High Voltage最新文献

When designing high-voltage elements, knowing if the corona effect will be present during their life cycle is relevant. Therefore, designers consider several prediction criteria based on physical features related to breakdown discharge principles to predict the corona effect. The introduced practical set-up consists of a concave cone electrode with a hemispheric tip above a plate to evaluate selected corona prediction criteria. The hemisphere has a fixed diameter of 7 mm, and the electrode separation ranges from 2.5 to 39 cm. Information about the corona mode inception under different voltage sources was extracted using an intensified charge-coupled device camera and a partial discharge metre. The prediction criteria were connected to a specific corona mode depending on the main discharge structure behind its development. The average deviation between these criteria and experimental results was around eight percent. Underlying assumptions in criteria are also discussed in light of the experimental results.

With the increasing penetration of DC systems into the high voltage and medium voltage power industry, DC voltage is becoming common for distribution/transmission and to supply different typologies of loads. The allowable extent of AC ripple superimposed to DC, and its effect on insulation ageing, is a long-term discussed topic. The most harmful phenomenon causing extrinsic ageing acceleration and insulation system premature failure is partial discharges (PD); thus, the risk of incepting PD due to AC ripple could become a primary issue for electrical asset equipment reliability. In this work, the impact of AC sinusoidal ripple on insulation system life and reliability is dealt with, considering both intrinsic and extrinsic ageing but focusing on the latter, that is, the PD aspect. Experiments are performed to assess how the jump voltage (due to AC ripple) and the DC component impact on PD activity in terms of amplitude and repetition rate. For the first time, the correlation between the magnitude of jump voltage associated with ripple and PD inception is established, shedding a light on the allowable ripple extent which does not impact significantly on ageing and premature insulation failure. This approach can provide straightforward tools for design specification and ageing inference of insulation systems.

Effective monitoring and early warning of overhead power lines are critical for ensuring power supply stability and personnel safety. However, existing methods that rely solely on single vision or light detection and ranging (LiDAR) have limitations such as numerous invalid alarms, hardware equipment constraints, and lack of real-time monitoring capabilities. This article proposes an intelligent monitoring method for safety distance of the powerline corridor based on heterogeneous sensor information. Firstly, an optimised single-stage detector is designed to detect safety hazard objects. In the detector, the backbone network and the sample matching strategy are improved; in addition, the learning strategy is adjusted. Next, the pose transformation relationship is obtained through the visual matching of prior LiDAR information and images. The back projection transformation of 2D–3D is achieved according to the relationship. Finally, a monocular camera-based end-to-end distance measurement scheme is proposed by combining 2D object information with coordinate transformation relationships. The scheme is applied to the distance measurements from hazard objects to powerlines. The experimental results show that the optimisation method improves the detection accuracy and reduces the computational complexity of the model. Also, case experiments with continuous frame data verify the effectiveness of the safety distance monitoring scheme.