Explore the vast range of titles available.

Sewer pipeline defect detection is a critical task for ensuring the normal operation of urban infrastructure. However, the sewer environment often presents challenges such as multi-scale defects, complex backgrounds, lighting changes, and diverse defect morphologies. To address these issues, this paper proposes a lightweight cross-scale feature fusion model based on YOLOv8. First, the C2f module in the backbone network is replaced with the C2f-FAM module to enhance multi-scale feature extraction capabilities. Second, the HS-BiFPN module is adopted to replace the original structure, leveraging cross-layer semantic fusion and feature re-weighting mechanisms to improve the model’s ability to distinguish complex backgrounds and diverse defect morphologies. Finally, DySample is introduced to replace traditional sampling operations, enhancing the model’s ability to capture details in complex environments. This study uses the Sewer-ML dataset to train and evaluate the model, selecting 1,158 images containing six types of typical defects (CK, PL, SG, SL, TL, ZW), and expanding the dataset to 1,952 images through data augmentation. Experimental results show that compared to the YOLOv8n model, the improved model achieves a 3.8% increase in mAP, while reducing the number of parameters by 35%, floating-point operations by 21%, and model size by 33%. By improving detection accuracy while achieving model lightweighting, the model demonstrates potential for application in pipeline defect detection.

Long-term corrosion and external disturbances can lead to defects in sewer pipes, which threaten important parts of urban infrastructure. The automatic defect detection algorithm based on closed-circuit televisions (CCTV) has gradually matured using supervised deep learning. However, there are different types and sizes of sewer pipe defects, and relying on human inspection to detect defects is time-consuming and subjective. Therefore, a few-shot, accurate and automatic method for sewer pipe defect with localization and fine-grained classification is needed. Thus, this study constructs a few-shot image-level dataset of 15 categories using the sewer dataset ML-Sewer and then presents a collaborative localization network based on weakly supervised learning to automatically classify and detect defects. Specifically, an attention refinement module (ARM) is designed to obtain classification results and high-level semantic features. Furthermore, considering the correlation between target regions and the extraction of target edge information, we designed a collaborative localization module (CLM) consisting of two branches. Then, to ensure that the network focuses on the complete target area, this study applies an image iteration module (IIM). Finally, the results of the two branches in the CLM are fused to acquire target localization. The experimental results show that the proposed model exhibits favorable performance in detecting sewer pipe defects. The proposed method exhibits prediction classification accuracy that reaches 69.76 % and a positioning accuracy rate that reaches 65.32 % , which is higher than the performances of other weakly supervised detection models in sewer pipe defect detection.

Exfiltration from sanitary sewers has been researched for many years because of its potential impact on shallow groundwater or surface water, but measurements of exfiltration in situ are rare. Most previous measurements of sanitary sewer exfiltration have been done in the laboratory, in the field using natural, chemical or pharmaceutical tracers or modeled. Relatively few studies have employed physical measurements of volume loss in field settings. Here, we design, test, and apply at a watershed scale, a new methodology for measuring volume loss from sanitary sewer pipes that are currently in use and under typical operating conditions (i.e., not pressurized). The measurement system works by: (1) isolating a section of sanitary sewer between maintenance holes using a sewer bypass or equivalent, (2) introducing roughly 4,200 L of water at a controlled rate into the upstream inspection hole so that pipes remain one-third to one-half full, (3) using vacuum pumps to recover the introduced water at the downstream inspection hole, then (4) measuring differences in the volume from what was pumped into the inspection hole to what was recovered. This process is repeated up to six times to achieve a sensitivity of 0.95 L per experimental pipe segment. This technique was applied to 23 pipe segments of various ages and materials of construction that were selected to be a representative sample of the pipes throughout San Diego. Collectively, these pipes averaged averaged 3.78 × 10 −2  L/s-km exfiltration rates (95%CI: 4.96 × 10 −2 , 2.60 × 10 −2 ). Two of the pipe segments were infiltrating groundwater. Six pipe segments were not statistically different from zero (i.e., no exfiltration). There was no statistical difference between pipe segments of differing ages ( p = 0.5) or materials of construction ( p = 0.3). This study represents an initial effort at measuring exfiltration from in situ sanitary pipes. Future applications of this methodology should focus on method optimization, measurements at additional locations, and expanding measurements to collect data from additional types of pipe to better understand the geographic portability of the method and the relationship between exfiltration rates, pipe material, and pipe age.

Wastewater infrastructure systems deteriorate over time due to a combination of aging, physical, and chemical factors, among others. Failure of these critical structures cause social, environmental, and economic impacts. To avoid such problems, infrastructure condition assessment methodologies are developing to maintain sewer pipe network at desired condition. However, currently utility managers and other authorities have challenges when addressing appropriate intervals for inspection of sewer pipelines. Frequent inspection of sewer network is not cost-effective due to limited time and high cost of assessment technologies and large inventory of pipes. Therefore, it would be more beneficial to first predict critical sewers most likely to fail and then perform inspection to maximize rehabilitation or renewal projects. Sewer condition prediction models are developed to provide a framework to forecast future condition of pipes and to schedule inspection frequencies. The objective of this study is to present a state-of-the-art review on progress acquired over years in development of statistical condition prediction models for sewer pipes. Published papers for prediction models over a period from 2001 through 2019 are identified. The literature review suggests that deterioration models are capable to predict future condition of sewer pipes and they can be used in industry to improve the inspection timeline and maintenance planning. A comparison between logistic regression models, Markov Chain models, and linear regression models are provided in this paper. Artificial intelligence techniques can further improve higher accuracy and reduce uncertainty in current condition prediction models.

Defects in sewer pipes are the leading cause of soil erosion and road collapse, but the effect of the defect location is unclear. Laboratory model tests were conducted to investigate the erosion and loss of sandy soil adjacent to a defective sewer pipe. Five defect locations around a round pipe were examined with a physical model in the laboratory. Particle image velocimetry (PIV) shows that the erosion in the soil around the pipe is limited to a narrow region above the defect in the pipe, and the defect location affects the soil erosion and flow rates. The eroded soil region did not change when the defect was located in the top half of the pipe, whereas the eroded region expanded when the defect is located in the bottom half of the pipe. The eroded region was more than two times wider for a pipe with a defect at the invert than a defect at the crown. Flow rate analysis demonstrated that the soil and water flow rates are reduced when the location of the defect is changed from the crown to the invert, and the time it takes during erosion is increased. The ratio between the soil and water flow rates is constant until the erosion cavity reaches the defect. The soil/water flow rate ratio is not affected by the water level or the defect location and it only depends on the ratio between the defect size and the particle size. The ratio between the soil/water flow rates and the defect/particle sizes can be approximated by an exponential relationship which is consistent with experimental observations and predictions using the free-fall arch theory. An analytical model is proposed to estimate the rate of soil loss for cohesionless soil, and the soil flow rate can be calculated using an erosion cavity model based on the law of mass conservation. A kinetic model is extended to predict the soil velocity during erosion which is verified with laboratory measurements.

A coupled hydro-mechanical erosion model is presented that is used for studying soil piping and erosion void formation under practical, in-situ conditions. The continuum model treats the soil as a two-phase porous medium composed of a solid phase and a liquid phase, and accounts for its elasto-plastic deformation behaviour caused by frictional sliding and granular compaction. The kinetic law characterizing the erosion process is assumed to have a similar form as the type of threshold law typically used in interfacial erosion models. The numerical implementation of the coupled hydro-mechanical model is based on an incremental-iterative, staggered update scheme. A one-dimensional poro-elastic benchmark problem is used to study the basic features of the hydro-mechanical erosion model and validate its numerical implementation. This problem is further used to reveal the interplay between soil erosion and soil consolidation processes that occur under transient hydro-mechanical conditions, thereby identifying characteristic time scales of these processes for a sandy material. Subsequently, two practical case studies are considered that relate to a sewer system embedded in a sandy soil structure. The first case study treats soil piping caused by suffusion near a sewer system subjected to natural ground water flow, and the second case study considers the formation of a suffosion erosion void under strong ground water flow near a defect sewer pipe. The effects on the erosion profile and the soil deformation behaviour by plasticity phenomena are elucidated by comparing the computational results to those obtained by modelling the constitutive behaviour of the granular material as elastic. The results of this comparison study point out the importance of including an advanced elasto-plastic soil model in the numerical simulation of erosion-driven ground surface deformations and the consequent failure behaviour. The numerical analyses further illustrate that the model realistically predicts the size, location, and characteristic time scale of the generated soil piping and void erosion profiles. Hence, the modelling results may support the early detection of in-situ subsurface erosion phenomena from recorded ground surface deformations. Additionally, the computed erosion profiles may serve as input for a detailed analysis of the local, residual bearing capacity and stress redistribution of buried concrete pipe systems.

Microbial-induced concrete corrosion (MICC) has been recognized as one of the main factors causing damage to urban sewer pipelines. However, little is known about the neutralization and degradation of the concrete pipe during the early-stage MICC, which is a key point for the trenchless protection of sewer pipelines. In the present study, the corrosion behaviors of concrete pipe were simulated in a pilot-scale sewer system for 180 days and correlated to the change of microbial communities. The results revealed the post-corrosion characteristics in different spatial locations of the sewer during the early-stage MICC. Apart from the biogenic sulfuric acid attack and gypsum formation at the upper part (UP), the bottom part (BP) of the concrete pipe suffered severe degradation due to volatile fatty acids (VFAs) generated. Under the action of microbial metabolism, the decomposition of hydrates and pore coarsening occurred, resulting in decreased mechanical strength. In terms of microbiology, the dominating functional bacteria were fermentation bacteria (FB), such as Macellibacteroid es, Trichococcus , Longilinea , etc. FB played a major role in the production of VFAs, which would create suitable conditions for the subsequent development of microorganisms. The early fermentation processes were key factors contributing to concrete pipe corrosion, especially at BP. The relative abundance of sulfate-reducing bacteria ( Desulfovibrio ) and methanogenic archaea increased with exposure to age. The findings can provide a theoretical basis for the protection of urban concrete sewer pipelines.

As a basic infrastructure, sewers play an important role in the innards of every city and town to remove unsanitary water from all kinds of livable and functional spaces. Sewer pipe failures (SPFs) are unwanted and unsafe in many ways, as the disturbance that they cause is undeniable. Sewer pipes meet manholes frequently, unlike water distribution systems, as in sewers, water movement is due to gravity and manholes are needed in every intersection as well as through pipe length. Many studies have been focused on sewer pipe failures and so on, but few investigations have been done to show the effect of manhole proximity on pipe failure. Predicting and localizing the sewer pipe failures is affected by different parameters of sewer pipe properties, such as material, age, slope, and depth of the sewer pipes. This study investigates the applicability of a support vector machine (SVM), a supervised machine learning (ML) algorithm, for the development of a prediction model to predict sewer pipe failures and the effects of manhole proximity. The results show that SVM with an accuracy of 84% can properly approximate the manhole effects on sewer pipe failures.

An image-recognition-based diagnosis system of pipe defect types was established. 1043 practical pipe images were gathered by CCTV robot in a southern Chinese city. The overall accuracy of the system is 84% and the highest accuracy is 99.3%. The accuracy shows positive correlation to the number of training samples. Closed circuit television (CCTV) systems are widely used to inspect sewer pipe conditions. During the diagnosis process, the manual diagnosis of defects is time consuming, labor intensive and error prone. To assist inspectors in diagnosing sewer pipe defects on CCTV inspection images, this paper presents an image recognition algorithm that applies features extraction and machine learning approaches. An algorithm of image recognition techniques, including Hu invariant moment, texture features, lateral Fourier transform and Daubechies (DBn) wavelet transform, was used to describe the features of defects, and support vector machines were used to classify sewer pipe defects. According to the inspection results, seven defects were defined; the diagnostic system was applied to a sewer pipe system in a southern city of China, and 28,760 m of sewer pipes were inspected. The results revealed that the classification accuracies of the different defects ranged from 51.6% to 99.3%. The overall accuracy reached 84.1%. The diagnosing accuracy depended on the number of the training samples, and four fitting curves were applied to fit the data. According to this paper, the logarithmic fitting curve presents the highest coefficient of determination of 0.882, and more than 200 images need to be used for training samples to guarantee the accuracy higher than 85%.

The distributed fiber optic sensing technique (DFOST) provides an effective solution for long-term deformation monitoring of sewer pipelines under operation conditions due to occurrence of cracks. However, optical fibers attached on the pipeline wall might fail to capture all cracks as they are usually highly spatially scattered, resulting in misjudgment of pipe structural conditions. To address this challenge, this paper presents a novel deployment approach of distributed fiber optic monitoring system for pipeline cracks. First, the working principle of optical frequency domain reflectometry (OFDR) is explained. Second, the production of concrete pipe specimens with different optical fiber sensors deployment schemes is shown in detail. Finally, a comprehensive analyse of the results of visual inspection and fibre optic monitoring is completed. To evaluate the performance of this approach, experimental data from various cases were considered. The conclusion of this paper provides theoretical and empirical support for the application of distributed fiber optic sensors for pipeline cracks monitoring.