Mohammad Hesam Soleimani-Babakamali, Ismini Lourentzou, R. Sarlo
The curse of dimensionality (CD) brings difficulties in pattern recognition problems, such as those found in structural health monitoring (SHM). Dimensionality reduction techniques (DR) make data more manageable by reducing noise and noninformative portions. There exists a trade-off between CD and the loss of information due to the application of DR. Even though in supervised SHM, DR techniques are shown to be effective, for unsupervised SHM, the trade-off must be assessed due to the unknown data population of novel classes. This study assesses the trade-off concerning a novel method working with a raw frequency-domain feature, the fast Fourier transform (FFT). Different DR techniques are applied to the initial FFT-based feature to assess the trade-off, and detection results are compared. The results indicate that the loss of information can have detrimental effects, such as lowering the detection accuracy by 60% for the autoencoder-based DR. The accuracy reduction is present for all different DR techniques applied in the study; however, regularization lessens the accuracy decrements. This phenomenon indicates the assumption that novelties show themselves in less-vary portions of the baseline condition to be not true.
{"title":"DOES THE CURSE OF DIMENSIONALITY APPLY TO UNSUPERVISED SHM? INVESTIGATING THE TRADE-OFF BETWEEN LOSS OF INFORMATION AND GENERALIZABILITY TO UNSEEN STRUCTURAL CONDITIONS","authors":"Mohammad Hesam Soleimani-Babakamali, Ismini Lourentzou, R. Sarlo","doi":"10.12783/shm2021/36311","DOIUrl":"https://doi.org/10.12783/shm2021/36311","url":null,"abstract":"The curse of dimensionality (CD) brings difficulties in pattern recognition problems, such as those found in structural health monitoring (SHM). Dimensionality reduction techniques (DR) make data more manageable by reducing noise and noninformative portions. There exists a trade-off between CD and the loss of information due to the application of DR. Even though in supervised SHM, DR techniques are shown to be effective, for unsupervised SHM, the trade-off must be assessed due to the unknown data population of novel classes. This study assesses the trade-off concerning a novel method working with a raw frequency-domain feature, the fast Fourier transform (FFT). Different DR techniques are applied to the initial FFT-based feature to assess the trade-off, and detection results are compared. The results indicate that the loss of information can have detrimental effects, such as lowering the detection accuracy by 60% for the autoencoder-based DR. The accuracy reduction is present for all different DR techniques applied in the study; however, regularization lessens the accuracy decrements. This phenomenon indicates the assumption that novelties show themselves in less-vary portions of the baseline condition to be not true.","PeriodicalId":180083,"journal":{"name":"Proceedings of the 13th International Workshop on Structural Health Monitoring","volume":"62 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-03-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132377411","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Aims to address the issue that the degradation model may not accurately represent the underly true degradation physics in failure prognostics of miter gates, this paper presents a framework for degradation model correction using historical strain measurements. A stochastic gap growth model with uncertain model parameters is employed as the simplified degradation model to predict the gap evolution. A dynamic model discrepancy quantification framework is then proposed to correct the simplified model by representing the model bias term as a data-driven surrogate model. After that, a maximum likelihood estimation method is developed to estimate the parameters of the data-driven surrogate model using strain measurements. Additionally, the uncertainty in the model parameters of the simplified model is reduced using Bayesian method. The corrected and updated simplified degradation model is then employed for failure prognostics of a miter gate. Results of a case study show that the updated degradation model can accurately predict multi-step ahead gap growth while performing damage prognostics and remaining useful life estimation.
{"title":"ACCURACY IMPROVEMENT OF A DEGRADATION MODEL FOR FAILURE PROGNOSIS OF MITER GATES","authors":"Chen Jiang, M. A. Vega, Michael D. Todd, Zhen Hu","doi":"10.12783/shm2021/36357","DOIUrl":"https://doi.org/10.12783/shm2021/36357","url":null,"abstract":"Aims to address the issue that the degradation model may not accurately represent the underly true degradation physics in failure prognostics of miter gates, this paper presents a framework for degradation model correction using historical strain measurements. A stochastic gap growth model with uncertain model parameters is employed as the simplified degradation model to predict the gap evolution. A dynamic model discrepancy quantification framework is then proposed to correct the simplified model by representing the model bias term as a data-driven surrogate model. After that, a maximum likelihood estimation method is developed to estimate the parameters of the data-driven surrogate model using strain measurements. Additionally, the uncertainty in the model parameters of the simplified model is reduced using Bayesian method. The corrected and updated simplified degradation model is then employed for failure prognostics of a miter gate. Results of a case study show that the updated degradation model can accurately predict multi-step ahead gap growth while performing damage prognostics and remaining useful life estimation.","PeriodicalId":180083,"journal":{"name":"Proceedings of the 13th International Workshop on Structural Health Monitoring","volume":"57 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-03-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123781280","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Structural health monitoring (SHM) of bridges often involves machine learning algorithms, trained based on two independent learning strategies, namely unsupervised and supervised learning, depending on the type of training data available. When unsupervised learning strategy is employed, the algorithms are normally trained with data gathered from monitoring systems, corresponding to normal operational and environmental conditions. The lack of information regarding the dynamic response of the structure under extreme environmental and operational conditions, as well as under damage scenarios, may lead to flaws in the damage detection process, namely the rise of false indications of damage. In order to overcome this drawback, finite element models can be used as structural proxies to generate data that correspond to scenarios unlikely to be recorded by the monitoring systems, such as extreme temperatures or structural damage. The use of both monitoring and numerical data in the framework of a hybrid approach greatly improves the quality of the training process, as recently shown by the authors. The hybrid approach also enables the use of the supervised learning strategy if numerical data corresponding to damage scenarios are available. Therefore, this paper assesses the reliability of a hybrid approach for the supervised training of machine learning algorithms using numerical data corresponding to extreme temperatures and several damage scenarios. The damage scenarios comprise various degrees of settlement of a bridge pier and a landslide near the same pier. Monitoring data are used for the testing of the algorithms and for the initial calibration of the finite element model, which does not need to be exceedingly detailed, as the probabilistic variation of the uncertain parameters is taken into account. The procedure was applied to the Z-24 Bridge, a well-known benchmark consisting of one year of continuous monitoring and including progressive damage readings.
{"title":"HYBRID SUPERVISED MACHINE LEARNING APPROACH FOR DAMAGE IDENTIFICATION IN BRIDGES","authors":"M. Bud, M. Nedelcu, I. Moldovan, E. Figueiredo","doi":"10.12783/shm2021/36294","DOIUrl":"https://doi.org/10.12783/shm2021/36294","url":null,"abstract":"Structural health monitoring (SHM) of bridges often involves machine learning algorithms, trained based on two independent learning strategies, namely unsupervised and supervised learning, depending on the type of training data available. When unsupervised learning strategy is employed, the algorithms are normally trained with data gathered from monitoring systems, corresponding to normal operational and environmental conditions. The lack of information regarding the dynamic response of the structure under extreme environmental and operational conditions, as well as under damage scenarios, may lead to flaws in the damage detection process, namely the rise of false indications of damage. In order to overcome this drawback, finite element models can be used as structural proxies to generate data that correspond to scenarios unlikely to be recorded by the monitoring systems, such as extreme temperatures or structural damage. The use of both monitoring and numerical data in the framework of a hybrid approach greatly improves the quality of the training process, as recently shown by the authors. The hybrid approach also enables the use of the supervised learning strategy if numerical data corresponding to damage scenarios are available. Therefore, this paper assesses the reliability of a hybrid approach for the supervised training of machine learning algorithms using numerical data corresponding to extreme temperatures and several damage scenarios. The damage scenarios comprise various degrees of settlement of a bridge pier and a landslide near the same pier. Monitoring data are used for the testing of the algorithms and for the initial calibration of the finite element model, which does not need to be exceedingly detailed, as the probabilistic variation of the uncertain parameters is taken into account. The procedure was applied to the Z-24 Bridge, a well-known benchmark consisting of one year of continuous monitoring and including progressive damage readings.","PeriodicalId":180083,"journal":{"name":"Proceedings of the 13th International Workshop on Structural Health Monitoring","volume":"18 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-03-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125210502","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
J. Simon, J. Moll, V. Krozer, Thomas Kurin, A. Nuber, O. Bagemiel, Stefan Krause, V. Issakov
The present work describes the simulation procedure to determine an optimal sensor placement for RadCom (radar and communication) sensors operating in the frequency band from 57-63 GHz inside a wind turbine blade. Optimal placement means a full penetration and coverage of the blade as well as a communication path from every node to the blade’s root can be achieved. Furthermore, triple coverage is necessary to allow the localization of structural changes in the blade and its surface, such as ice aggretion. The sensors are partly applied to the surface and partly embedded in the core material of the rotor blade. In this way the blade can be monitored during the entire operation for structural health monitoring (SHM) purposes. The simulations take into account the transmission of waves, refraction, dispersion in the material and are based on material data obtained from measurements of rotor blade materials, as well as antenna data. The resulting sensor distribution is the basis for a prototype design of a 30 m long blade with embedded sensors for full-scale SHM testing. Since embedded sensors are not accessible after completion of the manufacturing process, the simulation results are key to the experiments success.
{"title":"OPTIMIZED PLACEMENT FOR EMBEDDED RADAR AND COMMUNICATION SENSORS IN WIND TURBINE BLADES","authors":"J. Simon, J. Moll, V. Krozer, Thomas Kurin, A. Nuber, O. Bagemiel, Stefan Krause, V. Issakov","doi":"10.12783/shm2021/36255","DOIUrl":"https://doi.org/10.12783/shm2021/36255","url":null,"abstract":"The present work describes the simulation procedure to determine an optimal sensor placement for RadCom (radar and communication) sensors operating in the frequency band from 57-63 GHz inside a wind turbine blade. Optimal placement means a full penetration and coverage of the blade as well as a communication path from every node to the blade’s root can be achieved. Furthermore, triple coverage is necessary to allow the localization of structural changes in the blade and its surface, such as ice aggretion. The sensors are partly applied to the surface and partly embedded in the core material of the rotor blade. In this way the blade can be monitored during the entire operation for structural health monitoring (SHM) purposes. The simulations take into account the transmission of waves, refraction, dispersion in the material and are based on material data obtained from measurements of rotor blade materials, as well as antenna data. The resulting sensor distribution is the basis for a prototype design of a 30 m long blade with embedded sensors for full-scale SHM testing. Since embedded sensors are not accessible after completion of the manufacturing process, the simulation results are key to the experiments success.","PeriodicalId":180083,"journal":{"name":"Proceedings of the 13th International Workshop on Structural Health Monitoring","volume":"14 12","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-03-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121008653","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
One challenge in establishing an effective structural health monitoring (SHM) system is the impact of environmental variability on damage identification. It is therefore, advantageous to consider any environmental effects when conducting sensor placement optimisation (SPO). One approach to this problem is to check the robustness of SPO technique to environmental variations and consider whether it is necessary to take account of these environmental factors as part of the optimisation process. This paper will study the robustness of an SPO method to variations in the ambient temperature of the structure. Two kinds of data, including the mode shapes and the Mahalanobis squared-distance (MSD), from tests on a glider wing structure are used as features for SPO separately. This structure was set up and tested in different health states across a series of controlled temperatures. The results show that the SPO results obtained via the mode shapes are robust to the temperature variation, while the SPO results corresponding to MSD are sensitive to temperature changes.
{"title":"ON ROBUSTNESS OF OPTIMAL SENSOR PLACEMENT TO ENVIRONMENTAL VARIATION FOR SHM","authors":"Tingna Wang, D. Wagg, R. Barthorpe, K. Worden","doi":"10.12783/shm2021/36350","DOIUrl":"https://doi.org/10.12783/shm2021/36350","url":null,"abstract":"One challenge in establishing an effective structural health monitoring (SHM) system is the impact of environmental variability on damage identification. It is therefore, advantageous to consider any environmental effects when conducting sensor placement optimisation (SPO). One approach to this problem is to check the robustness of SPO technique to environmental variations and consider whether it is necessary to take account of these environmental factors as part of the optimisation process. This paper will study the robustness of an SPO method to variations in the ambient temperature of the structure. Two kinds of data, including the mode shapes and the Mahalanobis squared-distance (MSD), from tests on a glider wing structure are used as features for SPO separately. This structure was set up and tested in different health states across a series of controlled temperatures. The results show that the SPO results obtained via the mode shapes are robust to the temperature variation, while the SPO results corresponding to MSD are sensitive to temperature changes.","PeriodicalId":180083,"journal":{"name":"Proceedings of the 13th International Workshop on Structural Health Monitoring","volume":"65 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-03-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121161511","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
G. Loubet, A. Sidibe, A. Takacs, J. Balayssac, D. Dragomirescu
This paper presents a cyber-physical system based on a wireless sensor network dedicated to structural health monitoring of reinforced concretes throughout their lifetime. This cyber-physical system is intended to implement a communicating reinforced concrete. Two types of nodes compose this WSN. The sensing node is fully wireless, can measure various parameters (such as temperature, relative humidity, mechanical strain, or resistivity), is battery-free, and is wirelessly and remotely powered and controlled via a radiative electromagnetic power transfer system by the second type of nodes, the communicating node. The communicating node connect the WSN to the digital world.
{"title":"BATTERY-FREE STRUCTURAL HEALTH MONITORING SYSTEM FOR CONCRETE STRUCTURES","authors":"G. Loubet, A. Sidibe, A. Takacs, J. Balayssac, D. Dragomirescu","doi":"10.12783/shm2021/36246","DOIUrl":"https://doi.org/10.12783/shm2021/36246","url":null,"abstract":"This paper presents a cyber-physical system based on a wireless sensor network dedicated to structural health monitoring of reinforced concretes throughout their lifetime. This cyber-physical system is intended to implement a communicating reinforced concrete. Two types of nodes compose this WSN. The sensing node is fully wireless, can measure various parameters (such as temperature, relative humidity, mechanical strain, or resistivity), is battery-free, and is wirelessly and remotely powered and controlled via a radiative electromagnetic power transfer system by the second type of nodes, the communicating node. The communicating node connect the WSN to the digital world.","PeriodicalId":180083,"journal":{"name":"Proceedings of the 13th International Workshop on Structural Health Monitoring","volume":"73 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-03-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121340145","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Recently, extensive studies have been performed for crack detection and segmentation using deep learning and computer vision techniques to accomplish autonomous bridge inspection. These deep network models are frequently trained with a large volume of parameters to ensure good performance. However, the robust applications under real-world situations of actual bridge inspection still face significant challenges. For example, false-positive recognitions of complex background disturbances excluded in the training sets are inevitable to exist. Besides, the real-time requirement for deploying large-volume deep networks in edge computing equipment is still challenging to achieve. This study establishes a lightweight semantic segmentation model for complex concrete crack segmentation in actual bridge inspection. First, the DeepLabv3+ model is adopted as the baseline, and the backbone module is replaced by MobileNetV2 instead of ResNet101. Second, the depthwise separable convolution, atrous convolution pyramid, and inverted residual modules are utilized to reduce convolutional parameters, expand receptive fields, and alleviate gradient vanishing, respectively. Third, the dataset is enhanced with negative disturbance examples, including straight-line-like structural edges and exposed rebars, to improve the model performance against false positives without additional labeling workload. Original images with different resolutions are first collected from actual bridges, and negative samples are further added to the dataset. A total of 4303 patches in 512 × 512 are generated by a sliding window, where 3443, 430, and 430 are randomly selected for training, validation, and test. Ablation experiments demonstrate the necessity and effectiveness of using MobileNetV2 instead of ResNet101 as the backbone and adding negative examples into the dataset. The results show that the mean intersection-over-union (mIoU) for crack segmentation in various real-world scenarios reaches 0.759. The recognition rate of false positives for complex background disturbances is effectually suppressed by introducing straight-line-like structural edges and exposed rebars into the dataset. Furthermore, the average time cost gains a significant reduction of 35.1% using the established lightweight crack segmentation model with only a slight drop on IoU of 0.017.
{"title":"LIGHTWEIGHT DEEP LEARNING MODEL OF SEMANTIC SEGMENTATION FOR COMPLEX CONCRETE CRACKS IN ACTUAL BRIDGE INSPECTION","authors":"Yang Xu, Yunlei Fan, Weidong Qiao, Hui Li","doi":"10.12783/shm2021/36273","DOIUrl":"https://doi.org/10.12783/shm2021/36273","url":null,"abstract":"Recently, extensive studies have been performed for crack detection and segmentation using deep learning and computer vision techniques to accomplish autonomous bridge inspection. These deep network models are frequently trained with a large volume of parameters to ensure good performance. However, the robust applications under real-world situations of actual bridge inspection still face significant challenges. For example, false-positive recognitions of complex background disturbances excluded in the training sets are inevitable to exist. Besides, the real-time requirement for deploying large-volume deep networks in edge computing equipment is still challenging to achieve. This study establishes a lightweight semantic segmentation model for complex concrete crack segmentation in actual bridge inspection. First, the DeepLabv3+ model is adopted as the baseline, and the backbone module is replaced by MobileNetV2 instead of ResNet101. Second, the depthwise separable convolution, atrous convolution pyramid, and inverted residual modules are utilized to reduce convolutional parameters, expand receptive fields, and alleviate gradient vanishing, respectively. Third, the dataset is enhanced with negative disturbance examples, including straight-line-like structural edges and exposed rebars, to improve the model performance against false positives without additional labeling workload. Original images with different resolutions are first collected from actual bridges, and negative samples are further added to the dataset. A total of 4303 patches in 512 × 512 are generated by a sliding window, where 3443, 430, and 430 are randomly selected for training, validation, and test. Ablation experiments demonstrate the necessity and effectiveness of using MobileNetV2 instead of ResNet101 as the backbone and adding negative examples into the dataset. The results show that the mean intersection-over-union (mIoU) for crack segmentation in various real-world scenarios reaches 0.759. The recognition rate of false positives for complex background disturbances is effectually suppressed by introducing straight-line-like structural edges and exposed rebars into the dataset. Furthermore, the average time cost gains a significant reduction of 35.1% using the established lightweight crack segmentation model with only a slight drop on IoU of 0.017.","PeriodicalId":180083,"journal":{"name":"Proceedings of the 13th International Workshop on Structural Health Monitoring","volume":"36 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-03-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122940210","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Wind energy generated through wind turbines is a critical contributor towards realizing a renewable energy economy. Maximizing the efficiency of power generation from wind turbines is required to meet target generation capacities. The leading edge of wind turbine’s blade is designed such that its smooth, aerodynamic surface will produce maximum power given the size specification of the turbine. As these blades age, they typically suffer from leading edge erosion, or LEE, which is the gradual erosion of the blade’s leading edge. Not only does LEE shorten a blade’s lifespan, but it also negatively affects performance, reducing annual energy production. Application of wind blade protection tape is a frequently used solution for LEE on the damaged area. Tape application requires a crew of technicians with a lift and is considered high-risk where one mistake can lead to fatal injury. The focus of this paper is to present a method of automating the wind blade protection tape application by using a UAV with an endeffector. Specifically, the end-effector is an automatic taping mechanism that will apply wind blade protection tape to the damaged area. The paper discusses the overall robotic arm design, topology optimization, and hardware components used to create the operation for the automatic taping mechanism. The end-effector is designed to dispense tape, extend to create contact to the surface, and cut to finish the application. The rest of the arm is used for the motion of tape application vertically starting from the bottom of the damaged region. The process is completed once the damaged area has been covered with protection tape and tape has been cut. The developed end-effector demonstrates the effectiveness of the taping operation, ultimately conserving the wind blade’s lifespan and decreasing the risk of human injury.
{"title":"DESIGN OF A ROBOTIC TAPING MECHANISM FOR UAV-BASED WIND TURBINE BLADE MAINTENANCE","authors":"Joshua Genova, Lakshay Gupta, V. Hoskere","doi":"10.12783/shm2021/36337","DOIUrl":"https://doi.org/10.12783/shm2021/36337","url":null,"abstract":"Wind energy generated through wind turbines is a critical contributor towards realizing a renewable energy economy. Maximizing the efficiency of power generation from wind turbines is required to meet target generation capacities. The leading edge of wind turbine’s blade is designed such that its smooth, aerodynamic surface will produce maximum power given the size specification of the turbine. As these blades age, they typically suffer from leading edge erosion, or LEE, which is the gradual erosion of the blade’s leading edge. Not only does LEE shorten a blade’s lifespan, but it also negatively affects performance, reducing annual energy production. Application of wind blade protection tape is a frequently used solution for LEE on the damaged area. Tape application requires a crew of technicians with a lift and is considered high-risk where one mistake can lead to fatal injury. The focus of this paper is to present a method of automating the wind blade protection tape application by using a UAV with an endeffector. Specifically, the end-effector is an automatic taping mechanism that will apply wind blade protection tape to the damaged area. The paper discusses the overall robotic arm design, topology optimization, and hardware components used to create the operation for the automatic taping mechanism. The end-effector is designed to dispense tape, extend to create contact to the surface, and cut to finish the application. The rest of the arm is used for the motion of tape application vertically starting from the bottom of the damaged region. The process is completed once the damaged area has been covered with protection tape and tape has been cut. The developed end-effector demonstrates the effectiveness of the taping operation, ultimately conserving the wind blade’s lifespan and decreasing the risk of human injury.","PeriodicalId":180083,"journal":{"name":"Proceedings of the 13th International Workshop on Structural Health Monitoring","volume":"40 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-03-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121545809","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
A novel approach is presented to conduct data-driven condition assessment in nuclear safety systems with the aid of deep learning. With the resurgence of nuclear energy due to the ever-increasing demand for electricity and carbon free power generation, ensuring safe operations at nuclear facilities is important. Nuclear safety systems, such as equipment-piping, undergo aging and subsequent degradation due to flow-accelerated erosion and corrosion. Conventional non-destructive techniques implemented during plant outages can take weeks and months to scan all the systems in their entirety. Continuous condition monitoring of such systems would result in lowering the maintenance costs along with extending the operating lifetime for a nuclear power plant. Additionally, the proposed framework should be able to detect minor degradation caused due to aging of nuclear facilities. Uncertainty in the degradation severity levels is also incorporated in the design of the condition assessment methodology. In this paper, the use of artificial intelligence (AI) algorithms as well as vibration-based health monitoring for degradation detection has been demonstrated. A simple equipment-piping system subjected to an external hazard, such as an earthquake, is selected as an application case study. A proof-of-concept is presented wherein the proposed framework utilizes the data collected from sensors to generate a machine learning data repository, demonstrates pattern recognition and feature extraction, explores the design of an artificial neural network (ANN), and develops a sensor placement strategy. The effectiveness of the proposed framework is demonstrated on a realistic primary safety system of a two-loop reactor plant. It is shown that the proposed post-hazard condition monitoring framework is able to detect degraded locations along with the severity levels with high degree of accuracy.
{"title":"DEEP LEARNING FRAMEWORK FOR POST-HAZARD CONDITION MONITORING OF NUCLEAR SAFETY SYSTEMS","authors":"Kaur Sandhu, Saran SRIKANTH BODDA, Abhinav Gupta","doi":"10.12783/shm2021/36253","DOIUrl":"https://doi.org/10.12783/shm2021/36253","url":null,"abstract":"A novel approach is presented to conduct data-driven condition assessment in nuclear safety systems with the aid of deep learning. With the resurgence of nuclear energy due to the ever-increasing demand for electricity and carbon free power generation, ensuring safe operations at nuclear facilities is important. Nuclear safety systems, such as equipment-piping, undergo aging and subsequent degradation due to flow-accelerated erosion and corrosion. Conventional non-destructive techniques implemented during plant outages can take weeks and months to scan all the systems in their entirety. Continuous condition monitoring of such systems would result in lowering the maintenance costs along with extending the operating lifetime for a nuclear power plant. Additionally, the proposed framework should be able to detect minor degradation caused due to aging of nuclear facilities. Uncertainty in the degradation severity levels is also incorporated in the design of the condition assessment methodology. In this paper, the use of artificial intelligence (AI) algorithms as well as vibration-based health monitoring for degradation detection has been demonstrated. A simple equipment-piping system subjected to an external hazard, such as an earthquake, is selected as an application case study. A proof-of-concept is presented wherein the proposed framework utilizes the data collected from sensors to generate a machine learning data repository, demonstrates pattern recognition and feature extraction, explores the design of an artificial neural network (ANN), and develops a sensor placement strategy. The effectiveness of the proposed framework is demonstrated on a realistic primary safety system of a two-loop reactor plant. It is shown that the proposed post-hazard condition monitoring framework is able to detect degraded locations along with the severity levels with high degree of accuracy.","PeriodicalId":180083,"journal":{"name":"Proceedings of the 13th International Workshop on Structural Health Monitoring","volume":"17 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-03-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122790711","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
R. Watson, Taiyi Zhao, Dayi Zhang, Mina Kamel, C. Macleod, G. Dobie, G. Bolton, Antoine Joly, S. G. Pierce, Juan Nieto
Use of Unmanned Aerial Vehicles (UAVs) for Structural Health Monitoring (SHM) has become commonplace across civil and energy generation applications with hazardous or time-consuming inspection processes. Expanding upon surface screening offered by non-contact remote visual inspection UAVs, systems are now beginning to incorporate contact-based Non-Destructive Evaluation (NDE) transducers to detect and monitor incipient sub-surface flaws. However, challenges to environmental interaction using conventional multirotor platform dynamics amid aerodynamic disturbances have frustrated efforts for stable and repeatable sensor placement. Herein, two distinct UAV systems are evaluated as a means to overcome these challenges. The first utilizes vectored thrust with a tri-copter layout. It may dynamically reorient dual-axis tilting propellers to directly effect interaction force and deploy drycoupled ultrasonic thickness measurement across omnidirectional targets. In static point and rolling scan measurement, laboratory tests demonstrate mean absolute error below 0.1 mm and 0.3 mm, respectively. The second UAV uses rigidly affixed multidirectional propellers to reverse and redirect its net thrust. Landing atop cylindrical structures it may crawl around their circumference, supporting itself without magnetic or vacuum adhesion. Arbitrary static position is maintained to within a mean deviation of 0.7 mm. Lastly, comparative discussion of each system informs strategies for further development of contact-based aerial SHM and its adoption to industrial practice.
{"title":"TECHNIQUES FOR CONTACT-BASED STRUCTURAL HEALTH MONITORING WITH MULTIROTOR UNMANNED AERIAL VEHICLES","authors":"R. Watson, Taiyi Zhao, Dayi Zhang, Mina Kamel, C. Macleod, G. Dobie, G. Bolton, Antoine Joly, S. G. Pierce, Juan Nieto","doi":"10.12783/shm2021/36236","DOIUrl":"https://doi.org/10.12783/shm2021/36236","url":null,"abstract":"Use of Unmanned Aerial Vehicles (UAVs) for Structural Health Monitoring (SHM) has become commonplace across civil and energy generation applications with hazardous or time-consuming inspection processes. Expanding upon surface screening offered by non-contact remote visual inspection UAVs, systems are now beginning to incorporate contact-based Non-Destructive Evaluation (NDE) transducers to detect and monitor incipient sub-surface flaws. However, challenges to environmental interaction using conventional multirotor platform dynamics amid aerodynamic disturbances have frustrated efforts for stable and repeatable sensor placement. Herein, two distinct UAV systems are evaluated as a means to overcome these challenges. The first utilizes vectored thrust with a tri-copter layout. It may dynamically reorient dual-axis tilting propellers to directly effect interaction force and deploy drycoupled ultrasonic thickness measurement across omnidirectional targets. In static point and rolling scan measurement, laboratory tests demonstrate mean absolute error below 0.1 mm and 0.3 mm, respectively. The second UAV uses rigidly affixed multidirectional propellers to reverse and redirect its net thrust. Landing atop cylindrical structures it may crawl around their circumference, supporting itself without magnetic or vacuum adhesion. Arbitrary static position is maintained to within a mean deviation of 0.7 mm. Lastly, comparative discussion of each system informs strategies for further development of contact-based aerial SHM and its adoption to industrial practice.","PeriodicalId":180083,"journal":{"name":"Proceedings of the 13th International Workshop on Structural Health Monitoring","volume":"214 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-03-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131536966","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：