Intelligent Hybrid Prediction Method of Reducing Zone Temperature in Shaft Furnace

The temperature of the reducing zone in the shaft furnace roasting process is a key controlled variable, but it´s hard to be measured continuously. In this instance, an intelligent hybrid prediction model is developed to predict the temperature punctually based on the neural networks and case-based reasoning. This model consists of five modules: a data collection and data processing module, a decision-making module, a prediction module, an online modified module and an effect estimating module. The whole model´s framework, the main function of each module and the implementation of the algorithms are discussed. Applications to a shaft furnace roasting process show that the model is effective in both normal and abnormal operating conditions. The obvious benefits of it are low maintenance expense, good real time character, high reliability and perfectly precision.