Optimal Operation Method Of Microgrid In Multi-Types Under The Implementation Of Generalized Demand Side Resources

The microgrid technology has become the core of the integration application of the distributed power generation system.Positive development on the user side and large-scale access of distributed power supply are the most effective ways to make use of the new energy locally,and take full advantage of the distributed power generation system.This dissertation studied the optimal operation problem of microgrid in multi-types from the generalized demand side resources alone with its classification,forecasting technique,the automated demand response technique and collaborative optimization.The microgrid on the user side includes generalized demand side resources,such as load resource,distributed generation resource,storage system resource,electricity price and demand response programme.An analysis is carried for the applicability of different types of resources.And a basic model of optimal operation of microgrid on the user side is proposed,which optimize for the generalized demand side resources in multiple scenarios.The simulation calculation and analysis of different kinds of generalized resources' cost-benefit influence level on the user side microgrid are carried out.And different kinds of resources' cooperation effects are studied,which establishes the theoretical foundation for the further study of optimal operation problem of microgrid.A combined short-term load forecasting model based on Empirical Mode Decomposition(EMD),Extended Kalman Filter(EKF)and Extreme Learning Machine with Kernel(KELM)is proposed to solve short-term load forecasting problem of micro-grid on the user side.The time-series data of micro-grid load with high randomness is gradually decomposed into a number of Intrinsic Mode Function(IMF)components through EMD.Two typical different prediction models—EKF and KELM— are adopted to predict different kinds of IMF components,and Particle Swarm Optimization(PSO)is used to realize the optimization of model parameters.A combined mode based on off-line parameter optimization,period parameters update and on-line prediction is proposed to treat with the micro-grid computational constraints on the user side.The model prediction accuracy,the stability of period updating and the calculation efficiency is verified through examples analysis of micro-grid of the user side with different types and capacity.The model is equipped with implementation conditions of embedded hardware development.A typical structure of island microgrid system includes generalized demand side resources like wind,photovoltaic,energy storage and diesel engine is proposed.Based on the functional analysis of different types of generalized resources,an optimal method and a masterslave charging strategy of storage system for island microgrid is proposed.On the base of the NSGA-? algorithm,the comparison of difference and advantage of the auxiliary power regulation of adjustable load,and the masterslave charging strategy separately are finished.And it can be verified that adjustable load could be applied in reducing the battery life loss and the cost of diesel engine fuel successfully.By comparing two charging strategies for energy storage system,the significant role of reasonably distributing the renewable energy's utilization pattern is verified.Combining the heuristic strategy and the intelligent optimization,a collaborative optimization operation model with flexible implementation is proposed to the residential microgrid on the user side.Based on the detailed classification method and the oriented modeling of residential controllable loads,mixed binary differential evolution algorithm is used to optimize different types of load operation mode and the distribution ratio of renewable energy.Linkage mechanism between real-time price,planning loads and user dissatisfaction is built in objective function,and a model solution containing double cycle synchronization and progressive population regeneration is proposed.Aiming at optimized result of actual residential microgrid,influence of operation mode of energy storage system on the optimization objective is analyzed,and the robustness of proposed model is verified through sensitivity analysis,which can provide data and theoretical support for the application and promotion of residential microgrid.Based on the analysis of flexibility and regularity of generalized demand side resources like commercial load and electric vehicle,a collaborative optimization operation mode for multiple commercial microgrid(two or more)is proposed aiming at commercial microgrid,in which,microgrid operating cost and dissatisfaction are taken as optimization objective.And a multi-objective optimal operation model including shiftable load,interruptible load and compensation rate,optimum contract capacity and imbalance electricity price is proposed as well.The relationship between load resource allocation,demand response program and multi-objective is built.The ability to implement automated demand response of commercial microgrid on the user side is improved and the applicability and economical efficiency of the double microgrids collaborative optimization containing the homologous interoperability is verified through simulation and analysis.While comparing to the independent optimization,collaborative optimization can greatly improve the economical efficiency.The effect of imbalance electricity price on objective function,microgrid,distribution network interaction power and the auxiliary power regulation is analyzed directing at optimizing data.