The user must specify the weekday on which 1st January falls for the simulation in order to determine the weekday pattern. This can be done in two ways. You can set the year by selecting ‘Same as for year’, in which case the weekday will be taken to be the weekday on which 1st January would fall for the specified year. Alternatively, you can explicitly specify the weekday to be used by selecting ‘Day’. Note, in particular, that ‘Same as for year’ does not specify the actual simulation year. Thus, one can specify a year from which the weekday on 1st January will be deduced and specify a leap year simulation, even if that year is not a leap year.
If you use the ‘Same as for year’ option, there is a further choice. If you tick ‘Take from weather file’ the year will be set from the year recorded on the simulation weather file. If you do not tick this box you can set the year explicitly. In either case, this year simply determines the weekday on which 1st January falls for the simulation.
Each of the options described above sets the weekday pattern for all days of the year. The weekdays are assumed to cycle round the sequence Monday, Tuesday, Wednesday... between January 1st and December 31st. Because the 365-day simulation year is 52 weeks and one day, December 31st and January 1st are assigned the same weekday, so the cyclic weekday pattern is broken at this point for simulations which cross the year end. The pattern is also broken in a similar way for leap years, where one year is 52 weeks and two days. However, one can preserve this pattern over the year end by ticking ‘Maintain weekday continuity across year end (with no holidays)?’
Note that if the user carries out a year-end-crossing simulation with this option enabled, then 1st January from the second year of the simulation will be associated with the specified weekday.
