Capacity Optimization Model And Algorithm For Grid-connected Microgrid Power Sources

With the effective utilization of renewable energies, flexible and intelligent controlcharacteristic, and their widely application in the improvement of power quality andsystem reliability, energy saving and environmental protection, microgrid graduallyattracts more and more attentions worldwide. Currently, there is a higher risk ofmicrogrid application due to the weak renewable energy industry, high installed cost ofdistributed generation and immaturity of microgrid technology. It has great significancefor planning and construction of microgrid to reasonably allocating microgrid powersources, obtaing the best social and economic benefits. Focusing on the capacityoptimization of grid-connected microgrid power sources, research works in this thesisare as follows:①Comprehensive considering the connectivity from load points to power sources,adequacy of microsource generation, and interaction between microgrid and distributionnetwork, this thesis proposes a path segmentation algorithm of reliability assessment fordistribution networks containing microgrids, which can be used to evaluate thereliability benefits of a microgrid. According to failure mode effects analysis, outageeffects type of load point in distribution network with microgrid is divided, and aconcept called load points failure excitation set is introduced; According toconnectedness between load points and power sources to classify paths, a discriminantmethod of elments in the load points failure excitation sets is put forward; Based on this,the classified paths are partitioned according to different types of switches, thus formingthe load points failure excitation sets and calculating reliability index. Using thealgorithm to assess the reliability of the distribution network with microgrid which ismodified by the main feeder F4of IEEE-RBTS BUS6, the example results show thatthe algorithm is feasible and practical.②Considering the sequential operating characteristics of microgrids, acost-benefit analysis model is established from different roles of systems, such as powersuplliers, distribution companies and load users. The model considers comprehensivecosts of installation, operation and maintenance, replacement, fuels and environment ofpower sources, as well as comprehensive benefits of energy saving, emission reduction,loss reduction, reliability improvement and delay of power grid investment. Reliabilitybenefits and loss reduction benefits of microgrid is calculated using the path segmentation algorithm of reliability assessment and the Monte Carlo sequential flowalgorithm for distribution network with microgrid respectively. Based on thecost-benefit analysis model, an economy evaluation model of microgrid is built usingthe net present value method. Example analysis results show that the proposed modelshave effectiveness and some practical engineering value.③An optimal capacity model of power sources of grid-connected microgrids ispresented based on the cost-benefit analysis principle, and an improved adaptive geneticalgorithm (IAGA) is used to solve the model. The model maximizes the net annualaverage net benefit as the goal, and considers constraint conditions of total installedcapacity, island operation adequacy, carbon emissions and so on. Combining operationmode of energy storage and micro gas turbine and the constraints included in the model,a generation method of initial population base on the dichotomy are formulated in orderto overcome the lack of using stochastic method to produce the initial population; In theobjective function，punishment items are introduced in order to convert constrainedproblem to unconstrained problem, and a constant is also added to prevent the fitnessfor negative, so as to construct the fitness function; Based on the adaptive linearadjustment method, crossover rate and mutation rate are adjusted to improve thealgorithm performance. Example analysis indicates that the model is able to fully reflectthe comprehensive application value, and obtain best economy of investment plan, andthe IAGA has better convergence properties than the GA.