Claudio Annicchiarico

Design of a Semi Active Differential to Improve the Vehicle Dynamics

Suppressing or limiting the differential action of the differential mechanism is the mostly adopted technique to avoid the skidding of a driving wheel of a vehicle riding on a poorly adherent surface. The devices carrying out this function unbalance the traction force distribution in the differential, generating a yaw torque acting on the vehicle as a secondary effect. If the unbalancing action is electronically controlled, this yaw torque can be used to affect the attitude of car as a torque vectoring technique.In this paper, a purpose built differential is presented and its technical features are highlighted, including the electrohydraulic actuation. Moreover, its torque vectoring capabilities are discussed, basing on the numerical simulation campaign performed deploying this device in a 7 DOFs model of a race car with low ground effect.The results of these simulations are compared with the behavior of the same vehicle equipped with a common passive locking differential, to show that the proposed one and its control logic (which relies on only measurable inputs) are able of improving the handling of the vehicle, in terms of both vehicle stability and linearity with the driver’s inputs. Therefore, this system could be considered as a completion of the common ESC (“Electronic Stability Control) systems to control the vehicle attitude when using the brake system is an inefficient solution.Copyright

Numerical and Experimental Testing of Composite Rings for Reciprocating Compressor Valves

Valves are one of the most critical class of components for the reliability of reciprocating compressors. Many failures of reciprocating compressor valves are due to valve ring failure. Investigating the failure mode of rings is complicated because of the ring composite material properties, which randomly and locally range from orthotropic to anisotropic due to the random disposition of the reinforcement, usually consisting of short fibers.In this paper, a cooperation between GE Oil & Gas and the University of Florence, the results of a numerical simulation campaign will be presented, along with its correlation with the experimental evidence arising from a purpose-built test rig. This work guided the design of a new material for rings, whose characteristics have been tailored based on the results of this experimental and numerical campaign.© 2013 ASME

Design and testing of an innovative electro-hydraulic actuator for a semi-active differential

The performance of semi-active differential systems is deeply influenced by the way in which they are actuated. Conventional semi-active systems are actuated by pressure-controlled hydraulic actuators that, despite their good dynamical behaviour, are relatively inefficient. In this work, the authors propose an innovative controlled hydraulic pump solution that is able to drastically improve the efficiency and to reduce the overall encumbrances, maintaining very high performance in terms of dynamical behaviour and corresponding frequency response. The authors investigate design criteria of the proposed solution, describe a simulation model and validate design and simulation models with experimental data.

An integrated artificial neural network–unscented Kalman filter vehicle sideslip angle estimation based on inertial measurement unit measurements

Vehicle dynamics stability control systems rely on the amount of so-called sideslip angle and yaw rate. As the sideslip angle can be measured directly only with very expensive sensors, its estimation has been widely studied in the literature. Because of the large non-linearities and uncertainties in the dynamics, model-based methods are not a good solution to estimate the sideslip angle. On the contrary, machine learning techniques require large datasets that cover the entire working range for a correct estimation. In this paper, we propose an integrated artificial neural network and unscented Kalman filter observer using only inertial measurement unit measurements, which can work as a standalone sensor. The artificial neural network is trained solely with numerical data obtained with a Vi-Grade model and outputs a pseudo-sideslip angle which is used as input for the unscented Kalman filter. This is based on a kinematic model making the filter completely transparent to model uncertainty. A direct integration with integral damping and integral reset value allows the estimation of the longitudinal velocity of the kinematic model. A modification strategy of the pseudo-sideslip angle is then proposed to improve the convergence of the filter’s output. The algorithm is tested on both numerical data and experimental data. The results show the effectiveness of the solution.

Development of a real-time steering system model for driving simulators

Driving simulators have boosted the vehicle design with the introduction of human beings in the simulation loop. For a realistic functioning, the steering system must provide an accurate behaviour, since the hand wheel is a crucial human interface. Despite a large diffusion of steering models, this paper deals with the creation of a specific solution for real-time applications, characterized by precise features as numerical stability and low computational cost. The proposed model is based on a physical structure and considers all the key phenomena, such as the system elasticities, the power steering effects and friction hysteresis, making the model more accurate in terms of steering wheel torque and lateral acceleration than other angle-driven models. Its two degrees of freedom design allows a proper behaviour of the power steering sub-model; another key aspect is the friction model: the use of the LuGre formulation greatly improves accuracy and stability in comparison to the lookup table friction models. Compared to the literature reference torque-driven model, it does not need the use of a torque sensor when implemented in driving simulators having an angle-driven formulation (the input of the steering wheel is its angle and the torque needed is its output), hence it is cheaper to implement; nevertheless, its accuracy is close to state-of-art reference. An original parametrization procedure is proposed since a generalized one is not available in literature; using a steering test-rig, all the model variables are defined. The validation phase combines offline and online simulations, assessing objectively and subjectively the model’s capabilities and showing accurate results in terms of steering wheel torque, lateral acceleration and steering feeling. In addition, a minor contribution of this paper shows how different analyses (steering effort evaluation, experimental data comparison or simulator feedback computation) require different output torques.

Tribologic Analysis, Wear Evolution and Torque Trend Estimation of an LSD Clutch Pack

The wear phenomenon in friction clutches involves a variation in the contact pairs characteristics, defining a decay of transmissible torque properties. For multidisk clutch packs installed in electroactuated differentials, the friction torque evolution allows to define how wear decay can influence the dynamic of the vehicle where the differential is installed in. A tribological analysis of a molybdenumcoated clutch pack was made under typical working conditions in terms of temperature, normal load and sliding speed. Friction coefficient and wear rate trends were outlined to define material characteristics through block-on-ring tests. Metallurgic studies on the specimens used in tribologic tests validated the base material used in the samples. Results from experimental analyses were used as input for a numerical study regarding wear and pressure evolutions during clutch engagement. The application of Reye’s theory and Archard’s law on a FE model of the clutch led to the definition of an algorithm to estimate the friction torque generated by the clutch system and the its lifetime. KEyWoRDS Archard, Clutch, FEM, Friction, LSD, Molybdenum, Reye, Torque, Wear