Conservation voltage reduction

[Brentst]

I'm working on a cvr project on a small rural cooperative. I was hoping someone might be able to share their experiences with conservation voltage reduction. Specifically I've been wondering:
1) do you control the voltage primarily from the regulators at the substation or do you control down line devices (regulators/capacitors)
2) what has your experience shown the best bang for the buck? Ive lowered voltage 1, 2 and 3% and found that for a short period I reduced demand by greater then 10%. Is that common?
3) Are there particular loads that seem to respond better than others? How do motor loads responds?
4) is there a time limit to the cvr? I have heard some people believe that using cvr for greater then 3hrs has no benefits and the load crawels back to normal levels.
5) how do you decide to use cvr? Temperature or load forecasting?

I appreciate any help you can provide!!

 

[magoo2]

We control voltage mostly from the substation level through LTC and switched capacitors.

My impression is that you can expect to get roughly a 1:1 situation, that is, getting 1% demand reduction for a 1% voltage reduction. If you're getting 3% demand change for a 1% voltage change, then you're getting additional load dropping off for other reasons and not simply as a response to the voltage reduction.

There's an EPRI report on CVR that you should be able to get through NRECA.

A lot of your other questions are a little difficult to answer. When you switch on CVR, you should be comparing it to a similar day in terms of temperature, humidity and energy usage. That's the only fair comparison.

What are you trying to achieve with your CVR program? Energy reduction? Demand reduction?

 

[Brentst]

Thank you for your response! Our biggest goal is to reduce demand. I did notice a better then 1:1 relationship as measured from the feeder regulators on a cold day. I imagined that a cold day would be the best day to do cvr. Do you recall the title of the epri report?

Thanks!

 

[davidbeach]

If you have resistive loads then cvr might be worth something. But that isn't generally the case.

Any load with a drive or switch mode power supply will function as a constant power load and your losses will go up as the current goes up in response to reduced voltage. Directly connected motors tend to exhibit constant power properties around nominal voltage also eliminating the possibility of gains from cvr.

Snake oil, I tell you, snake oil.

 

[dpc]

Reducing voltage to lower peak demand is not really CVR. In fact, it is sort of the opposite. A CVR scheme is based on lowering normal voltage on the feeder all the time. It generally involves implementing line drop compensation in the substation voltage regulation and lowering the setpoint voltage in the substation. I'm not really a believer, but there are studies that show some percentage of actual load reduction when voltage is reduced. It is not 1:1 for the reasons David mentions above. However, it is fair to say there is some actual reduction in energy consumption if the average voltage on the feeder is reduced. Of course, this also means reduced revenue for the utility. Under a CVR scheme, voltage is increased during peak loads, not reduced, in order to avoid dropping below the ANSI voltage window.

The problem with demand reduction schemes is that they have to be perfect. In my part of the world, there is a 12 month ratchet on peak demand. So if your demand reduction scheme misses one peak hour, you're screwed for the next year.

 

[magoo2]

Well you've gotten a reality check on CVR. All it says to me is there are different flavors to CVR. I don't think you need to employ it all the time. We look at the incremental cost curve of energy costs throughout each hour of the day. This tends to favor near peak times, but there is also an opportunity offpeak to reduce losses since transformer core losses are higher with higher voltage. You also need to be aware of payback where the thermostatically controlled loads like a/c, space heating and water heating will lose diversity and be on for longer periods if you reduce voltage for 3 hours or more.

The trick is to get some smart meters (smarties) so you can actually see what your voltage profile is. Common voltage regulation practice seems to assume a primary drop in the feeder, make allowance for distribution transformers and secondary and assume you've got it. What we've seen with our smarties is that some customer voltages may already be outside of range. And those customers with low voltages may be closer to the substation than you think. Without these meters, you really don't know how far you can go with your CVR.