fieldkit / caldor

Library for calibration system designed for field instrumentation

Caldor: The in-Situ Calibration System

Short Description

Caldor is an innovative in-situ calibration system designed for field instrumentation, allowing for convenient and accurate calibration using traceable standards. This system, developed by the FieldKit Project team at Conservify, integrates Human-Computer Interaction (HCI) principles to minimize errors and enhance data integrity in environmental sensing.

Features

• In-Field Calibration: Enables calibration in remote locations without needing lab facilities.
• Educational Resources: Provides comprehensive guidelines on calibration procedures.
• User-Friendly Interface: Simplifies the calibration process when paired with the intuitive app.
• Versatility: Compatible with various sensors like temperature, pH, conductivity, etc.
• Data Integrity: Ensures high-quality data with traceable calibration standards.

Getting Started

• Prerequisites:

  • Linux
  • Chrome
  • bash
  • git
  • curl
  • Flutter
  • Dart

• Installation:

  • sudo apt-get update -y && sudo apt-get install

• Initial Configuration: dart pub get or flutter pub get

Usage

To run: dart run To test: dart test or flutter test

To use the library you've provided for calculating and handling sensor calibrations, follow this step-by-step guide:

Step 1: Determine the Sensor Type

First, identify the type of sensor you need to calibrate. This could be, for example, modules.water.ph, modules.water.do, modules.water.ec, or modules.water.temp.

Step 2: Select the Appropriate Calibration Template

Use the CalibrationTemplate.forModuleKey method to retrieve a CalibrationTemplate that corresponds to your sensor type.

Example:

String sensorType = "modules.water.ph"; // Replace with your sensor type
CalibrationTemplate template = CalibrationTemplate.forModuleKey(sensorType);

Step 3: Collect Calibration Points

Gather a set of CalibrationPoint objects. Each point should include a Standard (which could be a FixedStandard, UnknownStandard, or UserStandard) and a corresponding SensorReading (which includes both uncalibrated and calibrated values, if available).

Example:

List<CalibrationPoint> points = [
  CalibrationPoint(
    standard: FixedStandard(4), // Replace with actual standard value
    reading: SensorReading(uncalibrated: 1.0, value: 4.0), // Replace with actual readings
  ),
  // Add more CalibrationPoint objects as required
];

Step 4: Calculate the Calibration Curve

Depending on the CurveType in your CalibrationTemplate (either linear or exponential), use the respective function (linearCurve or exponentialCurve) to calculate the calibration curve coefficients.

Example:

List<double> coefficients;
if (template.curveType == CurveType.linear) {
  coefficients = linearCurve(points);
} else if (template.curveType == CurveType.exponential) {
  coefficients = exponentialCurve(points);
} else {
  throw Exception("Unknown curve type: ${template.curveType}");
}

Step 5: Utilize the Calculated Coefficients

The coefficients list obtained from the previous step can now be used as needed in your application. This could be for sensor data correction, display, or further processing.

Step 6: Serialize Calibration Data (Optional)

If you need to serialize the calibration data (e.g., for storage or transmission), you can use the toBytes method from CurrentCalibration. Ensure to add all the calibration points to CurrentCalibration before serialization.

Example:

CurrentCalibration currentCalibration = CurrentCalibration(curveType: template.curveType);
for (var point in points) {
  currentCalibration.addPoint(point);
}
Uint8List serializedData = currentCalibration.toBytes();

Step 7: Deserialize Calibration Data (Optional)

To deserialize, you would typically use the proto.ModuleConfiguration to parse the bytes back into a meaningful structure. This process will depend on how you handle protobuf data in your application.

Additional Information

For more information on the Caldor package, including how to contribute, report issues, or get support, visit our GitHub repository.