In order to improve the rate and accuracy of wind power forecasting, the Least-Square Support Vector Machine method (LSSVM) is presented. LSSVM adopts equality constraints and defines the least-square system as the objective function, which can simplify the forecasting method to a large extent, as well as accelerate the rate of wind power forecasting. Through the analysis of the original load data, a reasonable choice on training set and test sample set is made in the simulation. Besides, many factors, such as, the temperature, wind direction, wind speed and power previous, are taken into consideration. The result shows that LSSVM is more effective than that of SVM.

To make effective fault diagnosis of grounding grid , a new method using Self-Adaptive Particle Swarm Optimization (SAPSO) is proposed. Firstly, the grounding grid can be handled as a resistive network to establish fault diagnosis equations. Then the objective function based on minimum energy principle is added to lower the ill-condition of diagnostic equation. Next, according to optimization techniques, a new method of SAPSO is proposed to solve the corrosion diagnosis equations. The method takes advantage of the high global searching ability of SAPSO to obtain the optimal solution to the diagnosis model. By means of the analysis of the simulation, the correctness and reliability of the method have been verified.

Electronic current transformers are more suitable for the development of power system compared with traditional electromagnetic current transformers. Rogowski coil current transformer is one of three electric current transformers. According to the measurement principle of Rogowski coils, the equivalent circuit of PCB Rogowski coils is analyzed. By using four PCB Rogowski coils combined, a PCB Rogowski coil current transformer is designed and tested. The results show that the designed PCB Rogowski coil transformer has good linearity and high sensitivity and measurement accuracy and it can meet the requirement of power system.

With GIS being widely used, partial discharge detecting and defect pattern recognition become more and more meaningful and important. To realize defects identification of partial discharge map in GIS, a novel method based on Radical Basis Function (RBF) neural network is proposed. Firstly, a model is constructed to simulate the discharge pattern map by the use of random function randint. Secondly, based on the model above, a lot of data which meet the condition can be collected to provide for pattern recognition. Then, a RBF network is introduced to identify the pattern recognition. It can be trained by using the data above. Finally, through changing training error, high correct rate can be got. These indicate that the method is effective.

The assessment of the overload capacity of transformer has a certain practical significance. In this paper, a temperature reverse extrapolation method is proposed to assess the overload capacity of transformer. Firstly, the top oil temperature is monitored by the online monitoring system. Secondly, the temperature distribution model and the calculation methods of hot spot temperature in the PTP7 (Power Transformers. Part 7: Loading guide for oil-immersed power transformers) guide are analyzed. Then, a new method called temperature reverse extrapolation which can calculate the overload factor of transformer is composed. And based on the overload factor, two meaningful data about overload capacity are obtained. Finally, an assessment system of transformer overload capacity based on the online monitoring is developed.