Font Size: a A A

Dynamic Characteristics Optimization Method Of Active Distribution Network Based On Synchronous Measurement And VSC Supplementary Control

Posted on:2022-02-20Degree:DoctorType:Dissertation
Country:ChinaCandidate:J HouFull Text:PDF
GTID:1522306815996799Subject:Electrical engineering
Abstract/Summary:PDF Full Text Request
With the integration of numerous renewable energy distributed generation facilities,distributed energy storage systems,and electric vehicle charging facilities into the distribution network,the distribution network changes from passive network into active network.And the structure and characteristics of the distribution network have changed fundamentally.Such changes bring great challenges to the operation optimization of the distribution network.Therefore,the concept of active distribution network(ADN),which is the distribution network with flexible topology using active control and active management,is proposed by experts and scholars to realize the efficient and optimal operation of the distribution network.In recent ten years,a lot of researches on ADN have been carried out at home and abroad,and fruitful results have been achieved.However,a large number of renewable energy distributed generation systems and energy storage systems are connected to the grid through voltage source converter(VSC)with its own controller.Also,affected by the uncertainty and intermittence of renewable energies,the dynamic characteristics of ADN become extremely complex.And the probability of adverse dynamic processes(frequent out of limit of node voltage and branch current,continuous or slow attenuation oscillation,etc.)of ADN is greatly increased.Therefore,if effective measures are not taken to improve the dynamic characteristics of ADN,the normal and stable operation of ADN will be threatened.In recent five years,the dynamic characteristics of ADN have been highly valued by scholars and experts at home and abroad,but the related research is still in its infancy.Against the above background,this dissertation focuses on the urgent need to solve the problems of bad dynamic processes of ADN,and explores the “Active Distribution Network Dynamic Characteristics Optimization Method Based on Synchronous Measurement and VSC Supplementary Control”.Based on the observability and controllability of hierarchically distributed physical system,this dissertation constructs a controlling system for ADN dynamic characteristic optimization and coordination.In order to provide new ideas and methods to improve the dynamic characteristics of ADN,this dissertation makes in-depth studies from the following aspects: the dynamic characteristics analysis of ADN,VSC supplementary controlling method based on current source injection(CSI),synchronous measurement technology for dynamic characteristic optimization of ADN,and artificial intelligence method for optimizing the dynamic characteristics of ADN based on the idea of double-layers coordinated control.The main research work completed and the corresponding innovative achievements obtained in this dissertation are as follows:(1)Based on the detailed mathematical models of VSC-based direct-drive wind power generation systems,photovoltaic power generation systems,and energy storage systems,the dynamic characteristics of VSC-integration system are deeply studied through eigenvalue analysis and time-domain simulation analysis,and the bad dynamic process of ADN with VSCintegration system is revealed.The time domain simulation model of an ADN in East China is established,and the dynamic characteristic simulation analysis of an ADN system in the demonstration project of the National Key Research and Development Program of China under Grant is carried out.The results show that weak damping and negative damping modes related to VSC controlling system in VSC-integration system might exist,and the voltage and power at VSC port will characterize the adverse dynamic process of continuous oscillation.In addition,the continuous oscillation of VSC-integration system will further propagate within the ADN and can take place on the branch lines of the distribution network,affecting the safe and stable operation of the ADN.(2)A CSI supplementary control method,which is suitable for high-power VSC in AND,is proposed to improve the dynamic characteristics of VSC-integration system.The controlling system structure and the design method of additional control link of the proposed CSI supplementary control are studied in depth,and the feasibility of this method is analyzed.By means of eigenvalue analysis,time domain simulation and closed-loop test of control hardware,the correctness and effectiveness of the proposed method are verified.The technical scheme of applying this method to the specific case of the demonstration project is introduced,and the wiring scheme of the CSI supplementary controller and the current source injection circuit of the additional control signal are designed in detail.The proposed CSI supplementary control method couples the three-phase AC additional control current signals to the current measurement loop of the VSC controlling system,optimizes the dynamic characteristics of the VSC-integration system by changing the dynamic process of the controlled object VSC.The proposed method discards the idea of applying additional control inside the VSC controlling system,significantly improves the implementation flexibility of additional control and provides a totally new method to improve the dynamic characteristics of VSC-integration system.(3)The synchronous measurement technology suitable for the optimal coordinated control of dynamic characteristics of ADN is studied,and a suspended energy harvesting synchronous measurement terminal(SEHSMT)with its own sensor is proposed.From the perspective of meeting the observability demand of ADN,the necessity of extensive synchronous measurement of ADN is analyzed.Based on theoretical analysis and time-domain simulation,the system structure of the proposed terminal,the technical scheme of sensor and recording,and the energy harvesting method based on current transformer are deeply investigated.The corresponding laboratory prototype is developed and laboratory tests are carried out to verify the effectiveness and correctness of the proposed SEHSMT.The terminal integrates the current sensor,power supply and the terminal body.The terminal is directly suspended and installed on the branch line of ADN.The terminal has the advantages of flexible installation,reasonable cost,and plug and play.Also,it meets the needs of low-cost and extensive monitoring of the dynamic process of ADN,and has outstanding engineering application value.The SEHSMTs deployed on a large scale and the existing micro synchronous phasor measurement units(μPMUs)of distribution network jointly monitor the operation state of ADN and record the dynamic process.Also,the SEHSMTs,μPMUs,and CSI controllers together constitute the bottom monitoring and control means of the optimal coordination control system for the dynamic characteristics of ADN.(4)Based on the bottom monitoring and control means,the upper controlling strategy of ADN dynamic characteristic optimization coordinated controlling system is studied in a deepgoing way.Two artificial intelligence methods for optimizing the dynamic characteristics of ADN are proposed.In order to solve the dynamic characteristic optimization problem under the condition of variable operation modes of ADN,based on CSI-VSC supplementary controller,an CNN-model training based on PSO acquisition samples(CTPAS)method to automatic optimize parameters of single CSI supplementary controller is proposed.A VSC adaptive supplementary controller(ASC)is implemented.In order to solve the problem of coordinated control of multiple supplementary controllers,based on the idea of double-layer coordinated control and a synchronous measurement system meeting the observability requirements,a coordinated control method of multiple supplementary controllers based on deep Q network(DQN)is proposed.In this dissertation,the principles and model construction methods of the above two methods are introduced in detail,and the effectiveness of the proposed methods are verified by numerical simulation.
Keywords/Search Tags:Active distribution network, Renewable energy distributed generation, Dynamic characteristic optimization, Supplementary control, Synchronous measurement, Coordination control, Deep reforcement learning
PDF Full Text Request
Related items