Processing content query requests for CSAF documents using a graphQL-based API

Abstract

 The issue of optimizing the combustion process is quite relevant for modern thermal power engineering and arises even more in the context of the restoration and reconstruction of the energy infrastructure of Ukraine. The use of computer vision tools will allow directly determining the qualitative characteristics of the gas combustion process in real time without relying on measurements of secondary parameters, the change in values ​​of which is quite inertial. The process of turbulent torch combustion itself, which occurs in industrial boiler units, is poorly studied and does not allow for its modeling in advance. Therefore, an important stage in the construction of any control and management system is the identification of the object by conducting experimental measurements, on the basis of which the parameters and forms of dependencies will be determined to obtain a mathematical model of the process. In the course of the current work, the process of identifying the methane torch combustion process inside the simulated installation of the boiler unit furnace (combustion chamber) is considered. According to the results of a series of experiments, it was possible to obtain the dependences of the characteristics of the visual manifestation of the combustion process, recorded by a video camera, on its current combustion mode, which was determined by the satisfaction of the stoichiometric ratio of gas and air supply to the burner with preliminary mixing of the gas-air mixture. In general, the dependences of the spectral composition of the radiation, the area of ​​the torch and its luminosity on the combustion mode were obtained.  The dependences are characterized by significant nonlinearity when the torch transitions to the mode of significant chemical underburning, but are easily linearized in the area of ​​interest, which is located near the point of satisfaction of the stoichiometric ratio. The results of this article demonstrate the suitability of the parameters obtained by computer vision for their use in traditional control systems. Taking into account the speed of obtaining such data, we can conclude that it is appropriate to create a control system based on combustion parameters completely obtained by computer vision methods and means.