基于混合模型的中央空调能源优化控制系统设计

发布时间：2018-01-14 02:24

本文关键词：基于混合模型的中央空调能源优化控制系统设计　出处：《北京交通大学》2015年硕士论文　论文类型：学位论文

【摘要】：在夏季随着人们生活水平的提高,中央空调耗电量也不断提高,中央空调耗电量占建筑物总耗电量的比例也在不断提高。为了降低能耗,许多单位对中央空调进行了节能改造,但仍然存在能效低下和冷量浪费严重的问题。针对这些问题本文研究了基于混合模型的中央空调能源优化控制系统,利用混合模型能够预测能耗和输出冷量关系的特点,规划了在消耗相同能量下最合理的目标温度曲线解决了冷量浪费的问题。同样利用混合模型在实时控制中协调优化了对空调系统各部分的控制,提高了空调系统的能效。本文的主要工作有：针对空调系统结构复杂,控制量、能耗和输出冷量之间关系难以预测的问题本文首先建立了带有待定系数的中央空调系统灰箱模型,然后利用空调系统实际运行数据对待定系数进行辨识建立了空调系统物理模型。针对物理模型精度比较差的问题,本文借助人工神经网络建立了中央空调系统混合模型,并对混合模型的预测效果进行了分析。针对基于混合模型的实时控制计算量大,系统快速性无法保障的问题,本文建立了基于混合模型预测的专家表,通过查表就可以实现对系统的控制。针对传统自动控制系统没有对能耗的预测,无法自动优化控制目标值,造成能源浪费的问题,本文设计了指定能耗下目标温度规划功能,既保证了空调环境下人体的舒适度又实现了节能。另外本文还对控制系统软件和硬件的核心部分进行了设计,经过模拟分析系统的节能效果在20%以上。
[Abstract]:In the summer, with the improvement of living standards, the central air-conditioning power consumption is also rising, the central air-conditioning power consumption accounted for the proportion of the total power consumption of buildings is also increasing. In order to reduce the energy consumption, many units of energy-saving of central air-conditioning, but there are still low energy efficiency and cooling capacity of serious waste problem. To solve these problems this paper studies the central air-conditioning energy optimization control system based on hybrid model, using the hybrid model is able to predict the relationship between the characteristics of energy consumption and output of the cold, the temperature curve of the most reasonable planning target in consumption under the same energy to solve the cold waste problem. Using the same hybrid model coordinated optimization control of all the parts in the air conditioning system in control system, improve the energy efficiency of air conditioning system.
The main work of this article is as follows:
In the air-conditioning system with complicated structure, control volume, the relationship between energy consumption and cold output is difficult to predict the problem this paper established a grey box central air conditioning system with undetermined coefficient model, and then use the air conditioning system in actual operation data to coefficient of air conditioning system, the physical model was established for identification.
Aiming at the poor accuracy of the physical model, a hybrid model of the central air-conditioning system is built by artificial neural network, and the prediction effect of the mixed model is analyzed.
In view of the fact that the real-time control based on the hybrid model is too large for computation and the speed of the system is not guaranteed, the expert table based on the mixed model prediction is established, and the control of the system can be realized by look-up table.
Aiming at the problem that the traditional automatic control system does not predict the energy consumption, it can not automatically control the target value and cause the waste of energy. In this paper, we designed the target temperature planning function under the specified energy consumption, which not only ensured the comfort of the human body in the air conditioning environment, but also realized the energy saving.
In addition, the core part of the software and hardware of the control system is designed, and the energy saving effect of the simulation analysis system is more than 20%.

【学位授予单位】：北京交通大学
【学位级别】：硕士
【学位授予年份】：2015
【分类号】：TB657.2;TP273

【参考文献】