| With the continuous advancement of the carbon peaking and carbon neutrality goals in China,as well as the rapid development of new energy technology and the constant upgrading and transformation of equipment,the distribution network is heading to the development orientation with“three high”characteristics,that is,high reliability,high initiative and high penetration of distributed power supply.However,when large-scale distributed power supply is joined up with the Active Distribution Network(ADN),it may trigger a series of problems such as fluctuating power grid load,reduced power quality,and increased operating costs.For instance,drastic change of power grid load,excessive voltage limit and other problems are caused by a large number of distributed photovoltaic power mutations.To solve the above problems,it is in urgent need to study the risk management and control measures after the distributed power supply is connected to the distribution network system.However,current researches fail to combine with field operational data and probe less into the management and control measures for weak links.In this regard,focusing on the safe and economic operation of the active distribution network,the risk control and management measures for the active distribution network under multi-smart terminal data fusion is carried out in this thesis from the aspects of data collection characteristics of multi-smart terminals of the distribution network,data fusion method,system and method assessing the comprehensive state of the active distribution network,risk management and control measures to improve distribution network reliability,etc.The major content of the study is listed as follows:(1)Various terminal data forms,data characteristics and application systems in the running process of the distribution network as well as the interaction situation of terminal system and data,etc.are explored and concluded,providing a multi-smart terminal data fusion scheme based on data center system,and verifying the advantages of data fusion by Weighted Least Squares(WLS)state estimation.Combining with the real operation data of the smart terminal of the distribution network,the study on the characteristics and influencing factors of distributed photovoltaic power generation under the active distribution network is accomplished and the overall load of the distribution network is divided into equivalent load and actual load.Meanwhile,the definition of the current trend change-over stage and change-over point of the transformer is proposed according to the return of transformers and lines caused by distributed photovoltaic,and the characteristics of the change-over point of the transformer current trend as well as the influence of power quality are studied.Finally,the new characteristics of the source-load behavior of the active distribution network under digital perception are analyzed.The analysis and research demonstrate that the large number of distributed power sources connected bring a series of stability problems such as load fluctuation and voltage limit to the distribution network,which need to be focused on more.(2)A method to assess the real-time state of the active distribution network is put forward on the basis of the data fusion on multiple smart terminals.Firstly,the system to assess the comprehensive state for the active distribution network which is different from the traditional assessment system is set up.The system fully integrates the smart terminal data and reflects the real-time system based on which the determination method of single index score on account of the Degree of Membership Function is proposed as the foundation for further study on the improvement of the objective weighting method.Then,the weighting method combining the subjective and objective based on the improved Maximum Diversification Portfolio Model is presented.The subjective and objective weighting method is optimized by the symmetry-scale subjective weighting method and the improved objective method,completing the overall design of the real-time state assessment method of the active distribution network.The symmetric scaling method improves the commonly used reciprocal scaling method,solving the problem of the default descending order arrangement of each index in the process of subjective empowerment and the problem of large index weight value variations.The actual calculated example has showed that the system and method to comprehensively assess state can effectively reflect the state of all aspects of the system.(3)A risk management and control model that can improve the reliability of the active distribution network is raised based on data fusion on multiple smart terminals.From the scratch,the equivalent load and photovoltaic power generation power of the active distribution network are predicted to reduce the actual load of the distribution network and effectively prevent the risk of power grid load fluctuation due to distributed photovoltaic.Next,a distribution network optimization model aiming at lowering the operating cost is established to optimize and adjust the access location and capacity of the distributed photovoltaic,hence lowering the operating cost while improving the operating power quality of the distribution system.Moreover,regarding the model optimization results as the basis,the weak degree index is deigned taking the smart terminal configuration into account and the method of identifying the weak factors and risk node line is put forward,which provides the adjustment direction for the optimal control of distributed photovoltaic.Finally,the actual operation of the distribution line is used to carry out example calculation,proving that the method proposed in this thesis manages to lower the operation cost of distribution network,improve its ability to deal with risks,and enhance its reliability in power supply. |
PDF Full Text Request