技术领域Technical Field
本发明涉及配电网故障定位技术领域，特别是基于融合模型和云边端协同的配电网故障定位方法及系统。The present invention relates to the technical field of distribution network fault location, and in particular to a distribution network fault location method and system based on a fusion model and cloud-edge-end collaboration.
背景技术Background technique
近些年来，新能源发电发展迅速，网中光伏发电和风力发电的比例越来越高，同时配电网的用电负荷增长迅速，对配电网供电的可靠性提出了更高的要求，因此，如何利用人工智能技术和边缘计算提高配电网故障定位的准确性，从而降低配电网的停电时间，对配电网来说是极其重要的。In recent years, renewable energy power generation has developed rapidly, and the proportion of photovoltaic power generation and wind power generation in the grid has become increasingly higher. At the same time, the power load of the distribution network has increased rapidly, which has put higher requirements on the reliability of the distribution network power supply. Therefore, how to use artificial intelligence technology and edge computing to improve the accuracy of distribution network fault location, thereby reducing the power outage time of the distribution network, is extremely important for the distribution network.
配电网的故障定位主要是搜寻配电网线路中发生故障的支路，如果配电网发生故障，首先需要找到发生故障的支路，然后将发生故障的支路进行隔离，对剩余的支路恢复供电。随着西电东送和电网特高压线路的建设，电网中远距离输电线路越来越多，这对配电网故障定位的准确性提出了更高的要求，尤其是位于远离市区的高山丛林之中的配电网线路，提高配电网故障定位准确性，可以降低故障抢修时间，提高配电网供电的可靠性。The fault location of the distribution network mainly searches for the faulty branch in the distribution network. If the distribution network fails, the faulty branch must be found first, then isolated, and the remaining branches restored. With the construction of West-East Power Transmission and UHV power lines, there are more and more long-distance transmission lines in the power grid, which puts higher requirements on the accuracy of fault location in the distribution network, especially for distribution network lines located in the mountains and jungles far away from the urban area. Improving the accuracy of fault location in the distribution network can reduce the time for emergency repair and improve the reliability of power supply in the distribution network.
目前现有技术中提出采用云边协同的思想进行配电网故障定位，但该方法需要在设备的边缘端进行故障数据特征提取，将故障数据上传至云端进行处理，在云端建立故障定位的模型，该方法在配电线路较为复杂时，云服务器需要连接的边缘端的数量非常多，会极大的增加服务器计算和存储的压力，会增大网络时延，影响故障定位的时效性。The current existing technology proposes to use the idea of cloud-edge collaboration to locate distribution network faults, but this method requires fault data feature extraction at the edge of the device, uploading the fault data to the cloud for processing, and establishing a fault location model in the cloud. When the distribution lines are more complex, the cloud server needs to connect to a large number of edge terminals, which will greatly increase the computing and storage pressure of the server, increase network latency, and affect the timeliness of fault location.
发明内容Summary of the invention
鉴于现有的基于融合模型和云边端协同的配电网故障定位方法及系统中存在的问题，提出了本发明。In view of the problems existing in the existing distribution network fault location method and system based on fusion model and cloud-edge collaboration, the present invention is proposed.
因此，本发明的目的是提供基于融合模型和云边端协同的配电网故障定位方法及系统，在配电线路较为复杂时，本发明通过端控制器采集配电网数据，利用改进的鲸鱼优化算法进行模型参数寻优，实现基于融合模型和云边端协同的配电网故障定位。Therefore, the purpose of the present invention is to provide a distribution network fault location method and system based on a fusion model and cloud-edge-end collaboration. When the distribution lines are relatively complex, the present invention collects distribution network data through the end controller, and uses the improved whale optimization algorithm to optimize the model parameters to achieve distribution network fault location based on a fusion model and cloud-edge-end collaboration.
为解决上述技术问题，本发明提供如下技术方案：In order to solve the above technical problems, the present invention provides the following technical solutions:
第一方面，本发明实施例提供了基于融合模型和云边端协同的配电网故障定位方法，其包括，利用端控制器采集当前时刻的配电网数据，端控制器将配电网数据按照通讯协议进行封装，将封装完成后的数据上传至边控制器，边控制器判断支路是否发生故障，对发生故障支路的数据进行预处理；边控制器根据设备定值对实时采集的电压对电流信息进行比较，将比较后的结果数据上传至云控制器，云控制器根据输入的线路信息生成配电网线路拓扑图并计算对应设备定值；将对应的定值设备编码下发至边控制器，建立对应的故障定位模型，云控制器接收到边控制器上传的故障数据后进行特征提取，输出配电网故障位置，通过故障定位模型展示故障位置信息。In the first aspect, an embodiment of the present invention provides a distribution network fault location method based on a fusion model and cloud-edge-end collaboration, which includes using an end controller to collect distribution network data at the current moment, the end controller encapsulates the distribution network data according to a communication protocol, and uploads the encapsulated data to an edge controller, the edge controller determines whether a branch has a fault, and pre-processes the data of the faulty branch; the edge controller compares the voltage and current information collected in real time according to the device setting, and uploads the compared result data to the cloud controller, the cloud controller generates a distribution network line topology map according to the input line information and calculates the corresponding device setting; the corresponding setting device code is sent to the edge controller, and a corresponding fault location model is established. After receiving the fault data uploaded by the edge controller, the cloud controller performs feature extraction, outputs the distribution network fault location, and displays the fault location information through the fault location model.
作为本发明所述基于融合模型和云边端协同的配电网故障定位方法的一种优选方案，其中：所述当前时刻的配电网数据包括三相电压、三相电流、零序电压以及零序电流，利用通信协议对采集到的数据进行封装，所述上传包括按照传输格式规定进行传输，所述预处理包括根据封装后的实时线路电流进行判断；As a preferred solution of the distribution network fault location method based on the fusion model and cloud-edge-end collaboration described in the present invention, wherein: the distribution network data at the current moment includes three-phase voltage, three-phase current, zero-sequence voltage and zero-sequence current, the collected data is encapsulated using a communication protocol, the uploading includes transmitting according to the transmission format regulations, and the preprocessing includes judging according to the real-time line current after encapsulation;
若实时线路电流大于设定电流定值或电压小于设定电压定值时，则设备发生故障，此时边控制器开始采集故障时刻前后两个周波的数据；否则设备未发生故障；If the real-time line current is greater than the set current value or the voltage is less than the set voltage value, the equipment fails, and the edge controller starts to collect data of the two cycles before and after the fault; otherwise, the equipment does not fail;
提取发生故障数据，采用插值法对空值进行补齐，计算对应的标幺值，具体步骤如下：Extract the fault data, use interpolation to fill the empty values, and calculate the corresponding per-unit value. The specific steps are as follows:
根据三相电流计算电流幅值，具体计算公式为：The current amplitude is calculated based on the three-phase current. The specific calculation formula is:
其中，Ia、Ib和Ic表示三相电流幅值，Irms表示电流幅值；Wherein, Ia , Ib and Ic represent the three-phase current amplitudes, and Irms represents the current amplitude;
根据三相电流计算电流相位角，具体计算公式为：The current phase angle is calculated based on the three-phase current. The specific calculation formula is:
其中，Iθ表示三相电流相位角，即电流的相位关系；Among them, Iθ represents the three-phase current phase angle, that is, the phase relationship of the current;
根据零序电流计算零序电流幅值，具体计算公式为：The zero-sequence current amplitude is calculated based on the zero-sequence current. The specific calculation formula is:
其中，I0rms表示零序电流的有效值，即零序电流的平均幅值，I0a、I0b和I0c表示三相零序电流幅值；Wherein, I0rms represents the effective value of zero-sequence current, that is, the average amplitude of zero-sequence current, I0a , I0b and I0c represent the amplitudes of three-phase zero-sequence current;
根据零序电流计算零序电流相位角，具体计算公式为：The zero-sequence current phase angle is calculated based on the zero-sequence current. The specific calculation formula is:
其中，I0θ表示零序电流的相位角，即零序电流的相位关系；Among them, I0θ represents the phase angle of the zero-sequence current, that is, the phase relationship of the zero-sequence current;
所述边控制器判断支路是否发生故障包括在边控制器中利用计算得到的三相电流幅值和三相电流相位角进行故障判断；The side controller determines whether a branch circuit is faulty, including using the calculated three-phase current amplitude and three-phase current phase angle to perform fault judgment in the side controller;
若三相电流幅值大于设定的电流阈值或大于电流相位角偏离正常范围时，则表明该支路存在异常电流，判断发生故障。If the three-phase current amplitude is greater than the set current threshold or greater than the current phase angle deviation from the normal range, it indicates that there is an abnormal current in the branch and a fault is determined to have occurred.
作为本发明所述基于融合模型和云边端协同的配电网故障定位方法的一种优选方案，其中：所述比较包括电流幅值比较和电流相位角比较；所述电流幅值比较包括计算各相电流波形的均方根值，根据均方根值计算三相电流的有效值，设定电流阈值进行比较；所述电流相位角比较包括计算电流相位角的整体偏差指标，根据整体偏差指标检查相位角是否偏离正常范围，具体步骤如下：As a preferred solution of the distribution network fault location method based on fusion model and cloud-edge collaboration described in the present invention, wherein: the comparison includes current amplitude comparison and current phase angle comparison; the current amplitude comparison includes calculating the root mean square value of each phase current waveform, calculating the effective value of the three-phase current according to the root mean square value, and setting the current threshold for comparison; the current phase angle comparison includes calculating the overall deviation index of the current phase angle, and checking whether the phase angle deviates from the normal range according to the overall deviation index. The specific steps are as follows:
计算各相电流波形的均方根值公式为：The formula for calculating the RMS value of each phase current waveform is:
计算三相电流的有效值公式为：The formula for calculating the effective value of three-phase current is:
其中，N表示采样点数，Iai、Ibi和Ici表示第i个采样点时的相电流值，表示相电流波形a的均方根值，/>表示相电流波形b的均方根值，/>表示相电流波形c的均方根值，I'rms表示三相电流的有效值，即各支路上提取的采样点值；Where N represents the number of sampling points, Iai , Ibi and Ici represent the phase current values at the i-th sampling point, represents the RMS value of the phase current waveform a, /> represents the RMS value of the phase current waveform b, /> represents the RMS value of the phase current waveform c, and I'rms represents the effective value of the three-phase current, i.e., the sampling point value extracted from each branch;
若I'rms＞设定的电流阈值时，则判断电网发生故障；若I'rms≤设定的电流阈值时，此时判定电流相位角比较，具体步骤如下：IfI'rms > the set current threshold, it is determined that the power grid has a fault; if I'rms ≤ the set current threshold, the current phase angle comparison is determined at this time. The specific steps are as follows:
计算电流相位角的整体偏差指标具体公式为：The specific formula for calculating the overall deviation index of the current phase angle is:
其中，N表示采样点数，δθn表示第n个采样点时的相位角偏差，Δθ表示电流相位角整体偏差指标，设定正常范围为[θmin，θmax]；Wherein, N represents the number of sampling points, δθn represents the phase angle deviation at the nth sampling point, Δθ represents the overall deviation index of the current phase angle, and the normal range is set to [θmin , θmax ];
若Δθ＞θmax-θmin时，则表示相位角偏离正常范围，若Δθ≤θmax-θmin时，则表示相位角未偏离正常范围。If Δθ＞θmax -θmin , it means that the phase angle deviates from the normal range, and if Δθ≤θmax -θmin , it means that the phase angle does not deviate from the normal range.
作为本发明所述基于融合模型和云边端协同的配电网故障定位方法的一种优选方案，其中：所述将对应的定值设备编码下发至边控制器包括将电流相位角整体偏差指标值进行下发至边控制器，所述建立对应的故障定位模型包括将下发至边控制器的电流相位角整体偏差指标值使用分类器进行分类，计算得到分类的目标值，使用带标签的故障数据进行训练，将L1正则化的混合特征进行选择，满足要求的特征向量，在目标值中增加惩罚项，获取目标函数，将目标函数绝对值之和变小，获取稀疏特征向量其中，/>表示混合特征对应的第i个样本数据，使用平方误差作为目标损失函数，目标函数公式定义为：As a preferred solution of the distribution network fault location method based on the fusion model and cloud-edge collaboration described in the present invention, wherein: the sending of the corresponding fixed value device code to the edge controller includes sending the current phase angle overall deviation index value to the edge controller, and the establishment of the corresponding fault location model includes classifying the current phase angle overall deviation index value sent to the edge controller using a classifier, calculating the classification target value, using labeled fault data for training, selectingL1 regularized mixed features, meeting the required feature vectors, adding a penalty term to the target value, obtaining the objective function, reducing the sum of the absolute values of the objective function, and obtaining a sparse feature vector Among them,/> Represents the i-th sample data corresponding to the mixed feature, using the square error as the target loss function, and the target function formula is defined as:
其中，ωp表示第p个特征的回归系数，K表示稀疏度控制参数，K越大，则模型越稀疏；Among them, ωp represents the regression coefficient of the pth feature, K represents the sparsity control parameter, and the larger K is, the sparser the model is;
使用梯度下降法求解上述目标损失函数方程，第K+1次迭代方程为：Use the gradient descent method to solve the above objective loss function equation. The K+1th iteration equation is:
其中，ω＝(ω1,ω2,...,ωn+q)和M是大于0的常数；Wherein, ω＝(ω1 ,ω2 ,...,ωn+q ) and M is a constant greater than 0;
若时，则方程通过如下公式进行计算为：like When , the equation is calculated by the following formula:
其中，和μp分别是ω(k+1)和μ的第p个分量，最后求解L1的归一化结果表示，将ωp的非零向量对应的特征被选择为最终的特征子集向量；in, and μp are the p-th components of ω(k+1) and μ respectively. Finally, the normalized result of solving L1 indicates that the features corresponding to the non-zero vector of ωp are selected as the final feature subset vector;
将特征子集向量转换为矩阵形式，利用电气量进行归一化处理，将矩阵中各值乘积生成对应的灰色图，具体电气量/>归一化处理公式为：Convert the feature subset vector into matrix form and use the electrical quantity Normalize and multiply the values in the matrix to generate the corresponding gray map. The specific electrical quantity/> The normalization formula is:
其中，表示归一化之后的数据，y表示归一化之前的数据，ymax和ymin分别表示最大值和最小值。in, represents the data after normalization, y represents the data before normalization, ymax and ymin represent the maximum and minimum values respectively.
作为本发明所述基于融合模型和云边端协同的配电网故障定位方法的一种优选方案，其中：所述特征提取包括将目标函数输入到配电网故障定位模型中进行学习，输出配电网故障定位模型训练后的数据，采用IWOA对定位模型的参数进行寻优，所述寻优包括基于改进的鲸鱼优化算法的初始化参数进行迭代次数，根据步长在寻优区间中随机生成的参数个体构成种群，将结果输入到XGBoost模型中，提取的故障特征为其中，有n样本，则K棵回归树集成故障预测结果为：As a preferred solution of the distribution network fault location method based on fusion model and cloud-edge collaboration described in the present invention, the feature extraction includes inputting the objective function into the distribution network fault location model for learning, outputting the data after the distribution network fault location model training, and using IWOA to optimize the parameters of the location model. The optimization includes iterating the number of times based on the initialization parameters of the improved whale optimization algorithm, randomly generating parameter individuals in the optimization interval according to the step size to form a population, and inputting the results into the XGBoost model. The extracted fault features are in, If there are n samples, the fault prediction result of K regression trees is:
其中，各函数fk是独立的一棵回归树，F为CART的集合，fk(xi)为在第K棵，CART上数据样本集i的预测值，为最终的预测结果，损失函数与惩罚函数相叠加的XGBoost目标函数定位为：Among them, each function fk is an independent regression tree, F is the set of CART, fk (xi ) is the predicted value of data sample set i on the Kth CART, For the final prediction result, the XGBoost objective function with the loss function and the penalty function superimposed is positioned as:
其中，yi表示样本的样本集的真实值，表示损失函数，Ω(fk)为正则项，T和W分别表示叶子节点的数量及权重，γ和λ为惩罚项；Among them,yi represents the true value of the sample set, represents the loss function, Ω(fk ) is the regularization term, T and W represent the number and weight of leaf nodes respectively, and γ and λ are penalty terms;
故障特征在训练后都会落在决策树的叶子节点上，根据叠加的XGBoost目标函数定位得到最终目标函数为：After training, the fault features will fall on the leaf nodes of the decision tree. According to the superimposed XGBoost objective function positioning, the final objective function is:
其中，Ij表示第j个叶子节点的样本集合，wj为二次函数。Among them,Ij represents the sample set of the jth leaf node, andwj is a quadratic function.
作为本发明所述基于融合模型和云边端协同的配电网故障定位方法的一种优选方案，其中：所述输出配电网故障位置包括根据最终目标函数输入的参数计算的配电网故障定位模型的准确率，返回适应度值，将本次迭代的个体适应度值和全局的适应度值进行比较，更新并记录全局的最优个体的适应度值；As a preferred solution of the distribution network fault location method based on the fusion model and cloud-edge collaboration described in the present invention, wherein: the output distribution network fault location includes the accuracy of the distribution network fault location model calculated according to the parameters input by the final objective function, returns the fitness value, compares the individual fitness value of this iteration with the global fitness value, and updates and records the global optimal individual fitness value;
若达到最大的迭代次数，则输出最佳的超参数组合，将寻优得到的最优的超参数组合输入到XGBoost模型中，输出定位结果；If the maximum number of iterations is reached, the best hyperparameter combination is output, the optimal hyperparameter combination obtained by optimization is input into the XGBoost model, and the positioning result is output;
若未达到最大迭代次数，则继续迭代，更新种群中各鲸鱼个体的位置和适应度值，更新种群中各鲸鱼个体的位置步骤如下：If the maximum number of iterations has not been reached, continue iterating to update the position and fitness value of each whale in the population. The steps for updating the position of each whale in the population are as follows:
计算非线性因子，具体计算公式为：Calculate the nonlinear factor. The specific calculation formula is:
其中，t为迭代次数；Where t is the number of iterations;
根据上式求得的非线性因子计算收敛因子A和摆动因子C，具体公式为：The convergence factor A and the swing factor C are calculated based on the nonlinear factor obtained from the above formula. The specific formula is:
A＝2ar-aA＝2ar-a
C＝2rC＝2r
其中，a由2到0线性变化，r为[0,1]的随机向量；Among them, a changes linearly from 2 to 0, and r is a random vector of [0,1];
若随机数P＜0.5，则按照下式更新搜索位置，具体公式为：If the random number P is less than 0.5, the search position is updated according to the following formula:
其中，X为鲸鱼个体的位置，b为螺旋包围常数，l为(-1,1)之间的随机数，D为当前个体到最优个体之间的距离；Among them, X is the position of the whale individual, b is the spiral enclosure constant, l is a random number between (-1,1), and D is the distance between the current individual and the optimal individual;
若随机数P≥0.5，且收敛因子A的绝对值大于1时，则按下式更新鲸鱼个体的位置，具体公式为：If the random number P ≥ 0.5, and the absolute value of the convergence factor A is greater than 1, the position of the individual whale is updated as follows:
否则，按照如下公式继续更新个体的位置，具体公式为：Otherwise, continue to update the individual's position according to the following formula:
当鲸鱼个体进行多项式的变异，根据如下公式进行限定，具体公式为：When the whale individual undergoes polynomial mutation, it is limited according to the following formula:
其中，μ为[0,1]上的随机数，ηm为分布指数，lk为位置下限，μk为位置上限。in, μ is a random number on [0,1], ηm is the distribution index, lk is the lower limit of the position, and μk is the upper limit of the position.
作为本发明所述基于融合模型和云边端协同的配电网故障定位方法的一种优选方案，其中：所述展示故障位置信息包括根据鲸鱼个体的位置和适应度值精准确定配电网故障位置，在配电网线路的拓扑图中展示故障位置，通知相关的运维人员进行处理。As an optimal solution for the distribution network fault location method based on fusion model and cloud-edge-end collaboration described in the present invention, the display of fault location information includes accurately determining the distribution network fault location according to the location and fitness value of individual whales, displaying the fault location in the topology map of the distribution network line, and notifying relevant operation and maintenance personnel to handle it.
第二方面，本发明实施例提供了基于融合模型和云边端协同的配电网故障定位系统，其包括：处理模块，通过利用端控制器采集当前时刻的配电网数据，端控制器将配电网数据按照通进行封装，边控制器判断支路是否发生故障，对发生故障支路的数据进行预处理；计算模块，其边控制器根据设备定值和实时采集的电压和电流信息进行比较，将比较后的结果数据上传至云控制器，计算对应设备定值；提取模块，其将对应的定值设备编码下发至边控制器，云控制器接收到边控制器上传的故障数据后进行特征提取，输出配电网故障位置，通过故障定位模型展示故障位置信息。In the second aspect, an embodiment of the present invention provides a distribution network fault location system based on a fusion model and cloud-edge-end collaboration, which includes: a processing module, which collects distribution network data at the current moment by using an end controller, and the end controller encapsulates the distribution network data according to the channel. The edge controller determines whether a branch has a fault and pre-processes the data of the faulty branch; a calculation module, in which the edge controller compares the device set value with the voltage and current information collected in real time, uploads the compared result data to the cloud controller, and calculates the corresponding device set value; an extraction module, which sends the corresponding set value device code to the edge controller, and the cloud controller performs feature extraction after receiving the fault data uploaded by the edge controller, outputs the distribution network fault location, and displays the fault location information through the fault location model.
第三方面，本发明实施例提供了一种计算机设备，包括存储器和处理器，所述存储器存储有计算机程序，其中：所述处理器执行所述计算机程序时实现上述的基于融合模型和云边端协同的配电网故障定位方法的任一步骤。In a third aspect, an embodiment of the present invention provides a computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: when the processor executes the computer program, it implements any step of the above-mentioned distribution network fault location method based on fusion model and cloud-edge-end collaboration.
第四方面，本发明实施例提供了一种计算机可读存储介质，其上存储有计算机程序，其中：所述计算机程序被处理器执行时实现上述的基于融合模型和云边端协同的配电网故障定位方法的任一步骤。In a fourth aspect, an embodiment of the present invention provides a computer-readable storage medium on which a computer program is stored, wherein: when the computer program is executed by a processor, any step of the above-mentioned distribution network fault location method based on fusion model and cloud-edge collaboration is implemented.
本发明有益效果为：本发明通过融合模型和云边协同的配电网故障定位方法，有效解决了在复杂配电线路中云服务器与边缘端连接数量过多的问题，减轻了云服务器的计算和存储压力，降低了网络时延，提高了故障定位的时效性。采用端、边和云三级协同的方式，通过端控制器实时采集数据，边控制器判断故障并进行预处理，云控制器计算设备定值和建立故障定位模型。利用改进的鲸鱼优化算法进行模型参数寻优，提高了模型的准确性和稳定性，系统中引入电流相位角整体偏差指标和基于XGBoost的故障特征提取，进一步提高了故障定位的精确度，通过优化参数和协同计算，本发明有效降低了云服务器的负担，为配电网故障定位提供了高效准确的解决方案。The beneficial effects of the present invention are as follows: the present invention effectively solves the problem of too many connections between cloud servers and edge terminals in complex distribution lines through a distribution network fault location method that integrates models and cloud-edge collaboration, reduces the computing and storage pressure of cloud servers, reduces network latency, and improves the timeliness of fault location. A three-level collaborative approach of end, edge, and cloud is adopted, with the end controller collecting data in real time, the edge controller determining faults and performing preprocessing, and the cloud controller calculating device constants and establishing a fault location model. The improved whale optimization algorithm is used to optimize model parameters, which improves the accuracy and stability of the model. The overall deviation index of the current phase angle and the fault feature extraction based on XGBoost are introduced into the system, which further improves the accuracy of fault location. By optimizing parameters and collaborative calculations, the present invention effectively reduces the burden on cloud servers and provides an efficient and accurate solution for distribution network fault location.
附图说明BRIEF DESCRIPTION OF THE DRAWINGS
为了更清楚地说明本发明实施例的技术方案，下面将对实施例描述中所需要使用的附图作简单地介绍，显而易见地，下面描述中的附图仅仅是本发明的一些实施例，对于本领域普通技术人员来讲，在不付出创造性劳动的前提下，还可以根据这些附图获得其它的附图。其中：In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following briefly introduces the drawings required for describing the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work. Among them:
图1为本发明一个实施例提供的基于融合模型和云边端协同的配电网故障定位方法及系统的流程示意图。FIG1 is a flow chart of a distribution network fault location method and system based on a fusion model and cloud-edge-end collaboration provided by an embodiment of the present invention.
图2为本发明一个实施例提供的基于融合模型和云边端协同的配电网故障定位方法及系统的配电网故障定位模型的流程示意图。Figure 2 is a flow chart of a distribution network fault location model of a distribution network fault location method and system based on a fusion model and cloud-edge-end collaboration provided by an embodiment of the present invention.
具体实施方式Detailed ways
为使本发明的上述目的、特征和优点能够更加明显易懂，下面结合说明书附图对本发明的具体实施方式做详细的说明，显然所描述的实施例是本发明的一部分实施例，而不是全部实施例。基于本发明中的实施例，本领域普通人员在没有做出创造性劳动前提下所获得的所有其他实施例，都应当属于本发明的保护的范围。In order to make the above-mentioned purposes, features and advantages of the present invention more obvious and easy to understand, the specific implementation methods of the present invention are described in detail below in conjunction with the drawings of the specification. Obviously, the described embodiments are part of the embodiments of the present invention, but not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary persons in the art without creative work should fall within the scope of protection of the present invention.
在下面的描述中阐述了很多具体细节以便于充分理解本发明，但是本发明还可以采用其他不同于在此描述的其它方式来实施，本领域技术人员可以在不违背本发明内涵的情况下做类似推广，因此本发明不受下面公开的具体实施例的限制。In the following description, many specific details are set forth to facilitate a full understanding of the present invention, but the present invention may also be implemented in other ways different from those described herein, and those skilled in the art may make similar generalizations without violating the connotation of the present invention. Therefore, the present invention is not limited to the specific embodiments disclosed below.
其次，此处所称的“一个实施例”或“实施例”是指可包含于本发明至少一个实现方式中的特定特征、结构或特性。在本说明书中不同地方出现的“在一个实施例中”并非均指同一个实施例，也不是单独的或选择性的与其他实施例互相排斥的实施例。Secondly, the term "one embodiment" or "embodiment" as used herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation of the present invention. The term "in one embodiment" that appears in different places in this specification does not necessarily refer to the same embodiment, nor does it refer to a separate or selective embodiment that is mutually exclusive with other embodiments.
本发明结合示意图进行详细描述，在详述本发明实施例时，为便于说明，表示器件结构的剖面图会不依一般比例作局部放大，而且所述示意图只是示例，其在此不应限制本发明保护的范围。此外，在实际制作中应包含长度、宽度及深度的三维空间尺寸。The present invention is described in detail with reference to schematic diagrams. When describing the embodiments of the present invention, for the sake of convenience, the cross-sectional diagrams showing the device structure will not be partially enlarged according to the general scale, and the schematic diagrams are only examples, which should not limit the scope of protection of the present invention. In addition, in actual production, the three-dimensional dimensions of length, width and depth should be included.
同时在本发明的描述中，需要说明的是，术语中的“上、下、内和外”等指示的方位或位置关系为基于附图所示的方位或位置关系，仅是为了便于描述本发明和简化描述，而不是指示或暗示所指的装置或元件必须具有特定的方位、以特定的方位构造和操作，因此不能理解为对本发明的限制。此外，术语“第一、第二或第三”仅用于描述目的，而不能理解为指示或暗示相对重要性。At the same time, in the description of the present invention, it should be noted that the directions or positional relationships indicated by the terms "upper, lower, inner and outer" are based on the directions or positional relationships shown in the drawings, which are only for the convenience of describing the present invention and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific direction, be constructed and operated in a specific direction, and therefore cannot be understood as limiting the present invention. In addition, the terms "first, second or third" are only used for descriptive purposes and cannot be understood as indicating or implying relative importance.
本发明中除非另有明确的规定和限定，术语“安装、相连、连接”应做广义理解，例如：可以是固定连接、可拆卸连接或一体式连接；同样可以是机械连接、电连接或直接连接，也可以通过中间媒介间接相连，也可以是两个元件内部的连通。对于本领域的普通技术人员而言，可以具体情况理解上述术语在本发明中的具体含义。In the present invention, unless otherwise clearly specified and limited, the terms "install, connect, connect" should be understood in a broad sense, for example: it can be a fixed connection, a detachable connection or an integral connection; it can also be a mechanical connection, an electrical connection or a direct connection, or it can be indirectly connected through an intermediate medium, or it can be the internal communication of two components. For ordinary technicians in this field, the specific meanings of the above terms in the present invention can be understood according to specific circumstances.
实施例1Example 1
参照图1和图2，为本发明第一个实施例，该实施例提供了基于融合模型和云边端协同的配电网故障定位方法，包括：Referring to FIG. 1 and FIG. 2 , a first embodiment of the present invention is provided, which provides a distribution network fault location method based on a fusion model and cloud-edge-end collaboration, including:
S1：利用端控制器采集当前时刻的配电网数据，端控制器将配电网数据按照通讯协议进行封装，将封装完成后的数据上传至边控制器，边控制器判断支路是否发生故障，对发生故障支路的数据进行预处理。S1: Use the end controller to collect the distribution network data at the current moment. The end controller encapsulates the distribution network data according to the communication protocol and uploads the encapsulated data to the edge controller. The edge controller determines whether a branch fails and pre-processes the data of the failed branch.
其中，当前时刻的配电网数据包括三相电压、三相电流、零序电压以及零序电流，利用通信协议对采集到的数据进行封装，上传包括按照传输格式规定进行传输，预处理包括根据封装后的实时线路电流进行判断；Among them, the distribution network data at the current moment includes three-phase voltage, three-phase current, zero-sequence voltage and zero-sequence current. The collected data is encapsulated using a communication protocol, and uploading includes transmitting according to the transmission format regulations. Preprocessing includes making a judgment based on the real-time line current after encapsulation;
若实时线路电流大于设定电流定值或电压小于设定电压定值时，则设备发生故障，此时边控制器开始采集故障时刻前后两个周波的数据；否则设备未发生故障；If the real-time line current is greater than the set current value or the voltage is less than the set voltage value, the equipment fails, and the edge controller starts to collect data of the two cycles before and after the fault; otherwise, the equipment does not fail;
提取发生故障数据，采用插值法对空值进行补齐，计算对应的标幺值，具体步骤如下：Extract the fault data, use interpolation to fill the empty values, and calculate the corresponding per-unit value. The specific steps are as follows:
根据三相电流计算电流幅值，具体计算公式为：The current amplitude is calculated based on the three-phase current. The specific calculation formula is:
其中，Ia、Ib和Ic表示三相电流幅值，Irms表示电流幅值；Among them, Ia, Ib and Ic represent the three-phase current amplitudes, and Irms represents the current amplitude;
根据三相电流计算电流相位角，具体计算公式为：The current phase angle is calculated based on the three-phase current. The specific calculation formula is:
其中，Iθ表示三相电流相位角，即电流的相位关系；Among them, Iθ represents the three-phase current phase angle, that is, the phase relationship of the current;
根据零序电流计算零序电流幅值，具体计算公式为：The zero-sequence current amplitude is calculated based on the zero-sequence current. The specific calculation formula is:
其中，I0rms表示零序电流的有效值，即零序电流的平均幅值，I0a、I0b和I0c表示三相零序电流幅值；Wherein, I0rms represents the effective value of zero-sequence current, that is, the average amplitude of zero-sequence current, I0a , I0b and I0c represent the amplitudes of three-phase zero-sequence current;
根据零序电流计算零序电流相位角，具体计算公式为：The zero-sequence current phase angle is calculated based on the zero-sequence current. The specific calculation formula is:
其中，I0θ表示零序电流的相位角，即零序电流的相位关系；Among them, I0θ represents the phase angle of the zero-sequence current, that is, the phase relationship of the zero-sequence current;
边控制器判断支路是否发生故障包括在边控制器中利用计算得到的三相电流幅值和三相电流相位角进行故障判断；The side controller determines whether a branch circuit is faulty, including using the calculated three-phase current amplitude and three-phase current phase angle to perform fault judgment in the side controller;
若三相电流幅值大于设定的电流阈值或大于电流相位角偏离正常范围时，则表明该支路存在异常电流，判断发生故障。If the three-phase current amplitude is greater than the set current threshold or greater than the current phase angle deviation from the normal range, it indicates that there is an abnormal current in the branch and a fault is determined to have occurred.
进一步的，利用端控制器采集当前时刻的配电网数据，包括三相电压(Ua＝230V、Ub＝230V、Uc＝230V)、三相电流(Ia＝50A、Ib＝45A、Ic＝48A)、零序电压(U0＝10V)以及零序电流(I0＝2A)，通过通信协议对采集到的数据进行封装，并按照传输格式规定上传至边控制器，在上传过程中，数据的传输格式符合规范，确保有效的通信。Furthermore, the end controller is used to collect the distribution network data at the current moment, including three-phase voltage (Ua=230V, Ub=230V, Uc=230V), three-phase current (Ia=50A, Ib=45A, Ic=48A), zero-sequence voltage (U0=10V) and zero-sequence current (I0=2A). The collected data is encapsulated through the communication protocol and uploaded to the edge controller according to the transmission format regulations. During the upload process, the data transmission format complies with the specifications to ensure effective communication.
在边控制器中，进行故障判断的具体步骤包括根据实时线路电流进行判断，若实时线路电流大于设定的电流定值(设定值为60A)或电压小于设定的电压定值(设定值为220V)，则判定设备发生故障，此时，边控制器开始采集故障时刻前后两个周波的数据，以获取更全面的信息用于故障定位。In the edge controller, the specific steps of fault judgment include judging based on the real-time line current. If the real-time line current is greater than the set current value (the set value is 60A) or the voltage is less than the set voltage value (the set value is 220V), the equipment is judged to be faulty. At this time, the edge controller starts to collect data of the two cycles before and after the fault to obtain more comprehensive information for fault location.
发生故障的支路数据在采集后进行预处理，包括采用插值法对空值进行补齐，并计算对应的标幺值，具体计算步骤涵盖了三相电流幅值、电流相位角、零序电流幅值、零序电流相位角的计算公式，边控制器利用计算得到的三相电流幅值(Ia＝52A、Ib＝48A、Ic＝50A)和相位角进行故障判断，若三相电流幅值大于设定的电流阈值(设定值为55A)或大于电流相位角偏离正常范围，则判定该支路存在异常电流，进而判断发生故障，比较本发明设置的数值与典型数值范围或标准值之间的关系如下表1所示：The data of the branch where the fault occurs is preprocessed after collection, including using the interpolation method to fill the null value and calculate the corresponding per-unit value. The specific calculation steps cover the calculation formulas of the three-phase current amplitude, current phase angle, zero-sequence current amplitude, and zero-sequence current phase angle. The edge controller uses the calculated three-phase current amplitude (Ia=52A, Ib=48A, Ic=50A) and phase angle to make a fault judgment. If the three-phase current amplitude is greater than the set current threshold (the set value is 55A) or greater than the current phase angle deviation from the normal range, it is determined that the branch has an abnormal current, and then it is determined that a fault occurs. The relationship between the values set by the present invention and the typical value range or standard value is shown in Table 1 below:
表1本发明设置数值与典型数值范围或标准值之间的关系数据表Table 1. Relationship between the numerical values set in the present invention and the typical numerical range or standard value.
本发明设置的示例数据与典型数值范围或标准值相匹配，确保在电力系统实际应用中的合理性，各项参数如三相电压、电流、零序电压以及零序电流均符合正常运行范围，电流定值、电压定值和电流阈值考虑了系统和设备规格，保证了本发明在实际电网环境下的有效性和准确性。The example data set in the present invention matches the typical numerical range or standard value to ensure rationality in actual application of the power system. Various parameters such as three-phase voltage, current, zero-sequence voltage and zero-sequence current are within the normal operating range. The current setting, voltage setting and current threshold take into account the system and equipment specifications, ensuring the effectiveness and accuracy of the present invention in the actual power grid environment.
S2：边控制器根据设备定值和实时采集的电压和电流信息进行比较，将比较后的结果数据上传至云控制器，云控制器根据输入的线路信息生成配电网线路拓扑图并计算对应设备定值。S2: The edge controller compares the device settings with the voltage and current information collected in real time, and uploads the comparison result data to the cloud controller. The cloud controller generates a distribution network line topology diagram based on the input line information and calculates the corresponding device settings.
其中，比较包括电流幅值比较和电流相位角比较；电流幅值比较包括计算各相电流波形的均方根值，根据均方根值计算三相电流的有效值，设定电流阈值进行比较；电流相位角比较包括计算电流相位角的整体偏差指标，根据整体偏差指标检查相位角是否偏离正常范围，具体步骤如下：Among them, the comparison includes current amplitude comparison and current phase angle comparison; the current amplitude comparison includes calculating the root mean square value of each phase current waveform, calculating the effective value of the three-phase current according to the root mean square value, and setting the current threshold for comparison; the current phase angle comparison includes calculating the overall deviation index of the current phase angle, and checking whether the phase angle deviates from the normal range according to the overall deviation index. The specific steps are as follows:
计算各相电流波形的均方根值公式为：The formula for calculating the RMS value of each phase current waveform is:
计算三相电流的有效值公式为：The formula for calculating the effective value of three-phase current is:
其中，N表示采样点数，Iai、Ibi和Ici表示第i个采样点时的相电流值，表示相电流波形a的均方根值，/>表示相电流波形b的均方根值，/>表示相电流波形c的均方根值，I'rms表示三相电流的有效值，即各支路上提取的采样点值；Where N represents the number of sampling points, Iai , Ibi and Ici represent the phase current values at the i-th sampling point, represents the RMS value of the phase current waveform a, /> represents the RMS value of the phase current waveform b, /> represents the RMS value of the phase current waveform c, and I'rms represents the effective value of the three-phase current, i.e., the sampling point value extracted from each branch;
若I'rms＞设定的电流阈值时，则判断电网发生故障；若I'rms≤设定的电流阈值时，此时判定电流相位角比较，具体步骤如下：If I'rms > the set current threshold, it is determined that the power grid has a fault; if I'rms ≤ the set current threshold, the current phase angle comparison is determined at this time. The specific steps are as follows:
计算电流相位角的整体偏差指标具体公式为：The specific formula for calculating the overall deviation index of the current phase angle is:
其中，N表示采样点数，δθn表示第n个采样点时的相位角偏差，Δθ表示电流相位角整体偏差指标，设定正常范围为[θmin，θmax]；Wherein, N represents the number of sampling points, δθn represents the phase angle deviation at the nth sampling point, Δθ represents the overall deviation index of the current phase angle, and the normal range is set to [θmin , θmax ];
若Δθ＞θmax-θmin时，则表示相位角偏离正常范围，若Δθ≤θmax-θmin时，则表示相位角未偏离正常范围。If Δθ＞θmax -θmin , it means that the phase angle deviates from the normal range, and if Δθ≤θmax -θmin , it means that the phase angle does not deviate from the normal range.
进一步的，电流幅值比较阶段，对各相电流波形的均方根值进行计算，假设采样点数为N，相电流波形分别为a、b和c，如果均方根值大于设定的电流阈值，假设设定值为40A，则判定电网发生故障。Furthermore, in the current amplitude comparison stage, the RMS value of each phase current waveform is calculated. Assuming the number of sampling points is N, the phase current waveforms are a, b and c respectively. If the RMS value is greater than the set current threshold, assuming the set value is 40A, it is determined that a grid fault has occurred.
在电流相位角比较阶段，计算电流相位角的整体偏差指标，假设采样点数为N，相位角偏差为δ，设定正常范围为±5°，若整体偏差指标大于设定的阈值，设定值为3°，则表示相位角偏离正常范围，判定发生故障，故障定位过程中所设置的关键参数数值与比较的数值范围和标准值进行详细对比，具体对比如下表2所示：In the current phase angle comparison stage, the overall deviation index of the current phase angle is calculated. Assuming that the number of sampling points is N, the phase angle deviation is δ, and the normal range is set to ±5°, if the overall deviation index is greater than the set threshold, the set value is 3°, it means that the phase angle deviates from the normal range, and it is determined that a fault has occurred. The key parameter values set in the fault location process are compared in detail with the comparison value range and standard value. The specific comparison is shown in Table 2 below:
表2与比较的数值范围和标准值的数据对比表Table 2 Data comparison table with the numerical range and standard value
本发明在故障定位中设置了关键参数数值，如电流均方根阈值、电流相位角偏差阈值和零序电流有效值，数值的设定旨在确保在正常运行条件下不会误判故障，同时在电流或相位角异常时准确判定故障，通过与比较数值范围和标准值对比，保证了系统在各种情况下的可靠性和准确性。The present invention sets key parameter values in fault location, such as the current root mean square threshold, the current phase angle deviation threshold and the zero-sequence current effective value. The setting of the values is intended to ensure that faults are not misjudged under normal operating conditions, and to accurately determine faults when the current or phase angle is abnormal. By comparing with the comparison value range and the standard value, the reliability and accuracy of the system in various situations are guaranteed.
S3：将对应的定值设备编码下发至边控制器，建立对应的故障定位模型，云控制器接收到边控制器上传的故障数据后进行特征提取，输出配电网故障位置，通过故障定位模型展示故障位置信息。S3: Send the corresponding fixed value device code to the edge controller and establish the corresponding fault location model. After receiving the fault data uploaded by the edge controller, the cloud controller extracts features, outputs the distribution network fault location, and displays the fault location information through the fault location model.
其中，将对应的定值设备编码下发至边控制器包括将电流相位角整体偏差指标值进行下发至边控制器，建立对应的故障定位模型包括将下发至边控制器的电流相位角整体偏差指标值使用分类器进行分类，计算得到分类的目标值，使用带标签的故障数据进行训练，将L1正则化的混合特征进行选择，满足要求的特征向量，在目标值中增加惩罚项，获取目标函数，将目标函数绝对值之和变小，获取稀疏特征向量其中，/>表示混合特征对应的第i个样本数据，使用平方误差作为目标损失函数，目标函数公式定义为：Among them, sending the corresponding fixed value device code to the edge controller includes sending the overall deviation index value of the current phase angle to the edge controller, establishing the corresponding fault location model includes using a classifier to classify the overall deviation index value of the current phase angle sent to the edge controller, calculating the classification target value, using labeled fault data for training, selecting theL1 regularized mixed features, meeting the required feature vectors, adding a penalty term to the target value, obtaining the objective function, reducing the sum of the absolute values of the objective function, and obtaining a sparse feature vector Among them,/> Represents the i-th sample data corresponding to the mixed feature, using the square error as the target loss function, and the target function formula is defined as:
其中，ωp表示第p个特征的回归系数，K表示稀疏度控制参数，K越大，则模型越稀疏；Among them, ωp represents the regression coefficient of the pth feature, K represents the sparsity control parameter, and the larger K is, the sparser the model is;
使用梯度下降法求解上述目标损失函数方程，第K+1次迭代方程为：Use the gradient descent method to solve the above objective loss function equation. The K+1th iteration equation is:
其中，ω＝(ω1,ω2,...,ωn+q)和M是大于0的常数；Wherein, ω＝(ω1 ,ω2 ,...,ωn+q ) and M is a constant greater than 0;
若时，则方程通过如下公式进行计算为：like When , the equation is calculated by the following formula:
其中，和μp分别是ω(k+1)和μ的第p个分量，最后求解L1的归一化结果表示，将ωp的非零向量对应的特征被选择为最终的特征子集向量；in, and μp are the p-th components of ω(k+1) and μ respectively. Finally, the normalized result of solving L1 indicates that the features corresponding to the non-zero vector of ωp are selected as the final feature subset vector;
将特征子集向量转换为矩阵形式，利用电气量进行归一化处理，将矩阵中各值乘积生成对应的灰色图，具体电气量/>归一化处理公式为：Convert the feature subset vector into matrix form and use the electrical quantity Normalize and multiply the values in the matrix to generate the corresponding gray map. The specific electrical quantity/> The normalization formula is:
其中，表示归一化之后的数据，y表示归一化之前的数据，ymax和ymin分别表示最大值和最小值。in, represents the data after normalization, y represents the data before normalization, ymax and ymin represent the maximum and minimum values respectively.
S3.1：特征提取包括将目标函数输入到配电网故障定位模型中进行学习，输出配电网故障定位模型训练后的数据，采用IWOA对定位模型的参数进行寻优，寻优包括基于改进的鲸鱼优化算法的初始化参数进行迭代次数，根据步长在寻优区间中随机生成的参数个体构成种群，将结果输入到XGBoost模型中，提取的故障特征为其中，有n样本，则K棵回归树集成故障预测结果为：S3.1: Feature extraction includes inputting the objective function into the distribution network fault location model for learning, outputting the data after the distribution network fault location model training, and optimizing the parameters of the location model using IWOA. The optimization includes the number of iterations based on the initialization parameters of the improved whale optimization algorithm, and randomly generating parameter individuals in the optimization interval according to the step size to form a population. The results are input into the XGBoost model, and the extracted fault features are in, If there are n samples, the fault prediction result of K regression trees is:
其中，各函数fk是独立的一棵回归树，F为CART的集合，fk(xi)为在第K棵，CART上数据样本集i的预测值，为最终的预测结果，损失函数与惩罚函数相叠加的XGBoost目标函数定位为：Among them, each function fk is an independent regression tree, F is the set of CART, fk (xi ) is the predicted value of data sample set i on the Kth CART, For the final prediction result, the XGBoost objective function with the loss function and the penalty function superimposed is positioned as:
其中，yi表示样本的样本集的真实值，表示损失函数，Ω(fk)为正则项，T和W分别表示叶子节点的数量及权重，γ和λ为惩罚项；Among them,yi represents the true value of the sample set, represents the loss function, Ω(fk ) is the regularization term, T and W represent the number and weight of leaf nodes respectively, and γ and λ are penalty terms;
故障特征在训练后都会落在决策树的叶子节点上，根据叠加的XGBoost目标函数定位得到最终目标函数为：After training, the fault features will fall on the leaf nodes of the decision tree. According to the superimposed XGBoost objective function positioning, the final objective function is:
其中，Ij表示第j个叶子节点的样本集合，wj为二次函数。Among them,Ij represents the sample set of the jth leaf node, andwj is a quadratic function.
进一步的，展示故障位置信息包括根据鲸鱼个体的位置和适应度值精准确定配电网故障位置，在配电网线路的拓扑图中展示故障位置，通知相关的运维人员进行处理。Furthermore, displaying the fault location information includes accurately determining the distribution network fault location based on the location and fitness value of the individual whales, displaying the fault location in the topological map of the distribution network line, and notifying relevant operation and maintenance personnel to handle it.
在一个优选实施例中，基于融合模型和云边端协同的配电网故障定位系统，该系统包括处理模块，通过利用端控制器采集当前时刻的配电网数据，端控制器将配电网数据按照通讯协议进行封装，边控制器判断支路是否发生故障，对发生故障支路的数据进行预处理；计算模块，其边控制器根据设备定值对实时采集的电压和电流信息进行比较，将比较后的结果数据上传至云控制器，计算对应设备定值；提取模块，其将对应的定值设备编码下发至边控制器，云控制器接收到边控制器上传的故障数据后进行特征提取，输出配电网故障位置，通过故障定位模型展示故障位置信息。In a preferred embodiment, a distribution network fault location system based on a fusion model and cloud-edge-end collaboration includes a processing module, which collects distribution network data at the current moment by using an end controller, and the end controller encapsulates the distribution network data according to a communication protocol. The edge controller determines whether a branch has a fault and pre-processes the data of the faulty branch; a calculation module, in which the edge controller compares the voltage and current information collected in real time according to the device constants, uploads the compared result data to the cloud controller, and calculates the corresponding device constants; an extraction module, which sends the corresponding constant device code to the edge controller, and the cloud controller extracts features after receiving the fault data uploaded by the edge controller, outputs the distribution network fault location, and displays the fault location information through a fault location model.
上述各单元模块可以硬件形式内嵌于或独立于计算机设备中的处理器中，也可以以软件形式存储于计算机设备中的存储器中，以便于处理器调用执行以上各个模块对应的操作。The above-mentioned unit modules may be embedded in or independent of the processor in the computer device in the form of hardware, or may be stored in the memory in the computer device in the form of software, so that the processor can call and execute the operations corresponding to the above-mentioned modules.
本实施例还提供一种计算机设备，适用于基于物联网的多源电网信息融合方法的情况，包括存储器和处理器；存储器用于存储计算机可执行指令，处理器用于执行计算机可执行指令，实现如上述实施例提出的基于物联网的多源电网信息融合方法。This embodiment also provides a computer device, which is suitable for the multi-source power grid information fusion method based on the Internet of Things, including a memory and a processor; the memory is used to store computer-executable instructions, and the processor is used to execute computer-executable instructions to implement the multi-source power grid information fusion method based on the Internet of Things proposed in the above embodiment.
在一个实施例中，提供了一种计算机设备，该计算机设备可以是终端，该计算机设备包括通过系统总线连接的处理器、存储器、通信接口、显示屏和输入装置。其中，该计算机设备的处理器用于提供计算和控制能力。该计算机设备的存储器包括非易失性存储介质、内存储器。该非易失性存储介质存储有操作系统和计算机程序。该内存储器为非易失性存储介质中的操作系统和计算机程序的运行提供环境。该计算机设备的通信接口用于与外部的终端进行有线或无线方式的通信，无线方式可通过WIFI、运营商网络、NFC(近场通信)或其他技术实现。该计算机设备的显示屏可以是液晶显示屏或者电子墨水显示屏，该计算机设备的输入装置可以是显示屏上覆盖的触摸层，也可以是计算机设备外壳上设置的按键、轨迹球或触控板，还可以是外接的键盘、触控板或鼠标等。In one embodiment, a computer device is provided, which may be a terminal, and includes a processor, a memory, a communication interface, a display screen, and an input device connected via a system bus. The processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The communication interface of the computer device is used to communicate with an external terminal in a wired or wireless manner, and the wireless manner can be implemented through WIFI, an operator network, NFC (near field communication) or other technologies. The display screen of the computer device may be a liquid crystal display screen or an electronic ink display screen, and the input device of the computer device may be a touch layer covered on the display screen, or a key, trackball or touchpad provided on the housing of the computer device, or an external keyboard, touchpad or mouse, etc.
综上，本发明通过融合模型和云边协同的配电网故障定位方法，有效解决了在复杂配电线路中云服务器与边缘端连接数量过多的问题，减轻了云服务器的计算和存储压力，降低了网络时延，提高了故障定位的时效性。采用端、边和云三级协同的方式，通过端控制器实时采集数据，边控制器判断故障并进行预处理，云控制器计算设备定值和建立故障定位模型。利用改进的鲸鱼优化算法进行模型参数寻优，提高了模型的准确性和稳定性，系统中引入电流相位角整体偏差指标和基于XGBoost的故障特征提取，进一步提高了故障定位的精确度，通过优化参数和协同计算，本发明有效降低了云服务器的负担，为配电网故障定位提供了高效准确的解决方案。In summary, the present invention effectively solves the problem of too many connections between cloud servers and edge terminals in complex distribution lines through a distribution network fault location method that integrates models and cloud-edge collaboration, reduces the computing and storage pressure of cloud servers, reduces network latency, and improves the timeliness of fault location. Using a three-level collaborative approach of end, edge, and cloud, the end controller collects data in real time, the edge controller determines the fault and performs preprocessing, and the cloud controller calculates the device constant and establishes a fault location model. The improved whale optimization algorithm is used to optimize model parameters, which improves the accuracy and stability of the model. The overall deviation index of the current phase angle and the fault feature extraction based on XGBoost are introduced into the system to further improve the accuracy of fault location. By optimizing parameters and collaborative calculations, the present invention effectively reduces the burden on cloud servers and provides an efficient and accurate solution for distribution network fault location.
实施例2Example 2
参照图1～图2，为本发明第二个实施例，该实施例提供了基于融合模型和云边端协同的配电网故障定位方法，为了验证本发明的有益效果，通过仿真实验进行科学论证。Referring to Figures 1 and 2, a second embodiment of the present invention is shown. This embodiment provides a distribution network fault location method based on a fusion model and cloud-edge collaboration. In order to verify the beneficial effects of the present invention, scientific demonstration is carried out through simulation experiments.
实验数据包括三相电流幅值(Ia、Ib、Ic)、三相电流相位角零序电流幅值(I0)以及零序电流相位角/>将设定的电流阈值(40A)和电流相位角范围(±5°)为基准，模拟了电网发生故障的情境。The experimental data include three-phase current amplitude (Ia, Ib, Ic), three-phase current phase angle Zero-sequence current amplitude (I0) and zero-sequence current phase angle/> Taking the set current threshold (40A) and current phase angle range (±5°) as the benchmark, the scenario of power grid failure was simulated.
实验结果表明，当三相电流幅值分别为30A、35A和45A，相应相位角分别为10°、12°和15°时，零序电流幅值为2A，零序电流相位角为5°时，边控制器判定电网发生故障。本发明在复杂配电线路中的故障定位方法的有效性，通过合理设定参数，成功实现了对电网故障的准确判断，采集到的具体数据如下表3所示：The experimental results show that when the three-phase current amplitudes are 30A, 35A and 45A respectively, and the corresponding phase angles are 10°, 12° and 15° respectively, the zero-sequence current amplitude is 2A, and the zero-sequence current phase angle is 5°, the edge controller determines that the power grid has a fault. The effectiveness of the fault location method of the present invention in complex distribution lines is successfully achieved by setting parameters reasonably, and the specific data collected are shown in Table 3 below:
表3实验过程中采集的具体数据表Table 3 Specific data collected during the experiment
数据代表了具有唯一性的不同电流和相位条件，用于验证本发明的故障定位方法在不同情况下的性能，例如，三相电流幅值分别为30A、35A和45A，相应的相位角分别为10°、12°和15°，以及零序电流幅值为2A，相位角为5°。上述数据的唯一性确保了全面的验证，与现有技术对比如下表4所示：The data represents different current and phase conditions with uniqueness, and is used to verify the performance of the fault location method of the present invention under different conditions, for example, the three-phase current amplitudes are 30A, 35A and 45A, the corresponding phase angles are 10°, 12° and 15°, and the zero-sequence current amplitude is 2A and the phase angle is 5°. The uniqueness of the above data ensures comprehensive verification, and the comparison with the prior art is shown in Table 4 below:
表4与现有技术对比表Table 4 Comparison with prior art
本发明相较于传统技术在配电网故障定位方面有明显优势，通过融合模型和云边端协同，减轻了服务器压力，降低了网络时延，精简了数据传输，提高了故障定位准确性，同时简化了技术实施难度。Compared with traditional technologies, the present invention has obvious advantages in locating distribution network faults. Through the fusion model and cloud-edge-end collaboration, it reduces server pressure, reduces network latency, streamlines data transmission, improves fault location accuracy, and simplifies the difficulty of technical implementation.
应说明的是，以上实施例仅用以说明本发明的技术方案而非限制，尽管参照较佳实施例对本发明进行了详细说明，本领域的普通技术人员应当理解，可以对本发明的技术方案进行修改或者等同替换，而不脱离本发明技术方案的精神和范围，其均应涵盖在本发明的权利要求范围当中。It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical solutions of the present invention, which should all be included in the scope of the claims of the present invention.
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410137739.0ACN118150938A (en) | 2024-01-31 | 2024-01-31 | Distribution network fault location method and system based on fusion model and cloud-edge collaboration |
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410137739.0ACN118150938A (en) | 2024-01-31 | 2024-01-31 | Distribution network fault location method and system based on fusion model and cloud-edge collaboration |
| Publication Number | Publication Date |
|---|---|
| CN118150938Atrue CN118150938A (en) | 2024-06-07 |
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202410137739.0AWithdrawnCN118150938A (en) | 2024-01-31 | 2024-01-31 | Distribution network fault location method and system based on fusion model and cloud-edge collaboration |
| Country | Link |
|---|---|
| CN (1) | CN118150938A (en) |
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN118535976A (en)* | 2024-07-24 | 2024-08-23 | 山东金煜电子科技有限公司 | Power distribution network fault analysis method and system based on Internet of things |
| CN119291392A (en)* | 2024-12-11 | 2025-01-10 | 国网冀北电力有限公司智能配电网中心 | A fast-locating distribution network fault management system and method |
| CN119596061A (en)* | 2024-11-20 | 2025-03-11 | 广东电网有限责任公司 | Fault state machine modeling method and fault processing method based on regularization method |
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN118535976A (en)* | 2024-07-24 | 2024-08-23 | 山东金煜电子科技有限公司 | Power distribution network fault analysis method and system based on Internet of things |
| CN119596061A (en)* | 2024-11-20 | 2025-03-11 | 广东电网有限责任公司 | Fault state machine modeling method and fault processing method based on regularization method |
| CN119291392A (en)* | 2024-12-11 | 2025-01-10 | 国网冀北电力有限公司智能配电网中心 | A fast-locating distribution network fault management system and method |
| Publication | Publication Date | Title |
|---|---|---|
| CN118150938A (en) | Distribution network fault location method and system based on fusion model and cloud-edge collaboration | |
| CN116245033B (en) | Artificial intelligent driven power system analysis method and intelligent software platform | |
| CN112215722B (en) | Dominant instability mode discrimination model construction method, dominant instability mode discrimination method | |
| CN116169675B (en) | Power system dynamic stability online evaluation method considering operation mode change | |
| CN108336739A (en) | A kind of Probabilistic Load Flow on-line calculation method based on RBF neural | |
| CN105354643A (en) | Risk prediction evaluation method for wind power grid integration | |
| CN109002926A (en) | The photovoltaic power generation quantity prediction model and its construction method of a kind of high accuracy and application | |
| CN110571926A (en) | intelligent power distribution network based on Internet of things technology and data model construction method thereof | |
| CN110910026B (en) | Cross-provincial power transmission line loss intelligent management and decision method and system | |
| CN115293454A (en) | Power system reliability rapid optimization method considering topological change | |
| CN118396366A (en) | Power grid operation risk prediction system and method based on graph convolution network | |
| CN113092934B (en) | Single-phase earth fault judgment method and system based on clustering and LSTM | |
| Shi et al. | A fault location method for distribution system based on one-dimensional convolutional neural network | |
| CN116821771A (en) | HVDC system fault diagnosis method and system based on ISTOA-CNN | |
| CN115118015B (en) | A power supply stability monitoring system based on fusion terminal | |
| CN113743460B (en) | Method, device and computer equipment for determining fault cause of power transmission line | |
| CN114996635A (en) | Power distribution network parameter determination method, device, equipment and storage medium | |
| CN119598248A (en) | Hydropower unit excitation power unit fault diagnosis method based on Attention-CNN and fault knowledge base | |
| CN118841949A (en) | Concentrator topology identification method based on SOM-BP, expansion module and concentrator | |
| CN118277808A (en) | DL-based high-proportion new energy power grid security risk assessment system and method | |
| Maharjan et al. | Probabilistic grid strength assessment of power systems with uncertain renewable generation based on probabilistic collocation method | |
| CN116090154A (en) | Distribution network topology and line parameter joint identification method, device and electronic equipment | |
| LU504569B1 (en) | A fast reliability evaluation method for the power system based on active constraints and its device | |
| CN120150122B (en) | Method and device for improving power system operation safety driven by model data hybrid | |
| CN116565872B (en) | AC power grid reliability rapid control method and device based on state similarity |
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| WW01 | Invention patent application withdrawn after publication | Application publication date:20240607 | |
| WW01 | Invention patent application withdrawn after publication |