Artificial Neural Network as a Prediction Tool in Agricultural Variables
暂无分享,去创建一个
Artificial Neural Network (ANN) technology is a form of artificial intelligence that learns by
processing representative data patterns through its internal architecture. ANN technology often
offers a superior alternative to traditional physical-based models, and excel at uncovering patterns or
relationships in data. It is also a powerful non-linear estimator which is recommended when the
functional form between input-output is unknown or it is not well understood but it is believed could
be nonlinear.
This paper show two applications of ANN for forecasting solar radiation and available dam storage.
In the first case ANN provided a good accurate forecasts for all period with an average square error
of 0.05% in the prediction. For the second case, ANN provide relatively accurate estimates of water
availability one month into the future in the Purisima dam located in the state of Guanajuato. The results suggest that additional input variables such as runoff, cropping patterns and groundwater
extractions may be necessary to increase ANN forecasting accuracy.
This feasibility study demonstrates that ANN technology has the potential to serve as a highly
accurate forecasting tool. Moreover, ANN technology can continuously be updated, as new data
become available, increasing its forecasting ability.