Piperazine pyrophosphate (PAPP) combined with melamine polyphosphate (MPP) was adopted to prepare a waterborne fire retardant intumescent coating (IC) for structural steel. Silicone acrylic emulsion was used as binder. In the 2-h torch test, PAPP/MPP-IC coating presented excellent fire resistance performance. The equilibrium temperature at the backside of the steel board decreased to 170 °C with protection of MPP/PAPP-IC, compared with 326 °C of APP/PER/MEL-IC. After 72-h water immersion, MPP/PAPP-IC could still provide sufficient thermal isolation, but APP/PER/MEL-IC failed the test. The water absorption of the MPP/PAPP coating was also reduced. The thermogravimetric analysis measured that the PAPP/MPP-IC had unique initial decomposition temperature of 296 °C and higher residue of 33.8 wt%, which demonstrated better thermal stability and fire retardancy in condensed phase. In addition, Scanning Electron Microscope (SEM) images illustrated that the structure of the carbon layer formed by MPP/PAPP-IC was dense, complete and consistent, indicating the improvement of mechanical strength and thermal isolation of the char. The synergistic effect between piperazine and phosphoric acid groups in MPP/PAPP contributed to the superior flame retardancy. Consequently, MPP/PAPP-IC was much more efficient than the traditional APP/PER/MEL-IC. This work provides a novel way for designing flame retardant coatings for structural steel with excellent comprehensive performance.
24 October 2013 in Science China Technological Sciences
In the process industry, online operators usually adjust the entire production process in time according to changes in certain key indicators. However, it is difficult to directly obtain the value of key indicators due to the complex structure and the numerous processes of the process industry. This paper proposes a method to achieve high precision on-line prediction of key indicators in the process industry. The method first performs data preprocessing of multi-source heterogeneous time series data involved in industrial processes based on the professional knowledge, which not only keeps the prediction error within 10%, but also reduces the prediction time. Then a model framework is constructed based on LSTM neural network and the error correction algorithm is proposed to improve prediction accuracy based on real-time error, which directly causes the error to drop by 3% -5%. At the end, a set of multi-mode online training strategies and related trigger conditions are designed to perform predicting online. Diesel hydrodesulfurization is a typical case in the process industry. The effectiveness of the proposed method is empirically studied by applying its actual data sets. Through the comparison with other traditional well-known forecasting models, and models optimized by adjusting parameters, the experimental results demonstrate that the method can achieve the great prediction performance in terms of both accuracy and stability.