A study of hybrid powertrains and predictive algorithms applied to energy management in refuse-collecting vehicles

There is an increasing demand for vehicles with lower environmental impact and higher fuel efficiency. To meet these requirements, powertrain hybridization has been introduced in both academia and industry during recent years. In this work we have carried out an in-depth analysis of the potential reduction in fuel consumption of refuse-collecting vehicles (RCV), based on different hybridization technologies. This study has been structured in four work packages: RCV energy model development, predictive algorithms aimed at energy management for RCV drive cycles, analysis of hybrid hydraulic powertrains for RCV and analysis of hybrid electric powertrains for RCV. RCV energy model development: the process of simulation starts with the RCV energy model development, which is based on the improvement of the classic approach (breaking down the energy consumption into aerodynamics, rolling resistance, road profile and inertias), and adding the RCV ancillaries? (lifter, compactor and unloading devices) consumption. The adjustments and validation methodology applied to the models is based on the use of low-cost hardware and post-processing data by use of GPS and cartography. Working with this method it has been empirically demonstrated that high accuracy energy consumption estimations are possible through the use of these models. Predictive algorithms aimed at energy management for RCV drive cycles: based on the principle that RCV drive cycles are repetitive, the typical RCV drive cycle has been modeled and its main characteristics parameterized. The model is separated into different drive cycles which are related to different power consumption modes. Based on this analysis two algorithms are presented, which can identify in which drive mode the vehicle is operating; these algorithms are based on deterministic principles and on artificial intelligence respectively. Also, an algorithm which can estimate the energy needed to finish the current trip is presented; this algorithm is based on information inferred from previous trips, statistics and infinitesimal calculus. Once the drive cycle of an RCV is analyzed and modeled, the working modes of a standard internal combustion engine on real drive cycles are presented and contrasted with fuel consumption maps. This study concludes that, because of the typical drive cycles of RCV, an ICE is oversized most of the time and tends to work in low efficiency points. Based on this information a basic alternative powertrain architecture is proposed Analysis of hybrid hydraulic powertrain for RCV: hybrid hydraulic powertrains are often used in heavy duty vehicles in which big power flows can be found. As RCV represent an application with big power flows, the performance of this technology applied to RCV has been analyzed. This study is based on the development of models for each of the components of the powertrain, and the simulation of the whole powertrain model on real routes. All the component models are based on information provided by the manufacturers, and the routes have been logged in real RCV drive cycles. The conclusion is that an important fuel saving can be achieved by the hydraulic hybridization of the powertrain. Analysis of hybrid electric powertrain for RCV: hybrid electric powertrains are often used in cars and light duty vehicles. As RCV represents a light to medium duty application the performance of this technology applied to RCV has been analyzed. As in the hybrid hydraulic case, this study is based on the development of models for each of the powertrain components, and the simulation of the whole powertrain model on real routes. Finally a comparison between the hybrid electric and the hybrid hydraulic is established, identifying the more efficient technology and the advantages and inconveniences of each. And establishing what would be, according to the author, the most interesting technological powertrain evolution in the mid and long term for this kind of vehicle.