| 基于深度确定性策略梯度的热力站一次侧优化控制 |
| |
| 引用本文： | 李琦,韩冰城.基于深度确定性策略梯度的热力站一次侧优化控制[J].科学技术与工程,2019,19(29):193-200. |
| |
| 作者姓名： | 李琦 韩冰城 |
| |
| 作者单位： | 内蒙古科技大学信息工程学院,包头,014010;内蒙古科技大学信息工程学院,包头,014010 |
| |
| 基金项目： | 国家自然科学基金项目（面上项目，重点项目，重大项目） |
| |
| 摘 要： | 针对热力站供热量与需求量不匹配的现象,提出一种基于深度确定性策略梯度(DDPG)的热力站一次侧优化控制方法。采用LSTM(long short term memory)算法对热力站进行建模,然后结合集中供热系统运行机理,使用DDPG控制算法对热力站一次侧供水流量序列求解。运用包头某热力站的大量历史工况数据,进行仿真实验,结果表明该方法的有效性,一定程度上实现了热力站的按需供热,提高热量的利用率。
|
| 关 键 词： | 深度确定性策略梯度 热力站 优化控制 长短时记忆网络 |
| 收稿时间： | 2019/3/26 0:00:00 |
| 修稿时间： | 2019/7/9 0:00:00 |
Primary Side Optimization Control of Heating Substations Based on Deep Deterministic Policy Gradient |
| |
| LI Qi and Han Bing-cheng.Primary Side Optimization Control of Heating Substations Based on Deep Deterministic Policy Gradient[J].Science Technology and Engineering,2019,19(29):193-200. |
| |
| Authors: | LI Qi and Han Bing-cheng |
| |
| Institution: | chool of Information Engineering, Inner Mongolia University of Science and Technology, |
| |
| Abstract: | Aiming at the phenomenon that the heat supply and demand of the heating substations are not matched, a primary side optimization control method based on Deep Deterministic Policy Gradient (DDPG) is proposed. The Long Short Term Memory (LSTM) algorithm was used to model the heating substation. Combined with the operation mechanism of the central heating system, the DDPG control algorithm was used to solve the primary side water supply flow sequence of the heating substation. The simulation experiment was carried out by using a large number of historical working conditions data of a heating substation in Baotou. The results show that the method can achieve the goal of on-demand distribution of heating substations and improve the heat rate. |
| |
| Keywords: | deep deterministic policy gradient heating substations optimized control long short term memory |
| 本文献已被 CNKI 万方数据 等数据库收录！ |
| 点击此处可从《科学技术与工程》浏览原始摘要信息 |
| 点击此处可从《科学技术与工程》下载免费的PDF全文 |
