Sivaji Chakravorti

Sivaji Chakravorti

Auto-generated profile, based on publicly available information

Sivaji Chakravorti

Jadavpur University

Research Focus & Metrics

Academic Career Length

14Years (2013-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
1 January 2020 in IEEE Transactions on Industrial Informatics
Verified
Journal Article
Closed access
BGBiswarup Ganguly
Biswarup Ganguly
SCSayanti Chaudhury
Sayanti Chaudhury
SBSubrata Biswas
Subrata Biswas
DDDebangshu Dey
Debangshu Dey
SMSugata Munshi
Sugata Munshi
SCSivaji Chakravorti
Sivaji Chakravorti
This paper presents a new convolutional neural network (CNN) topology using wavelet kernels to detect and discriminate single or multiple Partial Discharge (PD) locations in high voltage power apparatus with increased accuracy. The method is tested on an electrical equipment model with acoustic PD sensors. A cubical tank has been emulated in the laboratory representing the equipment under test and partial discharge sources have been placed at different positions along with the required data acquisition hardware. The present scheme eliminates the requirement of separate algorithms for feature extraction and classification of the acquired PD signals. Wavelet kernels of the CNN play a crucial role in feature learning, and the proposed CNN architecture as a whole, can classify the features in a supervised manner. The performance of the proposed scheme is compared with other existing methods using the same dataset. It is found that an overall accuracy of 97.64% is achieved by the proposed method, outperforming other existing methods by a significant margin of at least 5% in terms of accuracy. The developed module is a generic one and can be adapted for different high voltage electrical apparatus with similar topological structures; hence it can be used in various ways in power industry.

23 January 2018 in IEEE Transactions on Industrial Electronics
Verified
Journal Article
Closed access
SDSantanu Das
Santanu Das
Arpan Kumar PradhanArpan Kumar Pradhan
Arpan Kumar Pradhan
AKAyush Kedia
Ayush Kedia
SDSovan Dalai
Sovan Dalai
BCBiswendu Chatterjee
Biswendu Chatterjee
SCSivaji Chakravorti
Sivaji Chakravorti
This paper presents a novel technique to diagnose frequently encountered power quality (PQ) disturbance cases both single and mixed event types. Adopting existing approach to emulate signals containing different types of PQ disturbance events, a specially designed hardware setup has been developed in the laboratory. Captured signals from the developed event generator have been analyzed to extract useful information for diagnosis of various PQ disturbances. In the proposed method, a signal tracker capable of continuously tracking normal level of the signal has been employed to extract actual disturbance within it. A suitably designed detrended fluctuation analysis tool has been effectively used to diagnose the PQ disturbances. Implementation of the proposed technique shows the effectiveness in differentiating PQ events distinctly without much involving conventional analytical tools that result in minimum computational burden as compared to the existing methods. To establish the effectiveness and to get acceptability, the proposed method has been rigorously tested on significant number of captured signals containing different types of PQ disturbances and a satisfactory outcome has been observed.

14 October 2019 in IEEE Transactions on Instrumentation and Measurement
Verified
Journal Article
Closed access
DMDeepak Mishra
Deepak Mishra
ABArijit Baral
Arijit Baral
NHNasirul Haque
Nasirul Haque
SCSivaji Chakravorti
Sivaji Chakravorti
Utilities prefer non-invasive methods for assessing the condition of power transformer insulation. Analysis of Polarization-Depolarization Current (PDC) is one such popular method. One such analysis involves estimation of trapped charge released from the interfacial region of oil-paper insulation. Literature shows that such charges can be reliably used for diagnosis of transformer insulation. However, such analysis requires a complete profile of PDC. PDC measurement (an offline technique) takes a large amount of time (several hours) to complete. The magnitude of PDC data for a larger value of time is also sensitive to changes in environmental conditions and field noise as its magnitude is low. Hence, a reliable estimation of de-trapped charge may require numerous PDC measurements. This situation is not convenient for utilities as it prolongs shut down time. In the present paper, a method has been proposed which is capable of estimating de-trapping charge using PDC data measured for a short span of time. The proposed method is tested on data collected from several real life in-service transformers.

12 December 2014 in IEEE Sensors Journal
Verified
Journal Article
Closed access
SDSovan Dalai
Sovan Dalai
DDDebangshu Dey
Debangshu Dey
BCBiswendu Chatterjee
Biswendu Chatterjee
SCSivaji Chakravorti
Sivaji Chakravorti
KBKesab Bhattacharya
Kesab Bhattacharya
This paper presents a methodology of cross-wavelet transform aided Fischer linear discriminant analysis (FLDA)-based feature selection and classification for sensing simultaneous occurrence of multiple power quality disturbances. A linear support vector machine is used for classification of the extracted features as it suits well with FLDA. This scheme is implemented in a general purpose microcontroller as a standalone module and the performance of the standalone module for sensing simultaneous occurrence of multiple power quality disturbances is judged by both online and offline testing. Results show that the performance is comparable with the results reported in the literatures. Moreover, the scheme is immune to real life uncorrelated noises due to incorporation of cross spectrum analysis in the feature extraction phase. The present method is generic in nature and can be implemented for any other microcontroller-based applications addressing topologically similar problems.

14 January 2019 in IEEE Transactions on Power Delivery
Verified
Journal Article
Closed access
DMDeepak Mishra
Deepak Mishra
SDSaurabh Dutta
Saurabh Dutta
ABArijit Baral
Arijit Baral
NHNasirul Haque
Nasirul Haque
SCSivaji Chakravorti
Sivaji Chakravorti
Activation energy is popularly used for estimation of remaining life of transformer insulation. It is defined as the average rate of all reactions that happen with cellulose. Existing literature shows that activation energy of oil-paper insulation can be obtained from Polarization Depolarization Current (PDC) and Return Voltage Measurement (RVM) data that are measured at a specific temperature. It is practically difficult to ensure the same measurement temperature for both PDC and RVM data. On the other hand, PDC data and its analysis get influenced by de-trapping current. This de-trapping current is generated by ionic charge carriers that get freed from trap sites during PDC measurement process. Formation of these trap sites are related to physical, chemical reactions that happens at oil-paper interface. The present paper proposes a methodology which uses de-trapped charge, dislodged from deep and shallow traps, to assess insulation condition and for prediction of activation energy. Thus, eliminating the need of RVM data. The proposed method is tested using data collected from various real-life in-service transformers.

23 December 2020 in IEEE Transactions on Power Delivery
Verified
Journal Article
Closed access
SDSourav Dhara
Sourav Dhara
CKCHIRANJIB KOLEY
CHIRANJIB KOLEY
SCSivaji Chakravorti
Sivaji Chakravorti
A Partial Discharge (PD) monitoring system for Air Insulated Substations (AIS) has been developed for monitoring of partial discharge activities of various insulators and insulating media that are distributed over an electrical substation. The proposed system utilizes four UHF (ultra-high frequency) sensors, which are to be placed around the area of interest so that any impulsive UHF signal originated from PD activity can be captured and subsequently processed for detection and localization of multiple PD sources. To improve the localization accuracy and to improve the detection specificity, i.e., to eliminate false impulsive UHF signals originated from other sources, two novel approaches, namely, Nearest-Neighbour and Probabilistic Grid-Search are proposed, which utilize the physical layout of the AIS to identify most probable PD source locations. The proposed algorithm can accurately find the PD source locations even in the presence of high time-delay-of-arrival (TDOA) estimation error by making necessary corrections in the PD source location estimation. The proposed algorithm has been field trialed in a medium-size distribution substation. A detailed comparative study with the conventionally used PD source localization approach in different extreme situations has been performed with the help of simulation, laboratory as well as field-based testing. The obtained results are worthwhile for supporting the widespread implementation of the proposed method.