检索结果-内蒙古大学图书馆

Development of a Bias Compensating Q-Learning Controller for a Multi-Zone HVAC Facility

IEEE/CAA Journal of Automatica Sinica 2023年第8期10卷 1704-1715页

作者： Syed Ali Asad Rizvi Amanda J.Pertzborn Zongli Lin IEEE the Department of Electrical and Computer Engineering Tennessee Technological UniversityCookevilleTN 38505 USA the Mechanical Systems and Controls Group in the Building Energy and Environment Division at the National Institute of Standards and Technology(NIST) GaithersburgMD 20899 USA the Charles L.Brown Department of Electrical and Computer Engineering University of VirginiaCharlottesvilleVA 22904-4743 USA

We present the development of a bias compensating reinforcement learning(RL)algorithm that optimizes thermal comfort(by minimizing tracking error)and control utilization(by penalizing setpoint deviations)in a multi-zone heating,ventilation,and air-conditioning(HVAC)lab facility subject to unmeasurable disturbances and unknown *** is shown that the presence of unmeasurable disturbance results in an inconsistent learning equation in traditional RL controllers leading to parameter estimation bias(even with integral action support),and in the extreme case,the divergence of the learning *** demonstrate this issue by applying the popular Q-learning algorithm to linear quadratic regulation(LQR)of a multi-zone HVAC environment and showing that,even with integral support,the algorithm exhibits bias issue during the learning phase when the HVAC disturbance is unmeasurable due to unknown heat gains,occupancy variations,light sources,and outside weather *** address this difficulty,we present a bias compensating learning equation that learns a lumped bias term as a result of disturbances(and possibly other sources)in conjunction with the optimal control *** results show that the proposed scheme not only recovers the bias-free optimal control parameters but it does so without explicitly learning the dynamic model or estimating the disturbances,demonstrating the effectiveness of the algorithm in addressing the above challenges.

关键词： HVAC control optimal tracking Q-learning reinforcement learning(RL)

来源：评论

学校读者我要写书评

暂无评论

Model Predictive Control with Model Error Compensation by Koopman Approach 48

Model Predictive Control with Model Error Compensation by Ko...

引用

48th Annual Conference of the IEEE Industrial Electronics Society, IECON 2022

作者： Kanai, Masaki Yamakita, Masaki R&d Group Hitachi Ltd. Controls and Robotics Innovation Center Ibaraki Japan Tokyo Institute of Technology Department of Systems and Control Engineering Tokyo Japan

ISBN: (数字)9781665480253

ISBN: (纸本)9781665480253

This study proposes a method for high-performance model predictive control (MPC) by using a lifted bilinear model error model (MEM) based on the Koopman approach to compensate for errors in the prediction model. In MPC, more accurate prediction models can be expected to provide better control performance for the real plant. However, there often exist modeling errors in the prediction model, which may prevent MPC from achieving good performance. In this study, we identify the errors in the nominal model as an MEM, and formulate an MPC that integrates the nominal model and the MEM into an "error-compensated model"as a prediction model. We model the MEM with a lifted bilinear model that can represent nonlinear characteristics based on Koopman's theory. In addition, an efficient linear MPC algorithm is constructed by linearizing the MEM while maintaining high prediction accuracy. Numerical simulations demonstrate the effectiveness of the proposed method. © 2022 IEEE.

关键词： Model predictive control

来源：评论

学校读者我要写书评

暂无评论

Lifted Bilinear Model-based Linear Model Predictive Control with Scalability

引用

IFAC-PapersOnLine 2023年第2期56卷 9405-9410页

作者： Masaki Kanai Masaki Yamakita Controls and Robotics Innovation Center Research & Development Group Ltd. Hitachi Ibaraki Japan Department of Systems and Control Engineering Tokyo Institute of Technology Tokyo Japan

We propose a novel linear model predictive control (MPC) using a lifted bilinear model based on Koopman theory, which is computationally scalable against the dimension of the target system and the prediction horizon length. In MPC, the accuracy of the prediction model determines control performance, but it is a challenge to reduce the computational cost especially when considering nonlinearity of the model. To address this, a method has been proposed which represents the nonlinear input affine system as a lifted bilinear model and utilizes linear approximation and prediction error correction regarding the lifted state to achieve a low-computational-cost linear MPC with equivalent performance to nonlinear MPC. However, although the previous studies have shown its effectiveness for relatively low-order systems, it has not been applied to practical systems with higher dimensions. In this study, we extend the conventional method and propose a scalable linear MPC using a lifted bilinear model and apply it to higher-order nonlinear systems. In this paper, a quadrotor system operating in three-dimensional space is considered and its analytical lifted bilinear model is derived. In the formulation of linear MPC using the lifted bilinear model, an error correction method is newly introduced to feedback error for adjustment of the numerical relationships among the elements in the lifted state. The effectiveness of the proposed method is demonstrated through numerical simulations.

关键词： Model predictive control Koopman theory Lifting linearization Bilinear system

来源：评论

学校读者我要写书评

暂无评论

A Hybrid-based Clustering Approach for Fault Detection in HVAC systems

引用

IFAC-PapersOnLine 2024年第14期58卷 259-264页

作者： Hesam Hassanpour Amir H. Hamedi Prashant Mhaskar John M. House Timothy I. Salsbury Department of Chemical Engineering McMaster University Hamilton Ontario Canada These results were obtained during John House and Timothy Salsbury’s affiliation with the Building Management Systems Research Group Johnson Controls International plc

This paper presents a hybrid model-based fault detection strategy for heating, ventilation, and air conditioning (HVAC) systems, focusing on air handling units (AHUs). Addressing the substantial energy inefficiencies in commercial buildings due to undetected HVAC faults, this research combines first-principles knowledge with data-driven techniques to enhance fault detection accuracy. First-principles based residuals (differences between expected and observed behaviors) are integrated with data (temperature measurements in different locations of AHU) to perform principal component analysis (PCA) (pre-processing step). Pre-processed data (principal component scores) are then utilized to perform clustering analysis using K-means and DBSCAN approaches. The proposed approach is tested against two common faults in AHUs and its performance is evaluated compared to a purely data-driven method. The results indicate that the hybrid method, which synergizes residual knowledge from first-principles models with data, significantly outperforms the purely data-driven approach. This is demonstrated through performance analysis using metrics like the adjusted rand index (ARI) and normalized mutual information (NMI). The research underscores the potential of the hybrid method in improving fault diagnosis of HVAC systems, helping to conserve energy by ensuring efficient and reliable operation.

关键词： Fault Detection HVAC systems Principal Component Analysis (PCA) K-means DBSCAN

来源：评论

学校读者我要写书评

暂无评论

Approximate Model Predictive Control of Switched Affine systems Using Multitask Learning with Safety and Stability Guarantees

Approximate Model Predictive Control of Switched Affine Syst...

引用

IEEE Conference on Decision and Control

作者： Faiq Ghawash Morten Hovd Brad Schofield Department of Engineering Cybernetics Norwegian University of Science and Technology Trondheim Norway Beams Department Industrial Controls Systems Group at CERN Meyrin Switzerland

We study the problem of designing an approximate model predictive control (MPC) for discrete time switched affine systems. The MPC design for the switched affine system requires an online solution of a mixed integer program. However, the combinatorial nature of the mixed integer problems might require a large computational time limiting its applicability in real time scenarios. To this end, we propose a framework based on the multitask learning paradigm to approximate the solution of mixed integer MPC for switched affine systems. We also provide a computational method to overapproximate the reachable sets of the closed-loop system that helps to analyze the safety and stability of the system under the influence of the learned controller. Once trained offline, the resulting controller results in a solver free approach especially suited for implementation on resource constrained embedded hardware. We demonstrate the efficacy of the approach on a real world example of an induced draft cooling tower.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Additive Manufacturing of Ceramics opens the door to Components for Biomanufacturing Applications with high Functionality 6

Additive Manufacturing of Ceramics opens the door to Compone...

引用

6th CIRP Conference on BioManufacturing, BioM 2024

作者： Scheithauer, Uwe Holtzhausen, Stefan Wiemer, Hajo Fraunhofer Institute for Ceramic Technologies and Systems Group Additive and Hybrid Manufacturing Winterbergstr. 28 Dresden01277 Germany TUD Dresden University of Technology Institute of Machine Elements and Machine Design Virtual Product Development Group Advanced Geometric Modeling And Analysis Dresden01062 Germany TUD Dresden University of Technology Institute of Mechatronic Engineering Machine Tools Development and Adaptive Controls Department of Machine Data Processing Dresden01062 Germany

Significant advancements in production technology, materials, component design, and data science foster innovation across industries, including biomanufacturing. Ceramic materials, adapted additive manufacturing processes, and novel construction methods hold immense innovation potential due to ceramics' exceptional thermal, chemical, and mechanical properties. Additive manufacturing enables the creation of intricate ceramic components with unprecedented functionality. Hybridizing manufacturing technologies and materials facilitates direct integration of additional features like sensors and actuators. Despite their potential, ceramic materials remain underutilized. This article aims to spark a discourse on leveraging high-performance ceramic components in biomanufacturing, leveraging superior material properties, innovative design methods, additive manufacturing, process data analysis, and AI. © 2024 The Authors.

关键词： Ceramic materials

来源：评论

学校读者我要写书评

暂无评论

Development of a Hardware-in-the-loop Testbed for Laboratory Performance Verification of Flexible Building Equipment in Typical Commercial Buildings

arXiv

引用

arXiv 2023年

作者： Chen, Zhelun Lo, L. James Pertzborn, Amanda Grajewski, Gabriel Wen, Jin O'Neill, Zheng Calfa, Caleb Li, Yicheng Bushby, Steven T. Payne, W. Vance Fu, Yangyang Yang, Zhiyao The Department of Civil Architectural & Environmental Engineering Drexel University PhiladelphiaPA United States The Department of Mechanical Engineering Texas A&M University College StationTX United States The Mechanical Systems and Controls Group The HVAC&R Equipment Performance Group The Building Energy and Environment Division of the Engineering Laboratory The National Institute of Standards and Technology GaithersburgMD United States

The goals of reducing energy costs, shifting electricity peaks, increasing the use of renewable energy, and enhancing the stability of the electric grid can be met in part by fully exploiting the energy flexibility potential of buildings and building equipment. The development of strategies that exploit these flexibilities could be facilitated by publicly available high-resolution datasets illustrating how control of HVAC systems in commercial buildings can be used in different climate zones to shape the energy use profile of a building for grid needs. This article presents the development and integration of a Hardware-In-the-Loop Flexible load Testbed (HILFT) that integrates physical HVAC systems with a simulated building model and simulated occupants with the goal of generating datasets to verify load flexibility of typical commercial buildings. Compared to simulation-only experiments, the hardware-in-the-loop approach captures the dynamics of the physical systems while also allowing efficient testing of various boundary conditions. The HILFT integration in this article is achieved through the co-simulation among various software environments including LabVIEW, MATLAB, and EnergyPlus. Although theoretically viable, such integration has encountered many real-world challenges, such as: 1) how to design the overall data infrastructure to ensure effective, robust, and efficient integration;2) how to avoid closed-loop hunting between simulated and emulated variables;3) how to quantify system response times and minimize system delays;and 4) how to assess the overall integration quality. Lessons-learned using the examples of an AHU-VAV system, an air-source heat pump system, and a water-source heat pump system are presented. Copyright © 2023, The Authors. All rights reserved.

关键词： Synthetic apertures

来源：评论

学校读者我要写书评

暂无评论

Additive Manufacturing of Ceramics opens the door to Components for Biomanufacturing Applications with high Functionality

引用

Procedia CIRP 2024年 125卷 284-289页

作者： Uwe Scheithauer Stefan Holtzhausen Hajo Wiemer Fraunhofer Institute for Ceramic Technologies and Systems Group Additive and Hybrid Manufacturing Winterbergstr. 28 01277 Dresden Germany TUD Dresden University of Technology Institute of Machine Elements and Machine Design Chair of Virtual Product Development Group Advanced Geometric Modeling And Analysis 01062 Dresden Germany TUD Dresden University of Technology Institute of Mechatronic Engineering Chair of Machine Tools Development and Adaptive Controls Department of Machine Data Processing 01062 Dresden Germany

Significant advancements in production technology, materials, component design, and data science foster innovation across industries, including biomanufacturing. Ceramic materials, adapted additive manufacturing processes, and novel construction methods hold immense innovation potential due to ceramics’ exceptional thermal, chemical, and mechanical properties. Additive manufacturing enables the creation of intricate ceramic components with unprecedented functionality. Hybridizing manufacturing technologies and materials facilitates direct integration of additional features like sensors and actuators. Despite their potential, ceramic materials remain underutilized. This article aims to spark a discourse on leveraging high-performance ceramic components in biomanufacturing, leveraging superior material properties, innovative design methods, additive manufacturing, process data analysis, and AI.

关键词： ceramics: additive manufacturing component design data science biomanufacturing

来源：评论

学校读者我要写书评

暂无评论

Experimental Results of a Disturbance Compensating Q-learning Controller for HVAC systems

Experimental Results of a Disturbance Compensating Q-learnin...

引用

American Control Conference (ACC)

作者： Syed Ali Asad Rizvi Amanda J. Pertzborn Electrical and Computer Engineering Department Tennessee Technological University Cookeville USA Mechanical Systems and Controls Group Energy and Environment Division National Institute of Standards and Technology (NIST) Gaithersburg USA

ISBN: (数字)9781665451963

ISBN: (纸本)9781665494809

We present the experimental results of our recently proposed disturbance compensating Q-learning algorithm that learns the optimal setpoints for a building heating, ventilation, and air conditioning (HVAC) zone subject to external disturbances. Machine learning techniques such as reinforcement learning (RL) have gained considerable attention in HVAC controls research as these techniques do not require perfect knowledge of the HVAC dynamics to design optimal control. HVAC dynamics are subject to disturbances in the form of heat gain, weather variations, occupants, etc., which makes the learning task challenging as many of the mainstream RL (Q-learning) algorithms used in controls require the measurement or even manipulation of the disturbance during the learning phase. If not accounted for, the presence of unknown disturbances can result in estimation bias, causing suboptimality, and may also lead to instability in severe cases. To address this difficulty, we introduce a bias compensation mechanism in conjunction with integral control in Q-learning to learn the optimal control setpoints for HVAC equipment. The experimental results show the effectiveness of the proposed scheme.

关键词： Heating systems HVAC Q-learning Phase measurement Heuristic algorithms Buildings Optimal control

来源：评论

学校读者我要写书评

暂无评论

Assessing Human Worker Performance by Pattern Mining of Kinect Sensor Skeleton Data

SSRN

引用

SSRN 2022年

作者： Tran, Tuan-Anh Ruppert, Tamás Eigner, György Abonyi, János ELKH-PE Complex Systems Monitoring Research Group Department of Process Engineering University of Pannonia Veszprem Hungary Physiological Controls Research Center Biomatics and Applied Artificial Intelligence Institution John von Neumann Faculty of Informatics Obudai University Budapest Hungary Department of Mechanical Engineering School of Mechanical Engineering Vietnam Maritime University Haiphong Viet Nam Doctoral School of Applied Informatics and Applied Mathematics Obudai University Budapest Hungary

The human worker is an in-disposable factor of manufacturing processes. Traditional observation methods to assess their performance took time and relied on expert experience. Industry 4.0 technologies can innovate that process with data mining techniques. This paper proposes an approach to automatically assess worker performance with skeleton data from the commercial Microsoft Kinect sensor by applying pattern mining methods and supervised learning algorithms. A case study is performed on an assembly line producing electrical products to validate the recommended approach. By using supervised learning, the movements of workers in each workstation can be segmented, and the line performance can be assessed. The work movement motifs can be recognized with pattern mining. The mined results can be used to further improve the production processes in terms of work procedures, movement symmetry, body utilization, and other ergonomics factors for both short and long-term human resource development. The promising result motivates further utilization of other easy-to-adopt technology in the modern manufacturing environment of the forthcoming Industry 5.0, which facilitates the data-driven improvement in the human-related aspects, and shortens the ramp-up phase of establishing a new production line. © 2022, The Authors. All rights reserved.

关键词： Industry 4.0

来源：评论

学校读者我要写书评

暂无评论

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案：

请选择收藏分类：

通借通还

建议与咨询 留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案： 新增检索档案 确定 取消

请选择收藏分类： 新增自定义分类 确定 取消

通借通还

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

请选择保存的检索档案：

请选择收藏分类：