PFI TT: Intelligent quality assurance and integration tool for sewer inspection data

PFI TT：下水道检查数据的智能质量保证和集成工具

• 批准号: 2141184
• 负责人: Yongwei Shan
• 金额: $ 25万
• 依托单位: Oklahoma State University
• 依托单位国家: 美国
• 项目类别: Standard Grant
• 财政年份: 2022
• 资助国家: 美国
• 起止时间: 2022-02-01 至 2025-01-31
• 项目状态: 未结题
• 来源: https://www.nsf.gov/awardsearch/showAward?AWD_ID=2141184&HistoricalAwards=false
• 关键词: PFI; TT; Intelligent; quality; assurance

The broader impact/commercial potential of this Partnerships for Innovation - Technology Translation (PFI-TT) project is conduct quality monitoring of sewers for clear public benefit. Current practices in handling data incompatibility and quality issues are manual and time-consuming. In addition, these practices are customized for individual projects, a process hard to adapt to different projects. Moreover, the sewer inspection data remains fragmented within the utilities due to the lack of proper data management solutions, which makes data less accessible for asset management. The envisioned technology will help to combine sewer inspection data with standard PACP (Pipeline Assessment Certification Program) format, ensure data quality, and integrate historical data for easy and secured access so that cost-effective asset management plans can be developed based on true conditions. Each utility could save up to 80% of the time (on the order of months) for manual data quality assurance. With accurate and accessible sewer condition data, it is anticipated that at least 20% (approximately $600 million) of the annual spending on pipe rehabilitation can be saved for the entire industry. More importantly, improved sewer infrastructure condition can greatly reduce gastrointestinal (GI) illness emergency room visits induced by sanitary sewer overflow events. The proposed project is to 1) develop an artificial intelligence (AI)-enabled algorithm to resolve data incompatibility issues to maintain data fidelity and facilitate data exchange among different software platforms; 2) establish a holistic AI-enabled methodological framework to automatically detect and treat data quality issues, including data records inconsistency, incompleteness, and duplicates, caused by the interruption of inspection operations due to pipe defects or human errors; and 3) standardize municipal utilities’ practices in managing sewer condition data using blockchain-enabled technologies. This PFI-TT project will tailor the emerging AI/machine learning and blockchain technologies to develop an efficient data format translation algorithm, effective data quality assurance framework, and a blockchain-enabled sewer data management tool, with the objective to save engineers’ effort in data processing and render them more time for engineering decision making.This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.

该创新-技术转化伙伴关系（PFI-TT）项目的更广泛影响/商业潜力是对下水道进行质量监测，以实现明确的公共利益。 目前处理数据不一致和质量问题的做法是手工操作，而且很耗时。此外，这些实践是为单个项目定制的，这是一个很难适应不同项目的过程。此外，由于缺乏适当的数据管理解决方案，下水道检查数据在公用事业中仍然是分散的，这使得资产管理难以获得数据。该设想的技术将有助于将联合收割机下水道检测数据与标准PACP（管道评估认证计划）格式相结合，确保数据质量，并整合历史数据，以轻松安全地访问，从而可以根据真实情况制定具有成本效益的资产管理计划。每个实用程序都可以节省高达80%的时间（大约几个月）用于手动数据质量保证。有了准确和可访问的下水道状况数据，预计整个行业每年至少可以节省20%（约6亿美元）的管道修复费用。更重要的是，改善下水道基础设施条件可以大大减少卫生下水道溢流事件引起的胃肠道（GI）疾病急诊室就诊。建议的项目包括：1）开发一个人工智能算法，解决数据不兼容的问题，以保持数据的保真度，并促进不同软件平台之间的数据交换; 2）建立一个全面的人工智能支持的方法框架，以自动检测和处理数据质量问题，包括数据记录不一致，不完整和重复，由于管道缺陷或人为错误导致的检查操作中断造成的;以及3）使用区块链技术规范市政公用事业管理下水道状况数据的做法。这个PFI-TT项目将量身定制新兴的AI/机器学习和区块链技术，以开发高效的数据格式转换算法，有效的数据质量保证框架，以及支持区块链的下水道数据管理工具，旨在节省工程师在数据处理方面的精力，并使他们有更多的时间进行工程决策。该奖项反映了NSF的法定使命，并被认为值得通过使用基金会的知识价值和更广泛的影响审查标准进行评估。

项目成果

Transformer-enabled generative adversarial imputation network with selective generation (SGT-GAIN) for missing region imputation

• DOI: 10.1080/24725854.2023.2193257
• 发表时间: 2023-03
• 期刊: IISE Transactions
• 影响因子: 2.6
• 作者: Yuxuan Li;Zhangyue Shi;Chenang Liu

Developing a Data Quality Evaluation Framework for Sewer Inspection Data

开发下水道检查数据的数据质量评估框架

• DOI: 10.3390/w15112043
• 发表时间: 2023
• 期刊: Water
• 影响因子: 3.4
• 作者: Khaleghian, Hossein;Shan, Yongwei
• 通讯作者: Shan, Yongwei

A Generative Adversarial Network (GAN)-Assisted Data Quality Monitoring Approach for Out-of-Distribution Detection of High Dimensional Data

用于高维数据分布外检测的生成对抗网络 (GAN) 辅助数据质量监控方法

• 发表时间: 2023
• 期刊: IISE
• 作者: Slater, K.;Wang, Y.;Shan, Y.;Liu, C.
• 通讯作者: Liu, C.

Ensemble Generative Adversarial Imputation Network with Selective Multi-Generator (ESM-GAIN) for Missing Data Imputation

• DOI: 10.1109/case49997.2022.9926629
• 发表时间: 2022-08
• 期刊: 2022 IEEE 18th International Conference on Automation Science and Engineering (CASE)
• 作者: Yuxuan Li;Ayse Dogan;Chenang Liu

When AI meets additive manufacturing: Challenges and emerging opportunities for human-centered products development

• DOI: 10.1016/j.jmsy.2022.04.010
• 发表时间: 2022-07-01
• 期刊: JOURNAL OF MANUFACTURING SYSTEMS
• 影响因子: 12.1
• 作者: Liu，Chenang;Tian，Wenmeng;Kan，Chen
• 通讯作者: Kan，Chen