Framework for district heating optimization
(2016) In ISSN 0282-1990 MVK920 20161Department of Energy Sciences
- Abstract
- The objective of this thesis was to develop a flexible optimization framework for district heating networks (DHN) and to introduce pressure in the network models. This was done by creating a network represen- tation (NR) of the networks using the Python package NetworkX. Each node represents a customer or producer and the lines between the nodes are the corresponding pipes.
The method of the project was to first set up a large DHN using the NR based on geometries of the pipes and minimum and maximum supply temperatures for each customer. This model is then simulated to obtain different profiles for the network such as mass flows, loads and supply temperatures. Based on this information, the large network can be aggregated by using a... (More) - The objective of this thesis was to develop a flexible optimization framework for district heating networks (DHN) and to introduce pressure in the network models. This was done by creating a network represen- tation (NR) of the networks using the Python package NetworkX. Each node represents a customer or producer and the lines between the nodes are the corresponding pipes.
The method of the project was to first set up a large DHN using the NR based on geometries of the pipes and minimum and maximum supply temperatures for each customer. This model is then simulated to obtain different profiles for the network such as mass flows, loads and supply temperatures. Based on this information, the large network can be aggregated by using a method called “The German method”. This simplified network is then used for optimization, since large networks cannot be optimized due to the large computational time and memory consumption. This gives optimal trajectories of mass flow and supply temperature that are used as inputs in the original, complex network.
The results show that the framework is indeed possible to be used on real cases. By introducing pressure in the framework, one can introduce constraints on differential pressure on different types of nodes and still obtain optimal solutions for the producer unit. The framework is also flexible if the components to be changed contain the same parameters.
This framework could be a suitable tool to be used when performing production planning and model pre- dictive control (MPC) if a discretized optimization procedure is introduced. (Less) - Popular Abstract
- A flexible optimization framework for district heating networks could provide researchers a tool to test different system configurations and suppliers to improve their existing systems.
By introducing pressure in the system, the mass flows are distributed along the nodes, which determines the amount of heat delivered to each customer. The network representation of the district heating area can be created, simulated and optimized in different software programs.
Please use this url to cite or link to this publication:
http://lup.lub.lu.se/student-papers/record/8891193
- author
- Sharif, Jasir LU and Hamid, Mohammad
- supervisor
- organization
- course
- MVK920 20161
- year
- 2016
- type
- H2 - Master's Degree (Two Years)
- subject
- keywords
- District Heating, Network Representation, Model Predictive Control, optimization, simulation
- publication/series
- ISSN 0282-1990
- report number
- ISRN LUTMDN/TMHP-16/5373-SE
- ISSN
- 0282-1990
- language
- English
- id
- 8891193
- date added to LUP
- 2016-09-19 14:05:36
- date last changed
- 2016-09-19 14:05:36
@misc{8891193,
abstract = {{The objective of this thesis was to develop a flexible optimization framework for district heating networks (DHN) and to introduce pressure in the network models. This was done by creating a network represen- tation (NR) of the networks using the Python package NetworkX. Each node represents a customer or producer and the lines between the nodes are the corresponding pipes.
The method of the project was to first set up a large DHN using the NR based on geometries of the pipes and minimum and maximum supply temperatures for each customer. This model is then simulated to obtain different profiles for the network such as mass flows, loads and supply temperatures. Based on this information, the large network can be aggregated by using a method called “The German method”. This simplified network is then used for optimization, since large networks cannot be optimized due to the large computational time and memory consumption. This gives optimal trajectories of mass flow and supply temperature that are used as inputs in the original, complex network.
The results show that the framework is indeed possible to be used on real cases. By introducing pressure in the framework, one can introduce constraints on differential pressure on different types of nodes and still obtain optimal solutions for the producer unit. The framework is also flexible if the components to be changed contain the same parameters.
This framework could be a suitable tool to be used when performing production planning and model pre- dictive control (MPC) if a discretized optimization procedure is introduced.}},
author = {{Sharif, Jasir and Hamid, Mohammad}},
issn = {{0282-1990}},
language = {{eng}},
note = {{Student Paper}},
series = {{ISSN 0282-1990}},
title = {{Framework for district heating optimization}},
year = {{2016}},
}