Prediction of Boiler Heat-Conducting Oil Temperature Based on Multi-Modality Fuzzy Cognitive Maps

Accurate and interpretable time series prediction is of great significance in production, it can help people deal with boiler abnormalities timely and accurately, and provide guarantee for safe production. This paper proposes a time series prediction method based on multi-modality fuzzy cognitive maps with data preprocessing, which can soundly achieve the purpose. As the model is for solving practical problems, the experiment in paper uses real boiler data to improve objectivity. Firstly, the outliers in the dataset are replaced with reasonable data, then fuzzify the processed time series. Subsequently, the subsequences are chosen in the time series, and the fuzzy cognitive maps model is established on each subsequence. Finally, the model outputs of each subsequence are mixed as the prediction results and obtaining the forecast results. The experimental results contracted with Long Short-Term Memory (LSTM) show that the proposed model can well predict the changing trend of the actual industrial heat-conducting oil temperature of boiler.