This discussion has been locked.
You can no longer post new replies to this discussion. If you have a question you can start a new discussion

zone demand pattern versus customer demand pattern

A colleague is recommending using a demand pattern by zone instead of customer type.  I can see where this might work well for smaller zones that are mostly 1 type of customer, but in very large zones with a mix of residential, commercial and industrial customers I feel its a bad idea.  His thought is that as long as you calibrate it based on field data, 1 usage pattern is fine.  What about when planning for future growth?  Say there is an area that will be heavily residential in the future and you apply this zone pattern to it, then how reliable will it be?  I have my doubts on this approach.  Is there any experience doing it one way versus the other and the pros/cons?

Parents
  • We usually go for the middle road. We have quite large DMA's (they're more like service areas, with up to 100.000 connections per area). The connections in each area can be roughly split into 3 categories:

    * Large Industrial smart meters which send data frequently (every 15 minutes)

    * Small industrial/commercial meters which send data yearly

    * Residential meters which send data yearly.

    For the large industrial meters we use patterns that are unique to that meter (since we can get the pattern from the 15 minute data), for the smaller commercial & industrial meters we have a flat yearly base demand which we combine with patterns per customer type (the patterns we get from scientific research and past experiences). 

    The residential pattern however is derived from the zone pattern (DMA flow balance). For this we substract the demand of the large industrial meters & the small industrial/commercial meters from the zone demand. The result of this is used to create the residential pattern per zone/dma. This way the total demand per timestep in the model zone is always equal to the total zone demand you get from the field data while still differentiating between different customer types. 

    This approach might not be suitable to some DMA's (especially DMA's which have a large amount of large industrial smart meters, but a very small amount of meters that only have yearly data). For those DMA's we usually apply a customer type pattern for residential meters instead of a pattern derived from the zone demand pattern. 

  • Thanks for the feedback, that is basically how I'm used to doing it as well.  

Reply Children
No Data