In this work, Dissolved gas Analysis (DGA) has been implemented using soft computing models namely fuzzy logic and Adaptive Neuro fuzzy inference system (ANFIS). DGA has developed as an effective tool for the identification of transformer incipient faults. A number of standards and procedures have evolved over the years making DGA more reliable and user friendly. A comparative study of the two models has been developed based on their ability to circumvent the limitations of the IEC 599 standard, Rogers ratio method and Doernenburg's method. The models have been tested using a reported fault database for their diagnostic capability. Results presented in this paper clearly indicate the superiority of the ANFIS model over the fuzzy system. The ANFIS model presents a reliable system with relative high degree accuracy. ANFIS model being very simple to develop can obviate the limitations of conventional methods of transformer fault diagnosis using DGA.

A course on the emerging topic “Smart Sensors and Internet of Things (IoT)” has been designed, developed, and delivered from 21st March 2016 to 31st March 2016 in the Department of Electrical Engineering, Jamia Millia Islamia (Central University), New Delhi-110025, India. This course was organized under a program initiated by the Government of India, entitled “Global Initiative of Academic Networks” for higher education. The basic objective is to bring the internationally and nationally acclaimed scientists on a single platform to motivate young minds of India so that India's scientific and technological capacity reach global excellence. The course was designed to develop the technical skills for the students, the researchers, the engineers, and the faculties, who desire to address some of the issues faced by the people around the world in the fields of smart sensors and IoTs with applications in smart homes, smart cities, smart grids, smart environment, smart transport, and so on. The assessment of the course by the participants shows extreme usefulness of it.

The capacitive sensors are widely used for liquid level measurement in different applications, but the contact type capacitive sensors suffer from frequent maintenance and drift of the sensor output. The present paper presents a simple noncontact IDT fringing field capacitance sensor for liquid level measurement. The sensor can be made using flexible /non-flexible substrate, and it can be used to measure the liquid level of the vessel for both planar and non-planar surfaces. Theory of level measurement using the sensor has been discussed. Four different IDT sensor (1-n-1) configurations such as 1-1-1, 1-3-1, 1-5-1 and 1-11-1 have been designed, analysed, simulated (finite element method), and fabricated to measure water and insulating transformer oil levels from 0 to 40 mm. The analytical expressions are derived for 1-n-1 configuration to estimate the value of the capacitance and the maximum field penetration depth of the sensors. The capacitance value of all the four sensors varies almost linearly with the variation of liquid levels. Experimental results show that the sensitivity of the 1-5-1 IDT sensor is 40 fF/mm. Performances of the sensors were compared with other capacitive sensors reported in the literature. The sensor is capable to detect the water level with a resolution of 0.1 mm.

Efforts have been made by the researchers to enhance the sensitivity of a humidity sensor by selecting and fabricating different nanostructures thin films and optimizing their pore morphology. However, other characteristics, such as nonlinearity, hysteresis, reproducibility, response and recovery times, and selectivity, are equally important for real time measurement, monitoring, and controlling of humidity. It is reported that a low signal frequency is essential for enhanced sensitivity, but the effect of the signal frequency on other characteristics is rarely investigated. We have investigated the role of the signal frequency on the response characteristics of a nanostructure ceramic thin-film interdigitated electrode humidity sensor for the first time. Our studies show that although the sensitivity is high at low frequency, the performance of the sensor is highly nonlinear and frequency dependent. At high frequency above 100 kHz, the sensitivity is reduced, but other characteristics of the sensors are improved significantly. Improved characteristics are due to the reduced effect of double layer capacitance that plays a significant role for extreme sensitivity at low frequency. The effect of double layer capacitance is rarely considered for evaluating the response characteristics in the literatures. Results show that at 200-kHz frequency, nonlinearity and hysteresis are 9.39% and 16.77%, while at 100 Hz, these values are 5588% and 70.8%, respectively. Improve response characteristics will reduce the complexity of signal conditioning, particularly nonlinearity and hysteresis, the compensation of which is essential for almost all types of humidity sensors.

A microcontroller-based digital hygrometer system using a low-cost moisture sensor in the range of 3.7-100 ppm has been developed. The sensor is of capacitive type and consists of nanoporous thin film of alumina (γ - Al ₂ O ₃ ) dip coated in between two parallel gold electrodes onto an alumina substrate. The behavior of the sensor has been modeled with an object to develop a signal processing circuit to convert capacitance change into frequency. The detection electronics circuit is based on a relaxation oscillator whose output is suitable for interfacing with a digital device. The sensor has been characterized with the circuit and then output frequency is calibrated in terms of ppm. The accuracy of the digital hygrometer when compared with the commercial dew point meter is found to be ±1 PPM.