Reinforcement Learning Approach to Autonomous PID Tuning

Dogru, Oguzhan; Kirubakaran, V.; Ibrahim, Fadi; Wu, Yuqi; Sundaramoorthy, Arun Senthil; Huang, Biao; Xu, Shujie; Nixon, Mark; Bell, Noel

doi:10.23919/acc53348.2022.9867687

Cited by 4 publications

(3 citation statements)

References 9 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Reinforcement learning is a general learning framework that can be used to address general AI issues because intelligences interact with their environments in a manner similar to how humans do. Because of this, machine learning-based reinforcement learning is also referred to as a broad AI strategy [33].…”

Section: Models and Evaluation Methodsmentioning

confidence: 99%

Design and Application of Artificial Intelligence Technology-Driven Education and Teaching System in Universities

Zhang

2022

Computational and Mathematical Methods in Medicine

View full text Add to dashboard Cite

In recent years, many colleges and universities have been experimenting and exploring the evaluation of education and teaching system and have achieved certain results. In order to understand the quality of education and teaching system in colleges and universities, to improve the school conditions, and to promote the reform of teaching management, methods and means of evaluating the quality of education and teaching system in general higher education institutions are needed. Modern university education and teaching system should realize the combination of classroom teaching and practice teaching, and education and teaching system adopts the mode of the combination of on-campus practice and off-campus practice, so the design of teaching system is the key to the quality of teaching. Aiming at the current problem that talents cultivated by colleges and universities can hardly meet social demands in terms of engineering practice ability, innovation ability, and international competitiveness, this paper proposes the evaluation and adjustment of college education and teaching system driven by algorithms based on artificial intelligence (AI). By designing the teaching system of talent cultivation, and then establishing a quantitative and controllable quality assurance system for practical teaching, a new mechanism for the design of university education system is further explored. Specifically, the framework of the instructional system is built with the aid of an actor-critic algorithm in reinforcement learning, which assists in the design of the university education system, allowing students to truly understand, master and flex their knowledge, and strengthening the correct understanding of the students’ internal learning mechanisms. The practical teaching effect shows that the AI-driven instructional designs are more popular with contemporary students and have higher evaluation scores. The numerical experiment results also show the stability of the instructional design, overcoming the drawbacks of traditional manual subjectivity in the design. AI-driven college education and teaching system is conducive to cultivating students’ solid technical theoretical foundation. Therefore, through the AI-driven teaching system to strengthen the training of practical ability, so as to comprehensively improve students' comprehensive quality and innovation ability.

show abstract

Section: Models and Evaluation Methodsmentioning

confidence: 99%

Design and Application of Artificial Intelligence Technology-Driven Education and Teaching System in Universities

Zhang

2022

Computational and Mathematical Methods in Medicine

View full text Add to dashboard Cite

show abstract

“…These approaches learn a latent representation of each task, enabling the agent to simultaneously learn the context and the policy for a given task. This contrasts with other recent RL work on PID tuning, such as [17], where a process model is identified on a case-by-case basis and an RL agent is trained for a specific process.…”

Section: B Meta Reinforcement Learningmentioning

confidence: 91%

Meta-Reinforcement Learning for Adaptive Control of Second Order Systems

McClement

Lawrence

Forbes

et al. 2022

2022 IEEE International Symposium on Advanced Control of Industrial Processes (AdCONIP)

View full text Add to dashboard Cite

Meta-learning is a branch of machine learning which aims to synthesize data from a distribution of related tasks to efficiently solve new ones. In process control, many systems have similar and well-understood dynamics, which suggests it is feasible to create a generalizable controller through meta-learning. In this work, we formulate a meta reinforcement learning (meta-RL) control strategy that takes advantage of known, offline information for training, such as a model structure. The meta-RL agent is trained over a distribution of model parameters, rather than a single model, enabling the agent to automatically adapt to changes in the process dynamics while maintaining performance. A key design element is the ability to leverage model-based information offline during training, while maintaining a model-free policy structure for interacting with new environments. Our previous work has demonstrated how this approach can be applied to the industrially-relevant problem of tuning proportional-integral controllers to control first order processes. In this work, we briefly reintroduce our methodology and demonstrate how it can be extended to proportional-integral-derivative controllers and second order systems.

show abstract

“…• learning modifiers of the modifier adaptation scheme via GPs [6,11,148] • hybrid modeling for RTO models [194] • plant models for model-based control [52,67,95,105,112,150,194] • inverse models to provide control actions [65,67,88,145] • observer for parameter and state estimation [19,48,52,53,119,141] • self-tuning PID controllers [7,79,85,92,30,132,169] • solving static optimization of RTO using RL [140] • set-point optimization using RL [75,40] • imitation (supervise) learning [2] • RL-based [137,4] • RL for tuning PID controller [44] FIGURE 1.14 Selected, non-extensive, overview of works on machine-learning-supported control. Note that there is no clear-cut classification, as in some cases it is difficult to classify the methods into these categories.…”

Section: Controller Design Via Machine Learningmentioning

confidence: 99%

Machine Learning for Process Control of (Bio)Chemical Processes

Himmel¹,

Matschek²,

Kok³

et al. 2023

Preprint

View full text Add to dashboard Cite

The control of manufacturing processes must satisfy high quality and efficiency requirements while meeting safety requirements. A broad spectrum of monitoring and control strategies, such as model-and optimization-based controllers, are utilized to address these issues. Driven by rising demand for flexible yet energy and resource-efficient operations existing approaches are challenged due to high uncertainties and changes. Machine learning algorithms are becoming increasingly important in tackling these challenges, especially due to the growing amount of available data. The ability for automatic adaptation and learning from human operators offer new opportunities to increase efficiency yet provide flexible operation. Combining machine learning algorithms with safe or robust controls offers novel reliable operation methods. This chapter highlights ways to fuse machine learning and control for the safe and improved operation of chemical and biochemical processes. We outline and summarize both -learning models for control and learning the control components. We offer a general overview, including a literature review, to provide a guideline for utilizing machine learning techniques in control structures.

show abstract

Reinforcement Learning Approach to Autonomous PID Tuning

Cited by 4 publications

References 9 publications

Design and Application of Artificial Intelligence Technology-Driven Education and Teaching System in Universities

Design and Application of Artificial Intelligence Technology-Driven Education and Teaching System in Universities

Meta-Reinforcement Learning for Adaptive Control of Second Order Systems

Machine Learning for Process Control of (Bio)Chemical Processes

Contact Info

Product

Resources

About