Research On Modeling And Integrated Control Strategy Of Active Power Filter

With the development of technology, application of the non-linear loads becomes more and more popular in power system, which brings serious harmonic contamination. Conventionally, inductance–capacitance (LC) Passive Filter filters have been used to compensate these harmonic currents and improve input power factor. However, they have many disadvantages, such as long response time, lacking stability, and the risk of resonances. Compared with the passive filter, active power filter can compensate harmonics more effectively, with faster response speed, better regulation characteristics and almost unlimited compensation capacity. Therefore, the Active Power Filter plays an important role in improving power quality. It is of great significance in increasing power system security, reliability and economy, and ensuring the normal operation of electrical equipment and efficiency in industrial & agricultural production.Reasonable and effective control strategy can improve the compensation performance of the Active Power Filter. Traditionally, the most commonly employed control approaches for time-domain controllers are current control techniques, such as linear current control, digital deadbeat control, hysteresis control, and so on. On the other hand, frequency-domain techniques are based on Fourier analysis and on the linear control using typical I and typical II Linearization method. Therefore, when APF system is described using traditional control method, there are accompany with the effects of long time delay, highly sensitive to the control parameter, difficult to realize and other negative factors. Moreover, unable to reflect the actual situation of the non-linear systems, thus the control precision is greatly reduced.In this paper, the mathematical model for Active Power Filter is built, and an integrated control strategy of Active Power Filter is proposed. The research presented mainly on the following four parts:Firstly, referring to extensive references, the principle of active power filter is mastered, the research situation, the advantages and the disadvantages of the APF are investigated.Secondly, based on the sufficient study on the principle and characteristics of the active power filter, the model is built based on switching function combined with rotating coordinate transformation (dq transformation). The switching function includes two parts: the steady state value and the fluctuation value.Thirdly, the advantages and disadvantages of the available control methods in active power filter are studied in-depth. In view of the nonlinear and uncertainty during the operation of the APF, a novel control method based on Lyapunov function theory is proposed. Through controlling the steady state value and the fluctuation value of the switching function, respectively, the compensating signal is obtained. The control method proposed here has the following advantages: Harmonic detection is simple and with less calculation. Traditional APF control method based on the dq Transform generally uses two PI controllers to eliminate the influence of coupling, then analyze and design the system with the linearization method; therefore, an accurate model of the operating principle is unable to be obtained. The proposed control method does not depend on circuit parameters and can achieve unity power factor compensation. In addition, coupling can be eliminated during the calculation, which simplifies the circuit structure. Considering inaccurate reference value of the system, meanwhile, the proposed average error indicator, the range of the control parameter is determined.Finally, the verification of the research is carried on by the simulation model of the APF built using MATLAB SIMULINK. Meanwhile, the experimental results are presented to verify the feasibility of the control method.