A visual diagnosis method and system for mechanical faults

Abstract

The present invention is applicable to the technical field of mechanical fault identification and provides a method and system for visual diagnosis of mechanical faults. The present invention performs visual diagnosis planning on multiple working vulnerable structures; selects the current diagnosis structure and performs shooting path planning and control; extracts multiple size area images and calculates multiple horizontal intensity differences and multiple vertical intensity differences; compares and selects the area sizes and calculates the structural roughness index; performs preliminary diagnosis analysis and issues a fault diagnosis alarm when a critical fault occurs. The system can perform diagnostic shooting, calculate the horizontal intensity difference and vertical intensity difference of images of different sizes according to the size area requirements, select multiple size selection values, calculate the structural roughness index, and perform preliminary diagnosis analysis of structural faults, thereby achieving preliminary diagnosis of faults before the fault occurs, avoiding problems such as reduced production efficiency, product quality fluctuations, increased maintenance costs, and extended downtime.

Description

Technical Field

[0001] The present invention belongs to the technical field of mechanical fault identification, and in particular relates to a visual diagnosis method and system for mechanical faults. Background Art

[0002] Mechanical fault identification is the process of monitoring and analyzing the operating status of a mechanical system to determine if an anomaly or fault exists, and to identify the fault's type, location, cause, and severity. Its core goal is to proactively identify potential problems through a data-driven approach, avoiding equipment downtime or accidents and improving system reliability and maintenance efficiency.

[0003] Visual diagnosis of mechanical faults is one of the common mechanical fault identification technologies.

[0004] In the existing technology, visual diagnosis of mechanical faults is usually a passive application mode. Fault diagnosis and identification can only be performed when a mechanical structure fails, causing the diagnosis process to lag behind the time point when the fault actually occurs. At this time, the mechanical equipment is often in a state of serious damage, which will not only lead to reduced production efficiency and fluctuations in product quality, but may also trigger a chain reaction, increase maintenance costs, extend downtime, etc., seriously affecting the normal production of the enterprise. Summary of the Invention

[0005] The purpose of the embodiments of the present invention is to provide a method and system for visual diagnosis of mechanical faults, aiming to solve the technical problems existing in the prior art mentioned in the background technology.

[0006] The embodiment of the present invention is implemented as follows:

[0007] A method for visually diagnosing mechanical faults, the method specifically comprising the following steps:

[0008] Identifying multiple vulnerable working structures on the working machine equipment, obtaining historical fault data, performing visual diagnosis planning on the multiple vulnerable working structures, and determining multiple structural diagnosis cycles;

[0009] According to the plurality of structural diagnosis cycles, a current diagnosis structure is selected from the plurality of working vulnerable structures, and a shooting path is planned and controlled to obtain a current shooting image;

[0010] Extracting multiple size area images from the current captured image according to a preset size area requirement, and calculating multiple horizontal intensity differences and multiple vertical intensity differences in the horizontal direction and the vertical direction;

[0011] Comparing and selecting the area size according to the plurality of horizontal intensity differences and the plurality of vertical intensity differences, obtaining a plurality of size selection values, and calculating the structural roughness index;

[0012] According to the structural roughness index, a pre-diagnosis analysis is performed on the current diagnosis structure to determine whether it is in a critical fault state, and a fault diagnosis alarm is issued when it is in a critical fault state.

[0013] As a further limitation of the technical solution of the embodiment of the present invention, the steps of determining multiple vulnerable working structures on the working machine equipment, obtaining historical fault data, performing visual diagnosis planning on the multiple vulnerable working structures, and determining multiple structural diagnosis cycles specifically include the following steps:

[0014] Identify multiple vulnerable working structures on working machinery and equipment;

[0015] Acquiring historical fault data related to a plurality of said vulnerable working structures;

[0016] Analyzing the historical failure data to calculate the average failure frequencies corresponding to the plurality of vulnerable working structures;

[0017] Obtaining a working cycle of the mechanical equipment;

[0018] According to the equipment working cycle and the plurality of average failure frequencies, a visual diagnosis plan is performed on the plurality of working vulnerable structures to determine a plurality of structural diagnosis cycles.

[0019] As a further limitation of the technical solution of the embodiment of the present invention, the selecting a current diagnosis structure from the multiple working vulnerable structures according to the multiple structural diagnosis cycles, performing shooting path planning and control, and obtaining the current shot image specifically include the following steps:

[0020] Performing diagnostic matching according to the plurality of structural diagnostic cycles, and selecting a current diagnostic structure from the plurality of working vulnerable structures;

[0021] Performing binocular photography on the working mechanical equipment to obtain binocular photography data;

[0022] Perform diagnostic shooting planning based on the binocular shooting data and generate a diagnostic shooting path;

[0023] According to the diagnostic shooting path, diagnostic shooting control is performed to obtain the current shooting image.

[0024] As a further limitation of the technical solution of the embodiment of the present invention, extracting multiple size area images from the current captured image according to the preset size area requirement, and calculating multiple horizontal intensity differences and multiple vertical intensity differences in the horizontal direction and the vertical direction specifically includes the following steps:

[0025] Extract multiple size area images from the current captured image according to the preset size area requirements;

[0026] Performing grayscale processing on the plurality of size region images to generate a plurality of size grayscale images;

[0027] Calculating horizontal intensity differences corresponding to a plurality of grayscale images of the sizes in a horizontal direction;

[0028] According to the vertical direction, vertical intensity differences corresponding to the plurality of grayscale images of the sizes are calculated.

[0029] As a further limitation of the technical solution of the embodiment of the present invention, the calculation formula of the multiple horizontal intensity differences is:

[0030] ;

[0031] in, represents the horizontal intensity difference, Represents the size area requirement, Pixel Gray value at ;

[0032] The calculation formula for the multiple vertical strength differences is:

[0033] ;

[0034] in, Represents the vertical intensity difference.

[0035] As a further limitation of the technical solution of the embodiment of the present invention, the comparing and selecting the region size according to the multiple horizontal intensity differences and the multiple vertical intensity differences, obtaining multiple size selection values, and calculating the structural roughness index specifically include the following steps:

[0036] Obtaining basic size data of the currently captured image;

[0037] Determine the size selection criteria;

[0038] According to the size selection condition, the plurality of horizontal intensity difference values ​​and the plurality of vertical intensity difference values ​​are compared respectively to obtain a plurality of size selection values;

[0039] A structural roughness index is calculated based on the basic size data and a plurality of size selection values.

[0040] As a further limitation of the technical solution of the embodiment of the present invention, the calculation formula of the structural roughness index is:

[0041] ;

[0042] in, is the structural roughness index, For the corresponding conditions or When the maximum value The value of is the image width of the current captured image, The image height of the current captured image.

[0043] As a further limitation of the technical solution of the embodiment of the present invention, the pre-diagnosis analysis of the current diagnosis structure based on the structural roughness index to determine whether it is in a critical fault state, and issuing a fault diagnosis alarm when it is in a critical fault state specifically includes the following steps:

[0044] a critical roughness index matching the current diagnostic structure;

[0045] comparing the structural roughness index with the critical roughness index to determine whether the current diagnosis structure is in a critical fault state;

[0046] When the current diagnosis structure is in a critical fault, generating a diagnosis alarm signal;

[0047] According to the diagnosis alarm signal, a fault diagnosis alarm is performed.

[0048] A visual diagnosis system for mechanical faults, comprising a visual diagnosis planning module, a shooting planning control module, an intensity difference calculation module, a roughness index calculation module, and a pre-diagnosis processing module, wherein:

[0049] A visual diagnosis planning module is used to identify multiple vulnerable working structures on the working machine equipment, obtain historical fault data, perform visual diagnosis planning on the multiple vulnerable working structures, and determine multiple structural diagnosis cycles;

[0050] a shooting planning control module, configured to select a current diagnosis structure from the plurality of working vulnerable structures according to the plurality of structural diagnosis cycles, and perform shooting path planning and control to obtain a current shooting image;

[0051] An intensity difference calculation module is used to extract multiple size area images from the current captured image according to a preset size area requirement, and calculate multiple horizontal intensity differences and multiple vertical intensity differences in the horizontal direction and the vertical direction;

[0052] a roughness index calculation module, configured to compare and select region sizes according to the plurality of horizontal intensity differences and the plurality of vertical intensity differences, obtain a plurality of size selection values, and calculate a structural roughness index;

[0053] The pre-diagnosis processing module is used to perform pre-diagnosis analysis on the current diagnosis structure according to the structural roughness index, determine whether it is in a critical fault state, and issue a fault diagnosis alarm when it is in a critical fault state.

[0054] As a further limitation of the technical solution of the embodiment of the present invention, the intensity difference calculation module specifically includes:

[0055] An image extraction unit is used to extract multiple size area images from the current captured image according to a preset size area requirement;

[0056] A grayscale processing unit, configured to perform grayscale processing on the plurality of size region images to generate a plurality of size grayscale images;

[0057] a horizontal intensity difference calculation unit, configured to calculate horizontal intensity differences corresponding to the plurality of grayscale images of the sizes in a horizontal direction;

[0058] The vertical intensity difference calculation unit is used to calculate the vertical intensity differences corresponding to the plurality of grayscale images of the sizes in a vertical direction.

[0059] Compared with the prior art, the present invention has the following beneficial effects:

[0060] The embodiment of the present invention performs visual diagnostic planning for multiple vulnerable working structures; selects the current diagnostic structure and performs shooting path planning and control; extracts images of multiple size regions and calculates multiple horizontal intensity differences and multiple vertical intensity differences; compares and selects region sizes and calculates a structural roughness index; performs pre-diagnostic analysis, and issues a fault diagnosis alarm in the event of a critical fault. The system can perform diagnostic shooting, calculate the horizontal and vertical intensity differences of images of different sizes based on size region requirements, then select multiple size selection values, calculate the structural roughness index, and perform pre-diagnostic analysis of structural faults. This allows for pre-diagnosis of faults before they occur, avoiding problems such as decreased production efficiency, fluctuating product quality, increased maintenance costs, and extended downtime. BRIEF DESCRIPTION OF THE DRAWINGS

[0061] Figure 1 A flowchart of a method for visually diagnosing mechanical faults according to an embodiment of the present invention is shown;

[0062] Figure 2 A flowchart of visual diagnosis planning in a method provided by an embodiment of the present invention is shown;

[0063] Figure 3 A flowchart of shooting path planning control in the method provided by an embodiment of the present invention is shown;

[0064] Figure 4 A flowchart of calculating multiple horizontal intensity differences and multiple vertical intensity differences in the method provided by an embodiment of the present invention is shown;

[0065] Figure 5A flowchart of calculating the structural roughness index in the method provided by an embodiment of the present invention is shown;

[0066] Figure 6 A flowchart of performing pre-diagnostic analysis in the method provided by an embodiment of the present invention is shown;

[0067] Figure 7 The following is a diagram showing the application architecture of a visual diagnosis system for mechanical faults provided by an embodiment of the present invention;

[0068] Figure 8 The figure shows a structural block diagram of an intensity difference calculation module in a system provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0069] In order to make the purpose, technical solutions and advantages of the present invention more clearly understood, the present invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0070] It is understandable that in the existing technology, visual diagnosis of mechanical faults is usually a passive application mode. Fault diagnosis and identification can only be performed when a fault occurs in the mechanical structure, causing the diagnosis process to lag behind the time point when the fault actually occurs. At this time, the mechanical equipment is often in a state of serious damage, which will not only lead to a decrease in production efficiency and fluctuations in product quality, but may also trigger a chain reaction, increase maintenance costs, extend downtime, etc., seriously affecting the normal production of the enterprise.

[0071] To solve the above problems, an embodiment of the present invention discloses a visual diagnosis method and system for mechanical faults. The method determines multiple vulnerable working structures on a working mechanical device and obtains historical fault data, performs visual diagnosis planning on the multiple vulnerable working structures, and determines multiple structural diagnosis cycles. According to the multiple structural diagnosis cycles, a current diagnosis structure is selected from the multiple vulnerable working structures, and shooting path planning and control are performed to obtain a current shot image. According to preset size area requirements, multiple size area images are extracted from the current shot image, and multiple horizontal intensity differences and multiple vertical intensity differences are calculated in the horizontal and vertical directions. According to the multiple horizontal intensity differences and multiple vertical intensity differences, the area size is compared and selected, multiple size selection values ​​are obtained, and a structural roughness index is calculated. According to the structural roughness index, a preliminary diagnosis analysis is performed on the current diagnosis structure to determine whether it is in a critical fault, and a fault diagnosis alarm is issued when it is in a critical fault. It can perform diagnostic shooting, calculate the horizontal and vertical intensity differences of images of different sizes according to size area requirements, then select multiple size selection values, calculate the structural roughness index, and perform pre-diagnosis analysis of structural faults, thereby achieving pre-diagnosis of faults before they occur, avoiding problems such as reduced production efficiency, fluctuations in product quality, increased maintenance costs, and extended downtime.

[0072] Specifically, Figure 1 The flowchart of the visual diagnosis method of mechanical faults provided by an embodiment of the present invention is shown.

[0073] In a preferred embodiment of the present invention, a method for visually diagnosing mechanical failures comprises the following steps:

[0074] Step S101: determining a plurality of vulnerable working structures on a working machine, obtaining historical fault data, performing visual diagnosis planning on the plurality of vulnerable working structures, and determining a plurality of structural diagnosis cycles.

[0075] In an embodiment of the present invention, multiple vulnerable working structures on the working mechanical equipment are determined, and historical failure data about the multiple vulnerable working structures are obtained. By analyzing the historical failure data, the historical record period and the number of historical failures of the multiple vulnerable working structures are determined. Then, based on the historical record period and the multiple historical failure numbers, the average failure frequencies corresponding to the multiple vulnerable working structures are calculated, and the equipment working cycle of the working mechanical equipment is obtained. Taking the equipment working cycle as the basic cycle unit, visual diagnosis planning is performed on the multiple vulnerable working structures based on the multiple average failure frequencies, and multiple structure diagnosis cycles are determined.

[0076] It can be understood that the equipment working cycle is the period of time when the working mechanical equipment goes from the working state to the idle state under the regular working arrangement.

[0077] It can be understood that each structural diagnosis cycle is composed of one or more equipment working cycles; the greater the average failure frequency, the shorter the structural diagnosis cycle of the corresponding vulnerable structure.

[0078] Specifically, Figure 2 A flowchart of visual diagnosis planning in the method provided by an embodiment of the present invention is shown.

[0079] In another preferred embodiment of the present invention, the steps of determining multiple vulnerable working structures on a working machine and obtaining historical fault data, performing visual diagnosis planning on the multiple vulnerable working structures, and determining multiple structural diagnosis cycles specifically include the following steps:

[0080] Step S1011: Determine multiple vulnerable working structures on the working machine equipment.

[0081] Step S1012: Acquire historical fault data related to a plurality of the vulnerable working structures.

[0082] Step S1013: Analyze the historical fault data and calculate the average fault frequencies corresponding to the plurality of vulnerable working structures.

[0083] Step S1014: Acquire the equipment working cycle of the working mechanical equipment.

[0084] Step S1015: Perform visual diagnosis planning on the multiple vulnerable structures according to the equipment working cycle and the multiple average failure frequencies, and determine multiple structural diagnosis cycles.

[0085] Furthermore, the visual diagnosis method for mechanical failures further includes the following steps:

[0086] Step S102 : selecting a current diagnosis structure from the plurality of working vulnerable structures according to the plurality of structural diagnosis cycles, performing shooting path planning and control, and acquiring a current shooting image.

[0087] In an embodiment of the present invention, when the working mechanical equipment is in an idle state, the current diagnostic cycle corresponding to this time is selected from multiple structural diagnostic cycles, and according to the current diagnostic cycle, the corresponding current diagnostic structure is selected from multiple working vulnerable structures, and binocular shooting data is obtained by performing binocular shooting on the working mechanical equipment, and then the binocular shooting data is identified to obtain the structural position of the current diagnostic structure and the initial position of the diagnostic camera. Between the initial position and the structural position, path planning is performed to avoid obstacle collision interference, and a diagnostic shooting path is generated. Then, according to the diagnostic shooting path, the diagnostic camera is controlled to perform diagnostic shooting, so that after the diagnostic camera moves close to the current diagnostic structure, the current diagnostic structure is diagnostically photographed to obtain the current shot image.

[0088] Specifically, Figure 3 A flowchart of shooting path planning control in the method provided by an embodiment of the present invention is shown.

[0089] In another preferred embodiment of the present invention, selecting a current diagnosis structure from a plurality of working vulnerable structures according to a plurality of structural diagnosis cycles, performing shooting path planning and control, and obtaining a current shot image specifically include the following steps:

[0090] Step S1021 : performing diagnostic matching according to the plurality of structural diagnostic cycles, and selecting a current diagnosis structure from the plurality of working vulnerable structures.

[0091] Step S1022: perform binocular photography on the working mechanical equipment to obtain binocular photography data.

[0092] Step S1023: Perform diagnostic shooting planning based on the binocular shooting data to generate a diagnostic shooting path.

[0093] Step S1024: perform diagnostic shooting control according to the diagnostic shooting path to obtain the current shot image.

[0094] Furthermore, the visual diagnosis method for mechanical failures further includes the following steps:

[0095] Step S103 : extracting multiple size area images from the current captured image according to a preset size area requirement, and calculating multiple horizontal intensity differences and multiple vertical intensity differences in the horizontal direction and the vertical direction.

[0096] In an embodiment of the present invention, according to a preset size region requirement, multiple size region images are extracted from the current captured image, and then the multiple size region images are grayscaled to generate multiple size grayscale images. Then, intensity differences of the multiple size grayscale images are calculated in the horizontal and vertical directions to obtain horizontal intensity differences and vertical intensity differences corresponding to the multiple size grayscale images. Specifically, the calculation formulas for the multiple horizontal intensity differences are:

[0097] ;

[0098] in, represents the horizontal intensity difference, Represents the size area requirement, Pixel Gray value at ;

[0099] The calculation formula for multiple vertical intensity differences is:

[0100] ;

[0101] in, Represents the vertical intensity difference.

[0102] It is understood that in the embodiment of the present invention, the size area requirement is the size of the extracted size area image, which needs to meet .

[0103] Specifically, Figure 4 A flow chart of calculating multiple horizontal intensity differences and multiple vertical intensity differences in the method provided by an embodiment of the present invention is shown.

[0104] In another preferred embodiment of the present invention, extracting multiple size area images from the current captured image according to a preset size area requirement, and calculating multiple horizontal intensity differences and multiple vertical intensity differences in the horizontal and vertical directions specifically includes the following steps:

[0105] Step S1031 : extracting multiple size area images from the current captured image according to a preset size area requirement.

[0106] Step S1032: grayscale the plurality of size region images to generate a plurality of size grayscale images.

[0107] Step S1033: Calculate horizontal intensity differences corresponding to the plurality of grayscale images of the sizes in the horizontal direction.

[0108] Step S1034: Calculate vertical intensity differences corresponding to the plurality of grayscale images of the sizes in the vertical direction.

[0109] Furthermore, the visual diagnosis method for mechanical failures further includes the following steps:

[0110] Step S104: Compare and select the region size according to the multiple horizontal intensity difference values ​​and the multiple vertical intensity difference values, obtain multiple size selection values, and calculate the structural roughness index.

[0111] In the embodiment of the present invention, the basic size data of the current captured image is obtained and the size selection condition is determined. According to the size selection condition, multiple horizontal intensity differences and multiple vertical intensity differences are compared respectively, and the horizontal and vertical intensity differences are recorded. and When the maximum value The value of is obtained to obtain multiple size selection values, and then the structural roughness index is calculated according to the basic size data and the multiple size selection values. Specifically, the calculation formula of the structural roughness index is:

[0112] ;

[0113] in, is the structural roughness index, For the corresponding conditions or When the maximum value The value of is the image width of the current captured image, The image height of the current captured image.

[0114] Specifically, Figure 5 A flow chart of calculating the structural roughness index in the method provided by an embodiment of the present invention is shown.

[0115] In another preferred embodiment of the present invention, comparing and selecting the region size according to the plurality of horizontal intensity differences and the plurality of vertical intensity differences, obtaining a plurality of size selection values, and calculating the structural roughness index specifically include the following steps:

[0116] Step S1041: Obtain basic size data of the currently captured image.

[0117] Step S1042: Determine size selection conditions.

[0118] Step S1043: According to the size selection condition, compare the multiple horizontal intensity difference values ​​and the multiple vertical intensity difference values ​​respectively to obtain multiple size selection values.

[0119] Step S1044: Calculate a structural roughness index based on the basic size data and the multiple size selection values.

[0120] Furthermore, the visual diagnosis method for mechanical failures further includes the following steps:

[0121] Step S105 : performing a preliminary diagnosis analysis on the current diagnosis structure according to the structural roughness index to determine whether it is in a critical fault state, and issuing a fault diagnosis alarm if it is in a critical fault state.

[0122] In an embodiment of the present invention, the critical roughness index of the current diagnostic structure is matched from a preset critical index table, and by comparing the structural roughness index with the critical roughness index, it is determined whether the current diagnostic structure is in a critical fault state. Specifically, when the structural roughness index is greater than the critical roughness index, it indicates that the current diagnostic structure is in a critical state of imminent fault. At this time, it is determined that the current diagnostic structure is in a critical fault state, a diagnostic alarm signal is generated, and then a fault diagnosis alarm is performed according to the diagnostic alarm signal.

[0123] Specifically, Figure 6 A flow chart of performing preliminary diagnosis and analysis in the method provided by an embodiment of the present invention is shown.

[0124] In another preferred embodiment of the present invention, performing a preliminary diagnosis analysis on the current diagnosis structure based on the structural roughness index to determine whether it is in a critical fault state, and issuing a fault diagnosis alarm when it is in a critical fault state specifically includes the following steps:

[0125] Step S1051: Match the critical roughness index of the current diagnosis structure.

[0126] Step S1052: Compare the structural roughness index with the critical roughness index to determine whether the current diagnosis structure is in a critical fault state.

[0127] Step S1053: When the current diagnosis structure is in a critical fault state, a diagnosis alarm signal is generated.

[0128] Step S1054: Perform a fault diagnosis alarm according to the diagnosis alarm signal.

[0129] Further, Figure 7 The application architecture diagram of the visual diagnosis system for mechanical faults provided by an embodiment of the present invention is shown.

[0130] Specifically, in another preferred embodiment provided by the present invention, a visual diagnosis system for mechanical failures includes:

[0131] The visual diagnosis planning module 101 is used to determine multiple vulnerable working structures on the working machine equipment, obtain historical fault data, perform visual diagnosis planning on the multiple vulnerable working structures, and determine multiple structural diagnosis cycles.

[0132] In an embodiment of the present invention, the visual diagnosis planning module 101 determines multiple vulnerable working structures on the working mechanical equipment and obtains historical failure data about the multiple vulnerable working structures. By analyzing the historical failure data, the historical record period and the number of historical failures of the multiple vulnerable working structures are determined. Then, based on the historical record period and the multiple historical failure numbers, the average failure frequency corresponding to the multiple vulnerable working structures is calculated, and the equipment working cycle of the working mechanical equipment is obtained. With the equipment working cycle as the basic cycle unit, visual diagnosis planning is performed on the multiple vulnerable working structures based on the multiple average failure frequencies to determine multiple structure diagnosis cycles.

[0133] The shooting planning control module 102 is used to select a current diagnosis structure from the multiple working vulnerable structures according to the multiple structural diagnosis cycles, and perform shooting path planning and control to obtain a current shooting image.

[0134] In an embodiment of the present invention, when the working machinery and equipment is in an idle state, the shooting planning control module 102 selects the current diagnostic cycle corresponding to this time from multiple structural diagnostic cycles, and selects the corresponding current diagnostic structure from multiple working vulnerable structures according to the current diagnostic cycle, and obtains binocular shooting data by performing binocular shooting of the working machinery and equipment, and then identifies the binocular shooting data to obtain the structural position of the current diagnostic structure and the initial position of the diagnostic camera. Between the initial position and the structural position, a path planning is performed to avoid obstacle collision interference, and a diagnostic shooting path is generated. Then, according to the diagnostic shooting path, the diagnostic camera is controlled to perform diagnostic shooting, so that after the diagnostic camera moves close to the current diagnostic structure, the current diagnostic structure is diagnostically photographed to obtain the current shooting image.

[0135] The intensity difference calculation module 103 is used to extract multiple size area images from the current captured image according to a preset size area requirement, and calculate multiple horizontal intensity differences and multiple vertical intensity differences in the horizontal direction and the vertical direction.

[0136] In the embodiment of the present invention, the intensity difference calculation module 103 extracts multiple size region images from the current captured image according to the preset size region requirements, performs grayscale processing on the multiple size region images to generate multiple size grayscale images, and then calculates the intensity differences of the multiple size grayscale images in the horizontal direction and the vertical direction to obtain the horizontal intensity difference and the vertical intensity difference corresponding to the multiple size grayscale images. Specifically, the calculation formulas for the multiple horizontal intensity differences are:

[0137] ;

[0138] in, represents the horizontal intensity difference, Represents the size area requirement, Pixel Gray value at ;

[0139] The calculation formula for multiple vertical intensity differences is:

[0140] ;

[0141] in, Represents the vertical intensity difference.

[0142] Specifically, Figure 8 FIG. 1 shows a structural block diagram of the intensity difference calculation module 103 in the system provided by an embodiment of the present invention.

[0143] In another preferred embodiment of the present invention, the intensity difference calculation module 103 specifically includes:

[0144] The image extraction unit 1031 is configured to extract a plurality of size region images from the current captured image according to a preset size region requirement.

[0145] The grayscale processing unit 1032 is configured to perform grayscale processing on the plurality of size region images to generate a plurality of size grayscale images.

[0146] The horizontal intensity difference calculation unit 1033 is configured to calculate horizontal intensity differences corresponding to the plurality of grayscale images of the sizes in a horizontal direction.

[0147] The vertical intensity difference calculation unit 1034 is configured to calculate vertical intensity differences corresponding to the plurality of grayscale images of the sizes in a vertical direction.

[0148] Furthermore, the visual diagnosis system for mechanical failures also includes:

[0149] The roughness index calculation module 104 is configured to compare and select region sizes according to the plurality of horizontal intensity differences and the plurality of vertical intensity differences, obtain a plurality of size selection values, and calculate a structural roughness index.

[0150] In the embodiment of the present invention, the roughness index calculation module 104 obtains the basic size data of the current captured image and determines the size selection condition. According to the size selection condition, the roughness index calculation module 104 compares the multiple horizontal intensity differences and the multiple vertical intensity differences respectively, and records the horizontal and vertical intensity differences. and When the maximum value The value of is obtained to obtain multiple size selection values, and then the structural roughness index is calculated according to the basic size data and the multiple size selection values. Specifically, the calculation formula of the structural roughness index is:

[0151] ;

[0152] in, is the structural roughness index, For the corresponding conditions or When the maximum value The value of is the image width of the current captured image, The image height of the current captured image.

[0153] The pre-diagnosis processing module 105 is used to perform a pre-diagnosis analysis on the current diagnosis structure according to the structure roughness index, determine whether it is in a critical fault state, and issue a fault diagnosis alarm when it is in a critical fault state.

[0154] In an embodiment of the present invention, the pre-diagnosis processing module 105 matches the critical roughness index of the current diagnosis structure from a preset critical index table, and determines whether the current diagnosis structure is in a critical fault by comparing the structural roughness index with the critical roughness index. Specifically, when the structural roughness index is greater than the critical roughness index, it indicates that the current diagnosis structure is in a critical state of imminent fault. At this time, it is determined that the current diagnosis structure is in a critical fault, a diagnosis alarm signal is generated, and then a fault diagnosis alarm is performed according to the diagnosis alarm signal.

[0155] It should be understood that, although the various steps in the flow chart of each embodiment of the present invention are shown in sequence according to the indication of the arrows, these steps are not necessarily performed in sequence according to the order indicated by the arrows. Unless otherwise specified herein, the execution of these steps is not strictly limited in order, and these steps can be performed in other orders. Moreover, at least a portion of the steps in each embodiment may include a plurality of sub-steps or a plurality of stages, and these sub-steps or stages are not necessarily performed at the same time, but can be performed at different times, and the execution order of these sub-steps or stages is not necessarily performed in sequence, but can be performed in turn or alternately with at least a portion of other steps or sub-steps or stages of other steps.

[0156] Those skilled in the art will appreciate that all or part of the processes in the above-described method embodiments can be implemented by instructing the relevant hardware through a computer program. The program can be stored in a non-volatile computer-readable storage medium. When executed, the program can include the processes of the above-described method embodiments. Any reference to memory, storage, database, or other media used in the various embodiments provided herein may include non-volatile and / or volatile memory. Non-volatile memory may include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM), or flash memory. Volatile memory may include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in various forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link DRAM (SLDRAM), RAMbus direct RAM (RDRAM), direct RAMbus dynamic RAM (DRDRAM), and RAMbus dynamic RAM (RDRAM).

[0157] The above-described embodiments merely illustrate several implementations of the present invention, and while their descriptions are relatively specific and detailed, they should not be construed as limiting the scope of the present invention. It should be noted that a person skilled in the art would be able to make numerous variations and improvements without departing from the spirit of the present invention, all of which fall within the scope of protection of the present invention. Therefore, the scope of protection of the present invention shall be determined by the appended claims.

Claims

1. A visual diagnosis method for mechanical failure, characterized in that: The method specifically comprises the following steps: Identifying multiple vulnerable working structures on the working machine equipment, obtaining historical fault data, performing visual diagnosis planning on the multiple vulnerable working structures, and determining multiple structural diagnosis cycles; According to the plurality of structural diagnosis cycles, a current diagnosis structure is selected from the plurality of working vulnerable structures, and a shooting path is planned and controlled to obtain a current shooting image; Extracting multiple size area images from the current captured image according to a preset size area requirement, and calculating multiple horizontal intensity differences and multiple vertical intensity differences in the horizontal direction and the vertical direction; Comparing and selecting the area size according to the plurality of horizontal intensity differences and the plurality of vertical intensity differences, obtaining a plurality of size selection values, and calculating the structural roughness index; Performing a pre-diagnostic analysis on the current diagnosis structure according to the structural roughness index to determine whether it is in a critical fault state, and issuing a fault diagnosis alarm if it is in a critical fault state; The method of extracting multiple size area images from the current captured image according to the preset size area requirement and calculating multiple horizontal intensity differences and multiple vertical intensity differences in the horizontal direction and the vertical direction specifically includes the following steps: Extract multiple size area images from the current captured image according to the preset size area requirements; Performing grayscale processing on the plurality of size region images to generate a plurality of size grayscale images; Calculating horizontal intensity differences corresponding to a plurality of grayscale images of the sizes in a horizontal direction; Calculating vertical intensity differences corresponding to a plurality of grayscale images of the sizes in a vertical direction; The calculation formula for the multiple horizontal intensity differences is: ； in, represents the horizontal intensity difference, Represents the size area requirement, Pixel Gray value at ; The calculation formula for the multiple vertical strength differences is: ； in, represents the vertical intensity difference; The step of comparing and selecting the region size according to the plurality of horizontal intensity differences and the plurality of vertical intensity differences, obtaining a plurality of size selection values, and calculating the structure roughness index specifically includes the following steps: Obtaining basic size data of the currently captured image; Determine the size selection criteria; According to the size selection condition, the plurality of horizontal intensity difference values ​​and the plurality of vertical intensity difference values ​​are compared respectively to obtain a plurality of size selection values; Calculating a structural roughness index based on the basic size data and a plurality of size selection values; The calculation formula of the structural roughness index is: ； in, is the structural roughness index, For the corresponding conditions or When the maximum value The value of is the image width of the current captured image, The image height of the current captured image.

2. The visual diagnosis method for mechanical failure according to claim 1, characterized in that: The steps of determining multiple vulnerable working structures on the working machine equipment, obtaining historical fault data, performing visual diagnosis planning on the multiple vulnerable working structures, and determining multiple structural diagnosis cycles specifically include the following steps: Identify multiple vulnerable working structures on working machinery and equipment; Acquiring historical fault data related to a plurality of said vulnerable working structures; Analyzing the historical failure data to calculate the average failure frequencies corresponding to the plurality of vulnerable working structures; Obtaining a working cycle of the mechanical equipment; According to the equipment working cycle and the plurality of average failure frequencies, a visual diagnosis plan is performed on the plurality of working vulnerable structures to determine a plurality of structural diagnosis cycles.

3. The visual diagnosis method for mechanical failure according to claim 1, characterized in that: The step of selecting a current diagnosis structure from a plurality of working vulnerable structures according to the plurality of structural diagnosis cycles, and performing shooting path planning and control to obtain a current shot image specifically includes the following steps: Performing diagnostic matching according to the plurality of structural diagnostic cycles, and selecting a current diagnostic structure from the plurality of working vulnerable structures; Performing binocular photography on the working mechanical equipment to obtain binocular photography data; Perform diagnostic shooting planning based on the binocular shooting data and generate a diagnostic shooting path; According to the diagnostic shooting path, diagnostic shooting control is performed to obtain the current shooting image.

4. The visual diagnosis method for mechanical failure according to claim 1, characterized in that: The method of performing a pre-diagnosis analysis on the current diagnosis structure according to the structural roughness index to determine whether it is in a critical fault state and issuing a fault diagnosis alarm when it is in a critical fault state specifically includes the following steps: a critical roughness index matching the current diagnostic structure; comparing the structural roughness index with the critical roughness index to determine whether the current diagnosis structure is in a critical fault state; When the current diagnosis structure is in a critical fault, generating a diagnosis alarm signal; According to the diagnosis alarm signal, a fault diagnosis alarm is performed.

5. A visual diagnosis system for mechanical failures for executing the visual diagnosis method for mechanical failures according to any one of claims 1 to 4, characterized in that: The system includes a visual diagnosis planning module, a shooting planning control module, an intensity difference calculation module, a roughness index calculation module and a pre-diagnosis processing module, wherein: A visual diagnosis planning module is used to identify multiple vulnerable working structures on the working machine equipment, obtain historical fault data, perform visual diagnosis planning on the multiple vulnerable working structures, and determine multiple structural diagnosis cycles; a shooting planning control module, configured to select a current diagnosis structure from the plurality of working vulnerable structures according to the plurality of structural diagnosis cycles, and perform shooting path planning and control to obtain a current shooting image; An intensity difference calculation module is used to extract multiple size area images from the current captured image according to a preset size area requirement, and calculate multiple horizontal intensity differences and multiple vertical intensity differences in the horizontal direction and the vertical direction; a roughness index calculation module, configured to compare and select region sizes according to the plurality of horizontal intensity differences and the plurality of vertical intensity differences, obtain a plurality of size selection values, and calculate a structural roughness index; The pre-diagnosis processing module is used to perform pre-diagnosis analysis on the current diagnosis structure according to the structural roughness index, determine whether it is in a critical fault state, and issue a fault diagnosis alarm when it is in a critical fault state.

6. The visual diagnosis system for mechanical failure according to claim 5, characterized in that: The intensity difference calculation module specifically includes: An image extraction unit is used to extract multiple size area images from the current captured image according to a preset size area requirement; A grayscale processing unit, configured to perform grayscale processing on the plurality of size region images to generate a plurality of size grayscale images; a horizontal intensity difference calculation unit, configured to calculate horizontal intensity differences corresponding to the plurality of grayscale images of the sizes in a horizontal direction; The vertical intensity difference calculation unit is used to calculate the vertical intensity differences corresponding to the plurality of grayscale images of the sizes in a vertical direction.

Images