8月2日:香港城市大學(xué)田華明博士學(xué)術(shù)報(bào)告通知
報(bào)告題目 | 數(shù)據(jù)物理雙驅(qū)動(dòng)的巖土工程機(jī)器學(xué)習(xí)與數(shù)字孿生建模 |
報(bào)告人 | 田華明 博士 |
邀請(qǐng)人 | 李典慶 教授 |
時(shí)間 | 2024年8月2日(星期五)下午3:30 |
地點(diǎn) | 水資源國重大樓A區(qū)202會(huì)議室 |
報(bào)告人簡(jiǎn)介:
田華明博士,2021年12月獲武漢大學(xué)博士學(xué)位,現(xiàn)任香港城市大學(xué)博士后。主要從事巖土工程(如填海項(xiàng)目)數(shù)據(jù)分析、機(jī)器學(xué)習(xí)等領(lǐng)域研究,在工程地質(zhì)與巖土工程領(lǐng)域權(quán)威期刊《Journal of Geotechnical and Geoenvironmental Engineering》、《Canadian Geotechnical Journal》、《Engineering Geology》、《Acta Geotechnica》、《巖土工程學(xué)報(bào)》等發(fā)表多篇論文。
報(bào)告簡(jiǎn)介:
數(shù)字化轉(zhuǎn)型與數(shù)字孿生建模是巖土工程發(fā)展的必然趨勢(shì)。本報(bào)告將凝練巖土工程數(shù)字孿生建模的關(guān)鍵技術(shù)與挑戰(zhàn),提出數(shù)據(jù)與物理雙驅(qū)動(dòng)的機(jī)器學(xué)習(xí)方法,展示如何通過一體化、數(shù)字化和自動(dòng)化方式構(gòu)建高精度地質(zhì)模型,精準(zhǔn)映射物理原理至數(shù)字空間,并融合時(shí)變、多源、高維空間監(jiān)測(cè)數(shù)據(jù),實(shí)時(shí)、持續(xù)地提升模型預(yù)測(cè)能力。探討如何解決傳統(tǒng)機(jī)器學(xué)習(xí)大量訓(xùn)練數(shù)據(jù)依賴、時(shí)空預(yù)測(cè)外插能力不足、計(jì)算效率低下以及可解釋性差等難題。以澳大利亞、土耳其、美國、香港等實(shí)際工程案例,展示如何利用所提方法助力巖土工程數(shù)字化轉(zhuǎn)型與數(shù)字孿生建模。
8月8日【水科學(xué)講壇】第58講:北京航空航天大學(xué)潘翀教授學(xué)術(shù)報(bào)告通知
報(bào)告題目 | 機(jī)器學(xué)習(xí)輔助的流場(chǎng)測(cè)量技術(shù) |
報(bào)告人 | 潘 翀 教授 |
邀請(qǐng)人 | 季 斌 教授 |
時(shí)間 | 2024年8月8日(星期四)下午2:30 |
地點(diǎn) | 水資源國重大樓A區(qū)202會(huì)議室 |
報(bào)告人簡(jiǎn)介:
潘翀,北京航空航天大學(xué)航空科學(xué)與工程學(xué)院教授,國家杰出青年基金獲得者。主要從事先進(jìn)流場(chǎng)測(cè)量技術(shù)、壁湍流擬序結(jié)構(gòu)和和湍流控制相關(guān)研究,在Journal of Fluid Mechanics、Physics of Fluids、AIAA Journal、Experiment in Fluids等國內(nèi)外力學(xué)和航空航天權(quán)威期刊發(fā)表SCI論文80余篇,SCI他引1000余次。獲授權(quán)國家發(fā)明專利20余項(xiàng)、軟件著作權(quán)12項(xiàng)。曾獲國家技術(shù)發(fā)明二等獎(jiǎng)、教育部技術(shù)發(fā)明一等獎(jiǎng)、教育部自然科學(xué)一等獎(jiǎng)等。任中國空氣動(dòng)力學(xué)會(huì)理事、中國力學(xué)學(xué)會(huì)流體力學(xué)專業(yè)委員會(huì)委員、北京力學(xué)學(xué)會(huì)流體力學(xué)專業(yè)委員會(huì)副主任委員,《空氣動(dòng)力學(xué)報(bào)》和《氣動(dòng)研究與實(shí)驗(yàn)》副主編。
報(bào)告簡(jiǎn)介:
流體測(cè)量為認(rèn)識(shí)、把握復(fù)雜流動(dòng)的時(shí)空演化規(guī)律提供大數(shù)據(jù)。機(jī)器學(xué)習(xí)的最終目的是用已有數(shù)據(jù)取代未來實(shí)驗(yàn)/試驗(yàn)。在仍然需要流體測(cè)量實(shí)驗(yàn)/試驗(yàn)的現(xiàn)階段,如何用機(jī)器學(xué)習(xí)的方法來輔助測(cè)量并提升測(cè)量能力,是一個(gè)重要且有趣的機(jī)器學(xué)習(xí)應(yīng)用方向。本報(bào)告將匯報(bào)我們?cè)谶@一方向上的幾個(gè)嘗試,包括神經(jīng)網(wǎng)絡(luò)空間標(biāo)定模型、粒子三維重構(gòu)的NN-LOS算法、混合蟻群粒子匹配算法和機(jī)器學(xué)習(xí)增強(qiáng)的表面形貌測(cè)量方法。
8月13日:美國俄亥俄州立大學(xué)Ryan Winston副教授學(xué)術(shù)報(bào)告通知
報(bào)告題目 | Evaluating older bioretention performance: Soil development, pollution accumulation, and water quality performance |
報(bào)告人 | Ryan Winston 副教授 |
邀請(qǐng)人 | 張 翔 教授 |
時(shí)間 | 2024年8月13日(星期二)上午10:00 |
地點(diǎn) | 水資源國重大樓A區(qū)202會(huì)議室 |
報(bào)告人簡(jiǎn)介:
Ryan Winston,男,美國俄亥俄州立大學(xué)食品、農(nóng)業(yè)和生物工程系,土木、環(huán)境和大地測(cè)量工程系副教授。曾任俄亥俄州立大學(xué)食品、農(nóng)業(yè)和生物工程系的研究助理教授,北卡羅萊納州立大學(xué)推廣助理。美國土木工程師協(xié)會(huì)注冊(cè)專業(yè)工程師,《Sustainable Water in the Built Environment》副主編。
主要從事城市雨水徑流水文水質(zhì)研究以及雨水控制措施的效果研究。先后主持了多次綠色基礎(chǔ)設(shè)施相關(guān)會(huì)議,多項(xiàng)創(chuàng)新雨水管理重點(diǎn)項(xiàng)目,開展了雨水控制措施維護(hù)相關(guān)的檢查、維護(hù)課程。其團(tuán)隊(duì)的研究內(nèi)容主要包括“哥倫布藍(lán)圖”綠色基礎(chǔ)設(shè)施項(xiàng)目、綠色碼頭、生物滯留柱結(jié)構(gòu)及材料研究、酸性礦井排水研究等。
報(bào)告簡(jiǎn)介:
Bioretention is among the most frequently implemented green stormwater infrastructure practices. Beyond basic maintenance checks, however, little monitoring is done, despite functioning to clean stormwater for years. This seminar will focus on a field research study aimed at understanding pollutant accumulation and soil development in older bioretention cells. We visited 29 bioretention facilities in Ohio, Michigan, and Kentucky and sampled soils in the forebay (if present), and near the inlet, middle, and outlet of each bioretention cell. Further, three soil depths (0-5, 10-15, and 36-50 cm) were sampled to understand how pollutants accumulate spatially and with depth, with the ultimate goal of informing maintenance. The outcomes of this study include an understanding of phosphorus, metals, microplastics, PAHs, PCBs, alkylphenols, phthalates, and PFAS accumulation in older bioretention media. Further, we evaluate hydraulic function of the media using quick infiltration tests and relate the hydraulics to soil physical properties, including the development of an organic horizon on top of the filter media. Practical recommendations will be provided from the study including when and where to maintain field-scale bioretention systems.
8月13日:美國俄亥俄州立大學(xué)Joey Smith博士候選人學(xué)術(shù)報(bào)告通知
報(bào)告題目 | Soil evolution in aging bioretention cells in the USA and China |
報(bào)告人 | Joey Smith 博士候選人 |
邀請(qǐng)人 | 張 翔 教授 |
時(shí)間 | 2024年8月13日(星期二)上午11:00 |
地點(diǎn) | 水資源國重大樓A區(qū)202會(huì)議室 |
報(bào)告人簡(jiǎn)介:
Joey Smith,男,美國俄亥俄州立大學(xué)博士候選人,武漢大學(xué)高級(jí)訪問學(xué)者。在俄亥俄州立大學(xué)獲得生態(tài)工程和漢語語言工程碩士學(xué)位。博士研究結(jié)合水文學(xué)、土壤學(xué)和社會(huì)科學(xué),分析海綿城市綠色基礎(chǔ)設(shè)施對(duì)社會(huì)和自然環(huán)境的積極作用。
報(bào)告簡(jiǎn)介:
Bioretention cells are engineered stormwater controls essential for mitigating urbanization's adverse effects but require regular maintenance to ensure their continued hydraulic function. This study examines the accumulation of particulates and organic matter in over 50 bioretention cells, aged less than one to 16 years, in the Midwestern United States and 3 cells, aged three to five years, in Wuhan, China. The presentation reviews past studies on soil evolution within bioretention cells and their predictions of design lifespan, highlighting the reliability of laser diffraction particle size analysis over conventional methods. Findings indicate that mean particle size varies with depth and location, showing larger particles at the forebay/inlet and smaller particles at the middle/outlet. The mean surface intake rate was lowest at the inlet (777.2 mm/hr) and highest at the outlet (1389.4 mm/hr). Key factors influencing soil clogging in bioretention cells include the presence or absence of a forebay, percentage of plant coverage, and average mulch depth. This approach aims to guide maintenance practices effectively and ensure the long-term functionality of bioretention systems in urban stormwater management.
8月24日【水科學(xué)講壇】第59講:新加坡科技設(shè)計(jì)大學(xué)校長方國光院士學(xué)術(shù)報(bào)告通知
報(bào)告題目 | 數(shù)字巖土工程(Databases and data-centric geotechnics) |
報(bào)告人 | 方國光 教授 |
邀請(qǐng)人 | 李典慶 教授 |
時(shí)間 | 2024年8月24日(星期六)上午10:00 |
地點(diǎn) | 水資源國重大樓A區(qū)202會(huì)議室 |
報(bào)告人簡(jiǎn)介:
方國光,新加坡科技設(shè)計(jì)大學(xué)校長、鄭曾文講席教授,新加坡工程院(SAEng)院士,新加坡國家科學(xué)院(SNAS)院士,新加坡政府科學(xué)顧問委員會(huì)成員,新加坡-天津經(jīng)濟(jì)貿(mào)易理事會(huì)成員,新加坡民航局董事會(huì)成員,曾任新加坡國立大學(xué)高級(jí)副教務(wù)長、新加坡總理公署國家研究基金會(huì)副首席科學(xué)顧問。
方教授長期從事數(shù)字巖土工程和機(jī)器學(xué)習(xí)方法研究。主要學(xué)術(shù)榮譽(yù)包括:2005年和2020年兩次獲得美國土木工程師學(xué)會(huì)(ASCE)諾曼獎(jiǎng)?wù)拢?017年獲洪堡研究獎(jiǎng),2023年獲Harry Poulos Award獎(jiǎng),2024年獲Alfredo Ang Award,《Georisk》期刊的創(chuàng)刊主編和《Geodata and AI》期刊主編,新加坡注冊(cè)工程師和東盟特許專業(yè)工程師,等。
本次報(bào)告內(nèi)容是方教授于2024年7月10在澳洲巖土力學(xué)協(xié)會(huì)所做的Harry Poulos講座報(bào)告,今天首次在中國演講。
報(bào)告簡(jiǎn)介:
This lecture was first presented in the 2023 Harry Poulos Award and Lecture, 10 July 2024, Sydney
(https://australiangeomechanics.org/meetings/databases-and-data-centric-geotechnics/).
Research in data-centric geotechnics is accelerating as a result of tremendous advances in machine learning and AI. ChatGPT from OpenAI, Gemini from Google Deepmind, and other generative AIs have moved the divide between what a machine can do and what a human can do in a major way. There is near complete consensus that machine learning and AI have the potential to transform the way we work, live, and play in many fundamental ways. It is prudent for geotechnical engineers to understand and to explore the power and the impact of these new tools, particularly their value propositions to practice (Phoon and El-Din Anwar 2024).
Data is now considered to be an asset that is as valuable as our physical infrastructure. This advantage is not well appreciated by most engineers, although it is pivotal to digital transformation. Machine learning and AI universally depend on data for training and validation. Data-centric geotechnics is an emerging area that is underpinned by three elements: (1) data centricity, (2) fit for (and transform) practice, and (3) geotechnical context. The purpose of this lecture is to present the latest research findings in data-driven site characterization (one application area in data-centric geotechnics) to illustrate how data can support decision making in real world projects when data-driven methods are developed with the above elements in mind.