This paper proposes a neural-network-based system for ice motion prediction in the Gulf of St. Lawrence. A supervised neural network is trained to predict ice conditions at a given location and time using current ice charts, which are provided by the Canadian Ice Service. The input ice data is mapped to an output vector that gives the predicted ice conditions. The traditional non-modular feedforward neural network structure failed to map the required function, and hence, was modularized to give better prediction performance. Each neural module was responsible for the prediction of a 5x5 km area, while the ice characteristic of interest was the total concentration.
展开▼
