This study aims to effectively identify abnormal or non-compliant electricity consumption behaviors in electric vehicle charging stations, thereby enhancing the efficiency and accuracy of electricity management for these stations. Initially, the study analyzes the electricity consumption behavior characteristics of low-voltage charging stations, determining the differences in load characteristic curves between normal and abnormal electricity consumption states. Based on this, a clustering analysis algorithm is employed to extract load curve characteristics from operational charging stations and compare them with standard load curves to assess the presence of abnormal electricity consumption behaviors. Additionally, considering potential misjudgments arising from the “fast charging” and “slow charging” phases during the charging process, the concept of sliding difference linear fitting is introduced. This involves calculating the slope between each pair of 96-point load data points and using the number of slope changes to assist in the judgment of clustering analysis results. Through the aforementioned methods, users exhibiting abnormal electricity consumption behaviors have been successfully identified, providing technical support for the management of electricity consumption in charging stations.

Based on the characteristics of the fully domesticated chip relay protection device’s software and hardware platforms, and addressing the key and difficult points in the testing process, this paper proposes a multi-CPU internal communication testing method for fully domesticated chip relay protection devices. Utilizing a dedicated internal detection module, the method simulates and monitors various normal and abnormal data without altering the original wiring layout, to verify the stability of internal communications between CPU boards. Additionally, this paper presents key technologies of internal communication reliability testing and platform application testing, details test plans for critical applications such as platform self-check, intelligent IO self-check, abnormal reset functions, and memory overflow risks, and summarizes typical on-site problem verification and key testing items. Finally, by comparing and analyzing the differences in key technical indicators between the “nine unified” devices and fully domesticated chip relay protection devices, the study offers insights for the product development of fully domesticated chip relay protection devices.

Both sides of the high and low voltages of the auxiliary transformer, which is in the standby state for long time in the nuclear power plant, is still balanced after the single phase open on the high voltage side, posing a threat to the safe operation of nuclear power units. In order to detect the open-phase timely, the optical current transformer is selected to measure the current of the high voltage side accurately according to other electrical characteristics of the auxiliary transformer after phase open, and the corresponding identifying logic is developed according to different working conditions to determine the open-phase monitoring scheme. The setting method, resolution and measuring range of the optical current transformer are studied to understand its technical characteristics. The open-phase monitoring system is introduced from the aspects of hardware, software and logic. At last, maintenance strategy of regular inspection for optical current transformer is given.

For a 220kV substation with 35kV capacitive voltage transformer secondary voltage abnormal drop fault, this paper introduces the structure of capacitive voltage transformer and its working principle. Then, combined with fault phenomenon, diagnostic test data, equipment disintegration inspection, the cause of the fault as ferromagnetic resonance overvoltage causing insulation breakdown of the differential and angular coils is analyzed and determined. Finally, some suggestions are put forward for the infrared temperature measurement and secondary voltage monitoring in the daily operation and maintenance inspection to improve the reliability of capacitive voltage transformer operation.

Aiming at the problems of dense distribution of elements, similar elements and large number of small-sized elements in power dispatching control system diagrams, which lead to poor recognition effect, an improved you only look once v7 (YOLOv7) element recognition method for the power dispatching control system diagrams is proposed. Firstly, the lightweight dilated reparam block net with cross stage partial and efficient layer aggregation network (DRBNCSPELAN) module is used to replace the efficient layer aggregation network (ELAN) module in the backbone network to capture spatial patterns of different scales simultaneously. Secondly, an information-guided fusion module is proposed to replace the Concat in the neck network, and the sequeeze-and-excitation (SE) attention mechanism is integrated to enhance the global information interaction ability. Then, the minimum point distance intersection over union (MPDIoU) loss function is introduced to improve the recognition effect of the element bounding box. Finally, experimental validation is performed via the power dispatch control system diagram dataset. The results show that compared with the baseline model, the precision, recall and mean average precision of the proposed method are improved by 5.1 percentage points, 3.1 per-centage points and 3.5 percentage points respectively, which is helpful to achieve accurate recognition of elements in the power dispatching control system diagrams.

Aiming at the problem of the lack of historical data on arc faults in most photovoltaic power stations, this paper proposes a photovoltaic system series arc fault diagnosis method based on ultrasonic sensors and isolation forest after collecting arc ultrasonic signals and analyzing their characteristics. Firstly, arc ultrasonic signals are collected and their characteristics and advantages are analyzed. Secondly, the S-transform is used to convert the transient voltage signal of the ultrasonic wave during the occurrence of series arc faults to the time-frequency domain. Then, the Teager energy operator is used to amplify the spectral differences. Subsequently, the time-frequency entropy is used to extract the time-frequency domain features of arc faults. Finally, arc faults are diagnosed based on dynamic thresholds and isolation forest without the need for historical data. Experimental results show that the proposed method can accurately identify series arc faults, with a diagnosis accuracy rate of 97.25%, and has strong anti-interference ability.

Establishing the correlation between macrolevel indexes and power supply reliability is essential to analyzing the power supply reliability of distribution network on a large scale (provincial and municipal). However, the existing methods rely on the detailed distribution network topology, which is difficult to meet the requirements of regional power supply reliability analysis. This paper firstly establishes a macrolevel index system of power supply reliability for planning, construction, operation and management of distribution networks. Then a calculation method of power supply reliability is proposed by taking into account the automated isolation of distribution networks and pre-scheduled outages. On this foundation, the calculation method of power supply reliability based on the macrolevel index system is proposed by dividing the equivalence assessment sub-models and calculating the parameters of the sub-models. Finally, a case study is carried out to verify the proposed method. The results show that the relative deviation between the calculation results of the proposed method and the actual data is within 11%. The method proposed in this paper can analyze power supply reliability in the region by using only the macrolevel indexes, which provides a certain degree of accuracy and operability for the planning of distribution networks, reliability management and reliability target setting.

Wind power prediction plays a crucial role in ensuring the reliable integration of wind energy into the grid. This study proposes a novel hybrid model combining random forest (RF) and convolutional neural network (CNN), referred to as the RF-CNN model, specifically designed for short-term wind power prediction. The model integrates the advantages of RF integration technology, random selection of attributes, and CNN capturing the spatiotemporal characteristics of wind power, to enhance prediction accuracy and robustness. Firstly, by analyzing the analog equivalence between decision trees and CNNs, the theoretical basis for combining RF and CNN is established. Next, an evaluation system for wind power prediction models that includes root mean square error (RMSE), determination coefficient, and Spearman correlation coefficient is introduced. Finally, validatinos are conducted using three open-source wind power datasets from European wind farms. The results demonstrate that, compared to other five models, the RF-CNN model outperforms in all three datasets, thus confirming the model’s effectiveness and accuracy for wind power prediction.

The discharge rod is crucial for the safety of working personnel in substations. However, traditional discharge rods have long discharge times and low work efficiency. A rapid discharge rod is designed to achieve rapid discharge of residual voltage in equipment while avoiding damage to the equipment caused by excessive discharge current. On the basis of the traditional RC discharge circuit, inductive elements are added to form RLC and RL discharge circuits. By selecting RLC circuit or RL circuit through metal-oxide-semiconductor field-effect transistor (MOSFET), rapid discharge of equipments can be achieved, thereby improving the work efficiency of operators.

The failure of photovoltaic power generation equipment and various factors such as external environment lead to a large number of abnormal data during the power generation process. In order to improve the accuracy and efficiency of data processing, this paper proposes a distributed photovoltaic abnormal data identification method based on improved K-means algorithm and weighted dynamic time warping (WDTW). Firstly, the distributed photovoltaic power generation data is analyzed, and the abnormal data is preliminary eliminated by means of the simultaneous power mean method, and a photovoltaic data similarity day partitioning method based on improved K-means algorithm is proposed by normalizing the light intensity data. Secondly, considering the variability and complexity of photovoltaic data in the time dimension, a data similarity analysis method based on WDTW is proposed by introducing the best time period and threshold factor for identifying abnormal data. The similarity is used to calculate the contour coefficient, and the residual abnormal photovoltaic power generation data is culled twice. The simulation results show that the proposed method has significant advantages in identifying distributed photovoltaic abnormal data. Compared with the existing quartile method, 3-sigma method, and feature clustering method, the identification accuracy has been improved by 6.92%, 9.00%, and 8.12% respectively, while the computational complexity is reduced.