Current Search:   Zhou, Qun (x)

View All Items

  • CSV Spreadsheet
(1 - 10 of 10)
Data-Driven Modeling and Optimization of Building Energy Consumption
Title
Data-Driven Modeling and Optimization of Building Energy Consumption.
Creator
Grover, Divas, Pourmohammadi Fallah, Yaser, Vosoughi, Azadeh, Zhou, Qun, University of Central Florida
Abstract / Description
Sustainability and reducing energy consumption are targets for building operations. The installation of smart sensors and Building Automation Systems (BAS) makes it possible to study facility operations under different circumstances. These technologies generate large amounts of data. That data can be scrapped and used for the analysis. In this thesis, we focus on the process of data-driven modeling and decision making from scraping the data to simulate the building and optimizing the...
Show moreSustainability and reducing energy consumption are targets for building operations. The installation of smart sensors and Building Automation Systems (BAS) makes it possible to study facility operations under different circumstances. These technologies generate large amounts of data. That data can be scrapped and used for the analysis. In this thesis, we focus on the process of data-driven modeling and decision making from scraping the data to simulate the building and optimizing the operation. The City of Orlando has similar goals of sustainability and reduction of energy consumption so, they provided us access to their BAS for the data and study the operation of its facilities. The data scraped from the City's BAS serves can be used to develop statistical/machine learning methods for decision making. We selected a mid-size pilot building to apply these techniques. The process begins with the collection of data from BAS. An Application Programming Interface (API) is developed to login to the servers and scrape data for all data points and store it on the local machine. Then data is cleaned to analyze and model. The dataset contains various data points ranging from indoor and outdoor temperature to fan speed inside the Air Handling Unit (AHU) which are operated by Variable Frequency Drive (VFD). This whole dataset is a time series and is handled accordingly. The cleaned dataset is analyzed to find different patterns and investigate relations between different data points. The analysis helps us in choosing parameters for models that are developed in the next step. Different statistical models are developed to simulate building and equipment behavior. Finally, the models along with the data are used to optimize the building Operation with the equipment constraints to make decisions for building operation which leads to a reduction in energy consumption while maintaining temperature and pressure inside the building.
Show less
Date Issued
2019
Identifier
CFE0007810, ucf:52335
Format
Document (PDF)
PURL
http://purl.flvc.org/ucf/fd/CFE0007810
Physics-Guided Deep Learning for Power System Sate Estimation
Title
Physics-Guided Deep Learning for Power System Sate Estimation.
Creator
Wang, Lei, Zhou, Qun, Li, Qifeng, Qi, Junjian, Dimitrovski, Aleksandar, University of Central Florida
Abstract / Description
Conventionally, physics-based models are used for power system state estimation, including Weighted Least Square (WLS) or Weighted Absolute Value (WLAV). These models typically consider a single snapshot of the system without capturing temporal correlations of system states. In this thesis, a Physics-Guided Deep Learning (PGDL) method incorporating the physical power system model with the deep learning is proposed to improve the performance of power system state estimation. Specifically,...
Show moreConventionally, physics-based models are used for power system state estimation, including Weighted Least Square (WLS) or Weighted Absolute Value (WLAV). These models typically consider a single snapshot of the system without capturing temporal correlations of system states. In this thesis, a Physics-Guided Deep Learning (PGDL) method incorporating the physical power system model with the deep learning is proposed to improve the performance of power system state estimation. Specifically, inspired by Autoencoders, deep neural networks (DNNs) are utilized to learn the temporal correlations of power system states. The estimated system states are checked against the physics law by a set of power flow equations. Hence, the proposed PGDL approach is both data-driven and physics-based. The proposed method is compared with the traditional methods on the basis of accuracy and robustness in IEEE standard cases. The results indicate that PGDL framework provides more accurate and robust estimation for power system state estimation.
Show less
Date Issued
2019
Identifier
CFE0007871, ucf:52787
Format
Document (PDF)
PURL
http://purl.flvc.org/ucf/fd/CFE0007871
Online Neuro-Adaptive Learning For Power System Dynamic State Estimation
Title
Online Neuro-Adaptive Learning For Power System Dynamic State Estimation.
Creator
Birari, Rahul, Zhou, Qun, Sun, Wei, Dimitrovski, Aleksandar, University of Central Florida
Abstract / Description
With the increased penetration of renewable generation in the smart grid , it is crucial to have knowledge of rapid changes of system states. The information of real-time electro-mechanical dynamic states of generators are essential to ensuring reliability and detecting instability of the grid. The conventional SCADA based Dynamic State Estimation (DSE) was limited by the slow sampling rates (2-4 Hz). With the advent of PMU based synchro-phasor technology in tandem with Wide Area Monitoring...
Show moreWith the increased penetration of renewable generation in the smart grid , it is crucial to have knowledge of rapid changes of system states. The information of real-time electro-mechanical dynamic states of generators are essential to ensuring reliability and detecting instability of the grid. The conventional SCADA based Dynamic State Estimation (DSE) was limited by the slow sampling rates (2-4 Hz). With the advent of PMU based synchro-phasor technology in tandem with Wide Area Monitoring System (WAMS), it has become possible to avail rapid real-time measurements at the network nodes. These measurements can be exploited for better estimates of system dynamic states. In this research, we have proposed a novel Artificial Intelligence (AI) based real-time neuro-adaptive algorithm for rotor angle and speed estimation of synchronous generators. Generator swing equations and power flow models are incorporated in the online learning. The algorithm learns and adapts in real-time to achieve accurate estimates. Simulation is carried out on 68 bus 16 generator NETS-NYPS model. The neuro-adaptive algorithm is compared with classical Kalman Filter based DSE. Applicability and accuracy of the proposed method is demonstrated under the influence of noise and faulty conditions.
Show less
Date Issued
2017
Identifier
CFE0006858, ucf:51747
Format
Document (PDF)
PURL
http://purl.flvc.org/ucf/fd/CFE0006858
A Multiagent Q-learning-based Restoration Algorithm for Resilient Distribution System Operation
Title
A Multiagent Q-learning-based Restoration Algorithm for Resilient Distribution System Operation.
Creator
Hong, Jungseok, Sun, Wei, Zhou, Qun, Zheng, Qipeng, University of Central Florida
Abstract / Description
Natural disasters, human errors, and technical issues have caused disastrous blackouts to power systems and resulted in enormous economic losses. Moreover, distributed energy resources have been integrated into distribution systems, which bring extra uncertainty and challenges to system restoration. Therefore, the restoration of power distribution systems requires more efficient and effective methods to provide resilient operation.In the literature, using Q-learning and multiagent system (MAS...
Show moreNatural disasters, human errors, and technical issues have caused disastrous blackouts to power systems and resulted in enormous economic losses. Moreover, distributed energy resources have been integrated into distribution systems, which bring extra uncertainty and challenges to system restoration. Therefore, the restoration of power distribution systems requires more efficient and effective methods to provide resilient operation.In the literature, using Q-learning and multiagent system (MAS) to restore power systems has the limitation in real system application, without considering power system operation constraints. In order to adapt to system condition changes quickly, a restoration algorithm using Q-learning and MAS, together with the combination method and battery algorithm is proposed in this study. The developed algorithm considers voltage and current constraints while finding system switching configuration to maximize the load pick-up after faults happen to the given system. The algorithm consists of three parts. First, it finds switching configurations using Q-learning. Second, the combination algorithm works as a back-up plan in case of the solution from Q-learning violates system constraints. Third, the battery algorithm is applied to determine the charging or discharging schedule of battery systems. The obtained switching configuration provides restoration solutions without violating system constraints. Furthermore, the algorithm can adjust switching configurations after the restoration. For example, when renewable output changes, the algorithm provides an adjusted solution to avoid violating system constraints.The proposed algorithm has been tested in the modified IEEE 9-bus system using the real-time digital simulator. Simulation results demonstrate that the algorithm offers an efficient and effective restoration strategy for resilient distribution system operation.
Show less
Date Issued
2017
Identifier
CFE0006746, ucf:51856
Format
Document (PDF)
PURL
http://purl.flvc.org/ucf/fd/CFE0006746
Leaning Robust Sequence Features via Dynamic Temporal Pattern Discovery
Title
Leaning Robust Sequence Features via Dynamic Temporal Pattern Discovery.
Creator
Hu, Hao, Wang, Liqiang, Zhang, Shaojie, Liu, Fei, Qi, GuoJun, Zhou, Qun, University of Central Florida
Abstract / Description
As a major type of data, time series possess invaluable latent knowledge for describing the real world and human society. In order to improve the ability of intelligent systems for understanding the world and people, it is critical to design sophisticated machine learning algorithms for extracting robust time series features from such latent knowledge. Motivated by the successful applications of deep learning in computer vision, more and more machine learning researchers put their attentions...
Show moreAs a major type of data, time series possess invaluable latent knowledge for describing the real world and human society. In order to improve the ability of intelligent systems for understanding the world and people, it is critical to design sophisticated machine learning algorithms for extracting robust time series features from such latent knowledge. Motivated by the successful applications of deep learning in computer vision, more and more machine learning researchers put their attentions on the topic of applying deep learning techniques to time series data. However, directly employing current deep models in most time series domains could be problematic. A major reason is that temporal pattern types that current deep models are aiming at are very limited, which cannot meet the requirement of modeling different underlying patterns of data coming from various sources. In this study we address this problem by designing different network structures explicitly based on specific domain knowledge such that we can extract features via most salient temporal patterns. More specifically, we mainly focus on two types of temporal patterns: order patterns and frequency patterns. For order patterns, which are usually related to brain and human activities, we design a hashing-based neural network layer to globally encode the ordinal pattern information into the resultant features. It is further generalized into a specially designed Recurrent Neural Networks (RNN) cell which can learn order patterns in an online fashion. On the other hand, we believe audio-related data such as music and speech can benefit from modeling frequency patterns. Thus, we do so by developing two types of RNN cells. The first type tries to directly learn the long-term dependencies on frequency domain rather than time domain. The second one aims to dynamically filter out the ``noise" frequencies based on temporal contexts. By proposing various deep models based on different domain knowledge and evaluating them on extensive time series tasks, we hope this work can provide inspirations for others and increase the community's interests on the problem of applying deep learning techniques to more time series tasks.
Show less
Date Issued
2019
Identifier
CFE0007470, ucf:52679
Format
Document (PDF)
PURL
http://purl.flvc.org/ucf/fd/CFE0007470
Development of an Adaptive Restoration Tool For a Self-Healing Smart Grid
Title
Development of an Adaptive Restoration Tool For a Self-Healing Smart Grid.
Creator
Golshani, Amir, Sun, Wei, Qu, Zhihua, Vosoughi, Azadeh, Zhou, Qun, Zheng, Qipeng, University of Central Florida
Abstract / Description
Large power outages become more commonplace due to the increase in both frequency and strength of natural disasters and cyber-attacks. The outages and blackouts cost American industries and business billions of dollars and jeopardize the lives of hospital patients. The losses can be greatlyreduced with a fast, reliable and flexible restoration tool. Fast recovery and successfully adapting to extreme events are critical to build a resilient, and ultimately self-healing power grid. This...
Show moreLarge power outages become more commonplace due to the increase in both frequency and strength of natural disasters and cyber-attacks. The outages and blackouts cost American industries and business billions of dollars and jeopardize the lives of hospital patients. The losses can be greatlyreduced with a fast, reliable and flexible restoration tool. Fast recovery and successfully adapting to extreme events are critical to build a resilient, and ultimately self-healing power grid. This dissertation is aimed to tackle the challenging task of developing an adaptive restoration decisionsupport system (RDSS). The RDSS determines restoration actions both in planning and real-time phases and adapts to constantly changing system conditions. First, an efficient network partitioning approach is developed to provide initial conditions for RDSS by dividing large outage network into smaller islands. Then, the comprehensive formulation of RDSS integrates different recovery phases into one optimization problem, and encompasses practical constraints including AC powerflow, dynamic reserve, and dynamic behaviors of generators and load. Also, a frequency constrained load recovery module is proposed and integrated into the RDSS to determine the optimal location and amount of load pickup. Next, the proposed RDSS is applied to harness renewable energy sources and pumped-storage hydro (PSH) units by addressing the inherent variabilities and uncertainties of renewable and coordinating wind and PSH generators. A two-stage stochastic and robust optimization problem is formulated, and solved by the integer L-shaped and column-and-constraintsgeneration decomposition algorithms. The developed RDSS tool has been tested onthe modified IEEE 39-bus and IEEE 57-bus systems under different scenarios. Numerical results demonstrate the effectiveness and efficiency of the proposed RDSS. In case of contingencies or unexpected outages during the restoration process, RDSS can quickly update the restoration plan and adapt to changing system conditions. RDSS is an important step toward a self-healing power grid and its implementation will reduce the recovery time while maintaining system security.
Show less
Date Issued
2017
Identifier
CFE0007284, ucf:52169
Format
Document (PDF)
PURL
http://purl.flvc.org/ucf/fd/CFE0007284
DESIGN OF HIGH EFFICIENCY BRUSHLESS PERMANENT MAGNET MACHINES AND DRIVER SYSTEM
Title
DESIGN OF HIGH EFFICIENCY BRUSHLESS PERMANENT MAGNET MACHINES AND DRIVER SYSTEM.
Creator
He, Chengyuan, Wei, Lei, Sundaram, Kalpathy, Zhou, Qun, Jin, Yier, Zou, Shengli, University of Central Florida
Abstract / Description
The dissertation is concerned with the design of high-efficiency permanent magnet synchronous machinery and the control system. The dissertation first talks about the basic concept of the permanent magnet synchronous motor (PMSM) design and the mathematics design model of the advanced design method. The advantage of the design method is that it can increase the high load capacity at no cost of increasing the total machine size. After that, the control method of the PMSM and Permanent magnet...
Show moreThe dissertation is concerned with the design of high-efficiency permanent magnet synchronous machinery and the control system. The dissertation first talks about the basic concept of the permanent magnet synchronous motor (PMSM) design and the mathematics design model of the advanced design method. The advantage of the design method is that it can increase the high load capacity at no cost of increasing the total machine size. After that, the control method of the PMSM and Permanent magnet synchronous generator (PMSG) is introduced. The design, simulation, and test of a permanent magnet brushless DC (BLDC) motor for electric impact wrench and new mechanical structure are first presented based on the design method. Finite element analysis based on the Maxwell 2D is built to optimize the design and the control board is designed using Altium Designer. Both the motor and control board have been fabricated and tested to verify the design. The electrical and mechanical design are combined, and it provides an analytical IPMBLDC design method and an innovative and reasonable mechanical dynamical calculation method for the impact wrench system, which can be used in whole system design of other functional electric tools. A 2kw high-efficiency alternator system and its control board system are also designed, analyzed and fabricated applying to the truck auxiliary power unit (APU). The alternator system has two stages. The first stage is that the alternator three-phase outputs are connected to the three-phase active rectifier to get 48V DC. An advanced Sliding Mode Observer (SMO) is used to get an alternator position. The buck is used for the second stage to get 14V DC output. The whole system efficiency is much higher than the traditional system using induction motor.
Show less
Date Issued
2018
Identifier
CFE0007334, ucf:52135
Format
Document (PDF)
PURL
http://purl.flvc.org/ucf/fd/CFE0007334
Managing IO Resource for Co-running Data Intensive Applications in Virtual Clusters
Title
Managing IO Resource for Co-running Data Intensive Applications in Virtual Clusters.
Creator
Huang, Dan, Wang, Jun, Zhou, Qun, Sun, Wei, Zhang, Shaojie, Wang, Liqiang, University of Central Florida
Abstract / Description
Today Big Data computer platforms employ resource management systems such as Yarn, Torque, Mesos, and Google Borg to enable sharing the physical computing among many users or applications. Given virtualization and resource management systems, users are able to launch their applications on the same node with low mutual interference and management overhead on CPU and memory. However, there are still challenges to be addressed before these systems can be fully adopted to manage the IO resources...
Show moreToday Big Data computer platforms employ resource management systems such as Yarn, Torque, Mesos, and Google Borg to enable sharing the physical computing among many users or applications. Given virtualization and resource management systems, users are able to launch their applications on the same node with low mutual interference and management overhead on CPU and memory. However, there are still challenges to be addressed before these systems can be fully adopted to manage the IO resources in Big Data File Systems (BDFS) and shared network facilities. In this study, we mainly study on three IO management problems systematically, in terms of the proportional sharing of block IO in container-based virtualization, the network IO contention in MPI-based HPC applications and the data migration overhead in HPC workflows. To improve the proportional sharing, we develop a prototype system called BDFS-Container, by containerizing BDFS at Linux block IO level. Central to BDFS-Container, we propose and design a proactive IOPS throttling based mechanism named IOPS Regulator, which improves proportional IO sharing under the BDFS IO pattern by 74.4% on an average. In the aspect of network IO resource management, we exploit using virtual switches to facilitate network traffic manipulation and reduce mutual interference on the network for in-situ applications. In order to dynamically allocate the network bandwidth when it is needed, we adopt SARIMA-based techniques to analyze and predict MPI traffic issued from simulations. Third, to solve the data migration problem in small-medium sized HPC clusters, we propose to construct a sided IO path, named as SideIO, to explicitly direct analysis data to BDFS that co-locates computation with data. By experimenting with two real-world scientific workflows, SideIO completely avoids the most expensive data movement overhead and achieves up to 3x speedups compared with current solutions.
Show less
Date Issued
2018
Identifier
CFE0007195, ucf:52268
Format
Document (PDF)
PURL
http://purl.flvc.org/ucf/fd/CFE0007195
Analysis, Design and Efficiency Optimization of Power Converters for Renewable Energy Applications
Title
Analysis, Design and Efficiency Optimization of Power Converters for Renewable Energy Applications.
Creator
Chen, Xi, Batarseh, Issa, Zhou, Qun, Mikhael, Wasfy, Sun, Wei, Kutkut, Nasser, University of Central Florida
Abstract / Description
DC-DC power converters are widely used in renewable energy-based power generation systems due to the constant demand of high-power density and high-power conversion efficiency. DC-DC converters can be classified into non-isolated and isolated topologies. For non-isolated topologies, they are typically derived from buck, boost, buck-boost or forth order (such as Cuk, Sepic and Zeta) converters and they usually have relatively higher conversion efficiency than isolated topologies. However, with...
Show moreDC-DC power converters are widely used in renewable energy-based power generation systems due to the constant demand of high-power density and high-power conversion efficiency. DC-DC converters can be classified into non-isolated and isolated topologies. For non-isolated topologies, they are typically derived from buck, boost, buck-boost or forth order (such as Cuk, Sepic and Zeta) converters and they usually have relatively higher conversion efficiency than isolated topologies. However, with the applications where the isolation is required, either these topologies should be modified, or alternative topologies are needed. Among various isolated DC-DC converters, the LLC resonant converter is an attractive selection due to its soft switching, isolation, wide gain range, high reliability, high power density and high conversion efficiency.In low power applications, such as battery chargers and solar microinverters, increasing the switching frequency can reduce the size of passive components and reduce the current ripple and root-mean-square (RMS) current, resulting in higher power density and lower conduction loss. However, switching losses, gate driving loss and electromagnetic interference (EMI) may increase as a consequence of higher switching frequency. Therefore, switching frequency modulation, components optimization and soft switching techniques have been proposed to overcome these issues and achieve a tradeoff to reach the maximum conversion efficiency.This dissertation can be divided into two categories: the first part is focusing on the well-known non-isolated bidirectional cascaded-buck-boost converter, and the second part is concentrating on the isolated dual-input single resonant tank LLC converter. Several optimization approaches have been presented to improve the efficiency, power density and reliability of the power converters. In the first part, an adaptive switching frequency modulation technique has been proposed based on the precise loss model in this dissertation to increase the efficiency of the cascaded-buck-boost converter. In adaptive switching frequency modulation technique, the optimal switching frequency for the cascaded-buck-boost converter is adaptively selected to achieve the minimum total power loss. In addition, due to the major power losses coming from the inductor, a new low profile nanocrystalline inductor filled with copper foil has been designed to significantly reduce the core loss and winding loss. To further improve the efficiency of the cascaded-buck-boost converter, the adaptive switching frequency modulation technique has been applied on the converter with designed nanocrystalline inductor, in which the peak efficiency of the converter can break the 99% bottleneck.In the second part, a novel dual-input DC-DC converter is developed according to the LLC resonant topology. This design concept minimizes the circuit components by allowing single resonant tank to interface with multiple input sources. Based on different applications, the circuit configuration for the dual-input LLC converter will be a little different. In order to improve the efficiency of the dual-input LLC converter, the semi-active rectifiers have been used on the transformer secondary side to replace the low-side bridge diodes. In this case, higher magnetizing inductance can be selected while maintaining the same voltage gain. Besides, a burst-mode control strategy has been proposed to improve the light load and very light load efficiency of the dual- input LLC converter. This control strategy is able to be readily implemented on any power converter since it can be achieved directly through firmware and no circuit modification is needed in implementation of this strategy.
Show less
Date Issued
2019
Identifier
CFE0007612, ucf:52531
Format
Document (PDF)
PURL
http://purl.flvc.org/ucf/fd/CFE0007612
Improvement of Data-Intensive Applications Running on Cloud Computing Clusters
Title
Improvement of Data-Intensive Applications Running on Cloud Computing Clusters.
Creator
Ibrahim, Ibrahim, Bassiouni, Mostafa, Lin, Mingjie, Zhou, Qun, Ewetz, Rickard, Garibay, Ivan, University of Central Florida
Abstract / Description
MapReduce, designed by Google, is widely used as the most popular distributed programmingmodel in cloud environments. Hadoop, an open-source implementation of MapReduce, is a data management framework on large cluster of commodity machines to handle data-intensive applications. Many famous enterprises including Facebook, Twitter, and Adobehave been using Hadoop for their data-intensive processing needs. Task stragglers in MapReduce jobs dramatically impede job execution on massive datasets in...
Show moreMapReduce, designed by Google, is widely used as the most popular distributed programmingmodel in cloud environments. Hadoop, an open-source implementation of MapReduce, is a data management framework on large cluster of commodity machines to handle data-intensive applications. Many famous enterprises including Facebook, Twitter, and Adobehave been using Hadoop for their data-intensive processing needs. Task stragglers in MapReduce jobs dramatically impede job execution on massive datasets in cloud computing systems. This impedance is due to the uneven distribution of input data and computation load among cluster nodes, heterogeneous data nodes, data skew in reduce phase, resource contention situations, and network configurations. All these reasons may cause delay failure and the violation of job completion time. One of the key issues that can significantly affect the performance of cloud computing is the computation load balancing among cluster nodes. Replica placement in Hadoop distributed file system plays a significant role in data availability and the balanced utilization of clusters. In the current replica placement policy (RPP) of Hadoop distributed file system (HDFS), the replicas of data blocks cannot be evenly distributed across cluster's nodes. The current HDFS must rely on a load balancing utility for balancing the distribution of replicas, which results in extra overhead for time and resources. This dissertation addresses data load balancing problem and presents an innovative replica placement policy for HDFS. It can perfectly balance the data load among cluster's nodes. The heterogeneity of cluster nodes exacerbates the issue of computational load balancing; therefore, another replica placement algorithm has been proposed in this dissertation for heterogeneous cluster environments. The timing of identifying the straggler map task is very important for straggler mitigation in data-intensive cloud computing. To mitigate the straggler map task, Present progress and Feedback based Speculative Execution (PFSE) algorithm has been proposed in this dissertation. PFSE is a new straggler identification scheme to identify the straggler map tasks based on the feedback information received from completed tasks beside the progress of the current running task. Straggler reduce task aggravates the violation of MapReduce job completion time. Straggler reduce task is typically the result of bad data partitioning during the reduce phase. The Hash partitioner employed by Hadoop may cause intermediate data skew, which results in straggler reduce task. In this dissertation a new partitioning scheme, named Balanced Data Clusters Partitioner (BDCP), is proposed to mitigate straggler reduce tasks. BDCP is based on sampling of input data and feedback information about the current processing task. BDCP can assist in straggler mitigation during the reduce phase and minimize the job completion time in MapReduce jobs. The results of extensive experiments corroborate that the algorithms and policies proposed in this dissertation can improve the performance of data-intensive applications running on cloud platforms.
Show less
Date Issued
2019
Identifier
CFE0007818, ucf:52804
Format
Document (PDF)
PURL
http://purl.flvc.org/ucf/fd/CFE0007818