Introduction. Before repairing or reconstructing steel structures, it is necessary to obtain information about the strength capacities of the metal. The estimated service life of metal structures is tens of years, but it is known that the mechanical properties of the original metal change over time. Additionally, many facilities operate beyond these anticipated lifespans. As some researchers have noted, the challenge of obtaining such information is due to several factors. Firstly, in most cases, it is impossible to cut samples from existing structures. Secondly, the use of non-destructive testing methods needs to ensure sufficient accuracy in assessment. Thirdly, non-destructive testing may not be physically possible due to the design features of the object. Fourthly, survey work on the operating structure can be very laborious and expensive, requiring a reduction in volume and cost. Fifthly, when assessing the mechanical characteristics of the metal, it is important to apply an approach that guarantees the accuracy of results while minimizing work by utilizing previously obtained information on similar metals. Given these challenges, the development of a methodology that combines non-destructive testing with prior information is crucial.In non-destructive testing of structures, methods for qualitative assessment of the condition of metal or welded joints are used, such as ultrasonic, magnetic, and radiation techniques. There are also quantitative methods for evaluating mechanical characteristics, such as using portable hardness testers. However, most methods for assessing strength characteristics, such as yield strength and temporary tear resistance, are cumbersome and limited to laboratory settings.The methods of clarifying experimental information using a priori data by experts are conventionally divided into three categories:− according to the priority of the weight of a priori and experimental data;− extrapolation of past data to future periods;− based on Bayesian procedures.This article describes a non-destructive strength testing method based on indentation developed with the author's participation and repeatedly tested in actual surveys. The aim of this article is to justify the author's methodology to minimize the amount of required samples during survey work by combining non-destructive testing methods and Bayesian accounting for experimental information.Materials and Methods. The research plan involved analyzing experimental data on the mechanical properties of metals and developing an algorithm to minimize the number of samples of control objects. Before measuring, the metal of the structures was cleaned with a hand grinder. The method of non-destructive testing of the evaluation of mechanical characteristics according to the parameters of the impact insertion of the indenter into the surface under study was used. To minimize the amount of work, a Bayesian approach was used to reduce the variability of posterior values by utilizing additional experimental data on the mechanical characteristics of such steels. The material St3 of strength class KP 245 with yield strength of 245 MPa and tensile strength of 412 MPA was studied. Additional experimental data on this material's properties were available from a previously studied metal structure.Results. The method of non-destructive testing of the strength of metal in pipe structures has been implemented. This method used prior information obtained from previous surveys of similar materials. Based on a Bayesian approach, experimental and previous information was combined, in particular, the values of time resistance to rupture. A method for estimating the minimum required sample size of the examined structural elements was proposed provided there was minimal risk from an estimation error. As a result of calculations, it was shown that the use of such a technique was possible with a sample size of 2–3 elements.Discussion and Conclusion. The proposed methodology was developed based on an analysis of more than 20 surveys conducted to assess the strength of the existing metal structures. Using the non-destructive testing method, we were able to simultaneously determine the yield strength, tensile strength, elongation, and hardness. The article presents data on the values of tensile strength. It should be noted that although the duration of each measurement was 20–30 seconds, in some cases it took longer to inspect large structures, such as bridges, which could take weeks. The calculation performed using the proposed method, which combined experimental and pre-experimental information about one of the strength characteristics of steel, temporary tear resistance, showed the high efficiency and potential for further application in future surveys.
Read full abstract