Practical Approach to Prediction of Plant Technological Parameters with Missing Data
The novel approach how to determine the past and future time horizon for prediction of plant technological parameters with missing data is described. These situations frequently happen in real life when the whole spans of time data are missing due to e.g. plant outages. For this, the mutual information analysis has been proposed to use. The hierarchical training of ANNs has been applied to reduce the error of time series forecasting.
KeywordsMutual Information Time Horizon Time Series Forecast Chaotic Time Series Predictive Maintenance
Unable to display preview. Download preview PDF.
- 2.Bishop Ch.M. (1997) Neural Networks for Pattern Recognition. Clarendon Press.OxfordGoogle Scholar