Experimental and Machine Learning Approaches for Discharge Estimation in Arched Labyrinth Weirs

A weir is a hydraulic structure installed across or parallel to an open channel to regulate or measure flow discharge. Weirs are classified based on shape, discharge characteristics, crest width, and crest type. Among these, the arched labyrinth weir is characterized by its extended crest length, achieved through arching and notches, which enhances its hydraulic efficiency. This study presents a comprehensive analytical and experimental investigation into the discharge characteristics of arched labyrinth weirs in open channels. The artificial neural network model demonstrated exceptional predictive performance, achieving a mean squared error of 2.84×10−7, a mean absolute percentage error of 0.000191, and a correlation coefficient approaching 1, highlighting its accuracy and reliability in discharge estimation.