Research On Multi-time Scale Photovoltaic Stealing Identification Method

With the continuous rise of new energy power generation industry,large and small distributed photovoltaic power plants have been built into the grid in recent years,which has greatly promoted the rapid development of China,s photovoltaic industry.Although China's distributed photovoltaic power generation projects are developing rapidly and have potential,the problems that come with it are becoming more and more obvious.Since the implementation of the subsidy policy for photovoltaic power generation,some unscrupulous users have used the means of technology to make distributed photovoltaic meters record more power generation,so as to defraud high subsidies from the state subsidy policy.This kind of fraudulent subsidy is called photovoltaic power stealing.This not only causes waste of state subsidies,but also brings potential safety hazards to the operation of photovoltaic power plants,which greatly affects the normal development of the photovoltaic power generation industry.Considering the emergence of new methods of stealing electricity,power companies urgently need a comprehensive distributed photovoltaic power stealing supervision system.Therefore,this paper aims to ensure the green and healthy development of the grid photovoltaic industry,aiming at monitoring the illegal fraudulent profit behavior under the implementation of photovoltaic subsidies,and proposes a multi-scale photovoltaic stealing identification method based on the collaborative prediction model.Firstly,several external environnental factors that have a great influence on the photovoltaic output are screened out.The historical power generation data is used for analogy analysis to obtain the weight of the weather type coefficient.The photovoltaic output synergy prediction model is constructed to predict the photovoltaic power generation value,and whether the error meets the tampering screening threshold is analyzed.Secondly,for the distributed PV user data processing analysis,a three-layer photovoltaic stealing screening framework is established,and the determination indicators of photovoltaic stealing behavior are defined,which provides a normative definition for the entire photovoltaic stealing identification process.The decision process for real-time,short-term,medium-and long-term suspicion of photovoltaic power severing is also proposed,as well as qualitative considerations for the severity of steering by PV users.The paper studies the problem of fraudulent compensation in distributed photovoltaic power generation,and adopts quantitative and qualitative analysis to establish a more complete supervision process.As a new type of tampering mode,the future may be widely promoted by unscrupulous users.In order to maintain the normal order of the new energy market and the interests of the state and power companies,anti-stealing work will be imperative.