It is crucial to reduce the substantial line loss in distribution networks in power utility.It is the premise to develop a model to describe the loss characteristics that constitute the targeted and comprehensive loss mitigation decision.In this study,a general line loss probability density distribution model is proposed based on state identification in the power distribution area.Firstly,a line loss state identification method based on the C-vine Copula model is proposed to overcome the low accuracy of the existing models caused by ignoring the different characteristics of the different states when calculating the loss.Secondly,a general line loss rate probability density distribution model,including the optimal selection of the probability density function and parameter estimation of the proposed model,is presented to reflect different loss distribution characteristics under different line loss states of the power distribution area.Finally,the accuracy and validity of the proposed method is verified using the measured data across two years of a power distribution area in southwest China.
With the continuous development of the domestic photovoltaic industry and the increasing installed capacity of photovoltaics, fires occur frequently in photovoltaic power stations. The cause of fires is often caused by DC arc faults. Therefore, a reliable DC arc fault detection method for photovoltaic arrays is of great significance for ensuring the safe and reliable operation of photovoltaic power generation systems. The photovoltaic array is greatly affected by environmental factors, the output data is unstable, and the amplitude change of the fault arc data is also uncertain, which leads to the misjudgment of the DC arc detection method. Aiming at the DC series arc fault of photovoltaic array, this paper proposes a detection method of DC series arc fault of photovoltaic array based on random forest. Based on the Simlink platform, a simulation model of photovoltaic series arc fault is built. The simulation model is used to obtain the series arc fault data, and the difference between the normal string and the series arc fault string is compared. The DC series arc fault data is analyzed by statistical method and the feature quantity is extracted as the training set. The feature quantity is used to train the random forest model to establish the photovoltaic series fault diagnosis model. The string voltage data of each group collected by the inverter before and after the series arc fault of a photovoltaic project in a new energy enterprise in Xiamen are used for diagnosis. The inverter string data of the actual series arc fault verify that the method can effectively identify the DC series arc fault, and is less affected by environmental factors. The stability of the detection method is good.
In order to realize the optimal power flow under multiple time scales and multiple operation modes, the prediction results and power grid status are included in the optimization model simultaneously, and a power flow optimization strategy combining day-ahead and day-ahead optimization is proposed. According to the actual operation expectations, the objective function variable weight is adopted to adjust; The intra-day optimization combines the model predictive control system with super-short-term forecast, and plays the buffering role of energy storage to realize the tracking of scheduled distributed power output. In addition, give full play to the advantages of simplicity and high feasibility of knee search, and realize the global optimization of multi-objective optimization solution. The results show the practicability of this strategy.
The power grid load shows a trend of diversified development. With the continuous increase of the power grid construction scale, it is very important to optimize the power supply of the distribution network and improve the utilization of assets and equipment by using the peak valley complementary characteristics of multiple loads.In this paper, the peak-valley fluctuation characteristics of load are classified and analyzed. Cluster analysis is used to study the load classification with stable peak-valley characteristics in the complex load composition. Based on the peak-peak superposition and peak-valley complementary characteristics of the classified load, and considering the load space dimension, the multi-load optimization combination method for the whole period and time segment is proposed, and verified by simulation.
针对电力系统用户侧用电行为复杂、分析困难的问题,提出了一种基于多维增强特征提取(multi-dimensional enhanced feature extraction,MEFE)和群组归一化卷积神经网络(group normalization convolutional neural network,GN-CNN)的低压台区用户用电行为分析方法.首先,利用MEFE构建了基于用电用户的平均日用电量曲线、周内相邻天用电量皮尔逊相关系数和增强夜间用电量的增强型用户用电行为特征;其次,建立了一维GN-CNN的用户用电行为识别模型,实现了用户用电行为准确识别;最后,以江苏泰州某台区96点用电信息采集数据为输入,开展模型优化及方法对比试验,用电行为识别准确率为96.4%,可以为低压台区的电力调度、负荷预测、安全性评估等提供数据支撑.
