Research on Short-Term Low-Voltage Distribution Network Line Loss Prediction Based on Kmeans-LightGBM

Abstract

Due to the lack of data quality in real production environment, the traditional line loss calculation method cannot be applied, thus through the investigation of various information systems’ operation in power supply enterprises, a short-term low-voltage distribution network line loss prediction algorithm based on Kmeans-LightGBM is proposed. Operating data quality evaluation system of low-voltage distribution network was set up based on Hadoop platform, the feature dimensions were expanded by feature engineering, then those with no multicollinearity and high correlation with the line loss were selected, data normalization was again performed, Kmeans clustering algorithm was used to cluster the area and then, LightGBM algorithm was used to predict the classes within the area of line loss. Finally, the line loss of the numerical inverse normalization was found and validated with Beijing Power Grid of a low-voltage distribution network. By comparison, the model’s prediction accuracy is found to be higher than BPNN, FOA-SVR and traditional LightGBM.

Informatization of Accounting System of Electric Power Enterprises Based on Sensor Monitoring and Cloud Computing

“…Establishing and implementing an accounting information management system that combines modern information technology and computer technology. Information mainly refers to accounting information resources, and computer technology mainly focuses on accounting information processing, referred to as information processing processor, which is the core part of the information management system [11,12]. An accounting information system is a human-computer interaction management information system using modern information processing technology.…”

Design of photovoltaic-energy storage-load forecasting and optimal control method for distributed distribution network based on reinforcement learning

“…In winter, the photovoltaic system can receive solar energy in a slightly shorter time than in summer, and the photovoltaic power obtained in winter is also much less than in summer. As can be seen from Figure 3, the voltage fluctuation trend is larger when the method of reference [3] and the method of reference [4] are adopted. After the proposed method is used to control the energy storage power of the distributed distribution network, the voltage of the energy storage power of the distributed distribution network is always stable at 150V, which is equal to the rated voltage of the energy storage power of the distributed distribution network, indicating that the proposed method has higher control accuracy for the energy storage power of the distributed distribution network.…”

“…When the proposed method, reference [3] method and reference [4] method are used to optimize the photovoltaic-energy storage-load of the distribution network, the final optimization result is obtained after 100 iterations, as shown in Table 2. As can be seen from Table 2, the three methods have been iterated for the same number of times in the optimization process, that is, each method has been iterated for 100 times, and the accuracy of reference [3], reference [4] and the methods provided before optimization are 79.5%, 78.5% and 80.4% respectively. The accuracy of the three methods after optimization is 88.5%, 85.2% and 96.8% respectively.…”

“…The output power curve results of photovoltaic power generation equipment in summer and winter are obtained from the original data respectively, as shown in Figure 2. Then, compared with the proposed method, the method in reference [3] and the method in reference [4], the coordinated control accuracy of energy storage power in distributed distribution network is tested, and the results are shown in Figure 3. Finally, the power load curves of three scenes, namely, college area, residential area and factory area, are counted, and the results obtained in different time periods are shown in Figure 3.…”

“…Tang et al [3] proposed a short-term line loss prediction method for low-voltage distribution network based on Kmeans LightGBM, expanded the feature dimension through Feature engineering, selected data without multicollinearity and highly related to line loss, normalized the data again, and predicted the categories within the line loss area using LightGBM algorithm. Through comparison, it was found that the prediction accuracy of this model is higher than other methods.…”