Search

The article presents a non-destructive welded joints assessment (vibrodiagnostics) method, in which, with the use of the Internet, computer networking technologies and laser displacement sensors, the quality of welded steel plates may be monitored. The article also describes the possibilities of analysis of received vibrodiagnostics data on the base of diagnostic parameters analysed in function of time and frequency. The method described in the article allows remotely registering, analysing and diagnosing mechanical systems using vibrodiagnostics and information technologies. Modern, computerized methods exhibit efficiency and reliability of results – they are currently the best measuring tools that can be used in automatic monitoring (structural health monitoring - SHM) of marine structures. It is essential to make objective assessment of the results by NDT users, which is tightly connected with the expertise of norms and regulations of NDT. It may be helpful in this respect to monitor responses from a few sensors, arranged in a few measuring points. NDT results quite often trigger further analysis of durability of an object. The NDT of welded joints method as presented in the article allows remotely and in an IT-aided way analysing and diagnosing mechanical systems with the use of vibrodiagnostics method.

The authors of the article have been looking for a new parameters and dynamic characteristics, which can be applied to non-destructive testing of welded joints. All characteristics have been based on recorded data generated during the vibration tests of welded joints with and without failures. The article deals with the methods of assessing welding joints using 2D: time – frequency dynamic characteristics. A calculation procedure used for analysing simultaneous changes of the response modules, registered by acceleration sensors was presented. Vibrations amplitudes were transformed to a function of time and frequency (simultaneously) and presented over 2D time – frequency characteristics. The analyses of the characteristics were performed for a plate without any welded joint, for a plate with non-defected welded joint and for a plate with a welded joint defected by an edge bonding. Having analysed registered 2D time – frequency dynamic characteristics it can be noticed that presenting the responses analysed simultaneously over the time and frequency allows evaluating if examined system maintains non-linearity and, at the same time, it allows to indirectly assess the quality of the welded joint. The proposed measure parameters of the quality of a welded joint can be defined as a dispersion of colours on the obtained characteristics. The faults (and the vibration nonlinearity) of the welded joints is bigger if the dispersion is greater.

Strength of the propellers with the skewback greater than 25º has to be numerically analysed according to marine classification societies. The finite element method (FEM) is advised for that kind of calculations. Classical and typical propellers (skewback < 25º) may be designed on the base of empirical equations given by the societies. The minimal thickness of the propeller blade is determined by the equations. Each classification society has their own empirical formula. Sometimes, well-designed propeller for one society has not enough strength according to the other society. What is more, propellers designed according to the empirical formulas might be not optimal. Comparative analysis of the marine propeller’s blade strength has been described in the article. Calculations of the propeller’s blade thickness have been done by two international classification societies’ empirical formulas (ABS and DNV). The results have been compared with Finite Element Method calculations (NASTRAN program). The methodology of propeller static strength vibration analyses is presented. Numerical calculation methodology is based on solid-state mechanics with loadings determined by fluid mechanics calculations. Steady state and transient fluid flow of the propeller’s working conditions were taken into account. In order to determine the optimal modelling method of the propeller several different numerical models were compared, including free model of whole propeller and single blade with boundary conditions placed in the foot. The propeller optimization was the main target of the analyses. Propeller blade thickness might be reduced after FEM method analysis - the propeller mass saving can be achieved.

The safety, comfort of the crews, stability, economics of the equipment when ship operating is the leading requirement in the field of designing and manufacturing marine structure and machinery. As a result, all parts of the ships must be tested and inspected to meet the basic safety requirements of the shipping association. The design, manufacture, testing in the maritime field in general and shipbuilding sector in particular are expensive, time consuming: such as aerodynamic experiments of the engine, collision test, ship manoeuvring, vibration test and balance of deck beams, hull beams, hatch covers, shafts...thus experimental works are sometimes impossible. Along with the development of computer science, many numerical models and software programs have been developed to solve these difficult problems. There are many numerical modelling methods, starting with the finite difference method, the boundary element method, the finite element method, the no mesh method, the weight residue or the energy method. The Work will be limited to the analysis of the most popular numerical modelling method - finite element method using Patran and Nastran software. In the first step of our research, T-beam was analysed as a part of ship hull structure (thin-walled structure). The article goes into the analysis of the accuracy of selected numerical models for the natural vibration frequency of the T-beams mounted on the plate. After modelling, calculating the natural frequency of the T-beam using the Patran - Nastran software, the results were compared with the theoretical values. From that, we evaluate the dispersion and error of different numerical models and select the optimal numerical model. Optimal model will be used for modelling full ship hull with superstructure.

The article presents untypical diagnostic of the slow-speed, marine main engine. The engine was newly made-up and checking during mounting process and during sea trial. However, a leak under piston chamber of second cylinder along the piston rod appeared few days after the sea trial. The stuffing box exchange (with the piston rod’s regeneration) did not give expected results − the failure happened again. All geometrical and exploitation parameters was in acceptable range. The ship with main engine was fifth in the series of sister ships. The authors were asked for urgent expert opinion. Mix of different measurements was planned after formulation several hypothesis. Displacements of piston rod and cylinder, a stress level of main engine body, and vibrations level in different points of main engine was performed during short sea voyage. Not a single hypothesis can be considered as the main cause of the failure. It turns out, that a serious failure need not be caused by a single reason. According to snowball theory a sum of small effects, each of them affecting slightly the engine operation, can be a cause of serious failure. In the authors’ opinion, the failure was caused by a sum of relatively slight effects. The probably scenario of the failure process was enunciated. The recommendation for the engine project was formulated: some of the geometrical tolerances should be changed.

Among non-destructive testing methods of welded joints, a vibrodiagnostic method is one of the most promising in applications intended for the continuous monitoring of the structure. The accelerometer-recorded responses provide diagnostic information that requires mathematical processing to extract the essential features typical of the tested welded joints and to evaluate their execution. For this purpose, the proposed method was based on the determination of damping decrement in function of time. Due to the complexity of the responses run, a proposed method consisted of calculating the damping decrement using the response approximation with different functions. It has been shown that the changes analysis of damping decrement applied to welded plates enables the assessment of the quality associated with the weld. Attention was also paid to the elements relevant to the study. This work presents the algorithm and exemplary results obtained from the application of proposed method to several selected sample plates with different type of welds. A scientific apparatus, which purpose was to record dynamic characteristics of welded joints was also described.