查詢結果分析
相關文獻
- An Application of Artificial Neural Network to Predict Sorghum Yields
- Neural Network Procedures for Taguchi's Dynamic Problems
- A Fast and Efficient Competitive Learning Design Algorithm Based on Weight Vector Training in Transform Domain
- 專家系統振動訊號圖型判別之研究
- 反傳遞模糊類神經網路於流量推估之應用
- 類神經網路(Neural Networks)的種類及其在影像處理上的應用
- C++Fuzzy類神經網路物件導向發展系統之建立
- 臺灣汽保費率之估計--對數線性費率模式與類神經網路之比較
- 運用類神經網路於股價指數之套利--以日經225指數為例
- 使用類神經網路預估碳化鎢材料放電加工性能
頁籤選單縮合
題名 | An Application of Artificial Neural Network to Predict Sorghum Yields=神經網路應用於預測高粱產量 |
---|---|
作者姓名(中文) | 郭勝豐; | 書刊名 | 農業工程學報 |
卷期 | 41:2 1995.06[民84.06] |
頁次 | 頁29-35 |
分類號 | 434.156 |
關鍵詞 | 神經網路; 後勤函數; 平均差方; Neural network; Logistic function; Mean square error; |
語文 | 英文(English) |
中文摘要 | 利用神經網路的學習能力來預測非線性的物理現象是統計學上的新理論,一個三層(輸入、隱藏、輸出)的神經網路模式被發展來預測高梁的產量。此模式開始於學習輸入(田間土壤水分)及輸出(高梁產量)資料間的關係,並求得其間的係數,學習完成後,僅需輸入資料,便可預測輸出值。嘉南農田水利會學甲試驗站的田間試驗資料被用來測試此模式,結果顯示此模式能準確的學習土壤水分及高梁產量間的關係。然而預測1992年高梁產量的絕對百分比則高至15.64&及低於4.98%。 |
英文摘要 | A neural net work model (NNM) was developed to model sorghum yields. The model consists of a three-layer learning network with input, hidden and output layers. The back propagation method was used to conduct the training process to recognize the correspondence between inputs and outputs, where the inputs include weekly soil moisture and outputs are sorghum yields. After finishing the training process, the neural network model can be used to predict sorghum yields with only the input data and calibrated mode weights. The field experiment data, weekly soil moisture and related sorghum yields, with five treatments from 1990 to 1991, were used for training. Weekly soil moisture data from 1992 were used to predict sorghum yields. The training process is shown to correctly represent the relationship between weekly soil moisture and yields. The absolute error percentage was as high as 15.64% and as low as 4.98%. |
本系統之摘要資訊系依該期刊論文摘要之資訊為主。