考虑电量可实现性和启停功率轨迹的火电机组组合混合整数线性规划模型

An MILP Model for the Thermal Unit Commitment Problem With Feasible Energy Delivery and Start-up & Shut-Down Power Trajectories

为解决传统模型过于简化以致调度计划存在电量不可交付、产生系统频率偏差的问题,提出一种考虑电量可实现性和启停功率轨迹的火电机组组合混合整数线性规划（mixed-integer linear programming,MILP）模型,该模型引入一类0-1变量表示燃煤机组的运行状态,便于在其加热、升负荷、调度和降负荷4个阶段的逻辑判断和发电量计算;根据燃气燃油机组启停迅速的特点,对其运行状态重新建模;并支持冷、温、热等多种启动类型和1 h、15 min等多种调度时段长度。10～1 000机24时段系统的计算结果表明：所提模型更符合实际运行情况,可很好地解决电量不可交付问题,且具有较高的求解效率。

This paper proposes a mixed-integer linear programming （MILP） model for thermal unit commitment problem with feasible energy delivery and start-up ＆ shut-down power trajectories. The aim is to solve problems of unrealizable generation schedule in terms of energy delivery and system frequency deviation due to excessive simplification of traditional model. This model introduces a set of binary variables to represent operating states of coal-fired units, which facilitate logical judgment and electricity production calculation in four phases, i.e. warm-up, start-up ramp, dispatchable, and shut-down ramp. According to characteristics of rapid start-up and shut-down of gas-fired and oil-fired units, new operating states are modeled. This model also supports various start-up types of cold, warm and hot start and various dispatching time span of each interval such as 1 hour and 15 minutes. The results for systems ranging from 10 to 1000 units during 24 periods show that the proposed model fits actual running situation better and can not only solve unrealizable generation schedule in terms of energy delivery, but also is featured with higher solving efficiency.