Research On Key Technologies Of Power Distribution Network Operation Evaluation And Performance Control Based On Data Driven

In order to implement the digital transformation of power grid,the application of digital technology in the development of traditional power grid operation is an important way to support the construction of new power system as well as achieve "carbon peak" and "carbon neutralization" goals.The importance of power distribution network grows with the construction of new power system.More attentions have been devoted into the evaluation and performance control of power distribution network by data driving.However,some practical problems exists such as data management belonging to different departments,data analysis and mining disconnected from the operation mechanism of power grid equipment,researches limited to simulation analysis.As a result,the correlation of data sets is difficult to be fully explored.The effective utilization rate of data resources is low.The integration of data and services is insufficient.This dissertation focuses on power distribution network equipment evaluation and diagnosis based on data driving and improve the quality of power distribution network operation.The main contents are as follows:1.Research on key technologies of power distribution network equipment evaluation and rapid fault diagnosis is carried out.At present,the type of equipment status evaluation index is too simplistic to fully describe equipment characteristics and the weight obtained by the evaluation method is difficult to adapt to the deterioration of indicators.To deal with these problems,this dissertation proposes a method of power distribution network equipment state evaluation based on subjective and objective adaptive weighting.The objectivity of weight and adaptability to deterioration indexes are guaranteed by fuzzy triangular number weighting method based on subjective experience of experts and adaptive variable weighting method based on objective data respectively.The state evaluation of overhead transmission lines in power distribution network is realized based on this evaluation method.This dissertation proposes a rapid diagnosis method for wildfires in power distribution network transmission corridors based on characteristic signals and temporal and spatial distribution characteristics of wildfires to deal with remote diagnosis of power distribution network transmission corridor wildfires.The feasibility of rapid diagnosis model of power distribution network transmission corridor wildfires is verified by the mechanism of overhead transmission line wildfires and warning signs.Through the regression analysis of the temporal and spatial distribution characteristics of historical wildfires and the mining of characteristic signals based on the Apriori association rule algorithm,combined with the status evaluation results of power distribution network overhead transmission lines,the sectional diagnosis of power distribution network transmission corridor wildfires is realized.2.Research on key technologies of power distribution network operation quality evaluation and rapid fault diagnosis is carried out.The power distribution network operation quality evaluation has the characteristics of high frequency change of power flow,multiple operation scenarios and wide professional involvement.Therefore,based on the research results of power distribution network equipment state evaluation,this dissertation proposes a quality evaluation method of power distribution network in multi-operation scenario based on entropy weight and Bayes.To deal with the quality assessment of power distribution network multi-operation scenarios,multi-scenario evaluation system and entropy weight-Bayesian hierarchical analysis are constructed to assign weight.The characteristics of high frequency change of power flow and wide range of specialties can be satisfied.Aiming at the most frequent trip fault in power distribution network fault diagnosis,this dissertation proposes a rapid diagnosis method of trip fault based on biological sequence alignment.The classification decision analysis of power distribution network fault diagnosis is realized by establishing trip sequence template and biological sequence comparison algorithm,the rapidity of grid trip fault diagnosis is realized,and combined with the evaluation results of power distribution network operation quality.3.Research on regulation and control of power distribution network operation quality is carried out.On the basis of power distribution network operation quality evaluation results,in order to further optimize and improve the operation quality of power distribution network,the distribution network of a mega-city is taken as the research object.This dissertation proposes a power distribution network operation quality regulation control method based on surrogate model.To deal with the problem of adjusting only for a single grid operation index,the 110 kilo-volt distribution network is included in the proposed algorithm of power distribution network operation quality control,and the operation quality of the power distribution network is improved through the reconstruction of the topology of the 110 kilo-volt distribution network.4.The big data platform of power distribution network is designed and implemented.The existing research results of power distribution network equipment evaluation and diagnosis mostly stay in theoretical analysis.There is a lack of platform to realize real-time state evaluation and fault diagnosis of power distribution network to truly improve the operation of traditional power grid.The big data platform of power distribution network equipment is designed and constructed by combining power grid dispatching data network and data cloud.The platform is deployed at provincial and regional levels.Cross-region data interaction technology and distributed parallel processing technology are adopted to achieve real-time data acquisition in the production control area and management area of power grid.The platform is divided into four layers,including the data access layer,the storage layer,the analysis service layer,the function application layer.Two functions of the power distribution network equipment status evaluation and fault diagnosis,the power distribution network operation quality evaluation and fault diagnosis are realized.