Fuel Analytics

Owned by arkadi kaplon
Last updated: 05.12.24
4 min read

Lesson Highlights

Name of the lesson

Fuel Analytics

Purpose of the lesson

This document provides clear explanations and simplifies the underlying logic behind the various results users may encounter when analyzing their fuel data.

The following guide is designed to help you understand the fuel summary fields and utilize them effectively for improved data analysis, enhanced fuel dashboards, and increased confidence in data calculations.

Target Audience

End-users of the system such as technicians, service providers, back-office, organization managers, etc.

Operational Fuel Consumption per Hour

Galooli offers its clients three distinct methods for accessing fuel data related to operational consumption (not including fuel drops).

Below, you’ll find an overview of each option, along with recommendations on when to use these fields for optimal analysis.

Galooli Operational Consumption (Direct Reading + Estimations)

Location: Pro Fields - Summary >> Working - Generators >> Generator 1/2 Galooli Operational Consumption [Gallon US/H]

  • This field provides an adjusted assessment of fuel consumption when the detected data appears illogical or inconsistent with the generator’s KVA. This ensures reliable insights even when anomalies are present in the raw sensor data.

  • Calculation details: In case the direct reading of the fuel consumption is invalid (not in a predefined range), the estimation of the consumption is based on:

    • Final and initial fuel levels, considering drops and refuels.

    • Fuel consumption patterns observed in the generator’s operation. The average fuel consumption will be considered if the generator average operating load has not changed significantly (more than 10% change) compared to the previous week.

    • Estimated consumption derived from the generator’s load and KVA rating (load-based consumption)

Operational Consumption according to Fuel Sensor (Direct Reading)

Location: Pro Fields - Summary >> Working - Generators >> Generator 1/2 Fuel Sensor Operational Consumption [Gallon US/H]

  • This field captures a direct reading from the installed fuel sensor on-site, including potential spikes in data.

  • Calculation Details: The fuel level is measured every minute, and at the end of each day, the total fuel consumption in liters is calculated. This total is then divided by the generator's operating hours to determine the daily average fuel consumption rate

Load-Based Consumption (Estimation)

Location: Pro Fields - Summary >> Working - Generators >> Generator 1/2 Load-Based Fuel Consumption [Gallon US/H]

  • This field calculates estimated fuel consumption based on the generator's specific KVA rating and operating load. It provides a baseline for assessing whether the generator's fuel usage is reasonable and contributes to the "Data Quality" scores.

  • Calculation Details:

    • The operating load is calculated by (Avg Load Power / (Generator KVA × 0.8)). In case there is no generator power data available, a default of 60% load is used.

    • The estimation is using a different formula for each KVA, which was developed by operational expertise and manufacturer-provided data.

Fuel Drops

Galooli provides information about number and amount of fuel drops. To enable analysis, the user must configure a "Fuel Drop" event under "Fuel Events." Events

  • Total Number of Fuel Drops [Count]
    Location: Pro - Summary >> Working - Generators >> Generator 1/2 Fuel Drops
    This field displays the total number of fuel drop events detected, helping to track instances of abnormal fuel usage.

  • Total Fuel Drop [Gallon/Liters]
    Location: Pro - Summary >> Working - Generators >> Fuel Drop [Gallon US]
    This field calculates the total volume of fuel lost while the generator was not running. These drops could indicate potential issues, such as fuel theft or leakage, offering a critical tool for operational oversight.

To validate data quality and ensure only valid refueling events are considered, it is advised to use the "Fuel Data Quality - Possible Reasons" field and exclude rows with the following results:

  • "Irrational events count"

  • "Irrational events volume"

Refuels

Galooli provides information about number and amount of refuels. To enable analysis, the user must configure a "Tank Refueled" event under "Fuel Events." Events

  • Number of refuels
    Location: Pro - Summary >> Working - Generators >> Generator 1 Refuel.
    This field summarizes the total number of refuel events detected each day.

  • Total Amount [Gallon or Liters] of fuel refueled -
    Location: Pro - Summary >> Working - Generators >> Generator 1 Refuel [Gallon/Liter]
     This fields calculated total volume of fuel refueled within the selected time range.

To validate data quality and ensure only valid refueling events are considered, it is advised to use the "Fuel Data Quality - Possible Reasons" field and exclude rows with the following results:

  • "Irrational events count"

  • "Irrational events volume"

Fuel Data Quality

The Data fields section introduces two types data quality fields, "Data Quality - Possible Reasons" and "Low Fuel Data Quality - Description", which analyze fuel data reliability and offer insights into potential issues.
Each issue is explained with actionable recommendations, such as recalibrating sensors, verifying site configurations, or replacing faulty hardware.

  • Data Quality Score:

Location: AI-Based >> AI-Based Analysis - Fuel >> Fuel 1 Data Quality

Data quality is measured on a scale from 0% to 100%, offering a clear metric to evaluate reliability. Scores marked as "-99999" indicate cases where generator hours and fuel consumption are absent, making data quality analysis inapplicable.

  • "Data Quality - Possible Reasons":
    Location: AI-Based >> AI-Based Analysis - Fuel >> Fuel 1 Data Quality Possible Reasons
    This field summarizes all potential reasons impacting data quality across a report period. this provides our client a short explanation and spotlights reasons that the client can filter by using these data fields. This field is recommended for analyzing a large number of units over the report's time range.

Examples for possible reasons:

  • Disconnected fuel sensor

  • Low fuel level

  • Sensor stuck

  • Fuel consumed without engine hours

  • Deviation from the average CPH

  • Irrational CPH compared to load-based CPH

  • Spike

  • Irrational events volume

  • Irrational events count

  • "Low Fuel Data Quality - Description":
    location: AI-Based >> AI-Based Analysis - Fuel >> Low Fuel Data Quality - Description
    This field offers a detailed breakdown of issues from the last day of the report period, along with actionable recommendations. It provides clients with comprehensive explanations of all detected reasons and is ideal for conducting drill-down analyses on individual units.