Conferences
Rotor Fault Detection and Identification on a Quadcopter based on Statistical Data-driven Methods: Experimental Assessment via Flight Tests
Machine-Learning Framework for Online Probabilistic Rotor Fault Diagnosis via Strain Data under Varying Flight States
Proving the Correctness of Multicopter Rotor Fault Detection and Identification Software
Machine-learning Framework for Online Probabilistic Rotor Fault Detection, Identification, and Quantification on a Multicopter using Boom Strain Signals
Unified Statistical Framework for Rotor Fault Diagnosis on a Hexacopter via Functionally Pooled Stochastic Models
Rotor Fault Detection and Identification on a Hexacopter under Varying Flight States Based on Global Stochastic Models
Rotor Fault Detection and Identification for a Hexacopter Based on Control and State Signals via Statistical Learning Methods
Time Series Statistical Learning Methods for Multicopter Fault Detection and Identification
Fault Detection and Identification for Multirotor Aircraft by Data-Driven and Statistical Learning Methods
Rotor Fault Detection and Identification on a Hexacopter Based on Statistical Time Series Methods
Flexible Modelling of Treadle Pumps
Journal Articles
Dutta A., McKay,M.. Kopsaftopoulos F., Gandhi F.
American Institute of Aeronautics and Astronautics Journal, August 2021
Fault Detection and Identification for Multirotor Aircraft by Data-Driven and Statistical Learning Methods
In this paper we presented the introduction, investigation and critical assessment of three data-driven methods for rotor failure detection and identification in multicopter. These methods are based on aircraft attitude signals obtained from forward flight under turbulence and uncertainty. The knowledge based method exploits the system rigid-body dynamics insight under the different rotor failures to construct a decision tree that detects and identifies the rotor failure simultaneously by how the roll, pitch and yaw signals violate the statistical confidence limits immediately after failure. For the statistical time-series method, the development of stochastic time-series model and residual based statistical hypothesis testing are discussed. Here, fault detection in the transient response is followed by identification after the signals reach a stationary state post controller compensation with the healthy and the different faulty models respectively, in a two-step manner. The third method employs the healthy time-series model to extract a useful feature, the residual cross-correlation, as an input to a neural network trained to achieve simultaneous rotor failure detection and identification. The time-series assisted neural network is capable of taking decision throughout the entire flight and results in an accuracy of 99.3\% with very less computation time (< 0.03 s) making it the best alternative for online, real-time monitoring.
Dutta A., McKay,M.. Kopsaftopoulos F., Gandhi F.
Aerospace Science and Technology Journal, May 2021
Statistical Residual-Based Time Series Methods for Multicopter Fault Detection and Identification
This study presents the introduction, investigation and critical assessment of data-based statistical residual-based time series methods for rotor fault detection and identification in multicopters. A concise overview of statistical residual-based fault detection and identification (FDI) methods is provided based on scalar (univariate) and vector (multivariate) stochastic time series models. The FDI methods employed in this study are based on identified response-only parametric scalar (univariate) and vector (multivariate) autoregressive (AR) representations of multicopter attitude states (time series), as the external excitation is non-observable, and their corresponding model residuals obtained under the considered healthy and fault multicopter states. The comparative assessment of the effectiveness of several residual-based statistical FDI methods are presented in the face of external disturbances, namely three different levels of turbulence, and for different rotor fault scenarios. FDI methods based on Vector AutoRegressive (VAR) models exhibit improved performance compared to their scalar counterparts, as indicated by lower false alarms and missed fault rates. In the case of rotor fault identification (classification), the methods that are based on scalar AR models exhibit reduced rotor fault classification accuracy, while the VAR-based methods outperform their scalar counterparts and can achieve a fault identification accuracy of up to 99.6\%. The results of this study demonstrate the potential and effectiveness of residual-based time series methods in terms of fast, effective, and robust-to-uncertainties rotor fault detection and identification.
Dutta A., Khatait J.P., Saha S.K.,
Rural Technology  Development and Delivery,Â
Design Science and Innovation Springer,
Singapore, April 2019
Development of a Solar-Powered Treadle Pump
With the aim to reduce drudgery in operation and improve reliability, the design of a conventional feet-operated treadle pump have been modified into a solar-powered treadle pump which will be able to lift groundwater to suffice the irrigation needs during periods of scarce rainfall in an economic and environment friendly manner. Dynamic analysis has been performed for the proposed design after taking into account the hydraulic forces and frictional forces encountered while in operation. The results of dynamic simulation has helped us to design the various parts of it for proper functioning. The fabricated model has been tested to evaluate its discharge and efficiency.
​
It has been observed that a healthy adult can produce about 75 W for longer periods, which enables to lift 2−3 m3/h of water by a feet-operated treadle pump from a depth of 3−5 m at 50% efficiency. The solar-operated treadle pump can be powered by a maximum of 2 solar panels with capacity of 125 W for suction depth of 3 m and discharge of 3 m3/h. However, the installation of the solar panels and motor will add cost of about INR 20,000, making the total cost to be about INR 30,000. Considering the fact that the manual operation can be done not for more than 40 min owing to the fatigue of leg muscles, the solar operated one can be a good alternative. This is not only environment friendly but could be used for 4−6 h daily or more depending on the hours of peak sunshine. Therefore, the cost of installation can be shared by 4−5 marginal farmers to irrigate their lands with a nominal maintenance cost. The commercially available systems require diesel as fuel to operate, which not only incurs significant operating costs but also causes water and soil pollution due to leakage of oil.