Application of System Simulation to WCH Boiler Selection

Jan Hensen

University of Strathclyde, Glasgow

Karel Kabele

Czech Technical University in Prague

POSTER ONLY of paper presented at 'Building Simulation 97', Prague, Czech Republic

ABSTRACT

The paper reports the first results of an ongoing project aimed at generating design information /knowledge for wet central heating (WCH) refurbishment in multi-family houses in Central Europe. In that practical context, integral modelling and simulation of a building and its heating system is demonstrated. Given the underlying importance of the dynamic thermal interactions, building and plant are modelled at a high level of resolution using the ESP-r energy simulation environment.

BACKGROUND

A large proportion of the Czech housing stock consists of multi-family houses, very often with a central (solid fuel) boiler, gravity circulation, and almost no temperature control at the individual apartment level.

Nowadays very often refurbishment takes place, involving replacement of the boiler (usually gas or oil instead of solid fuel), incorporating a pump and either central thermostatic control or thermostatic valves, while the pipes and radiators remain the same.

The system designer now faces the challenge to select a suitable boiler in terms of lightweight or cast-iron, condensing or non-condensing, originally specified capacity (i.e. oversized) or re-calculated, etc.

Some of the factors that should be considered include:

It is very difficult to predict in a general sense the impact of these factors on:

The current work uses computer modelling and simulation to focus on a (solid fuel) boiler that needed to be changed to a modern natural gas type. The main questions are:

  1. to install a heavy-weight (cast iron) or a light-weight (steel or copper) boiler, and
  2. to size conventionally (i.e. 55 kW), equal to the total radiator capacity (i.e. 44 kW), or according to the total heat losses of the building (i.e. 32 kW).

MODELLING

The house in Prague.

Heating system schematic.

 

Boiler model parameters (base-case: conventional sized, heavyweight).

 

The aquastat of the boiler was set to 90 C. The boiler is ON/OFF controlled based on comparison of the temperature sensed in a reference zone with the set-point of the central thermostat.

 

 

ESP-r model of the building.

 

ESP-r model of the WCH system

 

Pump model parameters.

 

Radiator model parameters.

 

SIMULATIONS and RESULTS

Boiler configurations:

H = heavyweight; L = lightweight;

D = sized according prevailing design standards; O = over-sized.

 

Boiler code HO LO HD LD
total mass [kg] 1000 100 1000 100
fuel mass flow rate [kg/s] .00126 .00126 .00076 .00076
output [kW] 55 55 32 32

 

RESULTS:

 

Gas consumption [m3/day]

HO

LO

HD

LD

Winter

21.32

21.47

20.41

20.19

Spring

16.48

17.24

14.97

15.24

Ave.

18.90

19.35

17.69

17.71

%

100%

+2.4%

-6.4%

-6.3%

 

Operational time [hours/day]

HO

LO

HD

LD

Winter

4.70

4.73

7.50

7.42

Spring

3.63

3.80

5.50

5.60

Ave.

4.17

4.27

6.50

6.51

%

100%

+2.4%

+56%

+56%

 

Operational water-side efficiency [-]

HO LO HD LD

Winter

0.91

0.91

0.97

0.97

Spring

0.91

0.91

0.97

0.97

Ave.

0.91

0.91

0.97

0.97

%

100%

-0.17%

+6.9%

+6.8%

 

SIMULATION PERIOD:

A typical winter day (8 January) and a typical spring day (3 April), using a building-side computational time-step of 2 minutes, and a plant-side time-step of 1 minute.

 

 

 

CONCLUSIONS

In terms of the considered WCH heating system:

  1. Over sizing the boiler increases the gas consumption by about 10% with almost no effect on the indoor temperatures.
  2. A boiler 30% smaller than the installed radiator capacity, but in size conforms to the building design heat losses, is able to keep the indoor temperatures in the comfort range.
  3. In the case where the boiler is over-sized, a heavier boiler will have a lower gas consumption; in the current case about 3%. For a boiler sized according to heat losses, the influence of boiler mass is much smaller. However, during the winter the lighter boiler performs better, while during the spring the heavier boiler performs slightly better.
  4. In the current case the heating system is not very well balanced, resulting in instantaneous air temperature differences between the coolest and warmest zone of about 5 C, which is considerably higher than the normally accepted 3 C. For the rooms which are too warm this can easily be remedied by installing a simple valve. For the rooms which are too cold there is a need to introduce additional radiator capacity.
  5. Due to the high thermal inertia of the system (which is mainly caused by the radiators and the contained water, as opposed to the mass of the boiler) there is a definite need for an anticipating controller.

In terms of modelling and simulation:

  1. It is rather difficult to establish correct parameter values for various explicit plant component models.
  2. Currently there is no climatic reference year available for the Czech Republic.
  3. Although it is necessary for a study as described in this paper, it should be recognised that high resolution modelling of a WCH system involves a lot of resources in terms of setting up the model, verifying both the input and output data, and of analysing the results.

Future work:

  1. Incorporation of different WCH control systems, such as central control with anticipation, room-level control with thermostatic valves, etc.
  2. Investigation of the effect of intermittent heating as opposed to a continuous operation which has been assumed in the current paper.
  3. Verification of the plant component models in general, and the boiler model in particular. Ideally this would involve experimental work.

IN RETROSPECT:

Although various problems still require resolution, and despite the intensive effort required to effectively model a typical WCH system, we believe that integrated modelling and simulation of the building and plant is the way forward in addressing the related design and control problems of these systems.