MUMBAI, India, July 13 -- Intellectual Property India has published a patent application (202641084183 A) filed by Srm University- Ap on July 09, 2026, for Zernike Radial Neural Network-Based Adaptive Control System And Method For Dc-Dc Buck Converters.
Inventors include Chakravarty, Arghya; Gangula, Sasank Das; Averneni, Vijayasri; Kumar, Chavali Pavan; Nizami, Tousif Khan; and Momin, Mohammed.
The application for the patent was published on July 10, 2026, under issue no. 28/2026.
Abstract: A control system for a DC-DC buck converter (100) includes a switch (102) for controlling input current flow, a diode (104) for providing a freewheeling current path, an inductor (106) for energy storage and current smoothing, capacitor (108) for output voltage filtering, a voltage sensor (110) for measuring an output voltage, a current sensor (112) for measuring an inductor current, a constant power load (114) connected to an output, a resistive load (116) connected in parallel with the constant power load (114), and a controller (120). The controller (120) includes a Zernike radial neural network (122) having an input layer (124), a basis function layer (126) with Zernike radial polynomial basis functions, and a calculation layer (128). The Zernike radial neural network (122) receives state error signals and generates an estimated uncertainty signal representative of unknown load power demand. An adaptive backstepping controller (130) compensates for the estimated uncertainty signal and generates a control signal for regulating the output voltage. (Figure 6)
Disclaimer: Curated by HT Syndication.
