基于Q学习算法的燃煤机组深度调峰协调控制优化

Optimization of Coordinated Control for Deep Peak Shaving of Coal-Fired Units Based on Q-Learning Algorithm

常规的燃煤机组调峰协调控制方法以储能负荷控制为主,减少了节流损失,增加了煤耗量,影响深度调峰协调控制的经济性效果。因此,设计了基于Q学习算法的燃煤机组深度调峰协调控制优化方法。确定燃煤机组调峰协调控制优化参数,在满足调度中心给定负荷指令的基础上,将经济性指标作为协调控制优化的约束条件,简化燃煤机组深度调峰的协调步骤。基于Q学习算法控制燃煤机组深度调峰均衡负荷,令机组实际供电负荷之和与全厂总负荷指令相等,从而满足深度调峰协调控制的经济性需求。通过对比实验证实,文章所提优化方法的经济性能更高,能够应用于实际生活。

The conventional coordinated control method for peak shaving of coal-fired units mainly focuses on energy storage load control,reducing throttling losses,increasing coal consumption,and affecting the economic effectiveness of deep peak shaving coordinated control.Therefore,a coordinated control optimization method for deep peak shaving of coal-fired units based on Q-learning algorithm is designed.Determine the optimization parameters for peak shaving coordination control of coal-fired units,and on the basis of meeting the load instructions given by the dispatch center,use economic indicators as constraints for coordinated control optimization,simplifying the deep peak shaving coordination steps of coal-fired units.The Q-learning algorithm is used to control the deep peak shaving and load balancing of coal-fired units,so that the sum of the actual power supply loads of the units is equal to the total load command of the whole plant,so as to meet the economic needs of the coordinated control of deep peak shaving.Through comparative experiments,it is confirmed that the optimization method proposed in this paper has higher economic performance and can be applied to real life.