1 Introduction

Structural Health Monitoring(SHM),while ascertaining the current state of health of any structural system,also helps in predicting its remaining life.It is one of the efficient tools of periodic monitoring to prevent unprecedented failure and to plan emergency remedial measures.As offshore structures are of high strategic importance,preventive measures to safeguard these platforms from any damages are inevitable.This is a condition-based maintenance approach and not a conventional periodic maintenance approach(Chandrasekaran 2011).SHM using wireless sensor networks is useful to quantify the extent of damages as soon as they are identified,also advantageous when compared to that of a wired sensor networks.Reliability of SHM is improved with the appropriate design of sensor network,including the relevant sensor technologies and data processing algorithms(Chandrasekaran 2016a).Sensor technologies and damage detection algorithms combined with non-destructive evaluation showed better performance of health monitoring with lesser installation time and cost(Celebi 2002;Lynch et al.2004a,b;Chandarsekaran 2015a,b).Wireless networks use sensing elements,data transmission,computational modeling,and embodiment of processing abilities within the monitored structure,thus necessitating their use in naval vessels and offshore structures to improve operational safety(Swartz et al.2009;Chandrasekaran et al.2016;Chandrasekaran 2016b,2015a,b).Offshore accidents that result in unintended but substantial loss can also be controlled by initiating risk mitigation measures,if coupled with SHM tools(Chandrasekaran 2016b,2013);in addition to estimating severity of structures,residual life is also predictable.On the contrary,extensive length of cables,complexity of electromechanical systems,their interference with structural geometry,and untraceable failures of cables make wired SHM system inefficient.

Fig.1 SHM architecture

Measurements acquired using spatially distributed wireless sensors are effective in extracting damagesensitive features(Farrar and Worden 2007;Straser et al.2001).In case of extreme events such as earthquakes,blast,or explosions,SHM should be capable of rapid condition screening and should be capable of providing reliable information in real time.In real-time monitoring,SHM systems require ability to handle large volume of data and compress,normalize,and fuse the data to account for variability that occurs during continuous monitoring schemes.This is resolved by an intelligent integration of choice of sensors,their compatibility with the data acquisition system and processor and software capability to diagnose the acquired data without any procedurallapse.Use of analytical tools such as experimentally validated finite element models can be useful and efficient.

农村经济体制与经营管理司司长张红宇表示，对于清产核资工作已经制定了明确的时间表、路线图。2019年6月底前，重点是指导乡镇、村、组各级农村集体经济组织，全面清查核实资产，填写登记报表，逐级审核上报。2019年年底前基本完成清产核资，重点是组织开展监督检查和成果验收，农业农村部、中央农办将形成报告上报党中央、国务院。

Table 1 Structural details of the platform

Description Prototype Model 1:150 Mass of the structure/ton 400,000 0.13 Diameter of the Deck/m 100 0.6666 Diameter of the BLS/m 22.5 0.15 Length of the BLS/m 200 1.3333 Draft/m 163.57 1.1177 Ballast/ton 333,950 0.109 Length of the tether/m 470.84 1.3333

As in case of lab scale,damage identification,location and type of damage(large displacement or rotation),is prefixed;it is simple and easy to execute.However in real time,main difficulties arise in identifying the location of damage and type of damage caused.In the real-time monitoring,as the volume of data acquired will be practically very large,there exist a possibility of data loss due to stacking of data in packets continuously.Robust data reduction techniques are important in real-time monitoring,which should be capable to retain sensitivity of structural changes in the presence of environmental and operational variability.Further,data acquisition method in real-time monitoring is very vital.This is because as data can be measured under varying conditions,ability to normalize the data becomes vital to identify type of damage.However,in lab scale,such challenges are not significant as sensors and processing methods are pre-oriented to cater to the known type and volume of data.

惯性导航传感器具有不需要任何外界信息就能得到位姿参量的特点，在采煤机、机器人等领域实现了精确定位，但是惯导传感器航向角存在随时间产生累积误差的缺点[13-15]。因此，笔者运用惯导与地磁融合的组合惯导，该组合惯导采用三轴地磁传感器数据对惯导进行校正，克服了惯导传感器航向角随时间产生累积误差的难题。笔者构建了多传感器信息的煤矿悬臂式掘进机空间位姿监测系统，以期为实现悬臂式掘进机的自动化和智能化提供技术参考。

Fig.2 Sensors placed on the deck of BLSRP

Fig.3 Mounting underwater accelerometer on buoyant leg

Fig.4 Surge response of deck using wireless sensor

where τ is the center of the window in time and ε is the mean frequency of the window.ω(t- τ)e-jεt is the STFT,which traverses along the length of the signal.

Table 2 Specifications of sensors

Accelerometer description Wired Wireless Underwater 393B04 MPU6050 W 393B04 Type Integrated circuit piezoelectric MEMS Water resistant ICP accelerometer Number of axis 1 3 1 Range/g ±5 ±16(opted±2) ±5 Sensitivity 1 V/g 16,384 LSB/g 992 mV/g Noise performance/μg/√Hz 0.30 400 -

Fig.5 Surge response of deck using wired sensor

Fig.6 Comparison of deck responses in surge degrees-of-freedom

Fig.7 Comparison of deck responses in pitch degrees-of-freedom

Fig.8 Surge RAO of deck

Fig.9 Heave RAO of deck

Experimental investigations of BLSRP are carried out by deploying wireless sensor network;results are also compared with that of wired sensors for deriving vital discussions.In the present study,architecture of wireless sensor network and design of sensor node layout,base station,and communication protocol are discussed.While sensor configurations are arrived based on their adaptability to the proposed architecture,sensor network is optimized in terms of location and scalability(Lee et al.2016).The present study analyzes acceleration and displacements of the deck and buoyant legs in all active degrees-of-freedom using the proposed SHM scheme.Postulated failure cases are instigated by tether pull out;damage is identified by comparing dissimilarities between the recorded signals from a less number of sensors.The proposed SHM scheme thus limits the data size that needs to be processed and communicated.Alert Monitoring System(AMS),which is a part of the developed sensor network,triggers alert messages and email to the authentic users when the measured response exceeds the threshold value.

2 Design of SHM System Using Wireless Sensor Network

Fig.10 Pitch RAO of deck

Fig.11 Surge RAO of buoyant leg

Signal is decomposed into weighted combinations of sinusoids with different frequencies.Fourier transform correlates the signal with the basis function e-j2πft for specific value of frequency(f).This process is repeated for frequencies ranging from-∞to+∞.PSD of the signal represents the distribution of power across these frequencies.Global feature of the signal in the frequency domain is extracted by computing the PSD of the signal and is given by:

当然，有时候单看颜色也不能说明作者就有抄袭行为，因为有时作者会引用别人的文字或是引用作者自己曾经写过的文章。这种情况下，学术不端文献检测系统提供了百分比数据供参考，一是总文字复制比，另一个是去除引用文字复制比。一般来说，去除引用文字复制比是重要的参考数据。另外，学术不端文献检测系统还提供了疑似问题文章与数据库中已有文章的内容对比参照，将有相同文字的文章用两个窗口一起摊开，相同文字的内容用红色显示，方便检测者判断。

该电影产业大数据可视化Hybrid App整体采用Cordova框架，主体为Web View，iOS端由UIWebView或WKWebView实现，Android端由Webview或Crosswalk实现。系统的结构框架图如图1所示。

Fig.12 Heave RAO of buoyant leg

Fig.13 Pitch RAO of buoyant leg

Structural details of BLSRP are given in Table 1.Each buoyant leg is ballasted with weights to achieve the required draft and pretension in tethers.Ball joints allow superstructure to move in the displacement degrees-of freedom,but restrain the transfer of rotational degrees-of freedom from the legs to the deck;this isolates the deck and hence response is reduced under wave action.Scaled model of BLSRP is installed in the deep-water wave flume and sensors are placed on the deck and legs of the BLSRP as shown in Fig.2.Underwater accelerometer is mounted on the leg of the platform,well below the draft level as shown in Fig.3.Specifications of various sensors used in SHM architecture are given in Table 2.

SHM architecture used in the present study is shown in Fig.1.Sensor nodes will monitor the integrity of the structure and acquire and transmits data to the base station,and the server at the base station will further process the data.Threshold values are identified by benchmarking the average of previous records in non-critical conditions for the same structure.Under wave action,displacements are measured by the sensors,which shall be used to determine the damage level detected by the sensor node by comparing it with that of the threshold values.

However,for integrating the decision-making process with that of the SHM system,provision is made to enable changes in the threshold value by the end user at the client interface.If the acquired data is below the threshold limit,structure is considered to be in its healthy state.Otherw ise,processing unit will trigger the email alert and transmit the data to base station.Subsequently,server in the base station will process the data and perform detailed analysis.As a result,alert messages will be displayed in the user interface.The server will further trigger SMS to the registered mobile number associated with the SHM system.

随着教育的改革，小学英语的教学也应该得到更新，翻转课堂的教学模式在英语教学中能够使师生之间取长补短，互相学习，改善师生之间的关系，可以最大程度提高课堂学习效率和学生学习英语的热情。

Table 3 Comparison of deck response

Wave period/s Deck response amplitude operator Surge Heave Pitch Wired Wireless Diff in% Wired Wireless Diff in% Wired Wireless Diff in%1.6 1.3172 1.4878 12.9462 0.4391 0.3421 -22.0848 1.224 1.4596 19.2483 1.8 2.0054 2.0868 4.0562 0.4027 0.3315 -17.6758 1.4528 1.5512 6.77312 2 3.1719 3.9614 24.8902 0.8651 0.7946 -8.13999 1.7088 2.0957 22.6416 2.2 2.9116 3.2966 13.2231 0.8181 0.7624 -6.80387 1.28 1.4311 11.8046 2.4 1.6179 1.8528 14.5132 0.5727 0.6411 11.92661 0.6288 0.7822 24.3956 2.6 1.0922 1.2804 17.2307 0.4369 0.4471 2.328865 0.408 0.5176 26.8627 2.8 0.8575 1.0212 19.0909 0.6041 0.5492 -9.08846 0.2592 0.2849 9.91512 3 0.8725 0.9329 6.92307 0.8054 0.6709 -16.695 0.256 0.3063 19.6484

Table 4 Comparison of buoyant leg response

Wave period/s Buoyant leg response Surge Heave Pitch Wired Wireless Diff in% Wired Wireless Diff in% Wired Wireless Diff in%1.6 1.3427 1.4432 7.48815 0.10314 0.2226 18.88309 1.152 1.02441 11.0755 1.8 1.8121 2.0409 12.6222 0.1527 0.1871 22.58065 1.3792 1.112 19.3735 2 2.9332 3.4553 17.7966 0.2645 0.4404 14.9394 1.6176 1.79561 11.0045 2.2 2.3220 2.5772 10.8999 0.2268 0.2734 20.54054 1.0704 1.34668 25.8109 2.4 1.6895 1.5105 10.5932 0.1255 0.1856 47.90698 0.664 0.91225 37.3870 2.6 1.0922 1.0048 8 0.1164 0.1526 31.02941 0.4336 0.4286 1.15314 3 1.0738 0.9105 15.2083 0.1277 0.1571 23.03571 0.2912 0.3972 36.4011 3.2 0.9672 0.9189 5 0.1556 0.1868 20 0.1856 0.22699 22.3006

3 Experimental Model of BLSRP

To validate the wireless SHM system,response of the scaled model is obtained using both wireless and wired sensor system.Sensors are placed at appropriate locations on the deck and buoyant legs while the model is excited by a 10-cm wave height;wave period is varied from 1.6 to 3.4 s.Surge response of the deck under regular waves(10 cm,1.8 s)acquired using both wireless and wired sensors is shown in Figs.4 and 5,respectively.

Fig.14 Surge response of deck under normal load

Sensing unit used in the present study is MPU6050,a MEMS-based accelerometer and gyroscope module,comprising of tri-axial accelerometer,tri-axial gyroscope,and digital motion processor.It also features three,16-bit digital analog to digital converters(ADC)for digitizing the output from accelerometer and inclinometer.Full scale range(2 g)is opted at the sensitivity of 16,384 LSB/g.Sensing units,which are connected to the processor board through the GPIO pins,shall acquire the data and transmit it to the base station.There are several protocols available for sensor networking.Some of the commonly used protocols for wireless communication are(i)IEEE 802.11 including Wi-Fi,which has a range up to 10 km with Yagi antennas operating under ideal condition;(ii)IEEE 802.15.2 protocols for low power consumption data networks;and(iii)IEEE 802.15.4 protocol for the wireless sensor node ranges up to 300 m with high gain antennas.IEEE 802.11 protocol is used in the present study at an operating frequency range of 2.4 GHz,enabling data transmission between sensor nodes and the base station.It is interesting to note that transmitting at this frequency with the available standard protocol is a challenging issue under offshore environment.Hence,under the real offshore environment,the technology should support Voice over IP,broadband data,and video communication services for different topologies of sensors.Improvements in satellite,VSATs(very small aperture terminal),and antenna systems will enable it to meet the demand for higher bandwidth in offshore platforms.The VSAT services support primary business process,high quality voice calls,and broadband Internet services for offshore platforms.

Experimental investigations are carried out under normal sea state for regular waves.Acquired structural response of the deck and buoyant legs is used to fix the threshold values.Postulated failure cases are investigated by applying eccentric load to one of the buoyant legs,BLS 1(referred as postulated failure case 1,while tether pull out of BLS 1 is referred as postulated failure case 2).Postulated failure case 3 refers to the response of the platform under extreme waves;postulated failure modes are created to demonstrate the effectiveness of A lert Monitoring System(AMS).

Fig.15 Surge response of deck under postulated case 1

Fig.16 Surge response of deck under postulated case 2

4 Data Processing

Above equation can be interpreted as the expectation of the FT of the signal computed over an infinite time period.STFT extracts the local features of the signal both in time and frequency domains.This slices the signal into different segments using the window function ω(t)and each of these segments is subjected to FT.The window function is placed such that the center of the window coincides with the start of the signal and it traverses along the length of the signal and is given by:

Fig.17 Heave response of deck under normal load

Fig.18 Heave response of deck under postulated case 2

Design of SHM system involves integration between the hardware and software components,while sensor node is the primary component of the design.It includes a processing unit,a sensing unit,and a transceiver.Processing unit deployed in the current study is the Raspberry pi ARMv7,which is a low-cost computing device that allows integration of multiple input and output peripherals.Sensor unit and processor board are connected through the GPIO pins,the processor unit uses Inter-Integrated Circuits(I2C)and Serial Peripheral Interface(SPI)protocols to communicate with the sensing units.Processing unit operates on Linux-based Raspbian Wheezy Operating system,which is capable of both controlling and collecting logs from the sensing units.Extended memory is provided by adding SD cards externally to store the data until it is transmitted.While sensor nodes are supported by mobile power banks during experiments,power optimization is one of the most challenging factors in the field of SHM.Extending lifetime of the WSNs includes energy harvesting using solar,w ind,and vibration energy(Jahangiri et.al.2016).Ultra-power circuit boards operating on nano-watts power are other viable alternates(Lee et al.2016).

Fig.19 Comparison of heave response of deck and buoyant leg

Fig.20 Heave response of BLS 4 under normal case

Fig.21 Heave response of BLS 4 under failure case 1

Fig.22 Comparison of heave response of BLS 4(normal and extreme waves)

In frequency domain,data is processed to obtain power spectral density(PSD)by performing the fast Fourier transform(FFT)to extract the global features and short-time Fourier transform(STFT)to determine localization of frequency with respect to time.Signal considered in the present study is the measured time history of the acceleration response.For a time-varying function,Y(t),Fourier transform(FT)is given by:

One of the most common difficulties in real-time monitoring is correlating the measured response quantities.For example,vibration amplitude or frequency is to be correlated with degradation of the structural system;stiffness degradation and mass degradation result in different damage detection approaches.Based on the probabilistic damage of compliant offshore platforms under large displacements and tether pull out,researchers proposed effective methods of controlling the response;once SHM tools are in place,periodic monitoring can improve their safe operability(Chandrasekaran et al.2013).Offshore platforms under extreme waves are no exceptions from continuous monitoring as their downtime caused by unprecedented environmental loads shall challenge their functioning(Chandrasekaran and Yuvraj 2013).Wave-induced vibration control strategies are equally efficient and viable alternatives to ensure safe operability and structural integrity of offshore platforms(Zhang et al.2017;Moharrami and Tootkaboni 2014;Wang et al.2013;Li et al.2016).Structural integrity can also be assured using control methods in fixed offshore platforms;results show that even complaint offshore platform motion is controlled successfully in stiff degree-of-freedom,like heave(Patil and Jangid 2005;Alves and Baitsta 1999).Recent applications in health monitoring used wireless sensors with dual microcontrollers for experimental investigations;design of suitable data acquisition system to implement numeric algorithm is also demonstrated(Lynch et al.2004a,b;Wang et al.2007).Effectiveness of deploying wireless sensor networking to monitor various infrastructures is evident in the literature(Pérez et al.2011;Mollineaux et al.,2014;Spencer 2016;Wang et al.2013;Jalalpour 2016).Success of these applications intuited necessity and methods of implementing sensor networking to monitor offshore and coastal structures(Chandrasekaran andThailammai 2016).For example,integrity monitoring of offshore jacket platform is performed to locate the damages and estimate severity under noise-free measurement using wired sensors(Hossienlou et al.2016).Yu and Ou(2008)measured scaled response of offshore platform model using wireless sensors and show ed a close agreement with that of the analytical response.Hull monitoring system proposed by Swartz et al.(2009)deployed wireless sensors to acquire both acceleration and strain measurements,which are subsequently used to compute modal properties of the vessel.They showed overlap of measurements with that of wave frequencies but appeared distinctly clear at higher frequencies.Uncertainties that arise due to environmental loads during drilling and non-stationary responses that arise from continuous change in mass and stiffness characteristics are some of the real challenges in vibration measurements of offshore structures(Brownjohn 2007;Brow njohn et al.2011).Accounting for all complexities,the present study proposes a wireless sensor network for response measurement of offshore Buoyant Leg Storage and Regasification Platforms(BLSRP)under postulated failure.Preliminary experimental investigations carried out on scaled model of offshore tension leg platforms highlighted appropriate selection of wireless sensors and their networking geometry for data acquisition(Chandrasekaran and Thailammai 2016;Jahangiri et al.2016).Vital response parameters,namely deck response and dynamic tether tension variations that affect operational safety of LNG process plants and BLSRP,are well illustrated through detailed studies(Chandrasekaran and Lognath 2016);the current study uses those parameters as basis for health monitoring.

Fig.23 STFT of deck response in surge under normal case

Fig.24 STFT of deck response in surge under postulated failure(case 2)

5 Validation o f the SHM Design

The SHM system is installed on a scaled(1:150)model of a BLSRP.BLSRP is mainly used to process the LNG and export to the onshore facilities.Deck plate houses a stainless steel tank,which is supported by six buoyant legs.Deck of the platform rests on buoyant legs,which in turn are fixed to the seabed using taut-moored tethers.Deck and buoyant legs are isolated using ball joints(Chandrasekaran and Lognath 2017).

Fig.25 STFT of deck response in heave under normal case

Fig.26 STFT of deck response in heave under postulated failure(case 2)

Comparison of deck responses in both surge and pitch degrees-of-freedom is shown in Figs.6 and 7,respectively.Platform is excited under unidirectional wave;rotational degrees-of-freedom are also activated due to the differential heave motion of the platform.Pitch response is obtained using wired inclinometer;in case of wireless SHM,it is acquired using MPU6050,which is a combination of gyroscope and accelerometer.It is seen from the figures that peak frequency of surge response of the wireless sensor is obtained at 0.624 Hz and that of the wired sensor is at 0.56 Hz;computed error in the peak frequency is about 11.4%while that of the peak magnitude is about 14.44%.Pitch response shows an error of about 11.5%.Response amplitude operator(RAO)of deck and BLS 1,acquired by both wired and wireless sensors,is plotted and compared for surge,heave,and pitch degrees-of-freedom.Figures 8,9,and 10 show the comparison of RAO of the deck in surge,heave,and pitch degrees-of freedom while Figs 11,12,and 13 show that of the buoyant leg,respectively.Comparison of responses is highlighted in Tables3 and 4,respectively.It is seen from the tables that there is a good agreement between the data acquired by wired and wireless sensors.While the difference is mainly due to the sensitivity of the device,it shall also be attributed to the acquisition method and processing techniques.

Fig.27 STFT of heave response of BLS 4 under normal case

Fig.28 STFT of heave response of BLS 4 under postulated case 1

6 Postulated Failure Cases

Fig.29 Layout of proposed A lert Monitoring System

Fig.30 Comparison of deck response in surge

BLSRP is also analyzed under the postulated failure cases to assess the adaptability and efficiency of the proposed SHM network.Platform is investigated under the eccentric load and the tether pull out cases,respectively.Figure 14 shows the PSD of the deck response in surge degree-of-freedom,under normal loading while Figs 15 and 16 show the responses under postulated failure cases of 1 and 2,respectively.Response of the platform under postulated failure cases is acquired when the platform is subjected to regular waves(10 cm,2 s).It is seen from the figures that under normal case,peak frequency occurs at 0.52 Hz while it shifts to 0.56 Hz under damage case 1.Under the postulated failure of case 2,two peaks are seen at0.46 and 0.92 Hz,which is double of the earlier peak frequency.In addition,two minor peaks are also seen,one at a very low frequency and the other at 1.36 Hz,which is thrice that of the first peak frequency.Higher frequency responses of the deck show influence of response of buoyant legs on the deck motion due to their stiff connectivity in vertical plane.Heave response of the deck is shown in Figs.17 and 18 under normal and postulated failure case 2.It is seen from the figures that the first peak occurring at about 0.5 Hz does not shift even under the postulated failure but shows a significant increase in the magnitude.Reduction in the magnitude of the second peak occurring at about 1 Hz shows the influence of buoyant leg on the deck response in heave motion.Peaks occurring a thigher frequencies are attributed to the pitch motion of the deck,resulting from differential heave.Under the postulated failure of tether pull out,differential heave becomes predominant,which causes the first peak to magnify without any significant change in the frequency band.This behavior is clearly captured by the sensor network,deployed in the current study.Smaller peaks seen 2.5 and 3.5 Hz are not magnified and hence not transferred to the deck due to the isolation of superstructure from the base.

Fig.31 STFT of surge acceleration of BLS 1 under normal load

Fig.32 STFT of surge acceleration of BLS 1 under extreme waves

7 Response o f Buoyant Legs

Under no postulated failure,comparison of heave response of the deck and buoyant legs(4,5)acquired using underwater accelerometer is shown in Fig.19.Two sensors are deployed to measure the response on buoyant leg(BLS 4)just above and below the draft level.It is seen from the figure that both the sensors are capable of acquiring the response of buoyant leg in comparable accuracy.Peaks occurring at very low frequency and at 0.5 Hz show the coupled response of heave and surge degrees-of-freedom.Deck response is minimum in comparison to that of the buoyant leg,which is attributed to the presence of ball joint.

Fig.33 Email display of Alert Monitoring System

Figures 20 and 21 show PSD of heave response of BLS 4 for normal case and postulated failure case 1,when excited by(10 cm,2 s)regular waves.Under no postulated failure,first peak occurs at 0.52 Hz,which show s the maximum amplitude.Other successive peaks show coupling effect of buoyant legs with that of the deck response as the frequencies correspond to the buoyant legs in surge and heave modes.Under the postulated failure of tether pull out(case 2),sensors acquire response with significant magnitude at series of successive frequencies,indicating the effect of failure of buoyant leg.Variation in heave response in BLS 4 is magnified at all peak frequencies due to eccentric load added at the BLS 1,which is positioned diagonally opposite to BLS 4.The response variation due to eccentric loading in BLS 1 is not uniform in all the buoyant legs,which is seen by comparing the heave responses of BLS 4 under normal and extreme waves(Fig.22).

8 Localization of Frequencies Under Postulated Damage Scenarios

Localization of frequencies with respect to time is given by the STFT.Figures 23 and 24 show STFT of deck response in surge degree-of-freedom under normal loading and postulated failure of tether pull out(case 2),respectively.It is seen from the figures that there is no much variation in the plots of normal and damage case 2,except the occurrence of second peak at 0.91 Hz.Figures 25 and 26 show the deck response in heave degree-of-freedom under normal case and damage case 2.It is seen from the figures that amplitude values of the deck response are higher in comparison to that of the normal case,apart from a wide frequency distribution.

Figures 27 and 28 show STFTof heave response of BLS 4 under normal case and damage case 1.Variations in amplitude under normal case seen at 0.5,1.1,and 1.5 Hz are magnified under the postulated failure at all peak frequencies unlike in case of tether pull out.A wide spectrum of frequency distribution under the postulated damage case shows necessity of data acquisition for a wide bandwidth to capsulate efficient alert monitoring of the platform.

9 Alert Monitoring System

The processing of the indigenously designed A lert Monitoring System is shown in Fig.29.The model is excited by a regular wave of 16 cm wave height(corresponds to very severe sea state in proto type)to assess the efficiency of the proposed Alert Monitoring System.Under the extreme wave condition,response of the platform exceeds the threshold value in surge,pitch,and heave degrees-of-freedom as seen from Fig.30.Lim it values of the response in the active degrees-of freedom are user-defined and fixed based on the history of records for the normal condition and with design values.Figures31 and 32 show STFT of surge response under normal and damage case 3,respectively.As threshold value is exceeded,alert message is triggered.Proposed SHM system shall first trigger an alertemail to the user,whose screen shot is shown in Fig.33.Subsequently,data is transmitted to the base station;after further processing,SMS alert will be triggered by the web server as shown in Fig.34.Screen shot of the user interface is shown in Fig.35,displaying that the structure has exceeded threshold in surge,pitch,and heave responses.Detailed view of each sensor can be viewed on clicking the corresponding sensor on the graph shown in the user interface.Complete details of the value,exceeding threshold are obtained by clicking the alert view.

“广场舞”是一种集健身、娱乐、审美为一体，以大众为主体，配以富有节奏感的音乐，在广场、公园、院坝等宽敞场地开展的群众性健身舞蹈活动。它属于传统舞蹈与音乐的一种变奏式创编，包容性很强，既具有舞蹈审美的情趣，又具有健身运动的特点，适宜广大人群参与其中，所以，其可塑性、艺术性、简易性、普及性等特点明显。随着全民健身上升为国家战略，群众体育的发展步伐也随之加快，广场舞作为适应时代发展的创意健身舞蹈活动，展现了自身的发展优势。在健康中国2030体系推进过程中，需要广场舞等群众性体育项目参与建设，通过广场舞健身活动类创意打造“健康城市”名片，发挥社群价值，促使中老年人群广泛参与健身等[13]。

10 Conclusions

Design of SHM system using wireless sensor networking and its implementation in an experimental model is presented.Application of WSN in SHM systems ensures the health and safety of the platforms,minimizes the economic loss,reduces the system cost and installation time,ensures preventive maintenance approach instead of post damage repairs,and enables local data processing and efficient traceability of sensor failure.Closer agreement of response acquired by various sensors under wired and wireless SHM system validates the proposed design architecture of wireless SHM.Postulated damage cases enabled data acquisition by various sensors in the network,which is linked to Alert Monitoring System.Alert Monitoring System,proposed in the present paper,along with the network of wireless and wired sensor comprises SHM system for offshore platform,which is an indigenous design.Attempted study is a prime facie to improve functional safety of offshore production platforms apart from planning for a preventive maintenance,which is well demonstrated in a lab scale.

原发性甲状腺淋巴瘤(PTL)起源于淋巴细胞及其前体细胞，感染、免疫缺陷、物理或化学因素及遗传倾向等均可能与PTL的发生有关，但其确切的病因及发病机制目前并不明确。Watanabe等报道，与未患桥本甲状腺炎的人群相比，患有桥本甲状腺炎者发生PTL的风险相对较高[5]。屈兵等认为，桥本甲状腺炎可被认为是甲状腺癌的癌前病变，研究发现二者有内分泌功能失调及免疫功能缺陷的共同发病基础[6]。也有学者认为，PLT的发生是1个慢性过程，从慢性淋巴炎发展为黏膜相关组织淋巴瘤，最后转变为PLT[7]。

References

Alves RM,Batista RC 1999 Active/passive control of heave motion for TLP type of offshore platforms.International Offshore and Polar Engineering Conference(ISOPE),Brest,France,332-338.Document ID:ISOPE-I-99-050

Brownjohn JM(2007)Structural health monitoring of civil infrastructure.Philos Trans R Soc A Math Phys Eng Sci 365(1851):589-622.https://doi.org/10.1098/rsta.2006.1925

Brownjohn JM,De Stefano A,Xu YL,Wenzel H,Aktan AE(2011)Vibration-based monitoring of civil infrastructure:challenges and successes.J Civ Struct Heal Monit 1(3-4):79-95.https://doi.org/10.1007/s13349-011-0009-5

Celebi M(2002)Seismic Instrumentation of buildings,Special GSA.USGS Project,an administrative report,United States Geological Survey,Menlo Park,CA

Chandrasekaran S(2011)Risk assessment and management in offshore and petroleum industries.In:Key note address at Pre-conference workshop on International Conference on Asia Pacific HSE Forum on Oil,Gas and Petrochemicals,Fleming Gulf Conferences,p.81

Chandrasekaran S(2015a)Dynamic analysis and design of ocean structures.Springer,India

Chandrasekaran S(2015b)Advanced marine structures.CRC Press,Florida(USA)

Chandrasekaran S(2016a)Health,safety and environmental management for offshore and petroleum engineers.Wiley,U.K.

Chandrasekaran S(2016b)Offshore structural engineering:reliability and risk assessment.CRC Press,Florida,pp 14-987

Chandrasekaran S,Koshti Y(2013)Dynamic analysis of a tension leg platform under extreme waves.J Nav Archit Mar Eng 10:59-68.https://doi.org/10.3329/jname.v10i1.14518

Chandrasekaran S,Lognath RS(2016)Dynamic analyses of buoyant leg storage regasification platform(BLSRP)under regular waves:experimental investigations.Ships Offshore Struct12(2):227-232.https://doi.org/10.1080/17445302.2015.1131006

Chandrasekaran S,Lognath RS(2017)Dynamic analyses of Buoyant Leg Storage and Regasification platforms:numerical studies.J Mar Syst Ocean Tech 12:39-48.https://doi.org/10.1007/s40868-017-0022-6

Chandrasekaran S,Thailammai C(2016)Health monitoring of offshore structures using wireless sensor network:experimental investigations.Proc.SPIE 9804,Nondestructive Characterization and Monitoring of Advanced Materials,Aerospace,and Civil Infrastructure 2016,980416.https://doi.org/10.1117/12.2219208

Chandrasekaran S,Kumar D,Ramanathan R(2013)Dynamic response of tension leg platform with tuned mass dampers.J Nav Archit Mar Eng 10(2):149-156.https://doi.org/10.3329/jname.v10i2.16184

Chandrasekaran S,Thailammai C,Khader SA(2016)Structural health monitoring of offshore structures using wireless sensor networking under operational and environmental variability.Int JEnviron Chem Ecol Geol Geophys Eng 10(1):33-39

Farrar CR,Worden K(2007)An introduction to structural health monitoring.Philos Trans R Soc Lond A Math Phys Eng Sci 365(1851):303-315.https://doi.org/10.1098/rsta.2006.1928

Hosseinlou F,Mojtahedi A(2016)Developing a robust simplified method for structural integrity monitoring of offshore jacket-type platform using recorded dynamic responses.Appl Ocean Res 56:107-118.https://doi.org/10.1016/j.apor.2016.01.010

Jahangiri V,Mirab H,Fathi R,Ettefagh MM(2016)TLP strucLat Am J Solids Structtural health monitoring based on vibration signal of energy harvesting system.13(5):897-915.https://doi.org/10.1590/1679-78252282

Jalalpour M,Azarbayejani M,El-Osery A,Taha MR(2016)A new damage detection and tracking method using smartsensor network.JCiv Struct Heal Monit 6(2):291-301.https://doi.org/10.1007/s13349-016-0167-6

Lee Y,Blaauw D,Sylvester D(2016)Ultralow power circuit design for wireless sensor nodes for structural health monitoring.Proc IEEE 104(8):1529-1546.https://doi.org/10.1109/JPROC.2016.2547946

Li HN,Ren L,Jia ZG,Yi TH,Li DS(2016)State-of-the-art in structural health monitoring of large and complex civil infrastructures.J Civ Struct Heal Monit 6(1):3-16.https://doi.org/10.1007/s13349-015-0108-9

Lynch JP,Law KH,Kiremidjian AS,Carryer E,Farrar CR,Sohn H,Allen DW,Nadler B,Wait JR(2004a)Design and performance validation of a wireless sensing unit for structural monitoring applications.Struct Eng Mech 17(3-4):393-408

Lynch JP,Sundararajan A,Law KH,Kiremidjian AS,Carryer E(2004b)Embedding damage detection algorithms in a wireless sensing unit for operational power efficiency.Smart Mater Struct13(4):800-810.https://doi.org/10.1088/0964-1726/13/4/018

Moharrami M,Tootkaboni M(2014)Reducing response of offshore platformsto wave loadsusing hydrodynamic buoyant mass dampers.Eng Struct 81:162-174.https://doi.org/10.1016/j.engstruct.2014.09.037

Mollineaux M,Balafas K,Branner K,Nielsen P,Tesauro A,Kiremidjian A,Rajagopal R(2014)Damage detection methods on wind turbine blade testing with wired and wireless accelerometer sensors.EWSHM-7th European Workshop on Structural Health Monitoring.[hal-01022037]

Patil KC,Jangid RS(2005)Passive control of offshore jacket platforms.Ocean Eng 32:1933-1949.https://doi.org/10.1016/j.oceaneng.2005.01.002

Pérez CA,Jimenez M,Soto F,Torres R,López JA,Iborra A(2011)A system for monitoring marine environments based on wireless sensor networks.OCEANS,2011 IEEE-Spain,1-6.https://doi.org/10.1109/Oceans-Spain.2011.6003584

Spencer BF Jr,Jo H,Mechitov KA,Li J,Sim SH,Kim RE,Cho S,Linderman LE,Moinzadeh P,Giles RK(2016)Recent advances in wireless smart sensors for multi-scale monitoring and control of civil infrastructure.J Civ Struct Heal Monit 6(1):17-41.https://doi.org/10.1007/s13349-015-0111-1

Straser EG,Kiremidjian AS,Meng TH(2001)Modular,wireless damage monitoring system for structures.US Patent 6,292,108

Swartz RA,Zimmerman AT,Lynch JP,Brady TF,Rosario J,Salvino LW,Law KH(2009)Wireless hull monitoring systems for modal analysis of operational naval vessels.Proceedings of the International Modal Analysis Conference(IMAC)XXVII,57

Wang Y,Lynch JP,Law KH(2007)Aw ireless structural health monitoring system with multithreaded sensing devices:design and validation.Struct Infrastruct Eng 3(2):103-120.https://doi.org/10.1080/15732470600590820

Wang S,Yue Q,Zhang D(2013)Ice-induced non-structure vibration reduction of jacket platforms with isolation cone system.Ocean Eng 70(15):118-123.https://doi.org/10.1016/j.oceaneng.2013.05.018

Yu Y,Ou J(2008)W ireless sensing experiments for structural vibration monitoring of offshore platform.Front Electr Electron Eng Chin 3(3):333-337.https://doi.org/10.1007/s11460-008-0051-1

Zang B-L,Han Q-L,Zhang X-M(2017)Recent advances in vibration control of offshore platforms.Nonlinear Dyn 89(2):755-771.https://doi.org/10.1007/s11071-017-3503-4