关键词: 清洁家电, 视觉分析系统, 污渍检测, Logistic模型, PSO粒子群优化算法

Abstract: Stain cleaning efficiency on hard floors is not only a core concern for consumers when selecting floor washers, but also a key performance indicator in the formulation of relevant domestic and international standards. Based on research on machine vision-based stain detection, adopt the SCRAS (Surface Cleaning Rate Analysis System), refer to test standards IEC/ASTM 62885-6:2023 and GB/T 38048.6—2024, and carrie out cleaning tests for stubborn mustard sauce stains on ceramic tile surfaces using three types of floor washers, followed by regression modeling and analysis of experimental data. The test results reveal that the proportion of residual stains varying with cleaning cycles conforms to the Logistic curve, exhibiting a three-stage evolutionary trend: slow decline in the initial stage, sharp drop in the middle stage, and stabilization in the later stage. The Particle Swarm Optimization (PSO) algorithm is introduced to realize global optimization of model parameters, determining the cleaning speed coefficient and efficient cleaning interval of stain removal efficiency. The research can support the R&D and performance upgrading of decontamination capabilities for cleaning equipment and provide critical references for the improvement of test standards as well as the iteration and upgrading of inspection systems.

Key words: Household cleaning appliances, Visual analysis system, Stain detection, Logistic model, Particle swarm optimization algorithm