24小时热门版块排行榜

返回列表

【有奖交流】积极回复本帖子，参与交流，就有机会分得作者 AeroTraffic 的 20 个金币，回帖就立即获得 1 个金币，每人有 1 次机会

当前只显示满足指定条件的回帖，点击这里查看本话题的所有回帖

AeroTraffic

铜虫 (小有名气)

应助: 0 (幼儿园)
金币: 75.5
帖子: 203
在线: 47.4小时
虫号: 11001670

[交流] 【2022伦敦玛丽女王大学-CSC公派】Jun Chen老师招博士新生：运筹学/机器学习方向

【项目简介】
研究方向：运筹学、机器学习。
面向场景：通信网络。
课题名称：effective energy and infrastructure management through forecasting, interpretation and simulation-based optimisation.
联系方式：jun.chen@qmul.ac.uk（感兴趣的同学请从速联系 dr. jun chen）。
截止日期：2022年1月30日。
详情介绍请见以下英文描述，或通过链接下载：https://pan.baidu.com/s/1i1y59g5spp3ux-nv_0iogw 提取码: 2g1p

【额外福利】
成功申请者将加入qmul多学科研究团队，并与英国领先的电信公司密切合作；
将有机会加入图灵研究院（https://www.turing.ac.uk/）进行为期最长12个月的学习培训；
接受英国工程和自然科学研究委员会（epsrc）industrial mobility and doctoral training programme in data science and engineering的课程培训；
提供csc奖学金外的额外资金，用于参与国际会议、走访合作大学或工业伙伴等。

【学校概况】
qmul是英国伦敦大学四大核心学院之一，英国“常青藤联盟”罗素大学集团成员，英国科学与工程六校联盟成员，迄今共9位校友获诺贝尔奖。
usnews 2022：第110名；the 2022：第117名；qs 2022：第117名。
在英国与csc合作最久、支持力度最大的高校之一。

a csc studentship (2022) is available now:
project title: "effective energy and infrastructure management through forecasting, interpretation and simulation-based optimisation"
a csc studentship for one chinese student is available in areas of operational research and machine learning at the school of engineering and material sciences (sems), queen mary university of london (qmul).
communication providers account for 2% to 3% of total global energy demand, making them some of the most energy-intensive companies in their geographic markets, and alongside other utility providers they often operate some of the largest commercial fleets in their regions. existing energy models in the telecom industry have facilitated understanding of energy consumption at a more fine-grained detail. further improvement can be gained through better handling of heterogeneous data, nonlinear coupling between features and spatial/temporal dependence, and increased models' transparency for easy interpretation/integration with domain knowledge. furthermore, mapping energy consumption to energy efficiency for assets usage across telecom networks requires tools that can utilise large amounts of unlabelled and unbalanced data of overlapped boundaries. utility and communication providers are also moving towards electric vehicles (evs) for their engineering workforces, and they are therefore developing optimal recharging infrastructure and rollout strategies. further improvement/insights can be gained through evaluating evs' energy demands, and concurrent transit network planning and recharging scheduling, considering vehicle types, charging facilities and route characteristics. this project aims to develop such tools for supporting analysis, identifying and recommending energy saving/planning strategies.
building on our previous multi-objective interpretable fuzzy neural networks [1], but additionally incorporating preference-based evolutionary algorithms [2], long-short term mechanism and graph attention layers, a set of energy consumption models at different transparency levels will be elicited. building on our previous deep unsupervised learning approach [3], energy efficiency classification models that can boost the less represented categories and overcome the overlapping boundary problem will be developed. building on our previous multigraph generation approach [4-8], but additionally incorporating waypoints to represent route characteristics e.g. congestion/traffic lights/road gradient/speed limits, public transit networks will be created. building on the simplified longitudinal dynamics model [9], but additionally incorporating our energy-efficient driving generation approach [5, 8], energy demands of different evs on different routes employing different driving profiles (including regenerative braking) will be evaluated.

the proposed research will be set within this context and aims to answer the following key research questions:
1. given a set of noisy, nonlinear, dynamic and heterogeneous data sets containing e.g. measurements from iot devices, types and locations of equipment, meteorological and demographical datasets, filed historical energy consumption, and prior domain knowledge from specialist, how to accurately predict long-term energy consumption using supervised learning approaches, where the predictions are transparent to domain experts and provide uncertainty bounds.
2. given a large number of unlabelled/unbalanced energy data, how to uncover and describe their corresponding energy efficiency using unsupervised learning approaches so that further 'what-if' scenario-based analysis can be carried out, facilitating decision making.
3. given different evs, charging infrastructures and route characteristics, how to estimate energy requirements of evs using low-resolution data, and provide comprehensive planning decisions regarding the location and capacity of the charging stations and operational decisions regarding recharging schedules or the assignment of evs to chargers.

 csc will provide living expenses (up to 4 years) and one return flight ticket and qmul will provide a full tuition fee;
 successful applicants will join the multi-disciplinary research team at qmul and work closely with leading telecom companies in the uk;
 the student will have the opportunity to join the alan turing institute (https://www.turing.ac.uk/) for up to 12 months to boost their skills, grow their network and work alongside other turing researchers;
 the student will also benefit from training in data science and research provided by the epsrc industrial mobility and doctoral training programme in data science and engineering;
 additional funding will be available to cover site visits and dissemination of the results at international conferences, workshops and collaborative universities and industrial partners.

about the applicants:
 a top master or undergraduate student (top 5%) in at least two of the following areas: operational research, machine learning, energy consumption modelling, transportation engineering and control engineering.
 english minimum requirement: ielts 6.5 or equivalent english tests.
 preferably from a top chinese university (211 and 985).
 preferably with some decent publications.

the deadline for submitting your application to qmul is 30 january 2022. if you are interested in this studentship, please contact dr jun chen as soon as possible by email: jun.chen@qmul.ac.uk.
for more information on how to apply for this qmul-csc scholarship, please refer to this link:
https://www.sems.qmul.ac.uk/rese ... y-in-september-2022

reference:
[1] chen, j., & mahfouf, m. (2012). improving transparency in approximate fuzzy modeling using multi-objective immune-inspired optimisation. international journal of computational intelligence systems, 5(2), 322-342.
[2] weiszer, m., chen, j., stewart, p., & zhang, x. (2018). preference-based evolutionary algorithm for airport surface operations. transportation research part c: emerging technologies, 91, 296-316.
[3] li, b., du, w., zhang, y., chen, j., tang, k., & cao, x. (2021). a deep unsupervised learning approach for airspace complexity evaluation. ieee transactions on intelligent transportation systems.
[4] chen, j., weiszer, m., locatelli, g., ravizza, s., atkin, j. a., stewart, p., & burke, e. k. (2016). toward a more realistic, cost-effective, and greener ground movement through active routing: a multiobjective shortest path approach. ieee transactions on intelligent transportation systems, 17(12), 3524-3540.
[5] chen, j., weiszer, m., stewart, p., & shabani, m. (2016). toward a more realistic, cost-effective, and greener ground movement through active routing—part i: optimal speed profile generation. ieee transactions on intelligent transportation systems, 17(5), 1196-1209.
[6] weiszer, m., chen, j., & locatelli, g. (2015). an integrated optimisation approach to airport ground operations to foster sustainability in the aviation sector. applied energy, 157, 567-582.
[7] weiszer, m., chen, j., & stewart, p. (2015). a real-time active routing approach via a database for airport surface movement. transportation research part c: emerging technologies, 58, 127-145.
[8] zhang, t., ding, m., zuo, h., chen, j., weiszer, m., qian, x., & burke, e. k. (2018). an online speed profile generation approach for efficient airport ground movement. transportation research part c: emerging technologies, 93, 256-272.
[9] gallet, m., massier, t., & hamacher, t. (2018). estimation of the energy demand of electric buses based on real-world data for large-scale public transport networks. applied energy, 230, 344-356.

详情介绍有更新，新版PDF请见：https://pan.baidu.com/s/1_52HO3hnlz5JiUe4YuPMfQ 提取码: vivg
（原帖网盘链接已失效，请见谅）

[ Last edited by SongweiLIU on 2022-1-5 at 23:11 ]

回复此楼