Virtual Sensor Modeling via Feature Signal Filtering
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In complex systems, such as electronic or mechatronic devices and engines, the installation of sensors is hindered by limited space and unfavorable environmental conditions, leading to inaccurate sensor values due to changing conditions, resulting in high costs and inefficiencies.
Innovation Solution
A method involving feature signal filters is used to process measurement data from multiple sensors, employing frequency analysis and band pass filters to select relevant frequency ranges, allowing for the creation of a mathematical model that simulates target sensor signals, potentially dispensing with the need for physical target sensors by filtering out irrelevant signal components.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If sensors are installed to monitor system parameters, then measurement capability is improved, but installation space is consumed and environmental wear increases
Solution Approach 1:
The patent creates a virtual copy of the sensor through a mathematical model that replicates sensor behavior. Instead of installing physical sensors in difficult-to-access locations, the system uses feature sensors to capture data and generates a virtual sensor model that produces identical or complementary measurements, thereby eliminating the need for additional physical sensors in challenging environments
Solution Approach 2:
The patent introduces an intermediary mathematical model that translates data from feature sensors into target sensor signals. This model acts as a mediator between the physically accessible feature sensors and the target measurement locations that are difficult to access, enabling indirect measurement through the intermediary transformation
2Measurement precision
If sensors are installed in suitable locations, then measurement accuracy is improved, but environmental conditions cause wear and inaccurate values
Solution Approach 1:
The patent creates a virtual replica of the sensor that does not suffer from physical environmental wear. The mathematical model captures the essential sensor behavior and produces consistent measurements regardless of harsh thermal, mechanical, or chemical conditions that would degrade physical sensors over time
Solution Approach 2:
The patent transforms physical sensor measurements into mathematical representations by changing the domain from physical quantities to processed data. This parameter transformation allows the system to maintain measurement reliability by working with digital models rather than physical sensors exposed to degrading environmental conditions
3Measurement precision
If multiple sensors are used to monitor complex systems, then comprehensive measurement is improved, but system cost increases
Solution Approach 1:
The patent makes the feature sensors multi-functional by using them for both direct measurement and as input data for generating virtual sensor signals. Instead of installing separate sensors for each measurement type, the system leverages the universal capability of feature sensors to provide data that can be transformed into multiple types of sensor signals through the mathematical model
Solution Approach 2:
The patent generates virtual copies of sensor signals from feature sensor data, eliminating the need for multiple physical sensors. The mathematical model creates synthetic sensor readings that replicate the behavior of actual sensors, providing comprehensive system monitoring with fewer physical components
4Productivity
If calculation models are used to simulate sensor values, then virtual testing is improved, but matching accuracy with real sensor values deteriorates
Solution Approach 1:
The patent incorporates feedback mechanisms where the mathematical model continuously refines its predictions by comparing virtual sensor outputs with actual feature sensor measurements. This feedback loop adjusts the model parameters to improve matching accuracy between simulated and real sensor values, ensuring high precision in virtual testing
Solution Approach 2:
The patent performs preliminary processing of feature sensor data through frequency analysis and filter design before feeding it into the mathematical model. This preliminary action prepares the data in an optimized format that enhances the model's ability to accurately generate matching sensor values, improving virtual test precision before the actual simulation occurs
Data Source
AI summary
A method for determining feature signal filters for preparing signal measurement data sequences of a plurality of measurement variables for experimentally determining a mathematical model which maps model measurement data for at least one target signal sensor (7) on the basis of detected measurement data of a plurality of feature signal sensors (5) is disclosed. Further disclosed is a method for determining a mathematical model (16) which maps model measurement data for at least one target signal sensor (7) on the basis of detected measurement data of a plurality of feature signal sensors (5), wherein training input measurement data sequences (19) ascertained by the feature signal sensors (5) are mapped onto at least one training output measurement data sequence (18) ascertained by at least one target signal sensor (7).