Fault detection and diagnosis in synchronous motors using hidden Markov model-based semi-nonparametric approach

Early detection and diagnosis of faults in industrial machines would reduce the maintenance cost and also increase the overall equipment effectiveness by increasing the availability of the machinery systems. In this paper, a semi-nonparametric approach based on hidden Markov model is introduced for fault detection and diagnosis in synchronous motors. In this approach, after training the hidden Markov model classifiers (parametric stage), two matrices named probabilistic transition frequency profile and average probabilistic emission are computed based on the hidden Markov models for each signature (nonparametric stage) using probabilistic inference. These matrices are later used in forming a similarity scoring function, which is the basis of the classification in this approach. Moreover, a preprocessing method, named squeezing and stretching is proposed which rectifies the difficulty of dealing with various operating speeds in the classification process. Finally, the experimental results are provided and compared. Further investigations are carried out, providing sensitivity analysis on the length of signatures, the number of hidden state values, as well as statistical performance evaluation and comparison with conventional hidden Markov model-based fault diagnosis approach. Results indicate that implementation of the proposed preprocessing, which unifies the signatures from various operating speeds, increases the classification accuracy by nearly 21% and moreover utilization of the proposed semi-nonparametric approach improves the accuracy further by nearly 6%.     

Data aggregation in wireless sensor network using SVM-based failure detection and loss recovery

Abstract

In wireless sensor network, data aggregation can cause increased transmission overhead, failures, data loss and security-related issues. Earlier works did not concentrate on both fault management and loss recovery issues. In order to overcome these drawbacks, in this paper, a reliable data aggregation scheme is proposed that uses support vector machine (SVM) for performing failure detection and loss recovery. Initially, a group head, selected based on node connectivity, splits the nodes into clusters based on their location information. In each cluster, the cluster member with maximum node connectivity is chosen as the cluster head. When the aggregator receives data from the source, it identifies node failures in the received data by classifying the faulty data using SVM. Furthermore, a reserve node-based fault recovery mechanism is developed to prevent data loss. Through simulations, we show that the proposed technique minimises the transmission overhead and increases reliability.     

Optimization of bead geometry of submerged arc weld using fuzzy based desirability function approach

The present study highlights application of Taguchi’s robust design coupled with fuzzy based desirability function approach for optimizing multiple bead geometry parameters of submerged arc weldment. Fuzzy inference system has been adapted to avoid uncertainly, imprecision and vagueness in experimentation as well as in data analysis by traditional Taguchi based optimization approach. Detailed methodology and unique features of the proposed method has been highlighted through a case study. The said approach can efficiently be used in off-line quality control of any production process as well as automation of the process.     

Effects of time-inconsistent preferences on information technology infrastructure investments with growth options

Increasing information technology (IT) infrastructure spending and the capability of such projects to provide a platform for a firm to realize value from IT marks their importance. Effective management of IT infrastructure investments includes identification of embedded growth options in the infrastructure, and exercising them in a timely manner. Extant research has recognized that while managers could use real options thinking in IT investment management, managerial bias could affect the timing of option exercise and their realized value. We analyze the effect of time-inconsistent preferences of present-biased managers on the exercise time of real growth options and the realized value using a discrete time option valuation model. The results show that present-biased managers are more likely to exercise options early when the net payoffs are low, the option payoffs have high volatility, and the risk free discount rate is small. In addition, present biased managers are more likely to exercise a growth option early in its life when the project is performing well. We provide implications for practice and IT governance.     

Fuzzy modeling of system behavior for risk and reliability analysis

The main objective of the article is to permit the reliability analyst's/engineers/managers/practitioners to analyze the failure behavior of a system in a more consistent and logical manner. To this effect, the authors propose a methodological and structured framework, which makes use of both qualitative and quantitative techniques for risk and reliability analysis of the system. The framework has been applied to model and analyze a complex industrial system from a paper mill. In the quantitative framework, after developing the Petrinet model of the system, the fuzzy synthesis of failure and repair data (using fuzzy arithmetic operations) has been done. Various system parameters of managerial importance such as repair time, failure rate, mean time between failures, availability, and expected number of failures are computed to quantify the behavior in terms of fuzzy, crisp and defuzzified values. Further, to improve upon the reliability and maintainability characteristics of the system, in depth qualitative analysis of systems is carried out using failure mode and effect analysis (FMEA) by listing out all possible failure modes, their causes and effect on system performance. To address the limitations of traditional FMEA method based on risky priority number score, a risk ranking approach based on fuzzy and Grey relational analysis is proposed to prioritize failure causes.     

A simulation-based approach to study stochastic inventory-planning games

Non-cooperative decision-making problems in a decentralized supply chain can be characterized and studied using a stochastic game model. In an earlier paper, the authors developed a methodology that uses machine learning for finding (near) optimal policies for non-zero sum stochastic games, and applied their methodology on an N-retailer and W-warehouse inventory-planning problem. The focus of this paper is on making the methodology more amenable to practical applications by making it completely simulation-based. It is also demonstrated, through numerical example problems, how this methodology can be used to find (near) equilibrium policies, and evaluate short-term rewards of stochastic games. Short-term rewards of stochastic games could be, in many instances, more critical than equilibrium rewards. To our knowledge, no methodology exists in the open literature that can capture the short-term behaviour of non-zero sum stochastic games as examined in this paper.     

Performance analysis for queueing systems with close down periods and server under maintenance

A single server queue subject to maintenance of the server and the close down period is considered. We obtain explicit expressions for the transient probabilities of the system size, the server under maintenance state and the close down period. The time-dependent performance measures of the system and the probability density function of the first-passage-time to reach the maintenance state are discussed. The corresponding steady state analysis and key performance measures of the system are also presented. Finally, the effect of various parameters on system performance measures is demonstrated by a numerical example.     

Neotectonic study of Ganga and Yamuna tear faults,NW Himalaya,using remote sensing and GIS

The Ganga and Yamuna rivers emerge from the Himalayas along two major faults known as the Ganga and Yamuna Tear Faults respectively. The two major strike-slip faults transverse to the Siwalik range are clearly seen in satellite imagery of the Dehradun area. Earthquake records, landslide and recent changes in geomorphological features indicate that the area between the Main Boundary Thrust and the Main Frontal Thrust is tectonically active. An effort has been made to study the tectonic evolution and neotectonism of the Ganga and Yamuna tear faults. Spectral and spatial enhancement techniques have been employed to the digital data of IRS-1B LISS-I to delineate the lineaments and major faults of the area. Based on Mohr's theory, failure criteria and statistical analysis of remotely sensed lineament data, horizontal compressive stress values (S_Hmax) have been estimated at various sites of the study area. These data are found to be consistent with the published S_Hmax orientation determined from earthquake focal mechanism solutions. Active faults and lineaments have been extracted from the remotely sensed lineament data. Past earthquake data and depth to basement contour data have been used in an integrated approach with available Geographic Information System (GIS) techniques to reconstruct a present-day regional geodynamic model. Attempts have been made to investigate the genesis of Ganga and Yamuna Tear Faults and possible causes of recent tectonic activities of the area with the help of the proposed geodynamic model.     

Evaluation of satellite-altimetry-derived river stage variation for the braided Brahmaputra River

In this study, river stage variation derived from satellite altimetry was used to assess the water level, monthly discharge, and annual water yield at six virtual gauging stations at the braided reaches of the Brahmaputra River. The braided reaches of the river dynamically change their planform, thalweg line, and aggradation or degradation period. Stage records derived from the Envisat satellite of the European Space Agency and Topex/Poseidon of NASA/CNES were used for the period 2002–2010. Spatial interpolation and datum correction were applied on altimetry-derived river stage records before analysis. A correlation and error analysis between the in situ and satellite-altimetry-derived stages was carried out for these stations for both monsoon and non-monsoon seasons. Yearly optical satellite images were used for qualitative assessment of temporal variations in aggradation/degradation phases at the gauging stations. Using the pseudo-rating curve, discharges at two virtual gauging stations were estimated. The results show that the altimetry-estimated discharges are of good agreement with observed discharge for the monsoon months (June–September) as compared with the non-monsoon months (October–May). In order to assess the annual water yield variability, yearly variation in annual water yield from the altimetry data was also estimated and compared to that observed. The estimated annual water yields were 90% accurate. Similarly, the long-term averaged monthly discharge series estimated from satellite altimetry closely follows the temporal trend of that of the observed series.