DE102019132011A1 - Method for determining an engine parameter - Google Patents

Abstract

The invention relates to a method for determining an engine parameter, a computer program product and a device for performing the method.
The method for determining an engine parameter comprises the steps:
a) providing (S1) sound information (5) of an engine;
b) determining (S2) engine-characteristic features of the sound information (5) by means of an artificial neural network which is designed and set up to determine engine-characteristic features of different engine types and / or engine series;
c) Determining (S3) the engine parameter based on the engine characteristic features.

Description

The invention relates to a method for determining an engine parameter, a computer program product and a device for performing the method.

From the US008589115B2 A device and a method for estimating a torque and a rotational speed of a motor are known. The system includes a sound receiving device, a feature extraction device and an artificial neural network module.

1. a) Bereitstellen von Schallinformationen eines Motors;
2. b) Bestimmen von motorcharakteristischen Merkmalen der Schallinformationen mittels eines künstlichen neuronalen Netzwerkes, welches dazu ausgebildet und eingerichtet ist, motorcharakteristische Merkmale unterschiedlicher Motortypen und/oder Motorbaureihen zu bestimmen;
3. c) Bestimmen des Motorparameters anhand der motorcharakteristischen Merkmale.

The method according to the invention for determining an engine parameter comprises the steps:

1. a) providing sound information of an engine;
2. b) determining engine-characteristic features of the sound information by means of an artificial neural network which is designed and set up to determine engine-characteristic features of different engine types and / or engine series;
3. c) Determining the engine parameter based on the engine characteristic features.

When operating motorized machines, engine parameters, such as, for example, a speed and / or a torque of a motor, are typically monitored by means of a machine-integrated sensor system. The data obtained is often used for information purposes for machine users, to reduce fuel consumption or to avoid engine damage.

In the method according to the invention, engine parameters are determined on the basis of engine-characteristic features of sound information from engines. The determination is preferably carried out without the aid of machine-integrated sensors. An artificial neural network is used to determine the motor characteristic features. Because the artificial neural network is designed and set up to determine engine-characteristic features of different engine types and / or engine series, the method according to the invention enables engine parameters to be determined independently of the engine type and / or engine series.

The method according to the invention is preferably provided for determining engine parameters of vehicle-integrated engines. The motors are particularly preferably electric motors and / or internal combustion engines.

The computer program product according to the invention, when executed by a computer, causes the computer to carry out a method according to the invention for determining an engine parameter.

1. a) einen Sensor, der dazu ausgebildet ist, einen Schall zu messen und
2. b) einen Rechner, der dazu ausgebildet und eingerichtet ist, ein erfindungsgemäßes Computerprogrammprodukt auszuführen.

The device according to the invention for determining an engine parameter comprises

1. a) a sensor which is designed to measure a sound and
2. b) a computer which is designed and set up to execute a computer program product according to the invention.

The dependent claims describe further advantageous embodiments of the invention.

• eine schematische Darstellung eines Ausführungsbeispiels einer Vorrichtung zum Bestimmen eines Motorparameters,
• eine schematische Darstellung eines Ausführungsbeispiels eines Verfahrens zum Bestimmen eines Motorparameters und
• eine schematische Darstellung von Schallinformationen.

Preferred exemplary embodiments are explained in more detail with reference to the following figures. It shows

• 1 shows a schematic illustration of an exemplary embodiment of a device for determining an engine parameter,
• a schematic representation of an embodiment of a method for determining an engine parameter and
• a schematic representation of sound information.

In the exemplary embodiment, the device 1 (see a mobile phone to determine an engine parameter. The mobile phone 1 is located in a vehicle interior of a vehicle and is operated by a user. The vehicle has a four-cylinder petrol engine. The petrol engine is in operation and causes a noise emission that can be measured in the vehicle interior.

The mobile phone 1 includes a sensor 2 , which is designed to measure the noise emission of the gasoline engine. The sensor 2 is a microphone, which is in the mobile phone 1 is integrated. The mobile phone also includes 1 a screen 4 , which is designed to display an engine parameter. The mobile phone also includes 1 a calculator 3 , which is designed and set up to carry out an embodiment of a computer program product for determining an engine parameter.

The computer program product is a mobile application and prompts when it comes from the computer 3 running the calculator 3 to an embodiment of a method (compare ) to determine an engine parameter.

The process is started by the user using the mobile application on the mobile phone 1 activated.

In one step S0 of the method, the user provides engine-specific information. The specification is made by means of a manual entry on a control panel of the mobile phone 1 , In the exemplary embodiment, engine-specific information is the number of cylinders of the gasoline engine.

In a further step S1 the procedure becomes sound information 5 provided (compare ). It is provided on the basis of a single-channel sound measurement of airborne sound using the microphone 2 , In the exemplary embodiment, the measured sound information is 5 temporal courses of airborne sound pressures 6 ,

In a further step S1 - 2 the procedure is the sound information 5 edited. The preparation takes place by means of several algorithms that are generated by the computer 3 be carried out. A first algorithm transfers the time profiles of the airborne sound pressures 6 using a discrete Fourier transform into a discrete frequency spectrum 7 , The frequency spectrum 7 includes data points 8th of sound pressure amplitudes in a frequency range between 16 Hz and 300 Hz. The frequency resolution of the frequency spectrum is 4 Hz. A second algorithm squares the sound pressure amplitude values of each data point 8 x _i and divides the resulting value by the sum of the squared sound pressure amplitude values of all data points 8 Σx _i ² of the frequency spectrum 7 , Through the associated transfer of absolute values of the sound information 5 A computing effort of an artificial neural network in the step becomes relative values S2 of the procedure reduced.

In step S2 of the method become characteristic features of the sound information 5 determined by means of the artificial neural network. In the exemplary embodiment, engine-characteristic features are speed-dependent engine type and engine series characteristic ratios of squared sound pressure values. The artificial neural network is a multi-layer perceptron, which was trained using a variety of engine sound data from engine test bench measurements of different vehicles with different engine types and engine series. As a result, it is designed and set up to determine engine-characteristic features of different engine types and / or engine series.

In a further step S3 In the method, an engine parameter is determined based on the engine characteristic features. In the exemplary embodiment, the engine parameter is an engine speed. When determining, the number of cylinders of the petrol engine is taken into account as predetermined engine-specific information. The engine speed is determined using an algorithm that compares the engine characteristics with typical noise emissions from petrol engines with four cylinders.

In the exemplary embodiment, the sound information is provided S1 and determining S3 the motor speed is 50 ms.

In a further step S4 The procedure will display the engine speed on the screen 4 of the cell phone 1 displayed (compare ).

In a further exemplary embodiment of the method (not shown), a motor torque and a motor speed of an electric motor with three pole pairs are determined. In the exemplary embodiment, the number of pole pairs is motor-specific information which is obtained from sound information from the electric motor by means of a neural network. In the exemplary embodiment, the artificial neural network is a recurrent or feedback neural network (RNN).

QUOTES INCLUDE IN THE DESCRIPTION

This list of documents listed by the applicant has been generated automatically and is only included for the better information of the reader. The list is not part of the German patent or utility model application. The DPMA assumes no liability for any errors or omissions.

Patent literature cited

• US 008589115 B2 [0002]

Claims (10)

Method for determining an engine parameter, comprising the steps: a) providing (S1) sound information (5) of an engine; b) determining (S2) engine-characteristic features of the sound information (5) by means of an artificial neural network which is designed and set up to determine engine-characteristic features of different engine types and / or engine series; c) Determining (S3) the engine parameter based on the engine characteristic features. Procedure according to Claim 1 , The provision (S1) of the sound information (5) by measuring airborne sound and / or structure-borne sound. Procedure according to Claim 1 or 2 , steps a) to c) being carried out within a period of less than 500 ms. Method according to one of the preceding claims, engine-specific information being taken into account when determining (S3) the engine parameter. Procedure according to Claim 4 , the engine-specific information comprising a number of cylinders and / or a number of pole pairs. Procedure according to Claim 4 or 5 , comprising the step: specifying (S0) engine-specific information by a user. Method according to one of the preceding claims, comprising the step of: displaying (S4) the engine parameter. Method according to one of the preceding claims, comprising the step: preparation (S1-2) of the sound information (5) in order to reduce the computing effort of the neural network when determining motor-characteristic features. Computer program product which, when executed by a computer (3), causes the computer (3) to carry out a method for determining an engine parameter according to one of the preceding claims. Device (1) for determining an engine parameter according to one of the Claims 1 - 8th , comprising a) a sensor (2), which is designed to measure a sound and b) a computer (3), which is designed and set up, according to a computer program product Claim 9 perform.

Images