Detection and localization of partial discharge are very important in condition monitoring of power cables, so it is necessary to build an accurate recognizer to recognize the discharge types. In this paper, firstly, a power cable model based on FDTD simulation is built to get the typical discharge signals as training samples. Secondly, because the extraction of discharge signal features is crucial, fractal characteristics of the training samples are extracted and inputted into the recognizer. To make the results more accurate, multi-SVM recognizer made up of six Support Vector Machines (SVM) is proposed in this paper. The result of the multi-SVM recognizer is determined by the vote of the six SVM. Finally, the BP neural networks and ELM are compared with multi-SVM. The accuracy comparison shows that the multi-SVM recognizer has the best accuracy and stability, and it can recognize the discharge type efficiently.
To study the effect of fog-haze on the ionized field of high voltage direct current(HVDC) transmission lines, we proposed a computation method for the effects of fog-haze on the ionized field of HVDC transmission lines. Taking both the charging effects of haze particles and the influence of humidity on corona inception into account, we computed the field strengths and ion current densities in the vicinity of the ±800 k V ultra HVDC transmission lines under different haze pollution degrees using a finite element method. The results show that, in comparison to the values in fine condition, the total field strengths and ion current densities near ground increase with the pollution degree of hazy weather, and the growth rates of the respective maximum values for them are in a behavior of linearly increasing with respect to the pollution level at high pollution concentrations. Furthermore, the change of the ionized field with the pollution degree of hazy weather is more obvious for high humidity than that for the low one or for dry haze situation. The study suggests that the effect of humidity on corona inception and the charging of haze particles are the main reasons for the change of electrical field strengths and the ion flow densities near ground, and the decrease in ion mobility restricts the growth of ion flow densities under hazy weather. The study provides a method for analyzing the corona effect on HVDC transmission lines.
GIS has the advantage of compact structure, small size and high reliability, and now it is widely used in power system. But partial discharge (PD) in GIS will cause further deterioration of insulation and finally cause the huge losses of power system. So it is necessary to detect the PD signals in GIS. In this paper, firstly FDTD method is used to create models of GIS and calculate the propagation characteristic of electromagnetic wave inside the cylinder and draw the conclusions that the energy of PD signals is focused on GHz. So it is practicable to monitor the PD using the UHF method. Then PD monitoring system for GIS based on UHF method is introduced. The monitoring system can detect and monitor the real-time PD peak values and phase positions, and finally it calculated the spectrograms such as N-φ, H n (φ) and φ-Q-N. Using the spectrograms one can get the level of PD and then take the measures to handle the fault.
The distribution characteristics of space charge in the axial and radial direction of the cable at the stress cone were tested in paper. The 10kV XLPE cable was aging under electrical-thermal aging to explore the rule of the space charge distribution along the electric heating aging and the formation mechanism of the space charge accumulation at the stress cone. In the axial direction, the position of the maximum space charge density is at the inner edge of the cable stress cone, about 2cm. With the increase of aging time, space charge density at the stress cone is gradually increased and extend range. The accumulation of space charge from the outside to the stress cone is the cause of the space charge accumulation. The study of space charge accumulation at stress cone can help discover the essential explanations for the failure of cable accessories. It has practical significance for prolonging cable life and effectively diagnosing cable faults.
This study presents a combined model based on the exploratory factor analysis (EFA) and the least square support vector machine (LSSVM) to predict the contamination degree of insulator surface. Firstly, EFA method is utilised to reduce numerous influence factor variables of the insulator contamination into a few factor variables, which could decrease the complexity of the model. Then, regarding the above factor variables as new input variables, LSSVM model is established to predict the insulator contamination degree. In order to obtain the optimal predictive value, the non-dominated sorting genetic algorithm II is applied on the optimization of LSSVM model parameters. The proposed EFA-LSSVM combined model is compared with the models of LSSVM, back propagation neural network, and multiple linear regression on the model performance. Results indicate that the EFA-LSSVM combined model in this study effectively overcomes the shortcomings of the other three models mentioned above in computational time, prediction accuracy and generalization ability. Finally, the feasibility of the proposed model in predicting contamination degree of insulator surface is verified by adopting the radar map of the evaluation indexes of model performance.
The converter transformer is the most important piece of equipment in ultrahigh voltage direct current (UHVDC) transmission systems. The rectifying valve side of the converter transformer is dealing with composite voltages, containing both ac and dc components. Therefore, the characteristics of the partial discharge (PD) in oil-paper insulation of such rectifying valves are different from those of rectifying valves subjected to pure ac or dc voltage. In this study, theories of hydrodynamic drift-diffusion and bipolar charge transfer were used to build the needle-plate electrode model with a paperboard depression, and experiments were used to verify the feasibility of the simulation method. The electric field and net charge distribution that can reflect the PD were analyzed. Besides, the effects of instantaneous voltage, the proportion of ac-dc components, and depression parameters on the development characteristics of PD were also discussed. Simulation results show that applied voltage, interface charge density, and model parameters can all affect the electric field strength, which in turn influences the development of PD in the depression. This research reveals the process and influencing factors of PD during the electrical aging process, which not only helps to optimize the oil-paper insulation at the rectifying valve side of the converter transformer but also provides a basis for further explaining the PD mechanism.
A support vector regression (SVR) based on particle swarm optimization (PSO) is proposed to estimate the top oil temperature of transformer. This model establishes SVR model based on sample data such as ambient temperature, transformer load, and top oil temperature of transformer. The model analyzes the relationship between the top oil temperature of transformer and other factors, establishes the support vector hyperplane according to different influencing factors, and limits the prediction of the top oil temperature of transformer to a reasonable interval. According to the choice of penalty factor and relaxation factor of support vector machine, the error between this area and the actual oil temperature of the top layer of transformer is minimized, and the top-oil temperature prediction model has the highest prediction accuracy. PSO is used to optimize the penalty factor and relaxation factor in SVR model. The kernel function is improved by principal component analysis to optimize the support vector regression model. Compared with particle swarm optimization(pso) support vector machine(SVM), which considers the weight of data feature quantity, the prediction accuracy is higher.This model uses the advantages of support vector regression method, such as not requiring a large number of samples, not involving probability measure, and being able to deal with multi-dimensional influencing factors, etc.It can provide accurate top-oil temperature prediction results in case of insufficient short-term prediction data of transformer oil temperature or more dimensions of oil temperature related data collected.
'/%3e%3cuse xlink:href='%23a' transform='rotate(23.036 .093 25.536)'/%3e%3cuse xlink:href='%23a' transform='rotate(45.87 1.273 16.18)'/%3e%3cuse xlink:href='%23a' transform='rotate(69.945 .996 12.078)'/%3e%3cuse xlink:href='%23a' transform='rotate(20.66 -19.689 31.932)'/%3e%3c/svg%3e)
'/%3e%3cuse xlink:href='%23a' transform='rotate(144 450 300)'/%3e%3cuse xlink:href='%23a' transform='rotate(216 450 300)'/%3e%3cuse xlink:href='%23a' transform='rotate(288 450 300)'/%3e%3c/svg%3e)