IET Nanodielectrics最新文献

This contribution presents a two-step hybrid diagnostic approach, combining k-means clustering for subset formation, followed by subset analysis conducted by human experts. As the feature input vector has a significant influence on the performance of unsupervised machine learning algorithms, seven feature input vectors derived from traditional methods, including Duval pentagon method, Rogers ratio method, three ratios technique, Denkyoken method, ensemble gas characteristics method, Duval triangle method, and Gouda triangle method were explored for the subset formation stage. The seven proposed individual methods, corresponding to the seven feature input vectors, were implemented using a dataset of 595 DGA samples and tested on an additional 254 DGA samples. Furthermore, a combined technique based on a support vector machine was introduced, utilising the diagnostic results of the individual methods as input features. From training and testing, with diagnostic outcomes of 91.09% and 90.94%, the combined technique demonstrated the highest overall diagnostic accuracies. Using the IEC TC10 database, the diagnosis accuracies of the proposed diagnostic methods were compared to existing methods of literature. From the results obtained, the combined technique outperformed the proposed individual methods and existing methods used for comparison.

Silicone rubber (SiR) is commonly used in reinforced insulation parts for high-voltage direct current (HVDC) cable accessories due to its excellent insulation, elasticity, and high-temperature resistance. HVDC cable accessories always suffer from the local electric field concentration due to the electrical conductivity mismatch between reinforced insulation and main insulation, which can ultimately lead to electric breakdown. The non-linear conductive composites based on SiR have the ability to adaptively adjust the distribution of the electric field in cable accessories. This is expected to solve the problem of localised electric field concentration. The zinc oxide (ZnO) and glycidyl methacrylate (GMA) are used as fillers and grafted modifier respectively to improve the non-linear electrical conductivity of ZnO/SiR-GMA composites. The results indicate that grafting GMA can increase electrical conductivity of SiR, while doping ZnO filler enables SiR to have non-linear conductivity characteristics. The combination of doping and grafting modification of the composites achieves excellent non-linear conductive properties at lower ZnO filler content. Additionally, the mechanical properties of the modified composites are enhanced. The simulation results indicate that ZnO/SiR-GMA is the most effective material for homogenising the electric field when used as reinforced insulation for cable intermediate joints.

A key factor in ensuring the efficient and safe operation of power transformers is the early and accurate diagnosis of incipient faults. Among the tools available to achieve this goal, dissolved gas analysis (DGA) is widely used by power transformers' maintenance professionals. It is a preventive maintenance tool, used for condition monitoring, fault diagnosis and unplanned outage prevention. With the development of artificial intelligence (AI), many intelligent-based methods using AI tools have been proposed in the literature for DGA data interpretation. Although these methods achieve high diagnostic accuracies and improve DGA efficiency, they are generally complicated and the research documented in these publications is difficult to replicate. Traditional DGA-based methods are simple, easy to understand and implement, and widely used by power transformers' maintenance professionals. Many methods proposed in recent years overcome the limitations of the pioneer methods and are increasingly effective. The authors present a detailed and comprehensive literature review of the traditional DGA-based methods for mineral oil-immersed power transformer faults diagnosis. This review also addresses ways to improve the efficiency of the available traditional methods. Some pitfalls that need to be taken into account to improve the efficiency of the DGA-based diagnostic methods are also presented.

Self-healing polyurethane diacrylate (PUDA) composites with outstanding thermal conductivity were prepared with variable boron nitride (BN) and a small number of graphene nanoplatelets (GNPs) (1 wt%). The thermal conductivity of PUDA composites containing 1 wt% GNPs and 29 wt% BN reached 1.73 W/(m·K), showing 27.2% increment compared to PUDA composites only with 30 wt% BN. BN and GNPs mutually promoted the dispersion in the PUDA matrix, and the small number of GNPs was enough to bridge the gaps between BN, so that GNPs and BN exerted a synergistic enhancement effect on the thermal conductivity of PUDA composites with a synergistic efficiency of 1.343. The PUDA composites maintained the tensile strength of 8.21 MPa with good electrical insulation (4.55 × 10⁹ Ω cm). The high degrees of recovery of mechanical strength (>90%) and thermal conductivities (>84%) were also realised by the healing effect of the reversible DA reaction. The PUDA composites with excellent comprehensive properties show broad application potential in the thermal management of electronic devices.

High conductivity loss and low breakdown strength of traditional polypropylene (PP) film in the high-temperature environment are the key factors limiting the application of metallised film capacitors (MFCs). The effect of cyclic olefin copolymers (COCs) with different glass transition temperatures (Tgs) on the dielectric performance of PP/COC composites at high temperature are studied. The results showed that the addition of COC enhanced the intermolecular interaction force, which led to the reduction of defects in the films. In addition, COC with high Tg inhibits the movement of molecular chains and enhances the thermal stability of the film, which limits the transport of carriers at high temperatures. The conductivity and breakdown strength of the modified films (with a Tg of 134°C for COC) at 125°C are 91.6% lower and 45.7% higher, respectively, than that of pure PP. This method shows great potential in improving the dielectric properties of PP at high temperatures.

Film capacitors are playing an increasingly important role in power-related fields, driven by the continuous development of dielectric materials and practical needs. Long-term accumulation has also led to an increasing wealth of data related to film capacitors. Sonification opens up a new way for people to make good use of data from film capacitors. A framework for sonifying film capacitors data based on TwoTone is presented. Based on the analysis and discussion, it is clear that the sonification results can easily represent the monotonic variation pattern of film capacitors data. What's more, the sonification results increase the possibility that people pay attention to the changing trend of film capacitors data when there is no significant difference in the visual perception of the data. In addition to providing a new way of music generation of electrical equipment, the method proposed is expected to contribute to theory reference in typical scenarios, such as factory calibration of film capacitors, monitoring of film capacitor operation status, and presentation of statistical data of film capacitors' dielectric materials, which will help us to better understand the distribution characteristics of polymer films.