• Energy Research

Abstract This work reports the use of mid- and near-infrared spectroscopy (MIR and NIR) to predict the kinematic and dynamic viscosities of biodiesel–diesel blends. A partial least square regression (PLSR) modeling method was employed to develop the calibration models based on information from four commonly used biodiesel and three different commercial diesel fuels. For MIR spectroscopy, wavelengths in the fingerprint region of 550–1500 cm−1 were chosen for developing the model. The root mean square error of prediction (RMSEP) for kinematic viscosity and dynamic viscosity were 0.114 and 0.119 mm2/s, respectively, based on the validation set that consisted of 26 biodiesel–diesel blend samples made of six different biodiesel and three different diesel fuels. For the NIR spectroscopy, the PLSR model established using the spectral regions of 1100–1500 nm, 1600–1700 nm, and 1800–2200 nm obtained better results. The RMSEP were 0.070 mm2/s for kinematic viscosity and 0.062 mm2/s for dynamic viscosity prediction. The results indicated that both MIR and NIR can be used to accurately predict the viscosities of biodiesel–diesel blends, but better results can be obtained using NIR spectroscopy.

Biodiesel made from different source materials usually have different physical and chemical properties and the concentration of biodiesel in biodiesel-diesel blends varies from pump to pump and from user to user; all these factors have significant effects on performance and efficiency of engines fueled with biodiesel. To address these challenges, regressions based on near-infrared spectroscopy were developed for relatively inexpensive and rapid on-line measurement of the concentration and specific gravity of biodiesel-diesel blends. Methyl esters of five different oils — soybean oil, canola oil, palm oil, waste cooling oil, and coconut oil — and two different brands of commercial-grade No. 2 on-highway diesel and one brand of off-road No. 2 diesel were used in the calibration and validation processes. The predicted concentration and specific gravity of the biodiesel-diesel blends were compared with the actual values. The maximum and average root-mean-square errors of prediction (RMSEP) of biodiesel concentration were 5.2% and 2.9%, respectively, from the biodiesel type-specific regression. For the general regression, the RMSEP were 3.2% and 0.002 for biodiesel concentration and specific gravity predictions, respectively.