| At prensent,more and more nonlinear loads(including power electronic converters)are connected to the power system,and these nonlinear loads inject the harmonic and reactive current into the power system.Because of China’s agricultural three-phase four-wire electric supply system and the particularity of power load,the problem of power quality is often coexistence of many factors that interact with each other,It is also necessary to use a variety of measures at the same time in power quality control.The control strategies and detection algorithms of conventional three-level power quality control devices are mostly designed on the basis of three-phase power supply voltage symmetry and neutral-point potential balance of diode clamp three-level topology.However,in practical applications,the power supply voltage may be unbalanced or distorted by the influence of non-linear loads,and the voltage frequency may also be distorted.Migration and jump may occur,and the neutral point potential will also be offset due to zero sequence current.These problems all lead to the unsatisfactory effect of the traditional power quality comprehensive treatment device in the three-phase four-wire power system.In this dissertation,the power quality control device in three-phase four-wire distribution system is studied.The main work and innovation are as follows:1)In order to overcome the limitation of traditional three-phase three-wire harmonic detection algorithm in three-phase four-wire system,an adaptive neural network harmonic detection algorithm based on master-slave Kalman filter phase-locked loop is proposed to measure harmonic current and other non-fundamental positive sequence components in the system completely and accurately.Firstly,the master-slave unscented Kalman filter is used to calculate the accurate frequency and phase of the power supply voltage,then the adaptive neural network method is used to detect the non-fundamental positive sequence current in the three-phase four-wire power system,and the reference compensation current is calculated.2)For three-level or even multi-level voltage inverters,an improved space vector pulse width modulation(SVPWM)method is proposed,which uses a non-orthogonal coordinate system KL and five-segment SVPWM to replace the traditional SVPWM strategy.Because complex operations such as trigonometric functions are avoided,the calculation of sector judgment and switching function execution time is greatly reduced,and the control strategy can be implemented quickly,with strong compatibility and versatility.3)Aiming at the inherent neutral-point potential drift of three-level voltage inverters with medium capacitance,based on the stability of DC-side voltage and the fundamental cause of the neutral-point potential drift,the redundant vector of short and medium vectors in multi-level inverters is proposed to adjust the voltage difference between the upper and lower capacitors of DC-side,and the DC-side neutral-point voltage is realized.Effective bit control.4)A current controller based on Finite Control Set-Model Predictive Control(FCS-MPC)is designed.Combining with the improved SVPWM modulation strategy,this algorithm can track the reference current signal effectively without querying all the switching states in the case of large number of switching states of multi-level voltage inverters.It can also reduce the switching frequency and loss while reducing the amount of calculation.At the same time,the proposed value judgment function incorporates the balance factor of DC side capacitor voltage balance,which can solve the problem of unbalanced neutral point potential to a certain extent.5)In order to verify the effectiveness of the proposed control strategy and algorithm,a prototype of power quality integrated control device is designed and completed.The relevant experiments of filtering,reactive power compensation and three-phase unbalance are carried out.The experimental results show the effectiveness of the harmonic detection method and control strategy. |
PDF Full Text Request