No abstract

No abstract

FOREWORDAn extensive Round Robin test programme on compressive softening was carried out by the RILEM Technical Committee 148-SSC "Test methods for the Strain Softening response of Concrete". The goal was to develop a reliable standard test method for measuring strain softening of concrete under uniaxiat compression. The main variables in the test programme were the specimen slenderness hid and the boundary restraint caused by the loading platen used in the experiments. Both high friction and low friction loading systems were applied. Besides these main variables, which are both related to the experimental environment under which softening is measured, two different concretes were tested: a normal strength concrete of approximately 45 MPa and a higher strength concrete of approximately 75MPa. In addition to the prescribed test variables, due to individual initiatives, the Round Robin also provided information on the effect of specimen shape and size. The experiments revealed that under low boundary friction a constant compressive strength is measured irrespective of the specimen slenderness. For high friction loading systems (plain steel loading platen), an increase of specimen strength is found with decreasing slenderness. However, for slenderness greater than 2 (and up to 4), a constant strength was measured. The shape of the stress-strain curves was very consistent, in spite of the fact that each labora-tory cast its own specimens following a prescribed recipe. The pre-peak behaviour was found to be independent of specimen slenderness when low friction loading platens were used. However, for all loading systems a strong increase of (post-peak) ductility was found with decreasing specimen slenderness. Analysis of the results, and comparison with data from literature, showed that irrespective of the loading system used, a perfeet localization of deformations occured in the post-peak regime, which was first recognised by Van Mier in a series of uniaxial compression tests on concrete between brushes in 1984.Based on the results of the Round Robin, a draft recommendation will be made for a test procedure to measure strain softening of concrete under uniaxial compression. Although the post-peak stress-strain behaviour seems to be a mixture of material and structural behaviour, it appears that a test on either prismatic or cylindrical specimens of slenderness hid = 2, loaded between low friction boundaries (for example by inserting sheets of teflon between the steel loading platen and the specimen), yield.; reproducible results with relatively low scatter. For normal strength concrete, the closed-loop test can be controlled by using I the axial platen-to-platen deformation as a feed-back signal, ] whereas for high-strength concrete either a combination of axial] and lateral deformation should be used, or a combination of] axial deformation and axial load.

No abstract

No abstract

Often at the earliest stage of an engineering project, a preliminary optimization could be useful, in order to allow the designer to ascertain the envisaged performance of the system under development.Providing an efficient (analytical) tool to quickly define the Pareto-optimal set could be an extremely valuable chance to make the right design decision at the right time.The procedure proposed here to obtain the Pareto-optimal set in analytical form refers mostly to design problems described by a limited number of design variables and a limited number of objective functions and constraints.In the first part of the paper, the analytical derivation of the expression of the Pareto-optimal set for multi-objective optimization problems is dealt with.According to the knowledge of the authors, in the literature, very few papers exist on this topic and related issues. A survey of current continuous multi-objective optimization concepts and methods is presented in Ref. [19]. Some relevant contributions are given in Ref. [17] and Ref. [20] in which some new formulations of the Fritz John first order conditions are proposed and analyzed. In Ref. [30] first and second order conditions are proposed for a convex multi-objective problem via scalarization and in Ref.[1] some second order conditions are analyzed in detail. In Refs. [17,20,30,1,32,26] necessary and/or sufficient conditions are discussed but

This paper deals with the simulation of the tire/suspension dynamics by using recurrent neural networks (RNNs). RNNs are derived from the multilayer feedforward neural networks, by adding feedback connections between output and input layers. The optimal network architecture derives from a parametric analysis based on the optimal tradeoff between network accuracy and size. The neural network can be trained with experimental data obtained in the laboratory from simulated road profiles (cleats). The results obtained from the neural network demonstrate good agreement with the experimental results over a wide range of operation conditions. The NN model can be effectively applied as a part of vehicle system model to accurately predict elastic bushings and tire dynamics behavior. Although the neural network model, as a black-box model, does not provide a good insight of the physical behavior of the tire/suspension system, it is a useful tool for assessing vehicle ride and noise, vibration, harshness (NVH) performance due to its good computational efficiency and accuracy.

<div class="section abstract"><div class="htmlview paragraph">A multi-objective optimal design of a brushless DC electric motor for a brake system application is presented. Fifteen design variables are considered for the definition of the stator and rotor geometry, pole pieces and permanent magnets included. Target performance indices (peak torque, efficiency, rotor mass and inertia) are defined together with design constraints that refer to components stress levels and temperature thresholds, not to be surpassed after heavy duty cycles. The mathematical models used for optimization refer to electromagnetic field and related currents computation, to thermo-fluid dynamic simulation, to local stress and vibration assessment. An Artificial Neural Network model, trained with an iterative procedure, is employed for global approximation purposes. This allows to reduce the number of simulation runs needed to find the optimal configurations. Some of the Pareto-optimal solutions resulting from the optimal design process are analysed. They show high improvements of the performance indices with respect to a reference solution.</div></div>