“AI-Based Multi-Sensor Fusion for Real-Time Heavy Metal Detection in Water Using IoT”

Abstract: Water contamination by heavy metals poses a serious threat to human health and aquatic ecosystems. Conventional laboratory-based techniques such as Atomic Absorption Spectroscopy and Inductively Coupled Plasma Mass Spectrometry provide accurate results but are expensive and unsuitable for real-time field monitoring. This paper presents an enhanced IoT-enabled heavy metal detection system that combines colorimetric sensing using a TCS3200 color sensor with UV–Visible spectroscopic analysis for improved quantitative accuracy. Selective chemical reagents are used to induce characteristic color changes in the presence of zinc (Zn²⁺), copper (Cu²⁺), and nickel (Ni²⁺) ions. The RGB responses are captured using the TCS3200 sensor for rapid screening, while spectral absorbance data obtained from a compact UV–Visible spectrometer enables wavelength-based concentration estimation using the Beer–Lambert principle. An ESP32 microcontroller performs edge-level processing and transmits results to the ThingSpeak cloud platform for real-time monitoring and visualization. Experimental validation demonstrates improved sensitivity, selectivity, and concentration estimation capability compared to RGB-only detection. The proposed hybrid system offers a low-cost, portable, and scalable solution bridging IoT-based monitoring and spectroscopic analytical techniques.

Keywords: Heavy metal detection, UV–Visible Spectroscopy, TCS3200, IoT, Colorimetric analysis, Water quality monitoring, ESP32, Spectral absorbance.