Blockage Detection Research Articles

Enhancing the accuracy of blockage detection in centrifugal pump using the majority voting classifier on an unbalanced dataset

Blockage in a centrifugal pump can adversely impact its performance. In this study, an experimental facility is developed to simulate three types of blockages (suction, discharge, and simultaneous suction and discharge). To classify these faults and detect their severity, a methodology involving the application of majority voting classifier (MVC) to the pump’s discharge pressure signals is presented. An unbalanced dataset is constructed, where the number of samples for a specific blockage condition decreases with increasing fault severity. Statistical features, entropy features, and entropies meta features are extracted from the signal and ranked using XGBoost and minimum redundancy maximum relevance (MRMR). Subsequently, the optimal features are selected based on the best performing model (Linear Discriminant Analysis) among ten different models. Best four models are selected and ensembled to form MVC. Results show that MVC achieves an accuracy of 89.90% and 88.26% for features selected by XGBoost and MRMR, respectively. Finally, the unbalanced dataset is balanced using synthetic minority oversampling and it is shown that MVC achieves an accuracy of 100% on this balanced dataset for the features selected using both approaches.

A Novel Approach to Detecting Blockages in Sewers and Drains: The Reflected Wave Technique

Blockages in sewers and drains often result in overflows and flooding that cause significant environmental pollution and public health risks, particularly in hospitals, where the consequences can be catastrophic. Due to their low “visibility”, sewers and drains are inherently difficult to monitor and maintain, resulting in a reactive management approach whereby maintenance or repair is carried out only after a system failure has occurred. This paper investigates the feasibility of applying the reflected wave technique, a unique sonar-like monitoring approach capable of identifying changes in the geometry of closed-pipe conduits, as a means of proactive system monitoring. The technique uses a 10 Hz sinusoidal air pressure wave which is transmitted into the drainpipe. When the pressure wave encounters a system boundary, a reflection is generated which alters the measured test pressure response. Analysis of the reflections generated by a changed system boundary, such as the formation of a blockage, can provide information related to the location of that boundary within the system. An experimental setup was developed to simulate a horizontal drain using standard pipework of 100 mm diameter and 70 m length. The technique was able to detect applied blockages with cross-sectional coverage of 30% and 75%, and lengths ranging from 30 mm to 3000 mm. Accuracy was improved when the pressure sensor was positioned closer to the blockage. When the sensor was 3.4 m from the blockage, location estimates were very accurate (−2% to 3% error). At a 14 m distance from the blockage, the error increased to between 4% and 33%. The accuracy of blockage detection and location improved with increasing blockage cross-sectional area and length. Overall, the reflected wave technique could provide a potentially continuous monitoring solution for blockage detection in sewers and drains.

Acoustic Detection of Pipeline Blockages in Gas Extraction Systems: A Novel Approach

Gas extraction is crucial for coal mine safety, yet pipeline blockages by solid slag and water severely hinder efficiency and pose risks. Traditional detection methods are limited by rapid signal attenuation and noise interference. In this study, an acoustic detection technology is introduced for pipeline blockages, utilizing sensors at potential blockage points to collect sound wave data. Experiments with a scaled pipeline model reveal that slag blockages produce characteristic peaks in the 1200 Hz–2000 Hz range, while water blockages show peaks in the 1 kHz–2 kHz and 3.5 kHz–4.5 kHz bands. The longitudinal blockage intensity and extraction pressure significantly affect the sound pressure levels. A reliable fitting model predicts the blockage intensity based on acoustic signals, achieving high accuracy. This novel method enhances blockage identification, offering a non-invasive, cost-effective solution that improves coal mine safety and efficiency.

A Zero-Shot Learning Approach for Blockage Detection and Identification Based on the Stacking Ensemble Model.

A data-driven approach to defect identification requires many labeled samples for model training. Yet new defects tend to appear during data acquisition cycles, which can lead to a lack of labeled samples of these new defects. Aiming at solving this problem, we proposed a zero-shot pipeline blockage detection and identification method based on stacking ensemble learning. The experimental signals were first decomposed using variational modal decomposition (VMD), and then, the information entropy was calculated for each intrinsic modal function (IMF) component to construct the feature sets. Second, the attribute matrix was established according to the attribute descriptions of the defect categories, and the stacking ensemble attribute learner was used for the attribute learning of defect features. Finally, defect identification was accomplished by comparing the similarity within the attribute matrices. The experimental results show that target defects can be identified even without targeted training samples. The model showed better classification performance on the six sets of experimental data, and the average recognition accuracy of the model for unknown defect categories reached 72.5%.

Transitioning practices of water utilities from reactive to proactive: Leveraging Australian best practices in digital technologies and data analytics

Developments in technology and data analytics bring opportunities for water utilities to proactively adapt their practices to better manage pressing challenges, such as water losses, ageing assets, and urban water scarcity due to climate change. Despite the increase in commercial digital technologies for monitoring urban water, wastewater and stormwater systems, digital transformation in water utilities has been occurring much slower than in other businesses (e.g., the energy sector). Leveraging from Australian best practices, this study aims to facilitate water utility’s digital transformation by establishing a targeted research agenda and a continuous improvement process for the successful implementation, operation, and improvement of smart water technologies. Four case studies on 1) automated and digital water consumption metering, 2) early warning of water mains breaks, 3) early detection of sewerage blockages and spills, and 4) smart rainfall reuse and flood mitigation are used to exemplify the potential benefits and hurdles of adopting digital technologies in water utilities, while highlighting pressing areas needing further research and development. Key research areas identified include: 1) development frameworks to map water utilities capacity and needs for ranking priority areas of investment, 2) improvements in sensor hardware, 3) integration of big, multi-sourced data and cybersecurity, 4) increase in generality of smart data applications for various field conditions (i.e., reduce the need for or level of customization), and 5) data-informed proactive asset management and water pricing. Key barriers to be overcome include 1) lack of internal capacity to develop, operate and integrate smart applications, 2) the risk conservative culture of water utilities when facing innovation, and 3) lack of operation standards among water utilities to promote benchmarking. It is imperative that water utilities actively drive the digital transformation process by perpetuating collaborations with research centres and the private sector, while simultaneously increasing in-house capacity through investments in skilled personnel and training.

Experimental research on the reflection and response characteristics of pressure pulse waves for different gas pipeline blockage materials

AbstractNatural gas pipelines are exposed to complex production and transportation environments, leading to blockages of varying compositions and forms due to various factors. This paper proposes a pressure pulsing wave‐based pipe blockage detection method, and a series of pipeline blockage detection experiments of different materials are carried out, including hydrate blockage, ice blockage, paraffin blockage, and water blockage. Through the blockage detection calculation theory and the reflection and transmission characteristics during pressure wave emission, the blockage position, length, and section blockage rate under different blockage materials are obtained, and the influence of different blockage materials on the reflection and transmission signal characteristics of the pressure pulse wave is analyzed. The results show that the solid blocking material does not noticeably impact the reflectivity and reflective properties of pressure waves, and different cross‐sectional shapes have a noticeable impact on the reflectivity of the pressure wave. The reflected wave waveform of the liquid blockage material is visible but irregular. The average error of blockage location is 0.78%.

Feasibility of using the frequency-domain electromagnetic method to detect pipeline blockages: Experimental study

The normal function of modern society requires a vast number of buried pipelines with differing specifications, which could become blocked to varying degrees during routine operation. Consequently, a major technical municipal engineering problem is how best to accurately, nondestructively, and efficiently locate blockages in underground pipelines. The frequency-domain electromagnetic (FDEM) method has been widely used in geophysical exploration because of its features of nondestructive examination and high efficiency. Considering the application of FDEM to the problem of locating blockages in underground pipelines, numerical simulations and physical experiments are conducted on different pipeline types with different blocked states, different blockage substances, and different working configurations. Results demonstrate that FDEM is most applicable for the detection of blockages in insulated pipelines, especially when the single-ended charging method is used or when the blockage has reasonable impermeability because the induction signal exhibits a particularly obvious abrupt drop at the blockage position. However, owing to the low resistivity of metals and the current propagating to the surrounding ground, which makes it easier to form more conductive paths than blockages, the FDEM displays no obvious change in signal characteristics at the position of a blockage in a metal pipeline, which means that this method has certain challenges for detecting such blockages.

Deep learning for automated trash screen blockage detection using cameras: Actionable information for flood risk management

ABSTRACT Trash screens are used to prevent debris from entering critical parts of rivers. However, debris can accumulate on the screen and generate floods. This makes their monitoring critical both for maintenance and flood modeling purposes (e.g., local forecasts may change because the trash screen is blocked). We developed three novel deep learning methods for trash screen maintenance management consisting of automatically detecting trash screen blockage using cameras: a method based on image classification, a method based on image similarity matching, and a method based on anomaly detection. To facilitate their use by end users, these methods are designed so that they can be directly applied to any new trash screen camera installed by the end users. We have built a new dataset of labeled trash screen images to train and evaluate the efficiency of our methods, in terms of both accuracy and implications for end users. This dataset consists of 80,452 trash screen images from 54 cameras installed by the Environment Agency (UK). This work demonstrates that trash screen blockage detection can be automated using trash screen cameras and deep learning, which could have an impact on both trash screen management and flood modeling.

UOWC Integration with ROVs: A Fiber-Optic Approach for Enhanced Underwater Exploration

Underwater Optical Wireless Communication (UOWC) revolutionizes underwater exploration, leveraging seawater's unique property—reduced absorption of blue-green light. This offers superior advantages over traditional communication methods, providing higher bandwidth and lower latency. This paper explores the integration of UOWC with Remotely Operated Vehicles (ROVs), enhancing data exchange and command capabilities in challenging underwater environments. Seawater's unique properties empower UOWC to outperform acoustic and RF methods, crucial for real-time applications. The study utilizes advanced techniques like Fiber-Optic Communication, High Order Sliding Mode Control, and Hybrid Blockage Detection for ROV integration and control. While the abstract focuses on theory and design, it sets the stage for empirical validations, emphasizing the need for future research to quantify improvements in real-world underwater scenarios

Real-Time sanitary sewer blockage detection system using IoT

Sewer blockages and overflows have significant economic and environmental repercussions on communities. Thus, it is crucial to detect and remove sewer blockages prior to the occurrence of overflows. With the improvement in mobile networks and the development of high-quality and low-power sensors and loggers, wastewater network operators can now adopt monitoring devices enabled by the “Internet of Things” (IoT) technology for real-time monitoring. To this end, this paper studies the current state-of-the-art in sewer blockage management and introduces a novel methodology to monitor sanitary sewer blockages to prevent sanitary sewer overflows (SSO) with the least human interaction. The proposed system incorporates low-power level sensors and 4G telemetry for real-time monitoring of the manhole's sewage level to detect sewer blockages. The blockage detection methodology encompasses a set of decision rules and time series analysis to identify blockage events. The final blockage detection model offers a versatile capability to be implemented on sanitary sewer networks of different types with minimum computational costs.

Spectral-Based Proactive Blockage Detection for Sub-THz Communications

Spectral-Based Proactive Blockage Detection for Sub-THz Communications

Distributed Machine-Learning for Early HARQ Feedback Prediction in Cloud RANs

In this work, we propose novel HARQ prediction schemes for Cloud RANs (C-RANs) that use feedback over a rate-limited feedback channel (2 - 6 bits) from the Remote Radio Heads (RRHs) to predict at the User Equipment (UE) the decoding outcome at the BaseBand Unit (BBU) ahead of actual decoding. In particular, we propose a Dual Autoencoding 2-Stage Gaussian Mixture Model (DA2SGMM) that is trained in an end-to-end fashion over the whole C-RAN setup. Using realistic link-level simulations in the sub-THz band at 100 GHz, we show that the novel DA2SGMM HARQ prediction scheme clearly outperforms all other adapted and state-of-the-art schemes. The DA2SGMM shows a superior performance in terms of blockage detection as well as HARQ prediction in the no-blockage and single-blockage cases. In particular, the DA2SGMM with 4~bit feedback achieves a more than 200 % higher throughput in average compared to its best alternative. Compared to regular HARQ, the DA2SGMM reduces the maximum transmission latency by more than 72.4 %, while maintaining more than 75 % of the throughput in the no-blockage scenario. In the single-blockage scenario, DA2SGMM significantly increases the throughput for most of the evaluated Signal-to-Noise-Ratios (SNRs) compared to regular HARQ.

CNN-Based Blockage Detection and Beamforming Design in Dual-Band Communication Systems

CNN-Based Blockage Detection and Beamforming Design in Dual-Band Communication Systems

Preventive and Predictive CNN Based Solution for Pipeline Leak, Blockage and Corrosion Detection

Using Artificial Intelligence (AI) in Internet of Things (IoT) environments has shown great potential to transform many industries. As oil and gas industries continue to push for digital transformation, safety critical applications such as pipeline leak detection, corrosion and blockage detection could also benefit from the adoption of AI and IoT. In our day-to-day life, we come across pipeline accidents and their after-effects. Structural deterioration of pipes in urban distribution network has presented great challenges to the utilities all over the world. When the deterioration exceeds its resiliency, the pipes leak or burst, leading to significant socio-economic overhead. The harsh operating environment, extreme temperature and weather conditions increases the potential for corrosion induced damages to the pipelines. The extreme conditions where the pipelines are located also make it difficult to rely on human operators to physically monitor these pipelines and respond to observed leaks. We propose a preventive and predictive AI-based solution that can continuously monitor the health of a complex pipeline infrastructure. It is a complete solution where pipelines will be scrutinized for blockage, leakage, corrosion, and defects. We propose a novel approach in which video images from IoT cameras installed across various locations on the pipelines are continuously analyzed and a convolutional neural network model is implemented for detecting leaks from the pipeline. Flow sensors will be mounted on the hardware to measure the fluid flow rate. This approach is proposed to deliver greater benefits in terms of accuracy and efficiency.

AI-Driven High-Precision Model for Blockage Detection in Urban Wastewater Systems

In artificial intelligence (AI), computer vision consists of intelligent models to interpret and recognize the visual world, similar to human vision. This technology relies on a synergy of extensive data and human expertise, meticulously structured to yield accurate results. Tackling the intricate task of locating and resolving blockages within sewer systems is a significant challenge due to their diverse nature and lack of robust technique. This research utilizes the previously introduced “S-BIRD” dataset, a collection of frames depicting sewer blockages, as the foundational training data for a deep neural network model. To enhance the model’s performance and attain optimal results, transfer learning and fine-tuning techniques are strategically implemented on the YOLOv5 architecture, using the corresponding dataset. The outcomes of the trained model exhibit a remarkable accuracy rate in sewer blockage detection, thereby boosting the reliability and efficacy of the associated robotic framework for proficient removal of various blockages. Particularly noteworthy is the achieved mean average precision (mAP) score of 96.30% at a confidence threshold of 0.5, maintaining a consistently high-performance level of 79.20% across Intersection over Union (IoU) thresholds ranging from 0.5 to 0.95. It is expected that this work contributes to advancing the applications of AI-driven solutions for modern urban sanitation systems.

Experimental Investigation of Gas Transmission Pipeline Blockage Detection Based on Dynamic Pressure Method

The blockage of natural gas pipelines caused by solid deposits such as hydrates is one of the major safety risks to transmission pipeline systems. The key to effective blockage relief or removal is to determine the location and severity of the blockage. In recent years, the pressure pulse wave method has been considered as a practical detection method due to its fast response time, simplicity of operation, and extended detection distance. Nevertheless, the current implementation of this method in pipelines indicates that the accuracy in detecting blockages is very low. To improve the accuracy of the pressure wave blockage detection technique in our experiments, a series of experiments were carried out to detect and locate hydrate blockages in natural gas pipelines based on the pressure wave method using a separate pipeline system of 22 mm diameter and 106 m length. The experimental results show that the accuracy of the blockage location prediction based on the pressure pulse wave method is within 5%. Still, the blockage’s cross-sectional shape can significantly affect the intensity of the reflected wave, with a maximum prediction error of 35%.

An Autonomous Continuous Bladder Irrigation System.

Introduction and Objective: Continuous bladder irrigation (CBI) is used in a variety of clinical settings, including post-transurethral surgery and the emergency department. Currently, CBI administration relies on nurses to diligently monitor and switch irrigation bags, as well as titrate the inflow rate based on effluent color. Inappropriate administration can result in discomfort to patients, clot urinary retention, repeat injury to the pathologic or surgical site, extended hospital stays, and even operative management. Our objective was to create an autonomous CBI system that decreases the incidence of disrupted irrigation flow and monitors the outflow to alert clinicians of critical events. Methods: 3D printing and off-the-shelf microcontrollers were used to design a device to fit the needs identified by stakeholders at our institution. An in vitro model of the bladder was created to test our design. The mechanical, electrical, and software subsystems were adjusted accordingly to meet our design requirements. Results: Our in vitro CBI model was able to simulate routine CBI administration with sudden bleeding. Bovine blood was used to simulate the bleeding events. A device was created that met identified stakeholder needs. Accurate detection of critical bleeding events, catheter blockage, and empty irrigation bags were achieved. The device responds to bleeding appropriately by increasing the irrigation rate. When the catheter is blocked, it stops the irrigation and alerts the nurse. Our system accurately titrated the irrigation rate to match a set outflow blood level parameter, conserving irrigation and minimizing nursing workload. Continuous monitoring of CBI effluent was recorded. Conclusions: We anticipate our device will decrease the cognitive load on nurses in busy clinical settings and improve workflow. Moreover, the detection of critical events will likely decrease patient morbidity. Continuous monitoring of the CBI outflow may prove to be a new clinical decision-making tool for ongoing hematuria. Clinical trial is pending.

Detection of blockage of seed tube in seed drill by using lasor diode sensor

Detection of blockage of seed tube in seed drill by using lasor diode sensor

Fuzzy Recurrence Plots for Shallow Learning-Based Blockage Detection in a Centrifugal Pump Using Pre-Trained Image Recognition Models

Abstract Rags, dusts, foreign particles, etc., are the primary cause of blockage in the centrifugal pump and deteriorate the performance. This study elaborates an experimental and data-driven methodology to identify suction, discharge, and simultaneous occurrence of both blockages. The discharge pressure signals are acquired and denoised using CEEMD. The fuzzy recurrence plots obtained from denoised signals are attempted to classify using three pre-trained models: Xception, GoogleNet, and Inception. None of these models are trained on such images; thus, features are extracted from different pooling layers which include shallow features too. The features extracted from different layers are fed to four shallow learning classifiers: Quadratic SVM, Weighted k-nearest network, Narrow Neural network, and subspace discriminant classifier. The study finds that subspace discriminant achieves the highest accuracy of 97.8% when trained using features from second pooling of Xception model. Furthermore, this proposed methodology is implemented at other blockage conditions of the pump. The subspace discriminant analysis outperforms the other selected shallow classifier with an accuracy of 93% for the features extracted from the first pooling layer of the Xception model. Therefore, this study demonstrates an efficient method to identify pump blockage using pre-trained and shallow classifiers.

Performance evaluation of LSTM and Bi-LSTM using non-convolutional features for blockage detection in centrifugal pump

Blockages in the suction or discharge side of the pump adversely affect the pump’s performance by reducing the flow rate and head, increasing vibration, noise, and overheating. Thus, diagnosing pump blockages accurately and quickly is vital. In this study, a test rig is developed to detect blockages in the pump (suction, discharge, and simultaneous) at three levels of severity. Initially, a non-convolutional feature layer is developed to extract sets of statistical features (SF), entropy-based parameters, and Holder’s exponent (HE). Next, two sequential learners, LSTM and Bi-LSTM, are trained with statistical parameters and tuned using grid-search optimization. Model with least validation loss is chosen as the base model. Subsequently, the process is repeated with only non-linear features, combination of SF with HE, and combination of all three sets of features. Thus, eight different models are trained (four each for LSTM and Bi-LSTM). To eradicate the impact of random variations, these models are trained ten times and the average values of performance metrics (precision, recall, and F1score) are noted. It is found that LSTM trained with all sets of features is most accurate at 99.01% with F1 score of 98.95%. Although LSTM is superior, it is marginally ahead of Bi-LSTM, but LSTM consumes almost 40% lesser time for training. Toward the end, the ablation study of LSTM is performed to analyze the effects of the most influencing parameters. Thus, the present study emphasizes an automated and robust method for maintenance engineers to diagnose the blockages in the pump with severities of blockages.