You are using an outdated browser. Please upgrade your browser to improve your experience.
Javascript is disabled in your web browser. For full functionality of this site it is necessary to enable JavaScript.
This website is using cookies.
We use them to give you the best experience. If you continue using our website, we'll assume that you are happy to receive all cookies on this website.
x
CPOTE2020 logo
CPOTE2020
6th International Conference on
Contemporary Problems of Thermal Engineering
Online | 21-24 September 2020

Abstract CPOTE2020-1288-A

Book of abstracts draft
slider slider slider slider slider slider

Advanced operational optimization of the district heating network

Michał GUZEK, Warsaw University of Technology, Poland
Teresa KUREK, Warsaw University of Technology, Poland
Jakub BIAŁEK, Warsaw University of Technology, Poland
Michał WARCHOŁ, Warsaw University of Technology, Poland
Wojciech BUJALSKI, Warsaw University of Technology, Poland
Konrad ŚWIRSKI, Warsaw University of Technology, Poland
Konrad WOJDAN, Warsaw University of Technology, Poland

This paper examines an optimization system that allows controlling district heating network optimally. A simplified network model created with the use of machine learning techniques is used. An innovative algorithm, which iteratively runs fast calculations of the linear programming class, allows to select the parameters of heating plants (supply temperature and pressure), pumping station (pressure head), and decide on shutting down or starting peak load heating plant. The dedicated work environment of the solution is large-scale heating networks with multiple heat sources and a complex network structure. District heating systems are subject to increasing automation. One of its tasks is to improve economic indicators and, at the same time, increase the efficiency of the system, which is most often expressed by reducing heat losses. Still, in most systems with qualitative-quantitative regulation, which allows modifying the network supply temperature during the day, the decisions are made by the operator using the primary supply temperature heating curve, weather forecast, and mainly - own experience. Due to the necessity to take into account the high inertia of the heating network, these decisions are subject to a substantial safety margin, expressed as the excess of the supply temperature. By using computer calculations, this margin can be minimized, thus achieving economic and environmental benefits. This problem belongs to the Mixed Integer Nonlinear Programming class, and there are implementations of optimizers working on small radial networks. In the case of large heating networks with multiple rings, the convergence of a solution to such a complicated problem is challenging to achieve. In Europe, it happens that a heat distributor does not own heating plants, and optimizer in the paper is built from such a point of view. The paper proposes a method that allows to quickly optimize the operation of the large network in a short time horizon (48 hours), with the use of low computational expenditure due to the simplifications applied. The solution is scalable regardless of the size and complexity of the heating network.

Keywords: District heating, Cost optimization, Heat distribution, Machine learning, Network modeling