AqueSense Technologies

About Our Water Quality Monitoring System

Learn about the water quality parameters we monitor and the technologies implemented in our project.

Monitored Parameters

pH Level

pH is a measure of how acidic or basic water is. The range goes from 0 to 14, with 7 being neutral. pH affects many chemical and biological processes in water and is one of the most important parameters in water quality assessment.

Ideal Range

6.5 - 8.5

Effects

  • < 6.5: Acidic water can corrode pipes and fixtures, and may contain toxic metals
  • 6.5 - 8.5: Ideal range for most aquatic life and drinking water
  • > 8.5: Alkaline water can cause scale buildup and give water a bitter taste

Total Dissolved Solids (TDS)

TDS is a measure of all inorganic and organic substances dissolved in water. It indicates the general quality of water and its suitability for different uses.

Ideal Range

50 - 300 ppm

Effects

  • < 50 ppm: Very low mineral content, may taste flat and lack essential minerals
  • 50 - 300 ppm: Ideal range for drinking water with good taste and mineral content
  • > 500 ppm: High mineral content, may taste salty and cause scale buildup

Temperature

Water temperature affects many biological and chemical processes in water bodies. It influences the amount of oxygen that can dissolve in water, the rate of photosynthesis, and the metabolic rates of organisms.

Ideal Range

10 - 25°C

Effects

  • < 10°C: Cold water holds more oxygen but slows biological processes
  • 10 - 25°C: Ideal range for most aquatic life and biological processes
  • > 25°C: Warm water holds less oxygen and can accelerate algae growth

Conductivity

Electrical conductivity indicates the amount of dissolved solids in water. It is directly related to the concentration of ions in water and can be used as an indicator of water pollution.

Ideal Range

200 - 800 μS/cm

Effects

  • < 200 μS/cm: Low mineral content, typical of rainwater or distilled water
  • 200 - 800 μS/cm: Normal range for freshwater, good for most uses
  • > 800 μS/cm: High mineral content, may indicate pollution or saltwater intrusion

Turbidity

Turbidity measures the cloudiness or haziness of water caused by suspended particles. High turbidity can indicate the presence of microorganisms, sediments, or organic material that can affect water quality.

Ideal Range

< 5 NTU

Effects

  • < 1 NTU: Very clear water, excellent for drinking and aquatic life
  • 1 - 5 NTU: Slightly cloudy but still acceptable for most uses
  • > 20 NTU: Very cloudy water, harmful to aquatic life and requires significant treatment

Anomaly Score

Our system calculates an anomaly score that indicates how unusual the current readings are compared to normal patterns. This helps in early detection of potential issues in water quality.

Ideal Range

< 0.3

Effects

  • < 0.3: Normal readings, no significant anomalies detected
  • 0.3 - 0.6: Minor anomalies detected, may warrant monitoring
  • > 0.8: Critical anomalies detected, immediate action required

Classification Code

Based on multiple parameters, our system classifies water quality into different classes. This provides a comprehensive assessment of overall water quality.

Ideal Range

Class 1-2

Effects

  • Class 1: Excellent water quality, suitable for all uses
  • Class 2: Good water quality, suitable for most uses
  • Class 3-5: Fair to poor water quality, may require treatment

Our Prediction Model Implementation

Technical details about the machine learning approach we implemented for water quality prediction.

Prediction Model Architecture

Our prediction system implements time-series forecasting techniques to analyze historical data patterns and generate predictions for future water quality parameters.

Data Processing

Historical sensor data is collected and preprocessed to handle missing values and normalize readings.

Algorithm Implementation

We implemented exponential smoothing and regression techniques to identify trends and seasonal patterns.

Forecasting Engine

The model generates predictions with statistical confidence intervals to indicate reliability.

Analysis Module

An automated system interprets prediction results and generates insights about potential water quality changes.

Technical Implementation

  • Time-series analysis with exponential smoothing
  • Trend detection using regression algorithms
  • Statistical confidence intervals calculation
  • Feedback loop for model accuracy improvement
  • Configurable prediction horizons from 12 to 72 hours

Technical Outcomes

  • Accurate detection of parameter trend deviations
  • Proactive monitoring capabilities
  • Improved system response time to anomalies
  • Enhanced data visualization for trend analysis
  • Integration with the monitoring dashboard

Explore our project

Access the dashboard to see our water quality monitoring system in action.