THULab/oil013_sample

OIL-013 — Synthetic Production Time-Series Dataset (Sample) (TsFile)

Apache TsFile version of xpertsystems/oil013-sample.

Overview

A free, schema-identical preview of XpertSystems.ai's enterprise production time-series dataset for decline curve ML, artificial-lift optimization, workover-candidate prediction, and water-breakthrough forecasting. The sample covers 250 wells across 10 global basins and 8 asset types, simulated over 365 days, with 116,154 rows including 91,250 daily production records linked across 11 tables.

• Tables: 11 TsFile table(s), 116,154 rows in total.
  • artificial_lift: 13,250 rows, 2020-01-14 00:00:00 → 2022-12-25 00:00:00 (UTC)
  • daily_production: 91,250 rows, 2020-01-14 00:00:00 → 2022-12-25 00:00:00 (UTC)
  • downtime_events: 491 rows, 2020-02-04 00:00:00 → 2022-11-24 00:00:00 (UTC)
  • facility_constraints: 200 rows, 2020-01-01 00:00:00 → 2020-01-01 00:00:00.199000 (UTC)
  • flow_assurance_events: 47 rows, 2020-03-20 00:00:00 → 2022-09-09 00:00:00 (UTC)
  • injection_support: 40 rows, 2020-09-06 00:00:00 → 2022-08-10 00:00:00 (UTC)
  • production_allocations: 625 rows, 1970-01-01 00:00:00 → 1970-01-01 00:00:00.624000 (UTC)
  • production_labels: 3,250 rows, 2020-01-14 00:00:00 → 2022-12-21 00:00:00 (UTC)
  • reservoir_pressure: 6,750 rows, 2020-01-14 00:00:00 → 2022-12-25 00:00:00 (UTC)
  • stimulation_events: 1 rows, 2020-09-11 00:00:00 → 2020-09-11 00:00:00 (UTC)
  • wells_master: 250 rows, 1970-01-01 00:00:00 → 1970-01-01 00:00:00.249000 (UTC)

Schema (TsFile structure)

artificial_lift.tsfile

• Time (INT64, milliseconds) — the source record_date column.
• TAG: none — the table is stored as a single device.
• FIELD (8): lift_id (string), well_id (string), lift_type (string), esp_frequency_hz (double), pump_efficiency (double), motor_temperature_f (double), vibration_index (double), runtime_pct (double)

daily_production.tsfile

• Time (INT64, milliseconds) — the source production_date column.
• TAG: none — the table is stored as a single device.
• FIELD (16): prod_id (string), well_id (string), oil_rate_bopd (double), gas_rate_mscfd (double), water_rate_bwpd (double), water_cut_pct (double), gor_scf_bbl (double), choke_size_64ths (double), tubing_pressure_psi (double), casing_pressure_psi (double), uptime_fraction (double), cumulative_oil_bbl (double), cumulative_gas_mscf (double), cumulative_water_bbl (double), anomaly_flag (int64), anomaly_type (string)

downtime_events.tsfile

• Time (INT64, milliseconds) — the source start_date column.
• TAG: none — the table is stored as a single device.
• FIELD (6): event_id (string), well_id (string), downtime_type (string), duration_hours (double), lost_oil_bbl_est (double), root_cause_category (string)

facility_constraints.tsfile

• Time (INT64, milliseconds) — the source effective_date column.
• TAG: none — the table is stored as a single device.
• FIELD (7): constraint_id (string), facility_id (string), field_name (string), throughput_limit_bopd (double), gas_handling_limit_mscfd (double), water_handling_limit_bwpd (double), constraint_severity (string)

flow_assurance_events.tsfile

• Time (INT64, milliseconds) — the source event_date column.
• TAG: none — the table is stored as a single device.
• FIELD (8): flow_event_id (string), well_id (string), flow_assurance_type (string), scale_index (double), paraffin_index (double), sand_index (double), corrosion_risk_index (double), intervention_required_flag (int64)

injection_support.tsfile

• Time (INT64, milliseconds) — the source record_date column.
• TAG: none — the table is stored as a single device.
• FIELD (8): inj_id (string), injector_id (string), producer_id (string), field_name (string), injection_rate_bwpd (double), injection_pressure_psi (double), producer_response_lag_days (int64), correlation_score (double)

production_allocations.tsfile

• Time (INT64, milliseconds) — synthetic row index over the stable source order; this table has no clock column.
• TAG: none — the table is stored as a single device.
• FIELD (6): alloc_id (string), well_id (string), zone_name (string), allocation_pct (double), allocation_method (string), confidence_score (double)

production_labels.tsfile

• Time (INT64, milliseconds) — the source label_date column.
• TAG: none — the table is stored as a single device.
• FIELD (8): label_id (string), well_id (string), forecast_class (string), workover_candidate_flag (int64), water_breakthrough_flag (int64), steep_decline_flag (int64), lift_limited_flag (int64), next_90d_expected_decline_pct (double)

reservoir_pressure.tsfile

• Time (INT64, milliseconds) — the source test_date column.
• TAG: none — the table is stored as a single device.
• FIELD (6): pressure_id (string), well_id (string), reservoir_pressure_psi (double), bhfp_psi (double), pressure_drawdown_psi (double), pressure_test_quality (string)

stimulation_events.tsfile

• Time (INT64, milliseconds) — the source event_date column.
• TAG: none — the table is stored as a single device.
• FIELD (7): stim_id (string), well_id (string), event_type (string), frac_stage_count (int64), treatment_volume_bbl (double), expected_uplift_pct (double), actual_uplift_pct (double)

wells_master.tsfile

• Time (INT64, milliseconds) — synthetic row index over the stable source order; this table has no clock column.
• TAG: none — the table is stored as a single device.
• FIELD (24): well_id (string), api_like_well_code (int64), field_name (string), pad_id (string), basin_name (string), formation_name (string), asset_type (string), completion_type (string), lift_type (string), operator_class (string), first_production_date (string), lateral_length_ft (int64), proppant_lb_per_ft (double), true_vertical_depth_ft (int64), initial_oil_rate_bopd (double), initial_gas_rate_mscfd (double), initial_water_cut_pct (double), initial_gor_scf_bbl (double), nominal_decline_rate_annual (double), arps_b_factor (double), initial_reservoir_pressure_psi (double), reservoir_temperature_f (double), longitude (double), latitude (double)

Conversion notes

• One TsFile per source CSV (11 tables) for this oil-well production sample; each table is stored as a single device, so the rig/well/tank/sensor ids stay ordinary FIELDs.
• Time = the table's own timestamp column as INT64 milliseconds (event timestamps, daily record/production dates, or test dates). Master/inventory tables that carry no clock (rigs_master, sensors_master, wells_master, tank_master, the label tables and scada_telemetry) use a synthetic row index as Time.
• No column is dropped. Rows that share a timestamp are ordered by a 1 ms step in stable source order; daily tables keep their exact calendar day.

Usage

Install the Apache TsFile Python SDK (pip install tsfile) and read a converted file:

from pathlib import Path
from tsfile import TsFileReader

path = Path("artificial_lift.tsfile")
with TsFileReader(str(path)) as reader:
    schemas = reader.get_all_table_schemas()
    print("tables:", list(schemas))
    table_name = next(iter(schemas))
    table = schemas[table_name]
    columns = [column.get_column_name() for column in table.get_columns()]
    print("columns:", columns)
    field_names = [
        column.get_column_name()
        for column in table.get_columns()
        if column.get_column_name() not in {"Time", "time"}
    ]
    if field_names:
        with reader.query_table(table_name, field_names[:3], batch_size=1024) as result:
            batch = result.read_arrow_batch()
            if batch is not None:
                print(batch.to_pandas().head())

Source & license

• Original dataset: xpertsystems/oil013-sample
• Source card title / publisher: OIL-013 — Synthetic Production Time-Series Dataset (Sample)
• License: cc-by-nc-4.0 (as declared by the original dataset).