Education, Science, Technology, Innovation and Life
Open Access
Sign In
  1. Home
  2. Journals
  3. Advances in Computer, Signals and Systems
  4. Vol 10, Issue 2, 2026

Deep Reinforcement Learning-Based Adaptive PID Temperature Control Method for Scrap Aluminum Induction Heating System

Download as PDF

DOI: 10.23977/acss.2026.100207 | Downloads: 0 | Views: 79

Author(s)

Zhuoyue Xu 1, Jingwei Wang 1

Affiliation(s)

1 TUT Maritime College, Tianjin University of Technology, Tianjin, China

Corresponding Author

Jingwei Wang

ABSTRACT

To address the issues of temperature response lag, difficulty in real-time adjustment of control parameters, and insufficient system stability under complex operating conditions during induction heating of scrap aluminum, this paper proposes a proportional-integral-derivative (PID) temperature control method based on Proximal Policy Optimization (PPO) deep reinforcement learning. First, a first-order inertial pure time-delay mathematical model of the induction heating system is established to describe the thermal inertia and time-delay dynamic characteristics during the heating process. Second, a near-end strategy optimization algorithm is introduced into the proportional-integral-derivative (PID) controller parameter adjustment process. A reinforcement learning state space is constructed using the system temperature error, error rate of change, control output, and actual temperature. The proportional, integral, and derivative parameter corrections are used as action outputs to achieve online adaptive updating of the control parameters. Subsequently, a co-simulation experimental environment is built on the MATLAB/Simulink platform, and comparative analyses are conducted with traditional proportional-integral-derivative (PID) control, fuzzy proportional-integral-derivative (Fuzzy PID) control, and Sparrow Search Algorithm-optimized proportional-integral-derivative (SSA-PID) control method. Experimental results show that the proposed method can effectively reduce system temperature overshoot, shorten system rise time and settling time, and quickly recover the system's stable state under external disturbances, exhibiting good dynamic tracking performance and disturbance rejection capability. This research provides a new solution for intelligent temperature control of complex industrial induction heating systems.

KEYWORDS

Proximal Policy Optimization; Deep Reinforcement Learning; PID Control; Induction Heating

CITE THIS PAPER

Zhuoyue Xu, Jingwei Wang. Deep Reinforcement Learning-Based Adaptive PID Temperature Control Method for Scrap Aluminum Induction Heating System. Advances in Computer, Signals and Systems (2026). Vol. 10, No. 2, 59-72. DOI: http://dx.doi.org/10.23977/acss.2026.100207.

REFERENCES

[1] Majumdar S. An insight view of evolution of advanced aluminum alloy for aerospace and automotive industry: current status and future prospects[J]. Journal of The Institution of Engineers (India): Series D, 2024: 1-18.
[2] Trzepieciński T, Najm S M. Current trends in metallic materials for body panels and structural members used in the automotive industry[J]. Materials, 2024, 17(3): 590.
[3] Samberger S. New strategies to improve Recycling and reduce CO2-emission of Aluminum production and processing[J]. Journal of light metal welding, 2024, 62(3): 119-132.
[4] Yang Y. Supply potential, carbon emission reduction, energy conservation, and sustainable pathways for aluminum recycling in China[J]. Sustainable Production and Consumption, 2024, 50: 239-252.
[5] Yakubov V. Recycled aluminium feedstock in metal additive manufacturing: A state of the art review[J]. Heliyon, 2024, 10(5).
[6] Esteve V, Bellido J L, Jordán J. State of the art and future trends in monitoring for industrial induction heating applications[J]. Electronics, 2024, 13(13): 2591.
[7] Chen X. Research status and applications of dual-frequency induction heating power supply[J]. Electronics, 2024, 13(24): 4913.
[8] Chanda S. Comparative analysis of proportional, proportional–integral, and proportional–integral–derivative controllers for thermal regulation in a cylindrical cooling system with multiple O-Ring heat sources[J]. Physics of Fluids, 2024, 36(8).
[9] Lakman N A, Mokhtar Z A. Mobile application for practical structural engineering learning: A USE questionnaire-based student evaluation on usefulness, satisfaction and ease of use[J]. International Journal of Business Studies and Education, 2024, 2(1).
[10] Emara R. Artificial Intelligence Based Controller for a Temperature Control System[J]. Faculty of Engineering, The British University in Egypt, 2024.
[11] Jabari M. An advanced PID tuning method for temperature control in electric furnaces using the artificial rabbits optimization algorithm: M. Jabari et al[J]. International Journal of Dynamics and Control, 2025, 13(5): 175.
[12] Abualigah L. Particle swarm optimization algorithm: review and applications[J]. Metaheuristic optimization algorithms, 2024: 1-14.
[13] Priyadarshi R, Kumar R R. Evolution of swarm intelligence: a systematic review of particle swarm and ant colony optimization approaches in modern research[J]. Archives of Computational Methods in Engineering, 2025, 32(6): 3609-3650.
[14] Lee S, Chau N K, Choi S. Deep Reinforcement Learning-Based Real-time Temperature Control in Thermoelectric Heat Exchange System[J]. Case Studies in Thermal Engineering, 2026: 107633.
[15] Gharbi A. Intelligent HVAC Control: Comparative Simulation of Reinforcement Learning and PID Strategies for Energy Efficiency and Comfort Optimization[J]. Mathematics, 2025, 13(14): 2311.

Subscription
E-Mail Alert
Most popular papers
Call for Special Issue
Publication Ethics & OA Statement
Plagiarism Policy
Contact Us
Downloads: 49157
Visits: 1120971

Sponsors, Associates, and Links

All published work is licensed under a Creative Commons Attribution 4.0 International License.

Copyright © 2016 - 2031 Clausius Scientific Press Inc. All Rights Reserved.