VIEWS: 41 PAGES: 5 CATEGORY: Emerging Technologies POSTED ON: 10/12/2011 Public Domain
(IJCSIS) International Journal of Computer Science and Information Security, Vol. 9, No. 9, September 2011 Identifying Harmonics by Empirical Mode Decomposition for Effective Control of Active Filters During Electric Vehicle Charging B. V. Dhananjay and T. Senthil B.V. Dhananjay, Dr.T.Senthilkumar, Research Scholar, Professor, Automobile Engineering, Vinayaka Mission’s University, Bharathidhasan University, Salem,India Trichirapalli, India Abstract-This paper provides Hilbert Huang affecting charging are temperature, state of charge, Transform(HHT) method (an empirical mode plate area, impurities, gassing. decomposition(EMD)) for identifying the presence of Electric vehicles (EV) will become an harmonics during electric vehicle battery charging attractive alternative to internal combustion engine when harmonics are generated into the electric line, due to switching actions of the power electronics. vehicles in the event that their range can be Activation of the active filters based on the difference extended. One way to achieve this in the short term between load current and fundamental current is to provide a fast charger infrastructure. Such a measured from the line is done. By using active power structure would provide greater mobility for the EV filter (APF) injection of the required current to user, since during short stops (<1 hour) the EV minimize the harmonics is done. As part of batteries could be charged from typically 20 to 80 simulation, the accuracy of the HHT is above 95%. % of nominal charge. This would significantly By correctly recognizing the harmonics using HHT extend the EV range.Fast charger infrastructure and injecting the compensating current into the line, cost is high. Chargers adversely affect the grid the charging time of the battery can be reduced. The reduction in the charging time also depends on the power quality due to presence of power electronic battery condition. loads like diode rectifiers and thyristor bridge converters in the distribution network that result in Keywords-Hilbert Huang Transform; active power voltage distortion and current harmonics, (Akagi filter. 1996). High increase of problems in the electric power distribution networks due to the presence of I INTRODUCTION harmonics. Loads that use switching control with semiconductor devices are the main cause. One of the most important tools for correcting the lack of The battery is the primary source of electrical electric power quality are the active power filters energy. It stores chemicals. Two different types of (APF), (Udom et al. 2008). The objective of this lead in an acid mixture react to produce an work has been proving that back propagation electrical pressure. This electrochemical reaction neural networks, previously trained with a certain changes chemical energy into electrical energy. A number of distorted waveforms, are an alternative battery can be of primary cell, secondary cell, wet to the rest of the techniques used and proposed at charged, dry charged and low maintenance type. A the present time for controlling the APF's, as the fully charged battery contains a negative plate of ones based on the use of the Fast Fourier sponge lead(Pb), a positive plate of lead Transform (FFT). A large number of these control dioxide(Pbo2) and an electrolyte of sulphuric acid techniques are based on ANN’s, (Pecharanin et al. (H2So4) and water (H2o). During charging, sulphate 1994). leaves the plates and combines with hydrogen(H2) to become sulphuric acid (H2So4). Free oxygen combines with lead on the positive plate to form II MATERIALS AND METHODS lead dioxide. Gassing occurs as the battery nears full charge and hydrogen bubbles out at the negative plates, oxygen at the positive. Factors A Materials 109 http://sites.google.com/site/ijcsis/ ISSN 1947-5500 (IJCSIS) International Journal of Computer Science and Information Security, Vol. 9, No. 9, September 2011 Figure 1shows a three-phase diagram of M= (a + b) (1) an HHT controlled shunt APF. A load current 2 signal iLis acquiredand used by the ANN to obtain the distortion current waveform as reference signal Where a = Maximum_envelope and b = for the control of the APF.The power converter Minimum_envelope. injects the necessary compensation current iLin the 4. Obtain a new signal using the following power circuit, achieving thus a sinusoidal source equation: current. h 11 (t) = X(t) − M 11 (t) (2) Where h11(t) is called first IMF. Subsequent IMF’s had to be found if there are some overshoots and undershoots in the IMF. Hence, the envelope mean differs from the true local mean and h11(t) becomes asymmetric. In order to find the additional IMF’s, h11(t) is taken as the new signal. After nth iteration, we have: h1n (t) = h1(n −1) (t) − M1n (t) (3) Where M1n(t) is the mean envelop after the nth iteration and h1(n-1)(t) is the difference between the signal and the mean envelope at the (k-1)th iteration. 5. Calculate C2F as follows: C2F1 = IMFn (4) Figure 1 APF control using HHT Where IMFn = final IMF obtained B Methods C2F2 = IMFn + IMF(n −1) (5) Empirical Mode Decomposition (Huang) and Similarly, Hilbert Transform A signal can be analyzed in details for its frequency, amplitude and phase contents by using C2Fn = IMFn + IMF(n −1) + ....... + IMF1 (6) EMD followed by HT (Jayasree et al. 2010 and Stuti et al. 2009), The EMD produces the mono Where C2Fn is the original signal. components called IMFs from the original signal. In a given frame of signal, there can be many IMFs. Each IMF will contain a wave form of 6. Calculate F2C as follows: different amplitude. Hilbert Transform is applied F2C1 = IMF1 (7) on an IMF to obtain, IF and IA. It is mandatory that a signal be symmetric regarding the local zero mean, and should contain same number of extreme F2C 2 = IMF1 + IMF2 (8) and zero crossings. The steps involved in EMD of a signal X(t) with F2C n = IMF1 + IMF2 + ....... + IMFn (9) harmonics into a set of IMFs are as follows. 1. Identify all local maxima of X(t). Connect the Where F2Cn is the original signal. points using a cubic spline. The interpolated 7. Hilbert transform is applied for each curve obtained. The upper line is called the IMF and analytical signal is obtained. upper envelope (Maximum_envelope). A complex signal is obtained from each 2. Identify all local minima of X(t) connect the point using a cubic spline.. The lower line is IMF: called the lower envelope Analytic(I MF) = real(IMF) + imag(IMF) (10) (Minimum_envelope) obtained by cubic spline. 3. Compute the average by: 110 http://sites.google.com/site/ijcsis/ ISSN 1947-5500 (IJCSIS) International Journal of Computer Science and Information Security, Vol. 9, No. 9, September 2011 8. Instantaneous frequencies are obtained a simulation are presented where a step load change from analytical signal using occurs at time 60 ms. One additional resistance is connected in parallel with the load, increasing the IF = 0.5 × (angle( − X(t + 1) × conj(X(t − 1))) + π) (11) total load current. 2× π Figure 4 shows the EMD process. In the sample harmonics signal considered, only one 9. Instantaneous amplitudes are obtained instantaneous mode function is present. A flat from the analytical signal using the residue signal is also presented. This plot is only following for 1000 samples. This will be repeated for the remaining length of the signal. Figure 5 shows the IA = real(IMF) 2 + imag(IMF) 2 (12) extraction of different signals present from the fine level to coarse level. Similarly, Figure 5 shows the extraction of different signals present from the coarse level to fine level. Figure 7 presents the III EXPERIMENTAL SIMULATION instantaneous frequency present in every sample of the signal. Figure 8 presents the instantaneous amplitude present in every sample of the signal. Figure 9 and Figure 10 presents statistical values of instantaneous frequencies and instantaneous amplitudes. Based on the statistical values, the amount of harmonics will be estimated and appropriately, required compensating current will be injected into the line. V CONCLUSION A Hilbert Huang Transform method has been used at the control of a shunt active power filter. Based on the amount of harmonics recognition, the APF is activated. By correctly injecting the compensating current into the line, the charging time of the battery can be reduced. The circuit has to be verified with the implementation of HHT in real time for improved charging of the EV battery. The reduction in the charging time also depends on the battery condition. Figure 2 Power circuit with HHT controlling active power filter v=0.1pu,cy=50 The model of Figure2 has been created v=0.1pu,cy=50 using Matlab 10. Different sets of parameters have 400 v=0.1pu,cy=50 v=0.1pu,cy=50 been employed at the power circuit and APF. In 200 v=0.1pu,cy=50 most cases the reference current obtained by the V o l ta g e (V ) HHT controller was accurate enough to enable the 0 APF to compensate harmonic distortion. If an elevated content of high order harmonics were -200 present in the load current, the HHT controller helps in obtaining reference signal. -400 6 System parameter used: 4 1 Power circuit:Phase voltage = 220 VRMS , 2 0.5 Frequency = 50 Hz , Source resistance = 0.1 Ω, 0 0 Signal patterns Load resistance = 20 Ω, Load inductance = 30 mH Time(Sec) APF:Vdc = 500 V, R = 10 Ω, L = 30 mH, Switching frequency = 40 KHz. Figure 3 Sample plot of signals for harmonics IV RESULTS AND DISCUSSION More than 500 different harmonics waveforms (Figure 3) have been used in HHT analysis with different load changes.The results of 111 http://sites.google.com/site/ijcsis/ ISSN 1947-5500 (IJCSIS) International Journal of Computer Science and Information Security, Vol. 9, No. 9, September 2011 Empirical Mode Decomposition 50.1 s ig n a l 50.05 F re q u e n c y s p e c tru m In s ta n ta n e u o s 50 49.95 200 im f1 0 49.9 -200 49.85 0.015 0.02 0.025 0.03 0.035 0.04 Time (sec) 200 Figure 7 Instantaneous frequency re s . 0 231.5 -200 0.01 0.02 0.03 0.04 0.05 0.06 0.07 Figure 4 Empirical mode decomposition Am plitude 231 f2c s ig n a l 230.5 0.015 0.02 0.025 0.03 0.035 0.04 Time (sec) 200 Figure 8 Instantaneous amplitude f2 c 1 0 150 10 S td o f F re q u e n c y F re q u e n c y -200 100 M ean of 5 50 200 s ig n a l 0 0 0 0 1 2 3 0 1 2 3 Imf Imf -200 150 0.01 0.02 0.03 0.04 0.05 0.06 0.07 Maximum Frequency 1500 Minimum Frequency F re q u e n c y F re q u e n c y 100 N o rm o f Figure 5 Fine to coarse signals 1000 50 500 c2f 0 0 0 1 2 3 0 1 2 3 200 Imf Imf s ig n a l 0 Figure 9 Statistical values of the instantaneous frequencies -200 0.01 0.02 0.03 0.04 0.05 0.06 0.07 10 300 1 A m p litu d e A m p litu d e M ean of 200 S td o f c 2 f1 0 5 100 -1 0 0.01 0.02 0.03 0.04 0.05 0.06 0.07 0.08 0 0 0 1 2 3 0 1 2 3 500 Imf Imf s ig n a l 0 300 4000 Maximum Amplitude -500 A m p litu d e A m p litu d e 0 0.01 0.02 0.03 0.04 0.05 0.06 0.07 0.08 Minimum Amplitude N o rm o f 200 Figure 6 Coarse to fine signals 2000 100 0 0 0 1 2 3 0 1 2 3 Imf Imf Figure 10 Statistical values of the instantaneous amplitudes 112 http://sites.google.com/site/ijcsis/ ISSN 1947-5500 (IJCSIS) International Journal of Computer Science and Information Security, Vol. 9, No. 9, September 2011 REFERENCES [1] HAkagi., ”New Trends in Active Filters for Power Conditioning”, IEEE Trans. on Industry Applications, vol. 32, No 6, Dec. 1996, pp 1312-1322. [2] TJayasree., DDevaraj., RSukanesh.,”Power quality disturbance classification using Hilbert transform and RBF networks”, Neurocomputing , Vol 73 Issue 7-9, March, 2010, pp. 1451-1456. [3] NPecharanin., MSone., HMitsui., “An application of neural network for harmonic detection in active filter”,IEEE World Congress on Computational Intelligence.,IEEE International Conference on Neural Networks, Vol.6, pp. 3756-3760, 1994. [4] StutiShukla, S. Mishra, and BhimSingh,”Empirical-Mode Decomposition With Hilbert Transform for Power-Quality Assessment”, IEEE transactions on power delivery, Vol. 24, No. 4, October 2009. [5] Udom. Khruathep, SuittichaiPremrudeepreechacharn, YuttanaKumsuwan,“Implementation of shunt active power filter using source voltage and source current detection”, IEEE, pp.2364-2351, 2008. 113 http://sites.google.com/site/ijcsis/ ISSN 1947-5500