Study On Annual Power Supply-demand Balancing Considering Monthly Reservoirs Coordination

Power supply-demand balancing plays a major role in power system planning and future grid scheduling.The annual power supply-demand balancing is relatively long on the time scale,which can give full play to the long-term regulation performance of hydropower stations,and can fully consider the economic performance of long-term operation of power systems.Annual power supply-demand balancing can get scientific and reasonable decision plan on the basis of ensuring the balance of supply and demand of electricity.In China's traditional annual power balance model,monthly power balance was based on a given monthly hydropower generation and monthly load was represented by a single typical daily load.In order to adapt to the national energy-saving emission reduction strategy and the state grid company green power supply strategy,it is necessary to study new annual power supply-demand balancing model which further improves the resource utilization efficiency.The annual power supply-demand balancing model,a mixed integer nonlinear programming problem,is difficult to solve.So it is converted into a mixed integer linear programming model.And the research contents are as follows:The current annual power supply-demand balancing model first determines the monthly power generation of the hydropower station,and then balance the monthly supply and demand.There is a problem that failuring to give full play to the value of hydropower resources caused by monthly power supply-demand balancing.And the existing model has fewer considerations for thermal power unit constraints.This paper proposes a power supply-demand balancing model considering the monthly coordination of the reservoir and the typical single day of the month.The model is an annual overall optimization model.The model aims to minimize the annual electricity generation cost of the system.The model includes unit climb,minimum start-up time,and minimum downtime constraints,etc.The model unifies the annual operation plan of the hydropower station reservoir and the output of the hydro-fire unit on the typical day of each month.Finally,the feasibility,effectiveness and practicability of the model are verified by case study.For the current annual power supply-demand balancing model,monthly load is represented by a single typical day.There is a problem that the actual load description is rough.In this paper,a multi-typical daily load generation methods based on factor analysis and improved K-means clustering algorithm are proposed,and the annual power supply-demand balancing model considering reservoir monthly coordination and multi-day typical days in the month is proposed.Multi-typical daily generation method is based on factor analysis and improved K-means clustering algorithm.First,factor analysis is performed on the load to obtain a common factor.Secondly,the common factor is clustered by using the improved K-means clustering algorithm.Finally,the original load data is averaged according to the clustering classification result to obtain typical daily loads.The annual power supply-demand balancing model considers reservoir monthly coordination and multi-day typical days in the month.The monthly load is represented by a number of representative typical daily loads.The objective function is to minimize the annual power generation cost of the system,and considers the constraints such as unit climb,minimum start-up time and minimum downtime,and the continuation of the thermal power unit operating constraints between the typical daily loads.The model unifies the annual operation plan of the hydropower station reservoir and the output of the water and thermal unit on each typical day of each month.Finally,the the typical daily generation methods based on factor analysis and improved K-means clustering algorithm are validated and compared by using actual load data.The case study verifies that the model is more accurate by constructing examples.