APPLICATION OF MAGNETOELASTIC EFFECTS FOR STRESS ASSESMENT AND RISK MITIGATION IN CONSTRUCTIONS

Stress assessment in the steel constructions is crucial for risk mitigation. NDT methods are widely used in such applications. However, many of these methods are hazardous, due to the radiation or chemicals used. The proposed magnetoelastic method is neutral to the natural environment. In this paper, development of such a method of stress assessment in constructional steel samples is presented. Measurements of the magnetic hysteresis loops of the sample members were performed under varying mechanical load on the specially designed steel truss test stand, which allowed to obtain magnetoelastic Bm(σ)Hm characteristics. The results confirm the feasibility of the new method of NDT stress assessment in the steel truss constructions.


INTRODUCTION
Stress assessment in the steel constructions is very important for maintenance and risk mitigation during the given object lifetime.Widely used in such applications are various NDT methods, many of which are well developed and commercially available.However, many of these methods are hazardous for the environment and workers, for example, radiation method use very strong X-rays, dye penetrant inspection can utilize harmful chemical compounds.
There are also NDT methods neutral to the natural environment, one of them is the magnetic method based on the magnetoelastic effect.The magnetoelastic effect is also called the inverse magnetostrictive effect or Villari effect.It is connected with changes of flux density B (for given magnetic field H) under the influence of mechanical stresses σ in the material.It is most evident as the change of the shape of magnetic hysteresis loop for the given material.The extreme at the B(σ)H relation is known as the Villari point [1].Magnetoelastic effect may be utilized in the construction of robust mechanical stress/force sensors [2,3].
Moreover, this effect creates new possibilities of constructions monitoring and risk mitigation.On the basis of observation of changes in magnetic properties of constructional steel, value of stresses may be calculated.The changes of microstructure of constructional steel significantly influence its magnetic properties.These effects were previously successfully utilized for pipelines monitoring [4].On the other hand, the possibility of application of magnetoelastic Villari effect for stress assessment and risk monitoring in large constructions (such as trusses) was still not verified.Steel trusses are widely used for medium and big scale constructions, but without proper maintenance they are prone to catastrophic failures [5,6], which pose a risk to the local environment and human populations.It is especially important for energetic and civil engineering structures, such as power line pylons, bridges, etc.
In the paper investigation of the steel truss-type construction using the specially developed magnetoelastic method is presented.The main goal of the work is to develop NDT method of stress measurement in the freestanding steel structures.Utilization of the developed solutions could allow for stress assessment in the selected members of the investigated steel truss under mechanical load, without additional strain gauges.Experimental magnetoelastic measurements on the specially designed steel truss structure, with means to apply known stresses to selected truss members, are shown.

EXPERIMENTAL SETUP
In order to investigate the basic magnetic properties of the given constructional steel, three requirements have to be fulfilled.The first requirement is the closed magnetic circuit in the sample.Then the influence of the demagnetizing field on the measurements is greatly reduced, and the influence of the sample shape is nearly eliminated.The second condition is the uniform stress distribution along the whole magnetic circuit in the investigated sample.Acquiring this condition allows for the elimination of the stress influences canceling each other, which may happen when there are positive stresses in one part of the sample, and negative in another.The third, equally important condition is ensuring the distribution of the effective stresses parallel or perpendicular to the magnetic patch direction in the sample.
For the magnetoelastic tests, frame-shaped member sample was designed.The sample is presented in Figure 1.It is made of the 13CrMo4-5 constructional steel, widely used in the energetic industry.On the side columns of the sample, both sensing and magnetizing windings were made.It is best to wound magnetizing and sensing windings on both columns.Moreover, sensing winding should be located under the magnetizing winding, to reduce demagnetization effects.In the presented research, sample was wound with 260 turns of magnetizing winding (130 turns in each column), and 100 turns of sensing winding (50 turns in each column of the frameshaped sample).
The truss shown in Figure 2 was developed and used for testing, under varying mechanical load.Three central members of the constructed truss were the test samples.Their relatively smaller cross sections dimensions were chosen in order to carry out tests without damaging the rest of the members.The load was applied, and the stresses in the member samples were calculated on the basis of load, truss geometry, and sample members dimensions.The load was incrementally increased, and hysteresis loop measurements were done for the calculated stresses (Table 1).Eventually, one of the compressed member samples buckled, preventing the further measurements for higher stress values.Figure 6 presents the results of measurements of stress dependence of magnetic hysteresis loops of the K02 compressed sample.Figure 7 presents the experimental results of the magnetoelastic Bm(σ)Hm for the compressed K02 sample.Under the compressive stresses, value of the flux density B in the sample nonlinearly decreases.These changes are relatively higher for lower values of the amplitude of magnetizing field Hm.The reason is the same as for the tensile stresses.The obtained characteristics are consistent with the theoretical expectations, both for the compressive and tensile stresses.

Fig. 1 .
Fig. 1.Schematic diagram of the sample members, the middle section consist of two separate columns, wound by the measuring and magnetizing windings.

Fig. 2 .
Fig. 2. Schematic diagram of the experimental steel truss: A, B -support points; C -loading point; K01 -stretched sample member; K02, K03 -compressed sample members.The hysteresis loops measurements are done on the specially developed test stand, called the hysteresisgraph (Figure 3).The test stand is controlled by the PC equipped with NI USB 6525 data acquisition card.This card is controlled by NI LabView software, for real-time control as well as data processing.Voltage sine wave generated by the data acquisition card drives the Kepco BOP 36 voltage/current converter.The output of Kepco BOP 36 is connected to the magnetizing winding of the investigated sample.Measuring winding is connected to the Lakeshore 480 fluxmeter, which measures flux density B in the sample.Voltage output of the fluxmeter is connected to the data acquisition card.As a result, developed hysteresisgraph presents B(H) hysteresis loops under digitally controlled values of amplitude of magnetizing field, as well as for different frequencies and shapes of magnetizing signals.

Fig. 1 .
Fig. 1.Schematic block diagram of the magnetic hysteresis loop measurement system.The truss was supported on the bottom edge nodes.The mechanical load was exerted vertically by the oil hydraulic press on the upper central node of the truss.During the magnetoelastic measurements, three frame-shaped samples were installed into the truss.The truss was put under the mechanical.The sample K01 was stretched with tensile stresses and samples K02, K03 were treated with compressive stresses at the same time.The influence of the stresses on the shapes of the B(H) hysteresis loops was measured for amplitudes of magnetizing field H m equal to 350 A/m, 435 A/m, 655 A/m, 870 A/m, 1310 A/m and 2180 A/m.The frequency of the magnetizing signal was set to 0.1 Hz.

Figure 4
Figure 4 presents the experimental results of measurements of stress dependence of magnetic characteristics of frame-shaped member samples.Stress dependence of the shape of magnetic hysteresis B(H) loops may be observed for different values of amplitude of magnetizing field Hm.Changes of the basic magnetic parameters are evident: flux density, remanence, coercivity.From the utilitarian and technical point of view, changes in flux density are the most interesting.

Figure 5
Figure5presents the magnetoelastic B m (σ) Hm characteristics for the stretched K01 sample.Under the tensile stresses, value of the flux density B in the sample first increase, and then, after reaching the Villari point, it starts to decrease.The characteristics are incomplete, because of the compressed member sample buckling.Moreover, these changes are relatively higher for lower values of the amplitude of magnetizing field Hm.This occurs due to the fact, that for lower values of magnetizing field H m , participation of the magnetoelastic energy in the total free energy in the sample's material is significantly higher.

Fig. 5 .
Fig. 5. Influence of the tensile stresses on the flux density of the K01 sample under various magnetizing fields: H -magnetizing field; B -flux density, σ -tensile stress.

Table 1 .
Mechanical load applied to the truss.F -force exerted by the oil hydraulic press; σ -calculated mechanical stresses in the samples.