MATLAB scripts or models for grid frequency measurement in demand side management?

153 12

#1 21683808 18 Jan 2020 18:53

Josh McNelis Josh McNelis

Anonymous

Post #1
21683808 18 Jan 2020 18:53

I am student doing my dissertation on domestic demand side management through frequency measurement for grid stability. My plan is to make a prototype frequency measuring device apparently there are existing matlab scripts and models for frequency measuring circuits but I can't seem to find them does anyone know where I might find such a thing?
ADVERTISEMENT
#2 21683809 20 Jan 2020 23:03

Elizabeth Simon Elizabeth Simon

Anonymous

Post #2
21683809 20 Jan 2020 23:03

For a simple frequency measuring circuit, all you need is to convert the analog input into a square wave (can be done with a comparitor) and some way of measuring time between rising or falling edges to the desired accuracy. This gives you the period which is easy enough to convert to frequency.
You'll have to look at your microcontroller for the best way to measure the time but for low frequencies an interrupt and a timer may work. For higher frequencies, you can use a counter to detect a number of edges within a given time period.
ADVERTISEMENT
#3 21683810 21 Jan 2020 22:50

Josh McNelis Josh McNelis

Anonymous

Post #3
21683810 21 Jan 2020 22:50

Thanks for your response my circuit will consist of a voltage source, frequency measuring device, and loads (which can be turned on or off via circuit breakers to bring the frequency up or down as required) the arduino uno will see the frequency measurement and react accordingly, do you have any advice?
#4 21683811 22 Jan 2020 03:24

Elizabeth Simon Elizabeth Simon

Anonymous

Post #4
21683811 22 Jan 2020 03:24

Josh,How accurate does the frequency measurement need to be and how fast do you need to make the measurements?
With an arduino uno you will be somewhat limited by the 16 MHz clock frequency which will limit the resolution of the timer if you want to use the interrupt and time approach. Also, there may be some delay before the interrupt is processed that would affect accuracy.Another thing that will affect the accuracy is clock drift in the arduino (or nay other CPU you use). This is not important if you are looking for relative measurements but if you need accurate frequency, you're going to need an accurate time source and/or calibration. For what you are doing, relative measurements are likely good enough but you will need to make note of possible inaccuracies in your dissertation.
ADVERTISEMENT
#5 21683812 22 Jan 2020 05:17

PeterTraneus Anderson PeterTraneus Anderson

Anonymous

Post #5
21683812 22 Jan 2020 05:17

Why would domestic load changes affect the frequency? Is the effect large enough to be separated from simultaneous frequency changes do to other causes?
#6 21683813 22 Jan 2020 07:43

Elizabeth Simon Elizabeth Simon

Anonymous

Post #6
21683813 22 Jan 2020 07:43

Peter,Any load change on a generator driven power system will cause a frequency change due to the momentary change in generator speed caused by the change in load. The change is usually slight and momentary unless the change in load is a significant percentage of the total load. I think other types of power generation are subject to similar frequency changes but am not sure of the mechanism.
I assume that Josh has a small system where he can control what loads are applied and when to affect the stability of the system.
#7 21683814 22 Jan 2020 15:56

Josh McNelis Josh McNelis

Anonymous

Post #7
21683814 22 Jan 2020 15:56

Elizabeth,
You are correct, essentially I will be simulating how the grid reacts to frequency deviations, i.e. when the system frequency rises or falls past the given tolerances the grid must act by either turning on or off certain loads to bring the frequency back within it tolerance. The circuit I will be creating will hopefully be able to emulate this on a smaller simpler scale, obviously the national grids readings are very accurate and they act almost instantly so I was hoping for my circuit to be the same. Do you think this will be possible using an arduino?
ADVERTISEMENT
#8 21683815 22 Jan 2020 15:56

Josh McNelis Josh McNelis

Anonymous

Post #8
21683815 22 Jan 2020 15:56

See Elizabeth's response
#9 21683816 22 Jan 2020 23:16

Elizabeth Simon Elizabeth Simon

Anonymous

Post #9
21683816 22 Jan 2020 23:16

I work for a company who makes monitoring equipment used in the national grids.Most of the equipment uses 32 bit processors. Of course we are concerned with monitoring voltage and current in addition to frequency so there is a lot more processing involved. So it may be possible with an arduino but if I were doing it, I'd be looking at something a bit more powerful. And in your case, I think the biggest improvement would be going to a higher clock frequency on your CPU since that will give you better resolution on your frequency measurements.The Raspberry PI and Beaglebone Black are two that I'd look at. Learning how to use them may be a bit daunting unless you are familiar with Linux but that has the advantage of being able to capture and store raw data for later analysis.
There are also the Teensy boards and other similar ARM based boards that use the arduino IDE. Something like that would give you a step up in processing power without too much of a learning curve.
#10 21683817 27 Jan 2020 15:39

Josh McNelis Josh McNelis

Anonymous

Post #10
21683817 27 Jan 2020 15:39

Hi Elizabeth, thanks again for your response, I am required to use an arduino for this circuit, I will firstly try to create it on the software "Matlab/simulink" before eventually manually reproducing it. Do you have any tips on what the circuit would have to consist of? I will be using 3 phase loads to switch on and off to change the frequency accordingly, which means I will possibly be using circuit breakers for each load to switch them on and off, I will also be using a voltage source.
#11 21683818 28 Jan 2020 01:09

Elizabeth Simon Elizabeth Simon

Anonymous

Post #11
21683818 28 Jan 2020 01:09

In general, you want electrical isolation between the high voltage that you are measuring and controlling and your low voltage arduino circuit.
For the measurement side, you can use a transformer that provides isolation and also steps the voltage down to something your circuit can handle.For the control side, electro-mechanical relays provide isolation as well as a current and/or voltage boost as needed. If your circuit breakers inputs are isolated from the high voltage, you may not need relays for isolation but you will likely need a drive circuit of some kind.EEWeb had several articles a while back on relays and related topics.
#12 21683819 28 Jan 2020 13:06

Maurizio Di Paolo Emilio Maurizio Di Paolo E...

Anonymous

Post #12
21683819 28 Jan 2020 13:06

On EEWeb there are a lot of articles about several projects with Relay https://www.eeweb.com/search?q=relays&source=search-on-site please check your need. Thanks
#13 21683820 29 Jan 2020 19:35

leah melda leah melda

Anonymous

Post #13
21683820 29 Jan 2020 19:35

Students who seek Case Study Writing Services Online from the custom writing company are guaranteed of getting good grades for their Case Study Custom Writing Help and Online Case Study Writing Services that are free from grammatical errors.
Create an account, log in here. You will receive points by participating in discussions.
Join this discussion.

Install Elektroda application

Didn't find an answer? Ask Artificial Intelligence

*I agree to send the question to OpenAI, Anthropic PBC, Perplexity AI, Inc., Kagi Inc., Google LLC - owners of language models in order to prepare the best response. The companies may monitor and log information entered into the form.

*I agree to publicly display my question and answer. The question and answer will be publicly available to everyone. The process may take a few minutes. Upon completion, you will be redirected to the page with the answer.

Wait...(2min)

Topic summary

✨ The discussion focuses on developing a prototype frequency measuring device for domestic demand side management to maintain grid stability by monitoring frequency deviations and controlling loads accordingly. The measurement approach involves converting the analog voltage input into a square wave using a comparator, then measuring the time interval between edges to calculate frequency. An Arduino Uno is proposed for frequency measurement and load control, but its 16 MHz clock and interrupt latency limit measurement resolution and accuracy. Clock drift and timing calibration are also concerns for precise frequency measurement. Alternatives with higher processing power and clock speeds, such as Raspberry Pi, Beaglebone Black, or ARM-based Teensy boards compatible with the Arduino IDE, are suggested for improved resolution and data handling. The circuit design should include electrical isolation between high voltage and low voltage control circuits, typically achieved using transformers for measurement and electromechanical relays or drive circuits for load switching. Simulation in MATLAB/Simulink is recommended before hardware implementation. The system aims to emulate grid frequency response by switching three-phase loads via circuit breakers to stabilize frequency within tolerance limits.

FAQ

TL;DR: Use a comparator to square the mains sample, then time edge-to-edge with your MCU; Arduino Uno’s 16 MHz clock limits resolution, and “Any load change on a generator driven power system will cause a frequency change.” [Elektroda, Anonymous, post #21683813]

Why it matters:** This FAQ helps students and hobbyists build and simulate a safe, Arduino-based grid‑frequency monitor for demand‑side control.

Quick Facts

Arduino Uno clock: 16 MHz; interrupt latency and clock drift impact accuracy and timing. [Elektroda, Anonymous, post #21683811]
Simple method: comparator → square wave; measure period via edge timing or count edges in a gate window. [Elektroda, Anonymous, post #21683809]
Isolation is essential: use a sensing transformer and relay or isolated driver for breakers. [Elektroda, Anonymous, post #21683818]
More headroom: Teensy/Raspberry Pi/Beaglebone increase timing resolution and enable data capture. [Elektroda, Anonymous, post #21683816]
Small load steps shift frequency slightly; effects grow only with large system‑relative changes. [Elektroda, Anonymous, post #21683813]

How do I measure grid frequency with an Arduino?

Condition the mains sample safely, feed it to a comparator to create a clean square wave, and use the MCU timer to measure time between rising (or falling) edges. Convert period to frequency or count edges in a fixed time window. This approach is simple, fast, and works across wide frequency ranges if your timing resolution is adequate. [Elektroda, Anonymous, post #21683809]

What accuracy and speed can I expect from an Arduino Uno?

The 16 MHz clock sets base resolution (one tick ≈ 62.5 ns), but real accuracy depends on interrupt latency and the Uno’s clock drift. For relative changes, it works well; for absolute frequency, add calibration or a better timebase. “You’re going to need an accurate time source and/or calibration.” [Elektroda, Anonymous, post #21683811]

Do I need galvanic isolation for measurement and switching?

Yes. Isolate the measurement path with a small transformer that also steps voltage down. Isolate control paths with electromechanical relays or appropriately rated drivers for breaker coils. Even if a breaker’s input is isolated, you may still need a driver stage to meet coil current or voltage. [Elektroda, Anonymous, post #21683818]

Can small domestic loads really change grid frequency?

On a large interconnected grid, single‑home load changes cause only slight, momentary frequency shifts. In a small, controlled lab system, deliberate load steps are visible and useful for experiments. “Any load change … will cause a frequency change,” but magnitude depends on the system’s size and inertia. [Elektroda, Anonymous, post #21683813]

Is Arduino enough, or should I use a faster board?

Arduino can work, but higher‑clock MCUs or SBCs improve timing resolution and processing. Teensy, Raspberry Pi, or Beaglebone provide more headroom; Linux SBCs also simplify raw data capture for later analysis. If you’re constrained to Arduino, optimize timers and interrupts carefully. [Elektroda, Anonymous, post #21683816]

How should I model this in MATLAB/Simulink before building?

Create blocks for a voltage source, frequency measurement, controllable three‑phase loads, and breaker logic. Define frequency thresholds to trigger shedding or restoration. Validate timing and control logic in simulation, then port thresholds and logic to your Arduino for hardware testing. [Elektroda, Anonymous, post #21683817]

What is a comparator and why use it here?

A comparator outputs a digital high or low depending on the input’s instantaneous voltage. With an AC sample, it turns a noisy sine into a clean square wave with sharp edges. That clean edge makes timer‑based period or edge‑count measurement stable and repeatable. [Elektroda, Anonymous, post #21683809]

How do I switch three‑phase loads safely from a microcontroller?

Keep the MCU at low voltage. Use isolated relay interfaces or properly rated drivers to actuate contactors or breakers. Verify the breakers’ control inputs; some are isolated low‑voltage coils, others need specific drive currents and protection. Maintain isolation barriers between mains and logic. [Elektroda, Anonymous, post #21683818]

Quick how‑to: measuring frequency on Arduino Uno (3 steps)

Square the conditioned AC sample with a comparator and feed a digital pin.
Use an external interrupt to capture edges and read a hardware timer.
Compute frequency from period or count edges within a fixed gate time. [Elektroda, Anonymous, post #21683809]

How do I choose thresholds to shed or restore loads?

Define upper and lower frequency limits around nominal. When frequency drops below the lower limit, shed selected loads; when it rises above the upper limit, restore them. Your prototype can emulate grid logic by reacting “when the system frequency rises or falls past the given tolerances.” [Elektroda, Anonymous, post #21683814]

What’s an effective control loop for demand‑side response on a bench setup?

Measure frequency continuously, apply hysteresis on thresholds to avoid chatter, and switch loads in priority order. Log frequency and switching events to tune thresholds. Start with coarse steps, then refine timing and load sizes for stability and minimal overshoot. [Elektroda, Anonymous, post #21683810]

How can I log data for my dissertation?

Use a faster MCU or a Linux SBC to timestamp frequency, voltage, and switch states. Store raw data to files for replay and analysis. This makes it easier to validate controller behavior and produce publication‑quality plots. [Elektroda, Anonymous, post #21683816]

What is the Arduino Uno in this context?

It’s a 16 MHz microcontroller board suitable for timing edges and controlling relays. Its limited clock speed and non‑disciplined oscillator affect frequency resolution and absolute accuracy, so consider calibration or a higher‑quality clock if needed. [Elektroda, Anonymous, post #21683811]

What edge cases or failures should I plan for?

Interrupt storms or long ISRs can delay edge capture and distort period measurements. Oscillator drift skews absolute frequency. Mitigate by minimizing ISR work and calibrating timebase. Document these sources of error and their impact in your report. [Elektroda, Anonymous, post #21683811]