DC networks, power quality and plant reliability

The movement of electrical energy from generators to customers, through electricity networks, has historically been based on High Voltage Alternating Current (HVAC) technology. This has been a major success of the twentieth century, enabling reliable and stable energy supplies across the developed world. The technology dominated partly as a result of the ability to change voltage levels readily and efficiently using transformers. The alternative technology of High Voltage Direct Current (HVDC) has historically only been used for point-to-point links because of particular advantages in this situation. Now however, with the advent of power electronics, utilisation of HVDC systems is rapidly increasing across the world. This has been accelerated with the growth of renewable distributed energy supplies, such as offshore wind farms in the UK. As a result, local and international energy supplies are becoming dependent on HVDC. Consequently, the reliability of DC technologies is becoming critical as they become more embedded in supply networks. However, in comparison to AC systems, the understanding of insulation and plant reliability under HVDC is still in its infancy. At the same time, the working environment for DC plant is not well documented and, in reality, DC systems have AC ripple, impulses and voltage variation just as in any other system, and these time-varying waveforms are likely to control plant ageing and reliability.
This project comprises internationally leading researchers from The University of Manchester, The University of Strathclyde and Imperial College. They bring complementary expertise to form a unique team to address the problem. Prof Tim Green (Imperial) is an expert in the use of power electronics to enhance the controllability and flexibility of electricity networks; Prof Simon Rowland (Manchester) is an authority on ageing of high voltage insulation materials; and Prof Brian Steward (Strathclyde) has unique experience in condition monitoring and insulation diagnostics for high voltage systems. The project is designed to embed the work into the global community and in particular is linked to researchers in China where the largest systems are being developed.

This project will firstly identify the voltage profiles experienced by plant insulation in a real HVDC network or link, because in real systems the voltage on the network is not a constant, fixed value. The power converters that feed a DC network create intrinsic "noise" in the form of high frequency elements as part of their normal operation, and also create voltage disturbances in their responses to fault conditions and emergency overloads. Characterising these is the first step in the overall study of how DC power quality impacts the lifetime of HV insulation. The team will then, through laboratory exploration, develop life models for polymeric insulation subject to known levels of DC power quality. The focus will be on AC ripple over a wide frequency range. In addition, the influence of fast transient signals of varying levels and durations will be considered, as identified above. The third experimental theme is to develop tools for monitoring transient signals and power quality in a real DC cable setting, and enable subsequent interpretation. Finally, we will develop input for utility policy documents on acceptable DC power quality. We will also provide evidence for optimal insulation design for equipment manufacturers and asset management recommendations for utilities.

Through these means we hope to de-risk the UK's growing dependence on DC networks, and optimise equipment and system design and operation.

The project will ultimately benefit the whole of society through de-risking and optimising the long-term performance of DC networks. Given the global roll out of DC networks across the developing and developed world the potential reach of this project is therefore vast. In particular, connections of renewable energy, and enabling reliability through interconnection of national networks will support development of a low carbon future. The beneficiaries might be characterised as the consumer (individual people and organisations), electricity generators and network operators. It can be argued that enabling low carbon energy benefits everyone.

The mechanisms by which the project activity reaches the beneficiaries are principally through the utilities who connect supply to demand. However, the operation of the network and its associated plant, and the design of the plant are inextricably linked. To understand the restriction on power quality requirements, requires a knowledge of materials and equipment design. Thus the need for a holistic model of a material's response to different voltage and thermal stresses is required. The information generated in this project therefore needs to flow through material manufactures, equipment designers, system designers and system operators. The team on this project will aim to rapidly disseminate the knowledge generated to the wider technical community and described in our Pathways to Impact summary. This includes: an Experts Panel, which is constructed to be representative of the whole design, supply and operation chain; a one-day colloquium related to HVDC systems; a DEIS Technical Committee to be initiated to enable relevant IEEE Standards associated with HVDC insulation design, HVDC insulation system measurements and interpretation; and CIGRE Study Committees, for example B4 study committee on HVDC and Power Electronics.

Key benefits to utilities include understanding how tight power quality restrictions need to be on HVDC systems - any relaxation of restriction can reduce cost. Developing models of ageing also allows the reduction in reliability or asset life as a function of power quality to be identified. This provides tools to optimise asset management, through development of condition monitoring methods and prognostics to improve reliability and reduce cost. Key benefits for power electronic systems manufacturers include clarification of design requirements of converters; identifying improved designs; and identification of the opportunity of short-term ratings exploiting redundant sub-modules to raise voltage - balanced against the impact this has on cable lifetime. Material designers and equipment designers (such as cables, joints and terminations) will have a landscape allowing clear objectives for improvement of component reliability performance in terms of voltage exposure, enabling more reliable and efficient networks.

The key opportunity of this project is in bringing so many diverse views and needs together, thereby generating policies which optimise the whole network technically and financially.