A method of maintenance

Abstract

A kind of maintenance method.The maintenance method is used for conveying system, conveying system includes detection device, conveying belt and driving mechanism, driving mechanism is used to drive the movement of conveying belt, conveying belt is used to move workpiece, so that workpiece sequentially passes through multiple key positions;Detection device is used to detect the action of workpiece passing through multiple key positions;Maintenance method includes: obtaining the detection result of detection device, to determine the time length data required for workpiece to move between two key positions;Based on time length data, determine the characteristics of time length data;The characteristics of time length data are matched with the characteristics of the preset working performance of target component, and the future working performance of target component is predicted according to the matching result;Target component includes at least one of detection device, conveying belt and driving mechanism.The maintenance method can predict the future working performance of target component based on the time length data of workpiece, so as to carry out predictive maintenance, so that the working reliability of pipeline is higher.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of equipment maintenance, in particular to a maintenance method. BACKGROUND

[0002] In the era of Industry 4.0, factory automation assembly and test lines are getting longer and longer. Among them, the movement of workpieces in the line is usually realized through a flow line. Generally, workpieces can be loaded on carriers and move with the carriers. Moreover, the assembly and testing of workpieces are usually multi-process, and when automatic operation, the workpieces on the flow line need to flow and position in each processing station in turn.

[0003] The main working components of the flow line generally include motors, belts and photoelectric sensors, and these three working components have the risk of failure in use. The current scheme cannot predict the failure type and / or remaining service life of the working components in advance, resulting in low working reliability of the flow line. SUMMARY

[0004] The embodiments of the present application provide a maintenance method, which can predict the future working performance of a target component based on the time length data of workpieces, so as to perform predictive maintenance, so that the working reliability of the flow line is high.

[0005] In a first aspect, the embodiments of the present application provide a maintenance method for a conveying system, the conveying system comprising a detection device, a conveying belt and a driving mechanism, the driving mechanism being configured to drive the conveying belt to move, the conveying belt being configured to drive workpieces to move so that the workpieces pass through a plurality of key positions in turn; the detection device being configured to detect the actions of the workpieces passing through the plurality of key positions; the maintenance method comprising: obtaining the detection results of the detection device to determine the time length data required for the workpieces to move between two key positions; determining the features of the time length data based on the time length data; matching the features of the time length data with the features of the preset working performance of a target component, and predicting the future working performance of the target component according to the matching result; wherein the target component comprises at least one of the detection device, the conveying belt and the driving mechanism.

[0006] The maintenance method of the embodiments of the present application can predict the future working performance of a target component based on the time length data of a plurality of workpieces, so as to perform predictive maintenance, so that the working reliability of the flow line is high. For example, the remaining service life of the target component can be predicted, avoiding the occurrence of unpredictable long-time stop line when failure occurs, which helps to improve production efficiency and reduce maintenance cost, and for another example, the failure type of the target component can be predicted, which facilitates maintenance personnel to formulate appropriate maintenance strategies based on the failure type.

[0007] In a possible implementation, the plurality of key positions include an entering position, a to-position and a leaving position arranged in sequence along the conveying direction of the conveying belt; the entering position refers to a position where the workpiece enters a working area, the to-position refers to a processing position of the workpiece, and the processing position is located in the working area, and the leaving position refers to a position where the workpiece leaves the working area; the detection result of the detection device is acquired to determine the time length data required for the workpiece to move between two key positions, including: acquiring the detection result of the detection device detecting the action of the workpiece at the entering position and the action at the to-position to determine the entering time length of the workpiece, wherein the entering time length refers to the time length of the workpiece moving from the entering position to the to-position; and / or acquiring the detection result of the detection device detecting the action of the workpiece at the to-position and the action at the leaving position to determine the leaving time length of the workpiece; wherein the leaving time length refers to the time length of the workpiece moving from the to-position to the leaving position.

[0008] In a possible implementation, a plurality of workpieces are sequentially conveyed on the conveying belt; the detection result of the detection device is acquired to determine the time length data required for the workpiece to move between two key positions; including: sequentially acquiring the detection result of the detection device to determine the time length data required for each workpiece in the plurality of workpieces to move between two key positions.

[0009] In a possible implementation, the characteristic of the time length data is determined based on the time length data, including: determining a data set corresponding to a target workpiece in the plurality of workpieces based on the time length data of the plurality of workpieces; wherein the data set includes the time length data of the target workpiece and M workpieces before the target workpiece, M is a positive integer greater than or equal to 1, and the time length data in the data set corresponding to adjacent target workpieces does not overlap or partially overlaps; determining a deviation value corresponding to the target workpiece based on the data set, the deviation value including a standard deviation and / or a root mean square error; and determining the change of the deviation value over time as the characteristic of the time length data.

[0010] In a possible implementation, the preset working performance feature of the target component includes fault state features corresponding to different fault types, and each of the fault state features includes a variation of a deviation value of the time length data of the workpieces in a fault state over time when the plurality of workpieces are sequentially conveyed on the conveying belt; and the matching of the time length data feature with the preset working performance feature of the target component, and the prediction of the future working performance of the target component according to the matching result, include: matching the time length data feature with the plurality of fault state features, and predicting the fault type of the target component according to the matching result.

[0011] In a possible implementation, the preset working performance feature of the target component includes state features of a reference component of the target component in use cycles, the target component and the reference component are of the same type, each of the state features in the use cycles includes a normal state feature and a fault state feature; the fault state feature includes a variation of a deviation value of the time length data of the workpieces in a fault state over time when the plurality of workpieces are sequentially conveyed on the conveying belt, and the normal state feature includes a variation of a deviation value of the time length data of the workpieces in a normal state over time; and the matching of the time length data feature with the preset working performance feature of the target component, and the prediction of the future working performance of the target component according to the matching result, include: matching the time length data feature with the state features of the reference component of the target component in the use cycles, and predicting the remaining service life of the target component according to the matching result.

[0012] In a possible implementation, the target component includes at least two of the detection device, the conveying belt and the driving mechanism; and the matching of the time length data feature with the state features of a reference component of the target component in use cycles, and the prediction of the remaining service life of the target component according to the matching result, include: determining a component of the time length data feature corresponding to each of the target components based on the time length data feature; and predicting the remaining service life of each of the target components based on the component of the time length data feature corresponding to each of the target components and the state feature corresponding to each of the target components of the reference component of each of the target components in the use cycles.

[0013] In a possible implementation, the maintenance method further includes: determining an average time length, the average time length being an average value of time length data of L workpieces before a current workpiece in the plurality of workpieces, L being a positive integer; and outputting a warning signal in a case where an absolute value of a difference between the time length data of the current workpiece and the average time length is greater than a set value, the warning signal indicating that the target component is in a fault state.

[0014] In a possible implementation, the maintenance method comprises: matching, by a prediction model, the feature of the time length data with a feature of a preset working performance of the target component, and predicting the future working performance of the target component according to a matching result; wherein the input of the prediction model is the time length data or the feature of the time length data, and the output of the prediction model is the future working performance of the target component, and comprises the remaining service life of the target component and / or a failure type.

[0015] In a possible implementation, when the input of the prediction model is the time length data, the prediction model is trained in the following manner: a first sample set is obtained by the detection device, the first sample set being time length data corresponding to a reference component of the target component in the use cycle; the prediction model is trained based on the first sample set, so that the prediction model can determine a state feature of the reference component of the target component in the use cycle, to obtain an initial prediction model; a second sample set is obtained by the detection device, the second sample set being time length data of a plurality of workpieces conveyed on the conveying belt; the initial prediction model is trained based on the second sample set, so that the initial prediction model can determine a component of the feature of the time length data corresponding to the target component, and determine the remaining service life of the target component and / or the failure type according to the state feature of the reference component of the target component in the use cycle and the component of the feature of the time length data corresponding to the target component, to obtain a trained prediction model.

[0016] In a second aspect, an embodiment of the present application provides a maintenance device, the maintenance device being used in a conveying system, the conveying system comprising a detection device, a conveying belt and a driving mechanism, the driving mechanism being used to drive the conveying belt to move, the conveying belt being used to drive workpieces to move, so that the workpieces pass through a plurality of key positions in sequence; the detection device being used to detect actions of the workpieces passing through the plurality of key positions; the maintenance device comprising: a time length determination module, used to obtain a detection result of the detection device, to determine time length data required for the workpieces to move between two key positions; a feature extraction module, used to determine a feature of the time length data based on the time length data; and a prediction module, used to match the feature of the time length data with a feature of a preset working performance of a target component, and predict a future working performance of the target component according to a matching result; wherein the target component comprises at least one of the detection device, the conveying belt and the driving mechanism.

[0017] In a possible implementation, the plurality of key positions include an entering position, a to-position and a leaving position arranged in sequence along a conveying direction of the conveying belt; the entering position refers to a position at which the workpiece enters a working area, the to-position refers to a processing position of the workpiece, and the processing position is located in the working area, and the leaving position refers to a position at which the workpiece leaves the working area; the time length determining module is specifically configured to: acquire the action of the detection device detecting the workpiece at the entering position and the action of the detection device detecting the workpiece at the to-position, to determine an entering time length of the workpiece, where the entering time length refers to a time length required for the workpiece to move from the entering position to the to-position; and / or acquire the action of the detection device detecting the workpiece at the to-position and the action of the detection device detecting the workpiece at the leaving position, to determine a leaving time length of the workpiece, where the leaving time length refers to a time length required for the workpiece to move from the to-position to the leaving position.

[0018] In a possible implementation, a plurality of workpieces are sequentially conveyed on the conveying belt; and the time length determining module is specifically configured to: sequentially acquire detection results of the detection device, to determine time length data required for each workpiece in the plurality of workpieces to move between two key positions.

[0019] In a possible implementation, the feature extraction module is specifically configured to: determine, based on time length data of a plurality of workpieces, a data set corresponding to a target workpiece in the plurality of workpieces; where the data set includes time length data of the target workpiece and M workpieces before the target workpiece, M is a positive integer greater than or equal to 1, and time length data in the data set corresponding to adjacent target workpieces do not overlap or partially overlap; determine, based on the data set, a deviation value corresponding to the target workpiece, where the deviation value includes a standard deviation and / or a root mean square error; and determine, as a feature of the time length data, a change of the deviation value over time.

[0020] In a possible implementation, the feature of the preset working performance of the target component includes a fault state feature, and the fault state feature has a plurality of values corresponding to different fault types; when a plurality of workpieces are sequentially conveyed on the conveying belt, each fault state feature includes a change of a deviation value of time length data of the workpieces in a fault state over time; and the prediction module is specifically configured to: match the feature of the time length data with a plurality of fault state features, and predict a fault type of the target component according to a matching result.

[0021] In a possible implementation, the preset working performance feature of the target component includes state features of a reference component of the target component in use cycles, the reference component is of the same type as the target component, each state feature in the use cycles includes a normal state feature and a fault state feature, the fault state feature includes a variation of a deviation value of time length data of the workpieces in a fault state over time, and the normal state feature includes a variation of a deviation value of time length data of the workpieces in a normal state over time; and the prediction module is specifically configured to: match the time length data feature with the state features of the reference component of the target component in the use cycles, and predict the remaining service life of the target component according to a matching result.

[0022] In a possible implementation, the target component includes at least two of the detection device, the conveying belt and the driving mechanism; and the prediction module is specifically configured to: determine a component of the time length data feature corresponding to each target component based on the time length data feature, and predict the remaining service life of each target component based on a corresponding state feature of a reference component of each target component in the use cycle and the component of the time length data feature corresponding to each target component.

[0023] In a possible implementation, the maintenance device further includes a pre-warning module, the pre-warning module is configured to: determine an average time length, the average time length being an average of time length data of L workpieces before a current workpiece in the workpieces, L being a positive integer; and output a pre-warning signal in a case where an absolute value of a difference between the time length data of the current workpiece and the average time length is greater than a set value, the pre-warning signal indicating that the target component is faulty.

[0024] In a possible implementation, the prediction module is specifically configured to: match the time length data feature with the preset working performance feature of the target component by using a prediction model, and predict a future working performance of the target component according to a matching result, wherein an input of the prediction model is the time length data or the time length data feature, and an output of the prediction model is the future working performance of the target component, and includes the remaining service life of the target component and / or a fault type.

[0025] In a possible implementation, when the input of the prediction model is the time length data, the prediction model is trained in the following manner: a first sample set is obtained by the detection device, the first sample set being time length data corresponding to a reference component of the target component in the use cycle; the prediction model is trained based on the first sample set, so that the prediction model can determine the state feature of the reference component of the target component in the use cycle, to obtain an initial prediction model; a second sample set is obtained by the detection device, the second sample set being time length data of a plurality of workpieces conveyed on the conveying belt; the initial prediction model is trained based on the second sample set, so that the initial prediction model can determine the component of the feature of the time length data corresponding to the target component, and determine the remaining service life and / or the fault type of the target component according to the state feature of the reference component of the target component in the use cycle and the component of the feature of the time length data of the target component, to obtain the trained prediction model.

[0026] In a third aspect, an embodiment of the present application provides a control device, the control device comprising a processor and a memory; the processor and the memory are electrically connected; the memory is used for storing computer program code, the computer program code comprising computer instructions, when the processor executes the computer instructions, the control device executes the maintenance method described above.

[0027] In a fourth aspect, an embodiment of the present application provides a computer storage medium, the computer storage medium storing instructions, when the instructions run on a computer, the computer executes the method provided in the first aspect.

[0028] In a fifth aspect, an embodiment of the present application provides a computer program product comprising instructions, when the instructions run on a computer, the computer executes the method provided in the first aspect.

[0029] Other features and advantages of the present application will be described in detail in the following specific embodiment part. BRIEF DESCRIPTION OF DRAWINGS

[0030] The drawings used in the embodiments or prior art description are briefly introduced as follows.

[0031] Figure 1 It is a structural schematic diagram of a conveying system in an embodiment;

[0032] Figure 2 It is a structural schematic diagram of a conveying system in an embodiment; Figure 1

[0033] Figure 3 It is a flowchart of a maintenance method provided by an embodiment of the present application;​

[0034] Figure 4 for Figure 3 A schematic diagram illustrating the variation of a deviation value in the maintenance method shown;

[0035] Figure 5 for Figure 3 A schematic diagram of an exemplary maintenance method is shown.

[0036] Figure 6 for Figure 3 The graph showing the maintenance method is a curve of entry time versus number of runs;

[0037] Figure 7 This is a schematic diagram of the structure of a maintenance device provided in an embodiment of this application;

[0038] Figure 8 This is a schematic diagram of the structure of a control device provided in an embodiment of this application. Detailed Implementation

[0039] The technical solutions in the embodiments of this application will now be described with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments in this specification, and not all of them.

[0040] In the description of this specification, terms such as "one embodiment" or "some embodiments" mean that one or more embodiments of this specification include a particular feature, structure, or characteristic described in connection with that embodiment. Therefore, the phrases "in one embodiment," "in some embodiments," "in other embodiments," "in still other embodiments," etc., appearing in different parts of this specification do not necessarily refer to the same embodiment, but rather mean "one or more, but not all, embodiments," unless otherwise specifically emphasized.

[0041] In this specification, unless otherwise stated, " / " signifies "or," for example, A / B can mean A or B. "And / or" in this document merely describes the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent: A alone, A and B simultaneously, or B alone. Furthermore, in the description of the embodiments in this application, "multiple" refers to two or more.

[0042] In addition, in the description of the present specification, the terms "first", "second", "third" and the like are used only for the purpose of description, and cannot be understood as indicating or implying relative importance or implying the number of the technical features indicated. Therefore, the features defined as "first", "second" and the like can explicitly or implicitly include one or more of the features. The terms "include", "contain", "have" and their variants mean "including but not limited to", unless otherwise specifically emphasized.

[0043] Considering the cost requirements, the flow line generally uses a step motor or an AC motor for transmission, and these motors are open-loop control. The control system in the industrial computer cannot obtain motor feedback, and without feedback, there is no state detection, so generally predictive maintenance cannot be performed, and only responsive maintenance can be performed. Responsive maintenance is a solution that replaces and repairs after discovering that a working component fails, and cannot predict the failure type and / or the remaining service life of the working component in advance, resulting in low work reliability of the flow line.

[0044] In view of this, the embodiments of the present application provide a maintenance method, which can predict the working performance of the target component in the future based on the time length data of the workpiece, so as to perform predictive maintenance, so that the work reliability of the flow line is high. For example, the remaining service life of the target component can be predicted, avoiding unpredictable long-time stop line caused by failure, which helps to improve production efficiency and reduce maintenance cost; for another example, the failure type of the target component can be predicted, which facilitates maintenance personnel to formulate appropriate maintenance strategies based on the failure type. Wherein, the predictive maintenance refers to continuous state monitoring and fault diagnosis of the main components of the flow line, and according to the state development trend and possible failure mode of the working component, a predictive maintenance plan is prepared in advance.

[0045] Due to the high failure rate of the flow line, the maintenance time is long, and when the failure occurs, it will cause unpredictable stop line, and the stop line time is long, which not only affects the production efficiency, but also leads to high maintenance cost. The maintenance method of the embodiments of the present application realizes the predictive maintenance of the flow line according to the induction time without increasing the hardware, eliminates the unplanned stop line, predicts the components that need to be maintained, and can maximize the service life of the flow line motor, the conveyor belt and the detection device such as sensor.

[0046] Figure 1 The structure schematic diagram of the conveying system in an embodiment is shown in FIG. 1. As shown in FIG. 1, the conveying system includes a conveying belt 1, a motor 2, a sensor 3, a controller 4 and a workpiece 5. Figure 1As shown, the conveying system may include a conveying device 10, a drive mechanism 20, and a detection device 30. The conveying device 10 includes a conveyor belt 101 and a carrier 102 mounted on the conveyor belt, the carrier 102 being used to load workpiece P. The drive mechanism 20 is used to drive the conveyor belt 101, such as a belt drive, thereby moving the carrier 102 and the workpiece P loaded on the carrier 102. Exemplarily, the drive mechanism 20 may be a motor, such as a stepper motor or an AC motor. The detection device 30 is used to detect the movement of the carrier 102 carrying workpiece P as it passes by, and feeds this information back to the control system of the industrial control computer to determine the passing time of workpiece P. Exemplarily, the detection device 30 may be a photoelectric sensor.

[0047] Figure 2 for Figure 1 The diagram shows a section of the conveyor system. Figure 2 As shown, the detection device 30 may include an entry sensor 301, a position sensor 302, and a departure sensor 303, which are sequentially spaced and fixedly arranged along the conveying direction of the transmission belt 101. Thus, each of the multiple workpieces P can sequentially pass through the entry sensor 301, the position sensor 302, and the departure sensor 303. The control system records a first time T0 when the entry sensor 301 detects that the workpiece P has entered the working area, a second time T1 when the position sensor 302 detects that the workpiece P has reached the processing position, and a third time T2 when the position sensor 302 detects that the workpiece P has begun to leave the processing position. The interval between T2 and T1 is the dwell time of the workpiece P at the processing position. The control system also records a fourth time T3 when the departure sensor 303 detects that the workpiece has left the working area.

[0048] Figure 3 A flowchart illustrating a maintenance method provided for an example of this application. This maintenance method is used in a transmission system, which can be... Figure 1 and Figure 2 The conveying system shown includes a detection device, a conveyor belt, and a drive mechanism. The drive mechanism drives the conveyor belt, which in turn moves the workpiece, allowing it to sequentially pass through multiple key locations. The detection device detects the workpiece's movement through these key locations. Furthermore, maintenance can be performed by the control system in an industrial computer, such as... Figure 3 As shown, the maintenance method may include the following steps:

[0049] S301, Obtain the detection results from the detection device to determine the time required for the workpiece to move between two key positions.

[0050] The plurality of key positions include, in sequence along the conveying direction of the conveying belt, an entering position, a to-position, and a leaving position. The entering position refers to a position at which the workpiece enters the working area, the to-position refers to a processing position of the workpiece, the processing position is located in the working area, and the leaving position refers to a position at which the workpiece leaves the working area. The time length data can include a time length value and a time point at which the time length value is recorded.

[0051] Exemplarily, S301 can have, but is not limited to, the following three schemes:

[0052] Scheme 1: The detection of the workpiece at the entering position and the detection of the workpiece at the to-position are obtained by the detection device to determine the entering time length of the workpiece. The entering time length refers to the time length required for the workpiece to move from the entering position to the to-position, and the entering time length is represented by Ti, Ti=T1-T0.

[0053] Scheme 2: The detection of the workpiece at the to-position and the detection of the workpiece at the leaving position are obtained by the detection device to determine the leaving time length of the workpiece. The leaving time length refers to the time length required for the workpiece to move from the to-position to the leaving position, and the leaving time length is represented by To, To=T3-T2.

[0054] Scheme 3: The detection of the workpiece at the entering position and the detection of the workpiece at the to-position are obtained by the detection device to determine the entering time length of the workpiece, and the detection of the workpiece at the to-position and the detection of the workpiece at the leaving position are obtained by the detection device to determine the leaving time length of the workpiece. That is, scheme 3 is a combination of scheme 1 and scheme 2.

[0055] In addition, in some examples, a plurality of workpieces are sequentially conveyed on the conveying belt; S301 can include: sequentially obtaining the detection results of the detection device to determine the time length data required for each workpiece in the plurality of workpieces to move between the two key positions. That is, at this time, the time length data includes the time length data corresponding to the plurality of workpieces.

[0056] S302, determining a feature of the time length data based on the time length data.

[0057] In some examples, only one workpiece is conveyed on the conveying belt, and the time length data of the workpiece is the feature of the time length data.

[0058] In another example, a plurality of workpieces are conveyed on the conveying belt, and S302 can be performed according to the following process:

[0059] 1) Based on the time length data of the plurality of workpieces, a data set corresponding to a target workpiece in the plurality of workpieces can be determined, the data set including the time length data of the target workpiece and M workpieces before the target workpiece, M being a positive integer greater than or equal to 1.

[0060] Each data set includes a plurality of time length data arranged continuously, and the plurality of data sets can have, but are not limited to, the following two schemes:

[0061] Scheme 1: There is no overlap between the time length data in the data sets corresponding to adjacent target workpieces.

[0062] For example, M is 29, each data set includes 30 time length data, and the time length data of the 30th workpiece is taken as an example. The data set of the 30th workpiece includes the entering time length Ti of the 1st workpiece to the entering time length Ti of the 30th workpiece; the data set of the 60th workpiece includes the entering time length Ti of the 31st workpiece to the entering time length Ti of the 60th workpiece; the data set of the 90th workpiece includes the entering time length Ti of the 61st workpiece and the entering time length Ti of the 90th workpiece; and so on, so that the data sets corresponding to all target workpieces can be obtained, wherein the target workpiece is the 30nth workpiece, and n is a positive integer greater than or equal to 1.

[0063] Scheme 2: There is partial overlap between the time length data in the data sets corresponding to adjacent target workpieces.

[0064] For example, each data set includes M+1 time length data, and each workpiece corresponds to a data set starting from the (M+1)th workpiece, that is, the (M+1)th workpiece and the workpieces after it are target workpieces. The data set of the ith workpiece is a set formed by the time length data of the ith workpiece and the time length data of the M workpieces before the ith workpiece, M and i are positive integers, and i>M.

[0065] For example, M is 29, each data set includes 30 time length data, and each workpiece corresponds to a data set starting from the 30th workpiece. The data set corresponding to the 30th workpiece is a set formed by the time length data corresponding to the 1st workpiece to the 30th workpiece, the data set corresponding to the 31st workpiece is a set formed by the time length data corresponding to the 2nd workpiece to the 31st workpiece; and so on, so that the data sets corresponding to all target workpieces can be obtained, wherein the 30th workpiece and the workpieces after it are target workpieces.

[0066] 2) Determine the deviation value corresponding to the target workpiece based on the data set, and the deviation value includes the standard deviation and / or the root mean square error.

[0067] The standard deviation (standard deviation), also known as the mean square error (mean square error), is the average of the distance of each data from the average, which is the square root of the average of the sum of the square of the deviation. For the Xth workpiece, the data set corresponding thereto is {x1, x2, …, x N}, and the standard deviation corresponding to the Xth workpiece is: Where N represents the number of samples; x i Represents the i-th sample; μ represents x1 to x2. N The mean.

[0068] The root mean square error, also known as the standard error, is denoted by RMSE. In a finite number of measurements, for the Xth workpiece, the corresponding data set is {x1, x2, ..., x...}. N}, then the root mean square error corresponding to the Xth workpiece is: Where N represents the number of samples; x i Represents the i-th sample; y i These are the true values ​​for each sample.

[0069] Although the root mean square error (RMSE) is not significantly different from the standard deviation in its formula, they differ significantly in their physical meaning. The difference lies in the application scenario: the former involves a true value, measuring how much each data point deviates from the true value. In other words, the standard deviation (RMD) represents the relationship between a data series and the mean, while the RMSE represents the relationship between the data series and the true value. Therefore, the standard deviation measures the dispersion of a set of numbers, while the RMSE measures the deviation between observed values ​​and the true value.

[0070] 3) The variation of the deviation value over time is determined as a characteristic of the duration data. The following example uses Scheme 2 above and provides further reference. Figure 4 This will be introduced. Furthermore, in Scheme 2, each dataset includes 30 duration data points, meaning M is 29.

[0071] Figure 4 for Figure 3 This is a schematic diagram illustrating the variation of a deviation value in the maintenance method shown. (For example...) Figure 4 As shown, the horizontal axis represents the sequence number of the deviation value, and the vertical axis represents the deviation value. Taking the departure time To as an example, for the 30th workpiece, its corresponding data set includes the departure times To of the 1st workpiece to the 30th workpiece, and the deviation value of this data set is the first deviation value in the variation of deviation values ​​(i.e., the deviation value corresponding to sequence number 1); for the 31st workpiece, its corresponding data set includes the departure times To of the 2nd workpiece to the 31st workpiece, and the deviation value of this data set is the second deviation value in the variation of deviation values ​​(i.e., the deviation value corresponding to sequence number 2); for the 32nd workpiece, its corresponding data set includes the departure times To of the 3rd and 32nd workpieces, and the deviation value of this data set is the third deviation value in the variation of deviation values ​​(i.e., the deviation value corresponding to sequence number 3); and so on, the variation of deviation values ​​can be obtained.

[0072] in addition,Figure 4 The changes in the deviation values ​​shown can be considered an NP chart or NP control chart. A control chart, also called a control graph, is a statistically designed graph used to measure, record, and evaluate process quality characteristics, thereby monitoring whether the process is under control. Control charts are divided into two categories: control charts for measured values ​​and control charts for count values. Among them, count control charts include NP (for a fixed sample size of nonconforming items) charts.

[0073] In an NP control chart, upper control limits, lower control limits, and a center line can be set. Deviations between the upper and lower control limits are generally considered within the normal range, while deviations above the upper limit and below the lower limit are generally considered within the abnormal range. For example, in... Figure 4 In the control chart, the 9th and 20th deviation values ​​are above the upper control limit, indicating that the target component is malfunctioning. At this point, the target component can be repaired, or repair can be postponed. In some cases, the target component may automatically recover and resume normal operation. Repair can be carried out only when the deviation values ​​frequently fall outside the upper and lower control limits. Furthermore, in the control chart, the upper control limit, lower control limit, and center line can be straight lines or curves. In the scheme of this application, the upper control limit, lower control limit, and center line are generally curves. For example, in... Figure 4 In the diagram, the upper control limit, lower control limit, and center line are simply shown as straight lines.

[0074] S303, matching the characteristics of the duration data with the preset performance characteristics of the target component, and predicting the future performance of the target component based on the matching result; wherein, the target component includes at least one of a detection device, a conveyor belt, and a drive mechanism.

[0075] In some examples, the preset performance characteristics of the target component include fault state characteristics, which correspond to different fault types and have multiple fault state characteristics. When multiple workpieces are conveyed sequentially on the conveyor belt, each fault state characteristic includes the variation of the deviation value of the duration data of multiple workpieces in the fault state over time.

[0076] Matching the features of duration data with the preset performance features of the target component, and predicting the future performance of the target component based on the matching results, includes: matching the features of duration data with multiple fault state features, and predicting the fault type of the target component based on the matching results.

[0077] Exemplarily, the fault types can include sensor aging and inductive insensitivity or misinduction, conveyor belt aging and slippage, motor aging and weakened output torque, etc. Since the corresponding fault state features of different working components are generally different in the use cycle, and the corresponding fault state features of the same working component are generally different in different fault types, the working component about to fail and / or the fault type of the working component about to fail can be determined according to the fault state features.

[0078] In other examples, the features of the preset working performance of the target component include the state features of the reference components of the target component in the use cycle, the target component and the reference components are of the same type, and the state features in each use cycle include normal state features and fault state features; when a plurality of workpieces are sequentially conveyed on the conveyor belt, the fault state features include the change of the deviation value of the time length data of the plurality of workpieces in the fault state over time, and the normal state features include the change of the deviation value of the time length data of the plurality of workpieces in the normal state over time.

[0079] The "use cycle" is the time required for the target component to change from a normal working state to a state of stopping working due to at least one fault. The "reference components of the target component" are one or more working components of the same type as the "target component". Taking the "target component" as an electric motor, the corresponding time length data of the electric motor in the use cycle can be obtained in the following two ways, but not limited to: way 1 - obtaining the time length data of the workpieces conveyed by one electric motor of the same type in the use cycle; way 2 - obtaining the time length data of the workpieces conveyed by different electric motors of the same type in different stages of the use cycle, and then combining the time length data corresponding to the different electric motors to obtain the time length data of the electric motor in the use cycle, for example, obtaining the time length data of the workpieces conveyed by motor A in the first half of the use cycle, obtaining the time length data of the workpieces conveyed by motor B in the second half of the use cycle, and combining the time length data corresponding to motor A and the time length data corresponding to motor B to form the time length data of the "target component" - the electric motor in the use cycle.

[0080] The features of the time length data are matched with the features of the preset working performance of the target component, and the future working performance of the target component is predicted according to the matching result, including: the features of the time length data are matched with the state features of the reference components of the target component in the use cycle, and the remaining service life of the target component is predicted according to the matching result.

[0081] Further, the target components can include at least two of the detection device, the conveying belt, and the driving mechanism. First, based on the feature of the time length data, a component of the feature of the time length data corresponding to each target component is determined. Wherein, the component of the feature of the time length data is obtained by splitting the feature of the time length data, for example, when the feature of the time length data is the variation of the deviation value, a plurality of deviation values can form a deviation value variation curve, the target angles corresponding to different target components can be different, and the deviation value variation curve can be taken along the target angle to obtain a curve component (i.e. the component of the feature of the time length data) corresponding to each target component; in addition, the deviation value variation curve can also be directly split into different curve components according to the correspondence between the deviation value variation curve and the curve components of the plurality of target components, so as to obtain the component of the feature of the time length data corresponding to each target component, wherein the state feature of each target component in the use cycle can be converted into the curve component of the target component, and the curve components of the plurality of target components can be combined into a deviation value variation curve. Then, based on the state feature of the reference component corresponding to each target component in the use cycle and the component of the feature of the time length data corresponding to each target component, the remaining useful life of each target component is predicted. The determination of the remaining useful life will be specifically introduced below taking the feature of the time length data as the variation of the deviation value as an example.

[0082] Figure 5 For Figure 3 an exemplary schematic diagram of the maintenance method is shown. As Figure 5 shown, the equipment operating state can include the operating state of the detection device, the conveying belt, and the motor and other working components. Wherein, the curve cluster 1 is the variation of the deviation value corresponding to the detection device in the use cycle, the curve cluster 2 is the variation of the deviation value corresponding to the conveying belt in the use cycle, and the curve cluster 3 is the variation of the deviation value corresponding to the motor in the use cycle.

[0083] In addition, each curve cluster can include one or more than two curves, and different curves correspond to different fault types. Exemplarily, in Figure 4 , each curve cluster only shows one curve. Thus, the time required to be experienced between the end of the component of the variation of the deviation value of the same working component such as the motor when conveying a plurality of workpieces in a use time (i.e. the current state) and the end of the variation of the deviation value corresponding to the motor in the use cycle (i.e. the failure position) is the remaining useful life.

[0084] In other examples, the at least one curve cluster can include more than two curves. For example, the curve cluster 1 includes three curves, the first curve corresponds to the first sensor failure, the second curve corresponds to the second sensor failure, and the third curve corresponds to the third sensor failure. Thus, when determining the failure of the detection device, the curve corresponding to the variation of the deviation value in the first service time can be used to determine which sensor among the first sensor, the second sensor and the third sensor will fail and the remaining service life of the sensor.

[0085] In addition, the maintenance method can further include: S302', determining an average time length, the average time length being an average value of time length data of L workpieces before a current workpiece in the plurality of workpieces, L being a positive integer. In a case where an absolute value of a difference between the time length data of the current workpiece and the average time length is greater than a set value, outputting a warning signal, the warning signal indicating that the target component fails. And, S302' can be located before S302, or can be performed simultaneously with S302. Exemplarily, the current workpiece is T n , L can be 30 for example, and the set value can be 1s for example, so that the warning signal can be output in a case where ABS(T n -AVERAGE(T n-30 ~T n-1 ))>1, where ABS represents the absolute value, and AVERAGE(T n-30 ~T n-1 ) represents an average value of time length data of 30 workpieces before T n . Details will be described below with L being 30 as an example and referring to Figure 6 .

[0086] Figure 6 As shown in the maintenance method shown in Figure 3 , the running times can indicate the number of workpieces passing through. As shown in Figure 6 , taking the entering time Ti as an example, for the 31st workpiece, an average value of the entering time Ti of the first to the 30th workpieces can be calculated first, for example, 5.4s, and the average value is taken as the average time, then the absolute value Δ of the difference between the entering time Ti of the 31st workpiece and the average time is calculated to be 0.4s, which is less than 1s, so no warning signal needs to be output. For the 32nd workpiece, an average value of the entering time Ti of the second to the 31st workpieces can be calculated first, for example, 5.4s, and the average value is taken as the average time, then the absolute value Δ of the difference between the entering time Ti of the 32nd workpiece and the average time is calculated to be 0.0s, which is less than 1s, so no warning signal needs to be output.

[0087] Similarly, following the above method, it can be calculated that the absolute value Δ of the difference between the entry time Ti and the average time for workpieces 33 to 40 is less than 1 second. It can also be calculated that the absolute value Δ of the difference between the entry time Ti and the average time for workpieces 42 to 50 is less than 1 second; therefore, no warning signal needs to be output for any of them. For workpiece 41, the absolute value Δ of the difference between its entry time Ti and the average time is greater than 1 second, so a warning signal is output. At this point, the target component can be repaired, or it can be left unrepaired. In some cases, the target component can automatically return to normal operation. Repair should only be carried out when warning signals are frequently output.

[0088] Furthermore, the characteristics of the duration data can be matched with the preset performance characteristics of the target component through a prediction model, and the future performance of the target component can be predicted based on the matching results; wherein, the input of the prediction model is the duration data or the characteristics of the duration data, and the output of the prediction model is the future performance of the target component, including the remaining service life and / or fault type of the target component.

[0089] For example, when the input to the prediction model is duration data, the prediction model is trained in the following way:

[0090] First, a first sample set is acquired through a detection device. This first sample set consists of duration data for a reference component of the target component within its service life. For example, the first sample set may include duration data for multiple target components under different operating states. Operating states include normal operating states and fault operating states.

[0091] Next, the prediction model is trained based on the first sample set to enable it to determine the state characteristics of the reference component of the target component within its service life, thus obtaining the initial prediction model. Here, "state characteristics within the service life" can refer to "the changes in deviation values ​​within the service life."

[0092] In some examples, the target component includes a detection device. The duration data of the detection device during its service life (i.e., the process of the detection device going from normal operation, fault operation to shutdown) can be sequentially input into the prediction model. After processing, the changes in the deviation value of the detection device during its service life can be obtained.

[0093] In other examples, the target component includes a conveyor belt. The duration data of multiple workpieces during the service life of the conveyor belt (i.e., the conveyor belt sequentially goes from normal operation, fault operation to stopping operation) can be input into the prediction model. After processing, the change of the corresponding deviation value of the conveyor belt during its service life can be obtained.

[0094] In yet some examples, the target component includes a motor, and the time length data of multiple workpieces in a use cycle (i.e., the motor sequentially from normal operation, fault operation to stop operation) of the motor is sequentially input into the prediction model, and processing can obtain the change of the corresponding deviation value of the motor in the use cycle.

[0095] In addition, the change of the corresponding deviation value of different target components in the use cycle is generally different, and the change of the corresponding deviation value of the same target component under different fault types is also generally different. When processing the first sample set, various states can be recognized, and the characteristics of the normal operation state and the fault operation state are distinguished, and then the extracted characteristics can be used to train the machine learning model (i.e., the prediction model) to detect abnormalities, estimate the remaining useful life, and classify different types of faults.

[0096] Then, the second sample set is obtained by the detection device, and the second sample set is the time length data of multiple workpieces conveyed on the conveying belt. The second sample set is the time length data corresponding to the conveying system in a period of use.

[0097] Finally, the initial prediction model is trained based on the second sample set, so that the initial prediction model can determine the component of the feature of the time length data of the target component, and determine the remaining useful life and / or the fault type of the target component according to the state characteristics of the reference component of the target component in the use cycle and the component of the feature of the time length data of the target component, thereby obtaining the trained prediction model. Exemplarily, the "component of the feature of the time length data" can be the "component of the change of the deviation value of the time length data", and the trained model can be deployed and integrated into the control system of the industrial computer.

[0098] Taking the target component including the detection device, the conveying belt and the motor as an example, the initial prediction model is trained based on the second sample set, so that the initial prediction model can determine the feature of the time length data, and the feature of the time length data can be decomposed into the component of the feature of the time length data corresponding to the detection device, the component of the feature of the time length data corresponding to the conveying belt and the component of the feature of the time length data corresponding to the motor. In this way, the component of the feature of the time length data corresponding to the detection device is matched with the state characteristics thereof in the use cycle, so that the remaining working life of the detection device can be estimated; the component of the feature of the time length data corresponding to the conveying belt is matched with the state characteristics thereof in the use cycle, so that the remaining working life of the conveying belt can be estimated; and the component of the feature of the time length data corresponding to the motor is matched with the state characteristics thereof in the use cycle, so that the remaining working life of the motor can be estimated.

[0099] In summary, the workpiece / carrier has two consecutive flow times in a section of the assembly line, namely the entering time Ti and the leaving time To. The Ti and To are counted to obtain the time state that can be continuously monitored. The embodiments of the present application use the time sensed by the sensor for timing and statistical analysis. According to the development trend of the time state and the three failure modes, the remaining life of the workpiece to be repaired is predicted, which can maximize the use cycle of the three major workpieces of the assembly line. Without increasing the hardware cost, the preventive and predictive maintenance of the assembly line is carried out, the unplanned downtime is eliminated, and the production efficiency is improved and the maintenance cost is reduced.

[0100] In some examples, the maintenance method of the embodiments of the present application can include the following processes:

[0101] First, determine the failure of the workpiece. For example, the timing and statistical analysis of the time data recorded by the sensor can be performed, and the time data includes the entering time Ti and / or the leaving time To; calculate the difference ABS(T n -AVERAGE(T n-30 ~T n-1 ))>1, if the difference is more than 1s, a warning is given.

[0102] Second, classify the failure type through the prediction model, and predict the remaining service life of the target component. The prediction model is trained through a large number of time data control charts (such as the change of the deviation value) to identify different failure types, so as to not only determine which workpiece will fail, but also determine the failure type of the workpiece that will fail. Further, the remaining life of the three major workpieces is predicted by training and enhancing the model through machine learning.

[0103] For example, the prediction model can be trained in the following way: collect raw data, which includes time data of multiple workpieces in various normal and failure states; then, pre-process the raw data to identify various states and distinguish the characteristics of normal and failure states; then, use the extracted features to train the machine learning model to detect anomalies, classify different types of failures, and estimate the remaining service life; finally, the prediction model can be deployed and integrated into the control system for monitoring and maintenance.

[0104] In this way, the sudden mutation point that occurs accidentally can be warned, and the risk point can be identified in advance through the judgment formula. In addition, the prediction model is trained through a large amount of time data, which can detect anomalies according to the standard deviation or root mean square, analyze the failure type and estimate the remaining life, so as to realize predictive maintenance (preparing a predictive maintenance plan in advance).

[0105] Based on the method in the above embodiments, the embodiments of the present application further provide a maintenance device, wherein the maintenance device can be deployed on an industrial computer. Please refer to Figure 7 , Figure 7 FIG. 7 is a structural schematic diagram of a maintenance device provided by the embodiments of the present application. The maintenance device 700 is used in a conveying system, which comprises a detection device, a conveying belt and a driving mechanism. The driving mechanism is used to drive the conveying belt to move, and the conveying belt is used to drive the workpiece to move so that the workpiece passes through a plurality of key positions in sequence. The detection device is used to detect the action of the workpiece passing through the plurality of key positions. As shown in FIG. 7, the maintenance device 700 comprises a time length determination module 701, a feature extraction module 702 and a prediction module 703. Figure 7 The time length determination module 701 is used to obtain the detection result of the detection device to determine the time length data required for the workpiece to move between two key positions. The feature extraction module 702 is used to determine the feature of the time length data based on the time length data. The prediction module 703 is used to match the feature of the time length data with the feature of the preset working performance of the target component, and predict the future working performance of the target component according to the matching result; wherein the target component comprises at least one of the detection device, the conveying belt and the driving mechanism.

[0106] In a possible implementation, the plurality of key positions comprises an entering position, a to-position and a leaving position arranged in sequence along the conveying direction of the conveying belt; the entering position refers to the position of the workpiece entering a working area, the to-position refers to the processing position of the workpiece, and the processing position is located in the working area, and the leaving position refers to the position of the workpiece leaving the working area. The time length determination module 701 is specifically used to: obtain the action of the workpiece detected by the detection device at the entering position and the to-position to determine the entering time length of the workpiece, wherein the entering time length refers to the time length between the movement of the workpiece from the entering position to the to-position; and / or, obtain the action of the workpiece detected by the detection device at the to-position and the leaving position to determine the leaving time length of the workpiece; wherein the leaving time length refers to the time length between the movement of the workpiece from the to-position to the leaving position.

[0107] In a possible implementation, a plurality of workpieces are conveyed in sequence on the conveying belt. The time length determination module 701 is specifically used to: obtain the detection result of the detection device in sequence to determine the time length data required for each workpiece in the plurality of workpieces to move between two key positions.

[0108] In a possible implementation, a plurality of workpieces are conveyed in sequence on the conveying belt. The time length determination module 701 is specifically used to: obtain the detection result of the detection device in sequence to determine the time length data required for each workpiece in the plurality of workpieces to move between two key positions.

[0109] In a possible implementation, the feature extraction module 702 is specifically configured to: determine, based on the length data of the plurality of workpieces, a data set corresponding to a target workpiece in the plurality of workpieces; the data set includes the length data of the target workpiece and M workpieces before the target workpiece, M is a positive integer greater than or equal to 1, and the length data in the data set corresponding to adjacent target workpieces does not overlap or partially overlaps; determine, based on the data set, a deviation value corresponding to the target workpiece, the deviation value including a standard deviation and / or a root mean square error; and determine a change of the deviation value over time as a feature of the length data.

[0110] In a possible implementation, the feature of the preset working performance of the target component includes a plurality of fault state features corresponding to different fault types, and each fault state feature includes a change of a deviation value of the length data of the plurality of workpieces in a fault state over time when the plurality of workpieces are sequentially conveyed on the conveying belt. The prediction module 703 is specifically configured to: match the feature of the length data with the plurality of fault state features, and predict a fault type of the target component according to a matching result.

[0111] In a possible implementation, the feature of the preset working performance of the target component includes a state feature of a reference component of the target component in a use cycle, the target component and the reference component are of the same type, each state feature in the use cycle includes a normal state feature and a fault state feature, and the fault state feature includes a change of a deviation value of the length data of the plurality of workpieces in a fault state over time when the plurality of workpieces are sequentially conveyed on the conveying belt, and the normal state feature includes a change of a deviation value of the length data of the plurality of workpieces in a normal state over time. The prediction module 703 is specifically configured to: match the feature of the length data with the state feature of the reference component of the target component in the use cycle, and predict a remaining service life of the target component according to a matching result.

[0112] In a possible implementation, the target component includes at least two of a detection device, a conveying belt, and a driving mechanism. The prediction module 703 is specifically configured to: determine, based on the feature of the length data, a component of the feature of the length data corresponding to each target component; and predict, based on the component of the feature of the length data corresponding to each target component and a corresponding state feature of a reference component of each target component in a use cycle, a remaining service life of each target component.

[0113] In a possible implementation, the maintenance device 700 further includes a warning module 704, which is configured to: determine an average length, the average length being an average value of length data of L workpieces before a current workpiece in the plurality of workpieces, L being a positive integer; and output a warning signal in a case where an absolute value of a difference between the length data of the current workpiece and the average length is greater than a set value, the warning signal indicating that the target component is malfunctioning.

[0114] In a possible implementation, the prediction module 703 is specifically configured to: match the feature of the time length data with the feature of the preset working performance of the target component through the prediction model, and predict the future working performance of the target component according to the matching result; wherein the input of the prediction model is the feature of the time length data, the output of the prediction model is the future working performance of the target component, and includes the remaining service life of the target component and / or the fault type. In addition, when the input of the prediction model is the time length data, the feature extraction module 702 is merged into the prediction module 703, that is, the prediction module 703 has the function of the feature extraction module 702.

[0115] In a possible implementation, when the input of the prediction model is the time length data, the prediction model is trained in the following manner: obtaining a first sample set through the detection device, the first sample set being the time length data corresponding to the reference component of the target component in a use cycle; training the prediction model based on the first sample set, so that the prediction model can determine the state feature of the reference component of the target component in the use cycle, and obtain an initial prediction model; obtaining a second sample set through the detection device, the second sample set being the time length data of a plurality of workpieces conveyed on the conveying belt; training the initial prediction model based on the second sample set, so that the initial prediction model can determine the component of the feature of the time length data corresponding to the target component, and determine the remaining service life of the target component and / or the fault type according to the state feature of the reference component of the target component in the use cycle and the component of the feature of the time length data corresponding to the target component, thereby obtaining the trained prediction model.

[0116] It should be understood that the above maintenance device is used to execute the method in the above embodiments, the corresponding program modules in the maintenance device have similar implementation principles and technical effects to the description in the above method, and the working process of the maintenance device can refer to the corresponding process in the above method, which will not be described here.

[0117] In addition, based on the method in the above embodiments, the embodiments of the present application also provide a control device 800, Figure 8 The control device provided by the embodiments of the present application is shown in a structural schematic diagram. As Figure 8 shown, the control device 800 includes a processor 801 and a memory 802; the processor 801 and the memory 802 are electrically connected; the memory 802 is used to store computer program code, the computer program code includes computer instructions, when the processor 801 executes the computer instructions, the control device executes the above-mentioned maintenance method. Moreover, the number of processors and memories in the control device 800 is not limited by the present application. Wherein, the control device can be an industrial computer, and the above-mentioned pipeline maintenance method or prediction model can be deployed and integrated into the control system of the industrial computer for monitoring and maintenance.

[0118] Exemplarily, in Figure 8 The control device 800 further includes a communication interface 803 and a bus. The processor 801, the memory 802 and the communication interface 803 communicate with each other through the bus. The communication interface 803 can realize communication between the control device 800 and other devices or communication networks using a transceiving module such as but not limited to a network interface card and a transceiver. The bus can be a peripheral component interconnect (PCI) bus or an extended industry standard architecture (EISA) bus, etc. The bus can be divided into an address bus, a data bus, a control bus, etc. For ease of representation, Figure 8 In the above embodiments, only one line is used to represent the bus, but it does not mean that there is only one bus or only one type of bus. The bus can include a path for transmitting information between various components (for example, the memory 802, the processor 801, the communication interface 803) of the control device 800.

[0119] Based on the method in the above embodiments, the embodiment of the present application provides a computer readable storage medium, which stores a computer program. When the computer program runs on a processor, the processor executes the method in the above embodiments.

[0120] Based on the method in the above embodiments, the embodiment of the present application provides a computer program product, which runs on a processor. When the computer program product runs on the processor, the processor executes the method in the above embodiments.

[0121] It can be understood that the processor in the embodiments of the present application can be a central processing unit (CPU), and can also be other general-purpose processors, digital signal processors (DSPs), application specific integrated circuits (ASICs), field programmable gate arrays (FPGAs) or other programmable logic devices, transistor logic devices, hardware components or any combination thereof. The general-purpose processor can be a microprocessor or any conventional processor.

[0122] The method steps in the embodiments of the present application can be implemented by hardware, or by a combination of software and hardware executed by a processor. The software instructions can be composed of corresponding software modules, which can be stored in a random access memory (RAM), a flash memory, a read-only memory (ROM), a programmable read-only memory (PROM), an erasable PROM (EPROM), an electrically EPROM (EEPROM), a register, a hard disk, a mobile hard disk, a CD-ROM, or any other form of storage medium well known in the art. An exemplary storage medium is coupled to a processor, so that the processor can read information from the storage medium and write information to the storage medium. Of course, the storage medium can also be a component of the processor. The processor and the storage medium can be located in an ASIC.

[0123] In the above embodiments, all or some of the steps can be implemented by software, hardware, firmware or any combination thereof. When implemented by software, all or some of the steps can be implemented in the form of one or more computer programs. The computer programs are stored in a computer readable medium. The medium can be a computer readable storage medium or a computer readable transmission medium. The computer readable storage medium includes RAM, ROM, flash memory, EPROM, EEPROM, registers, hard disk, mobile hard disk, CD-ROM, DVD, etc. The computer readable transmission medium includes wires, cables, optical fiber, infrared data channel, etc. The computer readable medium is not limited to the computer readable storage medium and the computer readable transmission medium. The computer readable medium is any medium that can be accessed by a computer. The computer readable medium is a non-transitory computer readable medium. The computer readable medium does not include transitory computer readable medium, such as a modulated data signal, a carrier wave or the like.

[0124] In the above embodiments, the description of each embodiment has its own focus, and the parts not described or recorded in a certain embodiment can be referred to the relevant description of other embodiments. It should be understood that the size of the serial number of each step in the above embodiments does not mean the order of execution, and the execution order of each process should be determined according to its function and inherent logic, and should not constitute any limitation on the implementation process of the embodiments of the present application. Moreover, the various digital numbers involved in the embodiments of the present application are only for the convenience of differentiation in the description, and are not intended to limit the scope of the embodiments of the present application.

[0125] The basic principles of the present application are described above in combination with specific embodiments, but it should be pointed out that the advantages, advantages, effects and the like mentioned in the present application are only examples and not limitations, and these advantages, advantages, effects and the like cannot be considered as the must-have of each embodiment of the present disclosure. In addition, the specific details of the above disclosure are only for the purpose of example and for the purpose of understanding, and are not limited to the above specific details.

Claims

1. A maintenance method for a transport system, characterized in that The conveying system comprises a detection device, a conveying belt and a driving mechanism, the driving mechanism is used for driving the conveying belt to move, and the conveying belt is used for moving the workpiece to make the workpiece pass through a plurality of key positions in sequence; The detection device is used for detecting the action of the workpiece passing through the plurality of key positions; The maintenance method comprises: obtaining the detection result of the detection device to determine the time length data required for the workpiece to move between two key positions; Based on the time length data, the characteristics of the time length data are determined, specifically including: Based on the time length data of a plurality of workpieces, a data set corresponding to a target workpiece in the plurality of workpieces is determined; wherein the data set includes the time length data of the target workpiece and M workpieces before the target workpiece, M is a positive integer greater than or equal to 1, and the time length data in the data set corresponding to adjacent target workpieces does not overlap or partially overlaps; Based on the data set, a deviation value corresponding to the target workpiece is determined, and the deviation value includes a standard deviation and / or a root mean square error; The change of the deviation value with time is determined as the characteristics of the time length data; The characteristics of the time length data are matched with the preset working performance characteristics of the target component, and the future working performance of the target component is predicted according to the matching result; wherein the target component includes at least one of the detection device, the conveying belt and the driving mechanism, the preset working performance characteristics of the target component include the state characteristics of a reference component of the target component in a use cycle, the target component and the reference component are of the same type, each state characteristic in the use cycle includes a normal state characteristic and a fault state characteristic; when a plurality of workpieces are conveyed on the conveying belt in sequence, the fault state characteristic includes the change of the deviation value of the time length data of the plurality of workpieces in the fault state with time, and the normal state characteristic includes the change of the deviation value of the time length data of the plurality of workpieces in the normal state with time; the target component includes at least two of the detection device, the conveying belt and the driving mechanism; The matching of the characteristics of the time length data with the preset working performance characteristics of the target component, and the prediction of the future working performance of the target component according to the matching result, comprises: Based on the characteristics of the time length data, a component of the characteristics of the time length data corresponding to each target component is determined; Based on the corresponding state characteristics of the reference component of each target component in the use cycle and the component of the characteristics of the time length data corresponding to each target component, the remaining service life of each target component is predicted.

2. The method of maintenance according to claim 1, characterized in that, The plurality of key positions include an entering position, a to-position and a leaving position arranged in sequence along the conveying direction of the conveying belt; the entering position refers to the position of the workpiece entering a working area, the to-position refers to the processing position of the workpiece, and the processing position is located in the working area, and the leaving position refers to the position of the workpiece leaving the working area. The detection result of the detection device is acquired to determine the time length data required for the workpiece to move between the two key positions. The detection result of the detection device is acquired to determine the time length data required for the workpiece to move between the two key positions. The detection result of the detection device is acquired to determine the time length data required for the workpiece to move between the two key positions.

3. The method of claim 1, wherein, The detection result of the detection device is acquired to determine the time length data required for the workpiece to move between the two key positions. The detection result of the detection device is acquired to determine the time length data required for the workpiece to move between the two key positions.

4. The method of maintenance according to any one of claims 1-3, characterized in that, The detection result of the detection device is acquired to determine the time length data required for the workpiece to move between the two key positions. The detection result of the detection device is acquired to determine the time length data required for the workpiece to move between the two key positions. The detection result of the detection device is acquired to determine the time length data required for the workpiece to move between the two key positions.

5. The method of maintenance according to claim 4, characterized in that, The detection result of the detection device is acquired to determine the time length data required for the workpiece to move between the two key positions. The detection result of the detection device is acquired to determine the time length data required for the workpiece to move between the two key positions. The detection result of the detection device is acquired to determine the time length data required for the workpiece to move between the two key positions.

6. The method of maintenance according to any one of claims 1-3, characterized in that, The detection result of the detection device is acquired to determine the time length data required for the workpiece to move between the two key positions. The detection result of the detection device is acquired to determine the time length data required for the workpiece to move between the two key positions.

7. The method of maintenance according to claim 6, characterized in that, The detection result of the detection device is acquired to determine the time length data required for the workpiece to move between the two key positions. The detection result of the detection device is acquired to determine the time length data required for the workpiece to move between the two key positions. The detection result of the detection device is acquired to determine the time length data required for the workpiece to move between the two key positions. The detection result of the detection device is acquired to determine the time length data required for the workpiece to move between the two key positions. The detection result of the detection device is acquired to determine the time length data required for the workpiece to move between the two key positions. The detection result of the detection device is acquired to determine the time length data required for the workpiece to move between the two key positions. The detection result of the detection device is acquired to determine the time length data required for the workpiece to move between the two key positions. The detection result of the detection device is acquired to determine the time length data required for the workpiece to move between the two key positions. The detection result of the detection device is acquired to determine the time length data required for the workpiece to move between the two key positions. The detection result of the detection device is acquired to determine the time length data required for the workpiece to move between the two key positions. The detection result of the detection device is acquired to determine the time length data required for the workpiece to move between the two key positions. The detection result of the detection device is acquired to determine the time length data required for the workpiece to move between the two key positions. The detection result of the detection device is acquired to determine the time length data required for the workpiece to move between the two key positions. The detection result of the detection device is acquired to determine the time length data required for the workpiece to move between the two key positions. The detection result of the detection device is acquired to determine the time length data required for the workpiece to move between the two key positions. The detection result of the detection device is acquired to determine the time length data required for the workpiece to move between the two key positions. The detection result of the detection device is acquired to determine the time length data required for the workpiece to move between the two key positions. The detection result of the detection device is acquired to determine the time length data required for the workpiece to move between the two key positions. The detection result of the detection device is acquired to determine the time length data required for the workpiece to move between the two key positions. The detection result of the detection device is acquired to determine the time length data required for the workpiece to move between the two key positions. The detection result of the detection device is acquired to determine the time length data required for the workpiece to move between the two key positions. The detection result of the detection device is acquired to determine the time length data required for the workpiece to move between the two key positions. The detection result of the detection device is acquired to determine the time length data required for the workpiece to move between the two key positions. The detection result of the detection device is acquired to determine the time length data required for the workpiece to move between the two key positions. The detection result of the detection device is acquired to determine the time length data required for the workpiece to move between the two key positions. The detection result of the detection device is acquired to determine the time length data required for the workpiece to move between the two key positions. The detection result of the detection device is acquired to determine the time length data required for the workpiece to move between the two key positions. The detection result of the detection device is acquired to determine the time length data required for the workpiece to move between the two key positions. The detection result of the detection device is acquired to determine the time length data required for the workpiece to move between the two key positions. The detection result of the detection device is acquired to determine the time length data required for the workpiece to move between the two key positions. The detection result of the detection device is acquired to determine the time length data required for the workpiece to move between the two key positions. The detection result of the detection device is acquired to determine the time length data required for the workpiece to move between the two key positions. The detection result of the detection device is acquired to determine the time length data required for the workpiece to move between the two key positions. The detection result of the detection device is acquired to determine the time length data required for the workpiece to move between the two key positions. The detection result of the detection device is acquired to determine the time length data required for the workpiece to move between the two key positions. The detection result of the detection device is acquired to determine the time length data required for the workpiece to move between the two key positions. The detection result of the detection device is acquired to determine the time length data required for the workpiece to move between the two key positions. The detection result of the detection device is acquired to determine the time length data required for the workpiece to move between the two key positions. The detection result of the detection device is acquired to determine the time length data required for the workpiece to move between the two key positions. The detection result of the detection device is acquired to determine the time length data required for the workpiece to move between the two key positions. The detection result of the detection device is acquired to determine the time length data required for the workpiece to move between the two key positions. The detection result of the detection device is acquired to determine the time length data required for the workpiece to move between the two key positions. The detection result of the detection device is acquired to determine the time length data required for the workpiece to move between the two key positions. The detection result of the detection device is acquired to determine the time length data required for the workpiece to move between the two key positions. The detection result of the detection device is acquired to determine the time length data required for the workpiece to move between the two key positions. The detection result of the detection device is acquired to determine the time length data required for the workpiece to move between the two key positions. The detection result of the detection device is acquired to determine the time length data required for the workpiece to move between the two key positions. The detection result of the detection device is acquired to determine the time length data required for the workpiece to move between the two key positions. The detection result of the detection device is acquired to determine the time length data required for the workpiece to move between the two key positions. The detection result of the detection device is acquired to determine the time length data required for the workpiece to move between the two key positions. The detection result of the detection device is acquired to determine the time length data required for the workpiece to move between the two key positions. The detection result of the detection device is acquired to determine the time length data required for the workpiece to move between the two key positions. The detection result of the detection device is acquired to determine the time length data required for the workpiece to move between the two key positions. The detection result of the detection device is acquired to determine the time length data required for the workpiece to move between the two key positions. The detection result of the detection device is acquired to determine the time length data required for the workpiece to move between the two key positions. The detection result of the detection device is acquired to determine the time length data required for the workpiece to move between the two key positions. The detection result of the detection device is acquired to determine the time length data required for the workpiece to move between the two key positions. The detection result of the detection device is acquired to determine the time length data required for the workpiece to move between the two key positions. The detection result of the detection device is acquired to determine the time length data required for the workpiece to move between the two key positions. The detection result of the detection device is acquired to determine the time length data required for the workpiece to move between the two key positions. The detection result of acquiring a second sample set by the detection device, the second sample set being time length data of a plurality of workpieces conveyed on the conveying belt; training the initial prediction model based on the second sample set, so that the initial prediction model can determine a component of a feature of time length data corresponding to the target component, and determine the remaining service life and / or the failure type of the target component according to the state feature of the reference component of the target component in the use cycle and the component of the feature of the time length data of the target component, thereby obtaining a trained prediction model.

Images