Compressor state monitoring method, device and system

Abstract

The invention provides a compressor state monitoring method, device and system, and the method comprises the steps: filtering environment noise data from the audio data of a compressor in an operation state, and obtaining the operation audio data of the compressor; classifying the operation audio data to obtain audio data of each part of the compressor; determining an initial state monitoring result of each part of the compressor based on the audio data of each part; and based on the temperature data of each part of the compressor in the running state, correcting the initial state monitoring result of each part, and determining the state monitoring result of each part of the compressor. According to the compressor state monitoring method, device and system, the initial state monitoring result of each part can be accurately determined in real time based on the audio data of each part, and then the state monitoring result of each part can be accurately obtained in real time after the initial state monitoring result of each part is obtained based on the temperature data of each part.

Description

Technical Field

[0001] The present invention relates to the technical field of natural gas stations, and in particular to a compressor status monitoring method, device and system. Background Art

[0002] With the development of digitalization and intelligence, with the help of relevant remote management technologies, unmanned natural gas pipeline stations will become a future development trend. However, since there are no staff to conduct real-time inspections at unmanned natural gas stations, it is particularly important to ensure that abnormalities in the equipment in the station can be discovered in time, especially for compressor stations. The compressor is the most important large-scale equipment in the station. Timely detection of compressor abnormalities is of great significance to the safe operation of the gas pipeline. Existing compressor detection methods often rely on on-site detection by detection personnel, which cannot meet the needs of unmanned stations and cannot achieve real-time detection. Summary of the invention

[0003] The present invention provides a compressor status monitoring method, device and system, which are used to solve the defect in the prior art that the compressor status cannot be detected in real time by relying on monitoring personnel on site.

[0004] The present invention provides a compressor state monitoring method, comprising:

[0005] filtering out environmental noise data from audio data of the compressor in operation to obtain operation audio data of the compressor;

[0006] Classifying the operating audio data to obtain audio data of various parts of the compressor;

[0007] Determining initial state monitoring results of various parts of the compressor based on the audio data of various parts;

[0008] Based on the temperature data of each part of the compressor in the running state, the initial state monitoring results of each part are corrected to determine the state monitoring results of each part of the compressor.

[0009] According to a compressor state monitoring method provided by the present invention, filtering out environmental noise data from audio data of the compressor in operation to obtain the operation audio data of the compressor includes:

[0010] Based on the noise filtering model, filtering the environmental noise data from the audio data to obtain the operating audio data of the compressor;

[0011] The noise filtering model is trained based on sample audio data mixed with environmental noise and corresponding sample running audio data.

[0012] According to a compressor state monitoring method provided by the present invention, the initial state monitoring results of each part of the compressor are determined based on the audio data of each part, including:

[0013] Based on the state monitoring model, the initial state monitoring results of each part of the compressor are determined by applying the audio data of each part;

[0014] The state monitoring model is trained based on sample audio data of each part and sample initial state monitoring labels of the corresponding part.

[0015] According to a compressor state monitoring method provided by the present invention, based on the temperature data of each part of the compressor in the running state, the initial state monitoring results of each part are corrected to determine the state monitoring results of each part of the compressor, including:

[0016] Determine the correction weight based on the temperature data of each part and the temperature baseline; the temperature baseline refers to the curve of the temperature change of the compressor over time under normal working conditions;

[0017] Based on the correction weights, the initial status monitoring results of each part are corrected to determine the status monitoring results of each part of the compressor.

[0018] According to a compressor state monitoring method provided by the present invention, based on the temperature data of each part of the compressor in the running state, the initial state monitoring results of each part are corrected to determine the state monitoring results of each part of the compressor, and then the method further includes:

[0019] If the monitoring result of any part status is less than the threshold, the compressor status is marked as high risk;

[0020] After marking the compressor status as high-risk, determine whether the status monitoring results of the corresponding parts of the audio data for consecutive predicted times are greater than or equal to the threshold. If so, mark the compressor from high-risk to normal; if not, issue an early warning prompt.

[0021] According to a compressor state monitoring method provided by the present invention, based on the temperature data of each part of the compressor in the running state, the initial state monitoring results of each part are corrected to determine the state monitoring results of each part of the compressor, and then the method further includes:

[0022] If the status monitoring result of any part is abnormal, the camera is controlled to focus on the corresponding abnormal part.

[0023] According to a compressor status monitoring method provided by the present invention, the method further includes:

[0024] If it is detected that the methane concentration within the preset range of the compressor is greater than the preset concentration, an early warning prompt is issued.

[0025] The present invention also provides a compressor state monitoring device, comprising:

[0026] A denoising unit, used to filter out environmental noise data from audio data of the compressor in operation to obtain operation audio data of the compressor;

[0027] A classification unit, used for classifying the operation audio data to obtain audio data of each part of the compressor;

[0028] A prediction unit, used to determine the initial state monitoring results of each part of the compressor based on the audio data of each part;

[0029] The correction unit is used to correct the initial state monitoring results of each part based on the temperature data of each part of the compressor in the running state, so as to determine the state monitoring results of each part of the compressor.

[0030] The present invention also provides a compressor state monitoring system, comprising:

[0031] The compressor status monitoring device, audio monitoring device, temperature monitoring device, video monitoring device and gas leakage detection device as described above;

[0032] The audio monitoring device is used to collect audio data of the compressor in operation, the temperature monitoring device is used to collect temperature data of various parts of the compressor in operation, the video monitoring device is used to monitor the compressor by video, and the gas leakage detection device is used to detect the methane concentration of the compressor within a preset range. The present invention also provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements any of the above-mentioned compressor state monitoring methods when executing the computer program.

[0033] The present invention also provides a non-transitory computer-readable storage medium having a computer program stored thereon, and when the computer program is executed by a processor, the compressor state monitoring method as described in any one of the above is implemented.

[0034] The present invention also provides a computer program product, comprising a computer program, wherein when the computer program is executed by a processor, the compressor state monitoring method as described above is implemented.

[0035] The compressor state monitoring method, device and system provided by the present invention filter out environmental noise data from the audio data of the compressor in the operating state, and classify the operating audio data from which the environmental noise data is filtered out to obtain audio data of each part, thereby being able to accurately and in real time determine the initial state monitoring results of each part of the compressor based on the audio data of each part, and further, after the initial state monitoring results of each part are determined based on the temperature data of each part, the state monitoring results of each part of the compressor can be accurately and in real time obtained. BRIEF DESCRIPTION OF THE DRAWINGS

[0036] In order to more clearly illustrate the technical solutions in the present invention or the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0037] Figure 1 It is a flow chart of the compressor state monitoring method provided by the present invention;

[0038] Figure 2 It is a structural schematic diagram of a compressor state monitoring device provided by the present invention;

[0039] Figure 3 It is a structural schematic diagram of a compressor state monitoring device provided by the present invention;

[0040] Figure 4 is a flow chart of another compressor state monitoring method provided by the present invention;

[0041] Figure 5 It is a structural schematic diagram of the electronic device provided by the present invention. DETAILED DESCRIPTION

[0042] In order to make the purpose, technical solution and advantages of the present invention clearer, the technical solution of the present invention will be clearly and completely described below in conjunction with the drawings of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0043] Since there are no staff to conduct real-time inspections at unmanned natural gas stations, it is particularly important to ensure that abnormalities in the equipment in the station can be discovered in time, especially for compressor stations. The compressor is the most important large-scale equipment in the station. Timely detection of compressor abnormalities is of great significance to the safe operation of the gas pipeline. Existing compressor detection methods often rely on on-site detection by detection personnel, which cannot meet the needs of unmanned stations and cannot achieve real-time detection.

[0044] To this end, the present invention provides a compressor status monitoring method. Figure 1 Schematic diagram of the compressor status monitoring method provided by the present invention. Figure 1 As shown, the method comprises the following steps:

[0045] Step 110: Filter out the ambient noise data from the audio data of the compressor in operation to obtain the operation audio data of the compressor.

[0046] Here, the compressor is the compressor that needs to be monitored. The audio data of the compressor in operation refers to the noise data generated by the compressor in operation, which may include the noise generated by the airflow during the engine air filter intake and cylinder exhaust, the airflow noise caused by the cooling fan of the air cooler, the noise generated by the gas flowing in the pipeline, the pipeline surge noise, etc. The environmental noise data refers to the corresponding environmental noise of the compressor in operation, which may include wind, rain, thunder, animal calls, motor vehicle noise, human voices, metal knocking, walking and other environmental noise data.

[0047] In addition, the audio data may be mixed with environmental noise data, that is, the mixed environmental noise data may interfere with the state monitoring result of the compressor. Optionally, the environmental noise data in the audio data can be filtered out by a filter to obtain clean operating audio data, that is, the operating audio data does not include environmental noise data. Environmental noise data can also be filtered out from the audio data by a pre-trained model, which is not specifically limited in the embodiment of the present invention.

[0048] Step 120: Classify the operating audio data to obtain audio data of each part of the compressor.

[0049] Specifically, the operating audio data is clean data after filtering out environmental noise, but the operating audio data may include audio data from multiple parts of the compressor, such as the noise generated by the airflow during the engine air filter intake and cylinder exhaust, the airflow noise caused by the cooling fan of the air cooler, the noise generated by the gas flowing in the pipeline, the pipeline surge noise, etc. If the operating audio data includes audio data from multiple parts, the audio data from multiple parts need to be classified and stripped, so that the compressor status can be accurately detected based on the audio data from different parts.

[0050] Optionally, the running audio data may be input into a classification model, and the classification model may identify audio data corresponding to different parts in the running audio data, so as to realize the stripping of audio data of different parts in the running audio data.

[0051] Step 130: Determine initial status monitoring results of various parts of the compressor based on the audio data of various parts.

[0052] Specifically, the audio data of each part is used to characterize the operating status of each part of the compressor, that is, it can be understood that if the audio data of any part is abnormal, the operating status of the corresponding part of the compressor is abnormal; if the audio data of any part is normal, the operating status of the corresponding part of the compressor is normal.

[0053] Optionally, the compressor operating states corresponding to the audio data of different parts may be stored in a database, and then the compressor operating states may be matched in the database according to the categories corresponding to the audio data of each part in step 120. A model trained by a machine learning algorithm may also be used to determine the compressor operating state corresponding to the operating audio data, which is not specifically limited in the embodiment of the present invention.

[0054] Step 140: Based on the temperature data of each part of the compressor in the running state, the initial state monitoring results of each part are corrected to determine the state monitoring results of each part of the compressor.

[0055] Specifically, in some cases, the state of a certain part of the compressor is abnormal, but it is not reflected in the audio data of the corresponding part, that is, the audio data of the corresponding part is normal. At this time, the initial state monitoring result of the corresponding part determined based on the audio data of the corresponding part will have deviations. Therefore, it is necessary to correct the initial state monitoring results of each part in combination with the temperature data of each part when the compressor is in operation.

[0056] The temperature data can be understood as the real-time temperature data of the compressor in operation. If the deviation between the current temperature and the temperature baseline is large, it indicates that the probability of the compressor being abnormal is high. The temperature baseline can be understood as the curve of the temperature change over time of the compressor under normal operating conditions.

[0057] The compressor state monitoring method provided by the embodiment of the present invention filters out environmental noise data from the audio data of the compressor in the operating state, and classifies the operating audio data from which the environmental noise data is filtered out to obtain audio data of each part, thereby being able to accurately and in real time determine the initial state monitoring results of each part of the compressor based on the audio data of each part, and further, after the initial state monitoring results of each part are determined based on the temperature data of each part, the state monitoring results of each part of the compressor can be accurately and in real time obtained.

[0058] Based on the above embodiment, the ambient noise data is filtered out from the audio data of the compressor in the running state to obtain the running audio data of the compressor, including:

[0059] Based on the noise filtering model, the environmental noise data is filtered out from the audio data to obtain the operation audio data of the compressor;

[0060] The noise filtering model is trained based on sample audio data mixed with environmental noise and corresponding sample running audio data.

[0061] Specifically, the noise filtering model is trained based on sample audio data mixed with environmental noise and corresponding sample running audio data. Optionally, the noise filtering model can learn the relationship between the sample running audio data and the environmental noise data in the sample audio data, and identify the sample running audio data and the environmental noise data from the sample audio data.

[0062] After the noise filtering model is trained, the audio data is input into the noise filtering model, thereby filtering the environmental noise data from the audio data to obtain the operating audio data of the compressor.

[0063] Optionally, the SVR (Support Vector Regression) algorithm can be used to train the I1 data set (wind sound) to obtain the SI1 model for identifying wind sound; train the I2 data set (rain sound) to obtain the SI2 model for identifying rain sound; train the I3 data set (thunder) to obtain the SI3 model for identifying thunder; train the I4 data set (animal calls) to obtain the SI4 model for identifying animal calls; train the I5 data set (motor vehicle noise) to obtain the SI5 model for identifying motor vehicle noise; train the I6 data set (human talking) to obtain the SI6 model for identifying human talking; train the I7 data set (metal knocking sound) to obtain the SI7 model for identifying metal knocking sound; and train the I8 data set (walking sound) to obtain the SI8 model for identifying walking sound.

[0064] According to the SVR and Wiener fusion algorithms, the above 8 types of environmental noise are removed respectively (the number of types of environmental noise can be determined according to the actual situation, that is, the environmental noise can be multiple):

[0065] Assume that the observations x[0], x[1], …, x[N-1] contain ambient noise x[n] = y[n] + v[n], where x[n] represents ambient noise and y[n] represents actual audio data. Use the information of x[n] to make a linear estimate of y[n] Make Minimum.

[0066] Order error Then the variance ε is E(e[n]) 2 , the error function can be further written as:

[0067]

[0068] Furthermore, SI1 can be used instead of Calculate e1, and so on, to filter out the ambient noise in the audio data.

[0069] Based on any of the above embodiments, determining the initial state monitoring results of each part of the compressor based on the audio data of each part includes:

[0070] Based on the condition monitoring model, the audio data of each part is applied to determine the initial condition monitoring results of each part of the compressor;

[0071] Among them, the state monitoring model is trained based on the sample audio data of each part and the sample initial state monitoring labels of the corresponding part.

[0072] Specifically, the sample audio data may include normal sample audio data and abnormal sample audio data. For normal sample audio data, the corresponding sample initial state monitoring label is normal, and the corresponding label may be 1. For abnormal sample audio data, the corresponding sample initial state monitoring label is abnormal, and the corresponding label may be 0.

[0073] After the state monitoring model is trained, the audio data of each part is input into the state monitoring model to obtain the initial state monitoring results of each part of the compressor output by the state monitoring model.

[0074] Based on any of the above embodiments, based on the temperature data of each part of the compressor in the running state, the initial state monitoring results of each part are corrected to determine the state monitoring results of each part of the compressor, including:

[0075] The correction weight is determined based on the temperature data of each part and the temperature baseline; the temperature baseline refers to the curve of the compressor temperature changing over time under normal working conditions;

[0076] Based on the correction weight, the initial status monitoring results of each part are corrected to determine the status monitoring results of each part of the compressor.

[0077] Specifically, the temperature baseline refers to the curve of the temperature change of the compressor over time under normal operating conditions, which can be determined based on the historical temperature data in the temperature database. When the initial state monitoring results of each part are normal, the temperature baseline under normal conditions can be used to determine the deviation between the temperature data at the current moment and the temperature baseline under normal conditions, and the correction weight is determined based on the deviation. The larger the deviation, the smaller the correction weight. Similarly, when the initial state monitoring results of each part are abnormal, the temperature baseline under abnormal conditions can be used to determine the deviation between the temperature data at the previous moment and the temperature baseline under abnormal conditions, and the correction weight is determined based on the deviation. The larger the deviation, the smaller the correction weight.

[0078] After determining the correction weight, the correction weight can be multiplied by the initial state monitoring result of each part (the initial state monitoring result can be any value in the range of 0 to 1, the closer the value is to 0, the greater the probability that the compressor state is abnormal, and the closer the value is to 1, the greater the probability that it is normal) to obtain the state monitoring results of each part of the compressor.

[0079] Based on any of the above embodiments, based on the temperature data of each part of the compressor in the running state, the initial state monitoring results of each part are corrected to determine the state monitoring results of each part of the compressor, and then the following is further included:

[0080] If the status monitoring result of any part is less than the threshold, the compressor status is marked as high-risk;

[0081] After marking the compressor status as high-risk, determine whether the status monitoring results of the corresponding parts of the audio data for the consecutive predicted times are greater than or equal to the threshold. If so, the compressor is marked from high-risk to normal; if not, an early warning prompt is issued.

[0082] Specifically, the status monitoring result of each part can be any value in the range of 0 to 1. The closer the value is to 0, the greater the probability of abnormal compressor status, and the closer the value is to 1, the greater the probability of normal compressor status. If the status monitoring result of any part is less than the threshold value (such as less than 0.5), it indicates that the probability of abnormal status of the corresponding part is high, and the compressor status can be marked as high risk.

[0083] After marking the compressor status as high-risk, the preset number of consecutive audio data (such as 5 consecutive audio data) can be obtained, and it is determined whether the corresponding part status monitoring results of the preset number of consecutive audio data are greater than or equal to the threshold. If so, it indicates that the compressor can be removed from the high-risk state, and the compressor is marked from high-risk to normal. If not, an early warning prompt is issued.

[0084] It is understandable that when an early warning is issued, the linkage platform can automatically focus the camera on the abnormal part of the compressor, reminding the control room manager to arrange for professionals to manually re-inspect the compressor at the unmanned station.

[0085] Based on any of the above embodiments, based on the temperature data of each part of the compressor in the running state, the initial state monitoring results of each part are corrected to determine the state monitoring results of each part of the compressor, and then the following is further included:

[0086] If the status monitoring result of any part is abnormal, the camera is controlled to focus on the corresponding abnormal part.

[0087] Specifically, if the status monitoring result of any part is abnormal, the camera is controlled to focus on the corresponding abnormal part, so that a video or image of the corresponding abnormal part can be captured and transmitted back to the server, so that relevant personnel can obtain abnormal-related information in time through the transmitted video or image, and take corresponding measures in time.

[0088] Based on any of the above embodiments, the method further includes:

[0089] If it is detected that the methane concentration within the preset range of the compressor is greater than the preset concentration, an early warning prompt will be issued.

[0090] Specifically, considering the problem of natural gas leakage caused by compressor failure, the embodiment of the present invention will detect the methane concentration within the preset range of the compressor in real time. If the methane concentration is greater than the preset concentration, it indicates that there is a natural gas leak, and an early warning prompt will be issued. Optionally, when issuing an early warning prompt, the methane concentration can be prompted to provide a reference for the station maintenance strategy for relevant personnel, so as to avoid personal danger caused by relevant personnel rushing into the station to repair the compressor without knowing the on-site methane concentration.

[0091] The compressor state monitoring device provided by the present invention is described below. The compressor state monitoring device described below and the compressor state monitoring method described above can be referenced to each other.

[0092] Based on any of the above embodiments, the present invention further provides a compressor state monitoring device, such as Figure 2 As shown, the device comprises:

[0093] A denoising unit 210, configured to filter out ambient noise data from the audio data of the compressor in operation to obtain the operating audio data of the compressor;

[0094] The classification unit 220 is used to classify the operation audio data to obtain audio data of each part of the compressor;

[0095] A prediction unit 230, configured to determine initial state monitoring results of various parts of the compressor based on the audio data of the various parts;

[0096] The correction unit 240 is used to correct the initial state monitoring results of each part based on the temperature data of each part of the compressor in the running state, and determine the state monitoring results of each part of the compressor.

[0097] Based on any of the above embodiments, the present invention also provides a compressor status monitoring system, such as Figure 3 As shown, the system includes:

[0098] The compressor status monitoring device, audio monitoring device, temperature monitoring device, video monitoring device and gas leakage detection device as described above;

[0099] The audio monitoring device is used to collect audio data of the compressor when it is in operation, the temperature monitoring device is used to collect temperature data of various parts of the compressor when it is in operation, the video monitoring device is used to monitor the compressor via video, and the gas leakage detection device is used to detect the methane concentration of the compressor within a preset range.

[0100] Among them, the audio monitoring device may include a cross-shaped air sonar module, which is used to collect audio data of the compressor in operation. The temperature monitoring device can use infrared temperature measurement to measure the temperature in real time and establish a temperature database of the compressor shutdown part. The video monitoring device may include a visible light camera, an infrared camera, and a multi-degree-of-freedom pan-tilt head to monitor the compressor in real time, and work in conjunction with the audio monitoring device and the temperature monitoring device to achieve automatic focusing of the camera on the accident point under the rotation of the pan-tilt head. The gas leak detection device is used to detect the methane concentration near the compressor in real time. When the compressor fails and causes natural gas leakage, the gas leak detection device will transmit the concentration of natural gas at the compressor back to the control room in real time, providing a reference for the repair personnel to enter the station for maintenance strategy, so as to avoid personal danger caused by the repair personnel rushing into the station to repair the compressor without knowing the methane concentration on site.

[0101] In addition, the compressor status monitoring device is used to filter the audio data, remove environmental noise, and obtain real operating audio data; compare the measured temperature data with the historical temperature curve to analyze whether the current temperature is within the normal acceptable range; automatically control the rotation of the pan-tilt head according to the set rotation angle and rotation speed, and focus on different points of the compressor to monitor the compressor; if the status monitoring result is abnormal, the pan-tilt head is automatically rotated to the direction of the abnormal point, and the lens is focused on the abnormal point of the compressor; the measured operating audio data, temperature data, etc. are stored in the database module.

[0102] Figure 4 is a flow chart of another compressor state monitoring method provided by the present invention, such as Figure 4 As shown, the audio data of the compressor in operation is collected, and the noise filtering model is used to filter the ambient noise from the audio data to obtain the operation audio data; then, the classification model is used to classify and strip the operation audio data to obtain the audio data of each part. After the audio data of each part is obtained, the initial state monitoring results of each part of the compressor are determined by applying the audio data of each part based on the state monitoring model.

[0103] Next, based on the temperature data of each part and the temperature baseline, the correction weight is determined, and based on the correction weight, the initial state monitoring results of each part are corrected to determine the state monitoring results of each part of the compressor. Among them, the temperature baseline refers to the curve of the temperature change over time of the compressor under normal working conditions.

[0104] When the status monitoring result of any part is less than 0.5, the compressor status is marked as high-risk. After the compressor status is marked as high-risk, compare the status monitoring result of the corresponding part of the audio data collected in the next measurement batch to see if it is less than 0.5. If not, it indicates that the result of the next measurement batch is normal, then continue to collect audio data corresponding to the next measurement batch. If the status monitoring results of the corresponding parts of 5 consecutive batches are less than or equal to 0.5, the compressor is marked from high-risk to normal.

[0105] In addition, if the status monitoring result of any part is abnormal, the camera is controlled to focus on the corresponding abnormal part. If the methane concentration within the preset range of the compressor is detected to be greater than the preset concentration, an early warning prompt is issued.

[0106] Figure 5 is a schematic diagram of the structure of the electronic device provided by the present invention, such as Figure 5 As shown, the electronic device may include: a processor 510, a memory 520, a communications interface 530 and a communications bus 540, wherein the processor 510, the memory 520 and the communications interface 530 communicate with each other through the communications bus 540. The processor 510 may call the logic instructions in the memory 520 to execute the compressor state monitoring method, which includes: filtering the environmental noise data from the audio data of the compressor in the running state to obtain the running audio data of the compressor; classifying the running audio data to obtain the audio data of each part of the compressor; determining the initial state monitoring result of each part of the compressor based on the audio data of each part; and correcting the initial state monitoring result of each part based on the temperature data of each part of the compressor in the running state to determine the state monitoring result of each part of the compressor.

[0107] In addition, the logic instructions in the above-mentioned memory 520 can be implemented in the form of software functional units and can be stored in a computer-readable storage medium when sold or used as an independent product. Based on such an understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art or the part of the technical solution, can be embodied in the form of a software product, and the computer software product is stored in a storage medium, including a number of instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), disk or optical disk and other media that can store program codes.

[0108] On the other hand, the present invention also provides a computer program product, which includes a computer program stored on a non-transitory computer-readable storage medium, and the computer program includes program instructions. When the program instructions are executed by a computer, the computer can execute the compressor state monitoring method provided by the above-mentioned methods, and the method includes: filtering out environmental noise data from the audio data of the compressor in an operating state to obtain the operating audio data of the compressor; classifying the operating audio data to obtain audio data of various parts of the compressor; determining the initial state monitoring results of various parts of the compressor based on the audio data of each part; based on the temperature data of each part of the compressor in an operating state, correcting the initial state monitoring results of each part to determine the state monitoring results of each part of the compressor.

[0109] On the other hand, the present invention also provides a non-transitory computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, is implemented to execute the above-mentioned compressor state monitoring methods, the methods comprising: filtering out environmental noise data from the audio data of the compressor in an operating state to obtain the operating audio data of the compressor; classifying the operating audio data to obtain audio data of each part of the compressor; determining initial state monitoring results of each part of the compressor based on the audio data of each part; and correcting the initial state monitoring results of each part based on the temperature data of each part of the compressor in an operating state to determine the state monitoring results of each part of the compressor.

[0110] The device embodiments described above are merely illustrative, wherein the units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, they may be located in one place, or they may be distributed on multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the scheme of this embodiment. Ordinary technicians in this field can understand and implement it without paying creative labor.

[0111] Through the description of the above implementation methods, those skilled in the art can clearly understand that each implementation method can be implemented by means of software plus a necessary general hardware platform, and of course, it can also be implemented by hardware. Based on this understanding, the above technical solution is essentially or the part that contributes to the prior art can be embodied in the form of a software product, and the computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, a disk, an optical disk, etc., including a number of instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods described in each embodiment or some parts of the embodiments.

[0112] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A compressor status monitoring method, It is characterized in that include: filtering out environmental noise data from audio data of the compressor in operation to obtain operation audio data of the compressor; Classifying the operating audio data to obtain audio data of various parts of the compressor; Determining initial state monitoring results of various parts of the compressor based on the audio data of various parts; Based on the temperature data of each part of the compressor in the running state, the initial state monitoring results of each part are corrected to determine the state monitoring results of each part of the compressor.

2. The compressor status monitoring method according to claim 1, It is characterized in that The filtering out the ambient noise data from the audio data of the compressor in the running state to obtain the running audio data of the compressor includes: Based on the noise filtering model, filtering the environmental noise data from the audio data to obtain the operating audio data of the compressor; The noise filtering model is trained based on sample audio data mixed with environmental noise and corresponding sample running audio data.

3. The compressor status monitoring method according to claim 1, It is characterized in that The determining of the initial state monitoring results of each part of the compressor based on the audio data of each part includes: Based on the state monitoring model, the initial state monitoring results of each part of the compressor are determined by applying the audio data of each part; The state monitoring model is trained based on sample audio data of each part and sample initial state monitoring labels of the corresponding part.

4. The compressor status monitoring method according to claim 1, It is characterized in that The method of correcting the initial state monitoring results of each part based on the temperature data of each part of the compressor in the running state and determining the state monitoring results of each part of the compressor includes: Determine the correction weight based on the temperature data of each part and the temperature baseline; the temperature baseline refers to the curve of the temperature change of the compressor over time under normal working conditions; Based on the correction weights, the initial status monitoring results of each part are corrected to determine the status monitoring results of each part of the compressor.

5. The compressor status monitoring method according to claim 1, It is characterized in that Based on the temperature data of each part of the compressor in the running state, the initial state monitoring results of each part are corrected to determine the state monitoring results of each part of the compressor, and then the following is further included: If the monitoring result of any part status is less than the threshold, the compressor status is marked as high risk; After marking the compressor status as high-risk, determine whether the status monitoring results of the corresponding parts of the audio data for consecutive predicted times are greater than or equal to the threshold. If so, mark the compressor from high-risk to normal; if not, issue an early warning prompt.

6. The compressor status monitoring method according to claim 1, It is characterized in that Based on the temperature data of each part of the compressor in the running state, the initial state monitoring results of each part are corrected to determine the state monitoring results of each part of the compressor, and then the following is further included: If the status monitoring result of any part is abnormal, the camera is controlled to focus on the corresponding abnormal part.

7. The compressor status monitoring method according to claim 1, It is characterized in that The method further comprises: If it is detected that the methane concentration within the preset range of the compressor is greater than the preset concentration, an early warning prompt is issued.

8. A compressor status monitoring device, It is characterized in that include: A denoising unit, used to filter out environmental noise data from audio data of the compressor in operation to obtain operation audio data of the compressor; A classification unit, used for classifying the operation audio data to obtain audio data of each part of the compressor; A prediction unit, used to determine the initial state monitoring results of each part of the compressor based on the audio data of each part; The correction unit is used to correct the initial state monitoring results of each part based on the temperature data of each part of the compressor in the running state, so as to determine the state monitoring results of each part of the compressor.

9. A compressor status monitoring system, It is characterized in that include: The compressor status monitoring device, audio monitoring device, temperature monitoring device, video monitoring device and gas leakage detection device as claimed in claim 8; The audio monitoring device is used to collect audio data of the compressor when it is in operation, the temperature monitoring device is used to collect temperature data of various parts of the compressor when it is in operation, the video monitoring device is used to monitor the compressor via video, and the gas leakage detection device is used to detect the methane concentration of the compressor within a preset range.

10. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, It is characterized in that When the processor executes the computer program, the compressor state monitoring method according to any one of claims 1 to 7 is implemented.

Images