Hierarchical Control Of DC Microgrids Under Link Failure

In recent years,the control of renewable energy resources(RES)in microgrids(MG)is a growing region of research.In order to improve the reliability of the MG control,a distributed consensus-based control is utilized by spreading a communication network across the small distribution grid.Communication network is used to transfer the information within the connected nodes to converge the DC MG system response.However,communication and electrical network link islanding,and failure of information components affect the load imbalance between generation source nodes and load nodes.This thesis mainly studies the algorithm of communication based distributed secondary level control of DC MG which can detect islanding and utilize the hierarchical control scheme for achieving the steady operation,or stability of distributed operation.Hierarchical control scheme involves primary,secondary,and tertiary level controls.Primary level control is implemented locally at every distributed generating node.Whereas,the secondary and tertiary level controls can be implemented in centralized,de-centralized,or distributed fashion.In comparison with the isolated individual control of converters in a MG,Consensus-based control in DC MG provide better voltage regulation and harmonized energy generation/consumption.Communication based distributed secondary level control is more extensively used in DC MGs compared to centralized control.A distributed secondary level control technique is studied in detail to detect the communication,electrical network link failure,and islanding scenarios.The novel study is proposed to detect and maintain the operation of the DC MG system during communication islanding and electrical islanding.Recently,although a variety of robust control schemes have been proposed to improve system reliability,the failure of the communication links and electrical network links in MG control level remained an unsolved issue that may lead to one or more nodes being disconnected and operating as an island that may be of communication or electrical network island.Such types of islanding may also cause the unpredictable behavior of the system,and further destabilization may lead to a total system failure.This thesis aims to increase the reliability of the operation of the microgrid by modifying the control structure in the form of controllable communication islands and physical electrical network islands during link failures.Whereas,conventional control lacks the ability to sustain-system operation during communication and electrical network islanding scenario.Conventional control fails during the link islanding scenarios,so the thesis studies the control algorithms for sustaining the operation and working of the DC MG.In the proposed communication islanding study,it reduces the requirement for tertiary level control and maintains the operation of the system at secondary level control.Whereas,electrical network study is utilizing the same communication network for detection of electrical network failure without adding any further complexity and cost.Therefore,better load sharing and voltage regulation between the nodes of MG are achieved through proposed thesis study.The main work and innovations of this thesis are as follows:(1)If communication link failure occurs in DC MG,it may affect the balance of the generation and load demand.In order to overcome this problem,a constant reference voltage control method is proposed and adaptive constant reference voltage control method is proposed.These methods detect and observe the communication link failures within small-scale distributed grids.If the link failure is detected,then it modifies the control structure in order to meet the load demand and maintain reliable DC MG operation.The proposed controls are effective in maintaining DC MG operation as compared to conventional consensus control during communication link failure scenarios.(2)Communication link failures lead the DC MG towards the unstable operation,which may further result in the form of cascaded failure of the DC MG system.In order to solve this problem,multi-node averaging filter based secondary level control is proposed,which aims to improve the resilience of the DC MG control.The proposed control algorithm estimates the average reference signals for the neighbouring nodes,so as to minimize the deviation in each node reference signals.This results in the form of load sharing and voltage regulation between nodes of DC MG.The proposed multi-node averaging filter control have the ability to keep the main control system unaffected by communication link latencies or failures.The proposed control is more effective in maintaining the working of DC MG as compared to conventional consensus control during communication link failures and latencies.(3)Communication islanding is studied in detail for fast detection of islanding of multi-level DC MG in the complex networks.Ping based Echo detection algorithm and Cutset based control algorithm are proposed to detect communication islanding and maintain its operation.Meanwhile,the detection is done using a look-up table generated at every node.The look-up table is generated by utilizing the information about the connectivity of the connected nodes exchanged between nodes of MG over the communication network.Further the study investigates the islanding/de-islanding control algorithm for the DC MG,which can detect any communication islanding and also can rearrange the system back to its regular operation after the communication is restored.For this study,the proposed islanding/de-islanding control is used for the rapid detection of islanding and modify the control accordingly.After the fault is cleared,the proposed control restores the control back to its regular operation.These proposed distributed control algorithms are more effective in comparison with the conventional consensus control of the DC MG during communication islanding for reliable operation.(4)This study examines the electrical network islanding in the DC MG network.If electrical link failure occurs in DC MG,it may affect the balance between generation and load demand.The proposed electrical islanding detection algorithm uses signal injection method with the communication network to detect the electrical islanding.The proposed algorithm maintains the operation by rapid recognition of electrical islanding,which can be used to manage the generating sources to meet the load demand.Moreover,the performance and the effectiveness of the method is tested by case studies.