国网湖南省电力有限公司

The number of disassembly smart electricity meters caused by the reincarnation,replacement,failure,and demolition of electricity meters has increased dramatically,and it is of great significance to analyze the faults of disassembly smart electricity meters to reuse and improve the reliability of disassembly smart electricity meter.Based on the fault data of disassembly smart electricity meter in a province,this paper analyzes and summarizes the internal and external influencing factors of disassembly smart electricity meter faults.The correlation analysis is used to carry out the correlation analysis of fault historical data.Combined with meteorological,historical calibra-tion and re-calibration data,it is analyzed and clear that the non-compliance of the wire process and the non-com-pliance of the wiring material are the main reasons for the burnout meters.The strategy of improving the reliability of smart electricity meters is proposed to improve the quality of on-site assembly and copper-aluminum transition joints,and the measured data are compared and analyzed.The results show that after applying the proposed strate-gy,the burnout rate of smart electricity meters decreases from 82.73%to 34.55%,and the burnout rate of live wire terminals decreases from 81.44%to 35.01%,which not only directly reduces the burnout rate of low-voltage stations in the province,but also effectively improves the installation and operation and maintenance standardization of metering assets.

Against the background of the goal of “carbon peak and carbon neutrality”， it will be inevitable to form multi-agent cooperative alliance between pumped storage stations， wind power stations， and photovoltaic power stations driven by interests. However， the form of multi-agent cooperative alliance is not static and may break down during the evolution process. Evolutionary games are commonly used to study the cooperative operation of entities in the electricity market， so it can be used to study the long-term stability of cooperative alliances. However， as a non completely rational game method， it ignores the complexity and randomness of the game process. This article constructed a benefit coordination model based on evolutionary game theory， optimized the output of each agents， and solves the evolutionary game benefit matrix. Then， this article constructed tripartite determine evolutionary game model and stochastic evolutionary game model of wind power stations， solar power stations and pumped storage stations， revealing the mechanism of the formation and rupture of multi-agent cooperative alliance. Based on actual data from a certain province， this article concludes that it is feasible for pumped storage stations to participate as entities in the cooperative alliance in the future， and the cooperative alliance can be divided into initial stage， development stage， and rupture stage according to the proportion of new energy installed capacity； Under the influence of random interference factors， there is a possibility for solar power stations to withdraw from the alliance in advance； The stable cooperative alliance has shown excellent performance in ensuring the consumption of new energy. The above conclusions provide theoretical support for the participation of pumped storage power stations in profit oriented cooperative alliances， and provide direction for power grid companies to maintain the stability of cooperative alliances.

Accurately predicting the available capacity of lithium-ion batteries is critical to ensuring the safe operation of energy storage systems. Therefore, this paper proposes a method for predicting the capacity of lithium-ion batteries in energy storage systems based on transfer entropy and JS-BP neural networks. Based on an analysis of the information entropy of relevant parameters of the energy storage batteries, the health factors that have a significant impact on the available capacity of batteries are selected, and a prediction model for the available capacity of batteries is established by combining the selected healthy factors with the JS-BP neural network. Finally, a comprehensive analysis is carried out based on the aging datasets from NASA and the battery aging experimental platform, and the results show that the proposed method has a high prediction accuracy of battery capacity, and the error indicators MAE and RMSE are at a low level, which verifies the accuracy of the model.

[Objective] To address the problems of weak turn-to-turn short-circuit faults in dry-type air-core series reactors， which are difficult to recognize， and the lack of an early warning mechanism in traditional methods， this study proposes a multi-dimensional feature and intelligent algorithm fusion of an early fault diagnosis method. This method can overcome the lack of sensitivity of a single fault feature as it is easily interfered with by the noise of the fault leakage judgment. [Methods] First， the unbalance degree， power factor， zero sequence voltage， and characteristic impedance of the shunt capacitor bank are extracted as fault feature quantities， and their respective evolution laws after the fault are analyzed. Second， principal component analysis （PCA） is used to reduce the dimension and denoise the original data to eliminate interfering information. Subsequently， the denoised features with high saturation are input into the k-nearest neighbors （KNN） algorithm to construct a fault identification and classification model. Based on Maxwell， a field-circuit coupling model is established to generate single-turn， slight， and multi-turn short-circuit datasets； noise-free and 5% noise conditions are considered to verify the robustness of the algorithm. [Results] Simulation results show that the proposed method can achieve 100% recognition accuracy for minor turn-to-turn short circuits under both no noise and 5% noise， and manually adjusting the action threshold is not required. [Conclusions] This study realized high-precision early identification of weak faults through the three-stage architecture of “feature extraction-data noise reduction-intelligent classification.” The innovations include four-dimensional feature synergy to improve fault sensitivity， a PCA-KNN joint anti-noise mechanism； and an adaptive non-threshold discrimination system. The results provide a new idea for power-equipment condition monitoring， and the generalization ability of the model can be optimized by incorporating field data.

Pollution source enterprises are numerous and widespread. The production and pollution treatment processes of each enterprise vary， a lack of effective and uniform regulatory indicators and early warning systems are concerning. This creates problems， such as difficult supervision， poor real-time performance， and a large workload. This study proposes a method for identifying the environmental anomalies of enterprises based on electricity data mining. First， K-means clustering is used to identify the operating status of the equipment， and a model of the enterprise production line is constructed based on dynamic time-warping distance. Next， continuous and intermittent production lines are classified based on historical data statistics. Furthermore， the Fourier transform is used to identify the production cycle of the production line to establish a model of the environmental conditions suitable for the enterprise. Subsequently， the environmental condition identification method is proposed to identify the environmental conditions for continuous and intermittent production lines. Finally， the proposed method is validated using the monitoring data of actual pollution source enterprises. The electric power intelligent environmental protection platform developed based on the proposed method has been implemented in certain provinces， achieving suitable results. This platform enables the environmental protection department to grasp the situation of enterprise environmental protection， providing both technical means and data support.