Neural Network Sensor Compensation Method
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The existing sensor compensation processes, particularly for pressure sensors, are time-consuming and costly due to the need for numerous temperature and pressure setpoints, which also result in high energy consumption and a significant CO2 footprint.
Innovation Solution
A computer-implemented method using machine learning, specifically a neural network, to determine compensation coefficients for pressure sensors, reducing the number of setpoints required by using previously collected data and historical production line data to generate a look-up table or polynomial functions for faster compensation.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If traditional compensation process with multiple temperature cycles and setpoints is used, then sensor accuracy is improved, but manufacturing time and energy consumption increase
Solution Approach 1:
The patent applies preliminary action by collecting and storing sensor data from multiple sensors during their manufacturing process before compensation is needed. This historical data is stored in a database and later used by the neural network to determine compensation coefficients, eliminating the need for time-consuming temperature cycling during the compensation phase.
Solution Approach 2:
The patent replaces the mechanical/physical temperature cycling system with a computational system. Instead of physically heating and cooling sensors through multiple temperature cycles, a neural network processes previously collected data to calculate compensation coefficients, substituting physical thermal processes with information processing.
2Measurement precision
If traditional compensation process with multiple temperature cycles is used, then sensor accuracy is improved, but energy consumption and CO2 footprint increase
Solution Approach 1:
The patent replaces the energy-intensive thermal cycling system with a computational neural network approach. The neural network processes previously collected sensor data to determine compensation coefficients without requiring active heating or cooling, dramatically reducing energy consumption and associated CO2 emissions.
Solution Approach 2:
The patent uses data copies from previously manufactured sensors to compensate the current sensor. Instead of physically subjecting the sensor to multiple temperature cycles, the neural network analyzes copies of measurement data from the database to determine appropriate compensation coefficients, avoiding repeated energy-intensive thermal processing.
3Measurement precision
If high number of setpoints are used for compensation, then sensor accuracy is improved, but manufacturing cost and time increase
Solution Approach 1:
The patent performs preliminary data collection during the manufacturing process, storing measurement data from multiple sensors in a database. This preliminary action creates a rich dataset that the neural network can later process efficiently to determine compensation coefficients without requiring additional time-consuming measurement cycles.
Solution Approach 2:
The patent substitutes the mechanical process of generating numerous setpoints through temperature cycling with an automated neural network system. The neural network processes previously collected data to generate compensation coefficients, replacing the manual, time-intensive setpoint generation process with automated computational processing.
Data Source
AI summary
A computer-implemented method for compensating a sensor through machine learning during the manufacturing of the sensor is provided. The first step includes providing a plurality of sensors. The second step includes determining data that have been already collected during the manufacturing and/or compensating of the sensor or the plurality of sensors. The determined data is provided to a neural network configured to determine compensation coefficients. The determined compensation coefficients are saved within the sensor so that the sensor can output compensated sensor values with the help of the determined compensation coefficients.