Search

Nonlinear fluid dynamical systems, such as thermoacoustic systems, aeroelastic systems are archetypical complex systems involving state transitions upon a change in bifurcation parameter. These state transitions in any certain direction are always undesirable and can radically alter the operational paradigms associated with these systems. Hence, predicting the impending dynamical state is paramount for avoiding such undesirable transitions. The hitherto research so far focused largely on metric-based and model-based indicators to foretell an impending transition and is often fraught with difficulties when deployed in practicable scenarios. In this study, we assuage this end of concern by proposing a model-agnostic data-driven method for automated classification of the dynamical states of nonlinear fluid dynamical systems. By using recurrence plots we transform the time series pertaining to the dynamical states into images and subsequently employ a convolution neural network (CNN) to classify the generated images. This study also proceeds to present cross-domain classifications via a trained deep learning (DL) model and successfully classify the dynamical states of one fluid dynamical system (say, thermoacoustic) with the dynamical states of another fluid dynamical system (say, aeroelastic). The underlying methodology for the above is based on open set (OS) domain adaptation - inherent to transfer learning schemes. Towards enhancing the confidence levels of our proposed methodology, we carry out four cross-domain numerical experiments, wherein we consistently get about 94 - 98% accuracy.

Parkinson’s disease (PD) is a neurodegenerative disorder characterized by several motor symptoms such as resting tremor, muscular rigidity, slowness of movement and different speech impairments. PD is a kind of singular, multi-system disorder that gradually worsen over the time. In this research, we classify the neurological state of the patients with Parkinson’s disease (PwPD) according to the third section of the Movement Disorders Society - Unified Parkinson’s Disease Rating Scale (MDS-UPDRS-III) using multimodal bio-signal data. As PD advances from low to advanced state, PwPD finds it difficult in their speech production and irregularities in their gait patterns. Monitoring the chaotic nature of time series data corresponding to speech and gait biomarkers can provide insights into the progression of the condition across different stages. This work for the first time analyze PD in a complex system perspective while representing the biomarkers as complex networks. The time-series corresponding to speech and gait signals are represented separately, as a complex network using the visibility graph algorithm. The characterization of the different stages of PD is explored for each modalities using different network features. Performance evaluation shows that the results obtained using the multimodal configuration of speech and gait left foot signals outperform the state-of the-art method. Moreover, performance comparison with the unimodal counterparts proves the need for multimodal assessment of PD severity. The configuration ‘speech and gait left foot’ outperforms (in terms of accuracy) that of the unimodal by 32% in speech, 3% in gait left foot, 19% in gait right foot, and 3% in gait both feet.

Abstract Many systems found in nature are susceptible to tipping, where they can shift from one stable dynamical state to another. This shift in dynamics can be unfavorable in systems found in various fields ranging from ecology to finance. Hence, it is important to identify the factors that can lead to tipping in a physical system. Tipping can mainly be brought about by a change in parameter or due to the influence of external fluctuations. Further, the rate at which the parameter is varied also determines the final state that the system attains. Here, we show preconditioned rate induced tipping in experiments and in a theoretical model of a thermoacoustic system. We provide a specific initial condition (preconditioning) and vary the parameter at a rate higher than a critical rate to observe tipping. We find that the critical rate is a function of the initial condition. Our study is highly relevant because the parameters that dictate the asymptotic behavior of many physical systems are temporally dynamic.

Renewable energy sources in modern power systems pose a serious challenge to the power system stability in the presence of stochastic fluctuations. Many efforts have been made to assess power system stability from the viewpoint of the bifurcation theory. However, these studies have not covered the dynamic evolution of renewable energy integrated, non-autonomous power systems. Here, we numerically explore the transition phenomena exhibited by a non-autonomous stochastic bi-stable power system oscillator model. We use additive white Gaussian noise to model the stochasticity in power systems. We observe that the delay in the transition observed for the variation of mechanical power as a function of time shows significant variations in the presence of noise. We identify that if the angular velocity approaches the noise floor before crossing the unstable manifold, the rate at which the parameter evolves has no control over the transition characteristics. In such cases, the response of the system is purely controlled by the noise, and the system undergoes noise-induced transitions to limit-cycle oscillations. Furthermore, we employ an emergency control strategy to maintain the stable non-oscillatory state once the system has crossed the quasi-static bifurcation point. We demonstrate an effective control strategy that opens a possibility of maintaining the stability of electric utility that operates near the physical limits. [ABSTRACT FROM AUTHOR]

Electrocardiogram (ECG) signal is one of the most reliable methods to analyse the cardiovascular system. In the literature, there are different deep learning architectures proposed to detect various types of tachycardia diseases, such as atrial fibrillation, ventricular fibrillation, and sinus tachycardia. Even though all types of tachycardia diseases have fast beat rhythm as the common characteristic feature, existing deep learning architectures are trained with the corresponding disease-specific features. Most of the proposed works lack the interpretation and understanding of the results obtained. Hence, the objective of this letter is to explore the features learned by the deep learning models. For the detection of the different types of tachycardia diseases, the authors used a transfer learning approach. In this method, the model is trained with one of the tachycardia diseases called atrial fibrillation and tested with other tachycardia diseases, such as ventricular fibrillation and sinus tachycardia. The analysis was done using different deep learning models, such as RNN, LSTM, GRU, CNN, and RSCNN. RNN achieved an accuracy of 96.47% for atrial fibrillation data set, 90.88% accuracy for CU-ventricular tachycardia data set, and also achieved an accuracy of 94.71, and 94.18% for MIT-BIH malignant ventricular ectopy database for ECG lead I and lead II, respectively. The RNN model could only achieve an accuracy of 23.73% for the sinus tachycardia data set. A similar trend is shown by other models. From the analysis, it was evident that even though tachycardia diseases have fast beat rhythm as their common feature, the model was not able to detect different types of tachycardia diseases. The deep learning model could only detect atrial fibrillation and ventricular fibrillation and failed in the case of sinus tachycardia. From the analysis, they were able to interpret that, along with the fast beat rhythm, the model has learned the absence of P-wave which is a common feature for ventricular fibrillation and atrial fibrillation but sinus tachycardia disease has an upright positive P-wave. The time-based analysis is conducted to find the time complexity of the models. The analysis conveyed that RNN and RSCNN models could achieve better performance with lesser time complexity.