Assessment of horizontal bore drains performance in brown coal mines in the Latrobe Valley
- Authors: Perdigao, Cristhiana
- Date: 2021
- Type: Text , Thesis , Masters
- Full Text:
- Description: Horizontal bores are essential infrastructures for maintaining the stability of open-pit mine batters. The infiltration of water from large surface catchments during rain events and induced deformation caused by mining activities can cause the build-up of pore water pressures in mine batters, potentially leading to catastrophic slope failures. A field investigation unit containing a camera has been developed to survey long (>300m) horizontal bores. Features observed using the camera along the profile of horizontal bores are discussed. Water flow was quantified by flow meters. X-Ray Diffraction (XRD) was undertaken to investigate the water precipitates within the selected bores. Water flow temperature was recorded to test the hypothesis of a possibility to indicate whether a borehole was draining from the saturated zone or from the surface water through its temperature. The investigations have been conducted to determine the cause of change in the efficiency of horizontal boreholes and find a reliable measure to assess longevity and performance of horizontal drains. Bore efficiency has been defined as the bore functioning as a preferential path for water within the batter to be drained out to reduce the saturated zone and associated pore water pressures within the batter. The results suggest blockages and fractures inside the bores can be considered the leading cause of the change in the efficiency of a bore. Blockages occur because of sediment accumulation and because of coal chunks from internal wall collapses. Internal fractures affect efficiency when they become the water preferred path; thus, retaining water flowing within the batter. The bore’s longevity is considered the period of the bore is considered effective. Water flow measurement is suggested as a reliable measure to assess bores’ longevity.
- Description: Masters by Research
- Authors: Perdigao, Cristhiana
- Date: 2021
- Type: Text , Thesis , Masters
- Full Text:
- Description: Horizontal bores are essential infrastructures for maintaining the stability of open-pit mine batters. The infiltration of water from large surface catchments during rain events and induced deformation caused by mining activities can cause the build-up of pore water pressures in mine batters, potentially leading to catastrophic slope failures. A field investigation unit containing a camera has been developed to survey long (>300m) horizontal bores. Features observed using the camera along the profile of horizontal bores are discussed. Water flow was quantified by flow meters. X-Ray Diffraction (XRD) was undertaken to investigate the water precipitates within the selected bores. Water flow temperature was recorded to test the hypothesis of a possibility to indicate whether a borehole was draining from the saturated zone or from the surface water through its temperature. The investigations have been conducted to determine the cause of change in the efficiency of horizontal boreholes and find a reliable measure to assess longevity and performance of horizontal drains. Bore efficiency has been defined as the bore functioning as a preferential path for water within the batter to be drained out to reduce the saturated zone and associated pore water pressures within the batter. The results suggest blockages and fractures inside the bores can be considered the leading cause of the change in the efficiency of a bore. Blockages occur because of sediment accumulation and because of coal chunks from internal wall collapses. Internal fractures affect efficiency when they become the water preferred path; thus, retaining water flowing within the batter. The bore’s longevity is considered the period of the bore is considered effective. Water flow measurement is suggested as a reliable measure to assess bores’ longevity.
- Description: Masters by Research
Real-time concrete crack detection and instance segmentation using deep transfer learning
- Piyathilaka, Lasitha, Preethichandra, Daluwathu, Izhar, Umer, Appuhamillage, Gayan
- Authors: Piyathilaka, Lasitha , Preethichandra, Daluwathu , Izhar, Umer , Appuhamillage, Gayan
- Date: 2020
- Type: Text , Journal article
- Relation: Engineering Proceedings Vol. 2, no. 1 (2020), p.
- Full Text:
- Reviewed:
- Description: Cracks on concrete infrastructure are one of the early indications of structural degradation which needs to be identified early as possible to carry out early preventive measures to avoid further damage. In this paper, we propose to use YOLACT: a real-time instance segmentation algorithm for automatic concrete crack detection. This deep learning algorithm is used with transfer learning to train the YOLACT network to identify and localize cracks with their corresponding masks which can be used to identify each crack instance. The transfer learning techniques allowed us to train the network on a relatively small dataset of 500 crack images. To train the YOLACT network, we created a dataset with ground-truth masks from images collected from publicly available datasets. We evaluated the trained YOLACT model for concrete crack detection with ResNet-50 and ResNet-101 backbone architectures for both precision and speed of detection. The trained model achieved high mAP results with real-time frame rates when tested on concrete crack images on a single GPU. The YOLACT algorithm was able to correctly segment multiple cracks with individual instance level masks with high localization accuracy.
- Authors: Piyathilaka, Lasitha , Preethichandra, Daluwathu , Izhar, Umer , Appuhamillage, Gayan
- Date: 2020
- Type: Text , Journal article
- Relation: Engineering Proceedings Vol. 2, no. 1 (2020), p.
- Full Text:
- Reviewed:
- Description: Cracks on concrete infrastructure are one of the early indications of structural degradation which needs to be identified early as possible to carry out early preventive measures to avoid further damage. In this paper, we propose to use YOLACT: a real-time instance segmentation algorithm for automatic concrete crack detection. This deep learning algorithm is used with transfer learning to train the YOLACT network to identify and localize cracks with their corresponding masks which can be used to identify each crack instance. The transfer learning techniques allowed us to train the network on a relatively small dataset of 500 crack images. To train the YOLACT network, we created a dataset with ground-truth masks from images collected from publicly available datasets. We evaluated the trained YOLACT model for concrete crack detection with ResNet-50 and ResNet-101 backbone architectures for both precision and speed of detection. The trained model achieved high mAP results with real-time frame rates when tested on concrete crack images on a single GPU. The YOLACT algorithm was able to correctly segment multiple cracks with individual instance level masks with high localization accuracy.
- «
- ‹
- 1
- ›
- »