119MO Application of an artificial neural network for predicting the chemotherapy benefit of patients with gastric cancer after radical surgery

Qualitative and quantitative analysis of the influence of biodiesel fatty acid methyl esters on iodine value

Iodine value (IV) is an indicator to evaluate the degree of unsaturation (DU) of biodiesel. It reflects the biodiesel degradation and oxidation stability (OS) and also has an effect on viscosity, low-temperature flow properties (LTFP), and the combustion performance. To construct a theoretical system for the simultaneous optimization of LTFP and OS of biodiesel using IV, 52 measured experimental data are used to investigate the qualitative and quantitative relationship between IV and biodiesel composition. The relationships between biodiesel physicochemical properties and IV are investigated in this work. The qualitative analysis shows that the poly-unsaturated fatty acid methyl esters (FAMEs) contribute to an increase in IV, whereas saturated and mono-unsaturated FAMEs decrease IV. Multiple linear regression (MLR) and artificial neural network (ANN) are used to estimate IV from FAMEs. The correlation coefficient, root mean squared error (RMSE), and mean absolute percentage error (MAPE) are respectively 0.976, 2.45, and 1.76% for the MLR model and 0.983, 2.14, and 1.57% for the back propagation neural network (BPNN) model; these values indicate the high accuracy of these methods. The performances of the proposed models were compared with three existing IV prediction models and validated by another databank. The results indicate that the application of the developed BPNN model is better and more comprehensive. Additionally, a preliminary conclusion is that biodiesel with a low percentage of both long-chain saturated and poly-unsaturated FAMEs can have solidifying point (SP) and OS in the proper range. Biodiesel with a low IV is generally more combustible and efficient.