The Research Of Substation Grounding Grids Performance And Its Fault Diagnostic&Imaging System

Grounding grids are installed in substation and play a crucial role in groundingsafety. IEEE/GB has established safety parameters and guidelines to regulate the designof substation and maintain its safety. Safety parameters include ground impedance,ground potential rise, step and touch voltage; ground impedance can protect secondaryequipment of substation from the damage of fault currents or lightning strike, and theother parameters can guarantee the safety of staff. When grounding grids were corroded,safety parameter would exceed the limited threshold and threaten the security ofsubstation. As a result of micro-cell and macro-cell interaction in soil, a lot of groundinggrids shows corrosion or fracture phenomenon in old substations and those in heavypolluted area; therefore, lots of corroded grounding conductors need to be replaced. Ithas important practical significance to develop fault diagnostic system to accuratelylocate fault sites in substation. In this paper, it comprehensively analyses the substationgrounding grids and develops fault diagnostic and imaging platforms based on theresearch of grounding performance and fault diagnosis: it derives and the establishes theinversion model of soil parameters based on multi-layered complex image method andthe earth parameters are estimated by improved particle swarm optimization; it proposesthat nodal neighboring relationship can be used to reconstruct the unknown topology ofgrounding grids; a fault diagnosising platform integrated by a software and hardwaresuite is developed and utilized in the analysis of Zhongguan substation.The main innovation of the paper include the following:①The horizontal multi-layered earth model established by complex imagemethod is to be solved by using the improved particle swarm optimization. To overcomethe low efficiency of image method in solving soil parameter estimation, it adoptspolynomial summation form of Green’s function to construct apparent resistivity andcomplex image method is adopted to deduce its fast formula based on four-pin method;because the procedures in deriving the recurrence formula is complex and the conditionnumber Jacobian matrix apears too large to converge in Newton-Raphson algorithm,IPSO is adopted to solve multilayer soil parameters. It can provides a higher accuracyapproach in the estimation of multilayer soil parameters.②In order to employ electrical impedance imaging technology solving faultdiagnosis of grounding grid, the pretreatment for its integrity is necessary. Neighboring topology is firstly introduced to reconstruct the connections of the unknown groundinggrids; two neighbor interconnection rules and three supplementary rules are set toanalyze three situations: enough ground deflectors, insufficient ground deflectors andpositional shifted ground deflectors; with the current injected into grounding grid, thealgorithm to evaluate the feasible solution is to minimize the root mean square error(RMSE) of the measured and computed nodal voltage.③Aimed at overcoming the ill-conditioned inverse problems of grounding gridderived based on graph-theoretic field model and electrical impedance imagingprinciple, it proposes Newton Raphson method restricted by Tikhonov regularization toimprove fault diagnosis. According to the particularity of the grounding grid, threemodifications are followingly made:1) the approach is proposed based on the principleof inclusion and exclusion to automatically select the excitation current;2) utilizingpartition approach, the medium and large scale resistor model of grounding grids iscarried out to locate and identify failure position;3) ground deflectors with the positionoffsetted are considered as new nodes which are inserted into the primary topology ofgrounding grid. This part can effectively improve the efficiency of the choice ofexcitation current, improve the accuracy of nodal voltage measurement and faultlocation, so that electrical impedance imaging can be implemented in fault diagnosis ofsubstation grounding grid.④The fault diagnostic and imaging platform of grounding grids is developedbased on the principle of electrical impedance imaging: Hardware system canaccomplish four-electrode16-channel auto-cycle measurements and provide32bithigh-precision nodal voltage data, to avoid the interference of contact resistance on thenodal voltage measurement and improve measurement efficiency and accuracy;software system can locate and identify fault branch of grounding grid, and fufill soilparameters estimation and topology reconstruction of unknown grounding grid. Usingthis platform, a comprehensive analysis of zhongguan substation is done, the conclusionaccords with the actual situiation.