Investigating the Power Density of Power Electronics
Lead Research Organisation:
Cardiff University
Department Name: Sch of Engineering
Abstract
The power density of power electronics is defined by the power processed in a given volume or mass of converter (volumetric and gravimetric power densities respectively) where the converter is normally defined as the combination of power semiconductor devices, the input and output filters, the thermal management system and control electronics that together provide an interface between two or more electrical systems.
The power density of power electronics is an important consideration in modern engineering systems as in many applications it has become attractive to use electrical power distribution and actuation in place of mechanical, hydraulic or pneumatic systems to reduce overall system mass and complexity, along with improving efficiency and reliability. An understanding of the future capabilities of power electronics as well as the bottlenecks limiting increases in performance of power electronic systems will enable the UK to play a leading role in the development and application of this key technology.
Designing a converter is a complex engineering task that balances many interacting pressures: A designer must select from a range of possible designs and technologies, finding the optimal allocation of mass and volume for each component and within each component whilst meeting electrical and thermal specifications. It the hypothesis of this proposal that in order to reach a true power density optimum, the design process must consider system-wide electrical, mechanical and thermal problems simultaneously: It is not enough to design each component individually based on their electrical specification alone.
This research project will produce software that will optimise the power density of a converter subject to a particular electrical and thermal specification by considering all major aspects of the design under a global electrical, mechanical and thermal optimisation framework. By formulating a set of strict rules governing the design and placement of each component within the converter, the effect of basic design decisions and fundamental material properties on the overall power density of a converter can be explored.
The project will provide industry and academia with reliable and justified figures for best-in-class power density achievable now and into the future, identify technology bottlenecks into which application of additional resources bring large improvements in achievable power density, illustrate explicitly important trade-offs in modern converter design and provide converter optimisation tools that can be used and extended by academia and industry.
The power density of power electronics is an important consideration in modern engineering systems as in many applications it has become attractive to use electrical power distribution and actuation in place of mechanical, hydraulic or pneumatic systems to reduce overall system mass and complexity, along with improving efficiency and reliability. An understanding of the future capabilities of power electronics as well as the bottlenecks limiting increases in performance of power electronic systems will enable the UK to play a leading role in the development and application of this key technology.
Designing a converter is a complex engineering task that balances many interacting pressures: A designer must select from a range of possible designs and technologies, finding the optimal allocation of mass and volume for each component and within each component whilst meeting electrical and thermal specifications. It the hypothesis of this proposal that in order to reach a true power density optimum, the design process must consider system-wide electrical, mechanical and thermal problems simultaneously: It is not enough to design each component individually based on their electrical specification alone.
This research project will produce software that will optimise the power density of a converter subject to a particular electrical and thermal specification by considering all major aspects of the design under a global electrical, mechanical and thermal optimisation framework. By formulating a set of strict rules governing the design and placement of each component within the converter, the effect of basic design decisions and fundamental material properties on the overall power density of a converter can be explored.
The project will provide industry and academia with reliable and justified figures for best-in-class power density achievable now and into the future, identify technology bottlenecks into which application of additional resources bring large improvements in achievable power density, illustrate explicitly important trade-offs in modern converter design and provide converter optimisation tools that can be used and extended by academia and industry.
Planned Impact
An understanding of the future capabilities of power electronics as well as the bottlenecks limiting increases in performance of power electronic systems will enable the UK to continue to play a leading role in the development and application of this key technology.
The research will have a multi-fold impact in industries developing and using power electronic systems. The research will
1. Provide researchers and designers in industry and academia with reliable and justified figures for best-in-class power density achievable now and into the future as new technologies appear and mature, providing input into system studies and helping inform planning of long-term projects
2. Provide power density targets target for engineers actively involved with in the development of high performance power electronic systems
3. Inform those system design studies of the high-level effect of converter ratings, topology and technology on power density using a standardised approach.
4. Illustrate complex low-level design trade-offs, for example, what gains may be made by increasing nominal operating voltage or by relaxing converter output ripple limits. These types of specification-driven effects tend to be complex as they cause knock-on effects in both the design of the converter and the overall design of an electrical system.
5. Identify technology bottlenecks into which application of additional resources, either by industry or academia, may bring large improvements in achievable power density
6. Provide industry and academia with an example methodology for converter optimisation for incorporation into wider system optimisation tools
An RA will be employed for 12 months under the project and will gain experience in power electronic theory and design as well as the opportunity to engage with the industrial partners. A PhD student (funded by Cardiff University) will be trained partially under the project, increasing the UK skills base in this area.
The resulting software stack and documentation developed over the course of the project will be provided on a website hosted by Cardiff University and will be available for use by industry.
The research will have a multi-fold impact in industries developing and using power electronic systems. The research will
1. Provide researchers and designers in industry and academia with reliable and justified figures for best-in-class power density achievable now and into the future as new technologies appear and mature, providing input into system studies and helping inform planning of long-term projects
2. Provide power density targets target for engineers actively involved with in the development of high performance power electronic systems
3. Inform those system design studies of the high-level effect of converter ratings, topology and technology on power density using a standardised approach.
4. Illustrate complex low-level design trade-offs, for example, what gains may be made by increasing nominal operating voltage or by relaxing converter output ripple limits. These types of specification-driven effects tend to be complex as they cause knock-on effects in both the design of the converter and the overall design of an electrical system.
5. Identify technology bottlenecks into which application of additional resources, either by industry or academia, may bring large improvements in achievable power density
6. Provide industry and academia with an example methodology for converter optimisation for incorporation into wider system optimisation tools
An RA will be employed for 12 months under the project and will gain experience in power electronic theory and design as well as the opportunity to engage with the industrial partners. A PhD student (funded by Cardiff University) will be trained partially under the project, increasing the UK skills base in this area.
The resulting software stack and documentation developed over the course of the project will be provided on a website hosted by Cardiff University and will be available for use by industry.
People |
ORCID iD |
| Daniel Rogers (Principal Investigator) |
Publications
Choudhury K
(2019)
Steady-State Thermal Modeling of a Power Module: An $N$-Layer Fourier Approach
in IEEE Transactions on Power Electronics
Rogers D
(2016)
Digital Active Gate Drives using sequential optimization
| Description | To-date we have developed thermal and electrical models of sub-systems used in power electronic converters (filters, capacitor, inductor, semiconductors, thermal management system). We have also developed an overall optimization framework that links the sub-system models to produce complete converter designs. We are in the process of releasing this software to our project partners. |
| Exploitation Route | We have an ongoing engagement with industry, including Rolls Royce, Microsemi and IMV Japan. We will be holding a progress meeting in early December. Although funding for this project has finished, work is being actively continued by a PhD student recruited under the original project. |
| Sectors | Aerospace, Defence and Marine,Electronics,Energy |
| Description | Centre for Power Electronics Standard Proposal |
| Amount | £55,546 (GBP) |
| Funding ID | Optimisable system-level thermal models for power electronic converters |
| Organisation | Engineering and Physical Sciences Research Council (EPSRC) |
| Department | Centre for Power Electronics |
| Sector | Academic/University |
| Country | United Kingdom |
| Start | 01/2016 |
| End | 12/2016 |