Daniel Zakzewski
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Daniel Zakzewski has a total of 6 co-authors. Here are the top co-authors by number of shared publications:
- Alireza Khaligh
Maryland Power Electronics Laboratory, Department of Electrical and Computer Engineering, Institute for Systems Research, University of Maryland, College Park, MD, USA
6 shared publications - Rakesh Resalayyan
Department of Electrical and Computer Engineering, Maryland Power Electronics Laboratory, Institute for Systems Research, University of Maryland, College Park, MD, USA
5 shared publications - Arafat Hasnain
Department of Electrical and Computer Engineering, Maryland Power Electronics Laboratory, Institute for Systems Research, University of Maryland, College Park, MD, USA
3 shared publications - Yidi Shen
Department of Electrical and Computer Engineering, Maryland Power Electronics Laboratory, Institute for Systems Research, University of Maryland, College Park, MD, USA
3 shared publications - Ayooluwa Ajiboye
Maryland Power Electronics Laboratory, Department of Electrical and Computer Engineering, Institute for Systems Research, University of Maryland, College Park, MD, USA
1 shared publication - Benjamin Shani
Maryland Power Electronics Laboratory, Department of Electrical and Computer Engineering, Institute for Systems Research, University of Maryland, College Park, MD, USA
1 shared publication
Latest Publications
- 22 January 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Industrial Electronics
- Vol. PP (02780046) , 1-13
A 19-Level Hybrid Neutral Point Clamped Converter With Trinary Level Scaling and Model Predictive Control for Capacitor Balancing
Hybrid neutral-point clamped (HNPC) converters have received a wide-level of research focus in multilevel inverter applications due to their power quality and loss-improvement benefits. However, these converters are typically limited to a singular floating H-bridge (FHB) stage due to complex capacitor regulation and poor device utilization as compared with multilevel designs using flying capacitor stages. Therefore, this work proposes a 19-level topology consisting of a 3L-active neutral point clamped (ANPC) base cascaded with two FHB stages, thereby enabling several improvements to the conventional FHB-based HNPC converter. As opposed to traditional modulation techniques, which achieve nine-levels, the proposed method includes a novel use of trinary level scaling, which expands the number of output levels to 19 in the normal modulation range, and up to 23 in the extended modulation range. This results in major benefits such as the decreased switch voltage ratings, improved output power quality, and enhanced power density. Furthermore, this work introduces a joint-phase redundancy balancing technique combined with a multiobjective finite control-set model predictive control (FCS-MPC) algorithm aimed at achieving switching loss minimization and strict floating capacitor voltage regulation at the switching frequency level. The proposed FCS-MPC strategy simplifies the calculation of the available redundancies in the HNPC converter, enabling feasibility of implementation on a standard microcontroller with execution times compatible with switching-frequency synchronous control. To validate the proposed topology and control strategy, steady-state and transient experiments are conducted on a 2 kW HNPC prototype. The results clearly demonstrate significant performance improvements, thereby confirming the advantages of the proposed solution in practical applications.
- 22 January 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Industrial Electronics
- Vol. PP (02780046) , 1-13
A 19-Level Hybrid Neutral Point Clamped Converter With Trinary Level Scaling and Model Predictive Control for Capacitor Balancing
Hybrid neutral-point clamped (HNPC) converters have received a wide-level of research focus in multilevel inverter applications due to their power quality and loss-improvement benefits. However, these converters are typically limited to a singular floating H-bridge (FHB) stage due to complex capacitor regulation and poor device utilization as compared with multilevel designs using flying capacitor stages. Therefore, this work proposes a 19-level topology consisting of a 3L-active neutral point clamped (ANPC) base cascaded with two FHB stages, thereby enabling several improvements to the conventional FHB-based HNPC converter. As opposed to traditional modulation techniques, which achieve nine-levels, the proposed method includes a novel use of trinary level scaling, which expands the number of output levels to 19 in the normal modulation range, and up to 23 in the extended modulation range. This results in major benefits such as the decreased switch voltage ratings, improved output power quality, and enhanced power density. Furthermore, this work introduces a joint-phase redundancy balancing technique combined with a multiobjective finite control-set model predictive control (FCS-MPC) algorithm aimed at achieving switching loss minimization and strict floating capacitor voltage regulation at the switching frequency level. The proposed FCS-MPC strategy simplifies the calculation of the available redundancies in the HNPC converter, enabling feasibility of implementation on a standard microcontroller with execution times compatible with switching-frequency synchronous control. To validate the proposed topology and control strategy, steady-state and transient experiments are conducted on a 2 kW HNPC prototype. The results clearly demonstrate significant performance improvements, thereby confirming the advantages of the proposed solution in practical applications.
- 6 May 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Power Electronics
- Vol. 40 (9) , 12648-12660
Improved Single DC Source Hybrid Neutral Point Clamped Converter With 13 Levels
In this article, a 13-level circuit topology is proposed, which features a single dc source, extended linear modulation range, and a high ratio of output voltage levels to semiconductor devices. The proposed converter, which consists of a series-connected floating H-bridge (FHB) supplementing a five-level active neutral point clamped converter (5L-ANPC), can generate 13 pole voltage levels in the normal modulation region and 15 levels in the extended modulation region. The key performance improvements offered by the proposed topology are enhanced power quality, reduced losses, and increased power density. Whereas existing schemes with a similar circuit structure regulate the FHB to one-eighth of the dc-link voltage and only achieve nine levels, the proposed joint-phase redundancy balancing approach regulates the FHB stage to one-twelfth of the total dc-link voltage on a switching-frequency basis, thereby increasing output power quality while reducing the converter's cost and size without the need for multiple isolated supplies. Furthermore, the proposed modulation scheme also allows for a switching frequency reduction of the higher voltage blocking 5L-ANPC stage by shifting the high-frequency switching to the low voltage blocking FHB stage, thus reducing losses. This work also evaluates operation in the topology's extended linear modulation range, which is shown to be increased to 0.612 times the dc-link voltage (modulation index, $M = 1.223$) without limitations on load power factors, compared to the typical limit of to 0.577 times the dc-link voltage ($M = 1.154$) for nonboosting topologies such as 3L-ANPC and 5L-ANPC. The proposed converter uses boosting states enabled by the FHB in this extended modulation region, which increases the number of output levels to 15. The effectiveness of the proposed topology and capacitor balancing scheme is validated using experimental and analytical results. Based on experimental and analytical results, the proposed topology is shown to have improved power quality, efficiency, size, and weight relative to the existing solutions.
- 6 May 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Power Electronics
- Vol. 40 (9) , 12648-12660
Improved Single DC Source Hybrid Neutral Point Clamped Converter With 13 Levels
In this article, a 13-level circuit topology is proposed, which features a single dc source, extended linear modulation range, and a high ratio of output voltage levels to semiconductor devices. The proposed converter, which consists of a series-connected floating H-bridge (FHB) supplementing a five-level active neutral point clamped converter (5L-ANPC), can generate 13 pole voltage levels in the normal modulation region and 15 levels in the extended modulation region. The key performance improvements offered by the proposed topology are enhanced power quality, reduced losses, and increased power density. Whereas existing schemes with a similar circuit structure regulate the FHB to one-eighth of the dc-link voltage and only achieve nine levels, the proposed joint-phase redundancy balancing approach regulates the FHB stage to one-twelfth of the total dc-link voltage on a switching-frequency basis, thereby increasing output power quality while reducing the converter's cost and size without the need for multiple isolated supplies. Furthermore, the proposed modulation scheme also allows for a switching frequency reduction of the higher voltage blocking 5L-ANPC stage by shifting the high-frequency switching to the low voltage blocking FHB stage, thus reducing losses. This work also evaluates operation in the topology's extended linear modulation range, which is shown to be increased to 0.612 times the dc-link voltage (modulation index, $M = 1.223$) without limitations on load power factors, compared to the typical limit of to 0.577 times the dc-link voltage ($M = 1.154$) for nonboosting topologies such as 3L-ANPC and 5L-ANPC. The proposed converter uses boosting states enabled by the FHB in this extended modulation region, which increases the number of output levels to 15. The effectiveness of the proposed topology and capacitor balancing scheme is validated using experimental and analytical results. Based on experimental and analytical results, the proposed topology is shown to have improved power quality, efficiency, size, and weight relative to the existing solutions.
- 18 April 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Power Electronics
- Vol. 40 (9) , 13499-13513
Improved Feedforward Control of Active Power Decoupler for PV Microinverter With Nonsinusoidal Power Decoupling Capability
In single-phase power conversion systems, there is an inherent difference between the dc-side constant and ac-side oscillating power, and power decoupling is required especially in applications like photovoltaic (PV) inverters. Active power decoupling (APD) converters are often used to balance instantaneous power while reducing the size of dc link capacitors and improving the system reliability. Conventionally, APD converters are controlled to decouple only sinusoidal power at double-line frequency. However, the control is further complicated when nonsinusoidal power is delivered due to nonlinear circuit components, nonlinear loads, or grid distortions. This article proposes an improved feedforward APD control scheme with the capability of decoupling nonsinusoidal power, which has application in single-phase systems, particularly all types of PV microinverters. The proposed feedforward control includes the capacitor voltage calculation under arbitrary power harmonics and a dead-time compensation to increase the accuracy of duty prediction. The proposed algorithm can work as open-loop control to eliminate APD sensors and can serve as a feedforward term in closed-loop control to improve power decoupling performance, thereby simplifying the controller design. The proposed approach is experimentally verified using a dc-parallel APD prototype, a 50 V 250 W microinverter with nonlinear ac capacitor, and a diode rectifier load.
- 18 April 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Power Electronics
- Vol. 40 (9) , 13499-13513
Improved Feedforward Control of Active Power Decoupler for PV Microinverter With Nonsinusoidal Power Decoupling Capability
In single-phase power conversion systems, there is an inherent difference between the dc-side constant and ac-side oscillating power, and power decoupling is required especially in applications like photovoltaic (PV) inverters. Active power decoupling (APD) converters are often used to balance instantaneous power while reducing the size of dc link capacitors and improving the system reliability. Conventionally, APD converters are controlled to decouple only sinusoidal power at double-line frequency. However, the control is further complicated when nonsinusoidal power is delivered due to nonlinear circuit components, nonlinear loads, or grid distortions. This article proposes an improved feedforward APD control scheme with the capability of decoupling nonsinusoidal power, which has application in single-phase systems, particularly all types of PV microinverters. The proposed feedforward control includes the capacitor voltage calculation under arbitrary power harmonics and a dead-time compensation to increase the accuracy of duty prediction. The proposed algorithm can work as open-loop control to eliminate APD sensors and can serve as a feedforward term in closed-loop control to improve power decoupling performance, thereby simplifying the controller design. The proposed approach is experimentally verified using a dc-parallel APD prototype, a 50 V 250 W microinverter with nonlinear ac capacitor, and a diode rectifier load.
- 5 April 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Journal of Emerging and Selected Topics in Industrial Electronics
- Vol. 5 (4) , 1582-1592
Class II Ceramic Capacitor Voltage Characteristic Modeling and Compensation for AC-Connected Applications
Dielectrics used in class II multi-layer ceramic capacitors offer vastly higher dielectric constants than class I ceramics or film capacitors, leading to large volumetric and cost savings. However, class II ceramic capacitors undesirably tend to have varying capacitance depending on the voltage applied across it, referred to as voltage characteristic of capacitance (VCC). When class II ceramic capacitors are connected to an ac-source, VCC can contribute large harmonic content, adversely impacting power quality. If left unmitigated, additional filtering may be required to meet electromagnetic compatibility (EMC) standards, thus reducing volumetric benefits of class II ceramic capacitors. Furthermore, control systems relying on accurate feedforward require predictive control of such effects. This work introduces a compensation strategy to mitigate current distortions induced by VCC, leading to an enhancement in converter power quality without the need for additional filtering. To achieve satisfactory mitigation with minimal control burden, the paper also proposes a mathematical model describing the voltage characteristic. A design application is used to demonstrate the relevant trade-offs and the effectiveness of the proposed compensation approaches in enhancing power quality performance is validated through experimental testing.
- 5 April 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Journal of Emerging and Selected Topics in Industrial Electronics
- Vol. 5 (4) , 1582-1592
Class II Ceramic Capacitor Voltage Characteristic Modeling and Compensation for AC-Connected Applications
Dielectrics used in class II multi-layer ceramic capacitors offer vastly higher dielectric constants than class I ceramics or film capacitors, leading to large volumetric and cost savings. However, class II ceramic capacitors undesirably tend to have varying capacitance depending on the voltage applied across it, referred to as voltage characteristic of capacitance (VCC). When class II ceramic capacitors are connected to an ac-source, VCC can contribute large harmonic content, adversely impacting power quality. If left unmitigated, additional filtering may be required to meet electromagnetic compatibility (EMC) standards, thus reducing volumetric benefits of class II ceramic capacitors. Furthermore, control systems relying on accurate feedforward require predictive control of such effects. This work introduces a compensation strategy to mitigate current distortions induced by VCC, leading to an enhancement in converter power quality without the need for additional filtering. To achieve satisfactory mitigation with minimal control burden, the paper also proposes a mathematical model describing the voltage characteristic. A design application is used to demonstrate the relevant trade-offs and the effectiveness of the proposed compensation approaches in enhancing power quality performance is validated through experimental testing.
- 29 December 2023
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Transportation Electrification
- Vol. 10 (3) , 6087-6099
Hybrid Neutral Point Clamped Converter: Review and Comparison to Traditional Topologies
In applications such as transportation electrification there are strict requirements regarding efficiency, reliability, power density, and cost. Therefore, a diverse array of circuit topologies which can be selected and tailored for specific requirements is desirable. Hybrid topologies are particularly valuable in this regard because hybrid topologies can combine the attributes of traditional multilevel converter (MLC) topologies, such as cascaded H-bridge, flying capacitor (FC) converter, or three-level (3L) neutral point clamped (NPC) converter. A hybrid topology which has obtained substantial popularity combines NPC with FC to form the five-level active neutral point clamped (5L-ANPC) topology. The popularity of 3L-NPC and 5L-ANPC has been encouraging researchers to develop other hybrid MLC topologies. Therefore, to establish the present state of the art of hybrid NPC topologies, a literature review is conducted. Based on results of the literature review, several figures of merit are derived to objectively compare the system-level attributes of this class of power circuit topologies. These figures of merit are also compared to traditional topologies to demonstrate the utility of hybrid NPC topologies in terms of efficiency, power density and cost. This work will serve as a selection guide for future design efforts considering hybrid NPC topologies.
- 29 December 2023
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Transportation Electrification
- Vol. 10 (3) , 6087-6099
Hybrid Neutral Point Clamped Converter: Review and Comparison to Traditional Topologies
In applications such as transportation electrification there are strict requirements regarding efficiency, reliability, power density, and cost. Therefore, a diverse array of circuit topologies which can be selected and tailored for specific requirements is desirable. Hybrid topologies are particularly valuable in this regard because hybrid topologies can combine the attributes of traditional multilevel converter (MLC) topologies, such as cascaded H-bridge, flying capacitor (FC) converter, or three-level (3L) neutral point clamped (NPC) converter. A hybrid topology which has obtained substantial popularity combines NPC with FC to form the five-level active neutral point clamped (5L-ANPC) topology. The popularity of 3L-NPC and 5L-ANPC has been encouraging researchers to develop other hybrid MLC topologies. Therefore, to establish the present state of the art of hybrid NPC topologies, a literature review is conducted. Based on results of the literature review, several figures of merit are derived to objectively compare the system-level attributes of this class of power circuit topologies. These figures of merit are also compared to traditional topologies to demonstrate the utility of hybrid NPC topologies in terms of efficiency, power density and cost. This work will serve as a selection guide for future design efforts considering hybrid NPC topologies.
- 17 August 2023
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Power Electronics
- Vol. 38 (11) , 14628-14637
Sensorless Control for DC-Parallel Active Power Decoupling in PV Microinverters
Active power decoupling (APD) is a highly adopted technique for compact and reliable single-phase inverters to eliminate the undesired ac power drawn from the dc port. Typical operation of APD circuits requires the use of additional sensors measuring APD states to ensure proper decoupling. The additional voltage and current sensors add to system cost and hardware design complexity. This paper proposes a sensorless control approach that provides proper control of the APD without the use of additional sensors. The proposed predictive power decoupling control is analyzed together with the operating principle of the APD, and two types of error correctors are introduced to improve the control accuracy and reduce the reliance on the circuit component parameters. The proposed approach provides equally good power decoupling performance compared to traditional sensor-based control strategies while improving the dynamic response. The proposed approach is experimentally verified using a 30 - 60 V parallel APD circuit and 200 W microinverter hardware prototype.
- 17 August 2023
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Power Electronics
- Vol. 38 (11) , 14628-14637
Sensorless Control for DC-Parallel Active Power Decoupling in PV Microinverters
Active power decoupling (APD) is a highly adopted technique for compact and reliable single-phase inverters to eliminate the undesired ac power drawn from the dc port. Typical operation of APD circuits requires the use of additional sensors measuring APD states to ensure proper decoupling. The additional voltage and current sensors add to system cost and hardware design complexity. This paper proposes a sensorless control approach that provides proper control of the APD without the use of additional sensors. The proposed predictive power decoupling control is analyzed together with the operating principle of the APD, and two types of error correctors are introduced to improve the control accuracy and reduce the reliance on the circuit component parameters. The proposed approach provides equally good power decoupling performance compared to traditional sensor-based control strategies while improving the dynamic response. The proposed approach is experimentally verified using a 30 - 60 V parallel APD circuit and 200 W microinverter hardware prototype.