Lighting Control Optimization of BP Neural Network Based on Improved Sparrow Search Algorithm
This study addresses the challenge of intelligent lighting control algorithms prone to local op-tima by designing an improved sparrow search algorithm(ISSA).ISSA enhances optimization by integra-ting Tent chaotic population initialization,Cauchy variation,and historical memory function.The research introduces a BP neural network-based illumination model to predict lamp dimming coefficients.ISSA opti-mizes this model by refining neural network weights and thresholds,thereby boosting prediction accuracy.Benchmark tests of ISSA's performance and dimming optimization experiments across multiple algorithms were conducted.Compared to particle swarm optimization and the standard SSA,the ISSA-BP based ligh-ting control demonstrates superior precision in determining optimal dimming ratios.At the same time,it ful-fills average illuminance,uniformity,and glare standards,significantly enhancing energy efficiency.
intelligent lightingSSAchaotic population initializationCauchy variationhistorical memory function
NETL
NSTL
万方数据
