Search results

Filters

  • Journals
  • Authors
  • Keywords
  • Date
  • Type

Search results

Number of results: 17
items per page: 25 50 75
Sort by:
Download PDF Download RIS Download Bibtex

Abstract

The purpose of this work is to distinguish between Acoustic Emission (AE) signals coming from mechanical friction and AE signals coming from concrete cracking, recorded during fourteen seismic simulations conducted with the shaking table of the University of Granada on a reinforced concrete slab supported on four steel columns. To this end, a particular criterion is established based on the Root Mean Square of the AE waveforms calculated in two different temporal windows. This criterion includes a parameter calculated by optimizing the correlation between the mechanical energy dissipated by the specimen (calculated by means of measurements with accelerometers and displacement transducers) and the energy obtained from the AE signals recorded by low-frequency piezoelectric sensors located on the specimen. The final goal of this project, initiated four years ago, is to provide a reliable evaluation of the level of damage of Reinforced Concrete specimens by means of AE signals to be used in future Structural Health Monitoring strategies involving RC structures.
Go to article

Authors and Affiliations

Francisco A. Sagasta
Juan L. Torné
Antonio Sánchez-Parejo
Antolino Gallego
Download PDF Download RIS Download Bibtex

Abstract

Winglets are introduced into modern aircraft to reduce wing aerodynamic drag and to consequently optimize the fuel burn per mission. In order to be aerodynamically effective, these devices are installed at the wing tip section; this wing region is generally characterized by relevant oscillations induced by flights maneuvers and gust. The present work is focused on the validation of a continuous monitoring system based on fiber Bragg grating sensors and frequency domain analysis to detect physical condition of a skin-spar bonding failure in a composite winglet for in-service purposes. Optical fibers are used as deformation sensors. Short Time Fast Fourier Transform (STFT) analysis is applied to analyze the occurrence of structural response deviations on the base of strain data. Obtained results showed high accuracy in estimating static and dynamic deformations and great potentials in detecting structural failure occurrences.

Go to article

Bibliography

[1] K. Diamanti and C. Soutis. Structural health monitoring techniques for aircraft composite structures. Progress in Aerospace Sciences, 46(8):342–352, 2010. doi: 10.1016/j.paerosci.2010.05.001.
[2] C. Bockenheimer and H. Speckmann. Validation, verification and implementation of SHM at Airbus. In Proceedings of the 9th International Workshop on Structural Health Monitoring (IWSHM 2013), Stanford University, Stanford, CA, USA, pages 10–12, 2013.
[3] H. Speckmann and H. Roesner. Structual Health Monitoring: A contribution to the intelligent aircraft structure. In Proceedings of ECNDT 2006, 9th European Conference on NDT, Berlin, Germany, Sept. 2006.
[4] O. Shapira, S. Kedem, B. Glam, N.Y. Shemesh, A. Dvorjetski, N. Mashiach, J. Balter, R. Shklovsky, I. Sovran, N. Gorbatov, et al. Implementation of a fiber-optic sensing technology for global structural integrity monitoring of UAVs. In The 54th Israel Annual Conference on Aerospace Sciences, Tel-Aviv, Israel, 2014.
[5] R. De Oliveira, O. Frazão, J.L. Santos, and A.T. Marques. Optic fibre sensor for real-time damage detection in smart composite. Computers & Structures, 82(17):1315–1321, 2004. doi: 10.1016/j.compstruc.2004.03.028.
[6] E. Di Lorenzo, G. Petrone, S. Manzato, B. Peeters,W. Desmet, and F. Marulo. Damage detection in wind turbine blades by using operational modal analysis. Structural Health Monitoring, 15(3):289–301, 2016. doi: 10.1177/1475921716642748.
[7] I. Dimino and A. Calabrò. Structural damage identification by vibration parametres and fibre optic sensors. Czech Aerospace, 2009(3):33–41, 2009.
[8] S. Bhalla and C.K. Soh. Structural health monitoring by piezo-impedance transducers. I: Modeling. Journal of Aerospace Engineering, 17(4):154–165, 2004. doi: 10.1061/(ASCE)0893-1321(2004)17:4(154).
[9] S. Bhalla and C.K. Soh. Electromechanical impedance modeling for adhesively bonded piezotransducers. Journal of Intelligent Material Systems and Structures, 15(12):955–972, 2004. doi: 10.1177/1045389X04046309.
[10] A. De Fenza, A. Sorrentino, and P. Vitiello. Application of Artificial Neural Networks and Probability Ellipse methods for damage detection using Lamb waves. Composite Structures, 133:390–403, 2015. doi: 10.1016/j.compstruct.2015.07.089.
[11] R. Di Sante. Fibre optic sensors for structural health monitoring of aircraft composite structures: Recent advances and applications. Sensors, 15(8):18666–18713, 2015. doi: 10.3390/s150818666.
[12] H. Takeya, T. Ozaki, and N. Takeda. Structural health monitoring of advanced grid structure using multi-point FBG sensors. Proc. SPIE, 5762:204–211, 2005. doi: 10.1117/12.598759.
[13] H. Murayama, K. Kageyama, H. Naruse, A. Shimada, and K. Uzawa. Application of fiber-optic distributed sensors to health monitoring for full-scale composite structures. Journal of Intelligent Material Systems and Structures, 14(1):3–13, 2003. doi: 10.1177/1045389X03014001001.
[14] G. Fabbi, M. Ciminello, A. Mataloni, P. Perugini, A. Sorrentino, and A. Concilio. Filament wound solid rocket motor vessels strain measurement and potential Structural Health Monitoring through fiber optics. In The space Propulsion 201 Conference, Rome, Italy, 2-6 May 2016. Paper No. SP2016-3125185.
[15] M. Ciminello, I. Dimino, S. Ameduri, and A. Concilio. Fiber optic shape sensor for morphing wing trailing edge. In Proceedings of 26th International Conference on Adaptive Structures and Technologies (ICAST2015), pages 312–318, 14-16 Oct. 2015.
[16] J.R. Lee, C.Y. Ryu, B.Y. Koo, S.G. Kang, C.S. Hong, and C.G. Kim. In-flight health monitoring of a subscale wing using a fiber bragg grating sensor system. Smart Materials and Structures, 12(1):147, 2003. doi: 10.1088/0964-1726/12/1/317.
[17] A. De Fenza, G. Petrone, R. Pecora, and M. Barile. Post-impact damage detection on a winglet structure realized in composite material. Composite Structures, 169:129–137, 2017. doi: 10.1016/j.compstruct.2016.10.004.
[18] MD Nastran. Quick Reference Guide, 2011.
[19] K.O. Hill and G. Meltz. Fiber Bragg grating technology fundamentals and overview. Journal of Lightwave Technology, 15(8):1263–1276, 1997. doi: 10.1109/50.618320.
Go to article

Authors and Affiliations

Monica Ciminello
1
Angelo De Fenza
2 3
Ignazio Dimino
1
Rosario Pecora
2

  1. Italian Aerospace Research Center, Capua, Italy
  2. Department of Industrial Engineering – Aerospace Division, University of Naples “Federico II”, Naples, Italy
  3. NOVOTECH s.r.l. – Aerospace Advanced Technology, Naples, Italy
Download PDF Download RIS Download Bibtex

Abstract

Safety of dams and other hydraulic structures is a complex procedure that must consider the individual characteristics of each structure and provide an insight in the structural health at every stage of the structure’s life cycle. Failures of structures permanently or temporarily retaining water may cause large economic damage, environmental disasters, and loss of lives. An engineering design should, therefore, guarantee maximum security of such structures or maximize their reliability not only in ordinary operating conditions but also under extreme hydrological load. By performing structural heath monitoring (SHM), the safety can be optimized, including the performance and life expectancy of a structure by adopting an appropriate methodology to observe the identified failure modes for a selected dam type. To adopt SHM to hydraulic structures it is important to broaden the knowledge and understanding of the ageing processes on hydraulic structures, which can be achieved by laboratory testing and application and development of novel monitoring techniques, e.g., vibration monitoring. In Slovenia, we are increasingly faced with the problem of ageing of dam structures. At the same time, we are also faced with changes in the environment, especially with the variability in time-dependent loads and with new patterns of operation on dams used for hydropower, with several starts and stops of turbines happening on a daily basis. These changes can lead to a decrease in structural and operational safety of dams. In this paper we propose a methodology where the dynamic response of concrete dams is continuously monitored in few locations on the dam using accelerometers, while all significant structural members are measured in discrete time intervals using portable vibrometers. We focused on run-of-the-river dams, which are a common dam type in Slovenia. The pilot case for the system is lower Sava River with a cascade of 5 dams used for hydropower.
Go to article

Authors and Affiliations

Mateja Klun
1
ORCID: ORCID
Andrej Kryžanowski
1
ORCID: ORCID

  1. University of Ljubljana, Faculty of Civil and Geodetic Engineering, Jamova 2, 1000 Ljubljana, Slovenia
Download PDF Download RIS Download Bibtex

Abstract

The knowledge of the load in prestressed bolted connections is essential for the proper operation and safety of engineering structures. Recently, bolted joints have become an area of intensive research associated with non-destructive diagnostics, in particular in the context of wave propagation techniques. In this paper, a novel procedure of bolt load estimation based on the energy of Lamb wave signals was proposed. Experimental tests were performed on a single lap joint of two steel plates. Ultrasonic waves were excited and registered by means of piezoelectric transducers, while precise measurement of the bolt load was obtained by means of using the force washer transducer. Experimental tests were supported by the finite element method analysis based on Schoenberg’s concept. The results showed that the relationship between the bolt load and signal energy was strongly nonlinear and it depended on the location of acquisition points.

Go to article

Authors and Affiliations

R. Kędra
M. Rucka
Download PDF Download RIS Download Bibtex

Abstract

The present paper is devoted to the discussion and review of the non-destructive testing methods mainly based on vibration and wave propagation. In the first part, the experimental methods of actuating and analyzing the signal (vibration) are discussed. The piezoelectric elements, fiber optic sensors and Laser Scanning Doppler Vibrometer (SLDV) method are described. Effective detecting of the flaws needs very accurate theoretical models. Thus, the numerical methods, e.g. finite element, spectral element method and numerical models of the flaws in isotropic and composite materials are presented. Moreover, the detection of the damage in structures, which are subjected to cyclic or static loads, is based on the analyzing of the change in natural frequency of the whole structure, the change of internal impedance of the material and the change in guided waves propagating through the investigated structure. All these cases are characterized in detail. At the end of this paper, several applications of the structural health monitoring systems in machine design and operation are presented.

Go to article

Authors and Affiliations

Marek Barski
Piotr Kędziora
Aleksander Muc
Paweł Romanowicz
Download PDF Download RIS Download Bibtex

Abstract

The paper describes studies on the influence of humidity on the electrical resistance of a textile sensor made of carbon fibres. The concept of the sensor refers to externally bonded fibre reinforcement commonly used for strengthening of structures, however the zig-zag arrangement of carbon fibre tow allows for measuring its strain. The sensor tests showed its high sensitivity to the temperature and humidity changes which unfavourably affects the readings and their interpretation. The influence of these factors must be compensated. Due to the size of the sensor, there is not possible electrical compensation by the combining of “dummy” sensors into the half or full Wheatstone bridge circuit. Only mathematical compensation based on known humidity resistance functions is possible. The described research is the first step to develop such relations. The tests were carried out at temperatures of 10°C, 20°C, 30°C and humidity in the range of 30-90%.

Go to article

Authors and Affiliations

Marcin Górski
Rafał Krzywoń
Sofija Kekez
Download PDF Download RIS Download Bibtex

Abstract

With the developing technology and increasing construction, the importance of structural observations, which are of great significance in disaster management, has increased. Geodetic methods have been preferred in recent years due to their high accuracy and ease of use in Structural Health Monitoring (SHM) Surveys. In this study, harmonic oscillation tests have been carried out on a shake table to determine the usability of the Single Base and the Network Real-Time Kinematic (RTK) Global Navigation Satellite Systems (GNSS) method in SHM studies. It is aimed to determine the harmonic movements of different amplitudes and frequencies created by the shake table with 20 Hz multi-GNSS equipment. The amplitude and frequency values of the movements created using Fast Fourier Transform (FFT) and Time Series Analysis have been calculated. The precision of the analysis results has been determined by comparing the LVDT (Linear Variable Differential Transformer) data, which is the position sensor of the shake table, with the GNSS data. The advantages of the two RTK methods over each other have been determined using the calculated amplitude and frequency differences. As a result of all experiments, it has been determined that network and single base RTK GNSS methods effectively monitor structural behaviours and natural frequencies.
Go to article

Authors and Affiliations

Güldane Oku Topal
1
ORCID: ORCID
Burak Akpinar
1
ORCID: ORCID

  1. Yildiz Technical Universty, Istanbul, Turkey
Download PDF Download RIS Download Bibtex

Abstract

The condition monitoring of offshore wind power plants is an important topic that remains open. This monitoring aims to lower the maintenance cost of these plants. One of the main components of the wind power plant is the wind turbine foundation. This study describes a data-driven structural damage classification methodology applied in a wind turbine foundation. A vibration response was captured in the structure using an accelerometer network. After arranging the obtained data, a feature vector of 58 008 features was obtained. An ensemble approach of feature extraction methods was applied to obtain a new set of features. Principal Component Analysis (PCA) and Laplacian eigenmaps were used as dimensionality reduction methods, each one separately. The union of these new features is used to create a reduced feature matrix. The reduced feature matrix is used as input to train an Extreme Gradient Boosting (XGBoost) machine learning-based classification model. Four different damage scenarios were applied in the structure. Therefore, considering the healthy structure, there were 5 classes in total that were correctly classified. Five-fold cross validation is used to obtain a final classification accuracy. As a result, 100% of classification accuracy was obtained after applying the developed damage classification methodology in a wind-turbine offshore jacket-type foundation benchmark structure.
Go to article

Authors and Affiliations

Jersson X. Leon-Medina
1 2
ORCID: ORCID
Núria Parés
3
ORCID: ORCID
Maribel Anaya
4
ORCID: ORCID
Diego A. Tibaduiza
4
ORCID: ORCID
Francesc Pozo
1 5
ORCID: ORCID

  1. Control, Data, and Artificial Intelligence (CoDAlab), Department of Mathematics, Escola d’Enginyeria de Barcelona Est (EEBE),Campus Diagonal-Besòs (CDB), Universitat Politècnica de Catalunya (UPC), Eduard Maristany 16, 08019 Barcelona, Spain
  2. Programa de Ingeniería Mecatrónica, Universidad de San Buenaventura, Carrera 8H #172-20, Bogota, Colombia
  3. Laboratori de Càlcul Numèric (LaCàN), Department of Mathematics, Escola d’Enginyeria de Barcelona Est (EEBE), Campus Diagonal-Besòs
  4. Departamento de Ingeniería Eléctrica y Electrónica, Universidad Nacional de Colombia, Cra 45 No. 26-85, Bogotá 111321, Colombia
  5. Institute of Mathematics (IMTech), Universitat Politècnica de Catalunya (UPC), Pau Gargallo 14, 08028 Barcelona, Spain
Download PDF Download RIS Download Bibtex

Abstract

The hereby paper discusses the influence of cable length on the SHM systems with the use of vibrating wire dynamic measurements. Vibrating wire sensors are mainly used for measuring stable or slowly changing strains, e.g. system installed on Rędziński Bridge in Wroclaw. From some time applications of these sensors for measuring dynamic deformations are becoming popular. Such tests were conducted on STS Fryderyk Chopin. New solutions generate new problems. In this case: the operational stability of systems exciting wire vibrations. The structure of such sensors and the electric cables length has an essential influence on their operations, what is undertaken in the paper. The subject of investigations constitutes the measuring system based on self-exciting impulse exciter, for which impedance parameters of electric cables and of the vibrating wire sensor were the most essential. The mathematical model of the system, experimental verification of the model as well as the results of theoretical analyses at the application of electric cables of various lengths are presented in the paper.

Go to article

Authors and Affiliations

G. Cieplok
W. Karwowski
Ł. Bednarski
Download PDF Download RIS Download Bibtex

Abstract

The paper presents a concept and realization of monitoring system for the Silesian Stadium in Chorzow. The idea of the system lies in fusion of structure monitoring with a calibrated numerical FEM model [1]. The inverse problem is solved. On the base of measured selected displacements, the numerical FEM model of the structure combined with iterative method, develops the current snow load distribution. Knowing the load, we can calculate the forces and stresses in each element of the structure and thanks to this we can determine the safety thresholds and asses the owner. Test results and conclusions are presented.

Go to article

Authors and Affiliations

K. Zółtowski
M. Drawc
Download PDF Download RIS Download Bibtex

Abstract

The paper presents a method of structural monitoring with the use of angular displacement measurements performed with inclinometer devices. Inclinometer method is a solution free from the basic disadvantages of optical methods used commonly in structural monitoring, such as sensitivity to any type of visibility restrictions, pollution or influence of weather conditions. At the same time, with appropriate sensor parameters, a much better measurement accuracy is obtained than for typical optical methods and very low energy demand and moderate costs are achieved. Taking into account the above-mentioned issues, in the first stage an appropriate MEMS-type inclinometer sensor was selected, its laboratory tests were carried out and a method of the offset temperature drift correction, individual for each sensor, was developed.

Go to article

Authors and Affiliations

S. Wierzbicki
Z. Pióro
M. Osiniak
E. Antoszkiewicz
Download PDF Download RIS Download Bibtex

Abstract

Monitoring and structural health assessment are the primary requirements for performance evaluation of damaged bridges. This paper highlights the case-study of a damaged Reinforced Concrete (RC) bridge structure by considering the outcomes of destructive testing, Non-Destructive Testing (NDT) evaluations, static and 3D non-linear analysis methods. Finite element (FE) modelling of this structure is being done using the material properties extracted by the in-situ testing. Analysis is carried out to evaluate the bridge damage based on the data recorded after the static linear (AXIS VM software) and 3D non-linear analysis (ATENA 3D software). Extensive concrete cracking and high value of crack width are found to be the major problems, leading to lowering the performance of the bridge. As a solution, this paper proposes a proper Structural Health Monitoring (SHM) system, that will extend the life cycle of the bridge with minimal repair costs and reduced risk of failure. This system is based on the installation of three different types of sensors: Liquid Levelling sensors (LLS) for measurement of vertical displacement, Distributed Fiber Optic Sensors (DFOS) for crack monitoring, and Weigh in Motion (WIM) devices for monitoring of moving loads on bridge.
Go to article

Authors and Affiliations

Muhammad Fawad
1
ORCID: ORCID
Kalman Koris
2
ORCID: ORCID
Marek Salamak
1
ORCID: ORCID
Michael Gerges
3
ORCID: ORCID
Lukasz Bednarski
4
ORCID: ORCID
Rafał Sienko
5
ORCID: ORCID

  1. Silesian University of Technology, Faculty of Civil Engineering, ul. Akademicka 2A, 44-100 Gliwice, Poland
  2. Budapest University of Technology and Economics, Faculty of Civil Engineering, Muegyetem rkp. 3,1111 Budapest, Hungary
  3. University of Wolverhampton, Wulfruna St, Wolverhampton WV1 1LY, United Kingdom, UK
  4. AGH University of Science, Mechanical Engineering and Robotics, ul. Mickiewicza 30, 30-059 Kraków, Poland
  5. Cracow University of Technology, Faculty of Civil Engineering, ul. Warszawska 24, 31-155 Kraków, Poland
Download PDF Download RIS Download Bibtex

Abstract

Time series models have been used to extract damage features in the measured structural response. In order to better extract the sensitive features in the signal and detect structural damage, this paper proposes a damage identification method that combines empirical mode decomposition (EMD) and Autoregressive Integrated Moving Average (ARIMA) models. EMD decomposes nonlinear and non-stationary signals into different intrinsic mode functions (IMFs) according to frequency. IMF reduces the complexity of the signal and makes it easier to extract damage-sensitive features (DSF). The ARIMA model is used to extract damage sensitive features in IMF signals. The damage sensitive characteristic value of each node is used to analyze the location and damage degree of the damaged structure of the bridge. Considering that there are usually multiple failures in the actual engineering structure, this paper focuses on analysing the location and damage degree of multi-damaged bridge structures. A 6-meter-long multi-destructive steel-whole vibration experiment proved the state of the method. Meanwhile, the other two damage identification methods are compared. The results demonstrate that the DSF can effectively identify the damage location of the structure, and the accuracy rate has increased by 22.98% and 18.4% on average respectively.
Go to article

Authors and Affiliations

Weijia Lu
1
ORCID: ORCID
Jiafan Dong
1
ORCID: ORCID
Yuheng Pan
1
ORCID: ORCID
Guoya Li
1
ORCID: ORCID
Jinpeng Guo
1
ORCID: ORCID

  1. Tianjin Chengjian University, Computer and Information Engineering Department, Tianjin, China
Download PDF Download RIS Download Bibtex

Abstract

The paper presents a method of structural monitoring using measurement of vertical displacements realized optically by horizontally directed laser beam. A measuring device with an integrated rangefinder and inclinometer sensor was developed. Inclinometer sensor are used to correct measurement results of the rangefinder in order to eliminate errors resulting from spatial position changes of the laser beam. Such a solution was adopted as an alternative to a more complex and demanding method, which is the stabilization of the laser beam orientation. The proposed inclinometric correction method allows in a simple and clear way to eliminate a serious problem of the displacement measurement method with a perpendicularly directed laser beam, which is inevitable in practice the lack of permanent stability of the measuring device position. The developed measuring device is wireless, both in terms of power supply and communication with other elements of the monitoring system. In order to verify the correctness of measurements carried out by the developed device, on site tests were carried out in two industrial-warehouse buildings with functioning monitoring systems using other measurement methods, earlier verified. The tests confirmed compliance with the indications of the existing system at a level completely sufficient for structural monitoring system purposes. The conducted research show that the proposed method of displacements measurement with inclinometric correction of errors, provides accurate and reliable results, allowing also to obtain additional information about the behaviour of the structure in the place of installation of the measuring device.
Go to article

Authors and Affiliations

Stanisław Wierzbicki
1
ORCID: ORCID
Zbigniew Pióro
2 3
Marcin Osiniak
3

  1. Warsaw University of Technology, Faculty of Civil Engineering, Al. Armii Ludowej 16, 00-637 Warsaw, Poland
  2. Warsaw University of Technology, Faculty of Electronics and Information Technology, Nowowiejska15/19, 00-665 Warsaw (retired professor)
  3. WiSeNe Sp. z o.o., Taneczna 27, 02-829 Warsaw, Poland
Download PDF Download RIS Download Bibtex

Abstract

The paper presents a method of structural monitoring with the use of angular and linear displacement measurements performed using inclinometer and laser measuring devices. The focus is mainly on the inclinometer measurement method, which is a solution free from the basic disadvantages of optical methods, such as sensitivity to any type of visibility restrictions, pollution or influence of weather conditions. Testing of this method was carried out in practical application in an wireless monitoring system, installed in a large-area industrial building. The measurement results performed using the inclinometers were compared with simultaneous measurements of linear displacements performed with the use of proven methods based on laser rangefinders. The research and analysis show that the method of measuring angular displacements using the inclinometers with MEMS sensors of appropriate quality is a very good, better than typical optical methods, solution of structural monitoring systems that allows to obtain accurate and reliable results.

Go to article

Authors and Affiliations

Stanisław Wierzbicki
ORCID: ORCID
Zbigniew Pióro
Marcin Osiniak
Edward Antoszkiewicz
Download PDF Download RIS Download Bibtex

Abstract

Structural health monitoring (SHM) of bridges is constantly upgraded by researchers and bridge engineers as it directly deals with bridge performance and its safety over a certain time period. This article addresses some issues in the traditional SHM systems and the reason for moving towards an automated monitoring system. In order to automate the bridge assessment and monitoring process, a mechanism for the linkage of Digital Twins (DT) and Machine Learning (ML), namely the Support Vector Machine (SVM) algorithm, is discussed in detail. The basis of this mechanism lies in the collection of data from the real bridge using sensors and is providing the basis for the establishment and calibration of the digital twin. Then, data analysis and decision-making processes are to be carried out through regression-based ML algorithms. So, in this study, both ML brain and a DT model are merged to support the decision-making of the bridge management system and predict or even prevent further damage or collapse of the bridge. In this way, the SHM system cannot only be automated but calibrated from time to time to ensure the safety of the bridge against the associated damages.
Go to article

Authors and Affiliations

Asseel Za'al Ode Al-Hijazeen
1
ORCID: ORCID
Muhammad Fawad
1 2
ORCID: ORCID
Michael Gerges
3
ORCID: ORCID
Kalman Koris
1
ORCID: ORCID
Marek Salamak
2
ORCID: ORCID

  1. Budapest University of Technology and Economics, Faculty of Civil Engineering, Muegyetem rkp. 3, 1111 Budapest, Hungary
  2. Silesian University of Technology, Faculty of Civil Engineering, ul. Akademicka 2A, 44-100 Gliwice, Poland
  3. University of Wolverhampton, Wulfruna St, Wolverhampton WV1 1LY, the United Kingdom
Download PDF Download RIS Download Bibtex

Abstract

Operational load monitoring (OLM) is an industrial process related to structural health monitoring, where fatigue of the structure is tracked. Artificial intelligence methods, such as artificial neural networks (ANNs) or Gaussian processes, are utilized to improve efficiency of such processes. This paper focuses on moving such processes towards green computing by deploying and executing the algorithm on low-power consumption FPGA where high-throughput and truly parallel computations can be performed. In the following paper, the OLM process of typical aerostructure (hat-stiffened composite panel) is performed using ANN. The ANN was trained using numerically generated data, of every possible load case, to be working with sensor measurements as inputs. The trained ANN was deployed to Xilinx Artix-7 A100T FPGA of a real-time microcontroller. By executing the ANN on FPGA (where every neuron of a given layer can be processed at the same time, without limiting the number of parallel threads), computation time could be reduced by 70% as compared to standard CPU execution. Series of real-time experiments were performed that have proven the efficiency and high accuracy of the developed FPGA-based algorithm. Adjusting the ANN algorithm to FPGA requirements takes some effort, however it can lead to high performance increase. FPGA has the advantages of many more potential parallel threads than a standard CPU and much lower consumption than a GPU. This is particularly important taking into account potential embedded and remote applications, such as widely performed monitoring of airplane structures.
Go to article

Authors and Affiliations

Waldemar Mucha
1
ORCID: ORCID

  1. Department of Computational Mechanics and Engineering, Silesian University of Technology, Konarskiego 18A, 44-100 Gliwice, Poland

This page uses 'cookies'. Learn more