The dissertation has a volume of 80 pages, the main part consists of an introduction, five sections, contains 29 figures, 25 tables, 4 annexes and 13 sources of literature.

The purpose of the research is to investigate the possibility of using the neural networks for decision-making process of mechanical impedance analysis.

The object of the research is the methods of machine learning for detecting flaws by mechanical impedance analysis of composite materials.
The subject of the research is the decision-making process based on the analysis of the informational parameters of the mechanical impedance defectoscope transducer.

In the first two sections of the dissertation, an analytical review of existing testing of composites, major flaws in them, and description of existing devices and systems that implement this method are carried out. Also the physical foundations of the mechanical impedance testing method is described.

The following sections reveal the process of forming an array of input data for research, the development and testing of the neural network, and graphs that show the dependencies of errors on the parameters of the neural network. To investigate the ability of the neural network to learn on the data and to further classification of the testing object’s area on the basis of the presence or absence of a defect, the creation and testing of the neural network was carried out using the Keras library based on the Python programming language, which confirmed the expediency of using this method of information processing in the mechanical impedance testing of compositional materials.

Research advisor: prof. Suslov E.

Download

All Masters Thesis

Abstract

The dissertation has a volume of 80 pages, the main part consists of an introduction, five sections, contains 29 figures, 25 tables, 4 annexes and 13 sources of literature.

The purpose of the research is to investigate the possibility of using the neural networks for decision-making process of mechanical impedance analysis.

The object of the research is the methods of machine learning for detecting flaws by mechanical impedance analysis of composite materials.

The subject of the research is the decision-making process based on the analysis of the informational parameters of the mechanical impedance defectoscope transducer.

In the first two sections of the dissertation, an analytical review of existing testing of composites, major flaws in them, and description of existing devices and systems that implement this method are carried out. Also the physical foundations of the mechanical impedance testing method is described.

The following sections reveal the process of forming an array of input data for research, the development and testing of the neural network, and graphs that show the dependencies of errors on the parameters of the neural network. To investigate the ability of the neural network to learn on the data and to further classification of the testing object’s area on the basis of the presence or absence of a defect, the creation and testing of the neural network was carried out using the Keras library based on the Python programming language, which confirmed the expediency of using this method of information processing in the mechanical impedance testing of compositional materials.

Hits: 1028

Actuality of theme

At present, special attention is paid to automation of the processes of measuring control of product parameters, since the level of automation largely determines the economic costs and the quality of management of various technological processes. When creating modern measuring systems it is necessary to focus on the use of specialized devices, which in turn introduces certain additional difficulties in the design of automated control devices for many product parameters.

In a number of industries it is necessary to control the homogeneity of the electrophysical properties of conductive materials in products of complex forms. The heterogeneity of the conductive material is most often determined by the presence of defects that arise in the manufacturing process in violation of the production technology. For detected defects it is important to know both the volume and their orientation in space (local or elongated along the axis or radius). Rejection of products with detected defects is carried out according to the size of detected heterogeneity of the controlled material of the product.

The task of control is that when the object is controlled relative to the converter on the elements of the signals to draw a conclusion about the location of the defect in the volume of the product, its nature and size.

The volume of the defect is determined, first of all, by the amplitude of the signal. Signal the specific element of the converter depends on its location relative to the defect. Case where the defect is placed directly under one of the elements - the simplest. The volume of the defect can be judged directly in the amplitude of the signal from this element. If the amplitude of the signal exceeds the set threshold level, the defect is inadmissible the volume and object of control must be rejected. But if the same defect of inadmissible volume is located between the elements of the converter, the amplitude of the signal of each of them will be less than in the previous case, and may not exceed the threshold level, and the product must be rejected. Therefore, the development of an algorithm for the processing of data obtained with the help of differential converters, which allows to determine the volume of the defect regardless of its location relative to the elements of the converter, an urgent task of non-destructive testing.

The purpose and objectives of the study

The purpose of the research is to develop an automated system of eddy current control and simulation of signal formation processes in the system object control - eddy current converter

  • Run simulation of eddy current fields in object control
  • To substantiate the general structure of the system of eddy current control of products of complex geometry

Object of study
The object is the process of automated non-destructive control of products of complex geometry by means of eddy current defectoscopy.

Subject
Methods and tools for automated eddy current control of products of complex geometry

Scientific novelty

  • The structure of the automated eddy current control is proposed, which combines the technical vision, eddy current control and industrial work into a single complex of means for the achievement of the sole purpose of defectoscopy of complex geometry products from conductive materials of complex shape and defect coordinate definitions
  • Modeling. Adapted the program to the task of calculating the parameters of field elements of vortex currents in the complex control object
    The general requirements for systems of robotic eddy current control of products are formulated.

Research advisor: prof. Kuts Y.

Download

All Masters Thesis

 

Hits: 632

This master's thesis consists of 90 pages, 40 illustrations, 25 tables and 18 literary references.

In the master's dissertation was carried out research of possibilities of construction of an ultrasonic system for control of bones of a forearm. A system has been proposed that includes six linear antenna arrays, active groups of linear antenna arrays can be used as phase antenna arrays.

Possible sound schemes in the system are investigated. Two operating modes of the system are proposed, the first is the definition of the presence of a crack, the second finding the size of the displacement. The study found that the most effective is the mirror-shadow method, but it greatly complicates the algorithm of the system, and it requires initial setup. Also, in this system, radiation in two directions simultaneously at different frequencies is proposed in order to increase the reliability and speed of control in the inclined beam. The modern element is analyzed, which satisfies the requirements of the system with the help of broadband composite sensors.

Hits: 627

The master's dissertation consists of the introduction and 5 sections, the conclusion and the list of used literature. The full volume is 95 pages, including 32 illustrations, 28 tables, 15 literary sources and 4 appendices.

Topicality of the topic: The temperature often use as an information parameter for the diagnosis of industrial equipment, it characterizes the state of this equipment. The temperature have also particular importance during control of technological processes. The accuracy of temperature regime often determines not only the quality of the product, but also strategically important opportunities for using it for specific purposes. Among the most common devices used to measure temperature, pyrometers can be called, which allow to get the value of temperature without contact at a given point. In some cases, one point for temperature measurement is not enougt. Often, there is a need for a complete thermogram of the object. For this purpose, a device such as a thermal camera can be used to visualize the temperature distribution on the object's surface. However, compared with a pyrometer, the thermal imager is a complex and, accordingly, an expensive device that does not always justify its value.

Hits: 630

This master's thesis consists of 85 pages, 50 illustrations, 35 stamps and 24 literary references.

In this dissertation features of the use of orthogonal methods of measuring signal parameters during the design of eddy-current flaw detectors based on microcontrollers are considered. The author analyzes the new structure of a flaw detector, which consists of one measuring channel, and implements an orthogonal measurement method, has very low power consumption, small dimensions and corresponding cost. Experimental investigations of the proposed structure of the eddy current flaw detector on a special experimental model were carried out, where simulation of possible defects that could result both in phase change and amplitude of the measurement signal with the help of the high-precision synthesizer of signals SDG102, firm SIGLENT was carried out, which confirmed the high metrological capabilities of this structure.

Hits: 645

  1. Redka M. Development of information technology of ultrasonic flow detection with increased relability
  2. Pohrebenko D. Methods and means for rehabilitation of the ankle joint
  3. Shynderuk T. Research of correlation method for acoustic signals location
  4. Ukrainets S. Acoustic testing technology using GSM

Page 23 of 26

АСНК КПІ ім. Ігоря Сікорського, 2021