Anomaly Detection Platform: Wipro Pipe Sleuth
Published on : Thursday 04-01-2024
The implementation of Wipro Pipe Sleuth Solution signifies a significant leap forward in pipeline management, says Rupesh Kumbhare
Wipro Pipe Sleuth Solution stands at the forefront of wastewater utility innovation, offering an advanced artificial intelligence-based automated sewer pipe condition assessment solution. Tailored for the wastewater utilities sector, this groundbreaking technology utilises sophisticated image processing and deep neural network algorithms to identify, grade, and score pipe anomalies in accordance with industry standards.
The Solution boasts a wide-ranging capability, automatically generating Pipeline Assessment and Certification Program (PACP) or Water Research Centre (WRC) compliant comprehensive inspection reports while accommodating over 50 anomalies. This comprehensive approach ensures that the assessment is accurate and adheres to established standards, providing water utility companies with a reliable tool for pipeline maintenance.
The solution includes the following features:
• Efficiently process multiple video files simultaneously.
• Allow users to visualise identified anomalies.
• Support diverse video formats for enhanced flexibility and Playback function for processed videos.
• Benefit users from the ability to review individual defects with annotations, facilitating a nuanced and detailed examination of the pipeline's condition.
• Streamline the assessment process by significantly reducing the manual effort and time traditionally required to review and code pipe assessment video scans.
Wipro Pipe Sleuth Solution delivers both efficiency and a comprehensive understanding of sewer pipe conditions, empowering water utility companies to make informed decisions on proactive maintenance and infrastructure management.
Wipro Pipe Sleuth Solution is typically deployed on the Edge Computing Platform. However, the architecture is flexible to accommodate Hybrid deployment scenarios (Edge + Cloud).
Business requirement:
Utility companies conduct annual inspections covering approximately 50,000 feet of water and sewer pipes, a crucial practice in managing sewer pipelines. Periodic assessments are vital to identify potential issues and ensure the overall integrity of the pipeline network. However, the conventional manual review process, often involving the inspection of hundreds of hours of video footage, poses significant challenges. It incurs high costs and is susceptible to errors due to the sheer volume of data.
The adoption of improved technologies for the assessment of pipeline videos is paramount. This advancement enables a more efficient and accurate review process, offering a quicker turnaround time for creating repair work orders. By leveraging AI capabilities in Wipro Pipe Sleuth, utility companies can streamline the analysis of inspection videos, reducing the time and resources required. This enhances the speed at which necessary repairs can be identified and implemented and increases the overall precision of the assessment process. Ultimately, integrating Wipro Pipe Sleuth in video review contributes to optimising pipeline management practices in the utility sector.
Utility companies, in their transition towards digital technologies to deliver business efficiencies and improve the manual review process of the Pipeline Assessment Videos, were error-prone, slow, and labor intensive. Utility Companies use an external contractor to conduct pipeline inspections. A camera-mounted rover is sent inside the pipeline to scan and record the video. An external operator maneuvers the camera and the rover and analyses the video while recording. The operator uses established PACP software to key in their observations and generate television inspection logs and summary reports. The report highlights problems and the location within the pipeline. These video recordings and PACP reports are submitted to the Utility QA/QC department for review. QA and QC staff/engineers review the video and the reports to create a final pipe quality assessment report. Repair crew work order is prioritised and scheduled based on the results of this pipe quality assessment.
Typically, Utility Companies conduct rover-based video scans yearly covering approximately 50-70 thousand feet of pipe. The scan process generates several hundred hours of video annually that is reviewed manually, a process that is cumbersome, time-consuming, and prone to errors.
Solution
Wipro has successfully developed a cutting-edge solution tailored for underground wastewater pipeline installations using Advanced Image processing and Artificial Intelligence (AI) algorithms. This innovative technology enables the automated processing of videos, identifying and classifying various anomalies, such as cracks, roots, and grease deposits. The implementation of Wipro Pipe Sleuth Solution signifies a significant leap forward in pipeline management, as it harnesses the power of automation to analyse videos offline, freeing the process from the constraints of manual inspection.
One of the key advantages of this solution is its capacity for automatic report generation, providing a comprehensive and detailed overview of the pipeline's condition. This feature expedites the assessment process and ensures accuracy and consistency in anomaly detection. Notably, a premier water utility company in the United States of America has embraced and deployed this solution, attesting to its efficacy and relevance in real-world applications. Adopting Wipro's advanced technology exemplifies a stride toward more efficient and sophisticated approaches to pipeline maintenance, marking a positive evolution in integrating AI in critical infrastructure management.
Business impact
The first roll-out covered 3,200 miles of pipes, covering 725 sq.miles. The liability impact on the company was also mitigated.
• Reduced review time per video by 80%: Approx a 10-minute video is analysed within 2 minutes for average health pipeline videos. Users can review the report and submit additional comments/instructions for the Repair Team.
• Improved accuracy: This solution extracts the potential frames and identifies the anomalies with >90% accuracy. The solution is trained on a large real-world dataset to improve accuracy.
• Automation by batch processing: The Q/A team could select several videos together and start the analysis to run overnight. The engineer can then review the report for accuracy.
• Compliance support: The solution supports 50+ different anomalies that are PACP or WRC compliant.
• Improved Asset Management: Utility companies can plan their Expenditure and Operation spend efficiently and with fewer complaints.
Rupesh Kumbhare, Practice Head, Wipro Engineering Edge, is an embedded software professional with proven experience in complex embedded project execution and delivery, expertise in voice over IP system software, DSP platforms and architecture. His specialties include: DSP system software, voice over IP firmware, real-time operating system, ability to understand and solve complex technical problems, and DSP algorithm development.
For a deeper dive into the dynamic world of Industrial Automation and Robotic Process Automation (RPA), explore our comprehensive collection of articles and news covering cutting-edge technologies, robotics, PLC programming, SCADA systems, and the latest advancements in the Industrial Automation realm. Uncover valuable insights and stay abreast of industry trends by delving into the rest of our articles on Industrial Automation and RPA at www.industrialautomationindia.in
