Advanced Machine Learning Techniques for Corrosion Rate Estimation and Prediction in Industrial Cooling Water Pipelines.

This paper presents the results of a study on data preprocessing and modeling for predicting corrosion in water pipelines of a steel industrial plant. The use case is a cooling circuit consisting of both direct and indirect cooling. In the direct cooling circuit, water comes into direct contact with the product, whereas in the indirect one, it does not. In this study, advanced machine learning techniques, such as extreme gradient boosting and deep neural networks, have been employed for two distinct applications. Firstly, a virtual sensor was created to estimate the corrosion rate based on influencing process variables, such as pH and temperature. Secondly, a predictive tool was designed to foresee the future evolution of the corrosion rate, considering past values of both influencing variables and the corrosion rate. The results show that the most suitable algorithm for the virtual sensor approach is the dense neural network, with MAPE values of (25 ± 4)% and (11 ± 4)% for the direct and indirect circuits, respectively. In contrast, different results are obtained for the two circuits when following the predictive tool approach. For the primary circuit, the convolutional neural network yields the best results, with MAPE = 4% on the testing set, whereas for the secondary circuit, the LSTM recurrent network shows the highest prediction accuracy, with MAPE = 9%. In general, models employing temporal windows have emerged as more suitable for corrosion prediction, with model performance significantly improving with a larger dataset.

Winterization of peanut biodiesel to improve the cold flow properties.

Biodiesel is susceptible to start-up and performance problems, consistent with its chemical composition, when vehicles and fuel systems are subjected to cold temperatures. In this work, a comprehensive evaluation of the crystallization behavior of different biodiesels was performed by measuring the cold filter plugging point (CFPP), cloud point (CP) and pour point (PP). Results were related to differential scanning calorimetry (DSC) thermograms. Peanut methyl esters in particular led to the most unfavorable properties due to the presence of long-chain saturated compounds (arachidic or C20:0, behenic or C22:0, and lignoceric or C24:0 acid methyl esters) approaching 6 wt.%. The cold flow properties may be improved with different winterization techniques to eliminate some of these compounds. In this work, various techniques are tested, and the best technique is found to be crystallization filtration using methanol, which reduces the CFPP from 17 degrees C to -8 degrees C with a biodiesel loss of 8.93 wt.%. Moreover, the cake from filtration, enriched with long-chain saturated methyl esters, can be used as phase change material (PCM) for thermo-regulated materials.

Influence of fatty acid composition of raw materials on biodiesel properties.

The aim of this work was the study of the influence of the raw material composition on biodiesel quality, using a transesterification reaction. Thus, ten refined vegetable oils were transesterificated using potassium methoxide as catalyst and standard reaction conditions (reaction time, 1h; weight of catalyst, 1 wt.% of initial oil weight; molar ratio methanol/oil, 6/1; reaction temperature, 60 degrees C). Biodiesel quality was tested according to the standard [UNE-EN 14214, 2003. Automotive fuels. Fatty acid methyl esters (FAME) for diesel engines. Requirements and test methods]. Some critical parameters like oxidation stability, cetane number, iodine value and cold filter plugging point were correlated with the methyl ester composition of each biodiesel, according to two parameters: degree of unsaturation and long chain saturated factor. Finally, a triangular graph based on the composition in monounsaturated, polyunsaturated and saturated methyl esters was built in order to predict the critical parameters of European standard for whatever biodiesel, known its composition.