太陽能發電效率與地理環境因子影響分析

Abstract

本研究旨在探討臺灣地區地理環境因子對太陽能發電效率之影響，聚焦於日照時數、氣溫、降雨量與月份等氣候變數，運用多元迴歸與逐步迴歸模型進行實證分析。研究資料來自H科技公司提供之2021年至2023年全臺各地太陽能發電站之每月平均發電量資料，並結合中央氣象署與農委會氣象開放資料平台所提供之對應區域氣候數據。研究結果顯示，日照時數與發電效率呈正相關，氣溫與降雨量則因非線性關係而對效率造成複合影響。透過加入變數平方項與月份虛擬變數，有效提升模型解釋力，並藉由逐步迴歸篩選出具顯著影響之關鍵氣候因子。進一步的共線性檢驗與殘差分析亦確認模型穩健性。本研究結果有助於提升太陽能設施之選址策略與季節性調度規劃，並可為未來再生能源設施部署與能源政策制定提供量化依據。

This study investigates the influence of geographic and climatic factors on solar power generation efficiency across Taiwan, focusing on key variables such as sunshine duration, temperature, rainfall, and seasonal variation. Utilizing multiple and stepwise regression models, the analysis is based on monthly average solar generation data from 2021 to 2023, provided by H Technology Company. These data are integrated with regional meteorological records from Taiwan’s Central Weather Administration and the Council of Agriculture. The results reveal that sunshine duration is positively correlated with solar efficiency, while temperature and rainfall exhibit nonlinear effects. By incorporating squared terms and monthly dummy variables, the regression models demonstrate improved explanatory power. Stepwise regression further identifies the most significant climate variables, and diagnostic checks, including multicollinearity and residual analyses, confirm model robustness. The results of this study will help improve the site selection strategy and seasonal scheduling planning of solar energy facilities, and provide a quantitative basis for the deployment of future renewable energy facilities and energy policy formulation.