A chiller plant with VSD chillers,VSD condenser water pumps,VSD cooling towers.Though each of the VSDs will help save energy,you would need a software programme to find the sweet spot( point where it consumes lowest possible energy for a given output.Has anyone used such plant optimization software?What are your impressions about its effectiveness?
Tek-Tips is the largest IT community on the Internet today!
Members share and learn making Tek-Tips Forums the best source of peer-reviewed technical information on the Internet!
-
Congratulations LittleInch on being selected by the Eng-Tips community for having the most helpful posts in the forums last week. Way to Go!
Chiller plant optimization software
- Thread starter SAK9
- Start date
EnergyProfessional
Mechanical
Haven't used it myself, but Trace has chiller plant built in.
- Thread starter
- #3
Trace is a design aid software.I am referring to a software that actually manages real life plant operation( more like BAS control software).Depending on real time operating conditions,it controls the whole plant in the most energy efficient manner.
- Thread starter
- #4
Here is a link to that explains the all variable speed plant operation.
as far as i know, such simulations exist only in manufacturers hands as proprietary knowledge.
they can offer you some chiller controller claiming that they possess algorithm that will conduct "live" optimization, and you can believe them, specify it and than monitor how successful it is.
that has mostly to do with VSD controls. i don't think hvac engineer can go so deeply in one such specialized field.
it is to be hoped, though, that in future some performance standards will be specified. currently we have manufacturers who claim how their new products makes revolution in efficiency, but what mostly matters, system efficiency is not elaborated much by anyone. this is becoming much more important with so many automated components in new chilled systems, but things still did not evolve.
they can offer you some chiller controller claiming that they possess algorithm that will conduct "live" optimization, and you can believe them, specify it and than monitor how successful it is.
that has mostly to do with VSD controls. i don't think hvac engineer can go so deeply in one such specialized field.
it is to be hoped, though, that in future some performance standards will be specified. currently we have manufacturers who claim how their new products makes revolution in efficiency, but what mostly matters, system efficiency is not elaborated much by anyone. this is becoming much more important with so many automated components in new chilled systems, but things still did not evolve.
Drazen's about right I think.
I've worked this problem on a control system a few times before, but not in the way you're thinking. With those variable-speed chillers, it's important to keep them in their sweet spot (about 30 percent load). Then let the cooling towers make the condenser water as cool as the chiller manufacturer allows. That's been the best sequence of operation in my experience, energy-wise.
In order to do a complete optimization, you'd need a kW meter on every piece of equipment -- tower fans, condenser pumps, every chiller. The program would have to "learn" the energy profile by logging and trending kW versus load. Then it would have to adjust its own sequence of operation to move toward that minimum-energy optimum point. That's a hairy program to write, I have not heard of it being done...
Best to you,
Goober Dave
Haven't see the forum policies? Do so now: Forum Policies
I've worked this problem on a control system a few times before, but not in the way you're thinking. With those variable-speed chillers, it's important to keep them in their sweet spot (about 30 percent load). Then let the cooling towers make the condenser water as cool as the chiller manufacturer allows. That's been the best sequence of operation in my experience, energy-wise.
In order to do a complete optimization, you'd need a kW meter on every piece of equipment -- tower fans, condenser pumps, every chiller. The program would have to "learn" the energy profile by logging and trending kW versus load. Then it would have to adjust its own sequence of operation to move toward that minimum-energy optimum point. That's a hairy program to write, I have not heard of it being done...
Best to you,
Goober Dave
Haven't see the forum policies? Do so now: Forum Policies
- Thread starter
- #7
Thanks Drazen and DRWeig.There seem to be a few out there at the moment with proprietary software like this:
DRWeig;Your logic is sound for variable speed chillers with variable speed cooling towers.I will run multiple towers at reduced speed to get lowest possible condenser water so that chiller can operate at lowest possible speed(whether on full load or part load).Things get tricky when you have a variable speed condenser water pump as well.If the chiller load is only 50% you can reduce condenser water flow as heat rejection is much lower.The question then is whether you reduce speed of both cooling tower fans and condenser water pumps in the same proportion as chiller load or go for some staged control( reduce tower fan speed first to get a desired temperature then reduce pump speed etc.) .With three variables a number of scenarios are possible with one affecting the other as well as the overall result.
You do not need individual KW meters as VSDs record instantaneous kw anyway.
DRWeig;Your logic is sound for variable speed chillers with variable speed cooling towers.I will run multiple towers at reduced speed to get lowest possible condenser water so that chiller can operate at lowest possible speed(whether on full load or part load).Things get tricky when you have a variable speed condenser water pump as well.If the chiller load is only 50% you can reduce condenser water flow as heat rejection is much lower.The question then is whether you reduce speed of both cooling tower fans and condenser water pumps in the same proportion as chiller load or go for some staged control( reduce tower fan speed first to get a desired temperature then reduce pump speed etc.) .With three variables a number of scenarios are possible with one affecting the other as well as the overall result.
You do not need individual KW meters as VSDs record instantaneous kw anyway.
SAK
How are you achieving condenser water pump flow? what will throttle the Pump VSD? no means to choke the flow really, even if one tower is shut-down (in case of multiple towers) water will still flow through the next open tower with higher velocity. There is really ne mneans to throttle the condenser water flow, thus VSD on CWP useless - AM I wrong?
On the topic: I tried asymetric chiller plant once, monitoring amp draw and comparing combination of chiller amp draw. with limited success.
the concept was to use say two 300-ton centrifugal chillers, one 500-ton chiller and one 125-ton screw chiller.
You try to match the load with best combination of chiller amp draw. say you have 430-ton load, you compare whether one 300-ton and one 125-ton screw at full load is more efficient than one 500-ton at part load.
Of course, you'd need to collect trending data on Amp draw at various loads to be able to do such comparison.
How are you achieving condenser water pump flow? what will throttle the Pump VSD? no means to choke the flow really, even if one tower is shut-down (in case of multiple towers) water will still flow through the next open tower with higher velocity. There is really ne mneans to throttle the condenser water flow, thus VSD on CWP useless - AM I wrong?
On the topic: I tried asymetric chiller plant once, monitoring amp draw and comparing combination of chiller amp draw. with limited success.
the concept was to use say two 300-ton centrifugal chillers, one 500-ton chiller and one 125-ton screw chiller.
You try to match the load with best combination of chiller amp draw. say you have 430-ton load, you compare whether one 300-ton and one 125-ton screw at full load is more efficient than one 500-ton at part load.
Of course, you'd need to collect trending data on Amp draw at various loads to be able to do such comparison.
- Status
Similar threads
- Question