Multivariate data analysis applied in Hot-Mix asphalt plants

Abstract

A multivariate data analysis was used with environmental data obtained from Hot-Mix asphalt (HMA) plants. The main objectives of the study were to describe and interpret the real “phenomena” that exist on HMA plants; also, to search for the best mixing conditions, i.e. production with less energy consumption and less emissions. Another important objective was to identify means to allow HMA plants to be designed and operated in better ways. Along the study, an approach of the manufacturing process is presented, allowing for an overall comprehension of the plant's operation.There is not only one parameter to be controlled and checked when HMA are being produced. In this context, a multivariate data analysis was carried out, and the main control parameters considered for prediction and regression analysis were Hot-Mix asphalt formula, fuel type, burner flow rate, fuel consumption, negative pressure of the drum, air temperature, moisture of aggregates, asphalt temperature and rate of production.“Global warming emissions”, “organic emissions”, and “energy consumption” were chosen as models of regression. Globally, plants powered by natural gas emit half the amount of CO2, nine times less NOx and eighteen times less CO than plants with fuel oil. Considering NOx and CO2 emissions, small plants are preferable over big plants. For organic emissions, if the asphalt temperature increases, TOC emissions are increased; however, if the rate of production increases, TOC emissions are decreased. Previously supposed as an important parameter, moisture of aggregates had no great importance in the statistical analysis. Likewise, the formula of HMA produced also had no great importance, despite many types of formulas studied.The main conclusions in regards to the means with which Hot-Mix asphalt plants can be designed and operated in better ways were: the manufacturing of hyper-mobile and small HMA plants as opposed to stationary large plants, and the use of burners with natural gas instead of fuel oil in all cases. The production of HMA plants, aiming at a cleaner production, means to operate asphalt plants always considering timespan in the long term. This encompasses undertaking daily, weekly and yearly analyses. Therefore, it is necessary to produce with low rate of production and low bitumen temperature, and to maintain a regular production in order to avoid peaks of production and unnecessary starts and stops. It is also interesting to increase the culture of heated stocks.Finally, this study shows that would be an advantage to use multivariate data analysis and then to estimate with very good accuracy emissions and energy consumption in HMA plants using current parameters of control.