Study On Optimal Operation And Management Of Distributed Energy Resources And Energy Storage Resources In Microgrid

With the continuous growth of the demand for electricity, centralized power grid, as a primary transmission, distribution network, also will be rapid development. However, the traditional power grid has some disadvantages, such as long construction period, high investment costs and instability while promoting economic development continuously. To solve this problem, the concept of microgrid, the integration of a variety of distributed power, load and storage devices, was proposed.As an independent autonomous management system, microgrid can achieve large scale access of Distributed Energy Resources(DERs) in main power grid. However, for DERs in the microgrid, different types will vary greatly in operating characteristics and their outputs were greatly affected by environment; for the energy storage devices in the microgrid, they have the disadvantages of high cost and short life. In addition, with the increasing penetration of EVs, the disordered charge and discharge have more impact on the main power grid and microgrid; these made the operating environment of microgrid become more and more complex, and brought a lot of technical problems for optimal operation of microgrid. Therefore, it was necessary to study the output characteristics and the effective control method of various resources of microgrid in the different states.In general, microgrid were grouped according to resource types, into three classes: DERs, Energy Storage Resources(ESRs) and Load Resources, in which DERs and ESRs were more important. In this thesis, DERs and ESRs of microgrid, as the research object, and its optimal operation were analyzed and researched. Maximum power point tracking(MPPT) of photovoltaic(PV) array in irregular shadow and virtual inertia control for variable speed wind turbines with high penetration were studied in the optimal operation and energy management of DERs; capacity optimization of hybrid energy storage system under grid connected condition and battery cycle life equalization in microgrid with distributed storage were studied in the optimal operation and energy management of energy storage. In addition, Electric Vehicles(EVs) as special dynamic energy storage resources, Vehicle to Grid(V2G) and real-time optimal scheduling of EVs were researched in the end of this thesis.Traditional MPPT methods could hardly find global maximum power point(MPP) in irregular shadow. To address this drawback, a new algorithm combining sliding mode variable structure control(SMVSC) with perturbation and observation(P&O) was proposed, besides, PI controller was applied to reduce system chattering. In order to verify the effectiveness of the proposed algorithm, the control effect was compared between the proposed method and the traditional P&O method in two cases which is the changes of illumination and shadow distribution. Compared with the traditional P&O method, the tracking time of the proposed algorithm was much shorter, and oscillation amplitude of the proposed algorithm was much less.To effectively use the "hidden" inertia of doubly-fed induction generators(DFIGs), a hybrid adaptive fuzzy control scheme that uses both rotating massof DFIGs and super-capacitor parallelling between rotor side converter(RSC) and grid side converter(GSC) as the virtual inertia source was proposed based on the comparison of different methods.The performance of the method was validated by real time simulation platform RT-LAB.The capacity optimization of hybrid energy storage system under grid connected condition was an optimization problem. In which, minimize one-time investment and operation cost in the whole life cycle as optimization objective, and reliability of power supply as the constraints. The simulated annealing algorithm was introduced to solve the problem that particle swarm optimization(PSO) algorithm has disadvantages such as earliness, easily trapped in local best. The improved simulated annealing particle swarm optimization(SAPSO) algorithm was proposed to be applied to capacity optimization of hybrid energy storage system. The validity of the algorithm was proved by the analysis and calculation of an example.In microgrid with distributed energy storage, the line impedance of each DER was different because of their differences in geographic location and equipment parameters, which caused the pressure of each battery to be different and the cycle life of each battery to be unbalanced.To solve this problem, the idea was inspired by V-shape formation of a flock of birds in this thesis. On the basis of analysis of battery aging model, the equalization of line impedance between each DER was achieved through adopting droop control based on virtual impedance and weighted factor for power rating method based on hierarchical control, and the cycle life of the batteries was ultimately extended.For vehicle to grid, V2 G mode based on microgrid was mainly researched, and multi-objective optimization model, in which minimum grid load fluctuation, maximum renewable energy utilization and maximum owners gain as the optimization objectives, was established. In order to solve the optimization model, variable threshold scheduling algorithm was proposed, and an improved algorithm named variable charge discharge rate scheduling algorithm was presented based on variable threshold scheduling algorithm to coordinate energy exchange between EVs and microgrid.The results showed the effectiveness of the two proposed algorithms.Real-time optimal scheduling of EVs include optimal scheduling for charging process of EVs in charging station and optimal scheduling model for path planning of EVs. For the former optimal scheduling, charging processes of EVs were divided into three parts according to queuing theory: input process, queuing process and service process. The queuing process and charging service process was mainly studied. Preemptive service algorithm and weighted factor algorithm for piecewise power dispatching is proposed to schedule the two processes, respectively. Using the software ExtendSim, the charging process of EVs was simulated, the results showed that the proposed algorithm was better than other algorithms. For the latter optimal scheduling, uncertain optimization theory was applied. Optimal scheduling model for path planning of EVs based on random expected value was built after analysing uncertain information of path planning. For the model, SA algorithm was introduced because of local convergence of PSO algorithm. The results proved that the improved hybrid intelligent algorithm enhanced the global convergence.