论文部分内容阅读
循环水余热回收系统中,热泵热网水出口温度在跟踪供热负荷需求时,在受驱动蒸汽量的调节的同时,往往易受热网回水、循环水等工况变化的影响,传统PID控制方式超调量大、负荷跟踪能力差。提出一种混沌变异克隆选择-径向基函数(CPCS-RBF)直接多步预测控制策略,以热泵热网水出口温度预测值与设定值差值为目标函数,利用CPCS优化算法求取目标函数最小时的驱动蒸汽最佳值。预测模型由2个RBF神经网络结合热泵现场运行数据构建,以提高热泵系统适应工况变化的能力;实验结果表明,该控制策略能综合学习热网回水温度、循环水温度等参数的变化,使驱动蒸汽调门超前动作,及时跟踪供热负荷需求变化的同时,适应发电负荷变化下排气余热量的波动,具有更好的节能效果和变工况适应能力。
Circulating water heat recovery system, the heat pump heat network water outlet temperature tracking heating load requirements, driven by the amount of steam in the adjustment at the same time, are often susceptible to heat network backwater, circulating water and other conditions change the impact of traditional PID control Large overshoot, load tracking ability is poor. A direct chaos mutation clonal selection-radial basis function (CPCS-RBF) direct multi-step predictive control strategy is proposed. Taking the difference between the predicted value and the set value of the outlet temperature of the heat pump heat network as the objective function, the CPCS optimization algorithm is used to calculate the target Optimal drive steam at function minimum. The prediction model is constructed by two RBF neural networks combined with field data of the heat pump to improve the ability of the heat pump system to adapt to the change of working conditions. The experimental results show that the control strategy can comprehensively study the change of parameters such as the return water temperature and circulating water temperature, So that the drive steam valve door move ahead of schedule, timely tracking changes in demand for heating load at the same time, to adapt to changes in power generation load exhaust heat fluctuations, with better energy saving and adaptability to conditions change.