Temporal information is crucial for detecting occluded instances. Existing temporal representations have progressed from BEV or PV (perspective view) features to more compact query features. Compared to these aforementioned features, predictions offer the highest level of abstraction, providing explicit information. In the context of online vectorized HD map construction, this unique characteristic of predictions is potentially advantageous for long-term temporal modeling and the integration of map priors. This paper introduces PrevPredMap, a pioneering temporal modeling framework that leverages previous predictions for constructing online vectorized HD maps. We have meticulously crafted two essential modules for PrevPredMap: the previous-predictions-based query generator and the dynamic-position-query decoder. Specifically, the previous-predictions-based query generator is designed to separately encode different types of information from previous predictions, which are then effectively utilized by the dynamic-position-query decoder to generate current predictions. Furthermore, we have developed a dual-mode strategy to ensure PrevPredMap's robust performance across both single-frame and temporal modes. Extensive experiments demonstrate that PrevPredMap achieves state-of-the-art performance on the nuScenes and Argoverse2 datasets. Code is available at https://github.com/pnnnnnnn/PrevPredMap.

The insulation of the medium-voltage direct current (MVdc) cable is prone to deterioration due to external and internal factors, which may lead to faults. Aiming to accommodate two different cable operation modes, in this article, the feasibility of decoupling transient measurements is first verified. An algorithm is presented for calculating the phase-mode transformation matrices used for signal decoupling. Then, the states of decoupled modal signals are analyzed to select the featured ones that reflect fault occurrence. A fault detection method is proposed based on the variation characteristics of featured modal signals. Finally, the experimental model of the MVdc cables is created using real-time digital simulation. The experimental results show that the method is independent of the fault condition, cable length, and operating mode. The average fault detection accuracy exceeds 99% with noises equal to or larger than 20 dB. The method applies to a wide range of fault resistances (1–3000 Ω).

Faults in underground cables are hard to be detected directly from external features such as sound and light due to the harsh operating environment. Moreover, the rising penetration of distributed generations further complicates the characteristics of cable faults. Facing the challenges, this article first aims to reveal cable fault features via analyzing the generation mechanism of the cable grounding wire current (GWC). The impact of photovoltaic on GWC is then discussed, and the essential differences between faulty and healthy cables are also studied. Finally, considering both the structure of distribution network and an optimal deployment of measurement points, a novel fault cable segment identification method based on the amplitude and polarity of the starting terminal GWC steady-state variation is proposed. Simulations based on a real low-resistance grounded distribution network are performed in diverse fault scenarios and show a near 100% accuracy for all cases. In addition, the proposed method demonstrated a time complexity ofO(896), showing a significant reduction in calculation time compared to common methods. These findings suggest that the proposed method is highly reliable and efficient for identifying faulty cable segments in active distribution networks.

Power distribution cable feeders are prone to faults due to the poor working environment and internal defects. Considering the structure and electrical parameters of the typical distribution three-core cable, a fault sensing method is proposed. The fault sensing in this article includes three aspects: fault cable detection, fault phase identification, and fault resistance estimation. The grounding line currents of cable feeders are first analyzed in different neutral grounding modes. Based on the amplitudes and directions of the grounding line currents, a fault feeder detection criterion is then presented. Finally, based on the equivalent circuit of the fault cable, an algorithm for identifying the fault phase is developed by estimating the fault resistance. The experimental model of the distribution cable feeders is created by real time digital simulation (RTDS). Various fault experiments are carried out. The results show that the neutral grounding modes and fault conditions have little effect on the proposed method.

With significant technological advancements in power grids, three-core cables are widely employed for their safety and other benefits. In general, three-core cables in medium-voltage distribution systems are laid underground during operation. As a result, the harsh working environment and complex fault conditions can bring massive challenges to fault location issues. To address the problem, the distinctive cable features are extensively considered in this article, including relatively larger capacitance, cable shield effects, and shield bonding methods. Initially, an equivalent cable model is developed to analyze the relationship between shield-grounding wire currents and the fault point. A single-phase grounding fault location method is then formulated based on the ratio of shield-grounding wire currents at both terminals. In power systems computer aided design/electro magnetic transient in DC system (PSCAD/EMTDC), a range of fault scenarios are simulated to verify the proposed method. In comparison to conventional similar approaches, the proposed method exhibits higher accuracy and robustness to fault resistance and diverse fault conditions. Finally, the effectiveness of the proposed method is validated on a real 10 kV distribution network experimental platform under several fault scenarios.

Faults may be generated in power networks as a result of bad weather, human activities, or some other factors. The complex structure of power network adds to the difficulty of fault location (FL). This paper presents a novel FL scheme by utilizing the information of the time differences of arrival (TDOAs) of modal traveling waves (MTWs) asynchronously sampled in the network. First, the fault area is determined by searching the minimum TDOA of MTWs. Then, by applying the fictitious fault point method in the fault area, the minimum accumulated difference is used to detect the fault line. Finally, the accurate FL is estimated by solving the objective function of the absolute distance between the FL and multiple FL candidates. Various simulations are carried out in the IEEE 30-bus system. The calculation results demonstrate that the method is not affected by fault impedance, inception angle, location, and noises to a certain extent.