Off-Highway Machine Fuel Performance Prediction Through Engine Data Analytics

Field performance of a machine is conventionally analyzed using tools of virtual validation such as physics-based simulations models. Machine performance test data is typically not incorporated for machine performance evaluation using these tools. The present work aims to demonstrate use of Data Analytics as a tool to analyze this data for predictive purpose. It aims at establishing numerical relationships of engineering interest within the data, which would otherwise be complex if done only using physics-based models. Engine operation data spanning over three months, comprising of more than sixty channels, of an off-highway machine is used for model development. Machine fuel burn rate is chosen as the dependent variable. Several independent variables such as Ambient temperature, Engine speed, Charge air pressure are chosen based on their correlation with the dependent variable and upon engineering interest. Linear regression models are developed which show good fit and correlation. The model demonstrates a high R^2 value suggesting high robustness in the choice of independent variables. The model established from training data set is compared with predictions obtained from the validation data set, in an attempt towards model validation. The model so developed is deployed to predict fuel performance, given a set of machine operating parameters. Model accuracy can be further improved by exploring other algorithms as well as several techniques of machine learning.