Copyright © 2026 Authors retain the copyright of this article. This article is an open access article distributed under the Creative Commons Attribution License which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

@article{202507,
        author = {N.D. Gaikwad and Sneha Rawool and Swara Pingale and Shiv Gathe and Ganesh Pawar and Aditya Bhale},
        title = {Piezoelectric Tile with Machine Learning integration for intruder (fraud) detection},
        journal = {International Journal of Innovative Research in Technology},
        year = {2026},
        volume = {12},
        number = {12},
        pages = {7319-7333},
        issn = {2349-6002},
        url = {https://ijirt.org/article?manuscript=202507},
        abstract = {Piezoelectric energy harvesting provides a practical and sustainable method for converting repetitive mechanical pressure from human footsteps into usable electrical energy for low-power applications. Conventional footstep power generation prototypes predominantly demonstrate basic voltage generation, whereas real-world public installations additionally demand mechanical durability, reliable sensor readings, and robust protection against false energy reporting and data manipulation. This paper presents a Smart Piezoelectric Tile System that integrates spring-assisted compression, a series-parallel piezoelectric sensor array, a bridge rectifier with energy storage, and a machine learning-based fraud and anomaly detection module. The proposed system harvests clean energy from footstep pressure and simultaneously validates whether measured voltage, current, step count, and storage behavior conform to physically realistic patterns. The prototype employs piezoelectric discs mounted beneath a spring-supported tile platform, a full-bridge rectifier, a filter capacitor, a rechargeable storage element, voltage and current sensing circuits, and an ESP32 microcontroller for data acquisition and logging. A machine learning inference layer analyzes features including voltage peak, current peak, pulse interval, battery state-of-charge delta, and tile identity to classify operational events as normal, low generation, loose connection, artificial tapping, bypassed storage, or abnormal impact. The key novelty of this work is the integration of clean energy harvesting with intelligent reliability monitoring, making the tile significantly more suitable for actual deployment than basic power-generation prototypes. The system maps directly to UN Sustainable Development Goal 7 (Affordable and Clean Energy) and SDG 9 (Industry, Innovation and Infrastructure). Expected outcomes include stable demonstration-level energy harvesting, structured anomaly classification, and improved maintenance insight for smart public infrastructure.},
        keywords = {piezoelectric energy harvesting; smart tile; footstep power generation; anomaly detection; fraud detection; Isolation Forest; Random Forest; ESP32 microcontroller; IoT smart infrastructure; sustainable energy; machine learning.},
        month = {May},
        }