2022 Fiscal Year Research-status Report
Social Energy System Design Incorporating AI and Lived Experience
| Project/Area Number |
22K18039
|
|---|
| Research Institution | Kyushu University |
|---|
Principal Investigator |
Chapman Andrew 九州大学, カーボンニュートラル・エネルギー国際研究所, 准教授 (60795293)
|
|---|
| Project Period (FY) |
2022-04-01 – 2024-03-31
|
| Keywords | environmental behavior / social equity / energy system design / machine learning / statistical analysis / survey |
|---|
| Outline of Annual Research Achievements |
Many research achievements have been made under this research plan. Firstly, the theoretical underpinnings of machine learning and prediction of environmental and social equity preferences of stakeholders via existing data sets (previously undertaken surveys) was established and reported in research currently under review at Applied Energy. The manuscript details multiple machine learning algorithmic approaches to energy system design preferences and the effects of imcreasing training data toward accuracy rates. Based on these findings, a national survey was designed and deployed, specifically to achieve a low-burden, high yield predictive data set which will be leveraged in the coming year. Currently this data is being used to underpin efforts toward a prediciton and visualization tool.
|
| Current Status of Research Progress |
Current Status of Research Progress
2: Research has progressed on the whole more than it was originally planned.
Reason
Underpinning theoretical work is complete on schedule, allowing for the leverage of findings and obtained data toward teh development of prediction and visualization tools. All envisaged activities for AY2022 were completed and it is expected that work on the tools can be proceeded smootly in AY2023.
|
| Strategy for Future Research Activity |
This year's activities will be separated into 3 phases - first, survey data analysis will be undertaken to establish linkages between behaviors and preferences. From these analyses a machine learning and prediciotn model will be established and subsequently tested at multiple workshops. Finally, the findings of these workshops and machine learning activities will inform the development of a publicly available prediciton and visualization tool and supporting academic publications.
|
