<?xml version="1.0" encoding="UTF-8"?><paper><paperId>PAP-001</paperId><title>Deep Learning for Smart Grid Fault Detection</title><conference>Journal of Applied Computing (Demo)</conference><abstract>We present a convolutional approach to fault detection in smart electrical grids, evaluated on three public datasets. The model detects line faults 40 percent faster than baseline methods while retaining precision above 0.95. This paper tests the Technology Fist publication pipeline end to end.</abstract><authors><author><name>Aisha Researcher</name><affiliation>Test University</affiliation></author></authors><doi>10.0000/pap-001</doi></paper>