Aninda Bhattacharya

System to monitor blade health in axial flow compressors

Detection of incipient cracks and crack propagation in rotor blades during operation of gas compressors is an important problem from the perspective of safety, reliability, availability, and maintenance of gas turbines. In this paper we present the architecture of a blade health monitoring system that has been deployed to monitor rotor blades of the gas compressors. The system operates continuously, 24×7, with minimal human intervention to monitor a fleet of gas compressors all around the world. The compressors are retrofitted with non-contact magnetic clearance sensors to generate Blade Passing Signals (BPS) during the rotation of the blades. These signals are acquired by a Data Acquisition (DAQ) box. The DAQ extracts the Time of Arrival (TOA) of the blades and stores them. The TOA data is transferred to a central system for feature extraction at regular intervals. The extracted features - natural frequencies of predetermined modes of vibration and static deflection are used to determine the health of the rotating blades and generate alarms for the Monitoring and Diagnostics (M&D) team. The M&D team monitors the health of blades using a web portal that provides visualization of data, features, and alarms.

Estimation of static deflection under operational conditions for blade health monitoring

Reliability and availability of gas turbines are critical to industry as they are used in a variety of applications. It is important to determine the health of these complex systems in near-real time and detect failures as early as possible. Cracks in compressor blades, used in gas turbines, if not detected early, could lead to catastrophic failures that would result in increased costs and extended outages. These cracks tend to induce changes in the static deflection of the blade. By tracking the changes in static deflection, it is possible to determine whether a blade is cracked or not. However, the static deflection of the blade, under field conditions, is also affected by operational parameters of the gas turbine. Therefore, estimating static deflection under field conditions and isolating the effects of crack from that of operational parameters is a significant task. This paper describes an algorithm for the estimation of static deflection under field conditions. It discusses some of the challenges involved and possible solutions for mitigating them. This algorithm can be used in conjunction with a crack propagation model to determine crack size in real-time.

Flow Control of Real-time Multimedia Applications in Best-effort Networks

One of the fastest growing segments of Internet applications are real-time multimedia applications, like Voice over Internet Protocol ( VoIP). Real-time multimedia applications use the User Datagram Protocol (UDP) as the transport protocol because of the inherent conservative nature of the congestion avoidance schemes of Transmission Control Protocol (TCP). The effects of uncontrolled flows on the Internet have not yet been felt because UDP traffic frequently constitutes only ∼20% of the total Internet traffic. It is pertinent that real-time multimedia applications become better citizens of the Internet, while at the same time deliver acceptable Quality of Service (QoS). Traditionally, packet losses and the increase in the end-to-end delay experienced by some of the packets characterizes congestion in the network. These two signals have been used to develop most known flow control schemes. The current research considers the flow accumulation in the network as the signal for use in flow control. The most significant contribution of the current research is to propose novel end-to-end flow control schemes for unicast real-time multimedia flows transmitting over best-effort networks. These control schemes are based on predictive control of the accumulation signal. The end-to-end control schemes available in the literature are based on reactive control that do not take into account the feedback delay existing between the sender and the receiver nor the forward delay in the flow dynamics. The performance of the proposed control schemes has been evaluated using the ns-2 simulation environment. The research concludes that active control of hard real-time flows delivers the same or somewhat better QoS as High Bit Rate (HBR, no control), but with a lower average bit rate. Consequently, it helps reduce bandwidth use of controlled real-time flows by anywhere between 31:43% to 43:96%. Proposed reactive control schemes deliver good QoS. However, they do not scale up as well as the predictive control schemes. Proposed predictive control schemes are effective in delivering good quality QoS while using up less bandwidth than even the reactive control schemes. They scale up well as more real-time multimedia flows start employing them.