Study On Anti-disturbance Methods And Their Application Technologies For Separating And Identifying AC Electric Signals Coupled By Heterogeneous Magnetic Circuits

Supervisory Control and Data Acquisition System(SCADA)is a key and essential facility for modern large-scale industrial systems.However,the data set collected by the traditional SCADA mainly includes the operation states of critical nodes and links,is challenging to satisfy the practical demands.Nowadays,the data acquisition platforms support many advanced applications such as operation monitoring,optimal scheduling,status assessment,trending early-warning,accident handling,fault diagnosis and equipment lifetime management,etc.Those new functions demand data acquisition platforms based on the conventional SCADA systems and the continuously expanded data acquisition subsystem with specific functions.Furthermore,from the perspective of evolution and development trends,the modern large industrial systems are updating with the usage of the new hardware devices and the application of advanced software technologies such as artificial intelligence and big data.Those further require the data acquisition sub-systems with higher acquisition density,more comprehensive coverage,and higher instantaneity synchronous acquisition functions,along with the new sensors and monitoring terminals operated under harsh environments,such as the significant harmonic distortion and high electromagnetic interference issues that existed in the high-power power electronic devices.Based on the background mentioned above,this dissertation investigates anti-disturbance methods for separating and identifying AC electric signals coupled via heterogeneous magnetic circuits.It proposes new solutions for typical engineering applications,i.e.,splitcore current transformer for multiple scenarios application in power systems and electromagnetic mold level detection system under mold electromagnetic stirring operation in a steel factory.Those researches are motivated and based on the engineering demands of one National Key Research and Development Program of China and one industrial project.The main contributions are summarized as follows:(1)Online identification of the signal transfer and conversion system parameters with single-frequency injected AC signal and variable heterogeneous magnetic circuit properties: based on the superposition principle of a linear system,a novel solution is proposed to obtain the signal transfer and conversion system parameters via injecting a controllable and measurable auxiliary signal with the frequency same as the measured signal’s frequency.This method converts the measurement system from a passive measurement system to an active one.For the split-core current transformer,online selfcorrection methods and the corresponding key technologies are presented.Also,aimed at the application of the larger-scale current monitoring of low-voltage lines in smart distribution networks,an application mode of the proposed method with a shared auxiliary winding loop and dada acquisition channel is provided to further reduce the application cost and raise the applicability of the larger-scale current monitoring of low-voltage lines in smart distribution networks.First,the operation principles and the implementation of the proposed method are studied.Then a prototype is made and tested,and the laboratory test results show the accuracy of the prototype achieves 0.2 class.(2)Online identification of the harmonic signal transfer and conversion system parameters considering distributed parameters and variable heterogeneous magnetic circuit properties: the effects of the distributed parameters,the non-linearity of the iron core,and the uncertain change of the gap length of the core on the harmonic transfer and conversion parameters of the split-core current transformers are analyzed,and the transfer model of the split-core current transformer is constructed.Then an online identification method of the harmonic transfer parameters of the split-core current transformer with an auxiliary winding is proposed,based on the injection of auxiliary square wave current signals.The proposed method also coverts the split-core current transformer to an active measurement system.Finally,the proposed method’s operation principle is theoretically analyzed;a prototype is implemented and tested in the laboratory.The test results show the significant effect of the proposed method on improving the harmonic measurement accuracy.Moreover,to meet the harmonic monitoring demands in the modern power systems with hybrid AC/DC transmission and the large-scale application of power electronic devices,a mounted harmonic current monitoring terminal with self-harvest and wireless communication is designed considering the economic and technical rationality and the applicability.(3)Fast separation and identification of the subsynchronous oscillation current based on signal transfer and conversion system with variable heterogeneous magnetic circuit properties: under the circumstance that subsynchronous oscillation keeps occurring due to the large wind power integration,a subsynchronous oscillation current identification method based on the state information of a frequency-sensitive circuit is proposed,aimed at overcoming the shortness that the conventional identification methods cannot ensure the accuracy and the response speed at the same time.Based on the observability principle of linear systems,the proposed method adds a frequency-sensitive circuit with a stage number not less than the number of the independent modes and then constructs an identification model based on the state information of the frequency-sensitive circuit.According to the identification model,a synchronous oscillation current identification algorithm is provided.The operating principle of the proposed algorithm is derived and explained theoretically;then,a prototype is implemented and tested via digital simulation and laboratory tests.The test results verify the accuracy of the proposed method to separate and identify the synchronous oscillation current.(4)The strong interference cancellation for active measurement system based on signal transfer and conversion system with variable heterogeneous magnetic circuit properties: to solve the issues that the electromagnetic mold level detection system usually fails under the mold electromagnetic stirring devices operating condition,this dissertation proposes a strong in-band interference cancellation method for active measurement systems.The proposed method contains three key parts: a coordinated modulator,an analogue narrow-band filter,and a coordinated digital filtering algorithm.Wherein the coordinated digital filtering algorithm utilizes the symmetry of the odd/even function waveforms,overcoming the effects of the estimating error of the interference signal frequency and then improving the identification accuracy of the measured signal.Compared with conventional approaches,the proposed method can effectively remove the effect of the strong in-band interference and can also ensure the instantaneity of the measurement system,leading to higher engineering applicability.The mathematical principle to cancel the in-band interference is derived and introduced in detail.Also,the cooperation of the key parts of the proposed method is shown.Moreover,an implementation scheme is designed for the real industrial application,and the simulations and laboratory tests for the prototype are carried out.The test results demonstrate that even under an extreme condition with SNR below-40 d B,the proposed method effectively removes the strong in-band interference with the frequency close to or equal to that of the measured signal and can also accurately extract the measured signal.