Composite dielectric structures with enhanced lifetimes

Lead Research Organisation: University of Manchester
Department Name: Electrical and Electronic Engineering

Abstract

For many years the engineering behind our power transmission and distribution networks has remained unchanged. One reason for this is that the reliability of electrical insulation systems, also found on our rail infrastructure, has proved extremely high. However much of the existing infrastructure is now very old, well beyond its design life. In addition, radical changes are now being implemented in power systems. These are largely driven by the desire to facilitate renewable (low carbon) energy supplies. Moreover as electricity replaces gas for heating homes, and electric vehicles replace petrol and diesel powered cars, more electricity will be consumed requiring greater densities of power transmission, particularly into our cities. These requirements necessitate higher electrical stresses on insulation and higher temperature ratings, whilst maintaining reliability. For these reasons improving our understanding of the failure mechanisms of insulation and designing improved insulation are both critical to the further development of our power systems. Reliability of insulation systems is also critical in power generation, rail networks and manufacturing industries.

This project will improve our understanding of insulation reliability and develop structured materials with improved performance. In particular a fundamental ageing mechanism in polymeric insulation known as electrical treeing will be studied in detail. Electrical tree growth is a mechanism of long-term failure in polymer insulation systems under high electrical stress and is a process leading to the development of an artefact which resembles a botanical tree. It consists of tubular hollow branches of up to tens of microns in diameter. The presence of a tree eventually leads to insulation failure.
The project will consider how the use of layered dielectrics can enhance insulation life, particularly in the presence of electrical trees. Recent feasibility work between Prof Rowland (University of Manchester) and Prof Choy (UCL) has shown that thin layers of polymers can change tree propagation times by an order of magnitude. Also novel techniques in Manchester have now enabled three-dimensional imaging of the treeing process to be generated. This has used unique experimental facilities and skills for sample preparation at UCL and the imaging capability at the University of Manchester including the Diamond Light Source X-ray facility.

Although this work is based on fundamental science, a key component of the project will be take the findings and develop a framework for improved layered dielectric structures. A route for future development of processes and products for the power networks, mass transit and power electronics industries will be developed. The commitment of external partners to generating this framework has been obtained to ensure technology transfer during the project life. Ultimately this work will contribute to better performing, lower cost and more robust electricity supplies.

Planned Impact

Sustaining reliable and affordable energy supplies is a major challenge in all countries. This has an impact on all aspects of life including heating, communication, transport and industrial/commercial competitiveness. There are also major technical challenges in preparing the UK's electricity networks for a low carbon future. Historically the reliability of electrical insulation systems, also found on our rail infrastructure, has proved extremely high. However much of the existing infrastructure is now very old, well beyond its design life. In addition, radical changes are now being implemented in power systems. These are largely driven by the desire to facilitate renewable (low carbon) energy supplies. Moreover if electric vehicles replace petrol and diesel powered cars and electricity replaces gas for heating homes, more electricity will be consumed requiring greater densities of power transmission, particularly into our cities. These requirements necessitate higher electrical stresses on insulation and higher temperature ratings, whilst maintaining reliability. For these reasons improving our understanding of the failure mechanisms of insulation and designing improved insulation are both critical to the further development of our power systems. Reliability of insulation systems is also critical in power generation, rail networks and manufacturing industries.

Better models of electrical tree growth and ageing in general are essential in managing existing plant, and the outputs from this work will enable Asset Managers to make more informed decisions concerning maintenance and replacement scheduling of existing equipment. This will support optimising the costs and reliability of the networks - a major issue across the globe. The development of improved dielectrics through appropriate filler loading and layering of polymers will provide new materials and design tools for equipment manufacturers. This in turn will increase reliability and reduce the size of high voltage plant, allowing the greater energy densities to be transmitted for future societal needs. Developments may also prove to be a key enabler of power electronics, and even low voltage applications if processing routes are developed for high reliability insulation.

The proposal is for basic research, but resource is planned, and support promised from key industries (network operators, a railway component manufacturer, a power system company, and an applied materials research institute) to enable the next stages of material, process and product development to be charted to ensure science generated is turned into engineering application, drawing in other interested parties as the program develops.

Publications

10 25 50
 
Description Manchester has made some significant advances in the modelling of partial discharges in electrical trees. This will enable us to interpret condition monitoring data of real systems in the electricity utilities with more confidence. This work is now published in a number of papers. We have also had published, and have several Journal papers accepted for publication, that share imaging work from measurements at the Diamond Synchrotron and the Henry Mosely Institute in Manchester. The advancement of imaging techniques is an important contribution in its own right. This has given unique data including the imaging of trees in filled materials for the first time. In addition, we have a comparative study of nano- and micro-filled epoxy materials. For the first time we have shown how electrical tree morphology is changed by micro-filler materials. We showed marked increase in longevity of trees with nano particles and in particular those with nano-particles treated by UCL have much improved distribution and longevity.

Following the original set of hypothesis and aims:
Hypothesis 1: Micro-filled epoxy resins slow tree growth by forcing tree branches to take
tortuous paths around filler particles. The quality of the filler/polymer interface impacts tree
growth rate and its morphology. In contrast, nano-filled epoxy resins slow tree growth by
modification of the polymer morphology.
Objective 1.1: to experimentally quantify in 3 dimensions the geometry and evolution of tree
growth in micro- and nano-filled epoxy resins and determine their impact on life expectancy.
Objective 1.2: to make open access on the internet, a set of 3D tree models in Solid Works™, directly imaged in well-controlled and characterised, filled and unfilled materials.

Both objectives were fully met. Considerable success was met understadning the interaction of tree channel and filler particles. Conferences have been used to promote the use of our data.

Hypothesis 2: Thin layers of controlled material, within bulk dielectrics, can radically improve the
life characteristics of bulk polymer insulation systems.
Objective 2.1: to characterise the growth of trees through thin layers of filler-modified polymer, set within the host material.
Objective 2.2: to develop a model of tree growth propagation through such layers, including
the effects of space charge trapping in the thin layers/interfaces.

Progress was made towards these objectives, but limited progress was made, due to difficulties controlling sample preparation.

Hypothesis 3: The inception of electrical trees is controlled by the physical interface between
the metal electrode and the polymer.
Objective 3.1: to characterise the earliest stages of electrical tree growth in epoxy resin, and
understand the responses to controlled modification of the interface.
Objective 3.2: to develop techniques to manage the metal-polymer interface, increasing tree
inception time, including the use of nano-composite layers.

Significant progress was made concerning understanding tree inception and the role of the needle polymer interface. Studies of issues around voids and low density regions at the needle tip have led to conference and journal publications. We now understand issues arising from needle tip geomtries, and this will influence all fututure work of this type and its interpretation.

Other Objectives:
Objective 4.1: to develop a framework for improved dielectric material design, by modification of electrode/polymer interface and by development of layered dielectric structures.
Objective 4.2: To develop with the power industry a route for future development of process
and products for the power networks, mass transit and power electronics industries.
Some good progress was made developing the framework. However little progress was made into the market as the technology is still early stage for manufacturers. However a new EPSRC funded project EP/T001232/1 is building on this platform and has much more dynamic involvement with industry - so the path is set and being followed for exploitation.
Exploitation Route We will continue to put the image files online to make these avaialble to modellers worldwide.We will continue to consult with our industrial advisory group concerning the route to market, including through Manchester's EPSRC-funded project EP/T001232/1 which is building on this platform and has much more dynamic involvement with industry - so the path is set and being followed for exploitation.
Sectors Energy,Transport

URL https://www.manchestertrees.com/
 
Description The underlying need for improved electrical power systems continues to grow. The push to net zero and international pressures on hydrocarbon resources have accelerated the drive to more intensive use of renewable sources. This in turn leads to a need for more reliability and higher density electrical power distribution. This project carried out basic research on high voltage insulation which underpins the distribution system. The development of improved dielectrics through appropriate filler loading and layering of polymers will provide new materials and design tools for equipment manufacturers is still seen as a key opportunity, including in power electronics, power networks, electrical vehicles and more-electric-aircraft applications. This in turn will increase reliability and increase the energy density of high voltage plant. A key contribution of this project has been development of a technique to visualise electrical tree growth in optically opaque polymers. This is an important step in providing the tools to understand key degradation mechanisms in practical insulation systems. In particular, we have shown in-situ the relationship between micro-filler particles and tree channel propagation. The use of both laboratory and synchrotron X-ray computed tomography (XCT) for imaging the bifurcated void structures presented by electrical trees in unfilled polymeric insulation was demonstrated previously by the investigators in Manchester. In this project, sub-micrometre resolution 3D reconstructions of electrical trees have been created using advanced XCT methods at the Diamond synchrotron light source. In highly filled materials (30% by weight) a radical change in tree growth behaviour is seen, leading to bush tree rather than branch tree growth. This technique can now be used in both laboratory experimentation and also in post-fault asset analysis. We have also used repeated laboratory XCT imaging of trees during their growth to develop an understanding the relationship between partial discharge activity and tree growth. This is key information, allowing the development of models of tree growth and providing data for models which will underpin tools for asset managers in the electricity supply industry. Our publications from this project on partial discharge mechanisms in tree structures are a further development feeding into asset management models. A further contribution was on confirming the importance of particle distribution in filled materials for high voltage application. A highly structured experiment enabled confirmation of the importance of good distribution of nanofiller on insulation performance. This was achieved through filler pre-treatment in the UCL laboratory.
First Year Of Impact 2020
Sector Energy
Impact Types Economic

 
Description Centre for Doctoral Training in Power Networks at the University of Manchester
Amount £86,208 (GBP)
Organisation University of Manchester 
Sector Academic/University
Country United Kingdom
Start 09/2015 
End 09/2020
 
Description DC networks, power quality and plant reliability
Amount £727,155 (GBP)
Funding ID EP/T001232/1 
Organisation Engineering and Physical Sciences Research Council (EPSRC) 
Sector Public
Country United Kingdom
Start 11/2019 
End 01/2023
 
Title XCT imaging of electrical trees 
Description The use of both laboratory and synchrotron X-ray computed tomography (XCT) for imaging the bifurcated void structures presented by electrical trees in polymeric insulation has been illustrated previously by the investigators in Manchester. However, that work was on unfilled polymers which are not representative of many thermoset systems. In particular, no method had been found for exploring the relationship between micro-filler particles and tree growth. In this project, sub-micrometre resolution 3D reconstructions of electrical trees are reported from X-ray Computed Tomography (XCT) using an advanced 'pink beam' at the Diamond synchrotron light source. This work, reported in 'Electrical Tree Growth in Microsilica-Filled Epoxy Resin' , S. Chen, Z. Lv, J. Carr, M. Storm, S. M. Rowland, IEEE Trans on Dielectrics and Electrical Insulation, vol. 27, no. 3, pp. 820-828, 2020, was used to identify the role of filler particles between 1 and 10 µm in size on tree channel propagation. In highly filled materials (30% by weight) a radical change in tree growth behaviour is seen, leading to bush tree rather than branch tree growth. This technique can now be used in both laboratory experimentation and also in post-fault asset analysis. 
Type Of Material Improvements to research infrastructure 
Year Produced 2020 
Provided To Others? Yes  
Impact This tool can be used to improve microfilled polyer insulation systems for high voltage insulation. That opportunity has yet to realised. 
URL https://www.diamond.ac.uk/Home.html;jsessionid=24A4C217C9FDA3DB20A9D66D895FA64A
 
Title Electrical Tree Web site 
Description A place where our imaging work can be obtained 
Type Of Material Database/Collection of data 
Year Produced 2016 
Provided To Others? Yes  
Impact Hopefully researchers form around teh world can use this as data for their work. 
URL https://www.manchestertrees.com/
 
Description Presentation to Xi'an Jioatong University in 2018 
Form Of Engagement Activity A talk or presentation
Part Of Official Scheme? No
Geographic Reach Local
Primary Audience Postgraduate students
Results and Impact A lecture on our imaging work.
Year(s) Of Engagement Activity 2018
 
Description Zijing Distinguished Lecture presented at Tzinghua University, China, October 2018 
Form Of Engagement Activity A talk or presentation
Part Of Official Scheme? No
Geographic Reach Local
Primary Audience Postgraduate students
Results and Impact Sharing research outputs with colleagues in Tzinghua University.
Year(s) Of Engagement Activity 2018