Developing the Model

Choice of Software

 

MS Excel was chosen as the modelling platform, because the team had prior training and confidence in using the software. In addition, this approach had the following advantages:

 

• Simplicity – Minimal training is required to operate and modify the model.

• Transparency – All steps and calculations can be critically observed and conclusions drawn about their reliability. This also adds educational value compared to commercial “black box” software.

• Customisation – All calculations can be modified depending on the system configuration being modelled and the assumptions made.

• Various outputs – graphs, summary statistics, etc can be generated and analysed within the software without exporting results.

• Cross-platform, unlike most commercial software.

 

Scope

 

When defining system boundaries, the focus was on the choice of input parameters. Parameters that have the most significant effect on plant performance and supply-demand matching were selected as inputs – a selection is shown in Table 1. At the same time it was ensured comparable data was available in literature for different plant types, so the model could be used in practice to compare these.

 

• Calculations were based on widely used high level operational parameters, but the underlying physical reasons and thermodynamical calculations were omitted.

• All inputs were assumed constant throughout plant operation.

• Some more specific assumptions were also made, which can be seen in the model.

Table 1. Inputs for the model

On the other hand, learning about the lack of data in the early stages of development directed the modelling process towards a more usable result. The requirements and assumptions of the model can be compared to the initial planning phases of installing a CHP plant in an industry. Then, as well as in the model, precise technologies may not yet have been chosen, meaning detailed specifications are unavailable and generic parameters need to be used. Approach First, the parameters described above were mapped, including the ways they interact with one another and how they influence plant operation. This network was then organised in the form of a flow diagram, where plant operation decisions are made by binary logic and calculations are based on that logic. Each node (single logical decision or calculation) on the diagram was numbered; modules performing a similar task were colour-coded. When the flow diagram was completed, the logic and calculations from the diagram were translated into Excel formulas. Each column in the calculator table represents one node on the diagram, following the same functions and numbering. The Excel calculator performs the calculations specified in the flow diagrams with a one hour timestep over the course of one year. The model was developed in stages, adding functional modules and moving from simpler to more complex. Peer validation was used in each step to review the logic. The results calculated by the model were compared against those calculated by hand, to ensure each module is functioning in the intended way.