A critical review on available models to predict engine fuel properties of biodiesel

The engine fuel properties of biodiesel are required for fuel quality assessment, spray and combustion modelling studies and engine optimization studies. Experimental measurement of biodiesel fuel properties over a broad range of temperatures is cost-intensive, laborious, time-consuming, and requires skilled labour. Thus, there have been sustained interests to develop models for predicting biodiesel fuel properties using various approaches for the past 50 years. The models developed for predicting fuel properties are based on various approaches that have advantages and limitations. The present review provides a critical analysis of approaches and models available in the literature to predict biodiesel properties that are important for fuel quality specifications, storage, handling, and durability of the fuel injection systems. The reliability of any empirical model depends on the number of data points considered in calibrating the model. The review articles available in the literature neither discussed the number of calibration samples nor the number of validation samples used for model development. Also, it is essential to validate the prediction models using additional data set which are not considered in calibrating the models. The present review addresses these limitations wherein the favourable features and limitations of models available to predict biodiesel properties are discussed in terms of reported deviations, validation method, model simplicity, applicable range and the number of data points used in calibrating and validating the models. Thus, the analysis provided therein helps to choose appropriate property prediction models based on the intended application. Further, poor oxidative stability is one of the significant drawbacks with biodiesel, and no literature discusses the models to predict oxidative stability of biodiesel. This limitation is also addressed in the present review. Biodiesel composition-based models are suggested to predict the calorific value provided a more comprehensive range, and the type of methyl esters are considered in the model calibration. Due to simplicity, estimating biodiesel density based on the predicted density of methyl esters and applying suitable mixing law is suggested. Accurate models to predict biodiesel flashpoint temperature and oxidative stability need to be developed considering their significance. Models involving composition-based indicators are recommended to predict biodiesel cold filter plugging point and kinematic viscosity due to model simplicity and better accuracy.