DETECTION OF FAULT IN PWM VOLTAGE SOURCE INVERTER FOR PMSM DRIVE SYSTEM 1 ANUTTAR J. SHENDE, 2 S.D.JOSHI 1,2 Dept. of Electrical Engineering P.E.S. Modern College of Engineering, Pune, Maharashtra, India E-mail: 1 anuttar.shende@gmail.com Abstract In this paper, a easy and low-cost fault identification and detection method for (PWM) pulse-width modulated and voltage-source inverter (VSI) for (PMSM) drive system is proposed. Fault in a power switch of PWM VSI varies the respective terminal voltages and causes the voltage distortions in each phase voltage. The proposed fault diagnosis method employs the (MRAS) model reference adaptive system techniques and needs no additional sensors or electrical devices to identify the fault situation and detect the faulty switch. The proposed technique has the features of fast identification time, easy structure, and being easily introduced to the existing control algorithms as a sub-routine without any major modifications. The experiments and MATLAB simulations are being carried out and the results show the effectiveness of the proposed method. Index Terms Fault Detection, Fault Diagnosis, Fault Identification, Pulse-Width Modulated Voltage-Source Inverter (PWM VSI). I. INTRODUCTION The (PMSM) permanent magnet synchronous motor is increasingly used in electrically powered wheelchairs, electric vehicles, military and medical applications, aerospace and nuclear power plants due to its advantages such as torque to inertia ratio, high power density, simple control and efficiency [3]. In these applications, due to any accident or fault can result in huge damages to the environments and human life, reliability of the machine drives is one of the most important factors to guarantee the safety of personals, continuous and high performance operation under even some accidents or faulty condition. Generally, when an accident or fault occurs, the drive system has to be stopped for non programmed maintenance schedule or emergency. Due to the high cost of unexpected maintenance, the development of a reliable system is the area of interest. The control system with less or zero effects from the faults is called a fault-tolerant controlsystem and it employs the following three tasks [11]: 1) Fault detection; 2) Fault identification; 3) Remedial actions. The fault detection is the method to check whether the system is healthy or not. The fault identification is performed after fault detection process, to identify the fault location, its type, and nature of the fault. Finally, the remedial action, also known as fault isolation, is the process to remove the faulty devices and reconfigure the system for a safe and continuous operation. Out of these three methods, the fault detection and fault identification are considered as an important process for the practical implementation and are often called as a fault diagnosis. [2]. Typically the motor drive systems consist of a power electronic converter, i.e., pulse-width modulated voltage source inverter (PWM VSI), and a motor. 1) Motor (PMSM) 2) Power converter (PWM VSI) 3) Connectors and wire 4) Sensors The connectors, and wires have very low failure rates compared to the remaining of the system. The connectors and wires are static and have low failure rates if selected and installed properly. Some of the machine faults caused by the winding insulation failure due to the excessive voltage or current stress can practically be removed because the line voltage surges are absorbed at the input side of the power converter and the current stresses are limited by the over current protection of the power converter [4]. In recent years, the sensor faults have been increasingly concerned in the literature works. The sensor faults including biased signal, loss of signal, incorrect gain, and unresponsive signal are mainly due to the broken or bad connections, bad communications, or some hardware or software malfunction. Therefore, if the connectors and wires are installed correctly, the failure rates of the sensor faults can be lowered. For these reasons, those faults are not considered.[12] - [13]. The statistical studies are introduced to show the components most susceptible to failures. These studies show that about 38% of failures in variable speed AC drives in industry are mostly found in power equipment and 53% in control circuits [17]. Also an industry based survey regarding the reliability in power electronic converters show that power devices, gate control circuits and capacitors are the most susceptible components [18]. Recently, another study shows that soldering, semiconductor and printed circuit board failures in the device modules totals about 60% of converter system failures [19]. Therefore, it is possible to conclude that a very high percentage of failures are reflected on power switch faults, since a failure in a gate control 111
circuit which results open-circuit fault in a switch. [20]. Power switch failures can be largely distributed as open circuit faults and short-circuit faults. Inverter typical protection methods include protection against the over-current or short-circuit, but they do not guarantee the protection against the open-circuit faults. Then, open-circuit faults can remain hidden or undetected for an extended period of time, tending to potential secondary faults in the converter. [1]. On the other hand, the power converter failures can be a critical factor to the overall system causing shutdowns; therefore, these require a high cost of unexpected maintenance. It is estimated that about 38% of all the failures are found in the power converter [14] and the most of faults are occurred to the power switches [15]. A voltage command is generated in response to a control algorithm and the VSI synthesizes this voltage command using the power switches, i.e., insulated-gate bipolar transistor (IGBT) and MOSFET, with the techniques such as the sinusoidal PWM or space vector. There could be a quite high chance of failure in the switching devices due to the high electrical and thermal stresses [16]. The failure of switching devices can take place in the form of open circuit or short circuit. An improper gate signal causes the short circuit fault so that both power switches in a leg of the VSI are turned ON. This results in the short circuit of the capacitor in the dc link that blows out the other components particularly switching devices. Therefore, the shortcircuit fault is one of the most severe accidents and the most important thing in the drive system, is to minimize the time between appropriate reaction and short-circuit fault initiation. Apparently, the control circuits are build so that a fast fault diagnosis characteristic to prevent the abnormal overcurrent and mostly the hardware based protection schemes are introduced. On the other hand the open-circuit fault, is often considered since it has the characteristic of less danger and slow response to the whole drive system compared with the short-circuit fault. The open-circuit fault maybe due to from the disconnection of a wire from the switching devices results to gate driver failure or thermic cycling. The opencircuit fault leads to the current imbalance in both the healthy and faulty phases results in the pulsating currents and torques, which highly degrades the driving performances. The opencircuit fault does not results system shutdowns and is not generally harmful to the machine drives and, but could lead to the secondary faults at the other components [4]. II BLOCK DIAGRAM AND ITS DESCRIPTION The block diagram of open circuit fault detection in PWM voltage source inverter for SM drive system is shown in fig 3.1 Fig 3.1: Block diagram of open circuit fault detection in PWM voltage source inverter for PMSM drive system It consist of: 1. Pulse generator- It is used to generate the PWM signals which is given to the gate terminal of MOSFET. 2. Inverter- It is used to convert DC into AC. This AC is then given to the motor via filter. 3. Filter- It is used to convert PWM pulses into sinusoidal wave. It is also used to remove the remaining distortion from the system. 4. Scope 1- It is used to see the results of inverter output in terms of voltage in normal condition as well in faulty condition. 5. Motor- Permanent magnet synchronous motor is used in this project. 6. Scope 3- It is used to see the results of in terms of voltage, current, speed and torque.in normal condition as well as in faulty condition. 7. Statcom- It is used to reduce the distortions to some extent of the faulty phase after the fault occurs. 8. Scope 2- It is used to see the results of STATCOM in terms of voltage in normal condition as well as in faulty condition. III. ANALYSIS OF THE OPEN-CIRCUIT FAULT IN THE PWM VSI Fig.2 shows the basic configuration of phase A leg of a three phase PWM System. When the system is under the normal condition as can be seen in Fig. 2(a), the terminal voltage of phase A, is determined by the phase current ias and the switching function Sa of QaU and QaL. If the switching function is 1, which means that when, QaU is turned ON and QaL is turned OFF, the terminal voltage of phase A is equal to half of Vdc, where Vdc is the DC voltage. If the switching function is 0, which means that QaU is turned OFF and QaL is ON, va0 = Vdc divided by 2. The possible terminal voltages according to the 112
switching function and the direction of phase current under the normal condition are Fig. 2. Basic configuration of phase A leg of a three-phase PWM VSI. (a) Normal system. (b) Open-circuit fault in the top switch QaU. Represented in TableI. Unlike the normal condition, an opencircuit fault in the top switch QaU results in changing the corresponding terminal voltage when the phase current ias is positive and the switching function Sa is 1, since the top switch QaU is not working properly. So that, the terminal voltage of phase A is not equal to Vdc/2,but equal to Vdc/2.The equivalent circuit after the open-circuit fault occurrence to the top switch QaU is shown in Fig. 2(b), and the corresponding terminal voltages are represented in Table II. TABLE I TERMINAL VOLTAGES OF PHASE AUNDER THE NORMAL CONDITION From the aforementioned analysis, it is observed that the phase voltages may have the voltage deviations in the steady state after the fault detected from the normal condition. Based on this fact, the proposed fault diagnosis method indirectly observes these voltage deviations using the analytical model of the PWM VSI and the fault diagnosis can be achieved. IV. SIMULATION MODEL The working of the experimental model is as follows: 1. Mainly the circuit consist of an inverter which consist of 6 MOSFETs. We can use any switch instead of MOSFET like BJT, IGBT, GTO, etc. We are using MOSFET because it is cheap as compared to other switches. 2. The PWM generator is used to give the PWM pulses to the gate terminal of MOSFET. MOSFET and internal diode in parallel with a series RC snubber circuit. When a gate signal is applied the MOSFET conducts and acts as a resistance (Ron) in both directions. If the gate signal falls to zero when current is negative, current is transferred to the antiparallel diode. 3. The inverter is basically used to convert DC into AC. The input of inverter is 440V dc and output is fed to the PMSM through LC filter. 113
We can change the magnitude of input dc voltage, by changing the magnitude of dc voltage, only the amplitude of the output gets changed. 4. LC filter is used to remove the remaining distortions from the inverter system. 5. The output which comes from inverter can be checked in scope 1. For this, three voltage measurement blocks are connected to the scope 1. 6. Here, three phase PMSM machine with sinusoidal back EMF or trapezoidal back EMF is used. Sinusoidal machine is modeled in dq rotor reference and trapezoidal machine is modeled in abc reference frame. Here we use sinusoidal back EMF. Stator windings are connected in wye to an internal neutral point. The parameters of the motor can be observed by referring the model. The mechanical input to the PMSM is mechanical torque. We can either use constant torque or step torque. Step torque is used to get the result in step form. Normally we used constant mechanical torque. The scope is attached to the motor which is used to see the result of the changes in voltage, current, torque and speed of the motor. There is a voltage measurement block attached to the scope to measure the voltage, Goto blocks which are used to send the signals to from blocks that have specified tag. If tag visibility in scoped then Goto tag visibility block must be used to define visibility of tag. Block icon displays the selected tag name. Before current Goto block there is sample and hold with one sample period delay. 7. Here extra block of three phase short circuit fault is used. We can use this block to program a short circuit fault between any phase to ground. 8. STATCOM is used only to reduce the distortions to some extent in particular phase after fault occurs. Scope 2 is used to see the results after applying the STATCOM. V. SIMULATION RESULTS Table 4.1 Different results of MATLAB model We can make the open circuit connection in two ways. Either we can remove the phases by opening the switches which are connected to gate terminal of MOSFET or we can open the phase connection with the help of three phase short circuit fault. The results which we get by opening the phases through switches and by opening the phases using short circuit fault are totally different. It is because open circuit fault using switches represent the fault to the inut of the inverter and open circuit fault using three phase short circuit fault to the output of the inverter. The following table shows different results related to this project which is as follows: The following waveform shows results for different cases: 114
1. Under normal condition 3. When phase B open by opening S3, S4 Fig.4.2 Three phase output of the inverter in normal condition. Fig.4.7 voltage waveform when phase B open circuited by removing switches S3, S4. Fig. 4.3 PMSM machine characteristics in terms of voltage, current, speed and torque in normal condition. 2. When phase A open circuited by opening S1, S2 Fig 4.8. Changes in motor current, voltage, speed and torque when phase B is Open Circuit fault by removing switches S3, S4 Fig.4.4 voltage waveform when phase A open circuited by removing switches S1, S2. Fig. 4.9 voltage waveform of statcom after phase B open by removing switches S3, S4. CONCLUSION Fig 4.5. Changes in motor current, voltage, speed and torque when phase A is Open Circuit fault by removing switches S1, S2. Fig. 4.6 voltage waveform of statcom after phase A open by removing switches S1, S2 The fault detection and identification is becoming moreand more important for industrial applications. Therefore, it is increasingly required to improve the fault diagnosis capabilities. In this project, a simple and low-cost open-circuit fault detection and identification method is presented. The proposed method can be well combined with the post fault actions which are the reconfigurations of the whole drive system to operate safely and continuously. In comparison with the previous existing fault diagnosis, the proposed method has simple structure and fast fault detection time. Also, it can be implemented without any extra devices such as voltage sensors and the computing effort is very small. The execution of the algorithm can be easily embedded in the existing systems without major modifications. To show the effectiveness of the proposed method, the simulations and experiments are carried out for the PMSM drive system. The simulation and experimental results 115
verify the validity of the proposed method and show that the proposed method gives the good performance and practical value. REFERENCES [1] Arafa S. Mohamed, Haitham Z. Azazy and Ashraf S. Zein El Din, Open-Circuit Fault Diagnosis Of Three-Phase Induction Motor Drive Systems, Journal of Electrical Engineering. [2] Shin-Myung Jung, Jin-Sik Park, Hag-Wone Kim, Kwan- Yuhl Cho, and Myung-Joong Youn, An MRAS-Based Diagnosis of Open Circuit Fault in PWM Voltage-Source Inverters for PM Synchronous Motor Drive Systems IEEE Transactions On Power Electronics, Vol. 28, No. 5, May 2013 [3] P. C. Krause, O. Wasynczuk, and S. D. Sudhoff, Analysis of Electric Machinery, IEEE Press: IEEE Power Eng. Soc., 1995. [4] D. Kastha and B. K. Bose, Investigation of fault modes of voltage-fed inverter system for induction motor drive, IEEE Trans. Ind. Appl., vol. 30, no. 4, pp. 1028 1038, Jul./Aug. 1994. [5] M.R. Ubale, R. B. Dhumale, S. D. Lokhande, Open switch fault diagnosis in three phase inverter using diagnostic variable method, IEEE Trans. Power Electron., vol no. 2, pp. 636 641, Dec. 2003. [6] Byoung-Kuk Lee. On the Feasibility of Four-Switch Three-Phase BLDC Motor Drives for Low Cost Commercial Applications: Topology and Control, IEEE Trans. On power electronics, vol. 18, no. 1, pp. 164-172, Jan 2003. [7] Jorge O. Estima and Antonio J. Marques Cardoso, A New Approach for Real-Time Multiple Open-Circuit Fault Diagnosis in Voltage-Source Inverters, IEEE Trans. On Industry Application, vol. 47, no. 6, pp. 2487-2494, Nov/Dec 2011. [8] M.D. Singh and K.B. Khanchandani power Electronics Second Edition, McGraw Hill Publication, pp 535 to 593. [9] B.L. Theraja and A. K. -Theraja AC and DC machine, S. Chand Publication vol.2, pp-1489,1551. [10] http://www.powerqualityworld.com/2011/09/statcom static-synchronous-compensator.html [11] R. L. A. Ribeiro, C. B. Jacobina, E. R. C. Silva, anda. M. N. Lima, Fault detection of open-switch damage in voltagefed PWM motor drive systems, IEEE Trans. Power Electron., vol. 18, no. 2, pp. 587 593, Mar. 2003. [12] P. A. Samara, G. N. Fouskitakis, J. S. Sakellariou, and S. D. Fassois, A statistical method for the detection of sensor abrupt faults in aircraft control systems, IEEE Trans. Ind. Control Syst. Technol., vol. 16, no. 4, pp. 789 798, Jul. 2009. [13] H. Berriri, M. W. Naouar, and I. Slama-Belkhodja, Easy and fast sensor fault detection and isolation algorithm for electrical drives, IEEE Trans. Power Electron., vol. 27, no. 2, pp. 490 499, Feb. 2012 [14] O. V. Thorsen and M. Dalva, A survey of the reliability with an analysis of faults on the variable frequency drives in industry, in Proc. Conf. Rec. 6th Eur. Power Electron. Appl. Conf., Sevilla, Spain, 1995, pp. 1033 1038. [15] H. Schwab, A. Klonne, S. Reck, and I. Ramesohl, Reliability evaluation of a permanent magnet synchronous motor drive for an automotive application, in Proc. Conf. Rec. 10th Eur. Power Electron. Appl. Conf., Toulouse, France, 2003, ISBN: 90-75815-07-7. [16] F. W. Fuchs, Some diagnosis methods for voltage source inverters in variable speed drives with induction machines A survey, in Proc. IEEE Ind. Electron. Conf., Nov. 2003, pp. 1378 1385. [17] Sethom, H.A. Ghedamsi, M.A.: Intermittent misfiring default detection and localisation on a PWM inverter using wavelet decomposition. In: Journal of Electrical Systems, Vol. 4, No. 2, 2008, p. 222 234. [18] Yang, S. at al.: An industry-based survey of reliability in power electronic converters. In: IEEE Transactions on Industry Applications, Vol. 47, No. 3, May/June 2011, p. 1441 1451. [19] Yang, S. at al.: Condition monitoring for device reliability in power electronic converters A review. In: IEEE Transactions on Power Electronics, Vol. 25, No. 11, November 2010, p. 2734 2752. [20] Freire N.A. Estima J.O., Marques Cardoso A.J.: Multiple open-circuit fault diagnosis in voltage-fed PWM motor drives using the current park's vector phase and the currents polarity. In: 8th IEEE International Symposium on Diagnostics for Electrical Machines, Power Electronics and Drives, September 2011, Bologna, Italy, p. 397-404. 116