Harnessing Hybrid Random Search Algorithms for Intelligent State Control in Technological Processes

This article presents a research focused on improving the control of stochastic search procedures within an ε-neighborhood surrounding current monitoring outcomes of control parameters in technological processes. The research aimed to enhance the effectiveness of control measures through the utilization of state-of-the-art Random Search (RS) technology. A mathematical structure was established to define an ε-neighborhood for manipulating parameters and setting boundaries for the search space. Various methods were explored for selecting manipulation parameters using operator-driven processes via Human-Machine Interface (HMI) tools. The RS-Control Module's functional structure accounted for the dynamics of controlled parameter evolution using a sliding observation window. The RS-Optimization Module's software implementation allowed for forecasting output parameters and assessing forecast accuracy using quality indicators. The implementation of the main program provided a user-friendly interface for adjusting process parameters, optimizing criteria, and monitoring control based on RS forecasts. The research demonstrated the effectiveness of the proposed approach in improving output and technological parameters in industrial processes.