Research On Three-phase Imbalance Control Of Distribution Network Based On Non-invasive Load Monitoring

The low-voltage distribution network is the basic unit of interaction between the power grid and users.With the improvement of people’s living standards,the amount of electrical appliances used by residential users have increased sharply.The random single-phase load switching has caused large fluctuations in the three-phase imbalance of the low-voltage distribution network.These phenomena not only affect the normal operation of the power grid,but also reduce the user’s comfort in electricity consumption and bring many problems.In view of the above problems,this thesis studied the three-phase imbalance control system,and realized the redistribution of the load on the three phases A,B,and C by adjusting the difference of the residents’ loads,so as to achieve the purpose of controlling the three-phase imbalance effectively.This thesis mainly studied the three-phase imbalance control problem based on nonintrusive load monitoring technology in low-voltage distribution network.Preliminary work introduced the research status of non-intrusive load monitoring technology and three-phase imbalance control methods respectively.And the harm of three-phase unbalance and the calculation method of three-phase unbalance were explained.In order to ensure the comfort of residents during commutation,an optimization method for obstacle avoidance commutation strategy was proposed.It was proposed to embed the non-intrusive load monitoring technology into the commutation strategy optimization problem to identify the operating status of the sensitive load of each residential user,and avoid commutation for users in the use of sensitive loads.In this way,more humanized control can be achieved.In the research of non-intrusive load monitoring technology,load identification is divided into two processes: load event identification and device state identification.The MK change point detection load event recognition algorithm based on double sliding windows and the load state recognition algorithm based on the weighted two-parameter HMM were proposed.In the state recognition,the occurrence probability of each device’s internal state in each time period was taken as the weight value into the algorithm.Through experimental analysis,the nonintrusive load monitoring algorithm proposed in this thesis can identify the user load more effectively with certain feasibility.In the research of commutation strategy optimization method,a commutation strategy optimization method based on improved multi-objective whale optimization algorithm was proposed.In order to solve the problem that the algorithm converges slowly and falls into the local optimum easily,the algorithm was improved in three aspects.The commutation strategy was optimized under the premise that the non-intrusive load monitoring technology generated the user’s sensitive load state penalty vector.The results showed that the commutation strategy optimization method in this thesis can optimize the commutation strategy effectively when the three-phase imbalance was controlled,and avoided the commutation operation for users who were using sensitive loads,and has certain feasibility.Based on the above research,a three-phase imbalance control system was designed.This thesis introduced the structure design and function of the system,and used UML modeling language to model the system.Finally,the system based on Java language on Myeclipse software was designed.Through experimental verification and result analysis,the three-phase imbalance control system proposed in this thesis can reduce the three-phase imbalance degree effectively.At the same time,it can also ensure the comfort of the residents and throw a light on the research of the control of three-phase imbalance.