| 摘要: |
| 为探索宏观农业生产系统预测的新方法，构建了基于人工神经网络的预测模型，利用1994-2003年的气象、经济、生产、投入、技术、价格各方面的数据对我国粮食生产进行了拟合分析，并预测了2004年粮食总产，预测的结果为46125．46万t。结果表明，与灰色系统相比，本文建立的模型具有90％以上的拟合精度，模型具有容错能力、联想能力和学习能力．可以用来尝试解决农业生产系统预测问题。 |
| 关键词: 人工神经网络 BP模型 农业生产系统 预测 |
| DOI：10.11841/j.issn.1007-4333.2004.05.116 |
| 修订日期：2004-04-26 |
| 基金项目:科技部十五攻关资助项目 (2 0 0 1BA5 13B3) |
|
| Prediction model of agricultural production based on artificial neural network |
|
|
|
| Abstract: |
| To explore new prediction methods of macro agricultural production system, one method of modeling the prediction of agricultural production based on BP (back propagation) model is established. Fitness analysis of foodstuff production was made and the total production of 461.2546 million ton in 2004 was forecasted using all the data of meteorology, economy, production, inputs, technology and prices. Compared with gray system method, the model has over 90% of precise fitness,tolerance on errors, association and studying capacity. This model can be applied to solve issue of prediction of agricultural production system. |
| Key words: artificial neural network,back propagation model,agricultural production system,forecast |
