Cold flow properties: Applying exploratory analyses and assessing predictive methods for biodiesel and diesel-biodiesel blends

[Display omitted] •Accuracy of several prediction methods of cold flow properties (CFPs) were assessed.•Exploratory analysis with 173 biodiesel data detected some esters related to CFPs.•Many biodiesel types were analyzed, such as methyl, ethyl, propyl, and butyl esters.•Due to lack of studies, predictions of diesel–biodiesel blends CFPs were assessed.•The used database comprises biodiesel from vegetable, residue and algae feedstock. With the necessity to mitigate climate change, fossil fuels have been replaced by biofuels. Currently, biodiesel, which is basically a mixture of esters used with diesel as a blend, is the most popular biofuel. As an advantage, biodiesel has lower and less harmful pollutant emissions compared to diesel, besides being renewable and biodegradable. However, its applicability at low temperatures requires caution due to crystallization. To measure biodiesel operability at low temperatures, cold flow properties (CFPs) such as cloud point (CP), pour point (PP), and cold filter plugging point (CFPP) are monitored. CFPs can be calculated based on physical properties or ester composition. Therefore, exploratory analyses were applied to investigate the influence of some esters on CFPs to develop new predictive methods with relevant esters. Also, existing methods that predict CFPs were assessed and their accuracy was compared by using the parameter Average Absolute Deviation (AAD). The accuracy of CFPs prediction was highly dependent on the biodiesel type and the suitable method. The most precise method for CP, PP, and CFPP prediction achieved AAD = 1.34 %, 1.22 %, and 1.16 %, respectively. For biodiesel-diesel blends new methods for CFP’s prediction were developed with AAD inferior to 1 %, similar to existing methods.