Demand-side Resources Aggregation Methods In Electricity Spot Market

The construction of electricity market provides an opportunity for the demand side to participate in the balance adjustment of power supply and demand,the demand-side resources are huge,scattered,and difficult to control,naturally they need to participate in the market through load aggregator.Load aggregation technology is the core competitiveness of load aggregator.With improvement of electricity market mechanism,enrichment of participating entities and transaction varieties,it’s necessary to develop load aggregation technology in accordance with the economic and physical characteristics of various demand-side resources.Aggregators suffer some practical problems,such as the influence of spot price signals with frequent fluctuations that are not taken into account by the aggregation strategy,and the difficulty in solving large-scale optimization problem.Regarding these problems,this thesis studies aggregation methods of demandside bidding curves and flexible feasible region in spot market.The main innovations are as follows:Firstly,this thesis presnet a decision-making optimization model for load aggregators based on China’s characteristics.For unschedulable users,the aggregator aggregates the bidding curves submitted by users,and considers that the aggregated curve reflects the user’s true price response characteristics,while meeting the requirements of the spot market;for schedulable users,the aggregator aggregate the feasible regionon of flexibility,and consider maximizing utilities of users.Secondly,this thesis presnet an aggregation method of demand-side bidding curves considering the uncertainty of electricity prices.The forecast errors of the electricity price point forecasting model are calculated.The kernel density estimation is used to fit the errors probability distribution to establish the electricity price probability distribution generation model.A model is introduced into the aggregation process of the load bidding curve,leading to more accurate solutions.The aggregation model is transformed into a constrained piecewise nonlinear programming by discretization deformation,and the solution was solved by a heuristic algorithm.We verify the effectiveness of proposed method by real electricity price data.Thirdly,this thesis presnet demand-side flexibility feasible region aggregation method considering time value.Firstly,a flexible feasibility region description model based on convex polyhedrons is established,and then zonotopes are used to simplify and approximate the convex polyhedrons,so as to realize the feasible region of schedulable user aggregates using Minkowski sum;The zonotope approximation method considering the time value guides the time slots with higher time value to perform more accurate feasible region fitting.The proposed method can achieve efficient aggregation of highdimensional,large-scale equipments’ feasible regions,so as to provide technical support for demand-side resources to provide multiple types of balance adjustment services.To sum up,for the problem of load aggregation,this thesis presnet the demand side bidding curve aggregation method considering the uncertainty of electricity price and the demand side flexibility feasible region aggregation method considering the time value respectively,which solved the technical bottlenecks such as the lack of consideration of the aggregation strategy of load aggregation technology and the difficulty of solving the aggregation model.Make the power system balance regulation more resources abundant at the same time to improve the user power efficiency.