CN117275520B - Voice recognition fault noise detection method, system and device and medium - Google Patents

Abstract

The invention discloses a method, a system, a device and a medium for detecting voice recognition fault noise, which relate to the technical field of fault diagnosis of electric fans or motors, wherein the detection method comprises the following steps: acquiring effective time length between occurrence of preset noise; judging two adjacent effective time length CT _n And CT _n+1 Is of the size of (2): if the phase difference is within the preset threshold value, the count CNT is accumulated ₀ The method comprises the steps of carrying out a first treatment on the surface of the If the phase difference exceeds a preset threshold value, the throwing duration CT is given out _n And count the number of throws DCT _x The method comprises the steps of carrying out a first treatment on the surface of the At the accumulated count CNT ₀ In the process, the thrown quantity DCT _x When the set number is reached, the CNT is counted ₀ Zero clearing when the count CNT ₀ And when the accumulation reaches the set upper limit, reporting the fault. The invention adopts voice recognition fault noise, recognizes preset friction noise according to the running rotating speed of the fan or the motor, carries out fault alarm, and timely detects the abnormal condition of the fault noise of the motor or the fan blade.

Description

A detection method, system, device and medium for speech recognition fault noise

Technical field

The present invention relates to the technical field of fault diagnosis of electric fans or motors, in particular to a detection method, system, device and medium for speech recognition fault noise.

Background technique

In the daily use of the fan, if the fan is operated at high speed for a long time, the motor will drag the load to conduct vibration, causing the fan blades to shake, which may cause the fan blades to twist and deform, causing friction between the fan blades and the casing, or causing the motor to Or bearing aging. Fan blade twisting failure, motor failure, bearing aging failure, etc. will produce mechanical failure noise. Traditional detection technology uses manual identification of mechanical failure noise and cannot detect abnormal noise and motor or fan blade failure in a timely manner. Failure to shut down in time for maintenance leads to related The fault situation worsens, and in severe cases, it causes the fan blades to twist, which greatly affects the user experience.

Contents of the invention

Aiming at the detection problem of traditional man-made mechanical fault noise, the present invention provides a detection method, system, device and medium for voice recognition of fault noise. By using voice recognition of mechanical fault noise, the abnormal situation of fault noise of the motor or fan blade can be detected in a timely manner. .

In order to achieve the above purpose, the present invention adopts the following technical solution: a speech recognition fault noise detection method, including the following steps:

Get the effective time between the occurrence of the preset noise;

Determine the size of two adjacent valid durations CT _n and CT _n+1 : if the difference is within the preset threshold, then the cumulative count CNT ₀ ; if the difference exceeds the preset threshold, throw out the duration CT _n , and count the thrown Quantity DCT _x ;

During the cumulative counting process of CNT ₀ , when the thrown quantity DCT _x reaches the set number, the counting CNT ₀ is cleared. When the counting CNT ₀ reaches the set upper limit, a fault is reported.

As a further improvement of the present invention, the obtaining the effective time period between the occurrence of preset noise includes:

Obtain the feedback information of the occurrence of the preset noise, analyze the time point of occurrence according to the feedback information, and record the time T _n ;

The time interval between the time T _n and the time T _n+1 when the preset noise occurs twice is recorded as the time length, and the time length that meets the conditions is marked as the effective time length CT _n .

As a further improvement of the present invention, obtaining feedback information on the occurrence of preset noise includes:

After powering on and entering the working state, the voice module starts to monitor the preset fault noise;

When preset noise occurs, the voice module outputs feedback information to the controller;

The controller obtains the feedback information and analyzes the time point at which it occurs, where the time point is carried in the feedback information or determined by the current time point when the controller receives the feedback information and starts a timer to determine the time point of the next feedback information. .

As a further improvement of the present invention, the marking of qualified durations as valid durations CT _n includes:

Obtain the current rotation speed at the time T _n or T _n+1 when the preset noise occurs;

Determine whether the time between time T _n and time T _n+1 is reasonable based on the current speed:

If it meets the preset effective value of the current rotation speed, determine the time period as the effective time period CT _n ;

If it does not meet the preset effective value of the current speed, throw out the time T _n and wait for the next time T _n+2 when the preset noise occurs, and continue the judgment.

As a further improvement of the present invention, the determination of the sizes of two adjacent valid durations CT _n and CT _n+1 includes:

Get the effective duration CT _n and CT _n+1 , and calculate the difference between the two valid durations |CT _n -CT _n+1 |;

Determine whether the difference |CT _n -CT _n+1 | exceeds the preset threshold dt:

If |CT _n -CT _n+1 |≤dt, then update count CNT ₀ =CNT ₀ +1;

If |CT _n -CT _n+1 |>dt, the throw duration is CT _n and the number of throws is updated DCT _x = DCT _x +1.

As a further improvement of the present invention, the detection method also includes:

In the process of accumulating count CNT ₀ , the duration CT _n is thrown and the quantity thrown DCT _x = DCT _x +1 is updated;

When the updated DCT _x reaches the set number, update CNT ₀ =0, where the set number is less than the set upper limit of the count CNT ₀ ;

Update DCT _x =0 and continue to wait for the next count.

As a further improvement of the present invention, the detection method also includes:

During the process of accumulating CNT ₀ , when the accumulated count of CNT ₀ reaches the set upper limit, a fault is reported, and at the same time, the number of throws is cleared, and DCT _x =0 is updated.

On the other hand, the present invention adopts the following technical solution: a speech recognition fault noise detection system, including:

The voice module is used to identify whether the fault noise belongs to the preset noise and output feedback information to the controller;

The controller determines the time point when the preset noise occurs based on the feedback information, and calculates the duration of the two occurrences of the preset noise;

The first judgment module is used to judge whether the duration meets the conditions and mark the valid duration CT _n ;

The second judgment module is used to judge the size of two adjacent valid durations CT _n and CT _n+1 and output the judgment result;

The counting module performs cumulative calculation on the count CNT ₀ and the thrown quantity DCT _x according to the judgment result;

The controller sets an upper limit based on the count CNT _0. When the set upper limit is reached, a fault report is triggered.

On the other hand, the present invention adopts the following technical solution: a speech recognition fault noise detection device, including a memory and a processor. The processor calls the control program stored in the memory to execute the above-mentioned one. Detection method for speech recognition failure noise.

On the other hand, the present invention adopts the following technical solution: a computer-readable storage medium, the computer-readable storage medium stores a computer program, and when executed by a processor, the computer program causes the processor to execute the above A detection method for speech recognition fault noise.

Compared with the existing technology, the present invention has the following beneficial effects:

The invention uses voice to identify fault noise, identifies preset friction noise according to the current running speed of the fan or motor, and issues a fault alarm to promptly detect abnormal fault noise in the motor or fan blades. It can effectively detect the distortion and deformation of the fan blades when the fan is running at high speed, causing friction between the fan blades and the casing, or the aging of the motor or bearings caused by long-term use, helping users find problems in time and shut down for maintenance.

Description of the drawings

In order to explain the technical solution more clearly, the drawings needed to be used in the implementation will be briefly introduced below. Obviously, the drawings in the following description are only some implementations of the present invention. For those of ordinary skill in the art, It is said that other drawings can be obtained based on these drawings without exerting creative labor.

Figure 1 is a schematic flow chart of the detection method of speech recognition fault noise.

Detailed ways

In order to have a clear and complete understanding of the technical solution, the present invention will be further described with reference to the embodiments and drawings. Obviously, the recorded embodiments are only some of the embodiments of the present invention, and those skilled in the art will not make any creative efforts without doing so. All other embodiments obtained fall within the scope of protection of the present invention.

It should be understood that, when used in this specification and the appended claims, the terms "comprises" and "comprises" indicate the presence of described features, integers, steps, operations, elements and/or components but do not exclude the presence of one or The presence or addition of multiple other features, integers, steps, operations, elements, components and/or collections thereof.

It should also be understood that the terminology used in the description of the invention is for the purpose of describing particular embodiments only and is not intended to be limiting of the invention. As used in this specification and the appended claims, the singular forms "a", "an" and "the" are intended to include the plural forms unless the context clearly dictates otherwise.

It will be further understood that the term "and/or" as used in the specification and the appended claims refers to and includes any and all possible combinations of one or more of the associated listed items. .

Embodiments of the present invention provide a voice recognition fault noise detection method, using a fan with a voice module function to realize timely detection of fan blade distortion faults, motor faults, bearing aging faults, etc. that will generate mechanical fault noise. . The voice module is used to identify mechanical fault noise, perform shutdown maintenance and display and report faults. Compared with traditional human-made mechanical fault noise recognition, it can detect abnormal noise and faults in motors or fan blades in a timely manner, improving the user experience.

This embodiment provides a speech recognition fault noise detection method, as shown in Figure 1, including the following steps:

Get the effective time between the occurrence of the preset noise;

Determine the size of two adjacent valid durations CT _n and CT _n+1 : if the difference is within the preset threshold, then the cumulative count CNT ₀ ; if the difference exceeds the preset threshold, throw out the duration CT _n , and count the thrown Quantity DCT _x ;

During the cumulative counting process of CNT ₀ , when the thrown quantity DCT _x reaches the set number, the counting CNT ₀ is cleared. When the counting CNT ₀ reaches the set upper limit, a fault is reported.

Through the above steps, by judging the size and frequency of occurrence of two adjacent valid periods, it is then judged whether the preset noise is caused by a fault. When the fan blades and motor rotate, when friction occurs, corresponding noise will appear periodically. Fault noise, this embodiment effectively determines the time between time points, thereby realizing fault reporting. It is suitable for scenarios such as fan blade friction to help users find and solve problems in time. It should be noted that the present invention does not place any form of restrictions on the technical means of recognizing noise in speech recognition. Technicians can set the recognition of noise by themselves according to actual needs, as long as the fault noise can be recognized by speech and output feedback information when it occurs. Can.

Those skilled in the art can understand that the preset noise in this embodiment is preset. When the speech module recognizes that the preset noise occurs, it can output feedback information that the preset noise occurs and identify the speech in which the preset noise occurs. The module needs to have good noise detection capabilities to ensure accurate identification of fault noise.

This embodiment can use a fan with a voice module function to promptly detect problems such as fan blade distortion, motor failure, bearing aging failure, etc. that may cause mechanical failure and abnormal noise without increasing the cost of mechanical components. This embodiment can also adjust parameters such as CT _n and CNT ₀ according to the actual situation of the entire machine to ensure accurate identification and reporting of faults.

For better understanding, in an optional embodiment, obtaining the effective time period between occurrences of preset noise includes:

Obtain feedback information that the preset noise occurs, analyze the time point of occurrence according to the feedback information, and record the time T _n . In some embodiments, after the fan or motor is powered on and enters the working state, the fan starts to rotate, and the voice module is turned on to monitor the preset fault noise. As explained above, the voice module can use existing technology to identify the fault noise. , by presetting the fault noise (that is, the preset noise in this embodiment), the voice module turns on monitoring to detect whether the preset noise occurs when the fan or mechanical component rotates.

When a preset noise occurs and is recognized by the voice module, the voice module outputs feedback information to the controller. After obtaining the feedback information, the controller analyzes the time point of occurrence and records the time T _n , where the time point can be the feedback output by the voice module. carried in the information, can also be determined by the current time point when the controller receives the feedback information, and at the same time, the timer is started to determine the time point of the next feedback information, and the time T _n+1 is recorded.

The time interval between the time T _n and the time T _n+1 when the preset noise occurs twice is recorded as the time length, that is, T _n+1 -T _n , and the time length that meets the conditions is marked as the effective time length CT _n . In some embodiments, marking a qualifying duration as a valid duration CT _n includes:

Obtain the current rotation speed at the time T _n or T _n+1 when the preset noise occurs; those skilled in the art can understand that different fans or motors have different rotation speeds at different gears, and correspondingly, the noise at different rotation speeds The performance is also different. Therefore, technicians can set the acquisition speed state by themselves according to actual usage requirements. It can be the current speed at time T _n or T _n+1 , or it can be a certain time from time T _n to T _n+1. The rotational speed or average rotational speed at a moment only needs to be pre-set according to the actual usage scenario, mechanical component structure, etc. and the reasonable duration conditions for the occurrence of noise at the corresponding rotation speed (this embodiment uses a preset effective value to judge whether the duration is reasonable).

Determine whether the time between time T _n and time T _n+1 is reasonable based on the current speed:

If it meets the preset effective value of the current rotation speed, determine the time period as the effective time period CT _n ;

If it does not meet the preset effective value of the current speed, throw out the time T _n and wait for the next time T _n+2 when the preset noise occurs, and continue the judgment. That is to say, when the preset noise occurs again, it is recorded as T _n+2 , and the time between time T _n+1 and time T _n+2 is continued to be judged based on the rotation speed at time T _n+1 or T _n+2 Whether it is reasonable or not, it will be executed in a loop.

In an optional embodiment, determining the sizes of two adjacent valid durations CT _n and CT _n+1 includes:

Get the effective duration CT _n and CT _n+1 , and calculate the difference between the two valid durations |CT _n -CT _n+1 |;

Determine whether the difference value |CT _n -CT _n+1 | exceeds the preset threshold dt, that is, whether the difference between the two valid durations CT _n and CT _n+1 is large. In some implementations, the judgment formula can also be CT _n -dt≤CT _n+1 ≤CT _n +dt. If the difference is large, it means that the two valid durations indicate that the identified fault noise is not closely related. If the difference is small, the number of valid durations will be further counted. When When the set upper limit is reached, it means that the preset noise of the fan or motor occurs continuously or periodically and intensively, which can be determined as an abnormal situation and be shut down for maintenance;

specifically:

If |CT _n -CT _n+1 |≤dt, then update count CNT ₀ =CNT ₀ +1, initial count CNT ₀ =0;

If |CT _n -CT _n+1 |>dt, the throwing duration is CT _n and the number of throws is updated DCT _x = DCT _x +1, and the initial number of throws is DCT _x =0.

In an optional embodiment, during the cumulative counting process of CNT ₀ , the duration CT _n is thrown and the thrown quantity DCT _x = DCT _x +1 is updated; when the updated DCT _x reaches the set number, CNT ₀ is updated. =0, where the set number is less than the set upper limit of the count CNT ₀ , update DCT _x =0, and continue to wait for the next count

When the accumulated count of CNT ₀ reaches the set upper limit, a fault is reported, the number of throws is cleared, and DCT _x =0 is updated.

The invention uses voice recognition to identify fault noise, identifies the preset friction noise based on the current running speed of the fan or motor, and performs a fault alarm to promptly detect abnormal fault noise in the motor or fan blades, helping users find problems in time and enable shutdown and maintenance. .

Through the description of the above embodiments, those skilled in the art can clearly understand that the method according to the above embodiments can be implemented by means of software plus the necessary general hardware platform. Of course, it can also be implemented by hardware, but in many cases the former is Better implementation. Based on this understanding, the technical solution of the present application can be embodied in the form of a software product in essence or that contributes to the existing technology. The computer software product is stored in a storage medium (such as a read-only memory (Read-Only Memory)). Memory (ROM for short), Random Access Memory (Random Access Memory (RAM for short), magnetic disk, optical disk), including a number of instructions to make a terminal device (can be a mobile phone, computer, server, or network device, etc.) Perform the methods described in various embodiments of this application.

An embodiment of the present invention also provides a speech recognition fault noise detection system. The detection system is used to implement the above embodiments and preferred implementations. What has already been explained will not be described again. As used below, the term "module" may be a combination of software and/or hardware that implements a predetermined function. Although the apparatus described in the following embodiments is preferably implemented in software, implementation in hardware, or a combination of software and hardware, is also possible and contemplated. This embodiment provides a speech recognition fault noise detection system, including:

The voice module is used to identify whether the fault noise belongs to the preset noise and output feedback information to the controller;

The controller determines the time point when the preset noise occurs based on the feedback information, and calculates the duration of the two occurrences of the preset noise;

The first judgment module is used to judge whether the duration meets the conditions and mark the valid duration CT _n ;

The second judgment module is used to judge the size of two adjacent valid durations CT _n and CT _n+1 and output the judgment result;

The counting module performs cumulative calculation on the count CNT ₀ and the thrown quantity DCT _x according to the judgment result;

The controller sets an upper limit based on the count CNT _0. When the set upper limit is reached, a fault report is triggered. Specifically:

Get the effective time between the occurrence of the preset noise;

Determine the size of two adjacent valid durations CT _n and CT _n+1 : if the difference is within the preset threshold, then the cumulative count CNT ₀ ; if the difference exceeds the preset threshold, throw out the duration CT _n , and count the thrown Quantity DCT _x ;

During the cumulative counting process of CNT ₀ , when the thrown quantity DCT _x reaches the set number, the counting CNT ₀ is cleared. When the counting CNT ₀ reaches the set upper limit, a fault is reported.

It should be noted that each of the above modules can be implemented through software or hardware. For the latter, it can be implemented in the following ways, but is not limited to this: the above modules are all located in the same processor; or the above modules can be implemented in any combination. The forms are located in different processors.

On the other hand, embodiments of the present invention also provide a speech recognition fault noise detection device, including a memory and a processor. The processor calls a control program stored in the memory to execute one or more of the above. A method for detecting speech recognition fault noise described in one embodiment.

Optionally, in this embodiment, the above-mentioned processor may be configured to perform the following steps through the control program:

Get the effective time between the occurrence of the preset noise;

Determine the size of two adjacent valid durations CT _n and CT _n+1 : if the difference is within the preset threshold, then the cumulative count CNT ₀ ; if the difference exceeds the preset threshold, throw out the duration CT _n , and count the thrown Quantity DCT _x ;

During the cumulative counting process of CNT ₀ , when the thrown quantity DCT _x reaches the set number, the counting CNT ₀ is cleared. When the counting CNT ₀ reaches the set upper limit, a fault is reported.

Optionally, for specific examples in this embodiment, reference can be made to the examples described in the above-mentioned embodiments and optional implementations, and details will not be described again in this embodiment.

Embodiments of the present invention also provide a computer-readable storage medium that stores a computer program, wherein the computer program is configured to execute a method as described in one or more of the above embodiments when running. Detection method for speech recognition failure noise.

Optionally, in this embodiment, the above-mentioned storage medium may be configured to store a computer program for performing the following steps:

Get the effective time between the occurrence of the preset noise;

Determine the size of two adjacent valid durations CT _n and CT _n+1 : if the difference is within the preset threshold, then the cumulative count CNT ₀ ; if the difference exceeds the preset threshold, throw out the duration CT _n , and count the thrown Quantity DCT _x ;

During the cumulative counting process of CNT ₀ , when the thrown quantity DCT _x reaches the set number, the counting CNT ₀ is cleared. When the counting CNT ₀ reaches the set upper limit, a fault is reported.

Optionally, in this embodiment, the above storage medium may include but is not limited to: U disk, read-only memory ROM, random access memory RAM, mobile hard disk, magnetic disk or optical disk and other various media that can store computer programs.

The steps of a method or algorithm described in connection with the embodiments disclosed herein may be embodied directly in hardware, in a software module executed by a processor, or in a combination of both. Software modules may reside in RAM memory, Flash memory, ROM memory, EPROM memory, EEPROM memory, registers, hard disk, removable disk, CD-ROM, or any other form of storage medium known in the art. An example storage medium is coupled to the processor such that the processor can read and write information from/to the storage medium. In the alternative, the storage medium may be integrated into the processor. The processor and storage media can reside in an ASIC. The ASIC can reside in the user terminal. In the alternative, the processor and storage medium may reside as discrete components in the user terminal.

The above disclosure is only one or more preferred embodiments of the present invention, which are used to help understand the inventive concept of the technical solution, and are not intended to limit the present invention in other forms. Those skilled in the art will make other equivalents based on the limited features of the present invention. Or the replacement scheme of conventional means still falls within the scope of the present invention.

Claims (9)

1. The method for detecting the voice recognition fault noise is characterized by comprising the following steps:

acquiring effective duration CT between occurrence of preset noise _n And CT _n+1 And calculates the difference |CT between the two effective durations _n -CT _n+1 |；

Judging the difference value |CT _n -CT _n+1 Whether or not the i exceeds a preset threshold dt:

if |CT _n -CT _n+1 If the I is less than or equal to dt, the count CNT is updated ₀ =CNT ₀ +1；

If |CT _n -CT _n+1 I > dt, throw-out duration CT _n And update the thrown quantity DCT _x = DCT _x +1；

At the accumulated count CNT ₀ In the process, the thrown quantity DCT _x When the set number is reached, the CNT is counted ₀ Zero clearing when the count CNT ₀ And when the accumulation reaches the set upper limit, reporting the fault.

2. The method for detecting speech recognition failure noise according to claim 1, wherein the obtaining the effective duration between occurrence of the preset noise comprises:

acquiring feedback information of occurrence of preset noise, analyzing occurrence time point according to the feedback information, and recording time T _n ；

Time T when preset noise is generated twice _n And time T _n+1 The time interval between the two is marked as the duration, and the duration meeting the condition is marked as the effective duration CT _n 。

3. The method for detecting speech recognition failure noise according to claim 2, wherein the obtaining feedback information of occurrence of the preset noise comprises:

after the power-on enters a working state, the voice module is started to monitor preset fault noise;

when the preset noise occurs, the voice module outputs feedback information to the controller;

the controller obtains feedback information and analyzes the time point of occurrence, wherein the time point is determined by the current time point carried in the feedback information or when the controller receives the feedback information, and a timer is started to determine the time point of the next feedback information.

4. The method for detecting speech recognition failure noise according to claim 2, wherein the time length mark meeting the condition is an effective time length CT _n Comprising the following steps:

acquiring time T when preset noise occurs _n Or T _n+1 A current rotational speed;

judging the time T according to the current rotating speed _n To time T _n+1 Whether the duration between them is reasonable:

if the preset effective value of the current rotating speed is met, determining the time length as the effective time length CT _n ；

If the preset effective value of the current rotating speed is not met, the moment T is thrown out _n Waiting for the next time T when the preset noise occurs _n+2 And (5) continuing to judge.

5. The method for detecting speech recognition failure noise according to claim 1, wherein the method further comprises:

at the accumulated count CNT ₀ In the process, the throwing time length CT _n And updates the thrown quantity DCT _x = DCT _x +1；

DCT when updated _x When the set number is reached, the CNT is updated ₀ =0, wherein the set number is less than the count CNT ₀ Is set to an upper limit;

updating DCT _x =0, and continues to wait for the next count.

6. The method for detecting speech recognition failure noise according to claim 5, further comprising:

at the accumulation CNT ₀ In the process, when CNT is counted ₀ When the accumulation reaches the set upper limit, reporting faults, clearing the thrown quantity, and updating DCT _x =0。

7. A system for detecting speech recognition failure noise, comprising:

the voice module is used for identifying whether the fault noise belongs to preset noise or not and outputting feedback information to the controller;

the controller determines a time point when the preset noise occurs according to the feedback information, and calculates the duration of the preset noise occurring twice;

a first judging module for judging whether the time length meets the condition and marking the effective time length CT _n ；

A second judging module for judging two adjacent effective time length CT _n And CT _n+1 And outputting a judgment result;

the counting module counts the CNT according to the judging result ₀ And number of throws DCT _x Performing accumulation calculation;

the controller counts CNTs according to the count ₀ And setting an upper limit, and triggering fault reporting when the upper limit is reached.

8. The utility model provides a detection device of speech recognition trouble noise, includes memory and treater, its characterized in that: the processor executes a method for detecting speech recognition failure noise according to any one of claims 1 to 6 by calling a control program stored in the memory.

9. A computer readable storage medium, characterized in that the computer readable storage medium stores a computer program which, when executed by a processor, causes the processor to perform a method of detecting speech recognition failure noise according to any of claims 1-6.