Induction motors inertia contribution 3

[Carlos Melim]

Good morning everybody.

Some grid stability studies neglect the demand side inertial contribution. Only compute the synchronous generators inertial contribution.

Other studies refer that, on the demand side, there are synchronous motors that must be considered.

The majority of electrical motors on the grid are asynchronous, induction type.

Do induction motors contribute to the grid inertia?

I know that if we install VFDs, they are decoupled from the grid and their inertial contribution is lost.

I didn´t find any document that addresses this subject so all your comments and expertise will be greatly appreciated.

Best regards.

Carlos Melim

 

[bacon4life]

Within the western portion of the USA, grid stability studies use a Composite Load Model that tries to account for dynamic behavior of typical loads, including the inertia of induction motors and the dynamic characteristics of VFDs. This model includes three separate polyphase motor equivalents to account for differing behavior of compressors vs fans vs pumps. Additionally, it includes modeling for single phase motors such compressor motors found in residential HVAC units. Including single phase motors has been quite important to studying fault-induced delayed voltage recovery (FIDVR).

In order to parameterize the regional model, each utility categorizes each circuit as rural, suburban, urban, industrial, residential, commercial, or a mix of these. For each of category of load, there are region specific estimates for the portion of expected loads (i.e. compressors, fans, pumps, resistance heat, VFD, lighting, etc) for each category based on the time of day, day of week, and season.

I think it is time to start building synthetic inertia into "smart" electronics, LED drivers, and perhaps some VFDs. Having TVs, computers, thermostats and lighting all start responding proportionally to frequency disturbances would be less disruptive than triggering entire substations to disconnect via Under Frequency Load Shedding (UFLS).

 

[waross]

I think that it is long past the time for the approvals of refrigeration equipment to require a delayed start.
Long ago and far away, when I worked in wye/delta land, we protected A/Cs and freezers and refrigerators from burnout with a relay and a push button. When the power failed, the relay would drop out.
The owner had to push the push button to reset the relay.
In later years, plug-in refrigeration protectors became available.
They would drop out on over voltage or under voltage.
There was a three minute delay on restarting.

Weekend outages were common.
Power was commonly restored one phase at a time.
On any system with a wye/delta transformer bank, when one phase is energized, the other two phases get about 50% voltage on the wye/delta back-feed.
Eventually the thermal trip would take them off-line.
The overwhelming majority of hermetic compressor burnouts occurred on Sunday afternoons, when power was restored after a planned outage.



Bill
--------------------
"Why not the best?"
Jimmy Carter

 

[mb3928]

waross, less true for commercial refrigeration, but FWIW many residential HVAC appliances, especially PTACs which get deployed in large numbers in multifamily and hotel buildings, include a controller that introduces a random delay-on-make to the compressor circuit after power loss.

 

[waross]

Grid stability and induction motor contributions may be apples and oranges.
Grid stability. If the grid voltage drops, the current demand of induction motors will increase. This may contribute to further voltage drops.
This is a stability issue.
Induction motor contribution to the grid:
Induction motors will start to contribute to the grid when the HZ/RPM of the grid drops below the RPM of the motor. In many instances of deteriorating conditions, the motor speed may decay faster than the grid frequency.
The grid may be load shedding before the frequency drops enough for induction motor contribution to be important.
The issue with induction motor contribution is in the event of faults when a fault effectively drops the grid frequency to zero Hz.


Bill
--------------------
"Why not the best?"
Jimmy Carter

 

[waross]

One effect may be in the case that the grid frequency decays faster than the motor speed decays, the motor demand on the grid may be less.
In real life I don't expect this to be the case very often and not with a large enough percentage of the motor base to be a concern.

Bill
--------------------
"Why not the best?"
Jimmy Carter

 

[bacon4life]

Induction motors respond to both the rate of change of frequency (i.e. inertia response) and to changes in steady-state frequency whereas VFDs always draw constant power. More detailed load models have evolved each time folks found that existing models failed to accurately predict power grid behavior. Within the WECC footprint, simple load models (i.e.% constant current, % impedance, % constant power) have little predictive value. Your mileage may vary when looking at other grids.

Going a bit off topic...induction motor behavior is also important when considering over-frequency events. For example, if the California-Oregon inter-tie trips off while Oregon generators are exporting lots of power to California, the northern Oregon island will speed up while the southern California island will slow down. BPA has a 1400 MW resistor to function as a dynamic brake to keep Oregon from accelerating too much. Induction motors with fan loads use more energy as the frequency goes up, so the fan loads actually provide negative damping to system.

Going slightly into the weeds... Although, I agree that induction motors very rarely export power except on rare occasions that the grid frequency declines very fast and very far (i.e. below 58.3 Hz), it is unfair say that motor inertia does not contribute to grid stability. Even without exporting power, high inertia motors actually start contributing to grid stability as soon a frequency deviations begins. As the system frequency drops, the slip will decrease and motor will temporarily draw less power from the grid.

 

[waross]

Hello, bacon4life.
I am under the impression that induction motors will contribute by reduced demand under two conditions.
1. The frequency decays faster than the motor speed.
2. The load reduces the demand with a drop in frequency and speed. (Good example, centrifugal fans)
Do I have this right or is this a learning day for me?

Bill
--------------------
"Why not the best?"
Jimmy Carter

 

[davidbeach]

I thought the brake was a mere 1200MW. Is there a good reference source?

I’ve seen the frequency plots of major events. Initial response in the PNW (Pacific North West) is rapid acceleration. Then the RAS schemes kick in, dumping generation and inserting the brake. The brake is only in for up to 30 cycles, but the time it’s dropped the frequency is below pre-event and the reserves can bring it back to normal. Way up, way down, and a return to normal. Fortunately it’s years between examples. Don’t even remember when the last good example occurred. Back of enveloped it once, came up with about $6.00 worth of power, wholesale, consumed by the brake but a huge impact.

 

[bacon4life]

Bill,
A induction motor will reduce demand any time the frequency declines, even when the system frequency change is very slow. Think about a perfectly efficient motor driving a constant power load. At constant frequency, the power taken from the grid is equal to the power consumed by the load. When the frequency changes, the power taken from the grid is the sum of power consumed by the load plus the power accelerating the rotor. Granted, for a very slow change in frequency, the magnitude of power accelerating the rotor will be quite small.

For hand calculating a first swing on a small system, ignoring motor inertia for slow frequency changes seems like a reasonable simplifying assumption. For the WECC regional model that runs for more than 1800 cycles, this effect is built into the differential equations the computer simulates.

David
This paper was a great reference ... in 1975

 

[waross]

Thanks bacon4life. I accept that.
I was typing when I should have been thinking and didn't clearly clarify my post.
I should have included a note that I was considering the inertial contribution to reduced demand, not the steady state reduction in demand.
I think that we are on the same page.
Your addition to my post, as originally written, is valid.
Thanks for keeping me honest.

Bill
--------------------
"Why not the best?"
Jimmy Carter

 

[davidbeach]

Gee, might have to update my calcs to say that the brake uses $6.25 worth of power instead of $6.00.