A method and device for monitoring abnormal sound of an induced draft fan in operation, and an electronic device

Abstract

The application provides a kind of induced draft fan operation abnormal sound monitoring method, device and electronic equipment, obtain the sound emitted in the operation process of induced draft fan, the sound includes: each sound signal collected by each sound pickup device in multiple sound pickup devices respectively;The non-average algorithm preprocessing is carried out to each sound signal, and each sound signal is integrated into one monitoring signal of each sound signal;The monitoring signal is processed to obtain a plurality of abnormal sound identification features;Utilize induced draft fan abnormal sound monitoring model to process a plurality of the abnormal sound identification features, obtain abnormal sound identification result.This application can greatly extend the monitoring time of induced draft fan, realize uninterrupted monitoring in full period, can continuously monitor the running state of various components of induced draft fan, greatly reduce the use cost of induced draft fan diagnosis system and the downtime of induced draft fan failure.

Description

Technical Field

[0001] This invention relates to the field of induced draft fan monitoring technology, and more specifically, to a method, device, and electronic equipment for monitoring abnormal noises during induced draft fan operation. Background Technology

[0002] The induced draft fan is one of the important auxiliary equipment in a thermal power plant boiler. Its working principle is to draw hot air from the boiler's combustion chamber, creating negative pressure within the boiler. Finally, the hot air is discharged into the atmosphere after passing through dust removal, desulfurization, and denitrification processes. Traditionally, induced draft fan monitoring relies on experienced personnel conducting periodic inspections to diagnose malfunctions. However, human monitoring is limited to periodic checks and cannot be performed long-term; furthermore, the accuracy of manual monitoring is relatively low. Summary of the Invention

[0003] To address the existing technical problems, embodiments of the present invention provide a method, device, and electronic equipment for monitoring abnormal noise during the operation of an induced draft fan.

[0004] In a first aspect, embodiments of the present invention provide a method for monitoring abnormal noise during the operation of an induced draft fan, comprising:

[0005] Acquire the sound emitted during the operation of the induced draft fan, the sound including: the sound signals collected by each of the multiple sound pickup devices in the multi-channel sound pickup device;

[0006] The audio signals are preprocessed using a non-average algorithm, and then combined into a single monitoring signal.

[0007] The monitoring signal is processed to obtain multiple abnormal noise identification features;

[0008] The abnormal noise monitoring model of the induced draft fan is used to process multiple abnormal noise identification features to obtain abnormal noise identification results.

[0009] Secondly, embodiments of the present invention also provide a device for monitoring abnormal noise during the operation of an induced draft fan, comprising:

[0010] The acquisition module acquires the sound emitted during the operation of the induced draft fan, including the sound signals collected by each of the multiple sound pickup devices.

[0011] The algorithm module performs non-average algorithm preprocessing on each audio signal and synthesizes the audio signals into a single monitoring signal.

[0012] The processing module processes the monitoring signal to obtain multiple abnormal noise identification features;

[0013] The results module uses the induced draft fan noise monitoring model to process multiple noise identification features and obtain noise identification results.

[0014] Thirdly, embodiments of the present invention provide an electronic device, including a bus, a transceiver, a memory, a processor, and a computer program stored in the memory and executable on the processor. The transceiver, the memory, and the processor are connected via the bus. When the computer program is executed by the processor, it implements the steps in the method for monitoring abnormal noise during operation of an induced draft fan as described in any of the first aspects above.

[0015] Fourthly, embodiments of the present invention also provide a computer-readable storage medium having a computer program stored thereon, wherein the computer program, when executed by a processor, implements the steps in the method for monitoring abnormal noise during operation of an induced draft fan as described in any of the first aspects.

[0016] In the solutions provided in the first to fourth aspects of this application, multiple sound signals emitted by the induced draft fan are acquired through a multi-channel sound pickup device. These multiple sound signals are merged into a single monitoring signal under the pre-processing of a non-average algorithm. The monitoring signal is then processed to extract multiple abnormal noise identification features. These abnormal noise identification features are input into the induced draft fan abnormal noise monitoring model to obtain the abnormal noise identification result of whether the induced draft fan has produced abnormal noise. Compared with the method of manually monitoring the induced draft fan periodically in related technologies, this method greatly extends the monitoring time of the induced draft fan, enabling uninterrupted monitoring of the induced draft fan at all times. It can continuously monitor the operating status of various components of the induced draft fan and promptly maintain and repair the induced draft fan when a fault is determined based on the abnormal noise identification result. This reduces the usage cost of the induced draft fan diagnostic system and the downtime of the induced draft fan due to faults.

[0017] To make the above-mentioned objects, features and advantages of the present invention more apparent and understandable, preferred embodiments are described below in detail with reference to the accompanying drawings. Attached Figure Description

[0018] To more clearly illustrate the technical solutions in the embodiments of the present invention or the background art, the accompanying drawings used in the embodiments of the present invention or the background art will be described below.

[0019] Figure 1 A flowchart illustrating a method for monitoring abnormal noise during the operation of an induced draft fan, provided in Embodiment 1 of the present invention, is shown.

[0020] Figure 2 This shows a schematic diagram of the structure of a fan operation noise monitoring device provided in Embodiment 2 of the present invention;

[0021] Figure 3The diagram shows a schematic of an electronic device for performing a method for monitoring abnormal noise during the operation of an induced draft fan, as provided in Embodiment 3 of the present invention. Detailed Implementation

[0022] In the description of this application, it should be understood that the terms "center", "longitudinal", "lateral", "length", "width", "thickness", "upper", "lower", "front", "rear", "left", "right", "vertical", "horizontal", "top", "bottom", "inner", "outer", "clockwise", "counterclockwise", etc., indicate the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings. They are used only for the convenience of describing this application and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation. Therefore, they should not be construed as limitations on this application.

[0023] Furthermore, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of technical features indicated. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more of that feature. In the description of this application, "multiple" means two or more, unless otherwise explicitly specified.

[0024] In this application, unless otherwise expressly specified and limited, the terms "installation," "connection," "linking," and "fixing," etc., should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral connection; they can refer to a mechanical connection or an electrical connection; they can refer to a direct connection or an indirect connection through an intermediate medium; and they can refer to the internal connection between two components. Those skilled in the art can understand the specific meaning of the above terms in this application according to the specific circumstances.

[0025] The induced draft fan is one of the important auxiliary equipment in a thermal power plant boiler. Its working principle is to draw hot air from the boiler's combustion chamber, creating negative pressure within the boiler. Finally, the hot air is discharged into the atmosphere after passing through dust removal, desulfurization, and denitrification processes. Traditionally, induced draft fan monitoring relies on experienced personnel conducting periodic inspections to diagnose malfunctions. However, human monitoring is limited to periodic checks and cannot be performed long-term; furthermore, the accuracy of manual monitoring is relatively low.

[0026] Based on this, the following embodiments of this application propose a method, device, and electronic equipment for monitoring abnormal noise during the operation of an induced draft fan. Multiple sound signals emitted by the induced draft fan are acquired through a multi-channel pickup device. These signals are then merged into a single monitoring signal through pre-processing using a non-average algorithm. The monitoring signal is further processed to extract multiple abnormal noise identification features, which are then input into the induced draft fan abnormal noise monitoring model to obtain an abnormal noise identification result indicating whether the induced draft fan is making abnormal noise. This significantly extends the monitoring time of the induced draft fan, enabling continuous monitoring of the fan at all times. It allows for continuous monitoring of the operating status of various components of the induced draft fan and timely maintenance and repair when a fault is determined based on the abnormal noise identification result. This reduces the operating cost of the induced draft fan diagnostic system and the downtime of the induced draft fan due to faults.

[0027] To make the principles behind the above-mentioned technical effects of this application more apparent and understandable, this application will be further described in detail below with reference to the accompanying drawings and embodiments.

[0028] Example 1

[0029] The execution entity of the abnormal noise monitoring method for induced draft fan operation proposed in this embodiment is a server.

[0030] See Figure 1 The schematic diagram shown illustrates a method for monitoring abnormal noise during the operation of an induced draft fan. This embodiment proposes a method for monitoring abnormal noise during the operation of an induced draft fan, which includes the following steps:

[0031] Step 100: Acquire the sound emitted during the operation of the induced draft fan, the sound including: the sound signals collected by each of the multiple sound pickup devices in the multi-channel sound pickup device.

[0032] In step 100 above, to monitor the operating status of the induced draft fan, multiple sound pickup devices need to be installed near the fan, with a certain distance maintained between each device and the fan. The distances between different sound pickup devices and the fan can be the same or different, but each device must be installed near a different component of the fan. This allows for the collection of sound signals emitted by different components within the fan.

[0033] Multiple sound pickup devices are connected to the aforementioned server, enabling the server to acquire the sound emitted during the operation of the induced draft fan.

[0034] Specifically, the aforementioned sound pickup devices include, but are not limited to, microphones, sound sensors, and acoustic devices; the sound signal collected is optimal when the sound pickup device is facing the direction of the exhaust fan. It should be noted that the aforementioned sound pickup device can be a single device or multiple devices arranged at equal intervals near the exhaust fan; if the sound pickup device is a single device, it means that only a single component of the exhaust fan needs to be monitored; if the sound pickup device is multiple devices, it means that the entire exhaust fan or some components need to be monitored, and the specific settings need to be configured according to different requirements. Therefore, this application embodiment does not limit the number of sound pickup devices.

[0035] When the induced draft fan is running, its internal components may make noise; these internal components include, but are not limited to: impeller, main shaft, exhaust wheel and coupling.

[0036] Step 101: Perform non-average algorithm preprocessing on the audio signals and synthesize the audio signals into a single monitoring signal.

[0037] In step 101 above, in order to merge multiple audio signals into one monitoring signal, the multiple audio signals need to be mapped into multiple mapped signals, and the multiple mapped signals are weighted and combined into one monitoring signal of each audio signal.

[0038] Among them, mapping conversion refers to using a pre-set mapping function to perform mapping conversion on multiple audio signals.

[0039] In one implementation, the mapping function is a monotonic function, for example, the mapping function is f(x). If the collected sound signal is x0, then the converted monitoring signal is y0, and y0 = f(x0).

[0040] For example, this mapping function can specifically be a logarithmic function. This mapping function can be a logarithmic function: f(x) = m × ln(1 + n) x ); m and n are coefficients, and m > 0, n > 1; for example, m = 0.23, n = 71, that is, the mapping function is f(x) = 0.23 × ln(1 + 71) x Where x is the quantity that needs to be mapped and transformed, that is, each audio signal in the multi-channel audio signal.

[0041] In order to perform weighted processing on the above multi-path mapped signals and synthesize the multi-path mapped signals into a single monitoring signal of each audio signal, the following steps (1) to (4) can be performed:

[0042] (1) Calculate the time-domain energy of each mapped signal in the multi-path mapped signal;

[0043] (2) Sort the multiple mapped signals according to the magnitude of their time-domain energy;

[0044] (3) Assign a first weight to the mapping signal with the highest time-domain energy ranking among the multiple mapping signals, and assign a second weight to the other mapping signals among the multiple mapping signals excluding the mapping signal with the highest time-domain energy ranking; wherein the first weight is greater than the second weight;

[0045] (4) The assigned mapping signals are weighted to obtain one monitoring signal of each audio signal.

[0046] In sub-step (1) above, the mapped signal is known. Substituting each mapped signal into the following time-domain energy formula, the time-domain energy corresponding to different mapped signals can be calculated:

[0047]

[0048] Where TE is the time-domain energy, s is the mapped signal, end is the end time of the mapped signal, beg is the start time of the mapped signal, and n is the length of the mapped signal.

[0049] In the above sub-step (2), after step (1), the time-domain energy corresponding to each mapped signal is calculated, and then all the mapped signals are sorted according to the magnitude of the time-domain energy.

[0050] In sub-step (3) above, the "first in time-domain energy sorting" refers to the largest time-domain energy value. The time-domain energies are sorted from largest to smallest and weighted accordingly. The mapping signal corresponding to the first time-domain energy value must be treated differently from other mapping signals. For example, the weight of the mapping signal with the first time-domain energy value can be set to 0.8, and the remaining mapping signals share the second weight of 0.2. Note that the weights of 0.8 and 0.2 are merely examples; the first and second weights can be set to other values, as long as the first weight is greater than the second weight. Therefore, this embodiment does not limit the specific values ​​of the first and second weights. At this point, all mapping signals have been weighted, and the weighted multi-channel mapping signals are combined and used as a single monitoring signal.

[0051] Step 102: Process the monitoring signal to obtain multiple abnormal noise identification features.

[0052] In this embodiment, step 100 is to acquire all sounds near the induced draft fan, and step 101 is to process all the acquired sounds and finally merge them into a single monitoring signal for further processing in subsequent steps. Step 102 is to extract a signal from the single monitoring signal that can determine whether the induced draft fan has malfunctioned; this signal is the abnormal noise identification feature of the induced draft fan. Therefore, to extract the abnormal noise identification feature from the monitoring signal, the following steps (A) to (C) can be performed:

[0053] (A) The monitoring signal is processed based on a filtering algorithm to generate a filtered signal;

[0054] (B) Perform Fourier spectrum transformation on the filtered signal to obtain the spectrum signal;

[0055] (C) Perform standard energy normalization processing on the spectrum signal to obtain multiple abnormal noise identification features.

[0056] The filtering algorithm mentioned in step (A) above is existing technology and will not be described again. In particular, the filtering frequency band involved above is 50Hz-20000Hz.

[0057] In step (C) above, obtaining the abnormal noise identification features through standard energy normalization processing can be achieved through the following steps (C1) to (C2):

[0058] (C1) Divide the spectrum signal to obtain multiple specific frequency bands after the spectrum signal is divided;

[0059] (C2) The spectrum signal is divided into multiple specific frequency bands, and the abnormal noise identification features of each specific frequency band are obtained.

[0060] In step (C1) above, after repeated testing, the voiceprint energy characteristics are most obvious when the spectrum signal is divided into 100 parts. For example, the spectrum signal is divided into 100 specific frequency bands, wherein, in order to divide the filter frequency bands as evenly as possible, the first specific frequency band is set separately, the band length of the first specific frequency band is set to 150Hz, and the band length of the remaining specific frequency bands is set to 200Hz.

[0061] The filtered frequency bands are divided into several specific frequency bands, including: the first specific frequency band is 50Hz-200Hz, the second specific frequency band is 200Hz-400Hz, the third specific frequency band is 400Hz-600Hz, ..., and the 100th specific frequency band is 19800Hz-20000Hz.

[0062] In step (C2), a specific frequency band is processed to obtain the abnormal noise identification characteristics of that specific frequency band. The aforementioned "processing" refers to first calculating the sum of energy for a specific frequency band according to the energy summation formula involved in step (C1), and then calculating the standard energy of all specific frequency bands. Therefore, to obtain the abnormal noise identification characteristics for each specific frequency band, the following steps (C21) to (C24) can be performed:

[0063] (C21) Calculate the sum of the energy of the first specific frequency band among the plurality of specific frequency bands;

[0064] (C22) Calculate the sum of the energy of each specific frequency band other than the first specific frequency band, and obtain the median of the sum of the energy of each specific frequency band other than the first specific frequency band.

[0065] (C23) Calculate the standard deviation of the sum of the energies of the first specific frequency band and the median, and determine the calculated standard deviation as the standard energy of multiple specific frequency bands;

[0066] (C24) The abnormal noise identification features for each specific frequency band are calculated using the following formula:

[0067]

[0068] Among them, IC j En represents the j-th abnormal sound identification feature. j SE represents the sum of energy in the j-th specific frequency band, and SE represents the standard energy.

[0069] In step (C21) above, the sum of the frequency band energy of each of the multiple specific frequency bands in the spectrum signal is calculated using the following formula:

[0070]

[0071] Among them, En j BEG represents the sum of energy in the j-th specific frequency band. j This represents the start frequency of the j-th specific frequency band, (BEG j +1) j This represents the second frequency in the j-th specific frequency band. END j This represents the termination frequency of the j-th specific frequency band.

[0072] For example, when j is 1, which refers to the first specific frequency band, as given above, the first specific frequency band is 50Hz-200Hz. Therefore, the sum of the energy in the first specific frequency band is:

[0073]

[0074] Wherein, En1 represents the sum of energy in the first specific frequency band, 501 represents the start frequency of the first specific frequency band as 50, and 2001 represents the end frequency of the first specific frequency band as 200.

[0075] Correspondingly, the process of summing the energy of the other 99 specific frequencies in the filter band is as shown above, and therefore will not be repeated.

[0076] Step 103: Use the induced draft fan noise monitoring model to process multiple noise identification features to obtain noise identification results.

[0077] In step 103 above, the induced draft fan noise monitoring model is a neural network model based on an artificial intelligence recognition algorithm. Specifically, the aforementioned artificial intelligence recognition algorithm is a multi-level neural network decision tree algorithm. Since the multi-level neural network decision tree algorithm is existing technology, its specific noise recognition feature identification process will not be repeated. Specifically, if the identification result is a known fault acoustic feature, the fault data and time are reported and added to the fault acoustic sample library. If the identification result does not match the fault acoustic features in the fault acoustic sample library, it is not reported.

[0078] In summary, this embodiment proposes a method for monitoring abnormal noise during the operation of an induced draft fan. Multiple sound signals emitted by the induced draft fan are acquired using a multi-channel pickup device. These signals are then merged into a single monitoring signal through pre-processing using a non-averaging algorithm. This monitoring signal is further processed to extract multiple abnormal noise identification features, which are then input into the induced draft fan abnormal noise monitoring model to obtain an abnormal noise identification result indicating whether the induced draft fan is making abnormal noise. This significantly extends the monitoring time of the induced draft fan, enabling continuous monitoring of its various components and allowing for timely maintenance and repair when a fault is identified based on the abnormal noise identification results. This reduces the operating cost of the induced draft fan diagnostic system and the downtime caused by fan malfunctions.

[0079] Example 2

[0080] This embodiment also proposes a device for monitoring abnormal noise during the operation of an induced draft fan, which is used to perform the method for monitoring abnormal noise during the operation of an induced draft fan proposed in Embodiment 1.

[0081] See Figure 2 The schematic diagram shown illustrates the structure of an induced draft fan noise monitoring device. This embodiment proposes an induced draft fan noise monitoring device, which includes:

[0082] The acquisition module 200 acquires the sound emitted during the operation of the induced draft fan, the sound including: the sound signals collected by each of the multiple sound pickup devices in the multi-channel sound pickup device;

[0083] Algorithm module 201 performs non-average algorithm preprocessing on each audio signal and synthesizes the audio signals into a monitoring signal of each audio signal;

[0084] Processing module 202 processes the monitoring signal to obtain multiple abnormal noise identification features;

[0085] Result module 203 uses the induced draft fan noise monitoring model to process multiple noise identification features to obtain noise identification results.

[0086] Furthermore, the algorithm module is specifically used for:

[0087] The multiple audio signals are mapped into multiple mapped signals, and the multiple mapped signals are weighted and combined into a single monitoring signal for each audio signal.

[0088] In summary, this embodiment proposes a device for monitoring abnormal noise during the operation of an induced draft fan. Multiple sound signals emitted by the induced draft fan are acquired through a multi-channel pickup device. These signals are then merged into a single monitoring signal through pre-processing using a non-averaging algorithm. This monitoring signal is further processed to extract multiple abnormal noise identification features, which are then input into the induced draft fan abnormal noise monitoring model to obtain the abnormal noise identification result indicating whether the induced draft fan is making abnormal noise. This significantly extends the monitoring time of the induced draft fan, enabling continuous monitoring of the fan's various components. Furthermore, when a fault is identified based on the abnormal noise identification result, timely maintenance and repair of the induced draft fan can be performed, reducing the operating cost of the induced draft fan diagnostic system and the downtime caused by fan failures.

[0089] Example 3

[0090] This embodiment proposes a computer-readable storage medium storing a computer program. When the computer program is run by a processor, it executes the steps of the induced draft fan operation abnormality monitoring method described in Embodiment 1. For specific implementation details, please refer to Method Embodiment 1, which will not be repeated here.

[0091] In addition, see Figure 3 The diagram shows the structure of an electronic device. This embodiment also proposes an electronic device, which includes a bus 300, a processor 301, a transceiver 302, a bus interface 303, a memory 304, and a user interface 305. The electronic device includes a memory 304.

[0092] In this embodiment, the electronic device further includes: one or more programs stored in the memory 304 and executable on the processor 301, configured to be executed by the processor to perform the one or more programs for steps (1) to (4):

[0093] (1) Acquire the sound emitted during the operation of the induced draft fan, the sound including: each sound signal collected by each of the multiple sound pickup devices in the multi-channel sound pickup device;

[0094] (2) Perform non-average algorithm preprocessing on each of the audio signals and synthesize each of the audio signals into one monitoring signal of each of the audio signals;

[0095] (3) The monitoring signal is processed to obtain multiple abnormal noise identification features;

[0096] (4) The abnormal noise monitoring model of the induced draft fan is used to process the multiple abnormal noise identification features to obtain the abnormal noise identification results.

[0097] Transceiver 302 is used to receive and send data under the control of processor 301.

[0098] The bus architecture (represented by bus 300) includes any number of interconnected buses and bridges, linking various circuits including one or more processors represented by processor 301 and memory represented by memory 304. Bus 300 can also link various other circuits such as peripheral devices, voltage regulators, and power management circuits, which are well known in the art and therefore will not be further described in this embodiment. Bus interface 303 provides an interface between bus 300 and transceiver 302. Transceiver 302 can be a single element or multiple elements, such as multiple receivers and transmitters, providing a unit for communicating with various other devices over a transmission medium. For example, transceiver 302 receives external data from other devices. Transceiver 302 is used to send data processed by processor 301 to other devices. Depending on the nature of the computing system, a user interface 305 may also be provided, such as a keypad, display, speaker, microphone, or joystick.

[0099] Processor 301 is responsible for managing bus 300 and general processing, such as running general operating system 3041 as described above. Memory 304 can be used to store data used by processor 301 during operation.

[0100] Optionally, the processor 301 may be, but is not limited to, a central processing unit, a microcontroller, a microprocessor, or a programmable logic device.

[0101] It is understood that the memory 304 in the embodiments of this application can be volatile memory or non-volatile memory, or may include both volatile and non-volatile memory. The non-volatile memory can be read-only memory (ROM), programmable read-only memory (PROM), erasable programmable read-only memory (EPROM), electrically erasable programmable read-only memory (EEPROM), or flash memory. The volatile memory can be random access memory (RAM), which is used as an external cache. By way of example, but not limitation, many forms of RAM are available, such as Static Random Access Memory (SRAM), Dynamic Random Access Memory (DRAM), Synchronous DRAM (SDRAM), Double Data Rate SDRAM (DDRSDRAM), Enhanced Synchronous DRAM (ESDRAM), Synchlink DRAM (SLDRAM), and Direct Rambus RAM (DRRAM). The memory 304 of the systems and methods described in this embodiment is intended to include, but is not limited to, these and any other suitable types of memory.

[0102] In some implementations, memory 304 stores elements such as executable modules or data structures, or subsets thereof, or extended sets thereof: operating system 3041 and application program 3042.

[0103] The operating system 3041 includes various system programs, such as the framework layer, core library layer, and driver layer, used to implement various basic business functions and handle hardware-based tasks. The application program 3042 includes various applications, such as a media player and a browser, used to implement various application functions. Programs implementing the methods of this application embodiment can be included in application program 3042.

[0104] The above description is merely a specific implementation of the embodiments of the present invention, but the protection scope of the embodiments of the present invention is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the technical scope disclosed in the embodiments of the present invention should be included within the protection scope of the embodiments of the present invention. Therefore, the protection scope of the embodiments of the present invention should be determined by the protection scope of the claims.

Claims

1. A method for monitoring abnormal noise during the operation of an induced draft fan, characterized in that, include: Acquire the sound emitted during the operation of the induced draft fan, the sound including: the sound signals collected by each of the multiple sound pickup devices in the multi-channel sound pickup device; The audio signals are preprocessed using a non-average algorithm, and then combined into a single monitoring signal. The monitoring signal is processed to obtain multiple abnormal noise identification features. The processing of the monitoring signal to obtain multiple abnormal noise identification features includes: processing the monitoring signal based on a filtering algorithm to generate a filtered signal; performing a Fourier spectrum transformation on the filtered signal to obtain a spectrum signal; dividing the spectrum signal to obtain multiple specific frequency bands; calculating the sum of the energy of a first specific frequency band among the multiple specific frequency bands; calculating the sum of the energy of each specific frequency band other than the first specific frequency band among the multiple specific frequency bands, and obtaining the median of the sum of the energy of each specific frequency band other than the first specific frequency band; calculating the standard deviation between the sum of the energy of the first specific frequency band and the median, and determining the calculated standard deviation as the standard energy of the multiple specific frequency bands; and calculating the abnormal noise identification feature of each specific frequency band using the following formula: in, Indicates the first A unique feature for identifying abnormal noises. Indicates the first The sum of the energy of a specific frequency band Represents standard energy; The abnormal noise monitoring model of the induced draft fan is used to process multiple abnormal noise identification features to obtain abnormal noise identification results.

2. The method for monitoring abnormal noise during the operation of an induced draft fan according to claim 1, characterized in that, The step of performing non-average algorithm preprocessing on the various audio signals to synthesize them into a single monitoring signal includes: The individual audio signals are mapped and converted into multi-channel mapped signals. The multi-channel mapped signals are then weighted and combined into a single monitoring signal from the individual audio signals.

3. The method for monitoring abnormal noise during the operation of an induced draft fan according to claim 2, characterized in that, The weighting process of the multi-path mapped signals to synthesize them into a single monitoring signal from each audio signal includes: Calculate the time-domain energy of each of the mapped signals in the multi-path mapped signals; The multiple mapped signals are sorted according to their time-domain energy. A first weight is assigned to the mapping signal that ranks first in time-domain energy among the multiple mapping signals, and a second weight is assigned to the other mapping signals among the multiple mapping signals excluding the mapping signal that ranks first in time-domain energy; wherein, the first weight is greater than the second weight; The assigned mapping signals are weighted to obtain one monitoring signal from each audio signal.

4. A device for monitoring abnormal noise during the operation of an induced draft fan, characterized in that, include: The acquisition module acquires the sound emitted during the operation of the induced draft fan, including the sound signals collected by each of the multiple sound pickup devices. The algorithm module performs non-average algorithm preprocessing on each audio signal and synthesizes the audio signals into a single monitoring signal. The processing module processes the monitoring signal to obtain multiple abnormal noise identification features; The processing module is specifically used to process the monitoring signal based on a filtering algorithm to generate a filtered signal; perform Fourier spectrum transformation on the filtered signal to obtain a spectrum signal; divide the spectrum signal to obtain multiple specific frequency bands; calculate the sum of the energy of a first specific frequency band among the multiple specific frequency bands; calculate the sum of the energy of each specific frequency band other than the first specific frequency band among the multiple specific frequency bands, and obtain the median of the sum of the energy of each specific frequency band other than the first specific frequency band; calculate the standard deviation of the sum of the energy of the first specific frequency band and the median, and determine the calculated standard deviation as the standard energy of the multiple specific frequency bands; and calculate the abnormal noise identification feature of each specific frequency band using the following formula: in, Indicates the first A unique feature for identifying abnormal noises. Indicates the first The sum of the energy of a specific frequency band Represents standard energy; The results module uses the induced draft fan noise monitoring model to process multiple noise identification features to obtain noise identification results.

5. The induced draft fan operation noise monitoring device according to claim 4, characterized in that, The algorithm module is specifically used for: The individual audio signals are mapped and converted into multi-channel mapped signals. The multi-channel mapped signals are then weighted and combined into a single monitoring signal from the individual audio signals.

6. An electronic device comprising a bus, a transceiver, a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the transceiver, the memory, and the processor are connected via the bus, characterized in that, When the computer program is executed by the processor, it implements the steps in the method for monitoring abnormal noise during operation of an induced draft fan as described in any one of claims 1 to 3.

7. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, it implements the steps in the method for monitoring abnormal noise during the operation of an induced draft fan as described in any one of claims 1 to 3.

Images