SHM data anomaly classification using machine learning strategies: A comparative study

Shieh-Kung Huang,Jau-Yu Chou,Yuguang Fu,Chia-Ming Chang 국제구조공학회 2022 Smart Structures and Systems, An International Jou Vol.29 No.1

Various monitoring systems have been implemented in civil infrastructure to ensure structural safety and integrity. In long-term monitoring, these systems generate a large amount of data, where anomalies are not unusual and can pose unique challenges for structural health monitoring applications, such as system identification and damage detection. Therefore, developing efficient techniques is quite essential to recognize the anomalies in monitoring data. In this study, several machine learning techniques are explored and implemented to detect and classify various types of data anomalies. A field dataset, which consists of one month long acceleration data obtained from a long-span cable-stayed bridge in China, is employed to examine the machine learning techniques for automated data anomaly detection. These techniques include the statistic-based pattern recognition network, spectrogram-based convolutional neural network, image-based time history convolutional neural network, image-based time-frequency hybrid convolution neural network (GoogLeNet), and proposed ensemble neural network model. The ensemble model deliberately combines different machine learning models to enhance anomaly classification performance. The results show that all these techniques can successfully detect and classify six types of data anomalies (i.e., missing, minor, outlier, square, trend, drift). Moreover, both image-based time history convolutional neural network and GoogLeNet are further investigated for the capability of autonomous online anomaly classification and found to effectively classify anomalies with decent performance. As seen in comparison with accuracy, the proposed ensemble neural network model outperforms the other three machine learning techniques. This study also evaluates the proposed ensemble neural network model to a blind test dataset. As found in the results, this ensemble model is effective for data anomaly detection and applicable for the signal characteristics changing over time.

Jerk-Constrained Time-Optimal Control of a Positioning Servo

Raymond Shieh,Yu-Sheng Lu 제어로봇시스템학회 2010 제어로봇시스템학회 국제학술대회 논문집 Vol.2010 No.10

This paper presents a jerk-constrained time-optimal control (JCTOC) method for a positioning servo. While high jerk, the derivative of acceleration with respect to time, can cause problems such as vibrations and high wear to mechanical parts, it is then important to limit the maximum jerk in many control applications. Many trajectory planning methods have been proposed, limiting the maximum jerk during the optimized planning of the desired path. Distinct from the previous path-planning methods that yield jerk-constrained ideal trajectories only, the approach proposed in this paper directly controls the jerk of the plant while guaranteeing minimum-time performance under the specified jerk constraint. Simulation studies of the JCTOC compared with the conventional PD controller have been conducted, demonstrating that the proposed control requires less control effort and also offers the time-optimal response with admissible jerk.

Therapeutic effect of Ferula assa-foetida oleo-gum resin in rats with letrozole-induced polycystic ovary syndrome

Amir Shieh,Seyyed Majid Bagheri,Maryam Yadegari,Davoud Javidmehr,Zeinab Farhadi The Korean Society for Reproductive Medicine 2022 Clinical and Experimental Reproductive Medicine Vol.49 No.4

Objective: Asafoetida is a gum derived from Ferula assa-foetida, which is used in traditional Iranian medicine to treat some reproductive system disorders. The effects of asafoetida on ovarian tissue, expression of certain genes associated with polycystic ovary syndrome (PCOS), and levels of liver, kidney, and blood cell factors after treatment in a rat model were investigated. Methods: Thirty rats were divided into five groups: normal, polycystic, and treatment with three doses of asafoetida (12.5, 25, and 50 mg/kg for 3 weeks after PCOS induction). PCOS was induced by letrozole at a dose of 1 mg/kg administered orally for 3 weeks. Blood samples were taken, and the ovaries were removed and prepared for histomorphometric examination. Liver and kidney parameters were measured. The mRNA expression levels of luteinizing hormone receptor, CYP11A1, adenosine monophosphate-activated protein kinase, adiponectin, and adiponectin receptors 1 and 2 were also measured by real-time polymerase chain reaction. Results: The levels of liver, kidney, and blood parameters did not significantly differ between the treatment groups and the control group. At doses of 25 and 50 mg/kg, ovarian histopathology, especially the thicknesses of the theca and granulosa layers, was significantly improved relative to the PCOS group. The expression of target genes also improved in the 25 and 50 mg/kg treatment groups. Conclusion: Asafoetida can be used to treat PCOS as a complementary approach to conventional therapies. Asafoetida appears to act by regulating and activating metabolic and ovarian cycle enzymes.

트랜스 아트 재단

SCHAUM/SHIEH,피터 몰릭(사진) 현대건축사 2019 월간 CONCEPT Vol.- No.238

Design of A Semi-spherical Microphone Array Based Sound Localization System

Ming-Yuan Shieh,Ming-Hung Tsai,Chin-Chien Chen,Jeng-Han Li 제어로봇시스템학회 2009 제어로봇시스템학회 국제학술대회 논문집 Vol.2009 No.8

This paper proposes a semi-spherical microphone array based sound localization system for a service robot. The hardware of the proposed system basically contains 12 capacitor microphones disposed in two layers on the semi-sphere of 19 cm diameter. It aims to estimate the degree relationship between the main speaker and the robot and provides the robot useful information for more effective human-robot interactions. The proposed system can determine the location of the voice according to energy information between the main speaker and robot not only in normal environment but also in blatant and/or reverberative spaces. The experimental results show that the proposed system has obtained satisfactory recognition efficiency, moreover, raised the robotic friendliness and adaptability.

Potential assessment of an innovative hybrid ventilator for building ventilation

Tzyy-hwang Shieh,Pei-Chi Chang,Che-ming Chiang,Chi-ming Lai 대한기계학회 2010 JOURNAL OF MECHANICAL SCIENCE AND TECHNOLOGY Vol.24 No.11

In this study, an innovative rooftop turbine ventilator powered by a hybrid of wind and photovoltaic energy, “Hybrid Ventilator” for short, was developed. The performance differences between Hybrid Ventilators and conventional ventilators were assessed through a series of experiments. Then, CFD (Computational Fluid Dynamics) simulations were applied to survey the building ventilation efficiency of this Hybrid Ventilator. The results show that, considering the ventilation quantity (rate), a Hybrid Ventilator provides approximately 4times the exhaust capacity of a conventional ventilator. All of the investigated space configurations that were examined via CFD simulations exhibited similar indoor airflow patterns and air velocity distributions (ventilation quality).

The Contribution of Religion to a More Comprehensive Environmental Education

Jennchyun Mark Shieh 인하대학교 다문화융합연구소 2020 다문화와 교육 Vol.5 No.2

The causes of contemporary environmental and ecological crisis are multiple and complex. This crisis cannot be resolved by science or technology alone. We need not only knowledge, economic, and technology, but also worldviews, ethos, and practices to reconnect humans, other species, and nature. A sustainable ecology relies on ideals and ways of life which include value, belief, worldview, ethical commitment, and pattern of life. The Enlightenment tradition prioritizes human sovereignty over non-human world and invokes individualism, materialism, and consumerism, which objectify, materialize, and instrumentalize the non-human world; detach human being from transcendental constraint on human desire; and seduce human being into short-term self-interested pursuit. They are the main determinants of modern way of living and the major causes of contemporary environmental and ecological crisis. Religious worldviews define the status of human in the world and the human-nature relation. Religious ethos can be biocentric and ecocentric ethics rather than self-centered ethics or materialistic worldview. Furthermore religion grounds its ethics on transcendental or sacred base and incorporates its ethical norms in the practice of everyday life, communal living, and even every spheres of human activities. Nevertheless, there are diverse religions and various models of environmental education. There are difference, even disputes, between science and religion in some aspects. In a situation of diversity there needs to be an open and inclusive public sphere and procedure of engaged mutual understanding for reaching a consensus or temporary mutual agreement concerning environmental education.

[Bronze Prize] 물의 여신

Ta-Lih Shieh,Yen-Shiuan Cheng,Cha-Ray Chu,Jimmy Day,Ying-Kai Huang 현대건축사 2001 월간 CONCEPT Vol.- No.33

DEVELOPMENT OF AN INTEGRATED CAD SYSTEM FOR VLSI DESIGN

Jiin Ming Shieh 대한전자공학회 1985 대한전자공학회 심포지엄 논문집 Vol.- No.-

DESIGN AND IMPLEMENTATION OF A PROCESS BASED MULTIUSER MULTITASKING OPERATING SYSTEM

Jong Jiann Shieh,Yu Wei Wang 대한전자공학회 1985 대한전자공학회 심포지엄 논문집 Vol.- No.-