An intelligent internet of things-based artificial intelligence framework for adaptive mechanical process control and real-time performance optimization
- 1Department of Mechanical Engineering, JNTUK
- 2Department of Mechanical Engineering, Annamacharya Institute of Technology and Sciences
Abstract
Proportional-Integral-Derivative (PID) and Model Predictive Control (MPC) are the conventional methods used for Industrial Process Control (IPC). But these methods often fail to maintain optimal performance when the operating environment is nonlinear and changes dynamically. Internet of Things (IoT) models help with connectivity and monitoring in industry. These models lack predictive decision-making for real-time adaptation, energy efficiency, and early fault detection. This study aims to develop an Artificial Intelligence-Enabled Control System (AI-ECS) that unifies real-time process management, predictive maintenance, and adaptive energy optimization for dynamic industrial environments. AI-ECS integrates IoT sensor networks with hybrid Random Forest (RF) and Long Short-Term Memory (LSTM) algorithms for predictive analytics. Further, real-time feedback control is achieved through dynamically tuned PID and MPC controllers running on an edge-computing platform. An experimental study was done on a continuous-flow industrial process testbed. The setup included multi-rate sensor sampling, Message Queuing Telemetry Transport (MQTT) communication, and containerized edge services. AI-ECS was compared with traditional PID, Advanced Process Control (APC), and IoT-only systems. AI-ECS reduced control error by 73.3% , measured as the relative reduction in Mean Absolute Percentage Error (MAPE) of the process variable relative to its setpoint. Energy Consumption (EC) decreased by 9.1% as well. Early Fault Detection (FD) also improved by a good margin. Anomalies were monitored over a 48 -hour pre-failure window, and the impending fault was detected up to 36 hours before failure. The system also presented faster disturbance rejection. Stability was verified across the 24 -hour and 30 -day operational windows of the testbed. The AI-ECS gives a reproducible, scalable, and energy-efficient model for intelligent industrial control. It provides a practical method for predictive, self-optimizing, and fault-tolerant manufacturing in accordance with the standards of Industry 4.0 by combining IoT-enabled sensing with adaptive decision-making driven by artificial intelligence (AI).
Related
- From Static Atlases to Intelligent Learning Systems: A Scoping Review of Imaging-Based Tools and Emerging Artificial Intelligence Applications for Facial Aesthetic Education and Training
- Artificial Intelligence in Photodynamic Therapy: From Molecular Design of Photosensitizers to Intelligent Delivery, Adaptive Dosimetry, and Personalized Prognostication of Results
- Artificial Intelligence‐Assisted Automated Optical Coherence Tomography Analysis for Percutaneous Coronary Intervention Planning: From Plaque Analysis to Precision Lesion Preparation and Stent Optimization
- CT-Referenced Diagnostic Performance and Complementary Error Patterns of Artificial Intelligence Versus a Single Emergency Physician in Chest Radiograph Interpretation: A Retrospective Paired Diagnostic Accuracy Study
- Evaluating the impact of social determinants of health on model performance and fairness in artificial intelligence for glaucoma progression prediction
BibTeX
@article{sadu2026intelligent,
title = {{An intelligent internet of things-based artificial intelligence framework for adaptive mechanical process control and real-time performance optimization}},
author = {Vijaya Bhaskar Sadu and Gopala Rao L. V. V. and Gopala Krishna A},
journal = {Scientific Reports},
year = {2026},
month = aug,
doi = {10.1038/s41598-026-67298-z},
url = {https://doi.org/10.1038/s41598-026-67298-z},
note = {CC BY},
}
RIS
TY - JOUR
TI - An intelligent internet of things-based artificial intelligence framework for adaptive mechanical process control and real-time performance optimization
AU - Vijaya Bhaskar Sadu
AU - Gopala Rao L. V. V.
AU - Gopala Krishna A
PY - 2026
DA - 2026/08/01
JO - Scientific Reports
DO - 10.1038/s41598-026-67298-z
UR - https://doi.org/10.1038/s41598-026-67298-z
AB - Proportional-Integral-Derivative (PID) and Model Predictive Control (MPC) are the conventional methods used for Industrial Process Control (IPC). But these methods often fail to maintain optimal performance when the operating environment is nonlinear and changes dynamically. Internet of Things (IoT) models help with connectivity and monitoring in industry. These models lack predictive decision-making for real-time adaptation, energy efficiency, and early fault detection. This study aims to develop an Artificial Intelligence-Enabled Control System (AI-ECS) that unifies real-time process management, predictive maintenance, and adaptive energy optimization for dynamic industrial environments. AI-ECS integrates IoT sensor networks with hybrid Random Forest (RF) and Long Short-Term Memory (LSTM) algorithms for predictive analytics. Further, real-time feedback control is achieved through dynamically tuned PID and MPC controllers running on an edge-computing platform. An experimental study was done on a continuous-flow industrial process testbed. The setup included multi-rate sensor sampling, Message Queuing Telemetry Transport (MQTT) communication, and containerized edge services. AI-ECS was compared with traditional PID, Advanced Process Control (APC), and IoT-only systems. AI-ECS reduced control error by 73.3% , measured as the relative reduction in Mean Absolute Percentage Error (MAPE) of the process variable relative to its setpoint. Energy Consumption (EC) decreased by 9.1% as well. Early Fault Detection (FD) also improved by a good margin. Anomalies were monitored over a 48 -hour pre-failure window, and the impending fault was detected up to 36 hours before failure. The system also presented faster disturbance rejection. Stability was verified across the 24 -hour and 30 -day operational windows of the testbed. The AI-ECS gives a reproducible, scalable, and energy-efficient model for intelligent industrial control. It provides a practical method for predictive, self-optimizing, and fault-tolerant manufacturing in accordance with the standards of Industry 4.0 by combining IoT-enabled sensing with adaptive decision-making driven by artificial intelligence (AI).
SN - 2045-2322
ER -