Research On Stochastic Robust Scheduling And Collaborative Optimization Strategy Of Regional Energy Internet

With the increasingly serious global energy crisis,environmental pollution and climate change,the development of renewable energy has become the consensus of the international community.Energy Internet has attracted more and more attention since it is an energy ecosystem with smart grid as the foundation,the Internet plus as a means and electricity as the main carrier.Regional energy Internet is one of the main forms of national energy Internet comprehensive pilot demonstration,which has taken the lead in the implementation.However,there are still many problems in the optimal dispatch of regional energy Internet.Firstly,the output of renewable energy is volatile and uncertain,which brings great risks to the security and stability of regional energy Internet.Therefore,how to deal with source-load uncertainties is an urgent problem to be solved.Secondly,regional energy Internet is composed of many devices,abundant energy types and rich load types such that the corresponding optimization problem will be difficult to solve since the problem has a large number of variables and complex constraints,which will put great computational pressure on the dispatching center.Finally,many users in regional energy Internet are equipped with distributed generation equipment,which can meet their own load demand,and the excess electric energy can be sold to reduce the cost of energy consumption.Each user is a stakeholder,and the interests of each stakeholder are in conflict and difficult to be coordinated.Aiming at solving the aforementioned three issues,i.e.the optimal dispatch of regional energy Internet considering multiple source-load uncertainties,the collaborative optimization of regional energy Internet,and the game-theoretic optimization and pricing mechanism of regional energy Internet with multiple stakeholders.Main contributions of this dissertation are as follows.(1)Aiming at addressing the issue that multiple source-load uncertainties in regional energy Internet affect the security and stability of power grid,a stochastic optimization approach of regional energy Internet based on the chance constrained programming method is proposed.This approach adopts the method of discretizing probability distribution function to solve the probability distribution of the superposition of multiple source-load uncertainties without the convolution calculation.According to the probability distribution,the chance constraints of multi-uncertainties can be transformed into the deterministic constraints.On this basis,the stochastic optimal scheduling strategy of regional energy Internet is obtained.The probability that the strategy meets the uncertainty constraints is greater than a certain preset confidence level such that the approach proposed can effectively adverse effects of source-load uncertainties in regional energy Internet.(2)In some cases,the probabilistic models of source-load uncertainties are unknown such that the stochastic programming of regional energy Internet is unrealistic.To address this issue,two robust optimization approaches of regional energy Internet are proposed,i.e.a model predictive control based multiple time-scale optimization approach and a two-stage robust optimization approach based on the column and constraint generation method.Multiple time-scale optimization approach adopts three steps,i.e.day ahead optimization,real-time prediction,intra day rolling optimization and feedback correction,to design the robust optimal strategy of regional energy Internet.The multiple time-scale optimization approach can effectively address the issue caused by source-load uncertainties in regional energy Internet.However,it requires that devices have fast response speed and the dispatching center has strong calculating power.Therefore,a two-stage robust optimization approach is proposed.In this approach,the scheduling model is decomposed into a master-problem-and-subproblem scheme.Considering that the subproblem is a bilevel max-min optimization problem with integer variables,it is extended to a trilevel problem,and further decomposed into the inner-master-problem-and-inner subproblem scheme.The two-stage robust optimization approach has good convergence and a wide range of applications.It is an effective method to solve source-load uncertainties of regional energy Internet.(3)Regional energy Internet has a lot of devices and rich types of energy such that the dispatch of regional energy Internet involves many variables and constraints which will cause complicated calculations.To address this issue,a distributed collaborative optimization approach is proposed.Based on the analytical target cascading method,this approach decomposes the regional energy Internet into small hierarchical systems,i.e.a parent system and multiple subsystems,decouples the parent system and subsystems,and sets the linking variables.Select the appropriate penalty function to promote the convergence of linking variables to obtain the optimal solution of each system,which is the approximation of the global optimal solution.The distributed collaborative optimization approach can effectively alleviate the computing pressure of the dispatching center and realize the optimal dispatch of regional energy Internet.(4)Regional energy Internet has multiple stakeholders and each stakeholder has its own optimization objective.To balance the conflict of interests of different stakeholders,a gametheoretic optimization model based on genetic pricing mechanism is proposed.The model is oriented to regional energy Internet with ”one energy supplier,multiple users”.The energy supplier and users are in a Stackelberg game,with the energy supplier as the leader and the user as the follower.In this game,the energy supplier has the authority to fix electricity prices such that it sets appropriate electricity prices via genetic-priced mechanism to ensure the maximization of its own interest.Users can buy electricity from the energy supplier or the power grid,and sell electricity to the energy supplier or the power grid to minimize the cost of energy consumption.The proposed game-theoretic optimization model and pricing mechanism of regional energy Internet can effectively balance the conflict of interests among the energy supplier and users,and is practical in the market planning of regional energy Internet.