FunSNM

New FunSNM tutorial now available: Infrastructure and Risk Analysis of Distributed Sensor Networks   Distributed sensor networks are at the core of modern monitoring, control, and optimization, from smart buildings and manufacturing to healthcare and smart grids.   In this new tutorial, Martin Koval, Czech Metrology Institute, walks through some of the key questions behind building robust, reliable, and trustworthy sensor networks: How do architecture, topology, and infrastructure differ? What roles do metrology, uncertainty, and traceability play? How can interoperability, data quality, and cybersecurity be ensured? And how can risks be assessed and managed throughout the network life cycle?   The tutorial was developed as part of the FunSNM project and provides practical insight into the infrastructure and risk considerations that support dependable sensor network operation. Watch the tutorial and explore how strong foundations in design and risk evaluation can improve the performance and trustworthiness of distributed sensor networks: https://lnkd.in/eG9HsjxS #SensorNetworks #Metrology #RiskAnalysis #DigitalTwin #Cybersecurity #DataQuality #SmartSystems EURAMET - The European Association of National Metrology Institutes The project Fundamental principles of sensor network metrology (FunSNM) is funded by the European Union. Views and opinions expressed are however those of the author(s) only and do not necessarily reflect those of the European Union or, EURAMET. Neither the European Union nor the granting authority can be held responsible for them. The project has received funding from the European Partnership on Metrology, co-financed from the European Union's Horizon Europe Research and Innovation Programme and by the participating states.

Looking forward to this upcoming online workshop on 6 May 2026 on advances in industrial thermometry and sensor network metrology. A great opportunity to hear from leading experts from across Europe on traceable temperature measurement, sensor networks, digitalization, and industrial applications. If you work with industrial measurements, sensing, or digital quality assurance, this is highly relevant. Free online event – registration required: https://lnkd.in/eRMDUH7e *** The project Fundamental principles of sensor network metrology (FunSNM) is funded by the European Union. Views and opinions expressed are however those of the author(s) only and do not necessarily reflect those of the European Union or, EURAMET. Neither the European Union nor the granting authority can be held responsible for them. The project has received funding from the European Partnership on Metrology, co-financed from the European Union's Horizon Europe Research and Innovation Programme and by the participating states.

Interested in practical tools for sensor network data analysis? In this video tutorial, Dr. Anupam Prasad Vedurmudi from Physikalisch-Technische Bundesanstalt (PTB), Germany, introduces the open-source agentMET4FOF / MET4FOF framework for sensor network applications. Using the open-source Appliance Energy Prediction dataset from Kaggle, the tutorial provides a step-by-step demonstration of how to build and analyse a sensor network simulation. Topics include: ▫️ Anomaly detection ▫️ Data aggregation ▫️ Sensor and monitor agents ▫️ The AgentNetwork object and its backends ▫️ Real-time visualisation via the web-based dashboard ▫️ Online machine learning and analysis under correlated conditions This tutorial is highly relevant for researchers, engineers, and practitioners looking for practical methods and open-source tools for working with sensor network data. Watch the video here: https://lnkd.in/eizKFETC #SensorNetworks #OpenSource #DataAnalysis #AnomalyDetection #MachineLearning #MET4FOF #agentMET4FOF #Tutorial #Research #Engineering The project Fundamental principles of sensor network metrology (FunSNM) is funded by the European Union. Views and opinions expressed are however those of the author(s) only and do not necessarily reflect those of the European Union or, EURAMET. Neither the European Union nor the granting authority can be held responsible for them. The project has received funding from the European Partnership on Metrology, co-financed from the European Union's Horizon Europe Research and Innovation Programme and by the participating states.

FunSNM has published an article in Precision magazine titled “Making sensor networks trustworthy”. The article highlights why trustworthy sensor data matters more than ever as low-cost sensors are deployed at increasing scale and explores the metrology challenges behind making sensor networks reliable in practice. Featuring examples from district heating, gas metering, air quality monitoring, high-temperature manufacturing, and smart buildings, it shows how issues such as drift, uncertainty, correlation and data quality can be addressed at network level. Read the article here (page 6-8):  https://lnkd.in/epkP8HEd The project Fundamental principles of sensor network metrology (FunSNM) is funded by the European Union. Views and opinions expressed are however those of the author(s) only and do not necessarily reflect those of the European Union or, EURAMET. Neither the European Union nor the granting authority can be held responsible for them. The project has received funding from the European Partnership on Metrology, co-financed from the European Union's Horizon Europe Research and Innovation Programme and by the participating states.

Have to share this superb summary of advancements in sensor network metrology from the latest edition of Precision (March 2026 pp6-8). Succinct and easy to read with useful references, I found this a very helpful overview. Those working in digital metrology are likely already aware of the FunSNM project and the significance of what it is trying to achieve, but I think this short article does a great job of articulating it simply and giving application examples, such as air quality monitoring across cities. 💡 If you can't read the whole article, the quotes below are the points I'd pick to highlight why this work is important - "In general, sensors have transitioned from expensive, specialised instruments to cheap, ubiquitous devices. That democratisation is good news because it gives us far more detail about the world than the old sparse measurement networks ever could. But there’s a catch. Low cost sensors drift, they’re affected by the environment, they’re deployed in different ways, and they sometimes fail in correlated ways. When you aggregate thousands of imperfect measurements, naively combining them can give a false sense of security. ... ... One implication of FunSNM’s approach is that you don’t always need better hardware to get better data. By carefully modelling how sensors relate to one another – exploiting redundancy, relative drift patterns and correlations – the project aims to lift the value of existing installed sensors with software and method improvements. For many applications, this approach is much more cost-effective than wholesale hardware replacement. It’s the difference between ‘buying a new sensor’ and ‘using statistics to know if the old one is still good enough’" #metrology #SensorNetworks

How can smart meters move beyond billing and become part of an intelligent sensor network for district heating? In the magazine Hot Cool no. 2 2026, partners in the FunSNM project present how physics-based modelling and advanced analytics can help transform remotely read heat meters into a self-calibrating sensor network. By combining meter data with a digital twin and an Unscented Kalman Filter (UKF), it becomes possible to estimate: ▫️ temperature sensor offset and drift ▫️ pipe insulation performance ▫️ heat loss in specific parts of the network This opens new opportunities for more targeted refurbishment, proactive meter maintenance, improved data quality, and better support for low-temperature district heating. A strong example of how the FunSNM project explores new ways to unlock value from existing infrastructure through sensor network intelligence. Read more in Hot Cool no. 2 2026: https://lnkd.in/eQw-S9FJ #FunSNM #SensorNetworks #DistrictHeating #SmartMeters #DigitalTwin #EnergyInnovation #ThermalNetworks The project Fundamental principles of sensor network metrology (FunSNM) is funded by the European Union. Views and opinions expressed are however those of the author(s) only and do not necessarily reflect those of the European Union or, EURAMET. Neither the European Union nor the granting authority can be held responsible for them. The project has received funding from the European Partnership on Metrology, co-financed from the European Union's Horizon Europe Research and Innovation Programme and by the participating states.  

How do you handle data quality and uncertainty in large‑scale or hard‑to‑access sensor networks, where traditional calibration is not feasible? In this FunSNM tutorial, Mads Johansen from FORCE Technology gives a practical, metrology‑informed overview of: ▫️ Key data quality challenges in sensor networks and IoT systems ▫️ Which data quality characteristics matter most for metrology ▫️ How alternative calibration and data‑driven methods can support uncertainty and traceability ▫️ How to use a structured, standards‑based process to monitor and improve data quality over time The tutorial is designed for researchers, engineers, metrologists, and practitioners working with sensor networks and IoT, as well as students entering this field. Watch the tutorial here: https://lnkd.in/eEyMh5Q6 #metrology #sensornetworks #calibration

In‑situ drift correction for multi‑wire Pt/Pt‑Rh thermocouples Are you working with high‑temperature industrial thermocouples in harsh, long‑duration applications where drift actually matters for your uncertainty budget? We’ve posted a new, deeply technical tutorial on in‑situ calibration of multi‑wire platinum–rhodium thermocouples, focused on correcting calibration drift caused by oxidation and oxide vaporisation. Who is this for? ▫️ Engineers and scientists working with high‑temperature industrial thermocouples. ▫️ Metrologists doing temperature calibration, drift correction, or related thermometry. ▫️ Researchers developing diagnostic or self‑calibrating temperature sensors. ▫️ Practitioners using Pt/Pt‑Rh (noble metal) thermocouples in long‑duration or harsh environments. What the tutorial covers We show how redundancy in multi‑wire Pt/Pt‑Rh structures can be exploited to estimate the true temperature using: ▫️ Physical modelling linking oxide vapour pressure, Rh mass‑fraction changes, and drift behaviour ▫️ Semi‑physical, optimisation‑based estimators ▫️ Purely data‑driven approaches using simple linear relationships between drifted thermocouple pairs You’re guided through the principles behind each method and how they can be implemented in practice. You can expect to learn: ▫️ Why Pt/Pt‑Rh thermocouples drift and how multi‑wire geometries enable self‑calibration ▫️ How physical modelling links oxide vapour pressure, Rh mass‑fraction changes, and drift behaviour ▫️ How simple linear relationships between drifted thermocouple pairs can be used to reconstruct true temperature ▫️ How to formulate and solve an optimisation‑based semi‑physical estimator ▫️ How data‑driven approaches such as co‑integration tests can identify stable thermocouple pairs for drift cancellation ▫️ How these approaches compare in terms of complexity, assumptions, and achievable uncertainty What we demonstrate ▫️ Experimental results from a five‑wire Pt/Pt‑Rh thermocouple dataset, showing real drift behaviour ▫️The physical‑model drift correction, using oxide vapour‑pressure and Rh‑fraction change relationships ▫️ A semi‑physical optimisation method for retrieving the true temperature ▫️ A purely data‑driven method relying on linearity between thermocouple pairs ▫️ Use of statistical tests (e.g., co‑integration) for selecting suitable thermocouple pairs ▫️ Comparison of corrected temperature estimates versus the drifted measurements, showing significant error reduction Watch the video tutorial here: https://lnkd.in/e-2j_-Vp

Join the ThermoSI-FunSNM Online Workshop 6 May 2026. We are pleased to invite you to the joint ThermoSI-FunSNM workshop: “Advances in industrial thermometry & enhancing sensor network metrology for industry and digitalization” Format: Online (free of charge, registration required)   The workshop brings together experts from NPL (UK), PTB (Germany), DTU & DTI (Denmark), VTT (Finland), and Beamex (Finland) to present and discuss: ▫️ The latest developments in traceable temperature measurement for industrial processes. ▫️ Fundamental sensor network metrology and measurement assurance in sensor networks. ▫️ Application areas such as heat treatment, district heating, gas distribution, waste incineration, and air quality monitoring. Highlights include: ▫️ Introduction to ThermoSI – Thermometry with embedded SI traceability for industrial applications. ▫️ Introduction to FunSNM – Fundamental principles of sensor network metrology. ▫️ Talks on SI‑traceable thermal imaging, Johnson noise thermometry, phosphor thermometry, AI approaches to thermometry, and digitalization. 🔗 Register here: https://lnkd.in/eRMDUH7e   We particularly welcome industrial users, partners, and stakeholders interested in reliable, traceable sensor networks and their role in digitalization.

New Physics-Based Model for Drift Correction in Platinum-Rhodium Thermocouples   Accurate high-temperature measurements are vital in many industries, but thermocouples are prone to drift over time, impacting reliability. @Jonathan Pearce and coworkers at The National Physical Laboratory (NPL) recently leveraged a physics-based model of platinum-rhodium thermocouple drift, enabling us to correct and significantly improve thermocouple measurements - especially in complex, multi-sensor networks. Read more about the model in an Open Access-paper by @Jonathan V Pearce in Metrologia: “A validated physical model of the thermoelectric drift of Pt-Rh thermocouples above 1200 °C https://lnkd.in/ehhet8xM   How does it work? The model predicts composition changes (and thus thermoelectric drift) from the vaporization of platinum and rhodium at high temperatures. This allows us to calculate a priori drift ratios between any two thermocouples in a network. With this knowledge, we can correct real-world measurements for drift as they occur - boosting both accuracy and confidence in our temperature data.   Key results: Validation: Applied to multiple data sets, the model-based correction closely matched in-situ drift measurements. Uncertainty quantification: We analyzed how uncertainties in material vapor pressure impact the correction, transparently communicating our confidence bands. See attached figures: Figure 1: Shows measured vs. corrected drift for several Pt-Rh thermocouples compared to true temperature. Figure 2: Illustrates correction uncertainty and validation against real data.   Why does it matter? Reliable, validated correction methodologies help drive better process control, safety, and efficiency - especially in high-tech, energy-intensive industries. We invite you to discuss, share experiences with thermocouple drift, or suggest collaboration opportunities. If you have insights, questions, or related challenges, please comment below or reach out to our project team. Let’s advance measurement science together! #thermocouples #instrumentation #materials #physics #RD #temperaturemeasurement