Research On Harmonic State Estimation And Location Of Harmonic Source In Distribution Network

With the increase of nonlinear load in distribution network,harmonic pollution is becoming more and more serious,which not only affects the quality of power,but also affects the economy,safety and smooth operation of power grid.In order to control harmonics,it is necessary to detect and locate the harmonic sources in the power grid.This paper uses the harmonic state estimation method to study the harmonic source location technology,and researches the quantitative estimation of the harmonic emission level,which provides a theoretical basis for the harmonic control,and has great significance for the safe and stable operation of the power grid.The harmonic source location method based on state estimation has integrity and high accuracy,which can be applied to the detection of multiple harmonic sources.According to the state estimation method,this paper proposes a harmonic source localization method based on orthogonal matching pursuit algorithm.Due to the sparsity of the distribution of harmonic sources in the system,the equation will be underdetermined,which makes it difficult to solve.To solve this problem,this paper uses compressed sensing method to realize compressed sampling and sparse representation of signal,and uses orthogonal matching pursuit algorithm to solve the problem,so as to realize the accurate location of harmonic source.This paper take the node harmonic current as the state quantity,and the branch harmonic current as the measurement quantity,estimates the harmonic current injected into the node,and realize the effective location of the harmonic source when the system is not globally observable to a certain extent.The simulation is carried out in the IEEE14 bus system.In the simulation analysis,changed the number and location of harmonic sources and considered the measurement error.The method is compared with the traditional least square method.The results show that the proposed method has high accuracy,can accurately get the distribution state of harmonic sources,and has good practicability and strong anti-interference ability.Based on the qualitative detection of harmonic sources within the space,complete the quantitative estimation of the harmonic emission level.This paper proposes a principal component linear regression method.First,sample the voltage and current signals at the common connection point,and then uses the principal component analysis method to establish a linear regression equation between the dependent variable and the principal component,and by solving the regression coefficient,obtains the harmonic impedance of the system side.And then according to the harmonic impedance find the harmonic emission level.Using Matlab software to simulate and analyze the method of this article.It turned out that the method can correctly calculate the harmonic emission level at the common connection point.And it has higher accuracy and more stability than the traditional binary regression method.Well,it provides a theoretical basis for the division of harmonic responsibilities.Finally,this paper designs a harmonic source location detection system based on method research,analyzes and designs the hardware architecture of the system,and completes the software development based on the Labview platform.The developed software processes the collected harmonic voltage and harmonic current data,calculates the injected harmonic current of the node and completes the estimation of the harmonic emission level,thereby realizing the positioning of the harmonic source.Last,this paper established an analog harmonic generation device in the laboratory to verify that the designed system has a better harmonic source location function,which has very important practical significance for the harmonic detection and control of the power grid.