摘要
针对光伏光热综合利用(PV/T)系统热电协调控制中组件温度控制非线性大惯性系统的温度控制问题,文中提出对PV/T组件进行短期温度预测,以使PV/T系统控制器根据短期预测情况提前动作,从而优化PV/T系统控制效果。文中分析了A类天气类型(晴天、晴间多云、多云间晴)下组件温度的变化情况,并结合RBF神经网络对组件温度数据建立预测模型。仿真实例分析结果表明,该预测方法在A类天气类型下预测精度较高,最大相对误差为13.22%,最大平均相对误差为3.6%。研究结果可为后续PV/T系统研究提供技术支撑。
Aiming at the non-linearity and large lag time of temperature control in PV/T system,a method of optimizing control by running the PV/T system controller ahead according to the shortterm forecasting of PV/T components is proposed. Based on the analysis of the temperature change of the PV/T module under A type( sunny,partly cloudy,cloudy) weather,a RBF neural network is used to establish the prediction model. The prediction method is used to predict PV/T components temperature under A type weather and the results show that the method had high prediction accuracy with the maximum relative error 13. 22% and the maximum average relative error 3. 6%,respectively. The method can provide technical support for the subsequent PV/T system research.
出处
《广西大学学报（自然科学版）》
CAS
北大核心
2017年第3期1035-1041,共7页
Journal of Guangxi University（Natural Science Edition）
基金
广西自然科学基金资助项目(2014GXNSFAA118372)
广西教育厅科研基金资助项目(2013YB015)
