Gas sensing device and method for operating a gas sensing device

Patent metadata

Field	Value
Publication number	US-11428658-B2
Application number	US-202016777065-A
Country	US
Kind code	B2
Filing date	Jan 30, 2020
Priority date	Feb 25, 2019
Publication date	Aug 30, 2022
Grant date	Aug 30, 2022

Abstract

A gas sensing device includes one or more chemo-resistive gas sensors; one or more heating elements for heating each of the gas sensors; a preprocessing block for filtering signal samples in order to generate filtered signal samples for each of the gas sensors; an information extraction block for generating representations for the filtered signal samples for each of the gas sensors based on dynamic characteristics of the received filtered signal samples of the respective gas sensor; and a decision making block for receiving the representations, wherein the decision making block includes a trained model based algorithm stage having an input layer and an output layer, wherein the decision making block includes trained models, wherein the decision making block creates sensing results based on output values of the output layer of the algorithm stage, and wherein the output values are created by using the trained models.

First claim

What is claimed is:

1. A gas sensing device for sensing one or more gases in a mixture of gases, the gas sensing device comprising:
one or more chemo-resistive gas sensors, wherein each of the gas sensors is configured for generating signal samples corresponding to a concentration of one of the one or more gases in the mixture of gases, wherein the one or more gas sensors are alternately operated in recovery phases and in sense phases, wherein the signal samples of each of the gas sensors are generated during the sense phases;
one or more heating elements for heating each of the gas sensors, wherein the one or more heating elements are brought to a first temperature during the recovery phases and to a second temperature during the sense phases, wherein the first temperature is higher than the second temperature;
a preprocessing block configured for receiving the signal samples from each of the gas sensors and for filtering the received signal samples in order to generate filtered signal samples for each of the gas sensors;
an information extraction block configured for receiving the filtered signal samples and for generating representations for the received filtered signal samples for each of the gas sensors based on dynamic characteristics of the received filtered signal samples of the respective gas sensor; and
a decision making block configured for receiving the representations, wherein the decision making block comprises a trained model based algorithm stage having an input layer and an output layer, wherein the decision making block comprises one or more trained models for the algorithm stage, wherein the representations for each of the gas sensors are input to the input layer of the algorithm stage, wherein the decision making block creates for each of the gas sensors sensing results based on output values of the output layer of the algorithm stage, wherein the output values for each of the gas sensors are created by using at least one of the one or more trained models at the algorithm stage so that the output values for each of the gas sensors depend on the representations of each of the gas sensors.

2. A gas sensing device according to claim 1 , wherein the gas sensing device comprises one or more auxiliary sensors, wherein each of the auxiliary sensors is configured for generating auxiliary signal samples corresponding to a physical quantity of operating conditions of the gas sensing device, and
wherein the one or more auxiliary sensors comprise
a first temperature sensor for generating first auxiliary signal samples of the auxiliary signal samples, which correspond to a temperature of the one or more heating elements, and/or a second temperature sensor for generating second auxiliary signal samples of the auxiliary signal samples, which correspond to an ambient temperature of the gas sensing device, and/or
a humidity sensor for generating third auxiliary signal samples of the auxiliary signal samples, which correspond to an ambient humidity of the gas sensing device.

3. A gas sensing device according to claim 2 , wherein the decision making block is configured for selecting one or more selected trained models from the one or more trained models based on the auxiliary signal samples of the one or more auxiliary sensors, wherein the output values for the one or more gas sensors are created by using the one or more selected trained models.

4. A gas sensing device according to claim 2 , wherein the preprocessing block is configured for receiving the auxiliary signal samples from each of the auxiliary sensors and for filtering the received auxiliary signal samples in order to generate filtered auxiliary signal samples for each of the auxiliary sensors,
wherein the information extraction block is configured for receiving the filtered auxiliary signal samples and for generating auxiliary representations for the received filtered auxiliary signal samples for each of the auxiliary sensors based on dynamic characteristics of the received filtered auxiliary signal samples of the respective auxiliary sensor, and
wherein the decision making block is configured for inputting the auxiliary representations for each of the auxiliary sensors to the input layer of the algorithm stage so that the output values for the one or more gas sensors depend on the auxiliary representations of each of the auxiliary sensors.

5. A gas sensing device according to claim 2 , wherein the pre-processing block comprises a defect detection stage configured for a detection of defects of the gas sensing device, wherein the detection is based on the auxiliary signal samples of one or of the more auxiliary sensors.

6. A gas sensing device according to claim 1 , wherein the algorithm stage comprises a neural network using the one or more trained models and/or a random decision forest using the one or more trained models.

7. A gas sensing device according to claim 1 , wherein the one or more trained models comprise one or more trained multiple-output models having a plurality of outputs, wherein for all of the gas sensors the output values for the respective gas sensor are created by using one of the trained multiple-output models at the algorithm stage, wherein each of the output values is created at a different output of the plurality of outputs.

8. A gas sensing device according to claim 1 , wherein for each of the gas sensors the one or more trained models comprise one or more trained single-output models having a single output, wherein the output values for different gas sensors of the gas sensors are created by using different single-output models of the trained single-output models at the algorithm stage.

9. A gas sensing device according to claim 1 , wherein the preprocessing block comprises a baseline manipulation stage configured for transforming the signal samples of each of the gas sensors into a relative resistance change according to a baseline of the signal samples of the respective gas sensor.

10. A gas sensing device according to claim 1 , wherein the preprocessing block comprises an artefact detection stage configured for a detection artefacts in the signal samples of each of the gas sensors, wherein the detection for each of the gas sensors is based on comparing of sequential signal samples of the respective gas sensor.

11. A gas sensing device according to claim 1 , wherein the preprocessing block comprises a moving mean filter stage configured for reducing noise in the filtered signal samples for each of the gas sensors.

12. A gas sensing device according to claim 1 , wherein the information extraction block comprises one or more feature extraction stages, wherein each of the feature extraction stages is configured for calculating a feature of one of the filtered signal samples for each of the gas sensors, wherein each of the features refers to the dynamic characteristics of the received filtered signal samples of the respective gas sensor.

13. A gas sensing device according to claim 12 , wherein the feature extraction stages comprise a derivative calculation stage configured for calculating a derivative of the one of the filtered signal samples for each of the gas sensors, wherein the derivative is one of the features of the one of the filtered signal samples.

14. A gas sensing device according to claim 12 , wherein the feature extraction stages comprise a phase space integral calculation stage configured for calculating a phase space integral of the one of the filtered signal samples for each of the gas sensors, wherein the phase space integral is one of the features of the one of the filtered signal samples.

15. A gas sensing device ac

Assignees

• Infineon Technologies Ag

Inventors

• Carbonelli Cecilia
• Carro Dominguez Manuel
• Zhao Jianyu

Classifications

• G06F18/24323

  Tree-organised classifiers · CPC title

• G06N3/0455

  Auto-encoder networks; Encoder-decoder networks · CPC title

• G06N3/09

  Supervised learning · CPC title

• G06N3/0499

  Feedforward networks · CPC title

• G06N3/04

  Architecture, e.g. interconnection topology · CPC title