Neuron Virtual Sensor

Neuron Virtual Sensors

We’ve been looking forward to offering our customers a solution that also includes virtual sensors – with countless applications. Well, now it’s here! Many of you have been asking for this product, which is now ready for launch. By using two or more physical sensors and a formula that calculates the relationship between them, you can easily gain entirely new insights into measurement points of great value for the business. Here, we’ll explain how and provide some examples of applications that illustrate how virtual sensors can be used.

Introduction to Neuron Virtual Sensors

A Neuron IoT sensor measures a value or state in the real world. However, the sensor’s value sometimes does not clearly convey the situation unless you recalculate it or compare it with data from other sensors.

A Neuron Virtual sensor uses a mathematical expression in combination with measurements from one or more physical sensor measurements to create a more meaningful value for the use case.

You can visualize and use the Neuron Virtual sensor just like any physical Neuron IoT sensor. Graphs show how the value changes over time. Rules and alerts notify you if the value exceeds or falls below user-selected thresholds. Additionally, you can transfer the Neuron Virtual sensor value into any other software for further analysis, logging, or action via various APIs.

Example 1 – Temperature difference

For some assets, a high or low temperature is not a clear indication of an issue, but a significant difference in temperature between two measurement points on the same asset will indicate an issue.

You can monitor the temperature difference by using two Neuron IoT sensors to measure temperatures and a Neuron Virtual sensor to calculate the temperature difference.

The formula for calculating the difference in temperature:

  • X = Temperature sensor 1
  • Y = Temperature sensor 2
  • Temperature difference:  

A sawmill requested this function to better control its band saws by monitoring the temperature difference between the tooth side and the back side of the saw blade.

To measure these two temperatures, you can use a Neuron Infrared 380 sensor ( This sensor can read the temperature of an object without touching it, which is very handy for fast-moving objects such as a saw blade.

The surrounding temperature affects the actual temperature of the saw blade, which increases when the saw is in use and rises even more with prolonged use. These variations do not indicate how well the saw performs. However, if the tooth side of the saw blade becomes significantly hotter than the back side, the wood is probably clamping the saw blade. Over time, this will damage the saw blade. By monitoring the temperature, you can detect this condition and take measures before the saw breaks down.

See also

Product brochure

Find the perfect sensor solution

We offer over 50 different sensor types. Explore the possibilities in our product brochure.

Example 2 – Deviation from average

If several assets should have similar temperatures, but the actual temperature can vary due to outside temperature, process steps, or other factors, it is helpful to receive a warning if one asset deviates significantly from the average temperature.

For this, a Neuron Virtual sensor calculates the deviance between each physical temperature sensor and the average temperature of all sensors.

The formula for calculating deviation in temperature from the average of three sensors:

  • A = Temperature sensor 1
  • B = Temperature sensor 2
  • C = Temperature sensor 3
  • Deviation from average:

A company requested this function to detect leakage from their cryogenic tanks. These tanks contain liquid nitrogen for very low-temperature storage of their products. If one of the tanks has a minor leakage, the stored products will be damaged. Detecting minor leakages by measuring temperatures inside the tank is not possible.

You can detect a very small leakage by measuring the temperature outside the tank because the surrounding air temperature will be lower due to the leaking nitrogen. Each tank will have a Neuron Temperature IP67 sensor installed (

The actual air temperature outside the tank will vary with weather and season. However, if the air temperature at one of the tanks is significantly lower than at the other tanks, it is probable that this tank is leaking.

Example 3 – Dew point measurements

In some industries, monitoring the dew point is important to prevent condensation and protect equipment, and in certain industries, it is crucial for safety reasons.

The dew point of a given body of air is the temperature at which it must become saturated with water vapor.

To measure the dew point temperature, you need a Neuron Humidity sensor (, which measures both air temperature and relative humidity. A Neuron Virtual sensor can calculate the dew point temperature.

If a material with a lower surface temperature than the dew point temperature is present, water vapor in the air close to the material will condense and form liquid water known as dew.

The formula for calculating dew point temperature:

  • R = Relative humidity in %
  • T = Air temperature in °C
  • Dew point temperature (approximate):
  • M = Material temperature in °C
  • Deviation from material to dew point temperature:

This function has been requested by a shipyard that welds hulls to be used by ocean-going ships. To deliver hulls that can operate safely at sea for decades, the welding wire needs to be of the highest standards. Any water on the welding wire will cause a deterioration of the quality of the welding wire, and thus also on the welding and the hulls.

If the temperature of the welding wire is always equal to the air temperature, dew will not be present. However, when the air temperature in the storage increases, it may take a while until the welding wire reaches the same temperature. In such situations, the welding wire temperature could be lower than the dew point temperature, and dew will occur.

You measure the air temperature and relative humidity with a Neuron Humidity sensor and use these parameters to calculate the air’s dew point temperature on the premises. You use a Neuron PT100 Surface Patch sensor ( to measure the surface temperature of the welding wire.

The system uses a virtual sensor to calculate the difference between material temperature and dew point temperature. The virtual sensor is set to alert if this difference falls below 0°C, indicating the presence of dew. This alert signals the need for actions or analysis before using the welding wire. The Neuron system, which stores all data from physical and virtual sensors in the cloud, also provides the necessary documentation to ensure compliance with quality standards regarding dew point monitoring.

Ready to try wireless sensors?

Please contact our Sales Manager for some good advice and a nice sensor- chat.

Example 4 – Tank volume in horizontal cylindrical tank

There are several ways to measure the filling height of a tank with Neuron IoT sensors.

Measuring filling height accurately measures the volume of liquid in a tank if the tank’s surface area is uniform at all heights. However, this is not the case in a horizontal cylindrical tank such as a drum lying on its side. You need more advanced calculations to determine the liquid’s volume.

You can monitor the filled volume of a horizontal cylindrical tank by using a Neuron IoT sensor to measure the filling height and a Neuron Virtual sensor to calculate the volume.

The formula for calculating the volume of horizontal cylindrical tank point temperature:

  • L = Length of container (height went standing up-right)
  • R = Radius of container
  • H = Filling height
  • Volume:

A workshop for buses and trucks requested this function to store oil, antifreeze, washer fluid, and other fluids in horizontal oil drums.

A sensor probe that is inserted into the oil drum is used to measure filling height. The sensor probe supplies a 4-20 mA output with a linear relationship to filling height. Typically, 4 mA will represent 0, and 20 mA will represent the maximum height measured by the sensor probe. The 4-20 mA output is connected to a Neuron Precision mA sensor (, which provides the measurement from the sensor probe to the Neuron Cloud.

A virtual sensor is used to calculate the filled volume.

Monitoring the tank’s volume makes it easier to detect when you are soon to run out and need to refill or order more.

See also:

Example 5 – Heat exchange rate in ventilation systems

Ventilation systems use heat exchangers to heat intake air by using energy from the exhausted air. You can recover up to 83% of the energy from the exhaust. This figure, known as the “Temperature Transfer Efficiency” or the “Heat Exchange Rate,” measures the efficiency of the heat exchanger.

Neuron Temperature IP67 sensors ( are used to measure temperatures at three points in the ventilation system, the exhaust air before and after the heat exchanger, and the intake air before the heat exchanger. The intake air before the heat exchanger very often equals outdoor temperature.

A virtual sensor is used to calculate the heat exchange rate.

The formula for calculating heat exchange rate:

  • B = Exhaust air temperature before heat exchanger (inside)
  • A = Exhaust air temperature after heat exchanger (outside)
  • I = Intake air temperature before heat exchanger (outside)
  • Heat exchange rate in percent:

This function was requested by a company that supplies and services ventilation systems. By monitoring the heat exchange rate, they will be able to advise their customers when the ventilation system requires maintenance.

Since heat exchangers wear out over time, it will be financially beneficial to invest in a new one after some years, when the cost of investment is lower than the additional cost of energy consumption caused by the old one. By monitoring this, the company can advise when it would be wise to replace heat exchangers.

See also:

Reduce electricity costs with sensors in ventilation systems (SWE version with ENG txt)

FAQs about Virtual Sensors

1. What is the difference between physical and virtual sensors?

    Physical sensors are hardware devices that measure physical quantities like temperature, humidity, pressure, or motion directly from the environment. They convert these physical measurements into electrical signals that can be processed by a computer or a monitoring system.

    Virtual sensors, on the other hand, are software-based models or algorithms that estimate or infer sensor data based on information from physical sensors and other data sources. Virtual sensors use mathematical models and data analytics to provide measurements that might not be directly accessible by physical sensors alone.

    2. What is an example of a virtual sensor?

      An example of a virtual sensor is an indoor air quality index (IAQI) estimator. Instead of directly measuring all possible pollutants, the virtual sensor might use data from physical sensors that measure temperature, humidity, CO2 levels, and particulate matter to estimate the overall air quality.

      3. How do you make a virtual sensor?

        Creating a virtual sensor involves several steps:

        • Data Collection: Gather data from physical sensors and other relevant sources.
        • Model Development: Develop a mathematical or machine learning model that can infer the desired measurements from the collected data.
        • Calibration and Validation: Calibrate the model using known data and validate its accuracy against actual measurements.
        • Integration: Integrate the virtual sensor into the system, ensuring it can process real-time data and provide accurate estimations or measurements.

        4. How to install a virtual sensor?

          Installing a virtual sensor typically involves the following steps:

          • Software Installation: Install the software that includes the virtual sensor model on the target system (e.g., a server, cloud platform, or embedded device).
          • Data Integration: Ensure the virtual sensor has access to the necessary input data from physical sensors and other sources.
          • Configuration: Configure the virtual sensor with the appropriate settings, such as data input formats, sampling rates, and thresholds.
          • Testing: Test the virtual sensor to verify it is functioning correctly and providing accurate outputs.

          5. What are the advantages of virtual sensors?

            Virtual sensors offer several advantages:

            • Cost-Effective: Reduce the need for multiple physical sensors, lowering hardware costs.
            • Flexibility: Easily adapt to changes in measurement requirements without needing new hardware.
            • Enhanced Insights: Provide complex measurements and insights that might be difficult or impossible to obtain with physical sensors alone.
            • Scalability: Scale more easily by deploying software updates rather than physical devices.

            6. What are online sensors?

              Online sensors are sensors that provide real-time data and are connected to a network, enabling continuous monitoring and immediate data processing. These sensors can transmit data to cloud-based platforms or other systems for instant analysis and action.

              7. What is a virtual proximity sensor?

                A virtual proximity sensor uses data from other physical sensors (e.g., GPS, accelerometers, and Wi-Fi signals) to estimate the proximity of an object or person. For instance, in a mobile device, it might use a combination of signal strength and motion data to determine how close the device is to another object.

                8. What is virtual light sensing?

                  Virtual light sensing involves using software algorithms to estimate light levels in an environment based on data from other sensors, such as cameras or ambient light sensors. This can be used in applications where direct measurement of light is either not feasible or needs to be supplemented with additional context, like in smart lighting systems that adjust based on both occupancy and estimated natural light levels.


                  Further reading