中原大學電機工程學系
Loading
專題成果首頁>專題成果
107學年度 >電力能源組
以人工智慧深度學習進行電力市場電價預測
指導老師:洪穎怡   組長:余舜華   組員:嚴少威、林芳竹
現代人對於金錢精確的掌握,對於公司或是家庭來說都需要有明確的支出開銷數據。因此精確的預測電價趨勢是必要的。本文用了多層感知器回歸、窗口方法的多層感知器、長期短期記憶多層神經網路、卷積神經網路、門控循環單元、長期短期記憶多層神經網路加上卷積神經網路、Wavenet此七種方法預測電價。此七種方法由神經網路和演算法所組成。神經網路是使用線性整流函數做為激活函數,演算法用於優化神經網路的權重和線性整流函數的參數。本文使用五種統計指標比較實際值與預測值的誤差,由此比較哪一種方法所預測出來的結果最精準。
Modern people's precise grasp of money requires clear expenditure data for companies or families. Therefore, accurate forecasting of electricity price trends is necessary. This article uses Multi-layer Perceptron Regression, Multi-layer Perceptron Using the Window Method, Long Short-Term Memory (LSTM), Convolution Neural Network (CNN), Gated-Recurrent-Unit (GRU), long-term short-term memory multi-layer neural network plus convolutional neural network (LSTM PLUS CNN), Wavenet these seven methods to predict electricity prices. These seven methods consist of neural networks and algorithms. The neural network uses a linear rectification function as an activation function, and the algorithm is used to optimize the weight of the neural network and the parameters of the linear rectification function. In this paper, five statistical indicators are used to compare the error between the actual value and the predicted value, so that which method predicts the most accurate results.
我們決定將壓電陶瓷晶片和鞋子做結合。
   
 
32023 桃園市中壢區中北路200號 No-.200, Zhongbei Rd., Zhongli City, Taoyuan County 320, Taiwan(ROC) TEL:03-2654801 FAX:03-2654899