Font Size: a A A

Research On Grounding Grid Topology Detection Based On Magnetic Field Method And Its Corrosion Diagnosis Method

Posted on:2022-04-27Degree:MasterType:Thesis
Country:ChinaCandidate:H P XieFull Text:PDF
GTID:2492306536466734Subject:Engineering (Electrical Engineering)
Abstract/Summary:
With the expansion of my country’s electric power scale,the electric power sector is facing a huge grounding grid detection task.In addition,with the overall advancement of UHV AC and DC lines and the gradual development of large cross-regional power grids,the effective grounding of power equipment will play a more important role in the safe operation of power systems.In order to ensure the safe operation of the power system in the future,the corrosion detection of the grounding grid has more important significance.my country’s substations have a very large base,and for the detection of grounding grids in old substations with a long operating life,the traditional excavation method is not only a heavy workload,but also the results of sampling testing are very blind and random.The current detection methods for grounding grids mainly include the method of detecting branch impedance based on electrical network theory.However,because old grounding grids often have problems such as inconsistent construction and design drawings,and inaccurate conductor positions,methods based on electromagnetic field principles can detect Conductor location.This paper combines the principle of electromagnetic field with the principle of electric network,exerts their respective advantages,uses the principle of magnetic field,proposes a method for detecting the topology of the grounding grid,provides preconditions for the electric network method,and then proposes a new grounding grid corrosion diagnosis based on the principle of electric network Method and verify its feasibility through experiments.The main contents of this paper are as follows(1)The basic principle of realizing the conductor position and depth detection of the grounding grid based on the magnetic field method is studied,the spatial magnetic field distribution of the conductor is calculated,and the peak value of the magnetic field is consistent with the position of the conductor,which verifies the feasibility of the principle of the magnetic field method.The conductor horizontal position and depth calculation method of the magnetic field differentiation method is proposed.Simulation and laboratory experiments verify the feasibility of magnetic field differentiation method for grounding grid detection.(2)A fast topology reconstruction method combining "circle method" and "paving method" is proposed,and the feasibility of this method is ve rified by simulation experiments.In the field conductor detection experiment of the grounding grid of the substation,the magnetic field information of the main branch of the grounding grid is detected by the magnetic field measuring device,and the local structure of the grounding grid is obtained by differential processing of the magnetic field information.It also illustrates the effectiveness of this method.(3)The corrosion diagnosis method based on the electrical network theory is studied,and the fault diagnosis model considering the multi-channel cyclic measurement method is established,which makes full use of the useful information of the reachable nodes.By comparing the grounding grid impedance inverse problem model with the neural network solution weights,a small batch gradient descent solution algorithm for grounding grids is proposed,the solution steps are proposed,and the robustness of the algorithm is verified.A method of introducing a greedy strategy and adding resistance constraints to reduce the degree of under-determination of the equation is proposed,and the simulation data is substituted into the hybrid algorithm for secondary calculations to obtain the detection data with good accuracy.At the same time,the reliability of the mixe d algorithm for corrosion diagnosis is verified by experiments by building a laboratory simulation grounding grid.
Keywords/Search Tags:grounding network, magnetic field method, differential processing, electrical network theory, mini batch gradient drop
Related items