Selena is a universal software for design and analysis of buildings and industrial structures. It allows to construct analytical models of any degree of complexity, perform strength and buckling analysis and optimize parameters of dynamic systems.

Static analysis can be performed once for a series of independent loadings. The result of static analysis are nodes displacements, forces in the beam elements, stresses in the beam, plates, shells and three-dimensional elements, the diagrams nodes equilibrium, supporting reaction of the construction. The results of static analysis can be displayed either on the individual loadings, or by combinations of loadings.

The calculation is performed in several iterations. In the first step, the static calculation for a given combination of loads (a request for the desired combination of loads shall be issued before the beginning of the analysis). Forces in structural elements, identified in this iteration, served as a parameter load for the next iteration, etc. The calculation can be extended to as long as the difference between the parametric load between successive iterations is more than a given error, or do not exceed the specified number of iterations.

The result of analysis (taking into account the longitudinal bending) are nodal displacements, forces in the beam elements, stresses in the breams, plates, shells and three-dimensional elements, nodes equilibrium diagrams, supporting reaction of the construction.

The linear buckling analysis permits to determine the buckling critical parameters of the global instability of the structure. The calculation is performed by iteration - all loads are multiplied by the same factor up to pass by the structure state of instability. The result of linear buckling analysis are the critical buckling parameter (the safety factor for stability), as well as buckling lengths and flexibilities of the beam elements.

Nonlinear analysis per stability it is advisable applied to structures with initial imperfections. The peculiarity of analysis of such structures is that the ratio between the longitudinal forces (stresses) in the loading process in the individual elements can vary considerably (the shells include to this category primarily). That is, in contrast to the linear analysis, forces (stresses) in the structure elements at the moment of loss of stability cannot be obtained by multiplying the forces (stresses), calculated at some initial state, on the same factor. Nonlinear buckling analysis specifies the stresses state of structure by recalculation it taking into accounts a longitudinal bending of elements at each stage of loading.

The peculiarity of beam structures is that its elements are almost always lose stability earlier than predicted by the linear analysis. The reason of this is the presence of small initial random distortions and attachment eccentricities of beams. Therefore, even if beam is central compressed, the bending moments appears, and if fiber stresses at some point exceed the yield strength of the material, the loss stability process begins to develop with catastrophic speed. The program offers a special mode analysis with a slightly curved beams. Depth of irregularities can be calculated either in accordance with Eurocode 3, or SNiP II-23-81, or CP 16.13330.2011. The result of analysis are the maximum fiber stress in the rods. Thus, to assess the stability of the beam element it is sufficient to compare obtained fiber stress with yield stress of the material from which beam element is made.

Practice shows that the virtually calculated by the program values of the buckling lengths can be used for no more than 5-10% of the rod elements included in the structure. For the remaining elements, the calculator must refer to various reference books and standards, which do not always cover all possible cases. Unfortunately, only a very experienced designer can know in this situation, understand when you can use the data provided by the analysis, or refer to reference books. Therefore, checking the local buckling of the rod elements often resembles fortune-telling on the coffee grounds.

Physically, the buckling length of the rod is the distance between a pair of neighboring nulls in the bending moment diagram of rod subjected to buckling. This, it is considered, an arbitrary rod is brought to the pinned ends rods. To calculate the buckling length of an individual rod, it suffices to solve for it the problem of buckling about eigenvalues under known boundary conditions of the rod. The last condition is the main problem of finding the buckling lengths - for the simplest types of rod fastening (hinged, cantilevered, etc.) these conditions are well known, while for an arbitrary rod, which is an element of a large structure, this is not an easy task. In general, in order to establish the effect of the rest of the structure on the rod, a procedure is required that is equivalent in cost to one static analysis. It is clear that if the structure is large enough, the task becomes almost unaffordable. Therefore, in order to somehow get closer to the solution of this task, they go to a trick - they make a calculation of the overall stability of the structure, proportionally changing the forces in all its elements. As a result of one such calculation, a critical force is obtained for each component of the structure, turning a blind eye to the fact that the boundary conditions at the ends of a concrete rod are provided by the reaction of a weakened structure - if the greater the forces in the elements of the structure, then weaker it resists to shifts and turns. As a consequence, the calculated lengths are always overstated.

Within the framework of this project, a unique algorithm has been developed, which makes it possible to very quickly find the buckling lengths of all rod elements of the structure corresponding to the real stress state. That is, for each rod, the real resistance of the remaining part of the structure at the actual stress state is taken into account. Thanks to this approach, the problem of "huge buckling lengths" disappears. The buckling lengths can be calculated independently of each other for both main section axes. The concept of "buckling length" can be quite naturally extended to unloaded and even stretchable elements. The calculated buckling lengths can be automatically transferred to the stage of checking and selecting of sections, which allows you almost completely automate this procedure.

The obvious disadvantage of this model is a significant increase in the size of the model (the number of variables going on the base description may significantly exceed the number of variables of the main structure). At present there are several mechanical models of the base allowing on the one hand take into account the distribution properties of the soil, and on the other - do not require large computational cost. One such model is the base model with two coefficients of soil reaction (two-parameter model). The elastic half- space model by the system of planar finite elements, whose properties depend on the underlying soil layers.

The eigenfrequencies and mode shapes can be determined in light of the longitudinal stress on the given combination of static loadings. If the calculation is performed in the regime of concentrated masses, the program will automatically reduce the masses of structural elements to the specified nodes. If necessary, the mass of structural elements can be added to these nodes.

The fundamental difference between systems with distributed masses from the systems with concentrated masses is that the spectrum of systems with distributed masses is unlimited. Therefore, to perform the calculation on the free vibration in distributed masses mode, you must first specify a frequencies range, in which will the eigenvalues will be searched. For calculation in distributed masses mode the program uses the qualitative method for determining the eigenvalues of the system.

Calculation mode allows you to "adjust" concentrated masses of the system, to previously assigned frequency spectrum. One of the possible applications of this mode - specify the masses of the system by the measured frequencies. But a more likely application is to fit frequencies of the system with lumped masses to the frequencies of the system with distributed masses.

A number of dynamic calculations may be executed only in the concentrated mass. On the other hand the distributed mass analysis makes it possible to determine the natural frequencies with any degree of accuracy. Thus, the definition of mass analysis (mass fitting) provides a convenient opportunity to move properly from systems with the distributed masses to the systems with concentrated masses.

As the effects may be a harmonic oscillation, inertial effects (the amplitude of the impact proportional to the square of the frequency), the impact of wind pulsation (as the amplitude of the dynamic effects selected wind pulsation component) and wind response (as the amplitude of the dynamic effects selected resonant load acting across the wind flow and resulting by the disruption of alternating vortices over the construction mast or tower type).

The calculation is performed by expanding by the forms of vibration. The individual coefficient of energy absorption may be done for each vibration mode.

The calculation of dynamic absorbers of vibration based on the assumption that the added mass interacts with each harmonic of vibrations separately. In this case the problem reduces to n independent tasks (n - number of dynamic degrees of freedom of the basic structure), each of which represents a system of two masses - the reduced mass of the main system to the point of damper suspension by the i-th form of natural oscillations and the mass of absorber. Since in practical problems damper is set to one of the frequencies of the main system, the influence of this particular mode will be dominant, and then the error of the absorber interaction with other modes at a given frequency can be neglected.

This mode of calculation is exploratory. It allows you to estimate the parameters of the dynamic absorbers before they are introduced into the construction. To check the absorber setting by precise methods you should use the option "Forced harmonic vibrations”."

This version allows you to perform dynamic analysis of structures on a range of effects, given its law changes over time. As effects can be selected as follows:

• ree oscillations with given initial static and dynamic conditions
• impulse
• impulse of finite duration
• triangular impulse
• suddenly applied load
• load, increasing linearly with time up to a certain point and then remains constant
• sinusoidal effect of finite duration
• sinusoidal effects of finite duration + suddenly applied force
• series of repetitive rectangular impulses
• sawtooth load
• harmonic excitation with amplitude increasing linearly with time up to a certain point and then remains constant
• harmonic excitation with frequency increasing proportional to the time and amplitude increasing with the square of time up to a certain point, after which the remaining constant (simulation of the pickup of the engine)
• load, defined by a polynomial in powers of time
• package of sinusoidal waves
• stationary random process, the spectral function is given by a polynomial in powers of frequency
• stationary random process, the spectral function is determined by the Davenport spectrum
• stationary random process, the spectral function is determined by the Panofsky spectrum
• seismic effect

Dynamic disturbance can be set either by amplitude values of concentrated forces applied to the nodes (the amplitude values of the forces are multiplied by a function of time, given by type of impact), or by the dynamic displacements of supports (a function of acceleration of the base is given in this case).

For stationary random processes (wind pulsations, seismic, etc.) the program itself implement a random process on the basis of its spectral density.

Thus, it is possible to build an arbitrarily long chain of solutions, each of which sequentially transfers its final state, as the initial conditions for a subsequent process.

It can be carried out calculations on a series of independent harmonic disturbances with different frequencies of exposure at the same time. Forced harmonic analysis allows taking into account the frequency-independent internal friction in the structure elements. It can be set individually the absorption of energy for each structural element. The dynamic vibration absorbers can be connected to the system in dynamic harmonic analysis.

This mode of analysis allows you to reduce the results of such measurements to some single given temperature. The next step may be issuing recommendations for bringing the structure into proper condition. This task is successfully solved by calculating the Rope tension adjustment mode, which allows you to build a rope adjustment program that provides reduce the system to a design state in a countable number of steps.