We present a novel decision-making framework for accelerated degradation tests and predictive maintenance that exploits prior knowledge and experimental data on the system’s state. As a framework for sequential decision making in these areas, dynamic programming and reinforcement learning are considered, along with data-driven degradation learning when necessary. Furthermore, we illustrate both stochastic and machine learning degradation models, which are integrated in the framework, using data-driven methods. These methods are presented as a valuable tool for designing life-testing experiments and for maintaining lithium-ion batteries.

Mopeds are popular means of transportation, particularly in southern Europe and in eastern and southern Asia. The relative importance of their emissions increases in urban environments which host large fleets of mopeds. In Naples, for example, mopeds make a considerable contribution to HC emissions (about 53%), although the percentage of mopeds (12.4%) in the total circulating fleet is lower than that of other vehicle categories [1]. This study presents a method for analysing the influence of kinematic parameters on the emission factors of mopeds during the “cold-start” and “hot” phases of elementary kinematic sequences (speed-time profiles between two successive stops). These elementary sequences were obtained through appropriate fragmentation of complex urban driving cycles. In a second step, we show how to estimate, for the whole cycle, the duration of the cold phase and the relevant time-dependence function. The method was applied to analyse the regulated exhaust emissions (CO, HC, NOX) of a 4-stroke moped (Euro 2) tested on a dynamometer bench. The experimental data were compared with ARTEMIS and COPERT emission factors, two of the emission models available in Europe to calculate road traffic emissions.

In internal combustion engines, the direct injection at high pressures produces a strong impact of the fuel on the combustion chamber wall, especially in small-bore sizes used for passenger cars. This effect is relevant for the combustion process resulting in an increase of the pollutant emissions and in a reduction of the engine performances. This paper aims to report the effects of the injection pressure and wall temperature on the macroscopic behavior and atomization of the impinging sprays on the wall.The gasoline spray-wall interaction was characterized inside an optically accessible quiescent chamber using a novel make ready Z-shaped schlieren-Mie scattering set-up using a high-speed C-Mos camera as imaging system. The arrangement was capable to acquire alternatively the schlieren and Mie-scattering images in a quasi-simultaneous fashion using the same line-of-sight. This methodology allowed complementing the Mie scattering images, adapting to the liquid phase, with the schlieren ones for the determination both of the liquid and vaporing phase during the single cycle. A single-hole axially disposed injector was used, 0.200 mm in diameter L/d=1.0, while the injection pressure and the wall temperature ranged between 5.0 - 20.0 MPa and ambient to 573 K, respectively. Optical investigations were carried out at atmospheric backpressure and ambient gas temperature. The images of the impact, resolved in the cycle, were processed by a customized algorithm able to catch the contours of the liquid phase and the vapor/atomized zone. Spatial and temporal evolutions were measured for both the phases in terms of width penetration and thickness growth.

Recently, the International Agency for Research on Cancer has assessed evidence that exposure to outdoor air pollution causes lung cancer and increases the risk of bladder cancer. Because air pollution in urban areas is mainly caused by transportation, it is necessary to evaluate pollutant exhaust emissions from vehicles and scooters during their real-world use. Laboratory tests, with the exception of the high repeatability of the results, often cannot explain the influence of some parameters, such as ambient conditions, traffic congestion, and road gradients. Therefore, experiments were conducted to study the effects of road gradients on the exhaust emissions of a medium displacement scooter, which represents a very popular means of travel in Southern Italy. First, the scooter was instrumented with a global positioning system (GPS) to acquire speed profiles and positions in an urban route of Napoli city, which is characterized by significant changes in the road gradient. The velocity profile has several missing points due to GPS signal loss over some stretches of road adjacent to tall buildings. The road gradient values were built into an algorithm to synchronize the missing data through Google elevation API based on a model of the same path with a complete set of data. Afterwards, some representative slope values for each kinematic sequence/driving cycle were evaluated. These new variables were found to contribute to the individual real-world driving cycles. The resulting real-world driving cycle was tested using a chassis-dynamometer to continuously simulate the exact road gradient. A series of experiments with and without the road gradient simulations were performed to evaluate the influence of the road gradient on the exhaust emissions of carbon monoxide, total hydrocarbons, nitrogen oxides, and carbon dioxide. This analysis was performed using a two-wheeler vehicle fuelled with commercial gasoline and two blends with a maximum bioethanol content of 20% vol. Gaseous emissions were correlated to the vehicle specific power (VSP), which is the most useful parameter for addressing the road gradient together with the kinematic characteristics of the driving cycle and the vehicle characteristics. Through a multivariate statistical approach, this type of gradient analysis can permit correlation of the emission profiles for a specific road position, and evaluate its influence on their behavior. A strict dependence on carbon dioxide emissions and fuel consumption exists for the VSP. The road gradient greatly increases the vehicle power demands, resulting in an increasing amount of fuel consumed for each kilometer driven