Modular approach for modeling a multi-energy district boiler

被引:7
作者
Eynard, Julien [1 ]
Grieu, Stephane [1 ]
Polit, Monique [1 ]
机构
[1] Univ Perpignan, ELIAUS Lab, F-66860 Perpignan, France
关键词
Multi-energy district boiler; Hot water distribution network; Modelling task; Hammerstein-Wiener model; Identification; White; grey and black boxes; POWER-PLANT; PREDICTIVE CONTROL; WIENER MODEL; PART I; IDENTIFICATION; OPTIMIZATION; SIMULATION; SYSTEMS; NETWORK; STORAGE;
D O I
10.1016/j.apm.2011.02.006
中图分类号
T [工业技术];
学科分类号
08 ;
摘要
The present paper deals with the modeling of a district boiler (city of La Rochelle, west coast of France), as part of the OptiEnR research project. This "multi-energy" boiler supplies domestic hot water and heats residential and public buildings, using mainly wood and sometimes fuel or gas if necessary. The OptiEnR research project focuses on optimizing the performance of the boiler. Its main objective is to minimize the use of fossil energy, stocking renewable energy during low-demand periods and using it when peak-demand is high. Because of both the complexity of the plant as a whole and the strong interactions between the sub-systems (the wood boiler, the gas-fuel oil boiler, the breaking pressure bottle, the cogeneration plant, the hot water distribution network), a modular approach has been proposed. According to what information is available, a combination of white, grey and black boxes (Hammerstein-Wiener models) has been used to carry out the modeling task. To answer for the lack of information, additional parameters were proposed and identified. The model has been first used in simulation during heating periods, with the aim of optimizing both the parameters of the boilers control systems and the use of wood, gas and fuel oil. Next, it will be used, when adding to the plant a thermal storage unit and implementing a model predictive controller, to improve its functioning, especially reducing the coverage rate of the fossil energy boiler. (C) 2011 Elsevier Inc. All rights reserved.
引用
收藏
页码:3926 / 3957
页数:32
相关论文
共 34 条