Research On Frequency Recovery And Low Frequency Load Shedding Strategy Of Microgrid Considering Controllable Load

With the continuous development of the economy and society,people pay more and more attention to environmental protection,and various new energy power generation is increasingly becoming an important part of the power grid.Clean energy generation represented by distributed generators(DG)has become an important direction for the development of power grids in the future.DG can access the distribution network nearby,avoiding the power loss and instability caused by long-distance large-scale transmission,and providing guarantee for the strong stability of the power grid.In order to avoid the volatility and randomness brought by DG access to the grid,the concept of microgrid(MG)came into being.The microgrid is a small energy management system.DG accesses the distribution network in the form of microgrid.While making full use of new energy,it can effectively smooth and control the fluctuation of DG and reduce the impact on the distribution network.At the same time,when the distribution network fails,the micro-grid can be disconnected from the network to enter the island operation at any time,which ensures the continuous power supply to the load in the network and greatly improves the reliability of the distribution network.With the introduction and development of the smart grid concept,people have put forward new requirements for the stability of the power grid and high-quality power supply.Self-healing is the basic connotation of smart grid.After the fault occurs,the grid can autonomously eliminate the fault or adjust it to avoid the impact of the fault is becoming the basic requirement for the development of modern power grid.How to achieve self-healing after the failure of the microgrid connected to the distribution network,avoiding the disassembly and collapse of the grid is becoming a hot research direction.Under Frequency Load Shedding(UFLS)is the last line of defense to ensure the safety and stability of the system.It is the last guarantee to ensure that the power grid does not collapse.It plays an irreplaceable role in the self-healing of the power grid.At present,various low-frequency load shedding schemes in large power grids have been relatively mature,but there are few studies on low-frequency load shedding algorithms for microgrids.The microgrid is very different from the large grid,and it is impossible to transplant the mature load shedding scheme of the large grid directly to the microgrid.At the same time,the demand side response has received more and more attention.The microgrid has a small capacity and the rotating spare capacity is not sufficient.If the load side can actively participate in the response and actively adjust,the frequency fluctuation can be stabilized and the power imbalance can be reduced.At present,the controllable research based on the load of electric water heaters,air conditioners and electric vehicles is in the ascendant.How to choose the controllable load suitable for participating in the low-frequency load shedding of the micro-grid and make the controllable load fully participate in the low-frequency load shedding,becoming a hot topic nowadays.research direction.Based on this,this paper establishes a set of low-frequency load shedding schemes for microgrid with controllable load participation.This scheme can fully adapt to the difference between microgrid and large grid,and take care of residents’ comfort in low frequency load shedding.While restoring the task,you can try to avoid the impact on the normal use of residents.This paper mainly does the following work:1.This paper first analyzes the load characteristics of household appliances,analyzes and compares whether different household appliances can participate in low-frequency load shedding,and finally selects electric water heaters as controllable loads to participate in the low-frequency load shedding of microgrids.Then,a single electric water heater is modeled and analyzed for its electrical characteristics.At the same time,in order to facilitate centralized control of large-scale electric water heaters,this paper first analyzes the main parameters affecting the operating characteristics of electric water heaters,and then uses self-organizing feature map neural network.(SOM)clusters water heaters to group electric water heaters with similar parameter characteristics to facilitate unified control of the control center.2.In order to predict the use of electric water heaters,this paper builds a predictive model for the use of electric water heaters based on long and short time memory network(LSTM),which can predict the number of water heaters that are being used at each time for the next 1-2 hours.This calculates the capacity of the water heater that can be deloaded by the grid in the next 1-2 hours.Based on this,this paper proposes a PID control-based electric water heater to participate in the frequency adjustment model.The PID controller collects the grid frequency information.When the frequency is abnormally fluctuating,the water heater that is not used is deloaded at this time,and the frequency is restored to the normal value.The model was verified by Matlab/Simulink platform.3.Based on the comparison between the low frequency load shedding of microgrid and the low frequency load shedding of large power grid,this paper establishes an adaptive load shedding algorithm based on reinforcement learning,which can accurately know the exact inertia constant of the system generator.Through the intensive learning of the microgrid,the accurate load shedding strategy is obtained,which effectively solves the problem that the power characteristics of the microgrid are different and the precise parameters cannot be obtained,and the good load shedding effect is obtained.This paper discusses in detail the construction of the reinforcement learning model.Combined with the above content,a complete micro-network emergency control algorithm with controlled load participation is established to ensure that the micro-network can recover to the normal frequency value by self-adjustment in the case of isolated network operation and large-scale frequency fluctuations,avoiding the system frequency,collapse.Finally,this paper builds a typical microgrid model based on Matlab/Simulink platform,and simulates the above control algorithm to verify the effectiveness of the emergency control scheme.