A coal mill operation monitoring system based on voiceprint recognition

Abstract

The application discloses a coal mill operation monitoring system based on voiceprint recognition and belongs to the technical field of coal mill operation monitoring, which comprises a platform end and a user end. The platform end comprises a critical analysis module. The critical analysis module is used for analyzing the coal mill information of users, determining a critical standard range and an abnormal voiceprint set data, and sending them to the corresponding user end. The user end comprises a voiceprint collection module, a standard debugging module and a monitoring analysis module. The voiceprint collection module is used for collecting the voiceprint information in a target environment in real time. The standard debugging module is used for setting and updating the monitoring segmentation standard of the coal mill in the target environment in real time. The monitoring segmentation standard comprises the critical standard range and an application standard range. The monitoring analysis module is used for monitoring and analyzing the coal mill, identifying the voiceprint information of the coal mill in real time, obtaining the monitoring segmentation standard, analyzing the voiceprint information in real time based on the monitoring segmentation standard, obtaining a monitoring analysis result, and processing the monitoring analysis result correspondingly.

Description

Technical Field

[0001] This invention belongs to the field of coal mill operation monitoring technology, specifically a coal mill operation monitoring system based on voiceprint recognition. Background Technology

[0002] Coal mills are critical equipment commonly used in industries such as power generation and cement, and their operating status directly affects the efficiency and safety of the entire production line. Traditional coal mill monitoring methods typically rely on physical inspections and manual patrols. These methods are not only time-consuming and labor-intensive, but also struggle to monitor the coal mill's operating status in real time and promptly identify potential faults. Therefore, how to achieve intelligent monitoring of coal mill operation has become a problem that needs to be solved.

[0003] Based on this, the present invention provides a coal mill operation monitoring system based on voiceprint recognition. Summary of the Invention

[0004] To address the problems of the above solutions, this invention provides a coal mill operation monitoring system based on voiceprint recognition.

[0005] The objective of this invention can be achieved through the following technical solutions:

[0006] A coal mill operation monitoring system based on voiceprint recognition, comprising a platform terminal and a user terminal;

[0007] The platform includes a critical analysis module;

[0008] The critical analysis module is used to analyze the user's coal mill information, identify the user's coal mill information, and match the corresponding critical standard range and abnormal soundprint set data from the preset platform reserve based on the coal mill information. The critical standard range is the soundprint recognition range of the coal mill during normal operation; the abnormal soundprint set data consists of the soundprint recognition range of the coal mill during abnormal operation and the abnormal problems corresponding to the soundprint recognition data within the soundprint recognition range; and the critical standard range and abnormal soundprint set data are sent to the corresponding user terminal.

[0009] Furthermore, the methods for establishing the platform's reserve repository include:

[0010] The platform collects various coal mills and gathers historical voiceprint data of the coal mills in real time. The historical voiceprint data includes voiceprint recognition data and coal mill status data. The historical voiceprint data is then classified according to the coal mill to form coal mill material data.

[0011] The coal mill material data is classified to form abnormal classification data and normal classification data of the coal mill; the voiceprint recognition data corresponding to the abnormal classification data and normal classification data are identified respectively, and integrated to form an abnormal voiceprint set and a normal voiceprint set, and the corresponding abnormal problems are marked for the voiceprint recognition data in the abnormal voiceprint set;

[0012] A critical standard range is generated based on the normal and abnormal voiceprint sets; a platform reserve is established based on the critical standard range and abnormal voiceprint set data corresponding to the coal mill.

[0013] Furthermore, methods for classifying coal mill material data include:

[0014] Establish a material classification model. The expression for the material classification model is:

[0015]

[0016] In the formula: (S, K) are the input data, S is the historical voiceprint data; K is the material classification standard; S→K means that the corresponding historical voiceprint data conforms to the material classification standard; the output data is the material classification value SF(S, K), and the material classification value is 1 or 0;

[0017] By analyzing the coal mill material data using a material classification model, the material classification value corresponding to each historical voiceprint data within the coal mill material data is obtained;

[0018] Historical voiceprint data with a classification value of 1 are grouped into one category;

[0019] Historical voiceprint data with a classification value of 0 are grouped into one category;

[0020] Determine abnormal and normal classification data based on the material classification standards.

[0021] The user terminal includes a voiceprint acquisition module, a standard debugging module, and a monitoring and analysis module;

[0022] The voiceprint acquisition module is used to acquire voiceprint information in the target environment in real time, and the target environment is the operating environment of the coal mill;

[0023] The standard commissioning module is used to set and update the monitoring segment standards of the coal mill in the target environment in real time. The monitoring segment standards include critical standard ranges and application standard ranges.

[0024] Furthermore, the methods for setting and updating monitoring segmentation standards include:

[0025] Determine the debugging duration, obtain the corresponding voiceprint information based on the debugging duration, identify the acquisition time corresponding to the voiceprint information, and obtain the coal mill detection status of the voiceprint information based on the acquisition time.

[0026] Based on the detection status of the coal mill, the acoustic fingerprint information is classified to form debugging classification data; the acoustic fingerprint set corresponding to the debugging classification data is identified; acoustic fingerprint values ​​are set for the corresponding acoustic fingerprint information in the acoustic fingerprint set; and the application standard range is set according to the acoustic fingerprint set and the acoustic fingerprint values.

[0027] Identify the critical standard range and formulate monitoring segmentation standards based on the critical standard range and the application standard range;

[0028] The system acquires the acoustic fingerprint information and detection status of the coal mill in real time, and updates the application standard range in real time based on the acoustic fingerprint information and coal mill detection status; it also updates the monitoring segment standards based on the updated application standard range.

[0029] Furthermore, the methods for setting the debugging duration include:

[0030] Set the unit duration and calculate the update rate of the application standard range within the corresponding unit duration in real time.

[0031] The update rate is compared with a threshold X1;

[0032] When the update rate is less than the threshold X1, end the debugging process and determine the debugging duration.

[0033] Continue debugging when the update rate is not less than the threshold X1.

[0034] Furthermore, the method for setting voiceprint values ​​for corresponding voiceprint information within the voiceprint set includes:

[0035] A real-time statistical time record table of corresponding voiceprint information in the voiceprint set is generated. The time record table consists of several unit records, and the unit records are marked as i, i = 1, 2, ..., n, where n is the number of unit records in the time record table.

[0036] The identification unit records the corresponding acquisition time and marks the acquisition time as TCi;

[0037] Real-time identification of the current time, marking the current time as TD;

[0038] The representative value of the corresponding voiceprint information within the voiceprint set is calculated based on the representative formula, which is:

[0039]

[0040] In the formula: DB is the representative value; μ is the adjustment coefficient, with a value range of [1.1, 1.2]; the unit of TD-TCi is days;

[0041] The representative value of the voiceprint information is marked as DBj, where j represents the corresponding voiceprint information, j = 1, 2, ..., m, and m is the number of voiceprint information in the voiceprint set;

[0042] The voiceprint value of the corresponding voiceprint information is calculated according to the voiceprint value calculation formula, which is:

[0043]

[0044] In the formula: SWj is the voiceprint value.

[0045] The monitoring and analysis module is used to monitor and analyze the coal mill, identify the acoustic fingerprint information of the coal mill in real time, obtain monitoring segmentation standards, perform real-time analysis of the acoustic fingerprint information based on the monitoring segmentation standards, obtain monitoring and analysis results, and perform corresponding processing based on the monitoring and analysis results.

[0046] Furthermore, methods for analyzing voiceprint information based on monitoring segmentation standards include:

[0047] The critical standard range and the application standard range corresponding to the monitoring segmentation standard are marked as LU and YU, respectively.

[0048] A monitoring and evaluation model is established, and its expression is as follows:

[0049]

[0050] In the formula: (SQ, LU, YU) are the input data, SQ is the voiceprint information; SQ∈YU indicates that the voiceprint information belongs to the application standard range; This indicates that the voiceprint information is not within the scope of the application standard; the output data is the monitoring evaluation value GP(SQ), which is 1, 2 or 3;

[0051] The monitoring and evaluation model is used to analyze the monitoring segmentation standards and voiceprint information to obtain the corresponding monitoring and evaluation values.

[0052] When the monitoring and evaluation value is 1, the evaluation is running normally.

[0053] When the monitoring and evaluation value is 2, the operation is considered abnormal, and the corresponding abnormal value is calculated.

[0054] When the monitoring and evaluation value is 3, an operational failure is assessed.

[0055] Furthermore, methods for calculating outliers include:

[0056] The voiceprint value of the identified voiceprint information is marked as SW0;

[0057] A reference voiceprint value is determined based on the application standard range, and the reference voiceprint value is the minimum voiceprint value corresponding to the application standard range.

[0058] Calculate the corresponding outlier value according to the outlier calculation formula, which is:

[0059] YZ = [1-(2 SW′-SW0 -1)]×100 ；

[0060] In the formula: YZ represents outliers; SW′ represents the baseline voiceprint value.

[0061] Compared with the prior art, the beneficial effects of the present invention are:

[0062] This invention enables real-time monitoring and fault diagnosis of the coal mill's operating status by extracting the acoustic signal characteristics during operation. Compared to traditional physical inspections and manual checks, this method offers higher accuracy and reliability, allowing for the timely detection of potential faults and improving production safety. It also enables remote monitoring and fault diagnosis, reducing the frequency and number of manual inspections and lowering maintenance costs. Furthermore, because the system can monitor the coal mill's operating status in real time, it can provide early warnings of potential faults, preventing equipment damage and production interruptions, further reducing economic losses for the enterprise. Attached Figure Description

[0063] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0064] Figure 1 This is a block diagram illustrating the principle of the present invention. Detailed Implementation

[0065] The technical solution of the present invention will be clearly and completely described below with reference to the embodiments. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of the present invention.

[0066] like Figure 1 As shown, a coal mill operation monitoring system based on voiceprint recognition includes a platform terminal and a user terminal;

[0067] The platform is used by the platform provider and includes a critical analysis module;

[0068] The critical analysis module is used to analyze the user's coal mill information, determine the critical standard range and abnormal sound pattern set data of the coal mill, and the critical standard range is the sound pattern recognition range corresponding to the normal operation of the corresponding coal mill; and send the critical standard range and abnormal sound pattern set data to the corresponding user terminal.

[0069] In one embodiment, the critical standard range can be set based on existing methods, such as by platform staff assisting users in setting it.

[0070] In one embodiment, the method for determining the critical criterion range includes:

[0071] The platform aggregates various coal mills currently available on the market and collects historical voiceprint data of each coal mill in real time. The historical voiceprint data includes voiceprint recognition data and coal mill status data. It can be collected by connecting to the user terminal after authorization, or it can be collected through other channels. The obtained historical voiceprint data is classified according to coal mills to form corresponding coal mill material data.

[0072] The coal mill data is categorized according to whether it is operating normally, resulting in abnormal and normal categories. Voiceprint recognition data corresponding to these categories is then identified and integrated into abnormal and normal voiceprint sets. The abnormal voiceprint sets are then labeled with corresponding abnormal issues, such as gear malfunctions; one voiceprint can correspond to multiple abnormal issues. Critical standard ranges are generated based on the normal and abnormal voiceprint sets, and these ranges can be verified and adjusted by platform professionals.

[0073] A platform reserve database was established based on the critical standard range and abnormal sound signature data corresponding to each coal mill.

[0074] Identify the user's coal mill information and match the corresponding critical standard range and abnormal sound pattern set data from the platform's reserve database based on the coal mill information.

[0075] In one embodiment, a method for classifying coal mill material data according to whether it is operating normally includes:

[0076] Establish a material classification model. The expression for the material classification model is:

[0077]

[0078] In the formula: (S, K) are the input data, S is the historical voiceprint data; K is the material classification standard, such as abnormal state or normal state, S→K means that the corresponding historical voiceprint data meets the material classification standard; the output data is the material classification value SF(S, K), and the material classification value is 1 or 0;

[0079] By analyzing the coal mill material data using a material classification model, the material classification value corresponding to each historical voiceprint data within the coal mill material data is obtained;

[0080] Historical voiceprint data with a classification value of 1 are grouped into one category;

[0081] Historical voiceprint data with a classification value of 0 are grouped into one category;

[0082] Determine abnormal and normal classification data based on the material classification standards.

[0083] The user terminal is for user use and includes a voiceprint acquisition module, a standard debugging module, and a monitoring and analysis module;

[0084] The voiceprint acquisition module is used to collect voiceprint information in the target environment in real time, mainly based on the corresponding acquisition equipment provided by the platform.

[0085] The target environment refers to the coal mill operation monitoring environment.

[0086] The standard commissioning module is used to set and update the monitoring segment standards of the coal mill in the target environment in real time. The monitoring segment standards include the critical standard range and the application standard range. The application standard range is equivalent to the range after narrowing the critical standard range.

[0087] In one embodiment, the method for monitoring the setting and updating of segmentation criteria includes:

[0088] The platform provider sets the debugging duration, which is the time period from the initial time to the application's corresponding standard range. This is used to improve the accuracy of the application standard range in the early stages, such as one week, three days, ten days, etc. The specific settings are determined by the platform provider.

[0089] During the debugging period, the system acquires voiceprint information collected by the voiceprint acquisition module, identifies the acquisition time corresponding to the voiceprint information, and obtains the corresponding coal mill detection status based on the acquisition time. The coal mill detection status can be uploaded by the user or determined by the detection system. Since the collected voiceprint information does not need to be analyzed in time, the already determined coal mill detection status can be obtained directly. The coal mill detection status includes normal detection and abnormal detection. The voiceprint information is classified according to the coal mill detection status, and the voiceprint information corresponding to normal detection is integrated into debugging classification data. The system identifies the voiceprint set corresponding to the debugging classification data. The system sets the corresponding voiceprint values ​​for the corresponding voiceprint information in the voiceprint set. The system sets the application standard range according to the voiceprint set.

[0090] Identify the critical standard range and formulate monitoring segmentation standards based on the critical standard range and the application standard range;

[0091] The system acquires the acoustic fingerprint information and detection status of the coal mill in real time, and updates the application standard range in real time based on the acoustic fingerprint information and detection status of the coal mill, that is, updates the acoustic fingerprint set and the corresponding acoustic fingerprint value.

[0092] The monitoring segmentation standards were updated based on the updated application standard scope.

[0093] In one embodiment, the debugging duration can also be set in the following ways:

[0094] The unit duration is set by the platform, such as one day; the application standard range is identified in real time. The initial application standard range is none. As long as the debugging is not over, a new application standard range will be formed based on the voiceprint information and the coal mill detection status. The application standard range will not be updated until the debugging period is over.

[0095] Calculate the update rate of the application standard scope within the corresponding unit time period, that is, the update rate of forming a new application standard scope within the unit time period. For example, if it is updated 10 times in one day, the update rate is 10 in 24 hours.

[0096] The obtained update rate is compared with the threshold X1;

[0097] When the update rate is less than the threshold X1, end the debugging process and determine the debugging duration.

[0098] Continue debugging when the update rate is not less than the threshold X1.

[0099] In one embodiment, a method for setting a voiceprint value for corresponding voiceprint information within a voiceprint set includes:

[0100] A real-time statistical time record table for corresponding voiceprint information within the voiceprint set is generated. The time record table is used to count the collection time of each voiceprint information acquisition. Each voiceprint information record in the time record table is marked as a unit record. That is, the time record table consists of several unit records. The unit record is marked as i, i = 1, 2, ..., n, where n is the number of unit records in the time record table.

[0101] The identification unit records the corresponding acquisition time and marks the acquisition time as TCi;

[0102] Real-time identification of the current time, marking the current time as TD;

[0103] The representative value of the corresponding voiceprint information within the voiceprint set is calculated based on the representative formula, which is:

[0104]

[0105] In the formula: DB is the representative value; μ is the adjustment coefficient, with a value range of [1.1, 1.2], and the specific adjustment coefficient is selected according to the user's needs; the unit of TD-TCi is days, such as TD-TCi equal to 36 hours, which is converted to 1.5 days;

[0106] The representative value of the corresponding voiceprint information is marked as DBj, where j represents the corresponding voiceprint information, j = 1, 2, ..., m, and m is the number of voiceprint information in the voiceprint set;

[0107] The voiceprint value of the corresponding voiceprint information is calculated according to the voiceprint value calculation formula, which is:

[0108]

[0109] In the formula: SWj is the voiceprint value.

[0110] In one embodiment, a method for setting the application standard range based on a voiceprint set includes:

[0111] Identify the voiceprint value of each voiceprint information in the voiceprint set; integrate voiceprint information with a voiceprint value greater than threshold X2 into a standard set, and set the application standard range according to the standard set.

[0112] The monitoring and analysis module is used to monitor and analyze the coal mill, identify the soundprint information of the coal mill in real time, obtain monitoring segmentation standards, analyze the identified soundprint information in real time based on the monitoring segmentation standards, obtain monitoring and analysis results, and perform corresponding processing based on the monitoring and analysis results, such as shutdown processing for operational faults, manual detection of abnormal warnings, etc., and pre-set corresponding processing measures according to the user's processing needs.

[0113] In one embodiment, the method for analyzing voiceprint information based on monitoring segmentation criteria includes:

[0114] The critical standard range and the application standard range corresponding to the monitoring segmentation standard are marked as LU and YU, respectively.

[0115] A monitoring and evaluation model is established, and its expression is as follows:

[0116]

[0117] In the formula: (SQ, LU, YU) are the input data, SQ is the voiceprint information; SQ∈YU indicates that the voiceprint information belongs to the application standard range; This indicates that the voiceprint information does not fall within the scope of the application standard. Existing voiceprint recognition technology will be used to determine whether it does. The output data is the monitoring evaluation value GP(SQ), which is 1, 2, or 3.

[0118] The monitoring and evaluation model is used to analyze the monitoring segmentation standards and voiceprint information to obtain the corresponding monitoring and evaluation values.

[0119] When the monitoring and evaluation value is 1, the evaluation is running normally.

[0120] When the monitoring and evaluation value is 2, the operation is considered abnormal, and the corresponding abnormal value is calculated.

[0121] When the monitoring and evaluation value is 3, an operational failure is assessed.

[0122] In one embodiment, the method for calculating outliers includes:

[0123] Identify the voiceprint value corresponding to this voiceprint information and mark it as SW0;

[0124] Identify the smallest voiceprint value corresponding to the application standard range and mark it as the baseline voiceprint value;

[0125] Calculate the corresponding outlier value according to the outlier calculation formula, which is:

[0126] YZ = [1-(2 SW′-SW0 -1)]×100 ；

[0127] In the formula: YZ represents outliers; SW′ represents the baseline voiceprint value.

[0128] The above formulas are all numerical calculations after removing dimensions. The formulas are obtained by software simulation based on a large amount of data and are closest to the real situation. The preset parameters and preset thresholds in the formulas are set by those skilled in the art according to the actual situation or obtained by simulation based on a large amount of data.

[0129] The above embodiments are only used to illustrate the technical methods of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical methods of the present invention without departing from the spirit and scope of the technical methods of the present invention.

Claims

1. A coal mill operation monitoring system based on voiceprint recognition, characterized by, The platform end and the user end are included; The platform end includes a critical analysis module; The critical analysis module is used for analyzing the coal mill information of a user, identifying the coal mill information of the user, matching corresponding critical standard ranges and abnormal voiceprint set data from a preset platform repository according to the coal mill information, the critical standard range being a voiceprint identification range of the coal mill in normal operation, the abnormal voiceprint set data being composed of a voiceprint identification range of the coal mill in abnormal operation and abnormal problems corresponding to the voiceprint identification data in the voiceprint identification range, and sending the critical standard range and the abnormal voiceprint set data to the corresponding user end; The user end includes a voiceprint collection module, a standard debugging module and a monitoring analysis module; The voiceprint collection module is used for collecting voiceprint information in a target environment in real time, the target environment being an operating environment of the coal mill; The standard debugging module is used for setting and updating monitoring segmentation standards of the coal mill in the target environment in real time, the monitoring segmentation standards including critical standard ranges and application standard ranges; The monitoring analysis module is used for monitoring and analyzing the coal mill, identifying voiceprint information of the coal mill in real time, obtaining monitoring segmentation standards, analyzing the voiceprint information in real time based on the monitoring segmentation standards, obtaining monitoring analysis results, and processing the monitoring analysis results accordingly.

2. A coal mill operation monitoring system based on voiceprint recognition according to claim 1, characterized in that, The method for establishing the platform repository includes: A platform party collects various coal mills, collects historical voiceprint data of the coal mills in real time, classifies the historical voiceprint data according to the coal mills to form coal mill material data of the coal mills, classifies the coal mill material data to form abnormal classification data and normal classification data of the coal mills, identifies voiceprint identification data corresponding to the abnormal classification data and the normal classification data respectively, integrates to form an abnormal voiceprint set and a normal voiceprint set, marks corresponding abnormal problems for the voiceprint identification data in the abnormal voiceprint set, generates critical standard ranges according to the normal voiceprint set and the abnormal voiceprint set, and establishes a platform repository according to the critical standard ranges and the abnormal voiceprint set data corresponding to the coal mills. The method for classifying the coal mill material data includes: A material classification model is established, and an expression of the material classification model is as follows:

3. A mill operation monitoring system based on voiceprint recognition as claimed in claim 2, wherein, In the formula, (S, K) is input data, S is historical voiceprint data, K is a material classification standard, S→K represents that the corresponding historical voiceprint data meets the material classification standard, and an output data is a material classification value SF(S, K), which is 1 or 0; The material classification model is used to analyze the coal mill material data, so as to obtain material classification values corresponding to each historical voiceprint data in the coal mill material data; The historical voiceprint data with the material classification value of 1 is classified into a category; The historical voiceprint data with the material classification value of 0 is classified into a category; The abnormal classification data and the normal classification data are determined according to the material classification standard. The method for setting and updating the monitoring segmentation standards includes: ​ 4. A coal mill operation monitoring system based on voiceprint recognition according to claim 1, characterized in that, ​ Determine the debugging duration, obtain the corresponding voiceprint information according to the debugging duration, identify the collection time corresponding to the voiceprint information, and obtain the mill detection state of the voiceprint information according to the collection time; Classify the voiceprint information according to the mill detection state to form debugging classification data; identify the voiceprint set corresponding to the debugging classification data; set the voiceprint value for the corresponding voiceprint information in the voiceprint set; set the application standard range according to the voiceprint set and the voiceprint value; Identify the critical standard range, and form the monitoring segmentation standard according to the critical standard range and the application standard range; Obtain the voiceprint information and the mill detection state of the mill in real time, and update the application standard range in real time according to the voiceprint information and the mill detection state; update the monitoring segmentation standard according to the updated application standard range.

5. A coal mill operation monitoring system based on voiceprint recognition according to claim 4, characterized in that, The setting method of the debugging duration includes: Set the unit duration, and calculate the update rate of the application standard range in the corresponding unit duration in real time; Compare the update rate with the threshold X1; When the update rate is less than the threshold X1, end the debugging and determine the debugging duration; When the update rate is not less than the threshold X1, continue the debugging.

6. A coal mill operation monitoring system based on voiceprint recognition as claimed in claim 4 wherein, The method for setting the voiceprint value for the corresponding voiceprint information in the voiceprint set includes: Real-time statistics of the time record table of the corresponding voiceprint information in the voiceprint set, the time record table is composed of a plurality of unit records, the unit record is marked as i, i=1, 2, …, n, n is the number of unit records in the time record table; Identify the collection time corresponding to the unit record, and mark the collection time as TCi; Identify the current time in real time, and mark the current time as TD; Calculate the representative value of the corresponding voiceprint information in the voiceprint set according to the representative formula, and the representative formula is: In the formula: DB is the representative value; μ is an adjustment coefficient, and the value range is [1.1, 1.2]; the unit of TD-TCi is day; Mark the representative value of the voiceprint information as DBj, j represents the corresponding voiceprint information, j=1, 2, …, m, m is the number of voiceprint information in the voiceprint set; Calculate the voiceprint value of the corresponding voiceprint information according to the voiceprint value calculation formula, and the voiceprint value formula is: In the formula: SWj is the voiceprint value.

7. A coal mill operation monitoring system based on voiceprint recognition as claimed in claim 1, wherein, The method for analyzing the voiceprint information based on the monitoring segmentation standard includes: Mark the critical standard range and the application standard range corresponding to the monitoring segmentation standard as LU and YU respectively; Establish a monitoring evaluation model, and the expression of the monitoring evaluation model is: In the formula: (SQ, LU, YU) is input data, SQ is voiceprint information; SQ∈YU indicates that the voiceprint information belongs to the application standard range; SQ∉YU indicates that the voiceprint information does not belong to the application standard range; the output data is a monitoring evaluation value GP(SQ), and the monitoring evaluation value is 1, 2 or 3. Analyze the monitoring segmentation standard and the voiceprint information through the monitoring evaluation model to obtain the corresponding monitoring evaluation value; When the monitoring evaluation value is 1, the evaluation is normal; When the monitoring evaluation value is 2, the evaluation is abnormal, and the corresponding abnormal value is calculated; When the monitoring evaluation value is 3, the evaluation is failure.

8. A coal mill operation monitoring system based on voiceprint recognition according to claim 7, characterized in that, The calculation method of the abnormal value includes: Identify the voiceprint value of the voiceprint information, marked as SW0; Determine the reference voiceprint value according to the application standard range, and the reference voiceprint value is the minimum voiceprint value corresponding to the application standard range; Calculate the corresponding abnormal value according to the abnormal value calculation formula, and the abnormal value calculation formula is: YZ = [1 - (2 SW′-SW0 -1)] x 100; In the formula: YZ is the abnormal value; SW' is the reference voiceprint value.

Images