1. The utility model provides a voiceprint integrated monitoring system for coal mine scene which characterized in that: the system comprises a voiceprint collector (1), an optical fiber voiceprint monitoring all-in-one machine (2), a cloud server (3) and a terminal device (4), wherein the voiceprint collector (1) is installed on a device to be tested in a coal mine scene, and the voiceprint collector (1) transmits collected voiceprint signals to the optical fiber voiceprint monitoring all-in-one machine (2);

a high-frequency collector (21) is arranged in the optical fiber voiceprint monitoring all-in-one machine (2), and the high-frequency collector (21) is used for receiving and processing the voiceprint signals collected by the voiceprint collector (1);

the voice print system is characterized in that a voice print engine module (31) is arranged inside the cloud server (3), the high-frequency collector (22) transmits collected voice print signals to the voice print engine module (31), and the voice print engine module (21) is used for identifying and analyzing voice print data transmitted by the high-frequency collector (22) and transmitting the processed voice print data to the terminal device (4) through the Ethernet.

2. The comprehensive voiceprint monitoring system applied to the coal mine scene is characterized in that: the voiceprint collector (1) is any one of an acousto-optic fiber audio sensor (11) and a voiceprint sensor (12).

3. The comprehensive voiceprint monitoring system applied to the coal mine scene is characterized in that: the optical fiber voiceprint monitoring all-in-one machine (2) processes monitored equipment voiceprint information, selects fault information and transmits the fault information to the cloud server (3), and the cloud server (3) identifies the selected fault information and analyzes the fault information so as to judge the working state of monitoring equipment and output the result on the terminal equipment (4).

4. The comprehensive voiceprint monitoring system applied to the coal mine scene is characterized by comprising the following components in parts by weight: the high-frequency collector (22) receives the audio of the voiceprint collector (1) and continuously generates an audio file, and the voiceprint engine module (21) receives the audio of the high-frequency collector (22), filters conventional sounds and reports fault audio.

5. The comprehensive voiceprint monitoring system applied to the coal mine scene is characterized in that: the optical fiber voiceprint monitoring all-in-one machine (2) is further internally provided with a fault alarm system (22), and the fault alarm system (22) is used for being connected with a control unit of equipment to be tested in a coal mine scene.

6. The comprehensive voiceprint monitoring system applied to the coal mine scene is characterized in that: the optical fiber voiceprint monitoring all-in-one machine (2) is characterized in that an optical fiber monitoring module (23) is further arranged inside the optical fiber voiceprint monitoring all-in-one machine (2), the optical fiber monitoring module (23) is used for monitoring fiber core vibration in an optical cable connected to equipment to be tested in a coal mine scene, collecting the strength of optical fiber vibration signals, carrying out sampling rate conversion on sampled sound, carrying out interpolation on original signals obtained by sampling to obtain signals, and enabling the audio sampling rate to be matched with a player.

7. The comprehensive voiceprint monitoring system applied to the coal mine scene is characterized in that: the voiceprint engine module (21) employs an Ai-VoiceCmfmc3.0 engine system for lossless conversion of voice formats and cross-channel recognition.

8. The comprehensive voiceprint monitoring system applied to the coal mine scene is characterized in that: the voiceprint engine module (21) performs weighted dimensionality reduction optimization based on the MFCC feature vectors, and identifies noise signals of equipment to be tested in a coal mine scene by applying a vector quantization algorithm.

9. The comprehensive voiceprint monitoring system applied to the coal mine scene is characterized in that: the terminal equipment (4) is any one or combination of a monitoring center, background display equipment, a database and an alarm.

Background

Mine accidents have become one of the most socially interesting problems at present. Many safety-related links are involved in the production process of coal mines. Any problem in these links will cause major accidents or even destructive attacks to the mine, such as explosion caused by coal mine gas leakage, coal mine bed rock water permeability, mine collapse and the like. Therefore, how to guarantee the safety production of the coal mine and how to monitor and early warn accidents in various important links and equipment are problems to be solved urgently.

Disclosure of Invention

The invention aims to provide a voiceprint comprehensive monitoring system applied to a coal mine scene so as to solve the problems in the background technology.

In order to achieve the purpose, the technical scheme of the invention is as follows:

a voiceprint comprehensive monitoring system applied to a coal mine scene comprises a voiceprint collector, an optical fiber voiceprint monitoring all-in-one machine, a cloud server and terminal equipment, wherein the voiceprint collector is installed on equipment to be tested of the coal mine scene, and transmits collected voiceprint signals to the optical fiber voiceprint monitoring all-in-one machine; the high-frequency collector is arranged in the optical fiber voiceprint monitoring all-in-one machine and is used for receiving and processing the voiceprint signals collected by the voiceprint collector; the high-frequency collector is used for collecting voiceprint signals, the voiceprint engine module is arranged in the cloud server, the high-frequency collector transmits the collected voiceprint signals to the voiceprint engine module, the voiceprint engine module is used for identifying and analyzing voiceprint data transmitted by the high-frequency collector, and the processed voiceprint data are transmitted to the terminal equipment through the Ethernet.

In the above scheme, the voiceprint collector is any one of an acousto-optic fiber audio sensor and a voiceprint sensor.

In the above scheme, the optical fiber voiceprint monitoring all-in-one machine processes monitored equipment voiceprint information, screens out fault information and transmits the fault information to the cloud server, and the cloud server identifies the screened-out fault information and analyzes the fault information, so that the working state of monitoring equipment is judged, and a result is output on the terminal equipment.

In the above scheme, the high-frequency collector receives the audio of the voiceprint collector and continuously generates an audio file, and the voiceprint engine module receives the audio of the high-frequency collector, filters conventional sounds, and reports a fault audio.

In the scheme, a fault alarm system is further arranged inside the optical fiber voiceprint monitoring all-in-one machine and is used for being connected with a control unit of the equipment to be tested in a coal mine scene.

In the scheme, the optical fiber acoustic print monitoring all-in-one machine is further internally provided with an optical fiber monitoring module, the optical fiber monitoring module is used for monitoring fiber core vibration in an optical cable connected to equipment to be tested in a coal mine scene, collecting the strength of optical fiber vibration signals, carrying out sampling rate conversion on sampled sound, interpolating original signals acquired by sampling to obtain signals, and enabling the audio sampling rate to be matched with a player.

In the above scheme, the voiceprint engine module adopts an Ai-Voice Cmfmc3.0 engine system, and is used for lossless conversion and cross-channel identification of Voice formats.

In the scheme, the voiceprint engine module performs weighted dimension reduction optimization based on the MFCC characteristic vectors, and identifies the noise signals of the equipment to be tested in a coal mine scene by applying a vector quantization algorithm.

In the above scheme, the terminal device is any one or a combination of a monitoring center, a background display device, a database and an alarm.

Compared with the prior art, the invention has the beneficial effects that: the distributed acoustic-optical fiber audio sensor or the patch type voiceprint sensor is configured for the monitored equipment, various voiceprint information of the equipment is collected through the sensor, and is transmitted to the optical fiber voiceprint monitoring all-in-one machine, the monitored equipment voiceprint information is processed by the optical fiber voiceprint monitoring all-in-one machine, the fault information is screened out and transmitted to the private/public cloud server, the voiceprint engine module in the cloud server identifies the screened fault information and analyzes the fault information, the working state of the monitoring equipment is judged, and the result is output to the terminal equipment. The voiceprint collector is installed on the device to be monitored in a coal mine scene and connected to the voiceprint comprehensive monitoring system, so that the safety state of the device is monitored in real time according to noise emitted by the device, and timely alarming is carried out when a fault is found.

Drawings

The disclosure of the present invention is illustrated with reference to the accompanying drawings. It is to be understood that the drawings are designed solely for the purposes of illustration and not as a definition of the limits of the invention. In the drawings, like reference numerals are used to refer to like parts. Wherein:

FIG. 1 is a schematic view of the overall structure of the present invention;

FIG. 2 is a schematic diagram of the algorithm of the voiceprint engine module of the present invention;

Detailed Description

In order to make the technical means, the creation features, the achievement purposes and the effects of the invention easy to understand, the invention is further described in detail with reference to the attached drawings. These drawings are simplified schematic views illustrating only the basic structure of the present invention in a schematic manner, and thus show only the constitution to which the present invention relates.

According to the technical scheme of the invention, a plurality of alternative structural modes and implementation modes can be provided by a person with ordinary skill in the art without changing the essential spirit of the invention. Therefore, the following detailed description and the accompanying drawings are merely illustrative of the technical aspects of the present invention, and should not be construed as all of the present invention or as limitations or limitations on the technical aspects of the present invention.

The technical solution of the present invention is further described in detail with reference to the accompanying drawings and examples.

Embodiment 1, as shown in fig. 1, a voiceprint comprehensive monitoring system applied to a coal mine scene includes a voiceprint collector 1, an optical fiber voiceprint monitoring all-in-one machine 2, a cloud server 3 and a terminal device 4. The voiceprint collector 1 is any one of an acousto-optic fiber audio sensor 11 and a voiceprint sensor 12, and a distributed acousto-optic fiber audio sensor 11 or a patch type voiceprint sensor 12 is installed on the coal mine equipment according to requirements. The voiceprint collector 1 is installed on a device to be tested in a coal mine scene, for example, in common coal mine production equipment such as a drainage device, a fan and a tape machine, and the voiceprint collector 1 transmits collected voiceprint signals to the optical fiber voiceprint monitoring all-in-one machine 2.

The optical fiber voiceprint monitoring all-in-one machine 2 processes the monitored equipment voiceprint information, selects fault information and transmits the fault information to the cloud server 3, and the cloud server 3 identifies the selected fault information and analyzes the fault information, so that the working state of the monitoring equipment is judged, and the result is output on the terminal equipment 4.

The optical fiber voiceprint monitoring all-in-one machine 2 is internally provided with a high-frequency collector 21, and the high-frequency collector 21 is used for receiving and processing the voiceprint signals collected by the voiceprint collector 1, so that the voiceprint signals can be conveniently identified by the cloud server 3. The high frequency collector 22 receives the audio of the voiceprint collector 1 and continuously generates an audio file, and the voiceprint engine module 21 receives the audio of the high frequency collector 22, filters conventional sounds, and reports fault audio.

The cloud server 3 may be of a private type or a public type, and is configured to analyze the reported specific fault type and distribute fault messages in real time. The cloud server 3 is internally provided with a voiceprint engine module 31, and the voiceprint engine module 21 adopts an Ai-Voice Cmfmc3.0 engine system for lossless conversion and cross-channel identification of Voice formats. The voiceprint engine module 21 has a single 8 channel to facilitate DSP processing. The input end can be connected with the input ends of a mobile phone 16k16bit, a flat plate 8k16bit, a Voip 8k16bit and a telephone 8k16bit, the 16k16bit can be uniformly output at the output end of the voiceprint engine module 21 through conversion to be connected with the voiceprint recognition engines of various manufacturers, and finally the voiceprint recognition engines are transmitted to the application program of the client through the Ethernet.

And the voiceprint engine module 21 performs weighted dimensionality reduction optimization based on the MFCC feature vector, and identifies the noise signal of the equipment to be detected in the coal mine scene by applying a vector quantization algorithm, so that higher identification accuracy and efficiency are obtained, and the system accuracy is more than 95%.

Specifically, referring to fig. 2, in voiceprint recognition, since a speech signal has short-term stability, a signal needs to be preprocessed to accurately extract a feature vector thereof, and a steady-state noise signal of a mining device also has short-term stability and needs to be preprocessed, but the noise signal of the mining device has different characteristics from the speech signal, so that the preprocessing method is different.

The pre-processing typically comprises two steps, framing and windowing. When a noise signal is framed, two frames generally overlap each other to ensure continuity between signals of two adjacent frames. The signal framing relationship that takes into account overlap can be expressed as:

M＝n-Lb/[L(1-b)]

wherein M is the number of frames; n is the noise signal length; l is the frame length; b is the overlap ratio.

20-30ms is generally taken as a frame in voiceprint recognition, and the length of an analysis frame of mining equipment is 500 ms. In addition, considering that the noise signal of the mining equipment under the same operation condition is stable and the continuity between two frames is good, the overlapping rate is set to be 40% for simple calculation. After preprocessing, discrete Fourier transform is needed to be carried out on the signals, Hamming windows are firstly applied to each frame signal and then the signals are transformed, so that the continuity of two ends of the signals is increased, and the distortion phenomenon caused by the Fourier transform is reduced. The MFCC coefficients are cepstral coefficients based on Mel-frequency domain, where Mel-frequency is a frequency domain transformed according to human auditory perception characteristics: and B (h) ═ 2595lg (1h/700), the framing signals respectively obtain a feature vector to form a feature vector group, the obtaining process comprises FFT (fast Fourier transform), Mel (Mel) filtering, logarithmic transformation and discrete cosine transformation, and the Mel filtering is realized by a filter bank consisting of a plurality of triangular band-pass filters. The number of the filters is p, p parameters mi (i 1,2, p) can be obtained after the signals are filtered, and the calculation formula is as follows:

wherein N is the number of FFT points; x (k) is FFT of the preprocessed frame signals; hi (k) is a filter parameter, which can be expressed as:

B(、f[i+1])-B(f[i])＝B(f[i])-B(f[i-1])

wherein f [ i ] in the formula is the center frequency of the triangular filter.

And after mi is obtained through calculation, taking a logarithm of the mi, performing discrete cosine transform, and obtaining c (i) which is the MFCC feature vector of the framing signal through calculation. When the noise signal of the mining equipment is analyzed, the high MFCC characteristic vector dimension can ensure that the noise signal characteristic can be fully extracted. However, too high dimensionality of the feature vectors can take a lot of time, with a consequent increase in computational complexity. In order to improve the calculation efficiency, the PCA algorithm is adopted to carry out dimensionality reduction and simplification on the high-dimensional MFCC feature vector obtained by calculation, and meanwhile, the noise feature of the monitored object is ensured to be accurately obtained.

The high frequency collector 22 transmits the collected voiceprint signal to the voiceprint engine module 31, and the voiceprint engine module 21 is configured to identify and analyze the voiceprint data transmitted by the high frequency collector 22, and transmit the processed voiceprint data to the terminal device 4 through the ethernet.

The terminal equipment 4 is any one or combination of a monitoring center, background display equipment, a database and an alarm, and is used for alarming the identified fault voiceprint signal when connected with the monitoring center; when a background display device such as a computer is connected, the device is used for displaying corresponding fault conditions, so that workers can find the fault conditions in time; when the system is connected with a database, the system is used for integrating data into a database center, and a worker can remotely check the equipment operation condition of the coal mine scene area through the database; when the alarm is connected, the voice print signal directly alarms when a fault is identified. When the alarm is implemented, the monitoring center can be directly connected with the alarm so as to alarm in time.

In the above scheme, the fiber voiceprint monitoring all-in-one machine 2 is further internally provided with a fault alarm system 22, and the fault alarm system 22 is used for connecting a control unit of the equipment to be tested in a coal mine scene. For example, a fault high-power system or device such as a gas detection device, a drainage device, a fan, and a tape machine is detected by the fault alarm system 22 when a control unit of the device has a problem, and then the fault alarm system 22 transmits fault information to the cloud server 3.

In the above scheme, the optical fiber voiceprint monitoring all-in-one machine 2 is also provided with an optical fiber interception module 23 inside, and the optical fiber interception module 23 is used for monitoring fiber core vibration in an optical cable connected to equipment to be tested in a coal mine scene. The method comprises the steps of adopting a Rayleigh scattering technology to obtain audio frequency causing optical fiber vibration, collecting the intensity of an optical fiber vibration signal, carrying out sampling rate conversion on sampled sound, carrying out interpolation on an original signal obtained by sampling to obtain a signal, and enabling the audio frequency sampling rate to be matched with a player. The light path system injects pulse light into the system according to a certain frequency, the light path returns Rayleigh scattered light at every moment, the photoelectric conversion circuit converts an optical signal into an electric signal, an early warning point is judged through an early warning algorithm, and the early warning point is displayed on a terminal map and a panoramic image.

The working principle of the optical fiber interception module 23 is as follows: when light is transmitted in the optical cable, Rayleigh scattered light is continuously transmitted backwards due to the action between photons and fiber core lattices. When vibration occurs outside, the fiber core in the optical cable can be deformed. The tiny vibration can also cause the fiber core to be weakly bent to cause the refractive index to change, the phase of the backward Rayleigh scattering light changes, when the signal light carrying the external vibration information is reflected back to the optical fiber sensor, the signal light is processed by an optical system to convert the weak phase change into light intensity change, after the light intensity change is subjected to photoelectric conversion and signal processing, the signal light enters a computer for data analysis, the nature and the type of an intrusion event are judged according to the analysis result, the length meter number of the cable sheath is judged according to the time difference between the detection light and the rayleigh scattering return light, and the physical position of an intrusion site is accurately judged according to a calibration system.

And the optical fiber interception module 23 is used for carrying out audio sampling reduction processing on a sampling point of a single point. The optical fiber vibration intensity signal of the listening point is stored in the memory, generates optical fiber vibration corresponding to the audio frequency and is reflected to the acquisition module through Rayleigh scattering. And filtering the signal of the signal stability background of the optical fiber vibration signal, and performing a filtering window function, wherein the filtering process is to accumulate the inverted sequence sum.

The fiber audio end detects fiber vibration caused by sound by Rayleigh scattering, and obtains audio by sampling rate conversion, preprocessing, filtering and encoding by using an audio restoring technology. Moreover, the optical fiber audio frequency can realize remote sound judgment, and only one optical fiber is needed at the listening point position without power supply, so that the system has safety and stability.

As described above, the rayleigh scattering audio reduction filtering in the present embodiment is performed and then encoded. The sound generated vibration signal generated by the corresponding position of the optical fiber can be preprocessed in advance, the strength of the sound vibration signal is used for establishing an energy signal time sequence, and the point-changing position is used as a sound source.

In practical tests, a vibration sampling signal of 5000Hz is injected into a 20-kilometer long optical fiber. According to the Nyquist theorem and the Shannon theorem, the audio frequency restoration of 2500Hz can be realized at most.

Embodiment 2, based on the voiceprint monitoring technical scheme in embodiment 1, the mine water pipe is monitored:

the mine water system comprises a plurality of transmission devices and kinetic energy devices such as water pipes, water pumps, generators and the like. By continuously monitoring the acoustic activity, each acoustic event in the pipeline is compared with the sound of the water pipe in a normal state, after environmental noise is filtered, the rest acoustic events contain the basic acoustic characteristics of leakage, and then detailed analysis and evaluation are carried out through signal processing. When the leakage amount of the water pipe is close to the critical value, the system gives an alarm, and the management unit arranges water cut-off, repair or replacement of the pipeline according to the alarm. When the leakage amount of the water pipe is close to the critical value, the system gives an alarm.

Embodiment 3, based on the voiceprint monitoring technical scheme in embodiment 1, kinetic energy devices such as a water pump and a motor in a mine scene are monitored:

the noise signals of the water pump and the motor in operation contain rich equipment information and are closely related to the internal structure, the operation state and the like of the water pump and the motor. When the leakage amount of the water pipe is close to the critical value, the system gives an alarm.

The optical fiber monitoring module 23 senses the vibration information of the equipment by using the optical fiber, and detects the rotating speed of the mobile equipment through information such as frequency, tone, audio characteristics and the like. When the engine rotates, the audio features of the engine correspond to the rotating speed, and audio information of various rotating speeds is collected to serve as a sample library. By means of collecting and analyzing the vibration wave audio information and comparing the vibration wave audio information with a sample library, online rotating speed monitoring is achieved, prejudgment and warning are conducted before a hydroelectric generating set breaks down, equipment halt caused by the faults of the hydroelectric generating set is avoided, and overhaul time of the hydroelectric generating set can be shortened.

Embodiment 4, the gas system is monitored based on the voiceprint monitoring technical scheme in embodiment 1:

the underground coal mine has various gas sources, wherein the combustible harmful gas comprises the following components: once the gases such as carbon monoxide, methane (biogas), hydrogen and the like are combusted, the underground safety can pose great threat and even cause explosion, so the integrity and the stability of the gas system play a very important role in guaranteeing the underground safety.

The method comprises the steps of monitoring a gas system and pipeline equipment in operation through a voiceprint monitoring technology, collecting noise emitted by the gas system and the pipeline equipment, effectively screening and analyzing a complex noise mechanism, identifying different equipment states, and performing prejudgment and alarm before the gas equipment breaks down.

The system monitors a gas system and pipeline equipment in operation, collects noise emitted by the gas pipeline and the equipment, and monitors the safety state of the gas pipeline in real time.

Embodiment 5, monitor the exhaust fan system based on the voiceprint monitoring technical scheme in embodiment 1:

the noise signal of the air draft equipment in operation contains rich equipment information and is closely related to the internal structure, the operation state and the like of the air draft equipment. The voiceprint monitoring technology is applied to air draft equipment, and a complex noise mechanism is effectively screened and analyzed through noise emitted by collecting equipment, so that different equipment states are identified, and a basis is provided for monitoring the equipment states.

The safety monitoring system monitors the running air draft equipment, and monitors the safety state of the air draft system in real time according to the noise sent by the system equipment.

Embodiment 6, the electric power explosion-proof system is monitored based on the voiceprint monitoring technical scheme in embodiment 1:

the voiceprint monitoring is carried out on the transformer and the power transformation cabinet in the electric power explosion-proof system, and voiceprint signals are obtained without being connected into the power transformer, so that the working state of the power transformer is not influenced, and the power transformer is not influenced in high voltage and strong electromagnetic fields. The voiceprint electric power online monitoring system integrates AI + Internet of things +5G, can provide 24-hour online detection for main equipment such as a transformer, a switch cabinet and a bus at present, and mainly provides timely early warning for various fault states of the transformer. And the fault phenomenon is displayed on a background of a monitoring center and a mobile phone APP of an inspection worker in real time, artificial intelligence deep learning is carried out on the fault phenomenon in an abnormal state, a change monitoring curve of the whole life cycle of equipment is provided for the inspection worker, and auxiliary study and judgment are made for a correct operation and inspection decision.

The system provides 24-hour online detection for main equipment such as transformers, switch cabinets and buses, and is mainly used for timely early warning various fault states of the transformers.

Embodiment 7, monitoring the tape machine based on the voiceprint monitoring technical scheme in embodiment 1:

the belt conveyor is not only used in the coal production and processing process, but also widely used in the links of coal mining, production, transportation and the like. However, during the use of the belt conveyor, along with the reasons of abrasion and aging of the equipment, various faults occur to the transmission roller and the carrier roller, such as: the transmission drum and the carrier roller are easy to damage and lock, and the phenomena of damage, locking and the like can be in hard conflict with the adhesive tape, overheat and age, so that the adhesive tape can be burnt seriously, and serious safety accidents are caused. In order to prevent safety hidden troubles caused by the faults of the transmission roller and the carrier roller, the transmission roller and the carrier roller need to be maintained and replaced in time when the transmission roller and the carrier roller are in faults, and the normal operation of the belt conveyor is ensured.

The distributed acoustic optical fiber audio sensor 11 is adopted to monitor the vibration of the carrier roller of the belt conveyor on line in real time, the optical fiber monitoring module 23 transmits vibration voice data collected in real time to the server, the running condition of the equipment is analyzed according to the audio frequency spectrum change rule, and a worker checks the running vibration condition of the equipment in real time at a remote monitoring end. When the equipment runs abnormally, the generated audio frequency spectrum changes, a neural network is used for judging a fault point, the fault condition is analyzed, and a remote monitoring interface receives a fault alarm, so that a worker can take maintenance measures according to the alarm.

The belt conveyor is erected: the optical cable is laid on the inner side of channel steel on one side of the belt conveyor, and is fixed on angle iron below the support when passing through the dragging and rolling support, 1 magnet is installed every 0.5 m, and the optical cable is fixed on the magnet by using a nylon cable tie. After the preliminary experiment is completed, the installation method is improved, and the clamp meeting the coal mine safety standard is designed.

The vibration of the carrying carrier roller of the belt conveyor is monitored in real time on line, the fault sound of the belt conveyor is monitored, the running condition of the equipment is analyzed according to the audio frequency spectrum change rule, and a worker checks the running vibration condition of the equipment in real time at a remote monitoring end.

Embodiment 8, the coal mine bedrock is monitored based on the voiceprint monitoring technical scheme in embodiment 1:

the exploitation of underground coal bed will cause the movement and deformation of upper complex rock stratum, especially in plain agricultural area the underground water level is higher, and ground direct source of flow discharges, and subside depression progressively forms ponding lake, and natural ecology and landform landscape are all destroyed. The subsidence area is a negative factor accompanying in the coal mining process, how to predict the change of the ecological environment in the subsidence range, reasonably planning and comprehensively treating the subsidence area, is an environmental problem which needs to be closely attached to in the coal mining process, and is also an important work in the coal mining development environment influence evaluation and design work.

Acquisition point arrangement position: the rock stratum acoustic emission sensor is stretched into the rock stratum within 0.5 meter, rock stratum vibration and stress change within the range of 50 meters can be monitored, pre-judgment of collapse and underground rivers (underground cavities) can be carried out, fault diagnosis is carried out on the internal environment of the coal mine rock stratum by using a voiceprint monitoring system, and geological disasters such as the underground rivers, the underground cavities and the like of the coal mine rock stratum are judged by emitting reflection of visible sound waves, so that water permeation and collapse accidents of the coal mine rock stratum are early warned.

By collecting and judging the voiceprint information of underground rivers, underground cavities and the like of the rock strata of the ore bed, the system platform monitors the real-time state of the rock strata of the coal mine.

Embodiment 9, based on the voiceprint monitoring technical scheme in embodiment 1, the downhole cable line is monitored:

in order to prevent the situation that the cable wire under the coal mine is dug to be broken or the channel is damaged by external force, the distributed optical fiber audio sensor 11 is applied to the power cable, the external force damage behaviors of the cable are timely found, positioned and early warned by carrying out 24-hour online vibration signal monitoring on the whole wire of the cable to be detected and combining a voiceprint signal analysis algorithm, and powerful technical support is provided for safe operation of the power cable, and management departments and personnel can timely find damage events and stop the damage events.

The state of the cable is found, positioned and pre-warned in time by carrying out 24-hour on-line vibration signal monitoring on the whole cable of the tested cable.

In summary, the distributed acoustic-optical fiber audio sensor 11 or the patch type voiceprint sensor 12 is configured for the monitored device, and various voiceprint information of the device is collected through the sensor and transmitted to the optical fiber voiceprint monitoring all-in-one machine 2, the monitored device voiceprint information is processed by the optical fiber voiceprint monitoring all-in-one machine 2, the fault information is screened out and transmitted to the private/public cloud server 3, the screened fault information is identified and analyzed by the voiceprint engine module 31 in the cloud server 3, so that the working state of the monitoring device is judged, and the result is output to the terminal device 4. By installing the voiceprint collector 1 on the device to be monitored in a coal mine scene and connecting the voiceprint collector to the voiceprint comprehensive monitoring system, the safety state of the device is monitored in real time according to noise sent by the device, and timely alarming is carried out when a fault is found.

It should be noted that, in this document, terms such as "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus.

The above-mentioned embodiments, objects, technical solutions and advantages of the present invention are further described in detail, it should be understood that the above-mentioned embodiments are only specific embodiments of the present invention, and are not intended to limit the scope of the present invention, and any modifications, equivalent substitutions, improvements and the like made within the spirit and principle of the present invention should be included in the scope of the present invention.