2020 3rd International Conference on Energy, Power and Environment: Towards Clean Energy Technologies
The energy storage power station (ESPS) is one of the most used energy storage system (ESS). It c... more The energy storage power station (ESPS) is one of the most used energy storage system (ESS). It can also be used to regulate the reactive power thus maintaining the micro-grid side voltage so no more static var generator needed. However, it requires a higher level controller to achieve that called coordinate control. The coordinate controller decides how much reactive power the ESPS should generate based on the voltage drop at the micro-grid side. And the reference reactive power signal is sent to the controller through a certain communication protocol, IEC61850-Generic Object-Oriented Substation Event (GOOSE) for example. The communication delay in such a protocol is ineligible. Such communication delay can seriously affect the performance of the reactive power control. To overcome this, Smith Predictor (SM) based delay compensation control is proposed and analyzed. Control hardware-in-the-loop results are presented to verify the effectiveness of the proposed method.
2016 IEEE PES Asia-Pacific Power and Energy Engineering Conference (APPEEC), 2016
Phasor Measurement Unit (PMU) has been applied on power system to improve the performance of powe... more Phasor Measurement Unit (PMU) has been applied on power system to improve the performance of power system state estimator for decades. It is anticipated that PMUs will observe distribution networks in future considering the contribution of PMUs to transmission networks. The advent of micro phasor measurement unit (μPMU) makes it possible to employ phasor measurement devices in distribution networks. Optimal PMU Placement (OPP) problem is to determine a minimal set of PMUs such that the whole system is observable. Since PMUs are mostly employed within transmission networks, most OPP algorithms are also designed for transmission networks. However, OPP algorithms designed transmission networks have not considered the radial topology of distribution networks. This paper investigated the different topology of transmission and distribution networks and proposed a customized OPP method for radial networks. The customized OPP method made use of radial topology of distribution networks to equivalently downsize the scale of the networks. Since the solution for OPP is usually difficult for large scale systems, the solution of OPP is accelerated by the customized OPP. The proposed method was tested on various IEEE distribution test feeders and transmission systems.
IECON 2011 - 37th Annual Conference of the IEEE Industrial Electronics Society, 2011
Inefficiency of the common lightings like incandescent, fluorescent and halogen lamps presents an... more Inefficiency of the common lightings like incandescent, fluorescent and halogen lamps presents an ample opportunity to retrofit high efficacy, long life-time and low maintenance LED solid-state lighting, which has significantly reduced in cost on an annual basis. LED lights and many electronic loads used within in-built environment are DC in nature. As such, in this paper, an energy-efficient low voltage
2017 IEEE Applied Power Electronics Conference and Exposition (APEC), 2017
The state of energy (SOE) of Li-ion batteries is a key indicator for the energy optimization and ... more The state of energy (SOE) of Li-ion batteries is a key indicator for the energy optimization and management of energy storage devices (ESDs) in electric vehicles (EVs) and smart grids. To improve the SOE estimation accuracy, a hybrid Li-ion battery model is presented in this study against the dynamic loads and the rate energy effects of the battery. Firstly, in order to take the advantages of both models, a 2nd order RC battery model is merged with an analytical kinetic battery model for accurately predicting the battery voltage characteristics and capturing the nonlinear rate energy effects to realize the high-fidelity SOE and runtime prediction. Secondly, a new method to separate the fast and slow dynamics of the 2nd order RC model is developed and presented with high-performance accuracy. Thirdly, commercial Li-ion batteries are tested at dynamic loads under various temperature to validate the effectiveness of the proposed model. The experimental results show high accuracy and reliability of the proposed battery model on the estimation of the battery SOE and the battery terminal voltage responses under dynamic loads.
2018 International Power Electronics Conference (IPEC-Niigata 2018 -ECCE Asia), 2018
Reliable online co-estimation of state of health (SOH) and state of charge (SOC) of Li-ion batter... more Reliable online co-estimation of state of health (SOH) and state of charge (SOC) of Li-ion batteries were of paramount importance for the realistic battery management system (BMS). This work aimed at bridging laboratory test and real-life battery operation data with a comprehensive analysis to provide a coherent and non-invasive approach based on probability density function (PDF), for building a pragmatic model to co-evaluate SOH and SOC of Li-ion batteries for smart grid applications. PDF results based on practical applications revealed that there was a prominent regularity of the voltage probabilities with regards to the SOH, which were exploited for setting up an online SOH evaluation scale gauge. Utilization of the load current of realistic smart grid further improve the practical generality of the proposed algorithm. The battery online SOC was determined afterwards based on the online OCV variation and the extracted SOH values. Commercial Li-ion batteries at arbitrary SOH and SOC level were tested to validate the effectiveness and robustness of the proposed algorithm, and the test results showed high accuracy and reliability of the proposed algorithm for co-evaluating SOH and SOC.
IAS '96. Conference Record of the 1996 IEEE Industry Applications Conference Thirty-First IAS Annual Meeting
This paper presents a physics-based electric arc model which can realistically simulate both stat... more This paper presents a physics-based electric arc model which can realistically simulate both static and dynamic characteristics of an arc load controlled by a power electronic circuit. The proposed model was developed from modifications of the classical Cassie and Mayr equations. The model equations have been expressed in a form suitable for incorporation into circuit simulators employing the nodal analysis method of equation solving. The model has been test implemented in the Saber circuit simulator. There has been a considerable degree of correspondence between the simulated and experimental results
Thermal comfort is a key consideration in smart buildings and a number of comfort models are avai... more Thermal comfort is a key consideration in smart buildings and a number of comfort models are available nowadays to evaluate the comfort level of occupants. However, the models are often complex and hardly interpretable for the developers and operators. Indeed, the model interpretations are beneficial in multifold such as for system inspection and optimization. In this paper, we propose an interpretable thermal comfort system to introduce interpretability to any black-box comfort models. First, we focus on the relationship between a model's input features and output comfort level. The feature impact on comfort is investigated and the impact patterns are shown to be diverse for different features. Second, we unveil the model mechanisms about the data processing inside the model by building the model surrogates based on the interpretable machine learning algorithms. The surrogates offer outstanding fidelity for simulating the actual model mechanisms and the interpretations based on the surrogates are intuitive and informative. Our interpretable comfort system can be integrated with the existing building management systems. Accordingly, we can ease building owner's concerns about adopting new black-box technologies and enable various smart building applications like smart energy management. Index Terms-Thermal comfort; smart building; smart city; deep learning; interpretable machine learning. I. INTRODUCTION T HERMAL comfort is an essential consideration of smart buildings. The impact of thermal comfort can be either positive or negative. Satisfied comfort ameliorates the occupant's well-being, happiness, and working efficiency [1], [2], while discomfort results in the sick building syndrome, absenteeism, and degraded cognitive performance [3]. Especially, modern people spend as much as 90% of their time in buildings [4]. As such, the Internet of things (IoT) system
2020 3rd International Conference on Energy, Power and Environment: Towards Clean Energy Technologies, 2021
This paper proposes the method to analyze the performance of wireless charging DC/DC resonant con... more This paper proposes the method to analyze the performance of wireless charging DC/DC resonant converters in various load conditions including the magnetic coupling effect. The equations for design performance analytical calculations are presented for series-series (SS) and series-parallel (SP) wireless charging DC/DC resonant converters. The equations can be applied in switching frequency sensitivity analysis at different magnetic strength and load resistance conditions to compare the performance of the resonant circuit topologies. The analytical calculation results can be used to size the voltage, current, and power rating of wireless charging resonant circuit components. The accuracy of the analytical calculation is validated by duty cycle control of wireless charging resonant converters in PLECS circuit simulation environment. Experimental results are obtained to support the conclusion drawn from the analytical solutions. The proposed method can be applied to study other wireless charging resonant converter topologies.
The state of energy (SOE) is a key indicator for the energy optimization and management of Li-ion... more The state of energy (SOE) is a key indicator for the energy optimization and management of Li-ion battery-based energy storage systems in the smart grid applications. To improve the SOE estimation accuracy, a Li-ion battery model is presented in this study against dynamic loads and battery ageing effects. Firstly, an electrical battery model is combined with an analytical model in order to take advantages of both models for accurate prediction of battery terminal voltage characteristics, SOE and remnant runtime. Secondly, a novel method to separate the fast and slow dynamics of the electrical battery model is de veloped, and its superior performance is presented. Thirdly, the effects of the battery initial SOC, load current rate and direction, operating temperature and ageing level are systematically scrutinized and involved into the proposed model for robust SOE and terminal voltage prediction. Commercial Li-ion batteries are then tested under dynamic loads and at an arbitrary battery ageing le vel to validate the effectiveness and robustness of the proposed model. The laboratory-scale experimental test results show superb accuracy and reliability of the proposed battery model for estimating battery SOE and terminal voltage under dynamic loads and battery ageing conditions.
Three co-estimation methods are compared for lithium-ion battery application. • A noise compensat... more Three co-estimation methods are compared for lithium-ion battery application. • A noise compensating method is exploited for co-estimation purpose. • Noise effects on different methods are analyzed with simulations and experiments. • The computing cost and tuning effort are discussed.
Reliable online estimation of state of charge (SOC) and capacity is critically important for the ... more Reliable online estimation of state of charge (SOC) and capacity is critically important for the battery management system. This paper presents a multi-timescale estimator to dually estimate the SOC and capacity for lithium-ion battery. The first-order RC model is used to simulate the dynamics of lithium-ion battery. Based on the battery model, the open circuit voltage (OCV) is timely updated with a simple OCV, the result of which is further corrected with the Kalman filter (KF). Then the SOC is inferred from the SOC-OCV look-up table. Meanwhile, a RLS-based capacity estimator is formulated to work simultaneously with the SOC estimation in the form of dual estimation. Different timescales are adopted for the dual estimator to improve accuracy and stability. Experimental results suggest that the proposed method estimates SOC and capacity in real time with fast convergence and high precision.
The high switching speed in a silicon carbide (SiC) metal-oxide-semiconductor field-effect transi... more The high switching speed in a silicon carbide (SiC) metal-oxide-semiconductor field-effect transistor (MOSFET) will aggravate the parasitic effects (di/dt and dv/dt) arising from the interaction with parasitic elements. In this project, a high-speed gate driver has been developed and optimised for the commercially available SiC MOSFET power module. The impact of various parasitic parameters on parasitic effects is initially evaluated. Then, an improved gate-assisted circuit is proposed with a local low-impedance path for both discharging and Cdv/dt currents. It allows maximised turn-off speed (dv/dt up to 36 V/ns) and minimised turn-off loss (reduction up to 70%). It also produces a reduction in electromagnetic interference. The gate voltage spike due to Cdv/dt current is reduced below the threshold voltage at various testing conditions.
Due to the voltage mismatch between the phase legs and the DC bus in Modular Multilevel Converter... more Due to the voltage mismatch between the phase legs and the DC bus in Modular Multilevel Converters (MMCs), the differential current in MMCs is inherently subjected to circulating even order harmonics. Repetitive control based active harmonic suppression methods can be adopted to eliminate such harmonics. Nevertheless, conventional repetitive controllers have a relatively slow dynamic response because all the sampled errors in the past one cycle have to be stored, which causes a response delay for one fundamental period. This paper proposes an improved repetitive control scheme that exclusively copes with even order harmonics based on the circulating current characteristics of MMC systems. The design details of the even harmonic repetitive control scheme according to the harmonics characteristics are provided. The proposed even-harmonic repetitive control scheme requires halved data memory to store error samplings and the delay introduced by the repetitive controller is also reduced. According to the frequency domain analysis, the even-harmonic repetitive control features faster convergence rate, greater low-frequency gains, higher crossover frequency, and higher tolerance against system frequency deviation, while possessing the same even-order harmonics suppression capability and stability as conventional ones. Simulation and experimental results are presented to show the steady-state harmonics suppression capability, dynamic response, and disturbance tolerance of the proposed even-harmonic repetitive control scheme.
h i g h l i g h t s Integrated online model identification and SOC estimate is explored. Noise va... more h i g h l i g h t s Integrated online model identification and SOC estimate is explored. Noise variances are online estimated in a data-driven way. Identification bias caused by noise corruption is attenuated. SOC is online estimated with high accuracy and fast convergence. Algorithm comparison shows the superiority of proposed method.
This paper presents the modeling and position-sensorless vector control of a dual-airgap axial fl... more This paper presents the modeling and position-sensorless vector control of a dual-airgap axial flux permanent magnet (AFPM) machine optimized for use in flywheel energy storage system (FESS) applications. The proposed AFPM machine has two sets of three-phase stator windings but requires only a single power converter to control both the electromagnetic torque and the axial levitation force. The proper controllability of the latter is crucial as it can be utilized to minimize the vertical bearing stress to improve the efficiency of the FESS. The method for controlling both the speed and axial displacement of the machine is discussed. An inherent speed sensorless observer is also proposed for speed estimation. The proposed observer eliminates the rotary encoder, which in turn reduces the overall weight and cost of the system while improving its reliability. The effectiveness of the proposed control scheme has been verified by simulations and experiments on a prototype machine.
2015 IEEE International Telecommunications Energy Conference (INTELEC), 2015
In recent years, there is growing interest in improving lithium-ion battery state monitoring. Thi... more In recent years, there is growing interest in improving lithium-ion battery state monitoring. This is due to growing concern over ensuring reliable and safe operation of lithium-ion based energy storage system especially in automotive and stationary e.g. UPS in data center applications. In this paper, battery state of charge estimation based on Extended Kalman Filter (EKF) and improved battery model considering hysteresis is presented. Hysteresis in battery is complex phenomenon and normally not considered in SOC estimation. However, the consideration is crucial especially in lithium iron phosphate (LFP) battery. Simplification of hysteresis in open circuit voltage (OCV) in the battery leads to higher percentage error in SOC estimation. To consider and model hysteresis in OCV, discrete Preisach model (DPM) is utilized to provide satisfactory improvement in SOC estimation.
IECON 2015 - 41st Annual Conference of the IEEE Industrial Electronics Society, 2015
This paper investigates the energy efficiency of Li-ion battery used as energy storage devices in... more This paper investigates the energy efficiency of Li-ion battery used as energy storage devices in a micro-grid. The overall energy efficiency of Li-ion battery depends on the energy efficiency under charging, discharging, and charging-discharging conditions. These three types of energy efficiency of single battery cell have been calculated under different current rates. The relations of the energy efficiencies with current rate are analyzed using non-linear curve fitting and polynomial equations. Moreover, the Li-ion battery used in the present study has been verified to have low cell imbalance when 10 cells are connected in series without any cell equalizers, and the charging-discharging energy efficiency of the battery pack has been obtained through hardware experiments.
2020 3rd International Conference on Energy, Power and Environment: Towards Clean Energy Technologies
The energy storage power station (ESPS) is one of the most used energy storage system (ESS). It c... more The energy storage power station (ESPS) is one of the most used energy storage system (ESS). It can also be used to regulate the reactive power thus maintaining the micro-grid side voltage so no more static var generator needed. However, it requires a higher level controller to achieve that called coordinate control. The coordinate controller decides how much reactive power the ESPS should generate based on the voltage drop at the micro-grid side. And the reference reactive power signal is sent to the controller through a certain communication protocol, IEC61850-Generic Object-Oriented Substation Event (GOOSE) for example. The communication delay in such a protocol is ineligible. Such communication delay can seriously affect the performance of the reactive power control. To overcome this, Smith Predictor (SM) based delay compensation control is proposed and analyzed. Control hardware-in-the-loop results are presented to verify the effectiveness of the proposed method.
2016 IEEE PES Asia-Pacific Power and Energy Engineering Conference (APPEEC), 2016
Phasor Measurement Unit (PMU) has been applied on power system to improve the performance of powe... more Phasor Measurement Unit (PMU) has been applied on power system to improve the performance of power system state estimator for decades. It is anticipated that PMUs will observe distribution networks in future considering the contribution of PMUs to transmission networks. The advent of micro phasor measurement unit (μPMU) makes it possible to employ phasor measurement devices in distribution networks. Optimal PMU Placement (OPP) problem is to determine a minimal set of PMUs such that the whole system is observable. Since PMUs are mostly employed within transmission networks, most OPP algorithms are also designed for transmission networks. However, OPP algorithms designed transmission networks have not considered the radial topology of distribution networks. This paper investigated the different topology of transmission and distribution networks and proposed a customized OPP method for radial networks. The customized OPP method made use of radial topology of distribution networks to equivalently downsize the scale of the networks. Since the solution for OPP is usually difficult for large scale systems, the solution of OPP is accelerated by the customized OPP. The proposed method was tested on various IEEE distribution test feeders and transmission systems.
IECON 2011 - 37th Annual Conference of the IEEE Industrial Electronics Society, 2011
Inefficiency of the common lightings like incandescent, fluorescent and halogen lamps presents an... more Inefficiency of the common lightings like incandescent, fluorescent and halogen lamps presents an ample opportunity to retrofit high efficacy, long life-time and low maintenance LED solid-state lighting, which has significantly reduced in cost on an annual basis. LED lights and many electronic loads used within in-built environment are DC in nature. As such, in this paper, an energy-efficient low voltage
2017 IEEE Applied Power Electronics Conference and Exposition (APEC), 2017
The state of energy (SOE) of Li-ion batteries is a key indicator for the energy optimization and ... more The state of energy (SOE) of Li-ion batteries is a key indicator for the energy optimization and management of energy storage devices (ESDs) in electric vehicles (EVs) and smart grids. To improve the SOE estimation accuracy, a hybrid Li-ion battery model is presented in this study against the dynamic loads and the rate energy effects of the battery. Firstly, in order to take the advantages of both models, a 2nd order RC battery model is merged with an analytical kinetic battery model for accurately predicting the battery voltage characteristics and capturing the nonlinear rate energy effects to realize the high-fidelity SOE and runtime prediction. Secondly, a new method to separate the fast and slow dynamics of the 2nd order RC model is developed and presented with high-performance accuracy. Thirdly, commercial Li-ion batteries are tested at dynamic loads under various temperature to validate the effectiveness of the proposed model. The experimental results show high accuracy and reliability of the proposed battery model on the estimation of the battery SOE and the battery terminal voltage responses under dynamic loads.
2018 International Power Electronics Conference (IPEC-Niigata 2018 -ECCE Asia), 2018
Reliable online co-estimation of state of health (SOH) and state of charge (SOC) of Li-ion batter... more Reliable online co-estimation of state of health (SOH) and state of charge (SOC) of Li-ion batteries were of paramount importance for the realistic battery management system (BMS). This work aimed at bridging laboratory test and real-life battery operation data with a comprehensive analysis to provide a coherent and non-invasive approach based on probability density function (PDF), for building a pragmatic model to co-evaluate SOH and SOC of Li-ion batteries for smart grid applications. PDF results based on practical applications revealed that there was a prominent regularity of the voltage probabilities with regards to the SOH, which were exploited for setting up an online SOH evaluation scale gauge. Utilization of the load current of realistic smart grid further improve the practical generality of the proposed algorithm. The battery online SOC was determined afterwards based on the online OCV variation and the extracted SOH values. Commercial Li-ion batteries at arbitrary SOH and SOC level were tested to validate the effectiveness and robustness of the proposed algorithm, and the test results showed high accuracy and reliability of the proposed algorithm for co-evaluating SOH and SOC.
IAS '96. Conference Record of the 1996 IEEE Industry Applications Conference Thirty-First IAS Annual Meeting
This paper presents a physics-based electric arc model which can realistically simulate both stat... more This paper presents a physics-based electric arc model which can realistically simulate both static and dynamic characteristics of an arc load controlled by a power electronic circuit. The proposed model was developed from modifications of the classical Cassie and Mayr equations. The model equations have been expressed in a form suitable for incorporation into circuit simulators employing the nodal analysis method of equation solving. The model has been test implemented in the Saber circuit simulator. There has been a considerable degree of correspondence between the simulated and experimental results
Thermal comfort is a key consideration in smart buildings and a number of comfort models are avai... more Thermal comfort is a key consideration in smart buildings and a number of comfort models are available nowadays to evaluate the comfort level of occupants. However, the models are often complex and hardly interpretable for the developers and operators. Indeed, the model interpretations are beneficial in multifold such as for system inspection and optimization. In this paper, we propose an interpretable thermal comfort system to introduce interpretability to any black-box comfort models. First, we focus on the relationship between a model's input features and output comfort level. The feature impact on comfort is investigated and the impact patterns are shown to be diverse for different features. Second, we unveil the model mechanisms about the data processing inside the model by building the model surrogates based on the interpretable machine learning algorithms. The surrogates offer outstanding fidelity for simulating the actual model mechanisms and the interpretations based on the surrogates are intuitive and informative. Our interpretable comfort system can be integrated with the existing building management systems. Accordingly, we can ease building owner's concerns about adopting new black-box technologies and enable various smart building applications like smart energy management. Index Terms-Thermal comfort; smart building; smart city; deep learning; interpretable machine learning. I. INTRODUCTION T HERMAL comfort is an essential consideration of smart buildings. The impact of thermal comfort can be either positive or negative. Satisfied comfort ameliorates the occupant's well-being, happiness, and working efficiency [1], [2], while discomfort results in the sick building syndrome, absenteeism, and degraded cognitive performance [3]. Especially, modern people spend as much as 90% of their time in buildings [4]. As such, the Internet of things (IoT) system
2020 3rd International Conference on Energy, Power and Environment: Towards Clean Energy Technologies, 2021
This paper proposes the method to analyze the performance of wireless charging DC/DC resonant con... more This paper proposes the method to analyze the performance of wireless charging DC/DC resonant converters in various load conditions including the magnetic coupling effect. The equations for design performance analytical calculations are presented for series-series (SS) and series-parallel (SP) wireless charging DC/DC resonant converters. The equations can be applied in switching frequency sensitivity analysis at different magnetic strength and load resistance conditions to compare the performance of the resonant circuit topologies. The analytical calculation results can be used to size the voltage, current, and power rating of wireless charging resonant circuit components. The accuracy of the analytical calculation is validated by duty cycle control of wireless charging resonant converters in PLECS circuit simulation environment. Experimental results are obtained to support the conclusion drawn from the analytical solutions. The proposed method can be applied to study other wireless charging resonant converter topologies.
The state of energy (SOE) is a key indicator for the energy optimization and management of Li-ion... more The state of energy (SOE) is a key indicator for the energy optimization and management of Li-ion battery-based energy storage systems in the smart grid applications. To improve the SOE estimation accuracy, a Li-ion battery model is presented in this study against dynamic loads and battery ageing effects. Firstly, an electrical battery model is combined with an analytical model in order to take advantages of both models for accurate prediction of battery terminal voltage characteristics, SOE and remnant runtime. Secondly, a novel method to separate the fast and slow dynamics of the electrical battery model is de veloped, and its superior performance is presented. Thirdly, the effects of the battery initial SOC, load current rate and direction, operating temperature and ageing level are systematically scrutinized and involved into the proposed model for robust SOE and terminal voltage prediction. Commercial Li-ion batteries are then tested under dynamic loads and at an arbitrary battery ageing le vel to validate the effectiveness and robustness of the proposed model. The laboratory-scale experimental test results show superb accuracy and reliability of the proposed battery model for estimating battery SOE and terminal voltage under dynamic loads and battery ageing conditions.
Three co-estimation methods are compared for lithium-ion battery application. • A noise compensat... more Three co-estimation methods are compared for lithium-ion battery application. • A noise compensating method is exploited for co-estimation purpose. • Noise effects on different methods are analyzed with simulations and experiments. • The computing cost and tuning effort are discussed.
Reliable online estimation of state of charge (SOC) and capacity is critically important for the ... more Reliable online estimation of state of charge (SOC) and capacity is critically important for the battery management system. This paper presents a multi-timescale estimator to dually estimate the SOC and capacity for lithium-ion battery. The first-order RC model is used to simulate the dynamics of lithium-ion battery. Based on the battery model, the open circuit voltage (OCV) is timely updated with a simple OCV, the result of which is further corrected with the Kalman filter (KF). Then the SOC is inferred from the SOC-OCV look-up table. Meanwhile, a RLS-based capacity estimator is formulated to work simultaneously with the SOC estimation in the form of dual estimation. Different timescales are adopted for the dual estimator to improve accuracy and stability. Experimental results suggest that the proposed method estimates SOC and capacity in real time with fast convergence and high precision.
The high switching speed in a silicon carbide (SiC) metal-oxide-semiconductor field-effect transi... more The high switching speed in a silicon carbide (SiC) metal-oxide-semiconductor field-effect transistor (MOSFET) will aggravate the parasitic effects (di/dt and dv/dt) arising from the interaction with parasitic elements. In this project, a high-speed gate driver has been developed and optimised for the commercially available SiC MOSFET power module. The impact of various parasitic parameters on parasitic effects is initially evaluated. Then, an improved gate-assisted circuit is proposed with a local low-impedance path for both discharging and Cdv/dt currents. It allows maximised turn-off speed (dv/dt up to 36 V/ns) and minimised turn-off loss (reduction up to 70%). It also produces a reduction in electromagnetic interference. The gate voltage spike due to Cdv/dt current is reduced below the threshold voltage at various testing conditions.
Due to the voltage mismatch between the phase legs and the DC bus in Modular Multilevel Converter... more Due to the voltage mismatch between the phase legs and the DC bus in Modular Multilevel Converters (MMCs), the differential current in MMCs is inherently subjected to circulating even order harmonics. Repetitive control based active harmonic suppression methods can be adopted to eliminate such harmonics. Nevertheless, conventional repetitive controllers have a relatively slow dynamic response because all the sampled errors in the past one cycle have to be stored, which causes a response delay for one fundamental period. This paper proposes an improved repetitive control scheme that exclusively copes with even order harmonics based on the circulating current characteristics of MMC systems. The design details of the even harmonic repetitive control scheme according to the harmonics characteristics are provided. The proposed even-harmonic repetitive control scheme requires halved data memory to store error samplings and the delay introduced by the repetitive controller is also reduced. According to the frequency domain analysis, the even-harmonic repetitive control features faster convergence rate, greater low-frequency gains, higher crossover frequency, and higher tolerance against system frequency deviation, while possessing the same even-order harmonics suppression capability and stability as conventional ones. Simulation and experimental results are presented to show the steady-state harmonics suppression capability, dynamic response, and disturbance tolerance of the proposed even-harmonic repetitive control scheme.
h i g h l i g h t s Integrated online model identification and SOC estimate is explored. Noise va... more h i g h l i g h t s Integrated online model identification and SOC estimate is explored. Noise variances are online estimated in a data-driven way. Identification bias caused by noise corruption is attenuated. SOC is online estimated with high accuracy and fast convergence. Algorithm comparison shows the superiority of proposed method.
This paper presents the modeling and position-sensorless vector control of a dual-airgap axial fl... more This paper presents the modeling and position-sensorless vector control of a dual-airgap axial flux permanent magnet (AFPM) machine optimized for use in flywheel energy storage system (FESS) applications. The proposed AFPM machine has two sets of three-phase stator windings but requires only a single power converter to control both the electromagnetic torque and the axial levitation force. The proper controllability of the latter is crucial as it can be utilized to minimize the vertical bearing stress to improve the efficiency of the FESS. The method for controlling both the speed and axial displacement of the machine is discussed. An inherent speed sensorless observer is also proposed for speed estimation. The proposed observer eliminates the rotary encoder, which in turn reduces the overall weight and cost of the system while improving its reliability. The effectiveness of the proposed control scheme has been verified by simulations and experiments on a prototype machine.
2015 IEEE International Telecommunications Energy Conference (INTELEC), 2015
In recent years, there is growing interest in improving lithium-ion battery state monitoring. Thi... more In recent years, there is growing interest in improving lithium-ion battery state monitoring. This is due to growing concern over ensuring reliable and safe operation of lithium-ion based energy storage system especially in automotive and stationary e.g. UPS in data center applications. In this paper, battery state of charge estimation based on Extended Kalman Filter (EKF) and improved battery model considering hysteresis is presented. Hysteresis in battery is complex phenomenon and normally not considered in SOC estimation. However, the consideration is crucial especially in lithium iron phosphate (LFP) battery. Simplification of hysteresis in open circuit voltage (OCV) in the battery leads to higher percentage error in SOC estimation. To consider and model hysteresis in OCV, discrete Preisach model (DPM) is utilized to provide satisfactory improvement in SOC estimation.
IECON 2015 - 41st Annual Conference of the IEEE Industrial Electronics Society, 2015
This paper investigates the energy efficiency of Li-ion battery used as energy storage devices in... more This paper investigates the energy efficiency of Li-ion battery used as energy storage devices in a micro-grid. The overall energy efficiency of Li-ion battery depends on the energy efficiency under charging, discharging, and charging-discharging conditions. These three types of energy efficiency of single battery cell have been calculated under different current rates. The relations of the energy efficiencies with current rate are analyzed using non-linear curve fitting and polynomial equations. Moreover, the Li-ion battery used in the present study has been verified to have low cell imbalance when 10 cells are connected in series without any cell equalizers, and the charging-discharging energy efficiency of the battery pack has been obtained through hardware experiments.
Uploads
Papers by King Jet Tseng