As an important part of the power system, the accuracy of the current calculation of the distribution network is of great significance for the operation stability analysis and optimization design of the power system. Traditional current calculation methods include Newton-Raphson method, PQ decomposition method, etc. Due to the difficulty in obtaining line parameters and complex source-load characteristics of distribution networks, it is difficult to meet the demand for accurate current calculation by applying the traditional current calculation methods directly. Aiming at the problems that the traditional trend calculation methods are not applicable to the radial structure, open-loop operation, and the large number of nodes and branches of distribution networks, this article proposes a decision tree algorithm based distribution network trend calculation method to realize the accurate trend analysis of distribution networks from a data-driven perspective. First, a dataset of distribution network trend data is obtained; then, based on the collected data, a decision tree model is trained to realize accurate distribution network trend analysis without the need of distribution network line parameters; finally, the trained model is applied to the IEEE33 node distribution network system, and the validity and accuracy of the model are verified by predicting the trend of this system.

Tianxiang Huang. Decision Tree Algorithm Based Tidal Flow Calculation for Distribution Networks. Academic Journal of Computing & Information Science (2024), Vol. 7, Issue 12: 9-15. https://doi.org/10.25236/AJCIS.2024.071202.

[1] Wu Xiaohui, Wang Zhongwen, Jiang Jiahe, et al. Research on the impact of distributed power supply on distribution network trend calculation [J]. Science and Technology Information, 2024, 22(13): 84-86.

[2] Zhang Wei, Wang Bin, Wang Xianli, et al. Non-iterative distribution network trend calculation based on BPNN [J/OL]. China Electric Power, 1-8[2024-08-14].

[3] Li Penghua, Song Zhuoran, Wu Wenchuan. Data-driven optimal tidal current model for distribution networks under partially observable conditions [J]. China Electric Power, 2023, 56(12): 51-57.

[4] Zhao Fei, Fan Xuejun, Li Yalou, et al. Iterative implicit linearization based mesh distribution network current calculation and its optimal current model [J]. Chinese Journal of Electrical Engineering, 2024, 44(06): 2197-2208.

[5] Long Yun, Lu Youfei, Zhao Hongwei, et al. A novel data-driven power system trend analysis method [J]. Intelligent Power, 2023, 51(09): 22-30.

[6] Zhao Enlai, Hu Yu, Guo Jun, et al. Linearized tidal current calculation of distribution networks considering load-voltage static characteristics and PV nodes [J]. Electrical Drives, 2024, 54(05): 56-63.

[7] Liu Huijia, Feng Ling, Ai Can. BPNN distribution network trend calculation based on GA-ADAM optimization [J]. Electronic Measurement Technology, 2023, 46(24): 84-92.

[8] Huang Rong. Joint iterative strategy for distribution network trend calculation based on Newton-Raphson method and P-Q decomposition [J]. Automation Application, 2023, 64(22): 55-57.

[9] Liu YZ. Autonomous Task Decision Making Method for Robot Ecosystem Based on Decision Tree Algorithm [C]// Tianjin Electronic Industry Association. Proceedings of the 2024 Annual Conference of Tianjin Electronics Industry Association. Tianjin Seven One Two Communication and Broadcasting Co. Ltd, 2024: 8.  

[10] Sheng Lijun, Chen Shiqi. Application of decision tree algorithm in the redundancy of ship autonomous cruise data[J]. Ship Science and Technology, 2024, 46(12): 157-161.