM modulation method and therefore the current sensor, results for current control units a high performance. The measurement current is directly coded in a “digital current value”, which can be decoded and used for the control unit. An AD-Converter is not more necessary for signal conversion, compared to the typical analog output current sensor based on the Hall-Effect. The coded signals can be transmitted potential- free with an optocoupler.

   A simple calorimetric setup for the accurate measurement of losses in power electronic converters
By Georgios DIMITRAKAKIS, Emmanuel TATAKIS, Anastasios NANAKOS [View]
[Download]

Abstract: This paper deals with the analysis and implementation of a low cost calorimetric setup, suitable for loss measurements in power electronic converters, for power ratings up to 30W. Main component of the device is an open Dewar vessel. A proper calibration method is suggested and the several error factors are spotted, so as to achieve an overall accuracy of 2\%. Application examples are demonstrated with loss measurements on magnetic components.

   Accuracy study of frequency converter estimates used in the sensorless diagnostics of induction-motor-driven systems
By Tero AHONEN, Jussi TAMMINEN, Jero AHOLA, Markku NIEMELÄ [View]
[Download]

Abstract: Modern frequency converters allow energy efficient operation of induction-motor-driven systems such as pumps and fans. In addition, they can be used as a sensorless monitoring unit for the pump and fan systems, as shown by previous studies. Sensorless diagnostics of induction-motor-driven systems relies on the accurate estimation of the rotational speed, shaft torque and shaft power of the motor, which is why all frequency converters are not suitable for this use. In this paper, accuracy of the rotational speed, shaft torque and shaft power estimates is studied by laboratory tests for two different frequency converters. Using the test results, estimation accuracy of a vector- and a direct-torque-controlled frequency converter are compared with each other. The effect of erroneous motor estimates on the pump flow rate estimation is also demonstrated in the paper.

   Analysis Of The Technology OF Cable Fault Detection And Location On-Line
By Chen WEI, Wang LI [View]
[Download]

Abstract: In this paper, two new cable fault detection and location on-line methods--Sequence Time Domain Reflectometry method (STDR) and Spread Spectrum Time Domain Reflectometry method (SSTDR) were introduced and analyzed. An improved frequency correlation algorithm for real-valued sequences was introduced to cable fault detection and location on-line; it greatly reduced the time and the storage consumption of the detection system. The simulation models of these two methods were established; the feasibility of these two methods and the algorithm was validated by simulation and then compared these two methods, Simulation and comparison results show that SSTDR is a universal cable fault detection and location on-line method.

   Evaluation of Machine Learning Techniques for Electro-Mechanical System Diagnosis
By Miguel DELGADO, Antoni GARCIA, Julio URRESTY, Jordi-Roger RIBA, Juan Antonio ORTEGA [View]
[Download]

Abstract: The application of intelligent algorithms, in electro-mechanical diagnosis systems, is increasing inorder to reach high reliability and performance ratios in critical and complex scenarios. In this context,different multidimensional intelligent diagnosis systems, based on different machine learningtechniques, are presented and evaluated in an electro-mechanical actuator diagnosis scheme. The useddiagnosis methodology includes the acquisition of different physical magnitudes from the system,such as machine vibrations and stator currents, to enhance the monitoring capabilities. The featurescalculation process is based on statistical time and frequency domains features, as well as timefrequencyfault indicators. A features reduction stage is, additionally, included to compress thedescriptive fault information in a reduced feature set. After, different classification algorithms such asSupport Vector Machines, Neural Network, k-Nearest Neighbors and Classification Trees areimplemented. Classification ratios over inputs corresponding to previously learnt classes, andgeneralization capabilities with inputs corresponding to learnt classes slightly modified are evaluatedin an experimental test bench to analyze the suitability of each algorithm for this kind of application.

   Ground fault detection method for three phase inverters based on the third harmonic
By Tommi KÄRKKÄINEN, Kimmo RAUMA, Pertti SILVENTOINEN [View]
[Download]

Abstract: A new ground fault detection method based on third harmonic injection and the measurement of third harmonic current is presented. As the third harmonic is a common mode signal in a three phase system, its existence in current is a sign of a ground fault.

   Implementing clamp on wireless torque measurement system for rotating shaft applications
By Ville NISKANEN, Jero AHOLA [View]
[Download]

Abstract: Typically mechanical torque measurement is required for the verification of simulation model of systems consisting of electrical drive, electric motor, load, and control system. It is also possible to detect faults in system by torque measurement. In this paper a wireless measurement system to measure mechanical torque and its oscillations on a shaft of electric motor is studied and developed. The operation of the system is verified by laboratory tests. The wireless torque measurement system is based on a strain gauge collar which can be installed around the shaft of the motor. Installation and the use of the wireless torque measurement system does not require any modifications to the system under measurement. In this paper inaccuracy boundaries of the wireless torque measurement system are determined. Comparison tests of the developed torque measurement system with a commercial system are carried out, and the results are analyzed.

   Line impedance estimation using model based identification technique
By Mihai CIOBOTARU, Vassilios G. AGELIDIS, Remus TEODORESCU [View]
[Download]

Abstract: The estimation of the line impedance can be used by the control of numerous grid-connected systems, such as active filters, islanding detection techniques, non-linear current controllers, detection of the on/off grid operation mode. Therefore, estimating the line impedance can add extra functions into the operation of the grid-connected power converters. This paper describes a quasi-passive method for estimating the line impedance of the distribution electricity network. The method uses the model based identification technique to obtain the resistive and inductive parts of the line impedance. The quasi-passive behaviour of the proposed method comes from the combination of the non-intrusive behaviour of the passive methods with a better accuracy of the active methods. The simulation results reveal the good accuracy of the proposed method.

   On the Role of the Shaft End in the Radio-Frequency Emission of Discharge Bearing Currents in Induction Motors
By Jero AHOLA, Annette MUETZE, Ville NISKANEN [View]
[Download]

Abstract: A novel non-intrusive radio-frequency-based method to detect discharge bearings currents occurring in variable speed drives had been presented previously. Further understanding of the emission characteristics of the discharge and of the electric machine it occurs in will widen its areas of application. To this aim, this paper addresses the modeling of the machine as an antenna, identifying the shaft ends to play a key role. Additionally, a first set of results on the use of location techniques to locate the discharge incident are presented.

   Predictive Zero-Crossing Detection Algorithm by Time Localised Iterative Least-Square Technique
By Biswajit BASU, Malabika BASU [View]
[Download]

Abstract: A time-localised iterative least square (TILS) based predictive zero-crossing detection (ZCD)algorithm for signals with slowly varying amplitudes and phases has been proposed in this paper.Useful for digital implementation, this technique predicts the zero crossing instant in advance sothat further control decision based on ZCD could be performed without delay. The analyticalformulation has been verified through experimental results.

   Stochastic model in Eddy Current Non Destructive Testing
By Charles HOMBERG, Thomas HENNERON, Stéphane CLENET [View]
[Download]

Abstract: In Non Destructive Testing (NDT), the system to scan experiences mechanical and thermalconstraints. The shape of crack in the system can not be predicted. A stochastic model for the shape ofthe crack has been combined with a numerical model based on the solution of the Maxwell equationsin quasistatics using the finite element method. In these conditions, a Monte Carlo Simulation Methodhas been introduced in a magnetoharmonic formulation used to solve a NDT problem.

EPE 2011 - DS3d: Topic 08: Measurements and Sensors
You are here: EPE Documents > 01 - EPE & EPE ECCE Conference Proceedings > EPE 2011 ECCE Europe - Conference > EPE 2011 - Topic 08: Measurements and Sensors > EPE 2011 - DS3d: Topic 08: Measurements and Sensors
	[return to parent folder]

	A digital platform for real-time simulation of power converters with high switching frequency By Kevin DE CUYPER, Michel OSEE, Frederic ROBERT, Pierre MATHYS	[View] [Download]
Abstract: In this paper, we propose a new implementation method for the rapid prototyping of power converters. AField-Programmable Gate Array emulates the temporal equations of the device while analog interfacesallow to use the platform to test control and protection systems in a nearly seamless way.A description of the converter based on an ideal on-off model for the power switches is used in orderto obtain multiple topologies for the same device. Each on these topologies is described by its ownequations and corresponds to a specific set of states of the switches. A topology selector based on thecontrol signals and the internal currents and voltages allows to find the correct equations in less than 10ns and a mid-point linear solver calculates the outputs in a few hundreds of nanoseconds. The solver also takes care of the passive switches (i.e. diodes) to ensure that they don’t go outside their current/voltage limit values.The whole process is described with a boost converter, and results show the great accuracy of our method even for high (> 100kHz) switching frequencies and sampling periods of less than 1μs.

	A fast estimation algorithm for low-frequency oscillations in power systems By Mebtu BEZA, Massimo BONGIORNO	[View] [Download]
Abstract: This paper deals with the application of the Recursive Least Square (RLS) algorithm for estimation oflow-frequency oscillations. A novel resetting method to improve the speed of response of the estimationwill be described. The effectiveness of the proposed method will be proven through simulation as wellas experimental verification.

	A Fully Automated Measurement Set-up for the Determination of the Reverse Recovery Behaviour of Ultra-Fast Diodes By Juergen STAHL, Daniel KUEBRICH, Thomas DUERBAUM, Janina PATZ, Anne-Christine LEICHT	[View] [Download]
Abstract: Often data sheets provide only poor information about the recovery behavior of ultra-fast diodes. Onthe other hand, existing diode models do not predict the real characteristic of all diodes. Nevertheless,due to its importance, this behavior needs to be known and therefore measured. For this purpose, afully automated measurement set-up for determining the reverse recovery characteristic of ultra-fastdiodes in an accurate manner was designed and is described here. All the data obtained is immediatelytransferred into MATLAB and therefore available for further calculation, model building and modelvalidation. Since the whole set-up is automated, a complete field of variations in the reverse voltage,the forward current, the temperature, and the rate of current drop di/dt can be easily applied to thetested diode. Hence, a lot of information can be obtained effortlessly. This uncomplicatedmethodology makes it readily available for circuit designers, allowing them to predict the contributionof the reverse recovery of rectifiers to the total losses more accurately. In addition, a real comparisonof different diodes at many operation points is possible.

	A new Digital Modulated Interface for a Precise and Robust Current Sensor By Abdoulkarim BOUABANA, Constantinos SOURKOUNIS	[View] [Download]
Abstract: Compared to conventional current sensors, the proposed topology is advantageous because of its high resolution throughout its full measurement range, thus reducing offsets to a minimum. This aim is reached through active offset compensation using an analog/digital chopper/de-chopper. A further advantage is the excellent suppression of 1/f noise and a low noise density (35 nV/√Hz) down to the DC-threshold. Common current sensors may be replaced with this technology whose one output is likewise analog. The power supply requirements are also far lower than for standard topologies as only a single supply is needed. The F-PWM modulation method and therefore the current sensor, results for current control units a high performance. The measurement current is directly coded in a “digital current value”, which can be decoded and used for the control unit. An AD-Converter is not more necessary for signal conversion, compared to the typical analog output current sensor based on the Hall-Effect. The coded signals can be transmitted potential- free with an optocoupler.

	A simple calorimetric setup for the accurate measurement of losses in power electronic converters By Georgios DIMITRAKAKIS, Emmanuel TATAKIS, Anastasios NANAKOS	[View] [Download]
Abstract: This paper deals with the analysis and implementation of a low cost calorimetric setup, suitable for loss measurements in power electronic converters, for power ratings up to 30W. Main component of the device is an open Dewar vessel. A proper calibration method is suggested and the several error factors are spotted, so as to achieve an overall accuracy of 2\%. Application examples are demonstrated with loss measurements on magnetic components.

	Accuracy study of frequency converter estimates used in the sensorless diagnostics of induction-motor-driven systems By Tero AHONEN, Jussi TAMMINEN, Jero AHOLA, Markku NIEMELÄ	[View] [Download]
Abstract: Modern frequency converters allow energy efficient operation of induction-motor-driven systems such as pumps and fans. In addition, they can be used as a sensorless monitoring unit for the pump and fan systems, as shown by previous studies. Sensorless diagnostics of induction-motor-driven systems relies on the accurate estimation of the rotational speed, shaft torque and shaft power of the motor, which is why all frequency converters are not suitable for this use. In this paper, accuracy of the rotational speed, shaft torque and shaft power estimates is studied by laboratory tests for two different frequency converters. Using the test results, estimation accuracy of a vector- and a direct-torque-controlled frequency converter are compared with each other. The effect of erroneous motor estimates on the pump flow rate estimation is also demonstrated in the paper.

	Analysis Of The Technology OF Cable Fault Detection And Location On-Line By Chen WEI, Wang LI	[View] [Download]
Abstract: In this paper, two new cable fault detection and location on-line methods--Sequence Time Domain Reflectometry method (STDR) and Spread Spectrum Time Domain Reflectometry method (SSTDR) were introduced and analyzed. An improved frequency correlation algorithm for real-valued sequences was introduced to cable fault detection and location on-line; it greatly reduced the time and the storage consumption of the detection system. The simulation models of these two methods were established; the feasibility of these two methods and the algorithm was validated by simulation and then compared these two methods, Simulation and comparison results show that SSTDR is a universal cable fault detection and location on-line method.

	Evaluation of Machine Learning Techniques for Electro-Mechanical System Diagnosis By Miguel DELGADO, Antoni GARCIA, Julio URRESTY, Jordi-Roger RIBA, Juan Antonio ORTEGA	[View] [Download]
Abstract: The application of intelligent algorithms, in electro-mechanical diagnosis systems, is increasing inorder to reach high reliability and performance ratios in critical and complex scenarios. In this context,different multidimensional intelligent diagnosis systems, based on different machine learningtechniques, are presented and evaluated in an electro-mechanical actuator diagnosis scheme. The useddiagnosis methodology includes the acquisition of different physical magnitudes from the system,such as machine vibrations and stator currents, to enhance the monitoring capabilities. The featurescalculation process is based on statistical time and frequency domains features, as well as timefrequencyfault indicators. A features reduction stage is, additionally, included to compress thedescriptive fault information in a reduced feature set. After, different classification algorithms such asSupport Vector Machines, Neural Network, k-Nearest Neighbors and Classification Trees areimplemented. Classification ratios over inputs corresponding to previously learnt classes, andgeneralization capabilities with inputs corresponding to learnt classes slightly modified are evaluatedin an experimental test bench to analyze the suitability of each algorithm for this kind of application.

	Ground fault detection method for three phase inverters based on the third harmonic By Tommi KÄRKKÄINEN, Kimmo RAUMA, Pertti SILVENTOINEN	[View] [Download]
Abstract: A new ground fault detection method based on third harmonic injection and the measurement of third harmonic current is presented. As the third harmonic is a common mode signal in a three phase system, its existence in current is a sign of a ground fault.

	Implementing clamp on wireless torque measurement system for rotating shaft applications By Ville NISKANEN, Jero AHOLA	[View] [Download]
Abstract: Typically mechanical torque measurement is required for the verification of simulation model of systems consisting of electrical drive, electric motor, load, and control system. It is also possible to detect faults in system by torque measurement. In this paper a wireless measurement system to measure mechanical torque and its oscillations on a shaft of electric motor is studied and developed. The operation of the system is verified by laboratory tests. The wireless torque measurement system is based on a strain gauge collar which can be installed around the shaft of the motor. Installation and the use of the wireless torque measurement system does not require any modifications to the system under measurement. In this paper inaccuracy boundaries of the wireless torque measurement system are determined. Comparison tests of the developed torque measurement system with a commercial system are carried out, and the results are analyzed.

	Line impedance estimation using model based identification technique By Mihai CIOBOTARU, Vassilios G. AGELIDIS, Remus TEODORESCU	[View] [Download]
Abstract: The estimation of the line impedance can be used by the control of numerous grid-connected systems, such as active filters, islanding detection techniques, non-linear current controllers, detection of the on/off grid operation mode. Therefore, estimating the line impedance can add extra functions into the operation of the grid-connected power converters. This paper describes a quasi-passive method for estimating the line impedance of the distribution electricity network. The method uses the model based identification technique to obtain the resistive and inductive parts of the line impedance. The quasi-passive behaviour of the proposed method comes from the combination of the non-intrusive behaviour of the passive methods with a better accuracy of the active methods. The simulation results reveal the good accuracy of the proposed method.

	On the Role of the Shaft End in the Radio-Frequency Emission of Discharge Bearing Currents in Induction Motors By Jero AHOLA, Annette MUETZE, Ville NISKANEN	[View] [Download]
Abstract: A novel non-intrusive radio-frequency-based method to detect discharge bearings currents occurring in variable speed drives had been presented previously. Further understanding of the emission characteristics of the discharge and of the electric machine it occurs in will widen its areas of application. To this aim, this paper addresses the modeling of the machine as an antenna, identifying the shaft ends to play a key role. Additionally, a first set of results on the use of location techniques to locate the discharge incident are presented.

	Predictive Zero-Crossing Detection Algorithm by Time Localised Iterative Least-Square Technique By Biswajit BASU, Malabika BASU	[View] [Download]
Abstract: A time-localised iterative least square (TILS) based predictive zero-crossing detection (ZCD)algorithm for signals with slowly varying amplitudes and phases has been proposed in this paper.Useful for digital implementation, this technique predicts the zero crossing instant in advance sothat further control decision based on ZCD could be performed without delay. The analyticalformulation has been verified through experimental results.

	Stochastic model in Eddy Current Non Destructive Testing By Charles HOMBERG, Thomas HENNERON, Stéphane CLENET	[View] [Download]
Abstract: In Non Destructive Testing (NDT), the system to scan experiences mechanical and thermalconstraints. The shape of crack in the system can not be predicted. A stochastic model for the shape ofthe crack has been combined with a numerical model based on the solution of the Maxwell equationsin quasistatics using the finite element method. In these conditions, a Monte Carlo Simulation Methodhas been introduced in a magnetoharmonic formulation used to solve a NDT problem.