Collaborative Optimal Scheduling For Bus-Branch Flexibility Resources Of Power Systems

To meet the requirements of low-carbon and sustainable development for power systems,the penetration of renewable energy generation in the transmission network,distribution network and microgrid levels is increasing constantly,bringing increasingly prominent volatility and uncertainty to power system operating states.Resources with flexible response characteristics in generation and load sides play an important role in meeting the challenges posed by strong volatility and uncertainty for secure and efficient operation of power systems.Highly frequent dispatch of bus-injection type flexiblility resources in generation and load sides will significantly increase the overload risk of relevant transmission lines.To address this issue,the capability of transmission lines to flexibly adjust power transfer on them should be fully exploited to improve the utility of bus-injection type flexibility resources.Comprehensive utilization and synergy of bus-injection type and branch-transfer type flexiblility resources have important theoretical significance and application value for ensuring the secure and efficient operation of power systems integrated with high-penetration renewable energy.The theme of this thesis is collaborative optimal scheduling for bus and branch flexibility resources of power systems,which centres on an important branch flexibltility resource,i.e.,dynamic line rating(DLR)technology.The main work and research results are as follows.(1)A day-ahead forecasting method is proposed for dynamic rating of power system lines.To address the conservatism of using static rating to determine the upper limit of line transfer capability,the DLR technology is applied to acquiring the dynamic ampacity of lines by monitoring operating states and weather conditions online.The method of recurrent neural network with attention mechanism is utilizedto forecast the weather conditions along the lines.Given the forecasting results of weather conditions,the dynamic ampacity and capacity of lines are calculated according to the heat balance equation.The effect of the DLR technology on improving the flexibility of relevant lines in adjusting power transfer is analyzed.Simulation results verify the superiority of the proposed day-ahead forecasting method.(2)A planning allocation method is proposed for branch flexibility resources in power systems.Importance sampling and Monte Carlo sampling are respectively performed in accordance with the probability distributions of renewable energy generation and load demand.With probability distributions of injection shift factors for each branch-bus pair calculated,the method of Latin hypercube sampling is employed to obtain the probability distributions of power transfer on the lines.The overload risk of each line is evaluated by setting its static rating as the baseline.The DLR technology should be deployed on the lines with significant overload risk to improve its flexibility.Simulation results of a test system verify the feasibility of the proposed planning allocation method.(3)A day-ahead optimal scheduling method is proposed for bus and branch flexibility resources in power systems.With the advancement of electric vehicle and controllable load technologies,flexibility from the load side can be extracted by guiding electric vehicle charging behavior and aggregating charging power demand,which supplements the flexibility from the generation side to constitute bus-injection type flexiblity resources.By using the day-ahead forecasting method,dynamic capacity of the lines equipped with the DLR technology can be obtained as their thermal stability limit,which essentially relaxes power transfer constraints of the lines.A day-ahead optimal scheduling model is built for bus and branch flexibility resources in power systems with the objective to minimize the total dispatch cost.Constraints of the model mainly include AC power flow equations,capacity limits of bus-injection-type flexibility resources,dynamic capacity limits of the lines equipped with the DLR technology,static capacity limits of the lines without the DLR technology.The uncertainties of relavant parameters are depicted by multiple scenarios by discretizing the continuous probability distributions of the parameters.Simulation results of a test system verify the effectiveness of the proposed optimal scheduling method.