Saturday, November 21, 2026
Science
No Result
View All Result
  • Login
  • HOME
  • SCIENCE NEWS
  • CONTACT US
  • HOME
  • SCIENCE NEWS
  • CONTACT US
No Result
View All Result
Scienmag
No Result
View All Result
Home Science News

Fuzzy Logic IoT System Classifies Water Usability in Real Time

November 21, 2026
in Science News
Martin Reynolds
By Martin Reynolds Scienmag Editorial Profile - Environmental Engineering
Reading Time: 6 mins read
0
Fuzzy Logic IoT System Classifies Water Usability in Real Time

Fuzzy Logic IoT System Classifies Water Usability in Real Time

65
SHARES
587
VIEWS
Share on FacebookShare on Twitter
ADVERTISEMENT

The study, published in 2026, introduces a multi-parameter monitoring framework designed to address the limitations of conventional laboratory-based water analysis, which is often time-consuming, expensive, and unsuitable for continuous monitoring. By integrating sensor technology with an embedded fuzzy inference system, the proposed architecture aims to provide immediate, practical decisions regarding water usability for various applications, ranging from drinking to industrial use.

The core of the system is an ESP32 microcontroller that serves as the central processing and communication unit. It interfaces with four specific sensors: a pH sensor to measure acidity or alkalinity, a turbidity sensor to determine water clarity via light scattering, a Total Dissolved Solids (TDS) sensor to quantify dissolved ions and minerals, and a DS18B20 temperature sensor. These sensors continuously monitor water parameters, forwarding data to the processing layer where signal conditioning and preprocessing occur. This preprocessing involves averaging multiple samples to reduce noise and normalizing the data to ensure all input parameters are mapped onto a common scale before being supplied to the fuzzy logic controller.

To handle the uncertainty and nonlinear relationships inherent in water quality parameters, the system employs a Mamdani Fuzzy Inference System (FIS). Unlike traditional threshold-based methods that rely on abrupt decision boundaries, the fuzzy logic approach uses adaptive membership functions and rule-based reasoning to represent gradual transitions between different water quality conditions. The FIS processes the conditioned sensor data simultaneously, evaluating the combined influence of pH, turbidity, TDS, and temperature. This method allows the system to interpret complex, overlapping physicochemical conditions that might confuse simpler binary classification models, providing a more nuanced assessment of water suitability.

The fuzzy knowledge base was constructed using expert knowledge and established water-quality guidelines, including recommendations from the World Health Organization (WHO), the Food and Agriculture Organization (FAO), and the US Environmental Protection Agency (EPA). The system utilizes a compact set of 30 representative fuzzy rules to cover the decision space while maintaining low computational complexity suitable for the ESP32 platform. These rules were derived from acceptable and critical limits reported in standard guidelines and adjusted through calibration experiments to ensure smooth overlap between adjacent linguistic variables. The inference process concludes with centroid defuzzification, which converts the aggregated fuzzy output into a crisp decision corresponding to one of five practical usability categories: drinkable, washable, farming, aquarium, and non-usable.

Experimental validation was conducted using 500 water samples collected from six representative sources over approximately six weeks. The sampling sources included municipal tap water, packaged drinking water, freshwater ponds, agricultural irrigation water, urban drainage water, and industrial wastewater. This diverse dataset covered domestic, drinking, aquaculture, agricultural, and wastewater-related conditions. Each sample was measured five consecutive times, with measurements averaged to ensure stability. The experimental setup was designed to be portable, low-cost, and energy-efficient, making it suitable for deployment in remote and industrial environments where continuous monitoring is required but laboratory access is limited.

The results demonstrated that the proposed fuzzy logic-based system provides improved classification accuracy and better adaptability compared to conventional threshold-based approaches. In the classification performance evaluation, the system achieved an overall accuracy of 96.4%, with a balanced accuracy of 96.12%. The precision, recall, and F1-score were 95.8%, 96.10%, and 95.9%, respectively. Notably, the system exhibited high specificity of 99.18% and a Matthews correlation coefficient (MCC) of 0.955, indicating strong agreement between predicted and actual classes. Approximately 17 of the 500 samples were misclassified, with most errors occurring between the washable-aquarium and washable-drinkable categories due to overlapping physicochemical characteristics. No critical confusion was observed between drinkable and non-usable categories, confirming the system’s ability to effectively separate clearly acceptable from highly contaminated water.

Statistical analysis using 10 repetitions of stratified five-fold cross-validation showed that the performance metrics were stable across different train-test partitions. The average Area Under the Curve (AUC) of 0.991 demonstrated excellent class separability. A sensitivity analysis of the membership-function boundaries revealed that the system remained robust to moderate variations in expert-defined parameters. Even with perturbations of up to ±15%, the accuracy remained above 94%, suggesting that the classifier is not excessively dependent on a single precise parameter setting. This robustness is crucial for practical applications where environmental conditions and sensor readings may fluctuate.

The IoT communication layer facilitates remote monitoring by transmitting sensor readings and classification results to a cloud-based platform via Wi-Fi. The ESP32 uses HTTP POST requests every 2 seconds to send data packets containing sensor values, timestamps, classification results, and device authentication tokens. Communication reliability tests using 1,000, 5,000, and 10,000 packets yielded packet-loss rates of 0.20%, 0.24%, and 0.28%, respectively. The average end-to-end response time, including sensor acquisition, fuzzy inference, cloud transmission, and dashboard update, was approximately 3.3 seconds. This near real-time performance enables users to monitor water quality remotely using any internet-connected device, with visual alerts activated for abnormal or non-usable water conditions.

Power consumption analysis indicated that the system operates efficiently, consuming approximately 0.8 W during normal operation. This low power demand supports the feasibility of battery-powered monitoring, with estimates suggesting that a 5000 mAh battery could support approximately 23 hours of continuous operation. The system’s low computational complexity and deterministic rule-based inference make it particularly suitable for embedded implementation on the ESP32, avoiding the need for extensive training datasets or high computational resources required by machine learning models.

When compared with five supervised machine-learning classifiers, including Decision Tree, Support Vector Machine, Artificial Neural Network, Random Forest, and XGBoost, the proposed FIS achieved competitive performance without the need for supervised training. The FIS significantly outperformed Decision Tree, Random Forest, and SVM classifiers, while showing no statistically significant difference from ANN and XGBoost. The primary advantage of the fuzzy logic approach is its interpretability and training-free operation, which simplifies deployment and maintenance. The system provides transparent decision logic based on linguistic rules, offering a favorable trade-off among classification accuracy, computational efficiency, power consumption, and embedded suitability.

The study highlights the potential of integrating fuzzy logic with IoT technologies for next-generation smart water quality monitoring. By converting combined sensor information into application-oriented usability categories, the system addresses the limitation of generic safe/unsafe indicators that may not provide sufficient practical decision support. The research demonstrates that a lightweight, interpretable, and training-free mechanism can be effectively implemented on low-cost embedded platforms for real-time water-usability assessment. Future work will focus on incorporating additional water-quality parameters, adaptive fuzzy intelligence, and predictive analytics, along with long-term field validation to address sensor aging and calibration drift.

Beyond the primary classification metrics, the research team conducted detailed validation of individual sensor performance against standard laboratory instruments. Across 50 comparison measurements gathered under varying conditions, the mean absolute error remained low across all monitored indicators. Turbidity exhibited the highest error among the four parameters because of suspended particles and light scattering effects, whereas pH and temperature readings demonstrated tight consistency with laboratory benchmarks. To establish short-term repeatability, each sample underwent five consecutive assessments under identical conditions, producing coefficients of variation below 4 percent across all sensors.

Sensor calibration procedures followed rigorous protocols prior to field testing. The pH sensor was calibrated using reference buffer solutions at pH 4.0, 7.0, and 9.2, showing a maximum deviation within 0.15 pH units. Calibration of the Total Dissolved Solids probe relied on standard conductivity solutions of 342 ppm and 1000 ppm, maintaining measurement error within 3 percent through integrated temperature compensation. Turbidity calibration incorporated distilled water at 0 Nephelometric Turbidity Units alongside prepared turbidity standards of roughly 50, 100, and 200 NTU mapped via linear regression, yielding an average error under 5 NTU. The DS18B20 temperature sensor, tested against a laboratory-grade digital thermometer across 15 to 40 degrees Celsius, deviated by no more than 0.5 degrees Celsius.

The physical testing protocol maintained controlled environmental constraints during data collection. Water evaluations took place between 9:00 AM and 4:00 PM at ambient conditions of 27 ± 3 degrees Celsius and 62 ± 8 percent relative humidity. To preserve water stability, researchers collected samples in clean 500-milliliter polyethylene containers, initiated testing immediately after collection, and rinsed sensor probes with distilled water between evaluations to prevent cross-contamination. Each measurement cycle incorporated a 10-second stabilization period before recording values at a 2-second sampling interval.

Detailed power measurements revealed that operational demand varied across system tasks. The embedded platform consumed approximately 0.64 W during active sensor acquisition, 0.67 W during fuzzy processing, 0.80 W during Wi-Fi transmission, and 0.49 W during idle monitoring. Because the Wi-Fi module activated periodically every 2 seconds with an active transmission window of roughly 250 milliseconds, its estimated duty cycle was 12.5 percent, demonstrating that wireless communication required more energy than on-chip fuzzy calculations. Based on the overall 0.8 W average draw, projected battery lifetimes reached 9 hours on a 2000 mAh battery and 46 hours on a 10,000 mAh battery.

Subject of Research: IoT-based fuzzy logic system for real-time multi-parameter water quality assessment and classification

Article Title: An intelligent IoT-based multi-parameter water quality assessment and classification system using fuzzy logic for real-time applications

Article References: Samanta, S., Panja, B., Kole, A., & Chakraborty, S. (2026). An intelligent IoT-based multi-parameter water quality assessment and classification system using fuzzy logic for real-time applications. Journal of Electrical Systems and Information Technology, 13(1), Article 87. https://doi.org/10.1186/s43067-026-00396-0

Image Credits: AI Generated

DOI: 10.1186/s43067-026-00396-0

Keywords: IoT water quality monitoring, Fuzzy logic inference, ESP32 microcontroller, Real-time water classification, Multi-parameter sensing, Water usability categories, Embedded systems, Cloud-based monitoring, Environmental engineering, Sensor calibration, Mamdani FIS, Low-cost water analysis, Smart water management, Edge computing, Hydrological assessment

Cite Scienmag News

Martin Reynolds. (November 21, 2026). Fuzzy Logic IoT System Classifies Water Usability in Real Time. Scienmag. https://scienmag.com/fuzzy-logic-iot-system-classifies-water-usability-in-real-time/

Martin Reynolds. "Fuzzy Logic IoT System Classifies Water Usability in Real Time." Scienmag, 21 November 2026, https://scienmag.com/fuzzy-logic-iot-system-classifies-water-usability-in-real-time/. Accessed 21 November 2026.

Martin Reynolds. "Fuzzy Logic IoT System Classifies Water Usability in Real Time." Scienmag. November 21, 2026. https://scienmag.com/fuzzy-logic-iot-system-classifies-water-usability-in-real-time/

Tags: cloud-based monitoringedge computingembedded systemsenvironmental engineeringESP32 microcontrollerFuzzy logic inferenceHydrological assessmentIoT water quality monitoringLow-cost water analysisMamdani FISmulti-parameter sensingReal-time water classificationsensor calibrationsmart water managementWater usability categories
Share26Tweet16
Previous Post

Room-Temperature Dual-Metal Frameworks Accelerate Photocatalytic Hydrogen Evolution and Hydrogenation

Next Post

Saudi Study Finds Strong Patient Preference for Direct Gynecological Cancer Disclosure

Related Posts

Social Media Addiction Links to Eating Psychopathology Through Body Shape Dissatisfaction
Psychology & Psychiatry

Social Media Addiction Links to Eating Psychopathology Through Body Shape Dissatisfaction

November 21, 2026
Verifiable Matchmaking Encryption Secures Cloud Electronic Health Records for Healthcare IoT
Technology and Engineering

Verifiable Matchmaking Encryption Secures Cloud Electronic Health Records for Healthcare IoT

November 21, 2026
Nitrogen Bridged Molybdenum Disulfide Anodes Deliver Outstanding Lithium Ion Storage Stability
Science News

Nitrogen Bridged Molybdenum Disulfide Anodes Deliver Outstanding Lithium Ion Storage Stability

November 21, 2026
Numerical Model Uncovers Seasonal Vortices and Surface Current Shifts in Persian Gulf
Science News

Numerical Model Uncovers Seasonal Vortices and Surface Current Shifts in Persian Gulf

November 21, 2026
Moral Injury Associates With Mental Health Struggles Among Humanitarian Aid Workers
Science News

Moral Injury Associates With Mental Health Struggles Among Humanitarian Aid Workers

November 21, 2026
Energy Consumption Drives Carbon Emissions Across Gulf Nations While Water Stress Alters Long-Term Trends
Agriculture

Energy Consumption Drives Carbon Emissions Across Gulf Nations While Water Stress Alters Long-Term Trends

November 21, 2026
Next Post
Saudi Study Finds Strong Patient Preference for Direct Gynecological Cancer Disclosure

Saudi Study Finds Strong Patient Preference for Direct Gynecological Cancer Disclosure

  • Mothers who receive childcare support from maternal grandparents show more optimized

    Mothers who receive childcare support from maternal grandparents show more parental warmth, finds NTU Singapore study

    27656 shares
    Share 11059 Tweet 6912
  • University of Seville Breaks 120-Year-Old Mystery, Revises a Key Einstein Concept

    1061 shares
    Share 424 Tweet 265
  • Bee body mass, pathogens and local climate influence heat tolerance

    682 shares
    Share 273 Tweet 171
  • Researchers record first-ever images and data of a shark experiencing a boat strike

    546 shares
    Share 218 Tweet 137
  • Groundbreaking Clinical Trial Reveals Lubiprostone Enhances Kidney Function

    531 shares
    Share 212 Tweet 133
Science

Embark on a thrilling journey of discovery with Scienmag.com—your ultimate source for cutting-edge breakthroughs. Immerse yourself in a world where curiosity knows no limits and tomorrow’s possibilities become today’s reality!

RECENT NEWS

  • Social Media Addiction Links to Eating Psychopathology Through Body Shape Dissatisfaction
  • Verifiable Matchmaking Encryption Secures Cloud Electronic Health Records for Healthcare IoT
  • Nitrogen Bridged Molybdenum Disulfide Anodes Deliver Outstanding Lithium Ion Storage Stability
  • Numerical Model Uncovers Seasonal Vortices and Surface Current Shifts in Persian Gulf

Categories

  • Agriculture
  • Anthropology
  • Archaeology
  • Athmospheric
  • Biology
  • Biotechnology
  • Blog
  • Bussines
  • Cancer
  • Chemistry
  • Climate
  • Earth Science
  • Editorial Policy
  • Marine
  • Mathematics
  • Medicine
  • Pediatry
  • Policy
  • Psychology & Psychiatry
  • Science Education
  • Science News
  • Social Science
  • Space
  • Technology and Engineering

Subscribe to Blog via Email

Enter your email address to subscribe to this blog and receive notifications of new posts by email.

Join 5,151 other subscribers

© 2026 Scienmag - Science Magazine

Welcome Back!

Login to your account below

Forgotten Password?

Retrieve your password

Please enter your username or email address to reset your password.

Log In
No Result
View All Result
  • HOME
  • SCIENCE NEWS
  • CONTACT US

© 2026 Scienmag - Science Magazine

Discover more from Science

Subscribe now to keep reading and get access to the full archive.

Continue reading