ADVISORY ON

STRUCTURAL HEALTH
MONITORING: THE
APPLICATION OF
SENSOR-BASED
APPROACHES

© Stock Studio Ae rials/S hut e rstock

——
TABLE OF CONTENTS

INTRODUCTION ...........................................................................................................................................................................................................................................................................................1

SECTION 1 – STRUCTURAL HEALTH MONITORING PRINCIPLES ..............................................................................................................................................1

Structural Health Monitoring Plan Questionnaire.................................................................................................................................................................................1
Sensor Specification Terminology...............................................................................................................................................................................................................................3

SECTION 2 – SENSOR-BASED MONITORING PLANS ..............................................................................................................................................................................4

Sensor Packages for Typical Commercial Vessels.....................................................................................................................................................................................4
Sensor Selection Considerations....................................................................................................................................................................................................................................6
Consideration for Selecting Sensor Sub-Types.............................................................................................................................................................................................7
Electrical Resistance Strain Gauge........................................................................................................................................................................................................................7
Fiber Optical Strain Gauge..............................................................................................................................................................................................................................................8
Accelerometers..............................................................................................................................................................................................................................................................................8
Pressure Transducer...............................................................................................................................................................................................................................................................8

SECTION 3 – COUPLING SENSOR DATA WITH ANALYSIS AND ANALYTICS MODELS .............................................................................9

SECTION 4 – ROLE OF ABS ........................................................................................................................................................................................................................................................ 11

APPENDIX A – ABBREVIATIONS AND ACRONYMS .................................................................................................................................................................................12

APPENDIX B – COMMONLY USED SENSOR TYPES .................................................................................................................................................................................13

APPENDIX C – COMMENTARY ON ACCELEROMETER AND PRESSURE TRANSDUCERS ....................................................................15

APPENDIX D – COMMENTARY ON COUPLING SENSOR DATA WITH ANALYSIS AND ANALYTICS MODELS ...........17

APPENDIX E – REFERENCES ..................................................................................................................................................................................................................................................21

Disclaimer:
While ABS uses reasonable efforts to accurately describe and update the information in this Advisory, ABS makes no warranties or representations as to its accuracy, currency or
completeness. ABS assumes no liability or responsibility for any errors or omissions in the content of this Advisory. To the extent permitted by applicable law, everything in this Advisory
is provided “as is” without warranty of any kind, either expressed or implied, including, but not limited to, the implied warranties of merchantability, fitness for a particular purpose,
or noninfringement. In no event will ABS be liable for any damages whatsoever, including special, indirect, consequential or incidental damages or damages for loss of profits, revenue
or use, whether brought in contract or tort, arising out of or connected with this Advisory or the use or reliance upon any of the content or any information contained herein.
——
INTRODUCTION
Increasingly, modern marine vessels and offshore units are being equipped with various tools for structural
health monitoring, operational assistance, and maintenance optimization. These tools collect data through
sensors and onboard instrumentation and provide status of structural health for awareness of the crew and
owners and operational optimization or for carrying out repairs/modifications to prevent further deterioration
or future failures.

This document focuses on the collection of data using sensors for the purpose of structural health monitoring.
Specifically, it addresses the monitoring of structural loads and/or responses using sensors to infer current
structural health status, predict future health states, and inform maintenance activity.

The sensor data can be used either directly in the raw format or processed to represent physical parameters, such
as pressure, tensile and compressive stresses, bending moments and deformation. Trending and threshold-based
alarms, in addition to the parameter monitoring, are typically an integrated function of sensor-based structural
health monitoring.

This advisory provides guidance on sensor-based structural health monitoring implementation through four key
sections:
Section One: Structural Health Monitoring Principles
Section Two: Sensor-Based Monitoring Plans
Section Three: Coupling Sensor Data with Analytics and Analysis
Section Four: ABS Role

——
SECTION 1: STRUCTURAL HEALTH MONITORING
PRINCIPLES
Structural health monitoring has been widely used for decades in various industries, such as aerospace, civil and
mechanical infrastructure, as well as early adoption in both the marine and offshore industries. Several industry
standards and publications discuss the application of sensor-based structural health monitoring for marine and
offshore assets.

IMO MSC/Circ.646 (June 1994), Recommendations for the fitting of hull stress monitoring systems requires that
the hardware and software of a hull stress monitoring system be type approved by an Administration, which in
practice is usually achieved via a certification of compliance issued by a recognized class society. Class societies
have also published technical standards and requirements for the use of various hull monitoring systems.

The Section sets out the key principles that need to be addressed when considering implementing a sensor-based
structural health monitoring program.

STRUCTURAL HEALTH MONITORING PLAN QUESTIONNAIRE

A systematic approach and careful planning will assist the various stakeholders for such systems better
contextualize the monitoring purpose, set proper expectations, explore the available technologies, and ensure
smooth implementation, operation, and delivery. A structural health-monitoring plan should cover sensors,
data acquisition, data usage, data processing, analysis and analytics models. Below are questions to assist the
stakeholder team to collect information, align expectations, and enhance common understanding.

Questions – Set 1: Define Implementation Goals

• What is the main purpose of the proposed structural health monitoring plan?
• What are the companies’ current practices on structural inspection, maintenance and integrity management?
Will the monitoring plan help support these aspects?
• What are the main concerns regarding the structural conditions, damages, and failure modes for the specific
vessel(s) being monitored given the historical issues with the vessel or vessel class?
• What are the anticipated operation and operational environment?

ABS | STRUCTURAL HEALTH MONITORING: THE APPLICATION OF SENSOR-BASED APPROACHES | 1

Questions – Set 2: Evaluate Technology Readiness Level

• What are the measurable physical variables reflecting the structural loads, capacity, and responses relevant to
the anticipated structural failure modes?
• What are the physical variable characteristics of the measured aspect of the structure, such as statics or
dynamics, range, the smallest meaningful change that needs to be captured, etc.?
• Will the sensor data be used/integrated with any engineering analysis and analytics models (e.g., finite element
analysis, operational modal analysis, machine learning algorithms, etc.) to derive structural health conditions
and assist in detecting anomalies?
• Does the sensor data need to be transferred on shore in real-time or near real-time for monitoring and further
analysis?
• Is there a need to retrieve or correlate data from other onboard systems to assist monitoring, analysis and
models? What data needs to be retrieved, and at what frequency?
• What are the tradeoffs between the investment and the added value of the structural health monitoring plan?
Will a techno-economic evaluation of the individual structural health monitoring plan be conducted?

Questions – Set 3: Define Sensor Type and Specification

• What sensor types are suitable for measuring the identified physical variables? Are the physical variables
directly measurable?
• What are the sensor specifications required for measuring the physical variables (e.g., range, sensitivity,
accuracy, response time, linearity, sensor-self noise, waterproof and environmental suitability, hazardous area
suitability, power supply, etc.)?
• Where should these sensors be installed? Are these installation locations accessible and feasible for the sensor
installation, inspection, and maintenance?
• How is the data transmitted and stored on board and/or on shore (wired or wireless, network topology)?

Questions – Set 4: Define Data Acquisition Device Specification

• What is the accuracy requirement for signal digitization (e.g. analog digital converter) to accurately reflecting
the physical variables?
• What are the compatibility requirements for the data acquisition devices (e.g., driver, bus, interface,
protocol, etc.)?
• What are the specification requirements for the data acquisition devices (e.g., electrical circuit design, noise
due to various sources, system accuracy, conversion time, waterproof and environmental suitability, hazardous
area suitability, power supply, and robustness)?
• How is the data from various sensors, data acquisition devices, and other data sources synchronized?
What is the synchronization requirement?

Questions – Set 5: Explore Analysis and Analytics Models

• What are the data quality requirements for the data to be suitable for analysis and models?
• What condition anomalies, such as overloading, excessive deformation, stress, and fracture, can be detected
directly from the data?
• What additional analysis and models are needed to extract more features and insights that cannot be derived
directly from the sensor data?
• How accurate and reliable are the analysis and models?
 
Questions – Set 6: Define Outcomes and Deliverables

• Who are the end users? Are they physically located on board, on shore, or both?
• What is the expected deliverable format and methods (reporting, visualization, dashboard, etc.)?
• What are key insights and information required to be included in the deliverables?
• What is the expected decision-making frequency? Is real-time needed?
• Is there any training required for the end users to interpret the deliverables?

It may not be practical to answer all the above questions during the initial planning and design stage, and the
answers may be refined as the project progresses. However, asking such questions up front can help stakeholders
better organize and manage the project and avoid costly rework.

2 | STRUCTURAL HEALTH MONITORING: THE APPLICATION OF SENSOR-BASED APPROACHES | ABS

SENSOR SPECIFICATION TERMINOLOGY
Sensors measure physical variations through a time-varying analog or digital signal. It is crucial to understand
the terminology that defines the characteristics of sensors and measuring instruments. International standards
(such as ISO/IEC GUIDE 98 series, ISO/IEC Guide 99, IEC 60050-300, IEC 62008) introduce terminology for measuring
instruments in terms of the performance characteristics and the expression of the uncertainty in measurement.
The following definitions are commonly referenced in sensor specifications:

• Accuracy (of Measurement): a qualitative concept describing the level of agreement between the result of a
measurement and the true value of the measurand, or quantity intended to be measured.
• Accuracy Class: the category of measuring instruments, all of which are intended to comply with a set of
specifications regarding uncertainty. Accuracy class is usually denoted by a number or symbol adopted by
convention. For example, accuracy classes of a thermometer are defined by IEC 60751:2008 Industrial platinum
resistance thermometers and platinum temperature sensors as Class AA; Class A, Class B and Class C.
• Error (of Measurement): the result of a measurement minus a true value of the measurand. Note that a true
value cannot be determined, so a unique “true” value is simply an idealized concept, and in practice a Reference
Quantity Value is used. ISO/IEC Guide 99:2007 may be referenced for calculating errors.
• Uncertainty (of Measurement): a non-negative quantitative parameter characterizing the dispersion of the
values attributed to a measurand, based on the information used. Two types of methods to evaluate uncertainty
are used: Type A evaluation by the statistical analysis of a series of observations, and Type B evaluation by
means other than the statistical analysis of a series of observations (for example, obtained from a calibration
certificate and/or the accuracy class of a verified measuring instrument).
• Relative Error: the ratio of the absolute error to a comparison value. This term can be seen in some product
specifications, conformity reports and verification certificates. In general, this “true value” approach concept is
used with a reference quantity value and/or a conventional true value together.
• Maximum Permissible Error (MPE): this represents extreme value of measurement error, with respect to
a known reference quantity value, permitted by specifications or regulations for a given measurement,
measuring instrument, or measuring system. MPE is usually given by the manufacturer in product
specification. For example, MPE = 0.05 mm for a caliper gauge specified based on absolute value and MPE = 0.1%
for a torque measurement based on relative error.
• (Measurement) Repeatability: the level of agreement among successive measurements of the same measurand,
carried out under the same conditions of measurement.
• (Measurement) Reproducibility: the level of agreement among measurements of the same value of a quantity,
when the individual measurements are made under different conditions of measurement.
• (Measurement) Precision: the level of agreement between indications or measured quantity values obtained
by replicate measurements on the same or similar objects under specified conditions. Precision is usually
expressed numerically by measures of imprecision, such as standard deviation, variance, or coefficient
of variation under the specified conditions of measurement. Precision is used to define repeatability and
reproducibility.
• Stability (of a Measuring Instrument): ability to keep its performance characteristics unchanged over time
when all other conditions remain the same. Stability may be quantified in several ways. For example, short-term
and long-term drifts can be used to quantify stability.
• Measurement Range (also called Measuring Range): a range defined by two values of the measurand, or
quantity to be supplied, within which the limits of uncertainty of the measuring instrument are specified.
• Sensitivity (of a Measuring System): the quotient of the change in an indication of a measuring system and the
corresponding change in a value of a quantity being measured. For example, gauge factor is the synonym of the
strain gauge sensitivity.
• Resolution: the smallest change in a quantity being measured that can be perceived.
• Discrimination Threshold: the largest change in a value of a quantity being measured that causes no detectable
change in the corresponding indication.
• Step Response Time: the duration between the instant when an input quantity value of a measuring
instrument or measuring system is subjected to an abrupt change between two specified constant quantity
values, and the instant when a corresponding indication settles within specified limits around its final steady
value.
• Sampling Rate: the number of analog-to-digital conversions per unit time; it is usually expressed in samples per
second.
• Drift: the change in the indication of a measuring instrument, generally slow, continuous, neither necessarily in
the same direction and nor related to a change in the measurand.

ABS | STRUCTURAL HEALTH MONITORING: THE APPLICATION OF SENSOR-BASED APPROACHES | 3

• Zero Offset: the magnitude of the output signal observed from a sensor or a measuring instrument, when the
input signal of the measurand is zero under a specified condition (typically in room-temperature condition).
• Frequency Bandwidth: the measure of a measuring instrument’s ability to pass an analog signal without
significant attenuation over a range of frequencies. For example, bandwidth is normally expressed in Hertz (Hz)
defined by -3 dB cut-off points of the frequency response function (FRF) of signal amplitude (i.e., the lower and
upper frequency points where the signal amplitude falls to −3 dB below the passband frequency). Bandwidth
is an important dynamic performance characteristic of a measuring chain in order to capture the correct
quantity value of a time-varying measurand.

——
SECTION 2: SENSOR-BASED MONITORING PLANS
Sensor packages and specifications should be developed according to the implementation goals, technology
readiness, and availability considering budget limitations. For a comprehensive sensor-based structural health
monitoring plan, the sensor data can be used to:
• Directly monitor the loads and structural responses, and alarm when overloading and other excessive measured
parameters or events are detected.
• Serve as a data source and input into the coupled analysis and models to derive structural health condition,
predict degradation and assist operational and asset integrity management decisions.
• Serve as high-fidelity data to validate and calibrate the analysis and analytics models in terms of model
parameters, analysis assumptions, and outcomes.

SENSOR PACKAGES FOR TYPICAL COMMERCIAL VESSELS

For structural health monitoring, the typically measured physical variables are those relevant to the dominant
loads and main structural failure modes, which vary for different vessel types due to their unique structural
arrangements, operational modes and environment. Accordingly, sensor package selection is typically vessel type
dependent. Table 1 summarizes some recommended sensor packages based on the measured physical variables
and vessel type for common commercial ship types. Furthermore, the redundancy for selected sensor packages as
discussed in Table 1 shall be assessed carefully in line with the intended purpose and data quality of the structural
health monitoring system.

These sensors and their corresponding measured physical variables in Table 1 can also be applied to non-ship
shape structures (e.g., Column Stabilized Unit, Self-Elevating Unit, Tension Leg Platform, Spar). For example, sensors
for position, wave and wind are valid for both ship and non-ship structures. However, the direct measurand and
its required instrumentation plan typically need to be developed on a case-by-case basis for a non-ship shape
unit due to the diversity of its structural arrangement and configuration. Appendix B provides guidance on
the selection of sensor types based on the common physical variables to be measured on marine and offshore
structures.

Hull Stress Monitoring

System Components
Signal Conditioning Unit

Display Unit

UPS

Zener Barrier Unit

Long Baseline Strain Gauges

Accelerometers

Bow Pressure Transducer

4 | STRUCTURAL HEALTH MONITORING: THE APPLICATION OF SENSOR-BASED APPROACHES | ABS

 Table 1: Recommended Structural Sensor Packages for Common Commercial Vessel Types

List of Sensor Abbreviations

ACC: Accelerometer
GNSS: Global Navigation Satellite System (e.g. GPS, GLONASS, Galileo, Beidou and other regional systems)
LBSG: Long based strain gauge
MRU: Motion Reference Unit
PT: Pressure transducer
SG: Strain gauge which can be either electrical-resistance or fiber optic type
TEMP: Temperature sensor for the structure temperature monitoring
WR: Wave radar
WS: Wind sensor for wind state monitoring, such as anemometer or automated weather station

n Optional/Recommended n Typically Required

General High
Container Bulk Ore Ro–Ro LNG Chemical Passenger
Direct Measurand Oil Tanker Cargo Speed
Carrier Carrier Carrier Ship Carrier Carrier Ship
Ship Craft

Vertical accelerations
ACC ACC ACC ACC ACC ACC ACC ACC ACC
at bow

Vertical, transverse
and longitudinal ACC
acceleration at bow

Transverse acceleration
ACC ACC ACC ACC ACC ACC
amidships

Vertical, transverse
and longitudinal
acceleration at ACC
longitudinal center of
gravity

Vertical, transverse
and longitudinal ACC
acceleration at stern

Ship motion (at center

MRU MRU MRU MRU MRU MRU MRU MRU MRU
of gravity)

Global longitudinal
stress amidships (port LBSG LBSG LBSG LBSG LBSG LBSG LBSG LBSG LBSG LBSG
and starboard side)

Global longitudinal
stress at quarter length
LBSG LBSG LBSG LBSG LBSG LBSG LBSG LBSG LBSG
fore and aft of midship
(port or starboard side)

Local transverse stress

at transverse deck strip SG
amidships

Global longitudinal
stress below neutral axis
LBSG LBSG LBSG
amidships (port and
starboard)

Double bottom bending

SG SG SG
stress

Bending/shear stress in
SG
pillar bulkheads

ABS | STRUCTURAL HEALTH MONITORING: THE APPLICATION OF SENSOR-BASED APPROACHES | 5

 General High
Container Bulk Ore Ro–Ro LNG Chemical Passenger
Direct Measurand Oil Tanker Cargo Speed
Carrier Carrier Carrier Ship Carrier Carrier Ship
Ship Craft

Global transverse stress

in wet deck between SG
each catamaran hull

Lateral loads at
bowflare or bottom near
PT/SG PT/SG PT/SG PT/SG PT/SG PT/SG PT/SG PT/SG PT/SG PT/SG
forward perpendicular
(slamming pressure)

Lateral loads at side

PT/SG PT/SG PT/SG PT/SG PT/SG PT/SG PT/SG PT/SG PT/SG
shell (wave pressure)

Lateral loads at the bow

PT/SG
door

Sloshing response of
liquid in tanks (sloshing PT/SG PT/SG PT/SG
pressure)

Structural temperature TEMP TEMP TEMP

Position, speed and

GNSS GNSS GNSS GNSS GNSS GNSS GNSS GNSS GNSS GNSS
course

Wave condition/sea
WR WR WR WR WR WR WR WR WR WR
state

Wind condition WS WS WS WS WS WS WS WS WS WS

Some of the recommended sensor packages in Table 1 have been adopted in Hull Condition Monitoring Systems
on vessels. If they meet requirements of ABS Guide for Hull Condition Monitoring Systems, the vessels fitted
with such sensors may be eligible for three optional class notations (i.e., HM1 – Motion Monitoring, HM2 – Stress
Monitoring, and HM3 – Voyage Data Monitoring). For example, a vessel fitted with one accelerometer (ACC) at
the bow, one MRU, one wave radar, six LBSGs, ten strain gauges (SGs) at ten selected local critical structural areas,
one GNSS device and one wind sensor (WS) in Table 1 may consider class notation HM1 (Slam Warning: ACS1,
Ship Motion: MOT1, Sea State: ST1), HM2 (Hull Girder Stress: HS6, Local Load Monitoring: LS10), HM3 (Navigation,
Wind: WD1).

SENSOR SELECTION CONSIDERATIONS

The selection of sensors and/or measuring instruments is governed by the measurand, such as the measuring
range and the bandwidth (also referred as frequency range) of the measurand. The following are common
considerations for sensor selection:
• Environmental condition, installation location of the sensor, and protective treatment (e.g., hazardous area,
ambient temperature, pressure, humidity, noise, corrosive acid, abrasive action, electromagnetic, neutron, and
radiation fields, etc.).
• Space limitations that may constrain the placement and location of the sensors.
• Adequate accuracy, range, and bandwidth for the measured physical variables.
• Sensor bandwidth, which as a rule of thumb is preferably three (3) times of the measurand for slow or moderate
periodic vibration and ten (10) times for impulse, shock, or sudden change measurements.
• Rise time capability, which is related to the measuring instrument’s bandwidth and is an important
specification parameter for transient measurement. As a rule of thumb, the system rise time is preferably three
(3) times faster than the measurand’s rise time.
• Wire/cable noise and resistance, and potential electromagnetic interference. Cables/wires should be installed,
secured and protected properly.
• Amplification that may be applied to increase measurement resolution and improve signal-to-noise ratio.
• Signal filtering that may be applied to remove external, high-frequency noise. The anti-aliasing filter must be
placed before the analog-to-digital converter

6 | STRUCTURAL HEALTH MONITORING: THE APPLICATION OF SENSOR-BASED APPROACHES | ABS

• The ground-loop effect that can be through proper design of measuring chain. A single-ended measurement
device is susceptible to ground loop and differential measurement device rejects ground-loop with common-
mode voltage rejection.
• Sensor, instrument, and cable/wire maintenance and re-calibration requirements.
• Life and performance degradation of the sensors and the installation method.
• Sensor location, installation, and wiring should satisfy relevant class rules, such as for water-tightness, cable
penetration, trip hazards, electromagnetic interference, et al.
• Installation location and method should not have significant impact on the asset’s normal operations, such as
cargo loading and unloading.
 
CONSIDERATION FOR SELECTING SENSOR SUB-TYPES
For certain sensor types such as strain gauges and accelerometers there are various sub-types that are more
suitable for certain applications than others due to their different working principles. The sub-types of commonly
used structural sensors are summarized below, and Appendix D provides information on commonly available
accelerometer and pressure transducers for reference.

ELECTRICAL RESISTANCE STRAIN GAUGE

Electrical resistance strain gauges convert the deformation into an electrical signal. The following should be taken
into consideration for electrical resistance strain gauge selection:
• Static performance characteristics (resistance, gauge factor, transverse sensitivity, temperature coefficient of
gauge factor, and thermal output).
• Long term creep and drift
• Specifications for strain measurement:
– Frequency range/bandwidth of an electrical resistance strain gauge is typically adequate for measuring
structural vibration in marine and offshore applications (such as slamming, sloshing, whipping, springing,
and ice load). The bandwidth of the strain measurement is typically determined by the strain gauge
instrument system rather than the strain gauges.
– The bandwidth of strain gauge instrument system can be generally categorized as:
• Static (slow varying): 0~200 Hz;
• Dynamic: 0~10 kHz;
• Super-dynamic: >200 kHz.
For ship and offshore vibration monitoring, a dynamic strain gauge is preferred.
• Signal conditioning for strain gauges:
– The main factors affecting strain gauge performance include bridge configuration, signal conditioning,
wiring, and data acquisition device.
– The resistance tolerance and strain induced by installation may generate initial offset voltage when no strain
is applied. Offset nulling should be applied to balance the bridge so that the output voltage is zero when no
strain is applied, and the bridge should be calibrated to verify the output to a known, expected value.
– Long lead wires can add resistance to the bridge, which adds an offset error and desensitizes the output
of the bridge.
– Three Wheatstone bridge configurations are commonly used:
• Quarter-bridge circuit is the simplest circuit and typically requires dedicated temperature
compensation.
• Half-bridge circuit is two times more sensitive than a quarter-bridge circuit and can cancel out the
temperature and uniform axial strain effect.
• Full-bridge circuit can cancel out the temperature and Poisson effect and generally offers the highest
sensitivity.
• Special consideration for measuring impulse/shock response:
– The gauge factors for static loads are valid for measuring strains caused by impulse/shock loads.
– The strain gauge installation method may affect the gauge performance for impulse and shock induced
strain measurement.
– The gauge length selection should be suitable for application.

ABS | STRUCTURAL HEALTH MONITORING: THE APPLICATION OF SENSOR-BASED APPROACHES | 7

FIBER OPTICAL STRAIN GAUGE
For a Fiber Bragg Gratings (FBGs) strain gauge, the gauge length is the length over which the applied strain is
averaged, converted and measured. This gauge length is usually not the same as the fiber Bragg grating length.
Compared with the electrical resistance strain gauge, the FBG strain gauge has the following characteristics:
• Requires a light source instead of electrical excitation.
• One fiberglass cable can connect several FBG gauges, reducing wiring.
• Is inherently immune to electromagnetic interference.
• Is intrinsically safe for applications in explosive atmospheres.
• Is made of sensor material that is resistant to mechanical failure at high-level vibration loads.
• Has lighter connection leads, as fiberglass is substantially thinner than copper conductors.
• Requires temperature compensation. A dummy FBG gauge is typically used for thermal compensation.
• Requires special static-dynamic FBG measuring instruments (optical interrogators) for vibration measurement.
• Can be paired with fiber optic sensors, which are good for temperature measurement. This allows the
measurement of the structure temperature and strain at the same time.

IEC 61757-1-1 provides more detailed requirements of strain sensors based on fiber Bragg gratings. There are many
FBG strain gauges and interrogators available in the market. Optical interrogators are typically more expensive
than the electrical ones, which may lead to the FBG solution being more costly than electrical.

ACCELEROMETERS
Various types of accelerometers are based on different physical principles and are suitable for different measuring
purposes. There are 2 types of accelerometers:
• AC-response accelerometer, which can only sense changes in acceleration (but not constant acceleration).
• DC-response accelerometer, which can sense both changes in acceleration, and constant acceleration.

Force-balanced accelerometers and servo-type accelerometers can measure DC components (i.e., constant
acceleration). Force-balanced accelerometers can be customized for low frequency and high sensitivity for high
precision measurements.

The target measurand and the potential usage of the measurement should be considered when choosing the
accelerometers to:
• Determine whether the DC-response (constant) is required for measurements
• Confirm the sensitivity, cross-axis sensitivity, and sensitivity temperature coefficient
• Confirm the measuring range
• Confirm the frequency response function / bandwidth
• Confirm the suitable temperature range

PRESSURE TRANSDUCER
Temperature and process medium of the measured fluid in contact with the sensing element must be considered
carefully when selecting pressure transducers. As pressure transducers must contact the process medium directly,
the mounting to the watertight members should be carefully planned and implemented. Watertight testing
is recommended after the installation. Alternatively, the pressure on watertight members can be measured
indirectly via strain gauges attached to the backing structures, which is a common practice for ice loading
monitoring.

8 | STRUCTURAL HEALTH MONITORING: THE APPLICATION OF SENSOR-BASED APPROACHES | ABS

——
SECTION 3: COUPLING SENSOR DATA WITH ANALYSIS
AND ANALYTICS MODELS
Engineering analyses (such as Finite Element Analysis) have been widely used in the design assessment of ships
and offshore structures, and there are many recognized standards and publications providing references and
requirements on such analyses (such as the ABS Guide for SafeHull-Dynamic Loading Approach for Vessels and
the ABS Guide for Dynamic Loading Approach for Floating Production, Storage and Offloading Installations). These
design analysis approaches and procedures provide insights into the vessel dominant loads and the corresponding
structural responses and failure modes, which can be used to guide the structural sensor package selection.
Table 2 summarizes the Dominant Loading Parameters (DLPs) recommended for design structural assessment of
common commercial ship types.

Table 2: Dominant Load Parameters (DLPs) vs. Vessel Type

Vessel Type Container

Tankers Bulk Carrier LNG Carriers FPSO
DLP Carriers

Vertical Bending Moment (VBM) X X X X X

Vertical Shear Force (VSF) X X X X

Torsion Bending Moment (TM) X X

Horizontal Bending Moment

X X
(HBM)

Horizontal Shear Force (HSF) X

Vertical Acceleration (VACC) X X X X X

Lateral Acceleration (LACC) X X X

Roll (ROLL) X X X X X

The accuracy and reliability of the analysis and analytics models reflect the uncertainties and assumptions of
the model inputs (e.g. loads, environments, structural gauging, etc.) as well as the analysis and model themselves
(e.g. model parameters, assumptions, approximation, etc.). Sensor-based full-scale measurements are typically
treated as high-fidelity data, which not only can be used for direct load and response monitoring and parameter
trending, but more importantly can be integrated with analysis and models through analysis/model calibration
and validation to enhance the analysis accuracy and reliability. The integration of sensor data with analysis and
models provides more accurate and reliable insights on the structural health and predicts potential damages for
both global hull and local critical areas. Table 3 lists recommendations on sensor packages for the calibration and
validation of global loads, analysis and model parameters. Furthermore, Appendix D provides considerations and
required effects on sensor data processing corresponding to specific application scenarios.

ABS | STRUCTURAL HEALTH MONITORING: THE APPLICATION OF SENSOR-BASED APPROACHES | 9

 Table 3: Sensor Package for Analysis and Model Calibration

Description of Symbol and Legend

XXX: Most Preferred n Basic level of data processing is required as defined in Table 9
XX: Moderately Preferred n Medium level of data processing is required as defined in Table 9
X: Less Preferred n Advanced level of data processing is required as defined in Table 9

Data Source Motion

Long Based Pressure
Reference Strain Gauge Accelerometer
Strain Gauge Transducer
Unit (SG) (ACC)
Quantity (LBSG) (PT)
(MRU)
of Interest

Vertical accelerations at bow XX XX X

Vertical Shear Force (VSF) XX XX X

Torsion Bending Moment (TM) XX XX X

Horizontal Bending Moment

XX XX X
Dominant Load (HBM)
Parameter
Horizontal Shear Force (HSF) XX XX X

Vertical Acceleration (VACC) XXX XX

Lateral Acceleration (LACC) XXX XX

Roll (ROLL) XXX XX

Ice Pressure XXX

Loading Sloshing Pressure XXX XX

Slamming Pressure XXX XX

Natural Frequency XX XXX XXX XXX

Modal Damping Ratio XX XXX XXX XXX

Finite Element
Model
Displacement Mode Shape XXX XXX

Strain Mode Shape XXX XXX

Stress at Sensor Location XXX XXX

Stress
Stress at Any Location XXX XXX XXX

Still Water Bending Moment XX XX X

Miscellaneous Vibration Dose Value XXX XXX

Motion Sickness Dose Value XXX XXX

Note: Arrangement of sensors refers to both number and mounting position of sensors.

10 | STRUCTURAL HEALTH MONITORING: THE APPLICATION OF SENSOR-BASED APPROACHES | ABS

——
SECTION 4: ABS ROLE
As a classification society, ABS has established technical requirements published in Rules and Guides for vessel
design, construction and survey. ABS conducts independent verification and validation through engineering
review and onboard/onsite survey to promote the asset’s safe operation. Sensor-based structural health monitoring
requires the installation of hardware and software on board the asset and may also involve integration and
interconnection with existing onboard systems that may impact the vessel’s safe and normal operation. In
addition, the monitoring system provides structural health awareness and condition anomalies, which can
potentially impact ABS inspection and survey decisions. Therefore, the safety, quality, accuracy and reliability of
the sensor-based structural health monitoring are of significant interest to classification societies.

ABS has offered optional class notations to recognize that the structural health monitoring system is in
compliance with ABS standards and requirements. The ABS Guide for Hull Condition Monitoring Systems
presents the need for fitting sensor-based hull condition monitoring systems and lists safety and performance
specifications for various types of systems, such as slamming warning, hull girder stress monitoring, and green
seas warning. Optional class notations HM1, HM2, or HM3 can be awarded for monitoring systems that satisfy
the Guide’s requirements. With the advent of data analytics capabilities and the coupling of operational data
and sensor measurements with analytics and analytics models, structural health monitoring is evolving into the
concept of the structural digital twin. The ABS Guide for Smart Functions for Marine Vessels and Offshore Units
offers the optional class notation SMART (SHM) to vessels with structural health monitoring capabilities based
on the coupling of sensor-based monitoring with analysis and analytics models for structural health condition
diagnostics and prognostics.

For sensor-based structural monitoring, ABS can provide not only the requirements and independent verification
and validation on system capability, safety and integrity, but also give functional and implementation
recommendations to assist stakeholders to obtain more value out of the implementation. With the proven
accuracy and reliability of the structural monitoring system, informed, targeted, condition and risk-based
alternative means of crediting class survey requirements can be implemented to enhance the asset safety with
less intrusiveness and reduced cost.

ABS | STRUCTURAL HEALTH MONITORING: THE APPLICATION OF SENSOR-BASED APPROACHES | 11

——
APPENDIX A: ABBREVIATIONS AND ACRONYMS
AC: Alternating Current
ASTM: American Society for Testing and Materials
DC: Direct Current
DGPS: Differential Global Positioning System
DLP: Dominant Load Parameters
DSP: Digital Signal Processing
EMA: Experimental Modal Analysis
FBG: Fiber Bragg Grating
FRF: Frequency Response Function
GPS: Global Positioning System
GNSS: Global Navigation Satellite System
IEC: International Electrotechnical Commission
IEEE: Institute of Electrical and Electronics Engineers
ISO: International Organization for Standardization
IEPE: Internal Electronic Piezoelectric
LBSG: Long Base Strain Gauge
LVDT: Linear Variable Differential Transformer
MEMS: Micro-Electro-Mechanical Systems
MRU: Motion Reference Unit
MSDV: Motion Sickness Dose Value
OMA: Operational Modal Analysis
PE: Piezoelectric
SHM: Structural Health Monitoring
VDV: Vibration Dose Value
VDR: Voyage Data Recorder

12 | STRUCTURAL HEALTH MONITORING: THE APPLICATION OF SENSOR-BASED APPROACHES | ABS

——
APPENDIX B: COMMONLY USED SENSOR TYPES
Table 4 lists common physical variables for marine and offshore assets and the commonly used sensor types to
measure or derive them.

Table 4: Common Physical Variables and Sensor Types

Physical Variable
Physical Variable Recommended Sensor Types
Measurement

Hull bending moment (vertical, horizontal and

Indirect Long based strain gauges (LBSGs)
torsional in longitudinal direction)

Long based strain gauges (LBSGs)

Sectional force (vertical and horizontal shear
Indirect
force)
Strain gauges (either electrical resistance or fiber optic type)

Accelerometer (vertical direction)

Slamming event detection Direct
Pressure transducer

Indirect Strain gauges (either electrical resistance or fiber optic type)

Pressure (slamming, sloshing, wave, etc.)
Direct Pressure transducer

Gyro heading/orientation Direct GNSS or from VDR

Rigid body motion (6 degrees of freedom) Direct Motion Reference Unit (MRU)

Inertia accelerometer, i.e., forced balanced type

Vibration - low frequency Direct
Accelerometer (ACC), e.g., piezo type, fiber optical, or
Micro-electro-mechanical systems (MEMs)

Accelerometer, e.g., piezo type, fiber optical, or

Vibration - high frequency Direct
Micro-electro-mechanical systems (MEMs)

Tension Direct Load cell - tension

Linear variable differential transformer (LVDT)

Direct
Structural displacement at a certain degree
High resolution differential global positioning system (DGPS)
of freedom
Indirect Integral based on accelerometer signal with DC component

Accelerometers

Structural deflection Indirect Strain gauges

Fiber optic sensors

Structural temperature Direct Temperature sensor e.g., thermal couple, fiber optic sensor)

Electrical strain gauge / rosette

Local strain/stress Direct
Fiber optic strain gauge

Strain gauges
Structural modal shapes and natural
Indirect
frequency
Accelerometers

Foundation fixities (dynamic) Indirect Accelerometers

Inclinometer or inertia accelerometers/gyros or loading

Inclination (heel, list, trim) Direct
computer data link

Global Navigation Satellite System (GNSS) or from Voyage data

Position Direct
recorder (VDR)

Speed Direct GNSS or from VDR

Course Direct GNSS or from VDR

Gyro heading/orientation Direct GNSS or from VDR

Crack initiation and propagation Direct Acoustic emission sensor, Electrochemical fatigue crack sensor

ABS | STRUCTURAL HEALTH MONITORING: THE APPLICATION OF SENSOR-BASED APPROACHES | 13

 Physical Variable
Physical Variable Recommended Sensor Types
Measurement

Crack Direct Guided wave sensor

Wind Direct Anemometer or Automated weather station

Wave/sea state Direct Wave radar

Acoustic doppler current profiler (ADCP)

Current Direct
Acoustic current meter

Conductivity-temperature-depth system (CTD) or

Water temperature Direct
Thermosalinographs

Salinity Direct CTD or Thermosalinographs

Ice concentration Direct Camera or Satellite image

Ice thickness Direct Camera via calibration

Air temperature Direct Thermometer or automated weather station

Air pressure Direct Barometer or Automated weather station

Water depth Direct Water depth sensor

Plate thickness Direct Point thickness sensor

Still water bending moment (SWBM) Indirect LBSGs or from loading computer

Power output and revolutions of propulsor(s) Direct Shaft torque meter, Shaft revolution counter

14 | STRUCTURAL HEALTH MONITORING: THE APPLICATION OF SENSOR-BASED APPROACHES | ABS

——
APPENDIX C: COMMENTARY ON ACCELEROMETER
AND PRESSURE TRANSDUCERS
There are various types of cost-effective accelerometers suitable for marine and offshore applications on the
market. In addition to the traditional inertia or piezo types, an emerging technology called Micro-Electro-
Mechanical Systems (MEMS) allows for producing inexpensive accelerometers in volume. Smart sensors, which
integrate microcontrollers (MCUs), digital signal processors (DSPs), application-specific integrated circuits (ASICs),
or field programmable gate arrays (FPGAs), are also commercially available. Table 5 compares the commonly
available accelerometer types and their potential applications for reference.

Table 6 provides an overview and comparison of the different pressure transducer types in terms of different
sensing technology and design philosophy. Low and/or medium range strain gauge type pressure transducers
(either electrical or FBG strain gauge type) or variable capacitance transducers are recommended when measuring
pressure from sea/waves (such as slamming) and liquid pressure (such as sloshing in tanks). Variable reluctance
transducers may be considered when extreme overpressure may be experienced and high degrees of accuracy and
critical stability over extended periods are required.

Table 5: Commonly Available Accelerometer Types [9, 20]

Sensing Technology Response Pros Cons Application

Inexpensive, small size Mobile and electronic devices for

and easy to integrate Poor signal to noise ratio, motion tracking and disk drive
Capacitive Micro- into electrical systems; limited bandwidth, and protection (for example, detecting
Electro-Mechanical DC Often come as surface- mostly restricted to small drops);
Systems (MEMS) mount devices that can acceleration levels (less Applicable to estimate displacement
be directly mounted to than 200 g) and velocity through integration over
printed circuit boards time.

Generally, not for lower frequency

Typically, low sensitivity;
and amplitude testing; Suitable for
Temperature
impulse/impact measurements which
Wide frequency compensation is required
have large frequency range and high
bandwidth; (some commercial
amplitude (for example, commonly
Measures down to zero products have
Piezoresistive (PR) DC used in automotive safety testing,
hertz so can be used for compensation internally
weapons testing, and higher shock
accurately calculating integrated);
range measurements);
velocity or displacement More expensive than
Applicable to estimate displacement
the capacitive MEMS
and velocity through integral over
accelerometers
time.

Widely used for test and

measurement;
Wide frequency
Not suitable for measuring Could be the first choice for most
bandwidth;
static accelerations and vibration measurements; Could be
Charge mode Good sensitivity;
low frequency vibrations; applied to extreme temperature
piezoelectric AC Easy installation;
Need special cabling to conditions (for example, -200°C to
(PE) Low noise levels;
shield from noise; Requires +640°C and beyond), such as turbine
Durable in hostile
a charge amplifier engine monitoring; Improper to
environments.
estimate displacement and velocity
through integral over time.

Widely used for tests and

measurements and could be the
Wide frequency first choice for most vibration
bandwidth; measurements;
Voltage mode Build-in microelectronic
Good sensitivity; Large operating temperature range
Internal Electronic circuit limits the ability
Easy installation; (for example, -40° to +125°C);
Piezoelectric (IEPE) to tolerate hostile
AC Low noise levels; Commonly used type for structural
(Charge mode PE environments when
Requires no special health monitoring purposes (for
with build-in charge compared to PE
cabling; example, high-rise buildings,
amplifier) accelerometers
Easily integrated with long-span bridges, and marine and
other systems. offshore structures);
Unable to estimate displacement and
velocity through integral over time.

ABS | STRUCTURAL HEALTH MONITORING: THE APPLICATION OF SENSOR-BASED APPROACHES | 15

 Table 6: Pressure Transducer Comparison and Overview [7]

Sensing Element
Design Category Sensing Element Remarks on Application
Configuration

For very low- or high-pressure measurements,

overpressure exposure, or high levels of vibration.
Linear Variable
Most commonly For measuring differential pressure of process
Differential
bellows, capsules, or media having high dielectric constants, especially
Transformers (LVDT)
Bourdon tubes liquid media.
Transducer
Accuracy and frequency response depend on
mechanical linkage and seal.

For very low-pressure measurements, overpressure

A bellows or Bourdon
exposure, high levels of vibration, stability and
Potentiometric tube is commonly
repeatability over extended periods of time, or
Transducer used as the sensing
extremely high-resolution requirements.
element
Frequency response depends on mechanical linkage.

Appropriate for measuring differential pressure of

Electrical Variable Capacitance
Diaphragm process media having high dielectric constants,
Pressure Transducer
especially liquid media.
Transducers

Well suited for measuring most process media,

especially if the core coil sensors are isolated from
the process media.
Variable Reluctance Well suitable for applications that include high shock
Diaphragm
Transducer or vibration levels, extreme overpressure, high degrees
of accuracy, or critical stability over extended periods.
All reluctance devices are affected by strong
magnetic fields.

Very effective in measuring changes in pressure.

Piezoelectric crystals
The piezoelectric crystals only produce an output when
Piezoelectric made of quartz,
they experience a change in load. With adequate signal
Transducer tourmaline, or
conditioners, they can also be used to perform static
ceramic material
measurements.

Fabry-Perot Two mirrors with

interferometers (FPI) one stator

Bragg Grating Diaphragm with Suitable for almost all applications.

Interferometer FBG Extremely sensitive and fit for high resolution
measurements.
Fiber-Optic
Recommended for environments where
Pressure
electromagnetic interference may be a concern.
Transducers
Is intrinsically safe and acceptable in
A pair of quartz hazardous environments.
resonators (one
vibrated due to
Quartz Resonators
pressure and the other
due to transducer
internal temperature)

Micromachined
Membrane/Diaphragm Membrane/Diaphragm
Deflection

16 | STRUCTURAL HEALTH MONITORING: THE APPLICATION OF SENSOR-BASED APPROACHES | ABS

——
APPENDIX D: COMMENTARY ON COUPLING SENSOR
DATA WITH ANALYSIS AND ANALYTICS MODELS
Vibration-based modal analysis approaches based on accelerometer measurements are well understood and have
been applied in various scenarios [8, 19]. In addition to the vibration-based approach, strain-based modal analysis
approaches based on strain gauge measurements have been investigated extensively [18, 21]. In strain-based
modal analysis, the mass-normalization of the displacement mode shapes and the strain mode shapes cannot
be performed with the purely strain-based Experimental Modal Analysis (EMA). The normalization (scaling)
problem for strain mode shapes can be addressed by employing an accelerometer in addition to the strain gauges.
In contrast to the vibration-based modal analysis, the strain-based modal analysis methodology has advantages
in sensing local damages with a high signal-to-noise due to the smaller sensor size and the working principles of
strain gauges. Strain-based modal analysis can be an alternative to vibration-based modal analysis in identifying
modal parameters (e.g., natural frequency, damping, and mode shape). Combining strain and vibration data is a
common practice for modal analysis [22].

Table 7 lists applicable sensors with popular vibration-based monitoring techniques and potential analysis and
models coupling. Table 8 lists the time series of vessel responses that can be obtained from structural sensors
and monitoring systems. Table 9 summarizes the application scenarios of the time series sensor data and the
corresponding required level of effort for signal processing.

Table 7: Sensor Recommendations and Potential Applications

Response
Sensor Package Option Application Example Use Case/Coupled Analysis and Model
Type

Fatigue damage direct calculation, ignoring mean

Fatigue AC
stress effect

Operational Modal Analysis (OMA) of wave induced global vibration, and

model updating and modal parameter tracking

Structural Natural frequency, damping ratio, and (unscaled) strain

AC mode shapes [#1]
Vibration (Cyclic)

Slamming/sloshing identification and load estimation

Strain gauge
(electrical or FBG)
Whipping/springing identification
with proper dynamic
strain instrument
Global load estimation through algorithms to interpret filtered sensor
data within wave frequency

Ice loading estimation

Strength (Static/
DC & AC
Quasi-static) Slamming/sloshing identification and load estimation to interpret filtered
sensor data within loading frequency.

Whipping/springing identification to interpret filtered sensor data within

wave frequency of concern

Human comfort evaluation, such as motion sickness dose value (MSDVZ)

and/or vibration dose value (VDV) based on filtered translational
accelerations

Motion Reference Unit Rigid Body

DC & AC Foundation fixity tuning that is based on OMA results of filtered
(MRU) Motion/Attitude
translational accelerations

Natural frequency, damping ratio, and (unscaled) displacement mode

shapes [#1]

ABS | STRUCTURAL HEALTH MONITORING: THE APPLICATION OF SENSOR-BASED APPROACHES | 17

 Response
Sensor Package Option Application Example Use Case/Coupled Analysis and Model
Type

Rigid Body Use algorithm to calculate the real-time 6 degree of freedom motion
DC & AC
Motion based on kinetics principle and linear algebra

Tilt Estimation DC & AC Tilt estimation (approximate)

OMA of filtered accelerations (such as wave induced global vibration)

and model updating and modal parameter tracking

Natural frequency, damping ratio, and (unscaled) displacement

mode shapes

Array of force-balanced Slamming/sloshing/whipping/springing identification to interpret filtered

accelerometer or sensor data within frequency of interest, that can correlate with strain
servo-type gauge measurement data to reduce inaccuracy
accelerometer [#3]
Structural
DC & AC Human comfort evaluation, such as motion sickness dose value (MSDVZ)
Vibration (Cyclic)
and/or VDV based on filtered translational accelerations

Estimates vibratory displacement/deformation of structure due

to wave load by double integral of acceleration and filtering
out the drift caused by integral of low frequency and constant
acceleration components, then conducts OMA of filtered vibratory
displacement/deformation (such as wave induced global response)
and model updating and modal parameter tracking
Output natural frequency, damping ratio, and (unscaled)
displacement mode shapes [#2]

Same as “Array of force-balanced accelerometer or

Array of MEMS Structural servo-type accelerometer”
Voltage mode IEPE Vibration AC IEPE has relatively high eco-technical advantage compared to
accelerometers [#4] (Cyclic) force-balanced accelerometer or servo-type accelerometer

Notes:
[#1] Both displacement mode shapes and strain mode shapes are related to structural vibration natural frequencies. Using a modal
decomposition and a modal expansion technique, it is possible to obtain the time series of dynamic displacement and stress
based on identified mode shapes respectively. The total displacement and stress should consist of both static and dynamic parts,
which can be found by adding the static part to the inferred dynamic part as an offset properly. Such total displacement and
stress can be used for strength and buckling check or extreme value analysis.

[#2] Some finite element software can only output displacement mode shapes. If that is a case, the strain can be obtained by imposing
displacement on to the finite element model.

[#3] MEMS Capacitive Accelerometers and MEMS Piezoresistive Accelerometers are economic alternatives of force-balanced
accelerometers or servo-type accelerometers, when large numbers of sensors are required.

[#4] When harsh operational environments are to be encountered, MEMS Charge mode piezoelectric (PE) accelerometers are
substituted for MEMS Voltage mode Internal Electronic Piezoelectric (IEPE) accelerometers. A combination of IEPE or charge
mode PE accelerometers with a few force-balanced accelerometers or servo-type accelerometers could offer more flexibility and
a high quality of required time series that consists of both static and dynamic parts, when a proper algorithm (sensor fusion) is
adopted. This may benefit the hull monitoring system design.

18 | STRUCTURAL HEALTH MONITORING: THE APPLICATION OF SENSOR-BASED APPROACHES | ABS

 Table 8: Time Series of Vessel Responses after Digital Signal Processing (DSP)

Applicable Direct/
Pass-Band Stop-Band
Signal Type Indirect Basic Filter Applied Remarks
Frequency Frequency
Measurand

Including all static and vibration

No filtering
components

Including static, temperature

Low-pass
0.3 Hz 0.4 Hz variation, and wave-induced
Strain/stress/ filtering
components
Motion/Translational
Vibration
acceleration due to
Dynamic
wave Wave-induced vibration and hull
high-pass 0.01 Hz
global vibration components
filtering

High-pass
0.4 Hz 0.3 Hz Hull global vibration components
filtering

Sea pressure due to Low -pass

5 Hz
slamming filtering

As slamming indicator, acceleration

Translational
Low -pass > 5 Hz is not due to wave induced
Shock acceleration due to 5 Hz
filtering response or wave induced global
slamming
vibration

Liquid pressure due to Low -pass

30 Hz
sloshing filtering

Note: the pass-band and stop-band frequency are for typical ship applications, and are for reference purpose only.

High-pass Filter

ABS | STRUCTURAL HEALTH MONITORING: THE APPLICATION OF SENSOR-BASED APPROACHES | 19

 Table 9: Required Effort Level for Sensor Data Processing

Effort Level Feature Extraction Feature Description

Maximum value

Minimum value

Mean value

Standard deviation
1. Statistics of time series data
i. No filtering Skewedness
ii. Low-pass filtering
Kurtosis
iii. Dynamic high-pass filtering
iv. High-pass filtering Mean zero crossing period (or mean crossing up count)

Maximum peak to peak value

Basic Number of observations used to calculate statistical parameters

Histogram of all the peaks

Histogram of all the troughs

Integrated energy of each event

2. Features for transient signal
Rise time of each event
(such as slamming, sloshing, impact)
Number of events during time interval

3. Probability distribution and threshold value

4. Threshold values and alarms for Alarms triggered when the measured value exceeds the threshold (or a given percentage of the
each channel threshold).

5. Fatigue damage estimation from

Rain flow counting of successive stress time series interval without overlap.
strain sensors

6. Parametric roll Can be done by MRU directly, for roll and pitch specifically.

7. Trend predictions for each channel Calculations for each time interval of each sensor are adopted for trend prediction.

8. Hull strain/stress

i. Raw strain Option 1: raw strain (all components, no filtering).

Medium Option 2: dynamic strain after high-pass filter in order to remove low cycle temperature effects (only
ii. Dynamic strain
dynamic).

Option 3: compounded strain by adding a strain offset corresponding to the static strain retrieved
iii. Compounded strain by filtered
from the loading computer at sensor location to the filtered strain without low cycle temperature
dynamic strain and loading computer
effects via a high-pass filter.

Option 4: compounded strain by summing dynamic strain (only dynamic) and the mean of the
iv. Compounded strain by filtered dynamic
low pass filtered measured strain (only static; where the mean is taken over a day to remove the
and static strain
temperature effects).

9. Modal analysis Recommended 20-30 min time series with sampling rate 20 Hz.

i. Natural frequency of global vibration

Array of individual accelerometers.
modes - displacement based

ii. Damping of global vibration modes-

Array of individual accelerometers.
displacement based

iii. Mode shape of global vibration modes -

Array of individual accelerometers.
displacement based

iv. Natural frequency of global vibration

Array of individual strain gauges (i.e. LBSG).
modes - strain based

v. Damping of global vibration modes-

Array of individual strain gauges (i.e. LBSG).
strain based

vi. Mode shape of global vibration modes -

Array of individual strain gauges (i.e. LBSG).
strain based

Event data trigger by predefined threshold as time series

A slamming warning level is to consider the whipping effect (without wave-induced response) from
Advanced the global deck sensors located amidships. The warning level based on whipping from global strain
10. Loads due to transient sea pressure
sensors in deck amidships is based on the critical situation where slamming occurs on one side of the
(slamming)
bow in bow quartering seas. This would be equivalent to a whipping response that is twice as high in
head seas at the same slamming magnitude when slamming occurs on both sides simultaneously. In
head seas the warning level should therefore be regarded as an early warning.

11. Global sectional forces Recommended 20-30 min time series with sampling rate 20 Hz.

Refer to hull strain/stress Option 1 of Table 9 and Option 4 of Table 9 to estimate the bending moment
whose proper estimate could be the mean value of the time series. The SWBM’s estimates can be
i. Still water bending moment (SWBM)
assembled according to loading conditions, and the statistics of same loading condition may be
computed.

ii. Dynamic sectional forces (moment Refer to hull strain / stress Option 1, Option 3 and Option 4 which can be used to define the DLPs
and forces) (based on extreme value) for ships.

12. Loads due to liquid motions in tanks (sloshing) Event data trigger by predefined threshold as time series.

13. Response due to operation in ice Event data trigger by predefined threshold as time series.

14. Comfort measurements by motion sickness dose

Calculated from translational accelerations from tri-axial accelerometer or high spec MRU.
value (MSDVZ) and/or Vibration dose value (VDV)

20 | STRUCTURAL HEALTH MONITORING: THE APPLICATION OF SENSOR-BASED APPROACHES | ABS

——
APPENDIX E: REFERENCES
1. ABS Guidance Notes on SafeHull Finite Element Analysis for Hull Structures
2. ABS Guide for Hull Condition Monitoring System
3. ABS Guide for Dynamic Loading Approach for Floating Production, Storage and Offloading (FPSO) Installations
4. ABS Guidance Notes on Smart Function Implementation
5. ABS Guide for SafeHull-Dynamic Loading Approach for Vessels
6. ABS Guide for Smart Functions for Marine Vessels and Offshore Units
7. ASTM F2070 − 00 (Reapproved 2017) Standard Specification for Transducers, Pressure and Differential, Pressure,
Electrical and Fiber-Optic.
8. Farrar, C. R., & Worden, K. (2012). Structural Health Monitoring: A Machine Learning Perspective. Chichester, West
Sussex, U.K.; Hoboken, N.J.: Wiley.
9. Hanly, S. (2016, March 11). Accelerometers: Taking the Guesswork out of Accelerometer Selection. Retrieved from
Mide Technology: https://blog.mide.com/accelerometer-selection
10. IEC 62008 Performance characteristics and calibration methods for digital data acquisition systems and relevant
software
11. IEC 60050-300 International Electrotechnical Vocabulary (IEV) Electrical and electronic measurements and
measuring instruments
12. IEC Guide 115 Application of uncertainty of measurement to conformity assessment activities in the
electrotechnical sector
13. IEEE Std 2700™-2017(E) IEEE Standard for Sensor Performance Parameter Definitions
14. ISO/IEC Guide 99 International vocabulary of metrology — Basic and general concepts and associated terms
(VIM)
15. ISO/IEC GUIDE 98-3 Uncertainty of measurement — Part 3: Guide to the expression of uncertainty in
measurement (GUM:1995)
16. ISO/IEC GUIDE 98-1 Uncertainty of measurement — Part 1: Introduction to the expression of uncertainty in
measurement
17. ISO/IEC GUIDE 98-4 Uncertainty of measurement — Part 4: Role of measurement uncertainty in conformity
assessment
18. Kranjc, T., Slavic, J., & Boltezar, M. (2014, April 29). A comparison of the strain and the classic experimental modal
analysis. Journal of Vibration and Control. doi:10.1177/1077546314533137
19. Rainieri, C., & Fabbrocino, G. (2014). Operational Modal Analysis of Civil Engineering Structures. New York:
Springer.
20. Ryden, B. (2018, Feb 15). Choosing the Proper Accelerometer for Measurement Success
and Sanity. Retrieved from FierceElectronics: https://www.fierceelectronics.com/components/
choosing-proper-accelerometer-for-measurement-success-and-sanity
21. Santos, F. L., Peeters, B., Desmet, W., & Goes, L. C. (2016). Strain-based experimental modal analysis: new concepts
and practical aspects. Proceedings of International Conference on Noise and Vibration Engineering 2016 (ISMA
2016) and International Conference on Uncertainty in Structural Dynamics (USD 2016).
22. Santos, F. L., Peeters, B., Vorst, R. V., Desmet, W., & Goes, L. C. (2014). The use of strain and mixed strain/acceleration
measurements for modal analysis. Proceedings of the 9th International Conference on Structural Dynamics,
EURODYN 2014. Porto, Portugal.

ABS | STRUCTURAL HEALTH MONITORING: THE APPLICATION OF SENSOR-BASED APPROACHES | 21

CONTACT
INFORMATION
NORTH AMERICA REGION
1701 City Plaza Dr.
Spring, Texas 77389, USA
Tel: +1-281-877-6000
Email: [email protected]

SOUTH AMERICA REGION

Rua Acre, nº 15 - 11º floor, Centro
Rio de Janeiro 20081-000, Brazil
Tel: +55 21 2276-3535
Email: [email protected]

EUROPE AND AFRICA REGION

111 Old Broad Street
London EC2N 1AP, UK
Tel: +44-20-7247-3255
Email: [email protected]

MIDDLE EAST REGION

Al Joud Center, 1st floor, Suite # 111
Sheikh Zayed Road
P.O. Box 24860, Dubai, UAE
Tel: +971 4 330 6000
Email: [email protected]

GREATER CHINA REGION

World Trade Tower, 29F, Room 2906
500 Guangdong Road, Huangpu District,
Shanghai China 200000
Tel: +86 21 23270888
Email: [email protected]

NORTH PACIFIC REGION

11th Floor, Kyobo Life Insurance Bldg.
7, Chungjang-daero, Jung-Gu
Busan 48939, Korea, Republic of
Tel: +82 51 460 4197
Email: [email protected]

SOUTH PACIFIC REGION

438 Alexandra Road
#08-00 Alexandra Point, Singapore 119958
Tel: +65 6276 8700
Email: [email protected]

© 2020 American Bureau of Shipping.

All rights reserved.

TX 02/19 19016