Uploading data
Accurate asset generation forecasts depend on historical generation data, so upload as much of it as you can. You upload generation data, and demand data for assets with a load component, on the Uploads page in the main sidebar.
Uploading historical generation data
Before uploading, check that the location ID of each asset is set correctly. For meter data, Tensor Cloud matches the uploaded data to assets by location ID.
- On the Uploads page, click "Upload data" in the top right.
- Choose "Meter data", "Monitoring data", or "Demand data". Meter data can be uploaded for solar, hydro, and biomass assets.
- For "Monitoring data" and "Demand data", choose the single asset the data belongs to. Only assets with a solar component are listed for monitoring data, and only assets with a load component for demand data. Then click "Next step".
- Add your files and click "Import". Each file can be up to 500 MB, and you can add up to 400 files at once.
Processing starts as soon as the upload finishes. The Uploads page refreshes automatically to show the result.
Tensor Cloud ignores files in a format not listed under supported data types, and files larger than 500 MB. They do not appear in your list of uploaded files. Meter data accepts CSV and XML files; monitoring and demand data accept CSV files only. If you have large amounts of data in another file format, contact Tensor support.
Processing status
Before and during processing, a file is shown as Uploading..., Waiting to process, or Processing. If the file itself could not be uploaded, it is marked as Uploading failed. Once processing finishes, the file has one of these statuses:
- Success: Tensor Cloud read the data and matched it with an asset in your workspace.
- Processing failed: Tensor Cloud could not read the data, or could not match it with any asset in your workspace.
- Partially processed: For data types that can hold data for several assets (confirmed and preliminary meter data), Tensor Cloud read the data but matched it with only some of the assets in your workspace.
Click a file in the list to open its File details panel. It shows the data type, the asset, who uploaded the file, a download link for the original file, and, if processing failed, the cause.
Fixing failed uploads
Tensor Cloud matches meter data with assets by location ID. If the location ID in a file differs from the one in the asset settings, the data cannot be matched, and the file is marked as Processing failed or Partially processed.
If a meter data file failed, first check that its location ID matches the one in the asset settings, then upload it again if needed. Re-uploading has no side effects: Tensor Cloud handles overlapping data and duplicate files automatically, so you can re-upload as often as you like.
Supported data types and file formats
Tensor Cloud supports the following data types. If your data is in a format that is not listed, contact Tensor support.
| Data type | File format | Description | Specification |
|---|---|---|---|
| Confirmed meter data | CSV | Confirmed meter data in 30-minute resolution, obtained from the TSO of each grid area. A file can contain data for several assets, matched by location ID, so it may be only partially processed. | - |
| Preliminary meter data | XML | Preliminary meter data in 30-minute resolution, obtained from the TSO of each grid area. A file can contain data for several assets, so it may be only partially processed. | OCCTO specifications |
| Laplace Systems monitoring data | CSV | Data downloaded from the Laplace Systems monitoring system in 1-minute or 30-minute resolution, for the single asset you select when uploading. | Laplace Systems specifications |
| Generic monitoring data | CSV | Generic upload format for monitoring data in 30-minute or 1-minute resolution, for the single asset you select when uploading. | Specification |
| Demand data | CSV | Demand data in 30-minute resolution for a single asset with a load component, which you select when uploading. | Specification |
Generic CSV format for monitoring data
Overview
Use this generic CSV format to upload monitoring data. Follow the specification below so that your files are processed correctly.
File format
- Encoding: UTF-8 without BOM
- File extension: .csv
- Column separator: Comma (
,) - Line breaks: LF (Unix/Linux) or CRLF (Windows)
General guidelines
- Include a header row with the exact column names below.
- Give all dates and times in JST (UTC+9), without a UTC offset.
- Do not include units in numeric fields.
- Trim leading and trailing spaces.
- Use the specified data format for each field.
- The optional fields are accepted but not currently used.
30-minute resolution
In 30-minute files, each value is the sum (or the average, where applicable) over the 30-minute interval starting at Date.
Required fields (30-minute)
| Field name | Romanization | Description | Format/Type | Example |
|---|---|---|---|---|
Date | Date and time in JST | String | 2024-01-31 15:30 | |
交流電力量(kWh) | kōryū denryokuryō | AC power (kWh) | Number | 25.3 |
Optional fields (30-minute)
| Field name | Romanization | Description | Format/Type | Example |
|---|---|---|---|---|
日射量(kWh/m2) | nissharyō | Solar radiation (kWh/m2) | Number | 123.0 |
気温(℃) | kion | Temperature (℃) | Number | 19.5 |
直流電圧(V) | chokuryū den'atsu | DC voltage (V) | Number | 123.0 |
直流電流(A) | chokuryū denryū | DC current (A) | Number | 123.0 |
直流電力量(kWh) | chokuryū denryokuryō | DC power (kWh) | Number | 123.0 |
交流電圧(V) | kōryū den'atsu | AC voltage (V) | Number | 123.0 |
交流電流(A) | kōryū denryū | AC current (A) | Number | 123.0 |
Sample CSV (30-minute)
Date,交流電力量(kWh)
2024-01-31 00:00,0.0
2024-01-31 00:30,0.0
2024-01-31 01:00,0.0
2024-01-31 01:30,0.0
Download sample CSV file (30-minute resolution)
1-minute resolution
In 1-minute files, each value is the instantaneous value at Date.
Required fields (1-minute)
| Field name | Romanization | Description | Format/Type | Example |
|---|---|---|---|---|
Date | Date and time in JST | String | 2024-01-31 15:30 | |
交流電力(kW) | kōryū denryoku | AC power (kW) | Number | 25.3 |
Optional fields (1-minute)
| Field name | Romanization | Description | Format/Type | Example |
|---|---|---|---|---|
日射強度(kW/m2) | nissha kyōdo | Solar radiation (kW/m2) | Number | 123.0 |
気温(℃) | kion | Temperature (℃) | Number | 19.5 |
直流電圧(V) | chokuryū den'atsu | DC voltage (V) | Number | 123.0 |
直流電流(A) | chokuryū denryū | DC current (A) | Number | 123.0 |
直流電力(kW) | chokuryū denryoku | DC power (kW) | Number | 123.0 |
交流電圧(V) | kōryū den'atsu | AC voltage (V) | Number | 123.0 |
交流電流(A) | kōryū denryū | AC current (A) | Number | 123.0 |
Sample CSV (1-minute)
Date,交流電力(kW)
2024-01-31 00:00,0.0
2024-01-31 00:01,0.0
2024-01-31 00:02,0.0
2024-01-31 00:03,0.0
Download sample CSV file (1-minute resolution)
CSV format for demand data
Upload demand data as CSV files in the following format. The file itself does not identify the asset: Tensor Cloud stores the data for the asset you select in the upload dialog, so each file holds data for a single asset.
- Encoding: UTF-8. Files in Shift-JIS are not recognized.
- Column separator: Comma (
,) - Header row: Must include the columns
測定日時(sokutei nichiji, measurement date and time) and平均需要[kW](heikin juyō, average demand). Other columns are ignored. - Resolution: 30 minutes. Files in any other resolution are not recognized.
- Order: Rows in ascending order of time.
Fields (demand data)
| Field name | Romanization | Description | Format/Type | Example |
|---|---|---|---|---|
測定日時 | sokutei nichiji | Start of the 30-minute time slot, in JST without a UTC offset | String | 2022-03-01 00:30:00 |
平均需要[kW] | heikin juyō | Average demand over the time slot (kW) | Number | 12.7 |
Sample CSV (demand data)
測定日時,平均需要[kW]
2022-03-01 00:00:00,12.7
2022-03-01 00:30:00,12.2
2022-03-01 01:00:00,11.1