Applied Mechanics and Materials Vols. 448-453

Abstract: Power load prediction is an important task for the electrical power system. The nonstationary, nonlinear and volatile characteristics of power load data make more difficult for the accurate load prediction. This paper presents a hybrid forecast algorithm based on wavelet transform and support vector machines for power load prediction. The hybrid algorithm firstly decomposed the load series to several subseries with obvious tendency by wavelet transform. Then these subseries are forecasted with least square support vector machines (LS-SVM), an extension of standard support vector machines, respectively. Finally these forecast results were reconstructed as the prediction of original power load series. The effective simulation results of above algorithm were testified based on a sample load series.

Abstract: In recent years, improved three-ratio is an effective method for transformer fault diagnosis based on Dissolved Gas Analysis (DGA). In this paper, a simple dynamic neural network named as diagonal recurrent neural network (DRNN) is used to resolve the online fault diagnosis problems for oil-filled power transformer based on DGA. Because of the characteristic of improved three-ratio boundary is lack of matching, fuzzy logic in fault diagnosis is presented also to deal with the data of the neural network inputs. DRNN is used to model the fault diagnosis structure, the fuzzy logic is used to improve the faults diagnose reliability. In addition, some cases are used to show the capability of the suggested method in oil-filled power transformers fault diagnosis.

Abstract: As the number of wind generation facilities in the power system is fast increasing, the research on available transfer capability (ATC) calculation with wind farms has great significance to system operation. With consideration of the uncertainty of the wind powers output, this paper proposes a probability computing method to study the ATC in wind power integrated system. This computing method of ATC is evaluated on non-sequential Monte Carlo simulation, and the ATC of every system state in random sampling is calculated by interior point method. The result shows that the model and algorithm is correct and effective.

Abstract: Taking into account the demands of multi-function meter with high reliability and low cost, a single-phase multi-function energy meter is designed based on the MSP430FE427. There is a multi-parameter measurement, multi-way communication and other functions for the MCU. Hardware circuit and software flow of the system are introduced. The current and voltage design, memory design and communication anti-interface design are described in detail in hardware circuit. Modular design scheme is adopted in the software design. Each module is relatively independent, so the program is easy to modify. It indicates that the energy measurement circuit owns high accuracy through analysis and comparison of test results.

Abstract: With the scale of the wind farm growing fast, its impact on the power system has become increasingly apparent. So the research has a significant meaning on the characteristics of dynamic stability of the power system which contains wind farms, and the stable operation of the large area interconnected power grid. In this paper it realized the application of the double-fed wind turbine grid model by using power system analysis software PSD-BPA. The analysis of the generator power Angle curve which indicate the state after the failure of N-1 shows as follows: with the wind farms integration on the grid, the damping ratio is decreased slightly but little change after the system failure of N-1, which is still live up to the standard of grid stability.

Abstract: In order to improve the control effect of the generator excitation controller, this paper proposes an excitation control based on objective holographic feedbacks and extended state observer. First, nonlinear factors of the nonlinear equation is switched to the state equation which contains the control input factors by objective holographic feedbacks. Then, constructing a second extended state observer tracks and compensates uncertain disturbance, simultaneously reduces system parameters of control law. Finally, MATLAB is used for simulation, the results show that nonlinear excitation controller that this paper designs can not only improve static and transient stability of system, but also improve the precision and stability of the terminal voltage control better.

Abstract: Although numerical simulation is an important method of researching dynamic frequency process, obvious deviations have been found between numerical simulation and the measured trajectory in many accidents. And the existing simulation model and parameters cannot describe the actual dynamic process of frequency accurately. Research was carried out on the influence of four parameters to the dynamic frequency process, which based on the WSCC system. The four parameters include the inertia constant of generator, generator frequency coefficient, dead band and turbine intermediate superheating coefficient. Northeast China power grid and measured data are used to verify the above research conclusion. Checking the dynamic frequency process simulation model and parameters can improve the accuracy of dynamic frequency process simulation on the base of the measured trajectory and the physical characteristics of the parameters. It can also give efficient foundation for the setting work of UFLS, overcoming the previous conservative operation mode and so on.

Abstract: A new kind of parallel processing unified power quality conditioner with only one inverter is introduced. The system PI control method has a problem that the DC transient component of supply current does decay because of the linking inductance between power supply and load. A system control method based on state variable feedback is proposed, which indirectly controls the load voltage by controlling the supply current. The proposed control strategy is verified by simulation using Matlab.

Abstract: The third frequency adjustment is the key link of the grid frequency regulation. The admission of high capacity and rejection of small capacity policy on thermal power cause the lack of capacity of the third frequency adjustment. Wind power brings about a certain impact to the frequency stability of the power system, particularly, increasing the difficulty of the third frequency adjustment. The introduction of nuclear power into the third frequency adjustment has been proposed after analyzing the characteristics of the energy generation of different forms. This paper formulate the control strategy with the core of wind power decomposition, generator units grouping in the third frequency adjustment, after elaborating the technical feasibility and economic feasibility, and, come to a conclusion that the participation of the nuclear in the third frequency adjustment would guarantee the stability of the frequency of the power system with this control strategy in the case of the introduction of wind power to the system, by simulation analysis.

Abstract: As the power grid blackout accidents occur frequently, the vulnerability analysis becomes an important topic in power system analysis. A new electrical dissection method for the power grid vulnerability assessment is presented in this paper. The method can effectively reflect the power transmission between generator and load on the terminal of each line, and quantify the using relationship with the branch transmission power limit. An IEEE 39-bus text system was used to testify this method and the results indicated the methods efficiency and rationality. This paper is supported by NSFC 51307019.