The article analyzes and defines the main trends in the development of decision support systems. It is shown that the most urgent is the development of decision-making systems based on data mining for the development and operation of technical systems and technological processes. In order to increase the efficiency of the micro-arc oxidation process, the developed structure of the decision support system in the process of obtaining oxide coatings is presented.
The term decision support system (DSS) appeared in the early 1970s [
According to [
The trend towards the prevalence of Internet-based decision support systems has
been clearly visible since the late 1990s. Modern systems of web recommendations
are characterized by the following main trends: the use of complex methods of
multicriteria decision-making; application of technologies of virtual and augmented reality;
expansion of the cognitive functions implemented in decision support systems;
increasing the list of opportunities for decision-makers [
DSS are widely used in various fields of application such as telecommunications, banking, trade, large construction, development and operation of complex technical systems, etc.
The task of developing intelligent systems as applied to technological production
processes is urgent [
To improve the performance of decision support systems, data mining based on data manning is used.
"Data manning - automated analysis of massive data sets" - data mining -
automated analysis of an array of data sets [
"Information - information (messages, data), regardless of the form of their presentation." Data have the property of integrity, which consists in accuracy and completeness.
Data mining allows you to automatically organize the search for data in large data
warehouses, to identify and establish relationships and patterns. Data mining is based
on the use of complex models and algorithms, including those performing data
segmentation, forecasting upcoming events with a certain probability. Data mining
algorithms automatically search for and identify patterns and trends in large data
warehouses. Data mining has the following basic properties [
Data mining can provide solutions to problems that cannot be analyzed through regular and reporting techniques and queries.
Data mining systems are applied in three main areas: ─ as tools for carrying out unique scientific research, development of intelligent technologies; ─ as tools in research and development work in the development of technical systems; ─ as a mass product for business application.
In order to increase the efficiency of the technological process for the synthesis of protective oxide coatings, it is important to create an intelligent system that makes it possible to reduce the technology development time by at least 2 times due to the automated selection of optimal technological modes, development of a methodology for the synthesis of oxide coatings with specified properties. Further, an intelligent decision-making system in the process of synthesizing protective oxide coatings by the method of micro-arc oxidation (MAO) is considered. 3
The specified intelligent system consists of three main parts: hardware, information support subsystem, software (Figure 2).
The basis of the information support subsystem is a knowledge bank (Figure 3), which contains a set of the following knowledge bases and databases: ─ knowledge base of the properties of oxide coatings obtained by the method of micro-arc oxidation; ─ knowledge base of the modes of the MAO process; ─ knowledge base of models of the relationship between the properties of the resulting coatings, depending on technological modes; ─ knowledge base of principles and methods for measuring parameters of oxide coatings, as well as technological modes; ─ knowledge base of measuring instruments, their technical and metrological characteristics; ─ knowledge base about the mechanism of surface modification by micro-arc oxidation; ─ knowledge base of the requirements for the properties of oxide coatings, depending on their field of application (medical purpose, instrumentation, mechanical engineering, etc.).
The software controls, configures, calibrates the hardware, and also includes an intelligent application for coatings' synthesis; client software to support the implementation of the MAO process. The software of the microcontroller, which controls the hardware part of the research support system; server software that configures the system were developed. In order to implement the methods proposed by the authors for the controlled synthesis of MAO coatings, an intelligent application has been created [12]. Research support for the MAO process is carried out using software that processes experimental data, calculates errors in measurement results, displays data in a form convenient for users to perceive (tables, graphs).
The hardware is a computer-controlled automated system for the synthesis of oxide coatings. In turn, the automated system consists of a process current source, a power supply unit, a measuring part, a microprocessor module, and a galvanic cell located in a protective enclosure (Figure 4).
The measuring part includes channels for real-time measurement of technological parameters (electric voltage, current, temperature, brightness of micro-discharges) of the micro-arc oxidation process and parameters of the synthesized oxide coating (thickness, porosity, etc.). The microprocessor module generates control signals both for the measuring channels and for the process current source. The microprocessor module includes digital signal synthesizer, galvanic isolation unit, a microcontroller (consists of an analog-to-digital, ADC and digital-to-analog converter, DAC), an 8-channel multiplexer and a UART port; USB-UART interface converter based on FT232RL microcircuit.
Thus, the automated measuring module as part of the decision support system, controlled by software, makes it possible to establish the relationship between the parameters of the micro-arc oxidation process and the properties of oxide coatings, to select the technological modes that ensure the synthesis of oxide coatings with the given properties.
An original structure of an intelligent system for controlled synthesis of MAO coatings, including a decision support system based on data mining received in real time from the output of the hardware was developed. In turn, the hardware part consists of the following elements: a technological current source (includes a power module and a switch); power supply for low-voltage electronics; measurement and control modules, USB-oscilloscope and galvanic cell.
The decision support system programmatically implements unique models of the interconnection of influencing factors and parameters of synthesized coatings, methods for choosing the optimal technological modes proposed by the authors. The proposed system made it possible to reduce the technology development time by at least 2 times, to minimize to 0.5% the relative error in measuring the electrophysical parameters of MAO coatings, which contributes to an increase in the efficiency of the micro-arc oxidation technological process. 6
The reported study was funded by RFBR according to the research project № 19-0800425. 11. Borikov, V.: Virtual electrolyte conductivity analyzer for microplasma oxidation process control. In: 17th Symposium IMEKO TC4 - Measurement of Electrical Quantities, 15th International Workshop on ADC Modelling and Testing, and 3rd Symposium IMEKO TC19 - Environmental Measurements, 54-58 (2010). 12. Golubkov, P., Pecherskaya, E., Karpanin, O., Safronov, M., Shepeleva, J.V., Bibarsova, A.: Intelligent automated system of controlled synthesis of MAO-coatings. In: Conference of Open Innovation Association, FRUCT, no. 8711874, 96-103 (2019).