Levchenko O. Automated control of the technological process for chemical milling using eddy current array probes

The dissertation is devoted to the development and research of an automated control system for the chemical milling process using eddy current array probes. Modern industrial production requires improved automation methods for production processes and advanced technical solutions to optimize technological conditions. Successful automation of industrial processes integrates achievements from various fields such as automatic control theory, information technologies, and non-destructive testing. A two-circuit automatic control system for chemical milling (CM) of aluminum alloy products was developed. The first circuit controls pickle temperature to maintain etching speed. The second circuit stops the process based on product thickness measurements using ECT. The concept of integrating ECA tools into APCS is further developed. ECA technology allows control of etching on large product surfaces, increases efficiency, optimizes resource use, and ensures high product quality. Advantages of ECT technology for chemical milling APCS include: Converting non-electrical parameters to electrical signals No mechanical contact with the object Allowing pickle access to the entire product surface Operation in a wide range of technological parameters Control based on product thickness measurements The potential of ECT for APCS use is not fully explored. Using ECA probes is promising for creating ECT devices, increasing testing performance and detailed analysis. This is achieved by reducing testing time, optimizing signal processing, adaptive change of excitation parameters, and applying modern data processing algorithms. Information parameters of ECA signals are amplitude, phase shift, and frequency, usually represented as hodographs. This creates difficulties for ECT automation in process control systems. Signal parameters are often sensitive to external influences, limiting their direct use for forming control actions. There is an urgent task to develop new automated methods and tools for ECT with ECA probes for APCS. A preliminary analysis of ECA signals identified unused resources of ECT for APCS use. The work analyzes recent publications on ECT automation, considers including ECT subsystems in APCS, and develops a multidimensional model of ECA signals using the discrete Hilbert transform (DHT) for complex signal analysis. The structural and logical diagram of ECA signal processing is considered. Using DHT in ECT-based control systems allows obtaining more digital information about objects, calculating signal characteristics, and expanding possibilities for forming control signals. Prototyping technology for ECT tools is considered. A prototype ECT system for chemical milling control was developed on the Red Pitaya platform. The system responds to material heterogeneities and generates control signals. The software architecture includes data processing algorithms. An ECA probe with flat spiral coils was developed using printed circuit board technology. A methodology for processing ECA signals, including normalization of complex transfer coefficients, was developed to increase sensitivity to technological parameters in APCS. Experimental studies confirmed the high performance of ECT technology with ECA probes. The effectiveness of ECA is evaluated using an efficiency coefficient accounting for reduced testing time and increased sensitivity. A prototype ECT system using ECA was created, implementing the proposed methodology. Tests on samples with various defects confirmed its effectiveness in flaw detection. Prospects for using neural network technologies for synthesizing automatic control systems based on ECT are substantiated. Using artificial neural networks can increase APCS efficiency in the absence of direct links between object characteristics and ECA signal parameters.