Development and performance assessment of a hybrid telemetry system for Indian tsunami buoy system

Transcription

1 doi: /ut Underwater Technology, Vol. 33, No. 2, pp , Development and performance assessment of a hybrid telemetry system for Indian tsunami buoy system Technical Paper R Sundar*, R Venkatesan, M Arulmuthiah, N Vedachalam and MA Atmanand National Institute of Ocean Technology, Chennai, India Received 30 March 2015; Accepted 4 September 2015 Abstract This paper describes a novel technique developed and demonstrated in an Indian tsunami buoy, for real-time data transmission from a moored surface buoy, using Indian National Satellite (INSAT) system and Inmarsat telemetry. Based on the identified safety reliability (SR) performances, the highly reliable Inmarsat telemetry link is configured to transmit the water level data during the tsunami mode. The INSAT telemetry with an SR of Safety Integrity Level 1 (SIL1) is configured for transmitting normal mode data and to maintain the SR of the Inmarsat telemetry link. To comply with the required SR performance, the buoy located Inmarsat transmitter is powered every 1hr to get the buoy position, and once in every 24hrs it sends the data to check the overall link in SIL4. Their proof test intervals are set at 1hr and 24hrs respectively. The response of the cost-effective hybrid system to a seismic event in 2014 gives the confidence on the hybrid telemetry system s reliable support to the Indian tsunami buoy system. Keywords: INSAT, Inmarsat, proof test interval, safety reliability, tsunami, Safety Integrity Level Acronym list ADDRESS advanced data reception and analysis system BPR bottom pressure recorder FRACAS failure reporting, analysis and corrective action system FIT failure-in-time HS&E health, safety and environment ITBS Indian tsunami buoy system INCOIS Indian National Centre for Ocean Information Services INSAT Indian National Satellite MSB moored surface buoys MTTF mean time to failure * Corresponding author. rsundar@niot.res.in MTTR NIOT NIOT-MCC NOAA OOS PTI SR SIL TDMA mean time to repair National Institute of Ocean Technology National Institute of Ocean Technology, Mission Control Centre National Oceanic and Atmospheric Administration ocean observation systems PFD: probability of failure on demand proof test interval safety reliability Safety Integrity Level time division multiple access 1. Introduction Tsunamis are an ever-present threat to lives in the 7500km-long India s coastline, where 30% of the national population resides. The Indian Ocean has two tsunamigenic zones: the Andaman-Sumatra trench and the Makran coast. Following the devastating 2004 tsunami event in South Asia, which resulted in the loss of more than 240,000 lives and property in 11 nations surrounding the Indian Ocean (Wijetunge, 2009), there has been an increasing concern on future tsunami threats. Hence, there is a rapid growth in the interest of tsunami detection and warning systems required for warning purposes. A reliable telemetry system is the linchpin for a successful ocean observation system (OOS), which involves offshore moored surface buoys (MSB), acquiring and transmitting timecritical oceanographic information to the shore centres in order to advance decisions during disastrous events such as cyclones and tsunamis (Venkatesan et al., 2013a). The OOSs in the National Institute of Ocean Technology (NIOT) have the mandate of developing, 105

2 Sundar et al. Development and performance assessment of a hybrid telemetry system for Indian tsunami buoy system operating and maintaining moored buoy observational and related telecommunication networks in the Indian waters since They have carried out nearly 750 moored buoy deployments for collecting meteorological, water surface and subsurface parameters from coastal to deep-sea locations. They have also provided the tsunami water level data to the tsunami warning centre from 2006 onwards (Venkatesan et al., 2013b). Five Indian tsunami buoy systems (ITBSs) transmit the real-time data to two centres the NIOT Mission Control Centre (NIOT-MCC) in Chennai, and the Indian National Centre for Ocean Information Services (INCOIS) warning centre in Hyderabad for data modelling and public notification in the event of a tsunami (Venkatesan et al., 2015a). The 24 7 manned NIOT-MCC receives the data from the tsunami buoys, using the customised advanced data reception and analysis software (ADDRESS). This software enables the data analysis and dissemination of the data to the decision-makers (Venkatesan et al., 2015b). The locations of the tsunami buoys forming part of the ITBS are shown in Fig 1, and more details are shown in Table 1. NIOT uses the Government of India approved Inmarsat telemetry for real-time data telemetry for all the moored tsunami buoys with two-way communications (Venkatesan et al., 2013). 2. Satellite telemetry systems for ocean monitoring Considering the reliability of satellite telemetry, globally moored buoy operators use the services of satellites, such as the Iridium (Leopold and Miller, 1993) and Inmarsat (Graff et al., 2006), for real-time data transmission from global OOSs. The Iridium satellite network provides global coverage featuring low transmission power, high data rates and low latency, and capable of providing two-way communication capability. However, it is not authorised in India and China. Instead, the International Maritime Organization (IMO) operated Inmarsat based satellite transmission is the most attractive option for deep-ocean MSB for reliable real-time data telemetry. The Inmarsat -C, which has 10 satellites positioned in the geo-stationary orbit, provides two-way communication, data storage and forward facility. It also features data acknowledgment between the satellite and the transmitter for every data transmission. The service, which offers communication at speeds between 600bps and 492kbps, is reliable and secure, and charges approximately US$ 0.15 for transmitting 32 bytes of data from the offshore MSB. The Government of India operated geo-stationary satellites, Indian National Satellite (INSAT) 3A and 3C (Zacharia et al., 2014), are provided free of cost to the Indian Ocean monitoring programme. Fig 1: Location of tsunami buoys in Indian waters 106

3 Underwater Technology Vol. 33, No. 2, 2015 Table 1: Details of the tsunami buoys in ITBS Buoy Id World Meteorological Organization ID Latitude (N) Longitude (E ) Depth in metres ITB ,966 ITB ,220 ITB ,725 ITB ,284 ITB ,050 However, they do not provide the data acknowledgement facility or the capability of working with time division multiple access (TDMA) technology. 3. Safety reliability requirements for tsunami buoy telemetry 3.1. Need for safety reliability For an effective tsunami early warning system, time is of essence. Braddock specifies the time constraint for the system as: t 1 +t 2 + t 3 t 4 (1) where t 1 is the detection time, t 2 is the assessment time, t 3 is the evacuation time and t 4 is the tsunami travel time to the shore (Braddock and Carmody, 2001). As t 4 is in the order of a few hours, the detection time t 1 should be as low as reasonably practicable, so as to have higher t 2 and t 3. Thus the systems installed in the challenging offshore environment and used for detecting an event should be highly reliable. As this is a critical safety system that is required to be available on demand, during the entire operating period, the SR of the system has to be as high as reasonably practicable IEC standard IEC (International Electrotechnical Commission, 2000) is a standard for implementing instrumented safety systems, using the principles of the Safety Integrity Level (SIL) taking into consideration the health, safety and environment (HS&E) aspects (Smith and Simpson, 2004; Yoshimura and Sato, 2008). Safety systems need to perform their intended operations on demand. SIL is defined based on the probability of failure on demand (PFD). SIL defines the degree of safety protection required by the process, and consequently the safety reliability (SR) of the system necessary to achieve the function. SIL has four levels, and the higher the number, the safer the function is considered to be. Table 2 describes the SIL levels, with the corresponding PFD. Proof test interval (PTI) is the periodic healthcheck test conducted on the safety system to confirm its operational healthiness, so that its meets Table 2: PFD and SIL levels as per IEC standards Safety integrated level (SIL) to to to to 10 5 Probability of failure on demand (PFD per year) the SR. Based on IEC 61508, the SIL rated system has to meet the following necessary condition: Number of demands on the system < 2 proof test frequency (2) 3.3. Computing SIL for tsunami telemetry Based on the IEC 61508, the SIL requirements for the application are determined by taking into consideration the risk consequence, alternative inplace safety system, human occupancy and the demand rate on the system. The methodology is applied for computing the SR of time-critical safety systems in offshore oil and gas production systems (Vedachalam, 2013; Vedachalam et al., 2014a; Vedachalam et al., 2014b). A similar approach is adopted to determine the SIL requirements for the tsunami buoy telemetry system. The risk matrix is used to evaluate the SIL level needed for the telemetry system and requires computation of the avoidance, occupancy and demand rate parameters. Based on the availability or unavailability of the alternative in-place safety protection system, the avoidance parameter F takes a value of 0 or 1. As this is the only system in the country, F is assigned 1. Based on the human occupancy, P takes the values of 2, 1 and 0, corresponding to the continuous, occasional and rare human presence, respectively, in the protection zone. As human presence is continuous in the Indian coastline, the value of P is assigned as 2. The demand rate parameter (W ) is computed as 9 based on the demand on the system during events, and is shown in Table 3. The value is assigned based on the historical data from the Indian Meteorological Department, where the system is demanded for operation during all seismic events. 107

4 Sundar et al. Development and performance assessment of a hybrid telemetry system for Indian tsunami buoy system Table 3: PFD and SIL levels as per IEC standards Demand rate Factor (W) W9 Often > 1/year 9 W8 Frequent 1/1 3 year 8 W7 Likely 1/ 3 10 year 7 W6 Probable 1/10 30 year 6 W5 Occasional 1/ year 5 W4 Remote 1/ year 4 W3 Improbable 1/ year 3 Table 4 shows the values assigned to the risk consequence parameters, which are catastrophic in all the defined three aspects. Having computed the values of P, F and W, the summed up values, i.e.12 ( ), are tabulated against the consequence factor in the risk graph matrix shown in Table 5. Thus, it is identified that the telemetry system used in a tsunami buoy should have an SR of better than the stringent SIL4. 4. Implementation of hybrid telemetry in tsunami buoy For carrying out effective water level variation detection and tsunami modelling, buoys in the ITBS are programmed to transmit water level data every 5min during the tsunami event mode, and up to 3hrs in an extended event mode (Arul Muthiah et al., 2011; Sundar et al., 2013). Data reception performance and evaluation tests were conducted for the first time in India on five offshore located meteorological MSB with INSAT telemetry. From these tests, the INSAT based telemetry link with a transmission frequency of every 3hrs has a data return mean time to failure (MTTF) of 151hrs and conforms to the SR of SIL1 (O Connor and Kleyner, 2012). The observed MTBF could be due to the inherent constraints with TDMA technology, absence of acknowledgment between the MSB transmitter and the satellite like Inmarsat, and data loss in other telemetry channels. With the identified level of MTTF and SR data, monitoring the water level changes during a tsunami event in ITBS MSB is not viable, as the system should meet SIL4. Thus, utilising the high reliability of Inmarsat telemetry and the identified SIL1 SR of the INSAT telemetry link, a hybrid telemetry system that is be reliable and cost-effective is demonstrated in the tsunami buoy ITB03 from August 2013 to March Implementation in ITB03 Considering the demanding SIL4 SR and economic requirements, a hybrid telemetry architecture using INSAT and Inmarsat telemetry has been implemented in ITB03. The hybrid system is configured to transmit the acquired water level data to the NIOT-MCC on an hourly basis in the normal mode through INSAT, and during a tsunami event through Inmarsat telemetry. The architecture of the same is shown in Fig 2. The communication architecture is configured with INSAT telemetry as the primary link and Inmarsat telemetry to be on hot redundancy waiting for the event. By means of this hybrid telemetry configuration, the overall SR of the system will be in SIL4 during an event mode, while during the normal mode the acquired data will be transmitted through the cost-effective INSAT telemetry link. The architecture of the same is shown in Fig 3. Table 4: PFD and SIL levels per IEC standards Personnel Environment health Financial Consequence* Catastrophic Catastrophic Catastrophic *Catastrophic, Extensive, Serious Table 5: Risk graph assignment matrix for SIL requirements Consequence F + P + W Severity level C 1,2 3,4 5,6 7,8 9,10 11,12 Catastrophic F NR SIL1 SIL2 SIL3 SIL4 > SIL4 Extensive E NR NR SIL1 SIL2 SIL3 SIL4 Serious D NR NR NR SIL1 SIL2 SIL3 Considerable C NR NR NR NR SIL1 SIL2 Marginal B NR NR NR NR NR SIL1 Negligible A NR NR NR NR NR NR NR No protective system required; SIL Safety Integrity level 108

5 Underwater Technology Vol. 33, No. 2, ,000 FIT INSAT modem Type 1: Normal mode Surface moored buoy system INMARSAT modem Type 2&3 event mode every 5GMT hour data Moored surface buoy INMARSAT-LES- NIOT through public network INSAT-INCOIS-VSAT link to NIOT 10,000 FIT 20,000 FIT 5,000,000 FIT 22,000 FIT Primary link Secondary link NIOT-MCC reception centre 110 FIT Normal mode Water level detection system Event mode bottom pressure recorder 400 FIT Fig 2: Reporting architecture in the hybrid telemetry system MSB IOR INMARSAT BPR VSAT INCOIS Land earth station and public network VSAT MCC-OOS,NIOT Fig 3: Configuration with INSAT and Inmarsat telemetry INMARSAT receiver As Inmarsat telemetry will be used as a safety telemetry system, maintaining the SR of the Inmarsat telemetry by implementing the PTI needs to be identified from applicable standards based on the failure-in-time (FIT) data (Ebeling, 2004). 5. Identification of proof test interval for safety reliability management The block diagram indicating the FIT of the subsystems in the implemented hybrid telemetry configuration is shown in Fig 4. Fig 4: Block diagram indicating FIT of the systems involved The failure rates for the subsystems involved in the MSB-INMARSAT-N_MCC link (Inmarsat telemetry link), and the MSB-INSAT-INCOIS-N_MCC link (INSAT telemetry link) are based on the published data on ITBS and INSAT telemetry, respectively. The values are shown in Table 6. The Inmarsat telemetry link, which is the redundant safety telemetry system, involves subsystems A, B, C, D, F and G (Table 6). Subsystems A, B and G are located in the offshore node, while C, D and F are located onshore, having provision for intervention and maintenance. Subsystems A and B are common for both the telemetry links, whose healthiness is confirmed through the hourly water level information data transmission done by using the INSAT telemetry link. Based on the FIT shown in Table 6, the PTI requirements for the MSB located Inmarsat transmitter (G) are computed using the Graphical Interface for Reliability Forecasting (GRIF) SIL tool (Total, 2013) and shown by the dashed line in Fig 5, where a PTI of 26.5hrs is required to maintain the SR of the MSB transmitter in SIL4. In order to maintain the SR, the Inmarsat transmitter is switched on every hour to get the MSB position transmitted through INSAT hourly, along with the water level data. Even though a PTI of 26.5hrs is enough for the transmitter, taking into Table 6: Subsystem FIT data for PTI determination Subsystem FIT in failures/billion hours References Bottom Pressure recorder (A) 400 Venkatesan et al., 2015a MSB (B) 20,000 Venkatesan et al., 2015a INMARSAT transmitter Unit (G) 23,500 Venkatesan et al., 2015a INMARSAT-LES-Public network with 12,600,000 Venkatesan et al., 2015a primary link to N_MCC (C) INMARSAT-LES-Public network with 220,000 Venkatesan et al., 2015a secondary link to N_MCC (D) MSB-INSAT-INCOIS-VSAT-N_MCC link (E) x 10 7 Sundar et al., 2013 N_MCC (F) 110 Venkatesan et al., 2015a 109

6 Sundar et al. Development and performance assessment of a hybrid telemetry system for Indian tsunami buoy system Probability of failure on demand (PFD) 1.2E-4 1.1E-4 1E-4 9E-5 8E-5 7E-5 6E-5 5E-5 4E-5 3E-5 2E-5 1E-5 SIL3 SIL Hour(s) HMI for INMARSAT transmission HMI for shore fiber link Fig 5: PTI computations for SIL maintenance consideration the time involved in restoring a faulty system (mean time to repair (MTTR) of about 7 days), a PTI of 1hr is implemented. This ensures that the SR of the Inmarsat transmitter unit is much better than SIL4. Fig 6 shows the hourly PTI data of the Inmarsat transmitter received at the NIOT-MCC. The right panel of the figure shows the data reception status. Green indicates the data received and red indicates data yet to be received from a buoy. This action is automatically updated by the ADDRESS software. Subsystems C, D and F are located onshore and have provisions for continuous monitoring and early restoration support during a failure. The solid line in Fig 5 shows the SR requirements computed using GRIF SIL tool for the data links C and D, which need a PTI of about 6.2hrs to have the PFD within SIL4. Fig 6: Inmarsat status transmitted to N_MCC hourly in ADDRESS 110

7 Underwater Technology Vol. 33, No. 2, 2015 Table 7: Required and implemented HMI for the hybrid telemetry subsystems Subsystem HMI for SR of SIL4 Computed Implemented MSB Transmitter 26.5 hrs 1 hr Primary and secondary 6.2 hrs 5 min links with N_MCC Overall link - 24 hrs In order to comply with the SR requirement, a watch dog program is implemented in the NIOT- N_MCC (F) and continuously monitors the integrity of the primary data link C and secondary data link D every 5min. It then reports the status to the N_MCC ADDRESS stations for advancing early restoration during an outage. During the C and D link outages, the system has to be restored back to operation with a period of 6.2hrs to avoid degradation of SR from SIL4, which sets the target for the MTTR for links C and D. Based on the C and D link failure records in the OOS failure reporting, analysis and corrective action system (FRACAS) register (Venkatesan et al., 2015a) for the years 2013 and 2014, the achieved MTTR was less than 6.2hrs, which was within the identified MTTR target. Further, to the implemented PTI for the MSB Inmarsat transmitter and shore based systems, an overall end-to-end SR test of the Inmarsat link is done every day at 5.00 GMT, by transmitting data through the Inmarsat. Table 7 shows the computed and implemented PTI for the Inmarsat telemetry and subsystems. 6. Real time performance of hybrid telemetry configuration The hybrid telemetry architecture was implemented, where the tsunami buoy ITB03 was deployed at location N and E in the Bay of Bengal on 18 Aug Fig 7 shows the MSB and the BPR deployed from the NIOT vessel Sagar Nidhi. Fig 8: Map indicating the location of the TB03 and the epicentre of the seismic event The deployed tsunami buoy with the hybrid telemetry was successful in detecting a water level change during a seismic event (see Fig 8), which had its epicentre at location N and E. During the event, MSB data transmission was switched from the INSAT to Inmarsat telemetry mode, and the data were received at the NIOT-MCC. The water level changes recorded by TB03 bottom pressure recorder (BPR) are shown in Fig 9, where a water level variation of 7.5cm was observed. The recorded water level variation was in close compliance with the sea water level changes recorded by station 23227, located N and E (maintained by INCOIS, India) during the same period (according to the National Data Buoy Center), which can be seen from Fig 10. The concept of hybrid telemetry involving INSAT and Inmarsat is demonstrated in the ITB03 tsunami buoy, which was able to perform the required on demand functionality. Also, by using the hybrid Water column height at station ITB03 (6N/88E) Image credit: NIOT/OOS metres /03 00 GMT 21/03 06 GMT 21/03 12 GMT 21/03 18 GMT 22/03 00 GMT 15 min 1 min 5 min Fig 7: BPR and MSB of the tsunami buoy ITB03 being deployed Fig 9: ADDRESS plot showing the water level changes in ITB03 111

8 Sundar et al. Development and performance assessment of a hybrid telemetry system for Indian tsunami buoy system metres /21 00 GMT telemetry, significant reduction (>96%) is obtained in the air time charges compared to the configuration using Inmarsat telemetry alone. 7. Summary and conclusion The hybrid telemetry system with INSAT as the primary link and Inmarsat as a redundant safety link is demonstrated in the Indian tsunami buoy. Based on the system reliability data and the identified SR SIL4 needs for the Inmarsat telemetry, a PTI of 1hr is implemented for the moored surface buoy transmitter, whose health status is transmitted along with the hourly INSAT transmitted water level data. A 6.2hr target mean restoration time for the primary and secondary data links with the NIOT-MCC is also identified. The successful response of the cost-effective telemetry system to a seismic event in 2014 gives confidence on the determination that the hybrid telemetry system is a reliable support to the ITBS. Acknowledgements We thank the Ministry of Earth Sciences, Government of India, for funding this project and the members of the National Expert Committee for evolving this program. We are grateful to Dr Jeyamani, Adviser, and the staff of INCOIS Hyderabad, for the establishment of the INSAT reception setup at INCOIS. We are also grateful to the directors of National Centre for Antarctic and Ocean Research (NCAOR), Goa and INCOIS Hyderabad for providing all the facilities and logistic support. We also thank the staff of the OOS group, Vessel Management Cell of the NIOT and the ship staff, for their excellent help and support on-board. References Water column height at station Image credit: NOAA/NWS/NDBC 03/21 06 GMT 03/21 12 GMT 03/21 18 GMT 15 min 1 min 15 sec Fig 10: Plot showing the water level fluctuation in INCOIS station /22 00 GMT Arul Muthiah M, Sundar R, Thamizh Mugilan A, Venkatesan R. (2011). Analysis on under water seismic event on June 12, 2010 recorded by indigenous tsunami early warning system. In: Proceedings of 2011 Conference of the Ocean Society of India, OSICON 2011, June, Chennai, India, 1 7. Braddock RD and Carmody O. (2001). Optimal location of deep-sea tsunami detectors. International Transactions in Operations Research 8: Ebeling CE. (2004). An introduction to reliability and maintainability engineering. New Delhi, India: Tata McGraw-Hill. 486pp. Graff T, Sprenke J and Bushnell M. (2006). Ocean systems test and evaluation program. Data communications plan report. National Oceanic and Atmospheric Administration (NOAA) Technical Report NOS CO-OPS 34., 189pp. Available at: Data_Communications_Plan_52506.pdf, last accessed <29 August 2015>. International Electrotechnical Commission (IEC). (2000). IEC Functional safety of electrical/electronic/ programmable electronic safety-related systems. Geneva: International Electrotechnical Commission, 12pp. Leopold RJ and Miller A. (1993). The Iridium communications system. Institute of Electrical and Electronics Engineers Potentials 12: 6 9. O Connor P and Kleyner A. (2012). Practical Reliability Engineering. West Sussex, UK: John Wiley & Sons, 512pp. Smith DJ and Simpson KGL. (2004). Functional Safety: A Straightforward Guide to Applying IEC and related standards, 2nd edition. New York: Elsevier Butterworth Heinemann, 280pp. Sundar R, ArulMuthiah M and Venkatesan R. (2013). Development of application software for real time tsunami buoy data reception. In: Proceedings of the National Symposium on Coastal Oceanographic Studies: Modelling and Observations (COSMOS), Kochi, India, 3 7. Total. (2013). Graphical Interface for Reliability Forecasting (GRIF) user manual version Available at: <last accessed 21 September 2015>. Vedachalam N. (2013). Review of challenges in reliable electric power delivery to remote deep water enhanced oil recovery systems. Journal of Applied Ocean Research 43: Vedachalam N, Ramadass GA and Atmanand MA. (2014a). Reliability centered modeling for development of deep water human occupied vehicles. Applied Ocean Research 46: Vedachalam N, Ramadass GA and Atmanand MA. (2014b). Review of technological advancements and HSE based safety model for deep-water human occupied vehicles. Marine Technological Society Journal 48: Vengatesan G, Muthiah MA, Upadhyay JS, Sundaravadivelu N, Sundar R and Venkatesan R. (2013). Real time, low power, high data rate and cost effective transmission scheme for coastal buoy system. In: International Symposium of Ocean Electronics (SYMPOL), Oct, Kochi, India: 1 6. Venkatesan R, Arul Muthiah M, Ramesh K, Ramasundaram S, Sundar R, and Atmanand MA. (2013a). Satellite communication systems for ocean observational platforms: societal importance and challenges. The Journal of Ocean Technology 8: Venkatesan R, Shamji VR, Latha G, Mathew S, Rao RR, Arul Muthiah M and Atmanand MA. (2013b). In situ ocean subsurface time-series measurements from OMNI buoy 112

9 Underwater Technology Vol. 33, No. 2, 2015 network in the Bay of Bengal. Current Science 104: Venkatesan R, Vedachalam N, Sundar R, Arul Muthiah M, and Atmanand MA. (2015a). Reliability assessment of Indian Tsunami buoy systems for societal importance. Underwater Technology 32: Venkatesan R, Amasundaram S, Sundar R, Lavanya R, Vedachalam N and Atmanand MA. (2015b). Reliability assessment of state-of-the-art real time data reception and analysis system for Indian seas. Marine Technology Society Journal 49: Wijetunge JJ. (2009). Field measurements and numerical simulations of the 2004 tsunami impact on the south coast of Sri Lanka. Ocean Engineering 36: Yoshimura I and Sato Y. (2008). Safety achieved by the safe failure fraction (SFF) in IEC IEEE Transactions on Reliability 57: Zacharia S, Seshasayanan R, Srinivasan R, Thamarai T, Sudhakar T, Rao RR and Atmanand MA. (2014). Design, development and validation of smart sensor drifting node with INSAT telemetry for oceanographic applications. Current Science 106: