Energy
18 columns
105 rows
8 formats

Industrial Process Load Disaggregation

This dataset provides granular, timestamped electrical consumption data for industrial facilities, disaggregated into motors, HVAC, compressed air, lighting, process heating, and ancillary loads using NILM algorithms. It enables facility managers, energy analysts, and researchers to identify energy-saving opportunities, monitor operational efficiency, and benchmark end-use consumption patterns across multiple sites.

last updated 2 Nov 2025
by GoMask

The brief that made it

Facility-level energy benchmarking and reporting

Sample rows

preview · 8 of 105 rows · all 18 columns

facility_idstring	total_consumption_kwfloat	address_statestring	facility_namestring	facility_typestring	address_streetstring	address_citystring	address_postal_codestring	address_countrystring	timestampdatetime	motors_consumption_kwfloat	hvac_consumption_kwfloat	compressed_air_consumption_kwfloat	lighting_consumption_kwfloat	process_heating_consumption_kwfloat	ancillary_consumption_kwfloat	nilm_methodstring	data_quality_flagboolean
FAC-0001	320.8	CA	Green Valley Foods	food processing	1024 Lincoln Ave	Fresno	93722	USA	2024-03-02T08:47:00	110.5	64.2	32.6	18.7	82.4	7.1	Sequence-to-sequence	true
FAC-0002	1450.7	TX	Horizon Manufacturing	manufacturing	215 Sunset Blvd	Houston	77018	USA	2024-03-14T19:36:00	990.1	166.2	102.5	49.3	122.8	19	Hierarchical Clustering	true
FAC-0003	730.2	GA	Blue Ridge Plastics	plastics	6750 Chandler Rd	Atlanta	30349	USA	2024-02-22T12:29:00	410.6	86.3	40.5	27.2	126.1	18.9	Unsupervised Learning	true
FAC-0004	0	OH	ChemCore Solutions	chemical	480 Industrial Park	Toledo	43612	USA	2024-03-05T23:15:00	0	0	0	0	0	0	Manual Tagging	true
FAC-0005	128.5	WI	Innovate Labs	laboratory	9009 Science Dr	Madison	53717	USA	2023-12-27T15:03:00	27	43.1	2.9	19.4	20.8	15.3	Matrix Factorization	true
FAC-0006	2720.6	WA	Evergreen Paper Mill	paper mill	1521 Maple St	Tacoma	98421	USA	2024-03-17T11:58:00	1812.2	188.4	265.6	93.3	278.7	82.4	Viterbi Algorithm	true
FAC-0007	510.4	NC	Textile Works	textile	700 Elm Ave	Charlotte	28217	USA	2023-11-12T09:20:00	212.1	76.9	22.5	29.6	139.7	19.6	Dictionary Learning	true
FAC-0008	104.7	OH	CleanFlow Water	water treatment	1440 River Rd	Cleveland	44108	USA	2024-01-23T21:34:00	36.5	19.2	3.1	16.7	18.8	10.3	Deep Neural Network	true
FAC-0009	392.2	CA	Techtron Electronics	electronics	6432 Innovation Blvd	San Jose	95131	USA	2024-03-09T06:12:00	188.5	63.4	14.6	21.5	80.1	12.4	Bayesian Inference	true
FAC-0010	485.1	CO	Precision Mechanical	mechanical	2040 Industrial Way	Denver	80239	USA	2024-02-16T13:19:00	223.7	70.5	10.7	24.2	111	17.8	Deep Neural Network	true
FAC-0011	82.3	IA	Harvest Fields	agriculture	39 Farmland Rd	Des Moines	50313	USA	2024-03-10T18:05:00	27.1	24.6	2.4	12.5	8.7	5.6	Convolutional Neural Network	true
FAC-0012	3140.3	PA	Liberty Steel	steel	4800 Liberty Ave	Pittsburgh	15201	USA	2024-03-18T05:44:00	2031.6	212.5	362.7	78.4	368.2	67.1	Hierarchical Clustering	true
FAC-0013	283.9	IN	PharmaGenix	pharmaceutical	7600 Health Park Dr	Indianapolis	46256	USA	2023-11-29T17:51:00	132.1	78.4	6.8	19.9	31.7	8.4	Dictionary Learning	true
FAC-0014	3562.9	AZ	Mountain Mining	mining	1834 Quarry Rd	Flagstaff	86001	USA	2023-12-03T02:23:00	2142.6	235.3	431.8	109.2	555.7	88	Viterbi Algorithm	true
FAC-0015	94.8	TX	Urban Construction	construction	3010 Cedar St	Austin	78745	USA	2024-01-20T09:11:00	36.3	8.6	13.2	7.8	2.7	6.9	Spectral Analysis	true
FAC-0016	248.2	PA	PackRight Packaging	packaging	5555 Market St	Philadelphia	19123	USA	2024-02-16T22:55:00	81.7	49.5	13.8	22.1	61.4	11.3	Bayesian Inference	true
FAC-0017	0	CA	Sunrise Dairy	dairy	2288 Country Ln	Chico	95926	USA	2024-03-13T17:21:00	0	0	0	0	0	0	Factorial Hidden Markov Model	true
FAC-0018	0.1	MO	Baker's Best	bakery	1201 Main St	St. Louis	63101	USA	2023-12-18T05:28:00	0	0	0	0.1	0	0	Hybrid	true
FAC-0019	415.3	CA	Fresh Fields Foods	food processing	2000 Walnut Ave	Modesto	95350	USA	2024-03-07T10:42:00	162.9	75.6	39.3	21.7	102.4	11.6	Deep Neural Network	true
FAC-0020	290.7	CA	PrimePack Foods	food processing	8808 Harvest Dr	Salinas	93906	USA	2024-02-09T16:40:00	92.3	53.9	27.6	19.9	83.2	7.6	Sequence-to-sequence	true

What the 105 rows show

from the 105-row sample

367.4median total_consumption_kw
14nilm methods
18facility types
132.1median motors_consumption_kw
64.2median hvac_consumption_kw
17.7median compressed_air_consumption_kw

total_consumption_kw105 rows, in bands of 500

Median 367.4, from 0.0 to 3,563.

address_state99 rows with a value · 6 left blank

CA13
OH8
TX6
NC6
CO5
IN5
WI4
WA4
NV4
IA3

18 columns by typefrom the column list below

string 9
float 7
datetime 1
boolean 1

Columns

18 columns in four groups

blueprint · 18 columns
column	type	description	example
Text 9 columns
`facility_id`	string	Unique identifier for the industrial facilityunique	FAC-0001
`facility_name`	string	Name of the industrial facilityoptional	Green Valley Foods
`facility_type`	string	Type of industrial facility (e.g., manufacturing, food processing)optional	food processing
`address_street`	string	Street address of the facilityoptional	1024 Lincoln Ave
`address_city`	string	City where the facility is locatedoptional	Fresno
`address_state`	string	State or province of the facilityoptional	CA
`address_postal_code`	string	Postal code of the facilityoptional	93722
`address_country`	string	Country where the facility is locatedoptional	USA
`nilm_method`	string	NILM (Non-Intrusive Load Monitoring) algorithm or method used for disaggregationoptional	Sequence-to-sequence
Numbers 7 columns
`total_consumption_kw`	float	Total electrical consumption in kilowatts at the given timestamp0 or more	320.8
`motors_consumption_kw`	float	Electrical consumption attributed to motors in kilowatts0 or more	110.5
`hvac_consumption_kw`	float	Electrical consumption attributed to HVAC systems in kilowatts0 or more	64.2
`compressed_air_consumption_kw`	float	Electrical consumption attributed to compressed air systems in kilowatts0 or more	32.6
`lighting_consumption_kw`	float	Electrical consumption attributed to lighting in kilowatts0 or more	18.7
`process_heating_consumption_kw`	float	Electrical consumption attributed to process heating in kilowatts0 or more	82.4
`ancillary_consumption_kw`	float	Electrical consumption attributed to ancillary loads in kilowatts0 or more	7.1
Dates and times 1 column
`timestamp`	datetime	Date and time of the load measurement	2024-03-02T08:47:00
True or false 1 column
`data_quality_flag`	boolean	Indicates if the data point passed quality checks (true = good, false = questionable)optional	true

Use it for

An energy dashboard
Total_consumption_kw and a breakdown of address_state. Excel, Power BI or Tableau.
What pattern can you find in 105 rows?
A class exercise
Hand out the rows and one question. Everyone works from the same 105 rows.
Facilities105
FAC-0001320.8
FAC-00021450.7
FAC-0003730.2
A software demo
Believable facilities with facility_name, facility_type and address_street to fill a screen in front of a buyer.

This dataset105 rows18 columns

Yours1,000 rows18 columnsaddress_street: UK only

blueprint · industrial-process-load-disaggregation

Behind this dataset

Same schema. As many rows as you need.

These 105 rows came out of a blueprint — 18 columns with generation rules behind each one. Open it in Data Factory to retune a column, add your own, wire in foreign keys, and run it at the size you actually need.

Rules it was built with

Facility ID and NAICS industry code
Total kW demand and disaggregated end-use breakdown
End-use categories: motors (30-40%), HVAC, compressed air, lighting, process heat
NILM algorithms: supervised learning, HMM, neural networks
Accuracy metrics for disaggregation (mean absolute error)
Energy efficiency opportunities identified per end-use
Benchmark consumption against similar facilities (Energy Star)
Retro-commissioning and behavioral recommendations

Rows1,000 rows10,000 rows100,000 rows1,000,000 rows

Open the blueprint in Data Factory

1 credit per row. Preview 20 rows free. Buy rows for the full run.

Exports: CSV, JSON, JSONL, Parquet, SQL, Excel, TSV, XML
Licence: yours to use, including commercially
API slug: industrial-process-load-disaggregation

What should your data show?

No signup. No card.

Sample rows

What the 105 rows show

Columns

Use it for

An energy dashboard

A class exercise

A software demo

Make it yours.

Same schema. As many rows as you need.

More in Energy

What should your data show?