participants
ITRI 工業技術研究院成立於1973年，以科技研發，帶動產業發展，創造經濟價值，增進社會福祉為任務；成立四十多年來，累積近3萬件專利，並新創及育成281家公司。
餘光能源團隊有十多年的太陽能經驗，在海內外協助建置、持有之太陽能電廠超過300MW以上，涵蓋家戶型、商業屋頂型與地面型。豐富的開發、建置與維運經驗，是你最值得信賴的團隊。
餘光秉持互信、互利，共同合作的理念，謀求業主、政府、餘光能源密切合作，共同創造三贏的局面！。
Net Zero Emissions has been an international consensus and a global goal. Currently 136 countries have participated and some countries have even reached the initial goals. Recently Taiwan has formulated four major strategies and expects to achieve zero carbon emissions by 2050. Increasing the proportion of renewable energy generation such as solar power and wind power is also an important direction for energy transformation.
This topic uses the historical data of solar power generation such as daily solar irradiation, installation capacity, etc., to predict the solar power generation at various places with machine learning and hope participants can find out the key factors affecting the amount of generation and make a positive impact for the details of future solar energy construction.
Reference： 2050 Net-Zero Emissions
Award eligibility: the score is lower than baseline (RMSE < 260) on Private Leaderboard
| Award name | Award money | |
|---|---|---|
| First place | 1 person | NTD $50,000 (Tax included) And TWS 100,000 cloud credits |
| Second place | 1 person | NTD $30,000 (Tax included) And TWS 80,000 cloud credits |
| Third place | 1 person | NTD $20,000 (Tax included) And TWS 50,000 cloud credits |
| Honorable mention | multiple persons | TWS 20,000 cloud credits |
| Continuous lead (Hold first place for longest time) | 1 person | TWS 30,000 cloud credits |
The time is based on UTC+8 as follows:
| Time | Event |
|---|---|
| 2022/04/20 | Registration and dataset download starts |
| 2022/06/21 23:59:59 | Upload deadline |
| 2022/06/23 | Private Leaderboard announcement and report submission starts |
| 2022/06/30 23:59:59 | Report submission deadline |
| 2022/07/06 | Awards announcement |
After the participants submitted the prediction result, the back-end of the system would process them in batches regularly to calculate the score. Evaluations are conducted by calculating the Root-Mean-Square Error, RMSE, of the actual value. The formula is as follows: $$RMSE = \sqrt{{1 \over n} \sum_{j=1}^{n} (y_i - \hat{y}_i) ^ 2}$$
