Multi-period MINLP model for optimising operation and structural changes to CHP plants in district heating networks with long-term thermal storage
Multi-period MINLP model for optimising operation and structural changes to CHP plants in district heating networks with long-term thermal storage
复制标题
DOI：
10.1016/j.enconman.2008.10.010
复制
发表时间：
2009-03
影响因子：
10.4
通讯作者：
Tor-Martin Tveit;Tuula Savola;Alemayehu Gebremedhin;C. Fogelholm
中科院分区：
文献类型：
--
作者：
Tor-Martin Tveit;Tuula Savola;Alemayehu Gebremedhin;C. Fogelholm
通过使用储热装置，可以将电力和热量的产生分离，并且还可以减少投资，因为储热装置可以用于覆盖高峰负荷时期。这项工作提出了一个MINLP模型，可用于分析具有长期储热的区域供热网络中热电联产电厂的新投资和长期运行。在这项工作中提出的模型包括热电联产电厂的非线性非设计行为以及储热的通用数学模型，而不需要固定温度和压力。该模型的表述方式使其适用于确定性MINLP求解。该模型是非凸的，因此局部求解不能保证全局最优性。为了减少得到较差的局部最优的概率，该模型需要在初始值随机变化的情况下进行多次求解。可以从模型中提取许多结果，例如年总利润、工艺选项的最佳选择、工厂的质量流量以及每个工厂的发电量。该模型的公式使其适用于确定性MINLP求解。
By using thermal storages it is possible to decouple the generation of power and heat, and it can also lead to an reduction in investments, as the storage can be used to cover the peak load periods. This work presents a MINLP model that can be used for analysing new investments and the long-term operation of CHP plants in a district heating network with long-term thermal storage. The model presented in this work includes the non-linear off-design behaviour of the CHP plants as well as a generic mathematical model of the thermal storage, without the need to fix temperatures and pressure. The model is formulated in such a way that it is suitable for deterministic MINLP solvers. The model is non-convex, and subsequently global optimality cannot be guaranteed with local solvers. In order to reduce the chance of obtaining a poor local optimum compared to the global optimum, the model should be solved many times with the initial values varying randomly. It is possible to extract a lot of results from the model, for instance total annual profit, the optimal selection of process options, mass flow through the plant, and generated power from each plant. The formulation of the model makes it suitable for deterministic MINLP solvers.