Method for performing an active profiling of a sound emitted by an engine and corresponding profiling system

• US 10,356,539 B2
• Filed: 12/21/2015
• Issued: 07/16/2019
• Est. Priority Date: 12/24/2014
• Status: Active Grant

• Current Assignee: Magneti Marelli S.p.A.
• Original Assignee: Magneti Marelli S.p.A.
• Inventors: Marcelli, Antonio, Armeni, Saverio, Nesci, Walter
• Primary Examiner(s): Flanders, Andrew C
• Application Number:
  US14/977,126
• Publication Number:
  US 20160192095A1
• Time in Patent Office: 1,303 Days
• CPC Class Codes: F01N 1/065 by using an active noise source, e.g. speakers
• CPC Class Codes: F01N 2900/0412 using pre-calibrated maps, tables or charts
• CPC Class Codes: F01N 9/00 Electrical control of exhaust gas treating apparatus monitoring or diagnostic devices for exhaust-gas treatment apparatus F01N11/00; conjoint electrical control of two or more combustion engine functions F02D43/00
• CPC Class Codes: F01N 9/005 using models instead of sensors to determine operating characteristics of exhaust systems, e.g. calculating catalyst temperature instead of measuring it directly
• CPC Class Codes: G10K 15/02 Synthesis of acoustic waves synthesis of speech G10L13/00
• CPC Class Codes: G10K 2210/12822 Exhaust pipes or mufflers
• CPC Class Codes: G10K 2210/3025 Determination of spectrum characteristics, e.g. FFT
• CPC Class Codes: G10K 2210/3032 Harmonics or sub-harmonics
• CPC Class Codes: G10K 2210/51 Improving tonal quality, e.g. mimicking sports cars
• CPC Class Codes: H04R 2499/13 Acoustic transducers and sound field adaptation in vehicles
• CPC Class Codes: H04R 29/00 Monitoring arrangements; Testing arrangements for hearing aids H04R25/30; detection of loudspeaker connection H04R5/04; sound-field adaptation dependent on speaker detection H04S7/308
• CPC Class Codes: Y02T 10/40 Engine management systems

Abstract

A method for performing an active profiling of the sound emitted by a vehicle engine including the steps of injecting second acoustic waves (MW), which are able to combine with the first acoustic waves (EW) in a merge point in said primary propagation path and to generate third resulting acoustic waves (RW); and generating a driving signal for the diffuser by regulating the amplitude of harmonics of the driving signal (SP) as a function of reference acoustic waves and of a workpoint of the engine, calculating equalization coefficients (α₁, . . . , α_M) to be applied to the amplitudes of said harmonics of the driving signal (SP) via a self-tuning procedure, and comparing the reference acoustic waves (TW) with a simulation (SRW) of the resulting waves. The simulation (SRW) is obtained by applying to a simulation of the second acoustic waves a model of the secondary propagation path.

Claims

1. A method for performing an active profiling of the sound emitted by a vehicle engine that issues first acoustic waves (EW) along a primary path of propagation of acoustic waves,

said method comprising the steps of:

injecting second acoustic waves (MW) via an acoustic-wave speaker, into a secondary propagation path that merges in said primary path, wherein said second acoustic waves (MW) combine with said first acoustic waves (EW) in a merge point in said primary propagation path and to generate third resulting acoustic waves (RW);

generating a driving signal (SP) for driving said speaker by regulating the amplitude of harmonics of said driving signal (SP) as a function of reference acoustic waves (TW) of a reference vehicle and of a workpoint (WP) of the engine;

calculating respective equalization coefficients (α

₁, . . . , α

_M) to be applied to the corresponding amplitudes of said harmonics of said driving signal (SP) via a self-tuning procedure, which comprises comparing said reference acoustic waves (TW) of a reference vehicle with a simulation (SRW) of said resulting waves (RW), said simulation (SRW) being obtained by applying a model (H) of the secondary propagation path to a simulation (MSW) of the second injected acoustic waves (MW);

generating basic harmonics (W₁, . . . , W_M) as a function of the current workpoint (WP) of the engine;

obtaining said equalization coefficients (α

₁, . . . , α

_M) as a function of the values of workpoint (WP);

obtaining equalized harmonics (EW₁, . . . , EW_M) as a function of said equalization coefficients (α

₁, . . . , α

_M);

adding together said equalized harmonics (EW₁, . . . , EW_M) to obtain a digital driving wave (DW); and

converting said digital driving wave (DW) into an analog signal corresponding to the driving signal (SP).

2. The method as set forth in claim 1, wherein said operation of generating basic harmonics (W₁, . . . , W_M) comprises generating, as a function of the engine r.p.m. (ER), a pre-set number (M) of basic harmonics (W₁, . . . , W_M) of a sinusoidal signal at the frequency (f) identified by said engine r.p.m. (ER).

3. The method as set forth in claim 1, wherein said self-tuning procedure comprises:

applying an FFT (Fast Fourier Transform) to the reference acoustic waves (TW), which in particular are originated by a reference vehicle, to obtain a reference spectrum (TTW), and extrapolating from said reference spectrum (TTW) amplitudes (TA₁, . . . , TAM) of reference harmonics;

injecting a simulation (MSW) of the second acoustic waves (MW) into the model (H) of the secondary path and obtaining corresponding simulated resulting waves (SRW), in particular for different values of equalization coefficients (α

₁, . . . , α

_M) that model the amplitudes of the second injected waves (MW);

adding the resulting waves (RW) to the reference waves (TW) and applying a transform (FFT) to a resulting sum wave (SW), to obtain a transformed sum wave (TSW);

evaluating and minimizing a cost function (CF) as a function of the amplitudes (TA₁, . . . , TA_M) of reference harmonics and of amplitudes (SA₁, . . . , SA_M) of the transformed sum wave (TSW), in particular as a function of their difference calculated as sum of the errors between the amplitudes (TA₁, . . . , TA_M) of reference harmonics and amplitudes (SA₁, . . . , SA_M) of the transformed sum wave (TSW), supplying as values of equalization coefficients (α

₁, . . . , α

_M) for a given value of workpoint (WP) the values that minimize said cost function (CF).

4. The method as set forth in claim 1, wherein said method further includes an identification procedure for identifying said model (H) of the secondary path, said procedure comprises calculating the parameters of a FIR (Finite Impulse Response) filter.

5. The method as set forth in claim 1, wherein said calculation of the parameters of a FIR filter comprises:

generation and injection, via the speaker, white noise (WN) into the secondary path, keeping the engine turned off, both acquiring the driving signal (SP) associated to the white noise (WN) as input signal (u) of the model (H) and acquiring the corresponding resulting waves (RW) at output from the primary path as output signal (y) of the model (H), repeating said operation of injection a plurality of times with different amplitudes of white noise (WN) at input, to obtain different instances of output signal (y); and

calculation of values of parameters (b_i) of the FIR filter of the model (H) of the secondary path, applying a regression as a function of said plurality of instances of inputs (u) and outputs (y), in particular applying an LMS (Least Mean Square) procedure.

6. The method as set forth in claim 1, wherein said workpoint (WP) comprises a value of engine r.p.m. (ER) and a value of gas-pedal percentage (GP), which are, in particular, issued by an electronic control unit of the vehicle.

7. A system for active profiling of the sound emitted by a vehicle engine that issues first acoustic waves (EW) along a primary path of propagation of acoustic waves, said system comprising a secondary path, which merges in said primary path of propagation, in which an acoustic-wave speaker injects second acoustic waves (MW), which combine with said first acoustic waves (EW) in a merge point in said primary path of propagation and to generate third resulting acoustic waves (RW), an active-sound-profiling unit that generates a driving signal (SP) for said speaker by regulating the amplitude of harmonics of said driving signal (SP) as a function of reference acoustic waves (TW) and of a workpoint (WP) of the engine, said system acting to carry out on one or more processors including:

calculating respective equalization coefficients (α

₁, . . . , α

_M) to be applied to the corresponding amplitudes of said harmonics of said driving signal (SP) via a self-tuning procedure, which comprises comparing said reference acoustic waves (TW) of a reference vehicle with a simulation (SRW) of said resulting waves (RW), said simulation (SRW) being obtained by applying a model (H) of the secondary propagation path to a simulation (MSW) of the second injected acoustic waves (MW);

generating basic harmonics (W₁, . . . , W_M) as a function of the current workpoint (WP) of the engine;

obtaining said equalization coefficients (α

₁, . . . , α

_M) as a function of the values of workpoint (WP);

obtaining equalized harmonics (EW₁, . . . , EW_M) as a function of said equalization coefficients (α

₁, . . . , α

_M);

adding together said equalized harmonics (EW₁, . . . , EW_M) to obtain a digital driving wave (DW); and

converting said digital driving wave (DW) into an analog signal corresponding to the driving signal (SP).

8. The system as set forth in claim 7, wherein said active-profiling unit comprises a harmonic-generation module, configured for generating, as a function of the engine r.p.m. (ER), a pre-set number (M) of basic harmonics (W₁, . . . , W_M) of a sinusoidal signal at a frequency (f) identified by said engine r.p.m. (ER), and a plurality of look-up tables, each look-up table, for a specific corresponding harmonic (W_i), associating to a value of workpoint (WP) a respective equalization coefficient (α

_i).

9. The system as set forth in claim 7, wherein said engine is an internal-combustion engine or an electric engine.