Answering the WHAT question

Create virtual sensors, estimators and classifiers to measure impossible things, replace sporadic lab measurements, estimate and classify KPIs and much more.

Learn from history to understand the present

aivis predict is all about the What. Learning dependencies and relationships on historical data, it creates predictors (independent software modules), that determine what is happening right now, what comes out if and what is the current situation.

These predictors can be used to predict, estimate or classify a certain quantity or state when fed with the latest data. They are very light-weight, independent software modules that can easily be implemented into any SW environment.

Applications

VIRTUAL SENSING – quality Assurance – error detection – root cause indicators – WASTE reduction

Up to now, creating virtual sensors, estimators or classifiers could only be done with heavy data science expertise, which was challenging and expensive. With aivis Predict you can create these now fast, simple and fully automated. No data science expertise needed.

Build Virtual Sensors

Datatype: Time-series

Available: Project

Virtual Sensors can bring significant benefits to your company. They can replace expensive or error-prone real sensors, minimize the need for lab measurements, or directly measure impossible or abstract parameters like quality or durability. To do so, they have to be implemented into your live data stream.

If you want to evaluate, if your historical data contains enough information to create a virtual sensor, you can use aivis Insights application ‘Find signal depedencies’.

Build estimators

Datatype: Tabular

Available: Project

Estimators are great at handling highly complex and non-linear relationships in your data. They act like look-up tables that return an estimated target value for multiple specific input values. They are used to estimate KPIs like the quality of a workpiece, physical properties like oil viscosity or concrete strength, and much more.

If you want to find out, if your historical data contains enough information to create an estimator, you can use aivis Insights application ‘Find influencing factors’.

Build classifiers

Datatype: Time-series & tabular

Available: Project

Use a classifier to figure out, if a certain quantity is still within its target range or is about to leave it. 

If you want to find out, if your historical data contains enough information to create an classifier, you can use aivis Insights application ‘Find influencing factors’.


Examples

VIRTUAL SENSING – QUALITY ASSURANCE – ERROR DETECTION – WASTE REDUCTION

Browse through our selection of different use cases, where aivis Insights has already generated huge value. Of course, those cases are only exemplary since aivis Insights is industry-agnostic and can be applied to countless other cases as well.

Paper making – Live measuring paper quality

Application: Build virtual sensors

During the continuous process of papermaking, paper quality is a critical process parameter. A direct and online determination of this parameter is impossible since a lab measurement is required. Since each measure takes about one hour from sampling to the result, the operator is flying blind, so to speak, until the next lab value is available.

Since aivis Insights already validated that the paper quality can be well predicted, aivis Predict could easily create a virtual sensor for the paper quality. The live quality prediction enabled the operator to continuously and instantaneously adjust the process, which lowered necessary safety margins, significantly reduced material consumption, and increased throughput.

Dead oil – Predicting oil viscosity

Application: Building estimators

The dead oil viscosity is a critical process parameter that heavily depends on the type of oil. Even slight differences in composition can dramatically impact the viscosity, making it almost impossible to address this issue with classical black oil correlations.

Since aivis Insights already validated that viscosity can be well predicted, aivis Predict could easily create an estimator for the viscosity. At each moment, this estimator can be fed with the current process parameters giving back the resulting oil viscosity.

Concrete – Measuring compressive strength

Application: Building estimators

The concrete compressive strength is a critical process parameter that determines the quality of the concrete. It depends highly nonlinear on age and ingredients. Usually, it has to be determined in the laboratory at great expense using a crushing test on concrete cylinders, which can take up to a month and is prone to human error.

aivis Predict created an estimator for the concrete strength to determine the compressive strength immediately and onsite from easily measurable process parameters.

Mining – Measuring Silica during flotation

Application: Build virtual sensors

In a flotation plant, minerals are processed to concentrate the iron ore. A critical process parameter is the percentage of Silica, which is the impurity of the iron ore. This value can only be determined with one hour delay by a lab measurement.

aivis Predict created a virtual sensor for Silica, which determines the iron ore’s impurity value from other well-known and easily measurable parameters. Based on this live prediction, the engineers can take corrective actions during the process to reduce impurity. Fewer impurities mean less ore goes to tailings where Silica is reduced in the ore concentrate. This reduces costs and effort and has a positive effect on the environment.

Liquid tank – Measuring temperature in the middle

Application: Build virtual sensors

A closed liquid tank has sensors to measure the temperature at the inlet and outlet. Besides, the temperature in the middle of the tank should be measured. This isn’t easy because the tank needs to be opened to place a sensor in the middle.

aivis Predict created a virtual sensor for the middle temperature using the data of one opened tank. It predicts the temperature in the middle based on the inflow and outflow of the tank. This sensor could be reused for all other tanks as well without opening them.

Nuclear fusion – Measuring reactor temperature

Application: Build virtual sensors

The temperature in a nuclear fusion reactor cannot be measured directly because it is too hot. However, many other values can be measured, such as pressure, density, etc.

Based on a simulation of the reactor, aivis Predict can create a virtual sensor for the temperature based on the other available parameters. The virtual sensor can then used for the real application.

Virtual Sensors

Virtual Sensors can bring significant benefits to your company. They can replace expensive or error-prone real sensors, minimize the need for lab measurements, or directly measure impossible or abstract parameters like quality or durability. Up to now, creating Virtual Sensors was hard and expensive. With aivis Predict, it can be done quick and simple.

Replacing real sensors

Sometimes physical sensors are too expensive or are located in an environment where they are prone to failure or have a short life expectancy.

Measure impossible things

Determine abstract parameters like quality or lifespan, which cannot be measured by any physical methods.

Sensor fusion

Reducing measurement errors of real measured values based on correlations with other physical measured values.

Less lab measurements

Create virtual sensors for live measuring parameters that otherwise can only be determined in the lab with a time delay.

Simulations

Determine non-measurable variables such as the temperature inside a reactor with a soft sensor that has been trained using simulation data.

Increase number of sensors

Increase the number of sensors to monitor your process by combining virtual and physical sensors.

Fill out the form to talk to our experts