Yuwono, M., Qin, Y., Zhou, J., Guo, Y., Celler, B.G. & Su, S.W. 2016, 'Automatic bearing fault diagnosis using particle swarm clustering and Hidden Markov Model', Engineering Applications of Artificial Intelligence, vol. 47, pp. 88-100.View/Download from: UTS OPUS or Publisher's site
© 2015 Elsevier Ltd. Ball bearings are integral elements in most rotating manufacturing machineries. While detecting defective bearing is relatively straightforward, discovering the source of defect requires advanced signal processing techniques. This paper proposes an automatic bearing defect diagnosis method based on Swarm Rapid Centroid Estimation (SRCE) and Hidden Markov Model (HMM). Using the defect frequency signatures extracted with Wavelet Kurtogram and Cepstral Liftering, SRCE+HMM achieved on average the sensitivity, specificity, and error rate of 98.02%, 96.03%, and 2.65%, respectively, on the bearing fault vibration data provided by Case School of Engineering of the Case Western Reserve University (CSE) which warrants further investigation.
Zhou, J., Guo, A., T. Nguyen, H. & Su, S. 2015, 'Intelligent Management of Multiple Access Schemes in Wireless Body Area Network', Journal of Networks, vol. 10, no. 2.View/Download from: UTS OPUS or Publisher's site
Zhou, J., Qin, Y., Kou, L., Yuwono, M. & Su, S. 2015, 'Fault detection of rolling bearing based on FFT and', Journal of Advanced Mechanical Design, Systems and Manufacturing, vol. 9, no. 5.View/Download from: Publisher's site
© 2015 The Japan Society of Mechanical Engineers. The rolling bearing carries a load by placing rolling elements between two bearing rings. It is a key device in the railway vehicles for monitoring work states to ensure high reliability and better performance of rotating machine. The states of rolling bearings can be detected by the measurement of vibration signals with effective process, features extraction and analysis. The propose of this paper is to establish an efficient and robust signal processing technique and classification mechanism to detect the fault of rolling bearing. Firstly Fast Fourier Transform is used to extract features and then these parameters are input into various classification schemes for accurate fault detection. Ensemble Rapid Centroid Estimation is proposed and then compared with Artificial Neural Network, and Principal Components Analysis. The simulation analyses the approaches of fault detection and the accuracy of identification.Then the linear performance of the data is proved by least square regularized regression.Finally various schemes are compared and analyzed to obtained the most efficient method for fault detection.
Zhou, J., Guo, A., Celler, B.G. & Su, S.W. 2014, 'Fault detection and identification spanning multiple processes by integrating PCA with neural network', Applied Soft Computing, vol. 14, no. A, pp. 4-11.View/Download from: UTS OPUS or Publisher's site
This paper proposes an effective fault detection and identification method for systems which perform in multiple processes. One such type of system investigated in this paper is COSMED K4b2. K4b2 is a standard portable electrical device designed to test pulmonary functions in various applications, such as athlete training, sports medicine and health monitoring. However, its actual sensor outputs and received data may be disturbed by Electromagnetic Interference (EMI), body artifacts, and device malfunctions/faults, which might cause misinterpretations of activities or statuses to people being monitored. Although some research is reported to detect faults in specific steady state, normal approach may yield false alarms in multi-processes applications. In this paper, a novel and comprehensive method, which merges statistical analysis and intelligent computational model, is proposed to detect and identify faults of K4b2 during exercise monitoring. Firstly the principal component analysis (PCA) is utilized to acquire main features of measured data and then K-means is combined to cluster various processes for abnormalities detection. When faults are detected, a back propagation (BP) neural network is constructed to identify and isolate faults. The effectiveness and feasibility of the proposed model method is finally verified with experimental data.
Wireless body area network (WBAN) collects significant signals of human body or environment information for health monitoring or professional services. But normal medium access protocols can hardly make a balance and ensure enough reliability of a network because there are specific features and service quality in WBAN applications. Contention access or fixed allocation of bandwidth cannot meet all nodes? requirements and may cause collisions and delay. Especially in emergency medical situations, some data must be transmitted immediately for accurate diagnosis and decision. The dropping of critical messages could possibly create life threatening results. In order to improve the reliability and efficiency of data transmission in WBAN, this paper proposes a fuzzy control medium access (FCMA) mechanism based on input parameters for performance gains. It controls the contention window in contention access period (CAP) and slots allocation in contention free period (CFP) according to nodes? status. Through simulation analysis, the improved performance of throughput, latency, and packets breakdown is demonstrated by efficient usage of bandwidth and avoidance of collision.
Zhou, J., Su, S.W., Guo, A. & Chen, W. 2012, 'Abnormalities Detection of IMU based on PCA in motion monitoring', Applied Mechanics and Materials, vol. 224, no. 1, pp. 533-538.View/Download from: UTS OPUS or Publisher's site
Inertial measurement units (IMU) are used as an affordable and effective remote measurement method for health monitoring in body sensor networks (BSNs) based on tracking peoples daily motions and activities. These inertial sensors are mostly micro-electro-mechanical systems with a combination of multi-axis combinations of precision gyroscopes, accelerometers, and magnetometers to sense multiple degrees of freedom (DoF).Unfortunately in the process of motion monitoring actual sensor outputs may contain some abnormalities, which might result in the misinterpretations of activities. In this paper, we use Principal component analysis (PCA) combined with Hotellings T2 and SPE statistic to detect abnormal data in the process of motion monitoring with IMU to ensure the reliability and accuracy in application. The simulated results prove this method is effective and feasible.
Zhou, J., Guo, A., Xu, J., Nguyen, H. & Su, S. 2014, 'A game theory control scheme in medium access for wireless body area network', IET Seminar Digest, International Conference on Wireless Communications, Networking and Mobile Computing, IET, China, pp. 404-409.View/Download from: UTS OPUS or Publisher's site
Wireless Body Area Network (WBAN) has been considered for applications in medical, healthcare and sports fields. Although there are several protocols for wireless personal area networks, specific features and reliability requirements in WBAN bring new challenges in protocol design. An appropriate control scheme in the MAC layer can make a significant improvement in network performance. Based on traffic priority and prior knowledge this paper proposes a game theoretical framework to smartly control access in contention period and contention free period as defined in IEEE 802.15.6 standard. The coordinator controls access probability of contention period based on users' priority in CSMA/CA and allocates suitable slots with strategies for best payoff based on link states in guaranteed time slots (GTS). The simulation results show the improved performance especially in heavily loaded channel condition when the optimal control mode is applied.
Zhou, J., Guo, A., xu, J., Su, S.W. & Celler, B.G. 2013, 'A reliable medium access mechanism based on priorities for wireless body sensor networks', Engineering in Medicine and Biology Society (EMBC), 2013 35th Annual International Conference of the IEEE, International Conference of the IEEE Engineering in Medicine and Biology Society, IEEE, Osaka, Japan, pp. 1855-1858.View/Download from: UTS OPUS or Publisher's site
Wireless body sensor networks (WBSN) provide health related information for monitoring or professional analysis by collecting various signals of human body or environment information with sensors. But different data acquired in many applications have different transmission requirements. The dropping of life-critical messages could possibly create life threatening results if the network is not reliable. To improve the reliability this paper proposes a novel reliable medium access mechanism (RMAM) which guarantees transmission of data with different priorities in less delay and energy consumption. The mechanism is designed and evaluated by Castalia. The improved performances of latency, packets breakdown and energy consumption are analyzed and depicted with comparison.
Zhou, J., Su, S.W. & Guo, A. 2012, 'Fault Detection and Identification of COSMED K4b2 based on PCA and Neural Network', WASET:International conference on Information, communication and Signal Processing, International conference on Information, communication and Signal Processing, WASET, Penang, Malaysia, pp. 729-734.View/Download from: UTS OPUS
COSMED K4b2 is a portable electrical device designed to test pulmonary functions. It is ideal for many applications that need the measurement of the cardio-respiratory response either in the field or in the lab is capable with the capability to delivery real time data to a sink node or a PC base station with storing data in the memory at the same time. But the actual sensor outputs and data received may contain some errors, such as impulsive noise which can be related to sensors, low batteries, environment or disturbance in data acquisition process. These abnormal outputs might cause misinterpretations of exercise or living activities to persons being monitored. In our paper we propose an effective and feasible method to detect and identify errors in applications by principal component analysis (PCA) and a back propagation (BP) neural network.