類神經網路在稻穀幹減率模式建立與分析上之應用

Application of Neural Network on Rice Moisture Content Removal Rate Modeling and Analysis

謝志誠(Jyh-Cherng Shieh)

《農業機械學刊》 2卷4期 (1993/12) Pp. 21-32

https://doi.org/10.30062/JAM.199312.0003

摘要

近年來已有不少研究者對不同品種的稻穀，以不同的方法探討熱風溫度、熱風濕度、送風量、每循環乾燥時間、乾燥時間與稻穀期初含水率等對稻穀在乾燥過程中含水率乾減速率的影響。
本文擬利用多層前饋式類神經網路模型來表示稻穀乾燥過程，並採反向傳遞演算法學習乾燥條件對稻穀含水率乾減率的關係，從而建立稻穀的乾減率模式，以量化了解乾燥條件對稻穀乾減速率的影響。所建立的模式除了可以用來分析不同乾燥條件與稻穀乾減率的關係外，並可用以推測在各種乾燥條件下為達到特定期末含水率所需的乾燥時間。

關鍵字

稻穀 ； 幹燥 ； 幹減速率 ； 類神經網路

並列摘要

The effects of drying conditions on rice moisture content removal rate have been extensively studied. The drying conditions consist of drying air temperature, drying air humidity, drying air flow rate, drying time per cycle, drying time, initial moisture content of grain, and the variety of rice.
An alternative approach that determines the moisture content removal rate model using neural network technology is suggested in this article. The experimental drying conditions and moisture content removal rate are used as input and desired patterns to estimate the connection weights of neural network. In this paper, a three-layer feed-forward neural network and back-propagation learning algorithm are proposed for the modeling of rice drying. The generated neural network model can be used to analyze the effects of drying conditions on rice moisture content removal rate and predict the needed drying time for a desired final moisture content.

並列關鍵字

Rice ； Drying ； Moisture Content Removal Rate ； Neural Network