No abstract

The authors look into the possibility of using artificial neural networks for predicting the deformation characteristics of steels (the parameters of the Basquin-Manson-Coffin strain-life curve equation) based on static strength and plasticity characteristics, by constructing four independent neural networks with different configurations of input and output data. The prediction of parameters of the Basquin-Manson-Coffin equation and the fatigue life calculations by means of artificial neural networks are demonstrated to provide a better accuracy in comparison to the available conventional methods. Keywords: artificial neural networks, parameters of the Basquin-Manson-Coffin strain-life curve equation, fatigue life, static strength and plasticity characteristics.Introduction. In Part 1 [1] we reviewed the currently available methods for assessing the strain-life curves (the parameters of the Basquin-Manson-Coffin equation) and revealed some drawbacks of their definition. Here we will study the possibility of predicting the parameters of the strain-life curves by means of artificial neural networks (ANN).ANN Concept Description. The neurocomputing methodology is a relatively new area of artificial intelligence, where attempts are made to simulate the structure and operation of biological neural systems such as the human brain by constructing ANN in a computer. The use of ANN is especially helpful where one encounters the problems for which the solution is not clearly formulated or where one has to reproduce some mechanism which is otherwise difficult to describe using the available physical models.The methodology of this approach with a comprehensive description of neural networks, considering the multidiscipline nature of this subject, was detailed in [2]. Here we will just outline the ANN background and look into the possibility of using ANN in solving the applied engineering problems.Artificial neural networks consist of simple interconnected elements referred to as the processing elements or artificial neurons, which act as microprocessors. Thus, a neuron represents an information processing unit in a neural network. Figure 1 shows a model of a neuron used as a basis of ANN [2], where three main elements can be identified:1. A set of synapses (or connections), each characterized by its peculiar weight or connection strength. In particular, a signal x j at the input to synapse j connected to neuron k is multiplied by weight w kj . As opposed to brain synapses, the synaptic weight of an artificial neuron can have either positive or negative values.2. The summator adds the input signals weighted relative to the corresponding neuron synapses. This operation can be described as a linear combination.

We describe the developed computerized data bank "Strength of Materials" which contains data on the physico-mechanical characteristics of numerous structural materials in a wide range of loading and operation conditions. The structure of the data bank and the description of its management system are presented. The data bank is a multipurpose data-processing complex intended for solving practical problems on assessment and justification of the life of main components, assemblies and units used in modern engineering structures by the criteria of strength, reliability and lifetime.Keywords: data bank "Strength of Materials," structural materials, bank structure, general information on grades of structural materials, special characteristics of structural materials, databank management system.The modern methodology for finding, justifying and extending the life of facilities, structures and equipment, for ensuring their reliability and safe operation provides for the necessity of using calculation-andexperimental methods for validating the life of main parts, assemblies and units of modern engineering objects with a considerable reduction in the extent of time-consuming and expensive experimental investigations. Important features of this methodology are special-purpose data banks containing information on general and special physico-mechanical characteristics of materials.

Based on the combination of the Fourier series expansions of variables with respect to the angular coordinate and a two-dimensional Wilkins algorithm, a numerical analytical method is developed to study the three-dimensional non-axisymmetric dynamics of multilayer thick-walled elastic spirally orthotropic cylinders of finite length. Verification of the proposed method is performed based on the comparison of the results with those obtained using the computational kernel of the LS-DYNA combined with the ANSYS/ED (an educational version of ANSYS finite element software). Some features of the dynamic behavior of a two-layer spirally reinforced cylinder under internal pulse loading produced by concentrated explosive charges are studied numerically.

The equations of two-dimensional dynamic theory of elasticity are used to study the domain of applicability of engineering one-dimensional analytic procedures developed earlier for the evaluation of the dynamic strength of cylindrical and conic matrices of finite length both with free and fastened ends subjected to stamping by high explosives. It is shown that, in view of the "buildup" effect leading, in some cases, to the violation of strength of the matrix, these procedures should be supplemented with correcting calculations performed according to the two-dimensional model for large periods of time. We propose new procedures of design-basis and verifying analyses for cylindrical and conic matrices of finite length subjected to stamping by high explosives.Keywords: cylindrical and conic thick-walled shells, matrix for pulsed stamping, charge of high explosives, Wilkins algorithm, dynamic strength, design-basis and verifying analyses.Introduction. Technological processes of stamping of axially symmetric thin-walled shells by high explosives are extensively used in machine building. In some cases, they have significant advantages over the other methods of treatment of the materials (production of large-sized components, products made of high-strength materials, etc.) [1].The main force element of the technological equipment aimed at stamping by high explosives is a matrix in the form of a thick-walled axially symmetric shell loaded with internal pulses of pressure.The stress-strain state of elastic infinitely long multilayer axially symmetric thick-walled cylinders whose inner surface is loaded by pulsed pressure is theoretically studied in [2-10]. Thus, in [2,4,8], approximate analytic solutions for a two-layer cylinder are obtained by using the condition of incompressibility of the material under the action of internal pressure in the form of a semisinusoidal pulse, a Heaviside function, and an exponential pulse. These loads are typical of stamping by press guns and gas-detonation presses, stamping with high explosives, and electrohydraulic stamping. The comparison of these solutions with the data of numerical calculations according to the characteristic-difference and Wilkins methods [7, 10] made it possible to establish the domain of applicability of the approximate dependences and develop the engineering (analytic) procedures of design-basis and verifying analyses [12] as well as the numerical procedure of verifying analysis for sufficiently long matrices of cylindrical and almost cylindrical shapes [11].The Branch Standard of the Ministry of Aircraft Industry of the USSR [12] specifies, in particular, the design-basis dependences and the procedures of design-basis and verifying numerical analyses of technological equipment [matrices for pulsed shaping of closed cylindrical shells, gently sloping shells of ogival and conic shapes, and matrices for stamping shells of all indicated shapes with circular stiffening elements (crimps, ridges, etc.)]. These procedures are based on the engineering methods for the sol...