Industrial robot fault diagnosis and recovery system

Abstract

The invention discloses an industrial robot fault diagnosis and recovery system, which relates to the field of industrial robot fault diagnosis and recovery and comprises a data acquisition module, a maintenance personnel registration module, a data analysis module, a fault classification module and an early warning module. According to the method, the historical faults of the industrial robot are analyzed, the fault indication value sets corresponding to different fault types are constructed, and related parameters in the operation process of the industrial robot are monitored and analyzed in real time, so that the problems that in the prior art, an alarm can only be given when the industrial robot breaks down, and a fault source cannot be accurately positioned are solved; the maintenance cost of an enterprise is reduced, and the efficiency of the enterprise is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of industrial robot fault diagnosis and recovery, and in particular to an industrial robot fault diagnosis and recovery system. Background Art

[0002] With the continuous improvement of industrial automation, industrial robots are increasingly used in production lines. However, due to the long-term high-intensity work of robots, various failures are prone to occur, causing considerable trouble to the production line.

[0003] Most of the existing technologies for fault diagnosis of industrial robots are based on threshold diagnosis, which can only identify that the industrial robot has a fault, but cannot identify the specific fault type. Maintenance personnel are required to go to the site for further confirmation, and the time for this confirmation will cause huge losses to the enterprise. In addition, the existing technology cannot find suitable maintenance personnel according to the specific fault type, which wastes manpower and material resources.

[0004] Therefore, an industrial robot fault diagnosis and recovery system is introduced. Summary of the invention

[0005] In view of this, the present invention provides an industrial robot fault diagnosis and recovery system to solve the problems raised by the above background technology.

[0006] The purpose of the present invention can be achieved through the following technical solutions: An industrial robot fault diagnosis and recovery system, comprising:

[0007] Data collection module: extracts historical fault information of industrial robots, takes the time point of each failure of the robot as the starting point, and obtains abnormal data packets within the set time interval before each failure of the industrial robot;

[0008] The acquired abnormal data packets are classified according to different fault types, which include motor abnormality, communication abnormality and mechanical abnormality; the three groups of abnormal data packets include motor abnormality data packets, communication abnormality data packets and mechanical abnormality data packets; the motor abnormality data packets, communication abnormality data packets and mechanical abnormality data packets are set to correspond to a group of data sets respectively;

[0009] Maintenance personnel registration module: used to collect information corresponding to maintenance personnel; the information includes the distance between the maintenance personnel and the industrial robot, the maintenance success rate of the maintenance personnel for each fault type, and the maintenance personnel's educational level; the educational level includes high school and below, technical secondary school, junior college, and undergraduate;

[0010] Data analysis module: used to analyze each abnormal data packet corresponding to each fault type to obtain the fault indication value corresponding to each fault type; the fault indication value includes the motor fault indication value DJC, the communication fault indication value TXC and the mechanical fault indication value JXC; and send the fault indication value to the fault classification module;

[0011] Fault classification module: used to store each group of fault indication values ​​corresponding to the historical fault information of the industrial robot, and divide each group of fault indication values ​​into three abnormal fault sets according to the different fault types, namely, the motor abnormal fault set corresponding to the motor abnormality, the communication abnormal fault set corresponding to the communication abnormality, and the mechanical abnormal fault set corresponding to the mechanical abnormality;

[0012] If the absolute value of the difference between a certain group of fault indication values ​​within a set time interval during the operation of the industrial robot and a certain group of fault indication values ​​in the corresponding abnormal fault set is less than a preset threshold, then a warning signal of the fault type corresponding to the abnormal fault set is triggered; the warning signal includes a motor abnormality warning signal, a communication abnormality warning signal and a mechanical abnormality warning signal;

[0013] Early warning module: executes corresponding steps based on the corresponding early warning signals received.

[0014] In some embodiments, the motor abnormality data packet, the communication abnormality data packet, and the mechanical abnormality data packet are set to correspond to a set of data sets, specifically:

[0015] The data set of the motor abnormality data packet is the robot motor temperature value in the set time interval and the working environment temperature value of the industrial robot; the data set of the communication abnormality data packet is the communication rate value in the set time interval; the data set of the mechanical abnormality data packet is the image information of the industrial robot's mechanical arm and the mechanical arm acceleration value.

[0016] In some embodiments, the motor abnormality data packet corresponding to the motor abnormality fault type is analyzed, specifically:

[0017] S1: Get the motor temperature value at each time node in the set time interval, and draw it into a line graph according to the time series. Preset the motor temperature value threshold, construct the threshold line corresponding to the motor temperature threshold on the line graph, mark the closed area surrounded by the threshold line and the line graph, and statistically sum the areas of the closed areas above and below the threshold line. The obtained values ​​are recorded as M1 and M2 respectively. Use the formula Gets overworked than MJ;

[0018] S2: In the line graph, the sum of the time when the motor temperature value is above the threshold line is counted and recorded as the high-temperature working time. The high-temperature working time and the set time interval are calculated by ratio, and the obtained value is recorded as the high-temperature working ratio MK;

[0019] S3: Obtain the ambient temperature values ​​of the industrial robot at each time node in the set time interval, take the average as the ring average, preset the ring average threshold of the industrial robot, subtract the ring average threshold from the ring average, so as to obtain the environmental index, preset each environmental index interval corresponding to each group of ambient temperature scores, match the environmental index interval corresponding to the environmental index, so as to obtain the ambient temperature score TR corresponding to the environmental index; the larger the ambient temperature score TR, the more suitable the ambient temperature is for the industrial robot to work;

[0020] S4: The excess work ratio MJ, high temperature work ratio MK, and ambient temperature score TR are normalized and then input into the formula: The motor fault indication value DJC is obtained, where a1, a2 and a3 are the weight influence factors corresponding to the ambient temperature score TR, excess working ratio MJ and high temperature working ratio MK respectively.

[0021] In some embodiments, the communication abnormality data packet corresponding to the communication abnormality fault type is analyzed, specifically:

[0022] L1: Obtain the network delay data value at each time node in the set time interval, calculate the network delay mean value N1 using the mean formula, and extract the network delay data value peak value as the network peak value N2;

[0023] L2: Obtain the broadband utilization rate at each time node in the set time interval, and use the mean formula to calculate the broadband utilization rate at each time node in the set time interval to obtain the average broadband rate KD1; extract the maximum value of the broadband utilization rate, preset the broadband utilization rate threshold, and subtract the broadband utilization rate threshold from the maximum value of the broadband utilization rate to obtain the broadband overload value KD2;

[0024] L3: Extract the network delay mean threshold N1 from the preset database 阈值 , Network peak threshold N2 阈值 , width average rate KD1 阈值 And broadband overload rate threshold KD2 阈值 , the obtained network delay mean N1 and network delay mean threshold N1 阈值 , Network peak value N2, Network peak threshold N2 阈值 , width average rate KD1, width average rate KD1 阈值 , broadband overload value KD2, broadband overload rate threshold KD2 阈值 After normalization, enter the formula The communication fault indication value TXC is obtained, where b1, b2, b3, and b4 are weighted influencing factors corresponding to the network delay mean value N1, the network peak value N2, the bandwidth average rate KD1, and the broadband overload value KD2, respectively.

[0025] In some embodiments, the mechanical abnormality data packet corresponding to the mechanical abnormality fault type is analyzed, specifically:

[0026] D1: Acquire the image of the industrial robot's mechanical arm at a set time interval, and preprocess the image of the industrial robot's mechanical arm, including enhancement and noise reduction;

[0027] D2: Divide the obtained robot arm image into x sub-parts, and number the robot arm images of each sub-part, where the number is represented by i, i=1,2...x; obtain the original images corresponding to each sub-part of the robot arm from the database, compare the obtained robot arm images of each sub-part with the corresponding original images, and calculate based on the perceptual hash algorithm to obtain the similarity between the robot arm images of each sub-part and the corresponding original images, which is recorded as the similarity score XS i ;

[0028] D3: Obtain the acceleration value of the industrial robot arm at each time node in the set time interval, preset the acceleration threshold of the industrial robot arm, count the number of acceleration values ​​exceeding the acceleration threshold in the set time interval, record it as the excess value GF, calculate the acceleration data using the standard deviation formula, and obtain the excess value JL. After normalizing the excess value GF and excess value JL, enter them into the formula The acceleration index JSD is obtained, where v1 and v2 are the weighted influence factors corresponding to the super value GF and the wave value JL respectively;

[0029] D4: Extract the corresponding maintenance times of each sub-part of the industrial robot arm from the historical maintenance data of the industrial robot, calculate the ratio of the maintenance times of each sub-part of the industrial robot to the total maintenance times, and get the maintenance factor μ i , preset each group of maintenance factors μ i The interval corresponds to the maintenance influence weight of each group, and the maintenance factor μ i Match with the preset maintenance factor intervals to obtain the maintenance factor μ i The corresponding maintenance impact weight is used as the maintenance impact weight of each sub-part of the robot arm as the similarity score XS i The weight influence factor λ i .

[0030] D5: The obtained similarity score XS i , the acceleration index JSD normalized and the weight influence factors corresponding to each part of the robot are substituted into the formula The mechanical fault indication value JXC is obtained, where dj is the weight influence factor corresponding to the acceleration index JSD.

[0031] In some embodiments, based on the received communication abnormality warning signal, corresponding steps are performed, specifically:

[0032] Obtain the difference between the motor fault indication value and the motor fault indication value threshold, preset the intervals of the difference between the motor fault indication value and the motor fault indication value threshold to correspond to the motor danger level, and the danger level is divided into slight, general and severe; match the intervals corresponding to the difference between the motor fault indication value and the motor fault indication value threshold to obtain the motor danger level, and perform corresponding processing according to the different motor danger levels:

[0033] When the danger level is slight: different power ranges are preset based on the different danger levels corresponding to the motor cooling fan and the motor, and based on the power ranges corresponding to the slight danger levels preset for the motor cooling fan and the motor, the motor cooling fan power and the motor power are adjusted to the preset ranges;

[0034] When the danger level is general: Based on the power ranges corresponding to the general danger levels preset for the motor cooling fan and the motor, the motor cooling fan power and the motor power are adjusted to the preset ranges, and the target position to be reached by the current target industrial robot is sent to the mobile terminal of the maintenance personnel closest to the target position, and the motor danger level is packaged and sent together, and the staff will make a secondary judgment and analysis on whether it is necessary to stop the machine for maintenance or shut down for rest;

[0035] When the danger level is severe: immediately shut down the industrial robot and monitor the motor temperature in real time. When the motor temperature drops to a safe temperature range, obtain the cooling area closest to the current industrial robot and control the industrial robot to automatically move to the nearest cooling area.

[0036] In some embodiments, based on the received communication abnormality warning signal, corresponding steps are performed, specifically:

[0037] The image information at the communication interface is obtained and preprocessed. The ratio of the dust area in the image to the total area of ​​the interface is obtained by using graphic processing technology, which is recorded as the dust impact degree. A dust impact degree threshold is preset. When the dust impact degree is greater than the preset dust impact degree threshold, the cleaning staff closest to the current industrial robot is immediately obtained, and the cleaning instruction and the location of the industrial robot are sent to the cleaning staff's mobile terminal, and the cleaning staff performs the cleaning operation;

[0038] When the dust impact is less than the preset dust impact threshold, a circle is drawn with the current position of the industrial robot as the center and the specified area as the radius to obtain the distance between each maintenance personnel and the industrial robot, the repair success rate of the communication fault, and the type of education within the circle; the education score corresponding to each type of education is preset; the education type of each maintenance personnel is matched with the corresponding education score, so as to obtain the education score of each maintenance personnel;

[0039] The distance between each maintenance personnel and the industrial robot, the repair success rate of communication failures, and the educational level score are marked as C1, C2, and C3 respectively;

[0040] The distance C1 from the industrial robot, the repair success rate C2 of the communication failure, and the educational background type C3 of each maintenance personnel within the circle are normalized and entered into the formula: The comprehensive maintenance index Wt of each maintenance personnel is obtained, where w1, w2 and w3 are the weighted influencing factors corresponding to the distance C1 between the maintenance personnel and the industrial robot, the repair success rate C2 of the communication failure and the maintenance personnel's educational score C3, respectively. The maintenance personnel with the largest comprehensive maintenance index Wt are selected as the target maintenance personnel, and the industrial robot fault information and the industrial robot's location information are packaged and sent to the target maintenance personnel terminal, who performs maintenance operations.

[0041] In some embodiments, based on the received mechanical abnormality warning signal, corresponding steps are performed, specifically:

[0042] Draw a circle with the current industrial robot as the center and the specified distance as the radius. Select the staff member with the highest comprehensive maintenance index within the circle, and send the fault information and the location of the industrial robot to their mobile terminal at the same time. The staff member will perform inspection and maintenance and decide whether to replace the robotic arm based on the actual situation.

[0043] Compared with the prior art, the present invention has the following beneficial effects:

[0044] The present invention analyzes the historical fault information of the industrial robot to obtain a set of fault indication values ​​corresponding to each fault type, sets a detection node during the operation of the industrial robot, analyzes the fault data in real time, and calculates and matches the fault indication value set accordingly, thereby enabling early warning of industrial robot faults and accurately locating the source of the fault, thereby solving the problem in the prior art that an alarm can only be triggered when a fault occurs in the industrial robot, and maintenance personnel are required to perform a secondary analysis of the specific type of the fault, which consumes the time cost of the enterprise.

[0045] The present invention solves the problems in a targeted manner by providing different solutions to different fault types, thereby solving the problems that only one solution or strategy is used in the prior art to deal with various faults, thereby causing low efficiency in problem handling and excessively long processing time. BRIEF DESCRIPTION OF THE DRAWINGS

[0046] Further details, features and advantages of the present application are disclosed in the following description of exemplary embodiments in conjunction with the accompanying drawings, in which:

[0047] Figure 1 It is a principle block diagram of the present invention. DETAILED DESCRIPTION

[0048] Several embodiments of the present application will be described in more detail below with reference to the accompanying drawings so that those skilled in the art can implement the present application. The present application can be embodied in many different forms and purposes and should not be limited to the embodiments described herein. These embodiments are provided to make the present application comprehensive and complete, and to fully convey the scope of the present application to those skilled in the art. The embodiments do not limit the present application.

[0049] Unless otherwise defined, all terms (including technical and scientific terms) used herein have the same meaning as commonly understood by those of ordinary skill in the art to which this application belongs. It will be further understood that terms such as those defined in commonly used dictionaries should be interpreted as having a meaning consistent with their meaning in the context of the relevant art and / or this specification, and will not be interpreted in an idealized or overly formal sense, unless explicitly defined as such herein.

[0050] See also Figure 1 As shown, an industrial robot fault diagnosis and recovery system includes: a data acquisition module, a maintenance personnel registration module, a data analysis module, a fault classification module, and an early warning module;

[0051] Data collection module: extracts historical fault information of industrial robots, takes the time point of each failure of the robot as the starting point, and obtains abnormal data packets within the set time interval before each failure of the industrial robot;

[0052] The acquired abnormal data packets are classified according to different fault types, which include motor abnormality, communication abnormality and mechanical abnormality; the three groups of abnormal data packets include motor abnormality data packets, communication abnormality data packets and mechanical abnormality data packets; the motor abnormality data packets, communication abnormality data packets and mechanical abnormality data packets are set to correspond to a group of data sets respectively;

[0053] Set the motor abnormal data packet, communication abnormal data packet and mechanical abnormal data packet to correspond to a set of data sets respectively, specifically:

[0054] The data set of the motor abnormality data packet is the robot motor temperature value in the set time interval and the working environment temperature value of the industrial robot; the data set of the communication abnormality data packet is the communication rate value in the set time interval; the data set of the mechanical abnormality data packet is the image information of the industrial robot mechanical arm and the mechanical arm acceleration value;

[0055] Maintenance personnel registration module: used to collect information corresponding to maintenance personnel; the information includes the distance between the maintenance personnel and the industrial robot, the maintenance success rate of the maintenance personnel corresponding to each fault type, and the maintenance personnel's educational level;

[0056] Data analysis module: used to analyze each abnormal data packet corresponding to each fault type to obtain the fault indication value corresponding to each fault type; the fault indication value includes the motor fault indication value DJC, the communication fault indication value TXC and the mechanical fault indication value JXC; and send the fault indication value to the fault classification module;

[0057] The motor abnormality data packet corresponding to the motor abnormality fault type is analyzed to obtain the motor fault indication value DJC, which is specifically:

[0058] S1: Get the motor temperature value at each time node in the set time interval, and draw it into a line graph according to the time series. Preset the motor temperature value threshold, construct the threshold line corresponding to the motor temperature threshold on the line graph, mark the closed area surrounded by the threshold line and the line graph, and statistically sum the areas of the closed areas above and below the threshold line. The obtained values ​​are recorded as M1 and M2 respectively. Use the formula Gets overworked than MJ;

[0059] The area above the line graph and the area enclosed by the line graph can be simply understood as the high-temperature working time, while the area below the line graph and the area enclosed by the line graph can be simply understood as the low-temperature working time. Therefore, the larger the MJ value, the longer the high-temperature time of the industrial robot motor.

[0060] S2: In the line graph, the sum of the time when the motor temperature value is above the threshold line is counted and recorded as the high-temperature working time. The high-temperature working time and the set time interval are calculated by ratio, and the obtained value is recorded as the high-temperature working ratio MK;

[0061] The high temperature working ratio MK can reflect the proportion of high temperature working time to the total working time. The larger the MK value, the more likely the motor is to fail.

[0062] S3: Obtain the ambient temperature values ​​of the industrial robot at each time node in the set time interval, take the average as the ring average, preset the ring average threshold of the industrial robot, subtract the ring average threshold from the ring average, so as to obtain the environmental index, preset each environmental index interval corresponding to each group of ambient temperature scores, match the environmental index interval corresponding to the environmental index, so as to obtain the ambient temperature score TR corresponding to the environmental index; the larger the ambient temperature score TR, the more suitable the ambient temperature is for the industrial robot to work;

[0063] The environmental index may have negative values, but points can still be assigned to negative values. For example, when the environmental index is 0, the ambient temperature score is 10 points, and when the environmental index is -1, the ambient temperature score is 11 points.

[0064] S4: The excess work ratio MJ, high temperature work ratio MK, and ambient temperature score TR are normalized and then input into the formula: The motor fault indication value DJC is obtained, where a1, a2 and a3 are the weight influence factors corresponding to the ambient temperature score TR, excess working ratio MJ and high temperature working ratio MK respectively.

[0065] The larger the calculated motor fault indication value DJC is, the greater the probability of motor failure is;

[0066] The communication abnormality data packet corresponding to the communication abnormality fault type is analyzed to obtain the communication fault indication value TXC, which is specifically:

[0067] L1: Obtain the network delay data value at each time node in the set time interval, calculate the network delay mean value N1 using the mean formula, and extract the network delay data value peak value as the network peak value N2;

[0068] The average network delay N1 can reflect the average level of network delay in the set time interval, while the peak network delay N2 can reflect the extreme situation of network delay in the set time interval. The smaller the values ​​of both, the better;

[0069] L2: Obtain the broadband utilization rate at each time node in the set time interval, and use the mean formula to calculate the broadband utilization rate at each time node in the set time interval to obtain the average broadband rate KD1; extract the maximum value of the broadband utilization rate, preset the broadband utilization rate threshold, and subtract the broadband utilization rate threshold from the maximum value of the broadband utilization rate to obtain the broadband overload value KD2;

[0070] The broadband rate is limited. The higher the broadband utilization rate, the lower the broadband rate allocated to the industrial robot. The average bandwidth rate KD1 represents the average level of broadband utilization in a set time interval, while the broadband overload value KD2 represents the extreme load level of the broadband. The larger the value, the worse it is.

[0071] L3: Extract the network delay mean threshold N1 from the preset database阈值 , Network peak threshold N2 阈值 , width average rate KD1 阈值 And broadband overload rate threshold KD2 阈值 , the obtained network delay mean N1 and network delay mean threshold N1 阈值 , Network peak value N2, Network peak threshold N2 阈值 , width average rate KD1, width average rate KD1 阈值 , broadband overload value KD2, broadband overload rate threshold KD2 阈值 After normalization, enter the formula The communication fault indication value TXC is obtained, where b1, b2, b3, and b4 are weighted influencing factors corresponding to the network delay mean value N1, the network peak value N2, the bandwidth average rate KD1, and the broadband overload value KD2 respectively;

[0072] By calculating the ratio of the obtained data to the respective set thresholds and then multiplying them by their respective weight influence factors, the different effects of four different factors on the communication fault indication value TXC are scientifically considered, so that the communication fault indication value TXC can more accurately reflect whether the industrial robot communication is stable;

[0073] The mechanical abnormality data packet corresponding to the mechanical abnormality fault type is analyzed to obtain the mechanical abnormality indication value JXC, which is specifically:

[0074] D1: Acquire the image of the industrial robot's mechanical arm at a set time interval, and preprocess the image of the industrial robot's mechanical arm, including enhancement and noise reduction;

[0075] D2: Divide the obtained robot arm image into x sub-parts, and number the robot arm images of each sub-part, where the number is represented by i, i=1,2...x; obtain the original images corresponding to each sub-part of the robot arm from the database, compare the obtained robot arm images of each sub-part with the corresponding original images, and calculate based on the perceptual hash algorithm to obtain the similarity between the robot arm images of each sub-part and the corresponding original images, which is recorded as the similarity score XS i ;

[0076] Similarity Score XS i The higher it is, the less wear and damage the corresponding part of the robot arm has;

[0077] D3: Obtain the acceleration value of the industrial robot arm at each time node in the set time interval, preset the acceleration threshold of the industrial robot arm, count the number of acceleration values ​​exceeding the acceleration threshold in the set time interval, record it as the excess value GF, calculate the acceleration data using the standard deviation formula, and obtain the excess value JL. After normalizing the excess value GF and excess value JL, enter them into the formula The acceleration index JSD is obtained, where v1 and v2 are the weighted influence factors corresponding to the super value GF and the wave value JL respectively;

[0078] D4: Extract the corresponding maintenance times of each sub-part of the industrial robot arm from the historical maintenance data of the industrial robot, calculate the ratio of the maintenance times of each sub-part of the industrial robot to the total maintenance times, and get the maintenance factor μ i , preset each group of maintenance factors μ i The interval corresponds to the maintenance influence weight of each group, and the maintenance factor μ i Match with the preset maintenance factor intervals to obtain the maintenance factor μ i The corresponding maintenance impact weight is used as the maintenance impact weight of each sub-part of the robot arm as the similarity score XS i The weight influence factor λ i ;

[0079] Maintenance factor μ i The larger the corresponding interval value, the smaller the corresponding maintenance impact weight;

[0080] D5: The obtained similarity score XS i , the acceleration index JSD normalized and the weight influence factors corresponding to each part of the robot are substituted into the formula The mechanical fault indication value JXC is obtained, where dj is the weighted influence factor corresponding to the acceleration index JSD;

[0081] Fault classification module: used to store each group of fault indication values ​​corresponding to the historical fault information of the industrial robot, and divide each group of fault indication values ​​into three abnormal fault sets according to the different fault types, namely, the motor abnormal fault set corresponding to the motor abnormality, the communication abnormal fault set corresponding to the communication abnormality, and the mechanical abnormal fault set corresponding to the mechanical abnormality;

[0082] If the absolute value of the difference between a certain group of fault indication values ​​within a set time interval during the operation of the industrial robot and a certain group of fault indication values ​​in the corresponding abnormal fault set is less than a preset threshold, then a warning signal of the fault type corresponding to the abnormal fault set is triggered; the warning signal includes a motor abnormality warning signal, a communication abnormality warning signal and a mechanical abnormality warning signal;

[0083] For different fault types, the threshold of the absolute value of the difference between the corresponding fault indication value and the data in the corresponding abnormal fault set should be set to different values, and the specific value can be adjusted by the operator according to the actual situation;

[0084] Early warning module: executes corresponding steps based on the corresponding early warning signals received;

[0085] When the received result is a motor abnormality warning signal: obtain the difference between the motor fault indication value and the motor fault indication value threshold, preset the intervals of the difference between the motor fault indication value and the motor fault indication value threshold to correspond to the danger levels of the motors, and the danger levels are divided into slight, general and severe; match the intervals corresponding to the difference between the motor fault indication value and the motor fault indication value threshold, thereby obtaining the danger level of the motor;

[0086] When the danger level is slight: different power ranges are preset based on the different danger levels corresponding to the motor cooling fan and the motor, and based on the power ranges corresponding to the slight danger levels preset for the motor cooling fan and the motor, the motor cooling fan power and the motor power are adjusted to the preset ranges;

[0087] When the danger level is general: Based on the power ranges corresponding to the general danger levels preset for the motor cooling fan and the motor, the motor cooling fan power and the motor power are adjusted to the preset ranges, and the target position to be reached by the current target industrial robot is sent to the mobile terminal of the maintenance personnel closest to the target position, and the motor danger level is packaged and sent together, and the staff will make a secondary judgment and analysis on whether it is necessary to stop the machine for maintenance or shut down for rest;

[0088] When the danger level is severe: immediately shut down the industrial robot and monitor the motor temperature in real time. When the motor temperature drops to a safe temperature range, obtain the cooling area closest to the current industrial robot and control the industrial robot to automatically move to the nearest cooling area.

[0089] When the received result is a communication abnormality warning signal: immediately obtain the image information at the communication interface and perform preprocessing, use graphic processing technology to obtain the ratio of the dust area in the image to the total area of ​​the interface, record it as the dust impact degree, preset the dust impact degree threshold, and when the dust impact degree is greater than the preset dust impact degree threshold, immediately obtain the cleaning staff closest to the current industrial robot, and send the cleaning instruction and the location of the industrial robot to the cleaning staff's mobile terminal, and let the cleaning staff perform the cleaning operation;

[0090] It should be noted that: since dust has a certain conductivity and micro-capacitance effect, the dust at the power interface can affect the communication strength of the industrial robot. The greater the dust impact, the stronger the impact and the lower the communication strength.

[0091] When the dust impact is less than the preset dust impact threshold, a circle is drawn with the current position of the industrial robot as the center and the specified area as the radius to obtain the distance between each maintenance personnel and the industrial robot, the repair success rate of the communication fault, and the type of education within the circle; the education score corresponding to each type of education is preset; the education type of each maintenance personnel is matched with the corresponding education score, so as to obtain the education score of each maintenance personnel;

[0092] The education score can be assigned 1 to 4 points according to high school, technical secondary school, junior college, and undergraduate degrees respectively; the education type of the maintenance personnel is matched with the preset education score to obtain the education score of each maintenance personnel;

[0093] The distance between each maintenance personnel and the industrial robot, the repair success rate of communication failures, and the educational level score are marked as C1, C2, and C3 respectively;

[0094] The distance C1 from the industrial robot, the repair success rate C2 of the communication failure, and the educational background type C3 of each maintenance personnel within the circle are normalized and entered into the formula: The comprehensive maintenance index Wt of each maintenance personnel is obtained, where w1, w2 and w3 are the weighted influencing factors corresponding to the distance C1 between the maintenance personnel and the industrial robot, the repair success rate C2 of the communication failure and the maintenance personnel's educational score C3, respectively. The maintenance personnel with the largest comprehensive maintenance index Wt are selected as the target maintenance personnel, and the industrial robot fault information and the industrial robot's location information are packaged and sent to the target maintenance personnel terminal, who performs maintenance operations.

[0095] When the received result is a mechanical abnormality warning signal: draw a circle with the current industrial robot as the center and the specified distance as the radius, select the staff member with the highest comprehensive maintenance index within the circle, and send the fault information and the location of the industrial robot to their mobile terminal at the same time. The staff member will conduct inspection and maintenance and determine whether the robotic arm needs to be replaced based on the actual situation.

[0096] The preferred embodiments of the present invention disclosed above are only used to help explain the present invention. The preferred embodiments do not describe all the details in detail, nor do they limit the invention to only specific implementation methods. Obviously, many modifications and changes can be made according to the content of this specification. This specification selects and specifically describes these embodiments in order to better explain the principles and practical applications of the present invention, so that those skilled in the art can understand and use the present invention well. The present invention is limited only by the claims and their full scope and equivalents.

Claims

1. An industrial robot fault diagnosis and recovery system, characterized in that: include: Data collection module: extracts historical fault information of industrial robots, takes the time point of each failure of the robot as the starting point, and obtains abnormal data packets within the set time interval before each failure of the industrial robot; The acquired abnormal data packets are classified according to different fault types, which include motor abnormality, communication abnormality and mechanical abnormality; the three groups of abnormal data packets include motor abnormality data packets, communication abnormality data packets and mechanical abnormality data packets; Set the motor abnormal data packet, the communication abnormal data packet and the mechanical abnormal data packet to correspond to a set of data sets respectively; Maintenance personnel registration module: used to collect information corresponding to maintenance personnel; the information includes the distance between the maintenance personnel and the industrial robot, the maintenance success rate of the maintenance personnel for each fault type, and the maintenance personnel's educational level; the educational level includes high school and below, technical secondary school, junior college, and undergraduate; Data analysis module: used to analyze each abnormal data packet corresponding to each fault type to obtain the fault indication value corresponding to each fault type; the fault indication value includes the motor fault indication value DJC, the communication fault indication value TXC and the mechanical fault indication value JXC; and send the fault indication value to the fault classification module; Fault classification module: used to store each group of fault indication values ​​corresponding to the historical fault information of the industrial robot, and divide each group of fault indication values ​​into three abnormal fault sets according to the different fault types, namely, the motor abnormal fault set corresponding to the motor abnormality, the communication abnormal fault set corresponding to the communication abnormality, and the mechanical abnormal fault set corresponding to the mechanical abnormality; If the absolute value of the difference between a certain group of fault indication values ​​within a set time interval during the operation of the industrial robot and a certain group of fault indication values ​​in the corresponding abnormal fault set is less than a preset threshold, a warning signal of the fault type corresponding to the abnormal fault set is triggered; The warning signals include motor abnormality warning signal, communication abnormality warning signal and mechanical abnormality warning signal; Early warning module: executes corresponding steps based on the corresponding early warning signals received.

2. The industrial robot fault diagnosis and recovery system according to claim 1, characterized in that: Set the motor abnormal data packet, communication abnormal data packet and mechanical abnormal data packet to correspond to a set of data sets respectively, specifically: The data set of the motor abnormality data packet is the robot motor temperature value in the set time interval and the working environment temperature value of the industrial robot; the data set of the communication abnormality data packet is the communication rate value in the set time interval; the data set of the mechanical abnormality data packet is the image information of the industrial robot's mechanical arm and the mechanical arm acceleration value.

3. An industrial robot fault diagnosis and recovery system according to claim 2, characterized in that: The analysis of the motor abnormality data packet corresponding to the motor abnormality fault type is specifically as follows: S1: Get the motor temperature value at each time node in the set time interval, and draw it into a line graph according to the time series. Preset the motor temperature value threshold, construct the threshold line corresponding to the motor temperature threshold on the line graph, mark the closed area surrounded by the threshold line and the line graph, and statistically sum the areas of the closed areas above and below the threshold line. The obtained values ​​are recorded as M1 and M2 respectively. Use the formula Gets overworked than MJ; S2: In the line graph, the sum of the time when the motor temperature value is above the threshold line is counted and recorded as the high-temperature working time. The high-temperature working time and the set time interval are calculated by ratio, and the obtained value is recorded as the high-temperature working ratio MK; S3: Obtain the ambient temperature values ​​of the industrial robot at each time node in the set time interval, take the average as the ring average, preset the ring average threshold of the industrial robot, subtract the ring average threshold from the ring average, so as to obtain the environmental index, preset each environmental index interval corresponding to each group of ambient temperature scores, match the environmental index interval corresponding to the environmental index, so as to obtain the ambient temperature score TR corresponding to the environmental index; S4: The excess work ratio MJ, high temperature work ratio MK, and ambient temperature score TR are normalized and then input into the formula: The motor fault indication value DJC is obtained, where a1, a2 and a3 are the weight influence factors corresponding to the ambient temperature score TR, excess working ratio MJ and high temperature working ratio MK respectively.

4. The industrial robot fault diagnosis and recovery system according to claim 3, characterized in that: Analyze the abnormal communication data packets corresponding to the communication fault type, specifically: L1: Obtain the network delay data value at each time node in the set time interval, calculate the network delay mean value N1 using the mean formula, and extract the network delay data value peak value as the network peak value N2; L2: Obtain the broadband utilization rate at each time node in the set time interval, and use the mean formula to calculate the broadband utilization rate at each time node in the set time interval to obtain the average broadband rate KD1; extract the maximum value of the broadband utilization rate, preset the broadband utilization rate threshold, and subtract the broadband utilization rate threshold from the maximum value of the broadband utilization rate to obtain the broadband overload value KD2; L3: Extract the network delay mean threshold N1 from the preset database 阈值 , Network peak threshold N2 阈值 , width average rate KD1 阈值 And broadband overload rate threshold KD2 阈值 , the obtained network delay mean N1 and network delay mean threshold N1 阈值 , Network peak value N2, Network peak threshold N2 阈值 , width average rate KD1, width average rate KD1 阈值 , broadband overload value KD2, broadband overload rate threshold KD2 阈值 After normalization, enter the formula The communication fault indication value TXC is obtained, where b1, b2, b3, and b4 are weighted influencing factors corresponding to the network delay mean value N1, the network peak value N2, the bandwidth average rate KD1, and the broadband overload value KD2, respectively.

5. The industrial robot fault diagnosis and recovery system according to claim 4, characterized in that: Analyze the mechanical abnormality data packets corresponding to the mechanical abnormality fault types, specifically: D1: Acquire the image of the industrial robot's mechanical arm at a set time interval and pre-process the image of the industrial robot's mechanical arm; D2: Divide the obtained robot arm image into x sub-parts, and number the robot arm images of each sub-part, where the number is represented by i, i=1,2...x; obtain the original images corresponding to each sub-part of the robot arm from the database, compare the obtained robot arm images of each sub-part with the corresponding original images, and calculate based on the perceptual hash algorithm to obtain the similarity between the robot arm images of each sub-part and the corresponding original images, which is recorded as the similarity score XS i ; D3: Obtain the acceleration value of the industrial robot arm at each time node in the set time interval, preset the acceleration threshold of the industrial robot arm, count the number of acceleration values ​​exceeding the acceleration threshold in the set time interval, record it as the excess value GF, calculate the acceleration data using the standard deviation formula, and obtain the excess value JL. After normalizing the excess value GF and excess value JL, enter them into the formula The acceleration index JSD is obtained, where v1 and v2 are the weighted influence factors corresponding to the super value GF and the wave value JL respectively; D4: Extract the corresponding maintenance times of each sub-part of the industrial robot arm from the historical maintenance data of the industrial robot, calculate the ratio of the maintenance times of each sub-part of the industrial robot to the total maintenance times, and get the maintenance factor μ i , preset each group of maintenance factors μ i The interval corresponds to the maintenance influence weight of each group, and the maintenance factor μ i Match with the preset maintenance factor intervals to obtain the maintenance factor μ i The corresponding maintenance impact weight is used as the maintenance impact weight of each sub-part of the robot arm as the similarity score XS i The weight influence factor λ i . D5: The obtained similarity score XS i , the acceleration index JSD normalized and the weight influence factors corresponding to each part of the robot are substituted into the formula The mechanical fault indication value JXC is obtained, where dj is the weight influence factor corresponding to the acceleration index JSD.

6. An industrial robot fault diagnosis and recovery system according to claim 5, characterized in that: Based on the received motor abnormality warning signal, corresponding steps are executed, specifically: Obtain the difference between the motor fault indication value and the motor fault indication value threshold, and preset the intervals of the difference between the motor fault indication value and the motor fault indication value threshold to correspond to the danger levels of the motors, which are divided into slight, general and severe; Matching the interval corresponding to the difference between the motor fault indication value and the motor fault indication value threshold, thereby obtaining the motor danger level; When the danger level is slight, different power intervals are preset based on the different danger levels corresponding to the motor cooling fan and the motor, and based on the power intervals corresponding to the slight danger levels preset for the motor cooling fan and the motor, the motor cooling fan power and the motor power are adjusted to the preset intervals; When the danger level is general, based on the power range corresponding to the general danger level preset by the motor cooling fan and the motor, the motor cooling fan power and the motor power are adjusted to the preset range, and the target position to be reached by the current target industrial robot is sent to the mobile terminal of the maintenance personnel closest to the target position, and the motor danger level is packaged and sent together, and the staff makes a secondary judgment and analysis on whether it is necessary to stop the machine for maintenance or shut down for rest; When the danger level is serious, the industrial robot is shut down immediately and the motor temperature is monitored in real time. When the motor temperature is detected to drop to a safe temperature range, the cooling area closest to the current industrial robot is obtained and the industrial robot is controlled to automatically move to the nearest cooling area.

7. An industrial robot fault diagnosis and recovery system according to claim 6, characterized in that: Based on the received communication abnormality warning signal, corresponding steps are executed, specifically: The image information at the communication interface is obtained and preprocessed. The ratio of the dust area in the image to the total area of ​​the interface is obtained by using graphic processing technology, which is recorded as the dust impact degree. A dust impact degree threshold is preset. When the dust impact degree is greater than the preset dust impact degree threshold, the cleaning staff closest to the current industrial robot is immediately obtained, and the cleaning instruction and the location of the industrial robot are sent to the cleaning staff's mobile terminal, and the cleaning staff performs the cleaning operation; When the dust impact is less than the preset dust impact threshold, a circle is drawn with the current position of the industrial robot as the center and the specified area as the radius to obtain the distance between each maintenance personnel and the industrial robot, the repair success rate of the communication fault, and the type of education within the circle; the education score corresponding to each type of education is preset; the education type of each maintenance personnel is matched with the corresponding education score, so as to obtain the education score of each maintenance personnel; The distance between each maintenance personnel and the industrial robot, the repair success rate of communication failures, and the educational level score are marked as C1, C2, and C3 respectively; The distance C1 of each maintenance personnel from the industrial robot, the repair success rate C2 of communication failure, and the academic qualification score C3 within the circle are normalized and entered into the formula: The comprehensive maintenance index Wt of each maintenance personnel is obtained, where w1, w2 and w3 are the weighted influencing factors corresponding to the distance C1 between the maintenance personnel and the industrial robot, the repair success rate C2 of the communication failure and the maintenance personnel's educational score C3, respectively. The maintenance personnel with the largest comprehensive maintenance index Wt are selected as the target maintenance personnel, and the industrial robot fault information and the industrial robot's location information are packaged and sent to the target maintenance personnel terminal, who performs maintenance operations.

8. An industrial robot fault diagnosis and recovery system according to claim 7, characterized in that: Based on the received mechanical abnormality warning signal, the corresponding steps are executed, specifically: Draw a circle with the current industrial robot as the center and the specified distance as the radius. Select the staff member with the highest comprehensive maintenance index within the circle, and send the fault information and the location of the industrial robot to their mobile terminal at the same time. The staff member will perform inspection and maintenance and decide whether to replace the robotic arm based on the actual situation.

Images