A conventional meter tells you how many units you used, which is exactly the information you cannot act on. This project clamps a current transformer around the incoming line, samples voltage and current together so real power and power factor are measured rather than assumed, streams the readings to a dashboard, and then goes one step further: it watches for step changes in the load and matches them against a library of appliance signatures, so the log says which appliance turned on rather than just that consumption rose. The identification is deliberately scoped to a small trained set of appliances, because that is what is honestly achievable from a single measurement point.
Not just the parts. Teams budget from a component list, then run out in week six on wire, connectors and the driver board they destroyed. These are the lines that catch them.
| Line | Cost |
|---|---|
Components and materials From the bill of materials | Rs 2,190 to Rs 6,450 |
Spares and replacements 15 % — you will destroy something | Rs 329 to Rs 968 |
Consumables and hardware Wire, connectors, headers, screws, heatshrink | Rs 263 to Rs 774 |
Perfboard and assembly | Rs 150 to Rs 400 |
Report, printing and binding Two or three bound copies, plots and the plagiarism check | Rs 400 to Rs 1,200 |
| Total | Rs 3,332 to Rs 9,792 |
| Each, split 3 ways | Rs 1,111 to Rs 3,264 |
Indicative bands last checked in August 2026 — a guide, not a quotation. Prices move and vary by seller, and anything you can borrow from the lab comes straight off the total. The 3.5 week figure assumes 8 person-weeks of work shared between 3; students past 4 add coordination rather than output, so the estimate stops improving there.
The same chain in words
The current transformer produces a small current proportional to the line current, and the burden resistor converts it into a voltage the ADC can read. A bias network lifts it to the middle of the ADC range so both half cycles are captured.
Voltage and current are sampled at the same instants. Multiplying sample by sample and averaging over whole cycles gives real power directly, which is the only honest way to get it — multiplying RMS volts by RMS amps gives apparent power, and the two are not the same on any real load.
Energy is the integral of real power over time, accumulated locally so a network outage never loses units.
When the real power steps by more than a threshold and then stays steady, the detector records the size of the step, the change in power factor, and the shape of the first second of the transient.
That signature is compared against the trained library. A resistive heater steps power up with power factor near one; a fridge compressor steps up with a poor power factor and a large brief inrush; an LED lamp array steps up a small amount with a distinctive distorted current waveform.
This is the chapter that separates an engineering project from an assembly job. Every value below carries the formula that produced it and the assumptions it rests on — check them against your own ratings before you use them, because your motor is not this motor.
R burden = desired peak voltage / peak secondary current
An SCT-013 rated 100 A to 50 mA measuring 30 A gives 15 mA RMS, which is 21.2 mA peak. For a 1.5 V peak swing around the bias point: 1.5 / 0.0212 = 70.7 Ohm. Choosing the burden for the actual expected current, not for the CT's full rating, is what gives the measurement its resolution.
Two equal resistors from the rail to ground, decoupled by a capacitor
The CT output is bipolar and the ADC cannot read below ground. The decoupling capacitor is what stops the bias point moving with the signal; leave it out and the readings distort at high current.
samples per cycle = fs / f
Enough to compute real power accurately for the fundamental and the low harmonics. Resolving the current waveform shape of an electronic load properly needs more, which is why the signature stage uses a higher rate burst.
P = average of (v x i) over whole cycles; S = Vrms x Irms
On a rectifier load the current is far from sinusoidal, so the ratio of these two is not the cosine of any phase angle. Computing both and quoting the ratio as power factor is correct; calling it cos phi is not.
Set above the noise floor of the measurement
Below this, mains fluctuation and small loads produce constant false events. It also honestly bounds the project: appliances below about 40 W will not be identified, and the report should say so rather than implying every load is detected.
samples per day = rate x seconds per day
That is why the device uploads averages and events rather than raw samples. Storing a year of one-second data for a single home is around thirty million rows, which is a database design decision, not an afterthought.
| Item | Qty | Unit cost |
|---|---|---|
Current transformer SCT-013-000, 100 A to 50 mA split core | 1 | Rs 400 to Rs 900 |
Voltage sensor module ZMPT101B, 250 V AC input | 1 | Rs 120 to Rs 260 |
Microcontroller board ESP32 devkit with WiFi | 1 | Rs 350 to Rs 750 |
Burden and bias components 68 Ohm 1 %, 10 kOhm pair, 10 uF decoupling | 1 | Rs 30 to Rs 90 |
Reference meter PZEM-004T, for calibration and accuracy verification | 1 | Rs 600 to Rs 1,400 |
Display 0.96 inch OLED or 16x2 LCD | 1 | Rs 140 to Rs 400 |
SMPS power supply 5 V, 1 A, enclosed | 1 | Rs 150 to Rs 350 |
Enclosure DIN rail ABS box with cable glands | 1 | Rs 250 to Rs 700 |
Wiring and terminals Terminal blocks, ferrules, cable ties | 1 | Rs 150 to Rs 400 |
Test appliances A heater, a fan and an LED lamp set for signature training Household items — usually costs nothing. | 1 | Rs 0 to Rs 1,200 |
Specifications are written as ratings rather than brands, so any equivalent part works. Where a line says to borrow from the lab, do — it is usually the largest number on the page, and the total drops to Rs 3,332 at the low end without them.
None of these blocks you — they are learnable inside a semester. Each one you do not have costs a couple of weeks, so count them into your plan.
Anything marked required is a blocker, not a difficulty. Arrange the access before you commit to the project, not after.
Use these when your guide says the scope is too small, or when you want something in the results chapter that nobody else in your batch will have.
The structure below is what a project report is marked against, with the mistake that costs marks in each chapter. There is no template to download and there will not be one: the writing is your work, and a site that did it for you would be helping you cheat rather than helping you pass.
A diploma report is typically 40 to 60 pages. The design and results chapters carry the marks; the literature survey can be short. A mini project report runs to about 40 to 50 pages, and the literature survey is usually brief. Guides expect one clear objective, met and measured. A final year report is normally 60 to 90 pages with a substantial literature survey, and many departments expect a paper submission alongside it.
The full report guide — front matter, conventions and page countsOne paragraph covering what the problem is, what you built, and what the result was. Written last, even though it appears first.
Loses marks: Describing the problem and the method but never stating a result. If there is no number in your abstract, it is not finished.
Why the problem matters, the context it sits in, and a clear statement of the objectives your project set out to meet.
Loses marks: Objectives written so vaguely that nobody can tell at the end whether they were met. Write objectives you can tick off in the conclusion.
What has already been done, what each approach achieved, and the specific gap your work addresses.
Loses marks: A list of paper summaries with no comparison and no gap. The survey exists to justify your design choice, so end it by saying what you chose and why.
The system as a whole: the block diagram, what each block does, and the signal or power path between them.
Loses marks: A block diagram copied from a datasheet or another report. Draw yours to match what you actually built, including the parts that are inconvenient.
Every component value and rating, with the formula, the assumptions and the arithmetic that produced it. This is the chapter that distinguishes an engineering project from an assembly job.
Loses marks: Component values with no justification. If a resistor is 10 kOhm, the report must say why it is not 1 kOhm.
The circuit as built, the PCB or wiring, the flowchart, and the parts of the code that carry the actual logic.
Loses marks: Twenty pages of pasted source listing. Include the flowchart and the few functions that matter, and put the full listing in an appendix.
What you measured, in tables and graphs, compared against what you predicted in the design chapter — including where they disagreed.
Loses marks: Photographs of the working model presented as results. A photograph is evidence that it exists, not evidence that it works. Results are measurements.
Whether each objective was met, what the measured performance was, and what the honest next step would be.
Loses marks: Future scope written as a wish list of unrelated features. It should follow from a limitation you actually hit.
Every source in a consistent style, plus datasheets, full code listings and the complete bill of materials.
Loses marks: Citing a blog or a video where a datasheet or a standard exists. Cite the primary source for anything you relied on.
The demonstration is half the assessment. These are the questions this project invites, and the answers are written the way you would give them out loud.
Because energy is the integral of real power, and real power is the average of the product of instantaneous voltage and current. Multiplying RMS voltage by RMS current gives apparent power, which is larger than real power whenever the load is reactive or distorted. Measuring only current and assuming a fixed voltage would overstate consumption on every motor load in the house.
A current transformer is a current source: its secondary tries to force a current proportional to the primary current. The burden resistor gives that current somewhere to flow and converts it into a measurable voltage. Without a burden the secondary is an open circuit, the core saturates and dangerously high voltages appear at the terminals.
Power alone is not enough, so the signature uses more dimensions: the power factor of the step, the shape and duration of the turn-on transient, and the harmonic content of the current. A 1000 W heater and a 1000 W motor look identical in real power but completely different in power factor and inrush. Where two loads genuinely match on every dimension, they cannot be separated from a single measurement point, and the report should say so.
It is inferring which individual appliances are running from a single measurement at the supply point, instead of instrumenting every circuit. The attraction is that one sensor covers a whole building. The difficulty is that the measurement is a sum, and separating a sum into its components is under-determined in general — which is why practical systems restrict themselves to a known set of appliances with distinct signatures.
Because the network is the least reliable part of the system. If energy is only integrated on the server, every WiFi outage becomes a gap in the meter reading. Accumulating locally in non-volatile memory and uploading the running total means an outage delays the data but never loses units.
Normally taken as a Diploma project (C-20, 6th sem) or B.E. mini project (5th/6th sem) or B.E. final year project (7th/8th sem) project. If yours is a different course, the extensions above scope it up and the simpler half of the design scopes it down — the arithmetic does not change. Back to all projects.