Narrow Results

There are many methods to evaluate the performance of an air pollution index forecasting model based on its forecasting results and history data. But, there is no method to judge if forecasting models can success or not in the future. Motivated by this finding, the idea of quality control charts in total quality management is used to judge the performance of an artificial neuron network (ANN) models in future forecasting. The new model consist of two parts, the first part is the well known ANN forecasting model, and the second part is its quality chart created on the basis of history error data. Results show that quality chart in this model can indicate its future performance in advance and the model can adjust itself accordingly. This makes it safe that it is always suitable for future usage. It performs well in terms of accuracy and reliability.

BP neural network can achieve arbitrary nonlinear mapping of the input to the output, so it is extensively adopted in intelligent control, image recognition, hydrological predicting and water-resource quantity evaluation, etc., has stronger features of mapping, classification, functional fitting. This paper chooses the water quality of Lanzhou section of Yellow river as example by use of BP model to forecast the water quality. It is well verified that BP network model can reach the purposes of early warning and predicting.