Abstract: Conventional mixed alkali titration analysis suffers from subjective endpoint determination at the first equivalence point and significant variability in triplicate measurements. This study employs Color-Discriminable Artificial Intelligence (CDAI) for endpoint detection, combining precise hardware control with intelligent algorithms to establish an objective, quantifiable dynamic color discrimination model for phenolphthalein and methyl orange indicators. The developed system effectively resolves endpoint determination challenges and improves measurement reproducibility. This work demonstrates an innovative intelligent titration methodology, offering a scalable solution for the digital transformation of chemical laboratory education.

Key words: Mixed Alkali, Titration, Python, Artificial Intelligence, Color-discrimination