Method and device for determining fault position of production line equipment, equipment and storage medium

Abstract

The invention discloses a method and device for determining a fault position of production line equipment, equipment and a storage medium. Acquiring equipment information of production line equipment in the production line; according to the equipment information, a fault diagnosis model of the production line equipment is determined by using the target fault tree, the fault diagnosis model comprises at least two fault levels, each fault level comprises at least one fault mode, and a dependency relationship exists between the adjacent fault levels; determining troubleshooting duration, troubleshooting cost and troubleshooting path complexity required by each type of fault mode; according to the troubleshooting duration, the troubleshooting cost and the troubleshooting path complexity, a troubleshooting path of the fault diagnosis model is determined, and the troubleshooting path is a troubleshooting path from the highest fault level to the lowest fault level; according to the troubleshooting path, the fault position of the production line equipment is determined, and the specific fault position of the production line equipment can be rapidly and accurately determined.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of fault diagnosis, and in particular to a method and device for determining the location of a fault in a production line, a storage medium, and a computer program product. BACKGROUND

[0002] With the rapid development of the Internet of Things technology, the demand for communication modules by terminal devices is growing exponentially. As a key hardware for realizing "Internet of Everything", the production, quality control, and post-maintenance of modules have formed a complete life cycle management system. The production link is the premise for ensuring the reliable delivery of modules, and any quality defects will directly affect the stability of downstream terminal products and user experience. Therefore, how to achieve fault positioning for a module production line is crucial.

[0003] A module production line is a complex system with nonlinearity, dynamics, and cascading effects. Traditional troubleshooting methods often rely on statistics of the failure rate of each product storage area to infer the root cause of the fault and locate the fault in the module production line. However, only by statistically analyzing the failure rate of each product storage area to infer the root cause of the fault, it is often only possible to provide a rough estimate of the failure probability, making it difficult to quickly and accurately locate the specific fault point in the module production line. SUMMARY

[0004] The embodiments of the present application provide a method and device for determining the location of a fault in a production line, a storage medium, and a computer program product, which can quickly and accurately locate the specific fault point in the module production line.

[0005] In a first aspect, the embodiments of the present application provide a method for determining the location of a fault in a production line, comprising: obtaining device information of a production line device in a production line; determining a fault diagnosis model of the production line device using a target fault tree according to the device information, the fault diagnosis model comprising at least two fault levels, each fault level comprising at least one type of fault mode, and there being a dependency relationship between adjacent fault levels; determining the troubleshooting duration, troubleshooting cost, and troubleshooting path complexity required for each type of fault mode; determining a troubleshooting path for the fault diagnosis model according to the troubleshooting duration, troubleshooting cost, and troubleshooting path complexity, the troubleshooting path being a troubleshooting path from the highest fault level to the lowest fault level; determining the location of the fault in the production line device according to the troubleshooting path.

[0006] In one possible implementation, the target fault tree is one of a plurality of fault trees; and before determining the fault diagnosis model of the production line device using the target fault tree according to the device information, the method further comprises: an objective selection manner for obtaining a target fault tree; selecting the target fault tree from the plurality of fault trees by using the objective selection manner.

[0007] In one possible implementation, the objective selection manner for obtaining a target fault tree comprises: obtaining a fault locating duration for locating a fault by each selected fault tree in a preset time period; obtaining a target time point for selecting a fault tree, a historical average fault locating duration at the target time point, and a production line change rate, the target time point being any time point in the preset time period; determining a random probability parameter according to the target time point, the historical average fault locating duration at the target time point, and the production line change rate; determining the objective selection manner for the target fault tree according to the random probability parameter.

[0008] In one possible implementation, the objective selection manner comprises selecting a target fault tree with a shortest fault locating duration; and the selecting the target fault tree from the plurality of fault trees by using the objective selection manner comprises: obtaining a fault locating duration for locating a fault by each selected fault tree in a preset time period; selecting a target fault tree with a shortest fault locating duration from the plurality of fault trees according to the fault locating duration.

[0009] In one possible implementation, the objective selection manner comprises randomly selecting a target fault tree; and the selecting the target fault tree from the plurality of fault trees by using the objective selection manner comprises: randomly selecting a target fault tree from the plurality of fault trees.

[0010] In one possible implementation, the determining the locating duration and the locating cost required for each fault mode comprises: obtaining the locating duration and the locating cost required for each fault mode in the lowest fault level; obtaining a first probability of occurrence of each fault mode in all fault levels, the first probability being obtained by a frequency of historical occurrence of the fault mode; determining the locating duration and the locating cost required for each fault mode other than the fault mode in the lowest fault level according to the first probability and the locating duration and the locating cost of each fault mode in the lowest fault level.

[0011] In one possible implementation, the determining the locating path complexity of each fault mode comprises: determining the locating path complexity of each fault mode according to a number of fault modes associated with each fault mode.

[0012] In one possible implementation, the troubleshooting path of the fault diagnosis model is determined according to the troubleshooting duration, the troubleshooting cost, and the troubleshooting path complexity, including: obtaining priorities of the troubleshooting duration, the troubleshooting cost, and the troubleshooting path complexity; for each type of fault mode, determining the troubleshooting path of the fault diagnosis model according to the priorities of the troubleshooting duration, the troubleshooting cost, and the troubleshooting path complexity.

[0013] In one possible implementation, the method further includes: obtaining historical fault modes; determining, according to the historical fault modes, a second probability of each type of fault mode in a highest fault level occurring under the condition that each type of fault mode in a lowest fault level occurs, to obtain a fault transition matrix, the target fault mode being any one of the historical fault modes; determining, under the condition that each type of fault mode in the lowest fault level occurs, a third probability of each type of fault mode in the highest fault level occurring; determining, according to the fault transition matrix and the third probability, a sequence of the probability of each type of fault mode in the historical fault modes occurring and a fault mode sequence corresponding to the sequence.

[0014] In one possible implementation, the method further includes: determining a set of minimal cut sets that cause the fault modes in the highest fault level to occur, each minimal cut set including at least one type of fault mode in the lowest fault level; determining, according to the set of minimal cut sets, a first fault probability function of each type of fault mode in the highest fault level; determining, according to the first fault probability function, a second fault probability function of all fault modes in the highest fault level; obtaining the troubleshooting duration, the troubleshooting cost, the optimization cost, the historical occurrence frequency, and the reliability requirement information of the production line equipment of each minimal cut set; determining, according to the troubleshooting duration, the troubleshooting cost, the optimization cost, and the historical occurrence frequency of each minimal cut set, a score of each minimal cut set by using a preset scoring rule; constructing an objective function according to the second fault probability function, the score of each minimal cut set, the reliability requirement information, and the relationship information between the product life; solving the objective function to obtain a target minimal cut set, the target minimal cut set including at least one type of fault mode; displaying the target minimal cut set to guide a user to optimize the production line equipment.

[0015] In one possible implementation, solving the objective function to obtain the target minimal cut set includes: The particle swarm optimization algorithm is used to solve the target function, and a target minimum cut set is obtained. The solving process of the particle swarm optimization algorithm is as follows: An iteration number and a value range of a learning factor are obtained. A selection function of the learning factor is fitted according to the iteration number and the value range of the learning factor, so as to select the learning factor in each iteration process according to the selection function; The fitness of the position reached in each iteration process is determined according to the learning factor, and a descending sequence of the fitness is determined, and the position represents a minimum cut set. A target fitness satisfying a preset reservation condition is obtained, the preset reservation condition is to reserve the first preset number of target fitnesses in the descending sequence, and the target fitness is randomly selected from the fitnesses in the second preset number of the descending sequence according to the fourth probability; A target position corresponding to the target fitness is determined, and the target position represents a target minimum cut set.

[0016] In one possible implementation, the method further includes: A production line model corresponding to the production line equipment is constructed, and the production line model includes personnel information, and the personnel information includes personnel position information; In a case where the personnel position information represents that the personnel is located outside the preset area, an abnormal alarm of a position corresponding to the personnel position information is triggered.

[0017] In one possible implementation, the production line model further includes equipment information; and the method further includes: According to the equipment information, a production line equipment image corresponding to the production line model is displayed, and a position where the production line equipment fails is marked in the production line equipment image.

[0018] In a second aspect, the embodiments of the present application provide a device for determining a fault position of production line equipment, including: An obtaining module is configured to obtain equipment information of production line equipment in a production line; A determining module is configured to determine a fault diagnosis model of the production line equipment by using a target fault tree according to the equipment information, the fault diagnosis model including at least two fault levels, each fault level including at least one type of fault mode, and there being a dependency relationship between adjacent fault levels; The determining module is further configured to determine an investigation time length, an investigation cost and an investigation path complexity required for each type of fault mode; The determining module is further configured to determine a fault investigation path of the fault diagnosis model according to the investigation time length, the investigation cost and the investigation path complexity, the fault investigation path being a fault investigation path from a highest fault level to a lowest fault level; The determining module is further configured to determine the location of the fault of the production line equipment according to the troubleshooting path.

[0019] In a third aspect, an electronic device is provided, and the device includes: A processor and a memory storing computer program instructions; the processor executes the computer program instructions to implement the method for determining the location of the fault of the production line equipment.

[0020] In a fourth aspect, a computer storage medium is provided, and the computer readable storage medium stores computer program instructions, and the computer program instructions are executed by a processor to implement the method for determining the location of the fault of the production line equipment.

[0021] In a fifth aspect, a computer program product is provided, and instructions in the computer program product are executed by a processor of an electronic device to enable the electronic device to execute the method for determining the location of the fault of the production line equipment.

[0022] The method, device, equipment, storage medium and computer program product for determining the location of the fault of the production line equipment provided in the embodiments of the present application obtain the equipment information of the production line equipment in the production line; according to the equipment information, the fault diagnosis model of the production line equipment is determined by using the target fault tree, the fault diagnosis model includes at least two fault levels, each fault level includes at least one type of fault mode, and there is a dependency relationship between adjacent fault levels; the troubleshooting time, troubleshooting cost and troubleshooting path complexity required by each type of fault mode are determined; according to the troubleshooting time, troubleshooting cost and troubleshooting path complexity, the troubleshooting path of the fault diagnosis model is determined, and the troubleshooting path is the troubleshooting path from the highest fault level to the lowest fault level; and the location of the fault of the production line equipment is determined according to the troubleshooting path. By using the fault tree, the fault level of the fault of the production line equipment and the fault mode of each fault level are determined, the troubleshooting path from the highest fault level to the lowest fault level in the fault diagnosis model is determined according to the troubleshooting time, troubleshooting cost and troubleshooting path complexity required by each type of fault mode, and the location of the fault of the production line equipment is determined according to the troubleshooting path, so that the specific location of the fault of the production line equipment can be quickly and accurately determined. BRIEF DESCRIPTION OF DRAWINGS

[0023] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings required to be used in the embodiments of the present application will be briefly introduced as follows, and other drawings can also be obtained by those of ordinary skill in the art without creative labor on the premise that they do not pay creative labor.

[0024] Figure 1is a flowchart of a method for determining a fault position of a production line device provided by an embodiment of the present application; Figure 2 is a schematic diagram of a fault tree of the FMA mode and the SMM mode provided by another embodiment of the present application; Figure 3 is a schematic diagram of a fault diagnosis model provided by yet another embodiment of the present application; Figure 4 is a schematic diagram of an AND gate and an OR gate provided by still another embodiment of the present application; Figure 5 is a flowchart of a method for determining a fault position of a production line device provided by still another embodiment of the present application; Figure 6 is a flowchart of a method for determining a fault position of a production line device provided by still another embodiment of the present application; Figure 7 is a flowchart of a method for determining a fault position of a production line device provided by still another embodiment of the present application; Figure 8 is a flowchart of a method for determining a fault position of a production line device provided by still another embodiment of the present application; Figure 9 is a flowchart of a particle swarm optimization algorithm provided by still another embodiment of the present application; Figure 10 is a structural schematic diagram of a device for determining a fault position of a production line device provided by still another embodiment of the present application; Figure 11 is a structural schematic diagram of an electronic device provided by still another embodiment of the present application. DETAILED DESCRIPTION

[0025] The features and exemplary embodiments of various aspects of the present application will be described in detail below, in order to make the purposes, technical solutions and advantages of the present application clearer, the present application will be further described in detail below in combination with the drawings and specific embodiments. It should be understood that the specific embodiments described herein are only intended to explain the present application, but not to limit the present application. The present application can be implemented without some of these specific details by those skilled in the art. The following description of the embodiments is only to provide a better understanding of the present application by showing examples of the present application.

[0026] It is to be noted that, in the present document, relational terms such as first and second and the like can be used solely to distinguish one entity or action from another entity or action without necessarily requiring or implying any actual such relationship or order between such entities or actions. Moreover, the terms "comprises", "comprising", or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but can include other elements not expressly listed or inherent to such process, method, article, or apparatus. An element proceeded by "comprises... a" does not, without more constraints, exclude the presence of additional identical elements in the process, method, article, or apparatus that comprises the element.

[0027] With the rapid development of Internet of Things technology, the demand for communication modules by terminal devices is growing exponentially. As the key hardware for realizing "Internet of Everything", the production, quality control and post-maintenance of modules have formed a complete life cycle management system. The production link is the premise to ensure the reliable delivery of modules, and any quality defects will directly affect the stability of downstream terminal products and user experience. Therefore, how to locate the fault of the module production line is crucial.

[0028] The module production line is a complex system with nonlinearity, dynamics and cascade effect. The traditional fault troubleshooting method can only rely on the statistical defective occurrence rate of each product storage area to infer the root cause of the fault, and locate the fault of the module production line. However, only by statistically inferring the root cause of the fault through the defective occurrence rate of each product storage area, a rough fault probability estimate can be provided, and it is difficult to quickly and accurately locate the specific fault point position in the module production line.

[0029] In order to solve the problems in the prior art, the embodiments of the present application provide a method, device, equipment, storage medium and computer program product for determining the fault position of a production line device. The embodiments of the present application determine the fault level of the production line device and the fault mode of each fault level through a fault tree, determine the fault troubleshooting path from the highest fault level to the lowest fault level in the fault diagnosis model according to the troubleshooting time, troubleshooting cost and troubleshooting path complexity required by each fault mode, and determine the position of the fault of the production line device according to the fault troubleshooting path, so as to quickly and accurately determine the specific position of the fault of the production line device.

[0030] The method for determining the fault position of the production line device provided by the embodiments of the present application will be described in detail below.

[0031] As shown in FIG. 1, the method for determining the fault position of the production line device provided by the embodiments of the present application includes the following steps S110-S150. Figure 1 As shown in FIG. 1, the method for determining the fault position of the production line device provided by the embodiments of the present application includes the following steps S110-S150.

[0032] S110, obtaining device information of the production line equipment in the production line.

[0033] The production line equipment includes a module production line equipment.

[0034] In some embodiments, a module production line is generally composed of several different stations, each of which is responsible for a corresponding function, such as a download station for firmware download and marking, a calibration station for radio frequency calibration and analysis, etc., so each station can correspond to different production line equipment. The device information of the production line equipment and the product information of the product corresponding to the production line equipment, such as test information, are uploaded to the factory manufacturing execution system (Manufacturing Execution System, MES) in real time, and the position information of the personnel corresponding to each station can be collected through a radio frequency identification tag (Radio Frequency Identification Tag, RFID). The device information, product information and personnel position information are stored in the factory corresponding database to realize the implementation control and data traceability of the production process. The device information of the production line equipment in the production line is obtained from the factory corresponding database.

[0035] Specifically, each production line equipment can be decomposed into components, elements, etc., and each production line equipment can also perform functions, movements, actions, etc. Therefore, the device information of the production line equipment can include component information, element information, function information, movement information and action information.

[0036] It can be understood that a function can be decomposed into at least one movement, and a movement can be decomposed into at least one action. Taking the battery cell stacking function as an example, it is decomposed according to function, movement and action as follows: The battery cell stacking function is decomposed into battery cell carrying movement, battery cell positioning movement and battery cell arrangement movement. The battery cell carrying movement is decomposed into the following actions: the end effector (such as a suction cup or a clamping jaw) of the mechanical arm first moves above the battery cell, then the suction cup adsorbs or the clamping jaw clamps the battery cell, then the mechanical arm carries the battery cell to the stacking position according to the predetermined trajectory, and finally the suction cup releases or the clamping jaw releases the battery cell. The battery cell positioning movement and the battery cell arrangement movement are similar and will not be described in detail here.

[0037] S120, determining a fault diagnosis model of the production line equipment by using a target fault tree according to the device information, the fault diagnosis model including at least two fault levels, each fault level including at least one type of fault mode, and there being a dependency relationship between adjacent fault levels.

[0038] In some embodiments, the target fault tree can be a fault tree of a Fault Mode Analysis (FMA) mode. According to the device information, a fault diagnosis model of the production line device is determined by using the fault tree of the FMA mode.

[0039] In some embodiments, the target fault tree can be a fault tree of a System Maintenance Management (SMM) mode. According to the device information, a fault diagnosis model of the production line device is determined by using the fault tree of the SMM mode.

[0040] In one example, the FMA mode and the SMM mode are as shown in Figure 2 . The FMA refers to logical decomposition of a device according to "function-motion-action", wherein an indivisible action is called a meta-action, and a component implementing the meta-action is called a meta-action unit, which is more suitable for mechanical and electrical products; the SMM refers to logical decomposition of a device according to "device-component-element", wherein an indivisible element is called a basic element, which is suitable for general electronic products. Figure 2 The system in the FMA mode and the SMM mode refers to a production line device. In the FMA mode, function, motion, and action respectively represent three event levels, such as top-level events, intermediate events, and basic events, which are called fault levels in the case of occurrence of faults at the event levels. Therefore, by using the fault tree of the FMA mode, a fault diagnosis model of the production line device is determined, and the fault diagnosis model includes fault levels respectively indicating occurrence of faults at the three event levels of function, motion, and action, and each fault level includes at least one fault mode indicating occurrence of a certain function, motion, or action at the three event levels of function, motion, and action. In the SMM mode, device, component, and element respectively represent three event levels, such as top-level events, intermediate events, and basic events, which are called fault levels in the case of occurrence of faults at the event levels. Therefore, by using the fault tree of the SMM mode, a fault diagnosis model of the production line device is determined, and the fault diagnosis model includes fault levels respectively indicating occurrence of faults at the three event levels of device, component, and element, and each fault level includes at least one fault mode indicating occurrence of a certain device, component, or element at the three event levels of device, component, and element.

[0041] Specifically, occurrence of each event is independent of each other. According to the fault tree, critical fault modes of all actions or elements are determined , , wherein represents the nth fault mode, and then each fault mode is sequentially taken as a top-level event, and is decomposed into basic events by using a logic gate , , i represents the number of basic events, i∈N+, and the structure of the fault diagnosis model is as shown in Figure 3 .

[0042] S130, determine the troubleshooting time, troubleshooting cost and troubleshooting path complexity required for each fault mode.

[0043] The advantage of the fault tree is simple, clear logical steps, and easy to troubleshoot, but the fault diagnosis model based on the production line equipment is complex, the tree chart is huge, and the sequential troubleshooting may take a particularly long time, so the retrieval efficiency needs to be improved. Based on this, in some embodiments, by setting multiple limiting conditions such as troubleshooting time, troubleshooting cost and troubleshooting path complexity, the appropriate branch in the fault diagnosis model is sorted and selected, which can greatly reduce the overall fault retrieval time.

[0044] In some embodiments, according to the historical troubleshooting time and historical troubleshooting cost of each fault mode, the troubleshooting time and troubleshooting cost required for each fault mode in the fault diagnosis model are determined. According to the number of nodes in the fault troubleshooting path of each node in the fault diagnosis model, the troubleshooting path complexity of each fault mode is determined. Wherein, the node represents the fault mode.

[0045] In some embodiments, in the above S130, the troubleshooting time and troubleshooting cost required for each fault mode are determined, including: Obtain the troubleshooting time and troubleshooting cost required for each fault mode in the lowest fault level; Obtain the first probability of each fault mode occurring in all fault levels, which is obtained by the frequency of historical occurrence of the fault mode; According to the first probability and the troubleshooting time and troubleshooting cost of each fault mode in the lowest fault level, determine the troubleshooting time and troubleshooting cost required for each fault mode other than the fault mode in the lowest fault level.

[0046] In this embodiment, each fault mode in the lowest fault level is a basic event failure, such as a component failure. And the troubleshooting time and troubleshooting cost of whether the component fails are determined according to the component properties, that is, fixed, and the troubleshooting time and troubleshooting cost required for each fault mode in the lowest fault level can be directly obtained from the recorded component properties. The troubleshooting time and troubleshooting cost required for each fault mode other than the fault mode in the lowest fault level (such as determining whether the intermediate event and the top event fail) need to be determined according to the probability of each fault mode in the lowest fault level and the troubleshooting time and troubleshooting cost required for each fault mode in the lowest fault level.

[0047] Specifically, the probability of the initial fault occurrence of each fault mode, i.e. each basic event, intermediate event and top event, is calculated from the lowest fault level of the fault diagnosis model upwards. The fault diagnosis model mainly involves two kinds of logic structures, and gates and or gates, such asFigure 4 as shown.

[0048] Basic events The probability of failure occurrence is represented by the historical failure occurrence frequency, as shown in equation (1): (1) wherein, represents the historical failure occurrence frequency of the ith basic event, and k represents the number of basic events.

[0049] The probability of occurrence of events with different logical structures is calculated as follows: equation (2) is for an OR gate, and equation (3) is for an AND gate: (2) (3) wherein, and both represent event occurrence (1 represents occurrence, and 0 represents non-occurrence).

[0050] Troubleshooting is performed from top to bottom, so the first probability of failure occurrence of basic events under different branch failure conditions needs to be calculated. The conditional probability of events with two kinds of logical structures is calculated as follows: equation (4) is for an OR gate, and equation (5) is for an AND gate: (4) (5) The troubleshooting duration and troubleshooting cost of each middle-level event or top-level event are calculated from bottom to top (starting from basic events and ending at top-level events). Equation (6) is for an OR gate, and equation (7) is for an AND gate: ; (6) ; (7) wherein, , respectively represent the troubleshooting duration of the middle-level event and the top-level event in an OR gate structure, and the adjacent lower-level event; , respectively represent the troubleshooting duration of the middle-level event and the top-level event in an AND gate structure, and the adjacent lower-level event; , respectively represent the troubleshooting cost of the middle-level event and the top-level event in an OR gate structure, and the adjacent lower-level event; , respectively represent the troubleshooting cost of the middle-level event and the top-level event in an AND gate structure, and the adjacent lower-level event; represents the first probability of failure occurrence of basic events in an OR gate structure.

[0051] The embodiments of the present application determine the troubleshooting time length and troubleshooting cost required for each type of fault mode through the above formulas (1)-(7), can first troubleshoot whether the event with a large fault occurrence probability, a short troubleshooting time length and a low troubleshooting cost has a fault, and improves the troubleshooting efficiency of the fault.

[0052] In some embodiments, in the above S130, the troubleshooting path complexity of each type of fault mode is determined, including: According to the number of fault modes associated with each type of fault mode, the troubleshooting path complexity of each type of fault mode is determined.

[0053] In this embodiment, the troubleshooting path complexity is mainly related to the subsequent path depth and breadth, and can be defined as the maximum number of troubleshooting steps, that is, the node number of each fault troubleshooting path remaining in the current fault troubleshooting path is accumulated. Assuming that there are n exploration paths for a node, the calculation formula (8) of the troubleshooting path complexity is as follows: (8) Among them, represents the node number of the i th fault troubleshooting path.

[0054] Through the number of fault modes associated with each type of fault mode, the embodiments of the present application can intuitively and accurately reflect the troubleshooting path complexity of each type of fault mode.

[0055] In some embodiments, the troubleshooting time length, the troubleshooting cost and the troubleshooting path complexity required for each type of fault mode are determined as the attributes of the fault mode. The attribute vector can be represented as , t represents the troubleshooting time length required for the fault mode, s represents the loss that the fault mode can cause, such as the troubleshooting cost, the troubleshooting cost can include the shutdown inspection fee, and z represents the troubleshooting path complexity of the fault mode.

[0056] S140, according to the troubleshooting time length, the troubleshooting cost and the troubleshooting path complexity, the fault diagnosis model is determined. The fault troubleshooting path is the fault troubleshooting path from the highest fault level to the lowest fault level.

[0057] In some embodiments, the weights of the troubleshooting time length, the troubleshooting cost and the troubleshooting path complexity are obtained, the troubleshooting time length, the troubleshooting cost and the troubleshooting path complexity are weighted and summed according to the obtained weights, and the score of each type of fault mode is obtained. When determining the next node of each node in the fault troubleshooting path, the node with the highest score is selected, and the fault mode represented by the node is the next troubleshooting object. According to the score, the objects to be troubleshoot are selected in turn, and finally the fault troubleshooting path of the fault diagnosis model is obtained. The fault troubleshooting path is the fault troubleshooting path from the highest fault level to the lowest fault level.

[0058] The higher the score, the lower the troubleshooting duration, troubleshooting cost, and troubleshooting path complexity.

[0059] In some embodiments, as shown in FIG. 13, S140 can include S141-S142. Figure 5

[0060] S141, obtaining priorities of the troubleshooting duration, the troubleshooting cost, and the troubleshooting path complexity.

[0061] In some embodiments, the priorities of the troubleshooting duration, the troubleshooting cost, and the troubleshooting path complexity are received as user inputs. The user can manually set the priorities of the conditions on the machine according to the actual situation, thereby improving the user experience.

[0062] S142, for each type of fault mode, determining a troubleshooting path of the fault diagnosis model according to the priorities of the troubleshooting duration, the troubleshooting cost, and the troubleshooting path complexity.

[0063] In this embodiment, a decision is made each time a branch node is selected. The general idea is to sort according to the high-priority condition, and then troubleshoot in order. If the decision results of two nodes are equal, the next priority condition is selected for sorting and decision-making. There are three priority conditions, which are the troubleshooting duration, the troubleshooting cost, and the troubleshooting path complexity.

[0064] Specifically, the priority conditions can be divided into a first priority condition, a second priority condition, and a third priority condition according to the size of the priority. The priority of the first priority condition is greater than the priority of the second priority condition, and the priority of the second priority condition is greater than the priority of the third priority condition. For example, the first priority condition is the troubleshooting duration, the second priority condition is the troubleshooting cost, and the third priority condition is the troubleshooting path complexity. Therefore, the priority of the troubleshooting duration is greater than the priority of the troubleshooting cost, and the priority of the troubleshooting cost is greater than the priority of the troubleshooting path complexity. The fault is troubleshooted from top to bottom.

[0065] ​Firstly, the top-level events are sorted in ascending order according to the first priority condition, the top-level events with the same first priority condition are sorted in ascending order according to the second priority condition, and the top-level events with the same second priority condition are sorted in ascending order according to the third priority condition, and finally an ascending sequence of the top-level events is obtained. The first top-level event in the ascending sequence of the top-level events, that is, the top-level event with the minimum troubleshooting duration, the minimum troubleshooting cost or the minimum troubleshooting path complexity, is selected for troubleshooting. Then, the intermediate events and the basic events are sorted according to the first priority condition, the second priority condition and the third priority condition in the same way as the sorting of the top-level events, and an ascending sequence of the intermediate events and an ascending sequence of the basic events are obtained. The first intermediate event in the ascending sequence of the intermediate events and the first basic event in the ascending sequence of the basic events are still selected. The selected top-level event, intermediate event and basic event can form a troubleshooting path. In the process of troubleshooting according to the troubleshooting path, it is determined whether the current troubleshooting path ends and whether it is necessary to backtrack to the previous branch node under the condition that it is determined that the selected event is not faulty. If the current troubleshooting path does not end, the troubleshooting continues. If the current troubleshooting path ends, the event at the bottom of the troubleshooting path is deleted from the ascending sequence of the events, and the troubleshooting path is determined again, and each event is sequentially subjected to troubleshooting until all faults are located.

[0066] The embodiments of the present application can determine the troubleshooting path of the fault diagnosis model according to the priority of the troubleshooting duration, the troubleshooting cost and the troubleshooting path complexity. Different priority sizes can be set for different production line devices, which is more in line with the requirements of the troubleshooting path of the production line device and can improve the fault positioning efficiency.

[0067] S150, determining the position of the fault of the production line device according to the troubleshooting path.

[0068] In some embodiments, the troubleshooting path includes a plurality of troubleshooting paths and an order of the plurality of troubleshooting paths. The position of the fault of the production line device is determined according to the troubleshooting path, including: sequentially determining the position of the fault of the production line device according to the troubleshooting path according to the order of the plurality of troubleshooting paths.

[0069] The position of the fault of the production line device is the position of an element of the production line device or an element action unit performing an action.

[0070] It can be understood that after the position of the fault of the production line device is determined each time, the device information of the production line device, the selected fault tree, the fault probability of each branch of the fault diagnosis model and the attributes of the events will change accordingly.

[0071] The embodiment of the application determines the fault level of the production line equipment failure through the fault tree, and the fault mode of each fault level, determines the fault troubleshooting path from the highest fault level to the lowest fault level in the fault diagnosis model according to the required troubleshooting time, troubleshooting cost and troubleshooting path complexity of each fault mode, and determines the location of the production line equipment failure according to the fault troubleshooting path, which can quickly and accurately determine the specific location of the production line equipment failure.

[0072] In some embodiments, the target fault tree is one of a plurality of fault trees; as Figure 6 As shown in S120 above, the method can further include S111-S112.

[0073] S111, obtaining a target selection manner of a target fault tree.

[0074] In some embodiments, the user can input the target selection manner of the target fault tree according to experience. Before obtaining the target selection manner of the target fault tree, the target selection manner of the target fault tree input by the user is received.

[0075] In some embodiments, the target selection manner is a target selection manner of the target fault tree randomly selected from a plurality of selection manners. Before obtaining the target selection manner of the target fault tree, a plurality of selection manners input by the user are received, and the target selection manner is randomly selected from the plurality of selection manners as the target selection manner of the target fault tree.

[0076] In some embodiments, S111 can specifically include: Obtaining the fault positioning time of the fault tree positioning fault each time in a preset time period; Obtaining the target time of the selected fault tree, the historical average fault positioning time of the target time, and the production line change rate, the target time being any time in a preset time period; Determining a random probability parameter according to the target time, the historical average fault positioning time of the target time, and the production line change rate; Determining the target selection manner of the target fault tree according to the random probability parameter.

[0077] In this embodiment, the target selection manner of the target fault tree determined at different times can be different through the setting of the random probability parameter. Since the module production line equipment is diverse in types and diversified in production processes, the best application model needs to be selected according to the actual application environment of the factory at different times, so the target fault tree can be selected through the method of reinforcement learning, and then the fault diagnosis model is determined. The specific selection process is as follows: First, set a learning period T, the fault tree of FMA mode and the fault tree of SMM mode are called F1 and F2 respectively. Each time a fault occurs, a selection of the fault tree is needed, that is, an action, and a real-time benefit, that is, the fault positioning time, is obtained. The goal of learning is to select the target fault tree with high fault positioning efficiency.

[0078] Assume that the target fault tree selected in a fault troubleshooting process is represented as , and the fault positioning time is called R. The value of a action at time t can be represented as , as shown in the following formula (9): ; (9) where i represents the time, R i represents the fault positioning time of positioning the fault using the target fault tree selected at the i time.

[0079] Set the obstacle removal action, and the overall strategy of the obstacle removal action is to select the fault tree with the maximum action value at the current time. In order to expand the exploration range in the early stage, a random probability parameter p is added, p ∈ (0, 0.5), that is, the probability of (1-p) selects the fault tree with the maximum action value at the current time, and the probability of p randomly selects the fault tree. The specific strategy function can be represented as . The calculation process of the random probability parameter p is shown in formula (10): ; (if , , else ) (10) , which is calculated as shown in formula (11): (11) , which is calculated as shown in formula (12): (12) where represents the current time, ∈ (0, T), in the learning period T, the exploration behavior should be increased in the early stage, and the exploration behavior should be reduced after the model tends to be stable in the later stage; represents the historical average fault positioning time, when the historical average fault positioning time is large, the exploration behavior should be increased, that is, the randomness of the target fault tree is increased; Indicates the production line change rate, the production line change includes personnel change, equipment change, product model change of production and production line number change, etc., whenever one of them changes, Just add 1, ={0,1,2,3,4}, when the production line change rate is large, the exploration behavior should be increased.

[0080] After T expires, or the number of obstacle removal actions reaches the maximum iteration number N, the target selection method is to select the fault tree with the maximum action value at the current time as the target fault tree. Wherein, N is a value set in advance according to actual situation.

[0081] Through the setting of random probability parameters, the embodiments of the application can select different target selection methods among multiple target selection methods of the target fault tree, dynamically select the target fault tree, break the selection logic of fixedly selecting the fault tree to generate the fault diagnosis model, let the low-frequency and easily ignored fault tree have a chance to be selected, improve the fault troubleshooting coverage, and avoid missing low-probability faults.

[0082] S112, selecting a target fault tree from the multiple fault trees by using the target selection method.

[0083] In some embodiments, the target selection method includes selecting a target fault tree that makes the fault positioning time length shortest; and the S112 can specifically include: Obtaining the fault positioning time length of each selected fault tree for positioning faults in a preset time period; According to the fault positioning time length, selecting a target fault tree that makes the fault positioning time length shortest from the multiple fault trees.

[0084] It should be noted that the target fault tree with the shortest fault positioning time length is the fault tree with the maximum value.

[0085] The embodiments of the application can improve the fault positioning efficiency by selecting a target fault tree that makes the fault positioning time length shortest.

[0086] In some embodiments, the target selection method includes randomly selecting a target fault tree; and the S112 can specifically include: Randomly selecting a target fault tree from the multiple fault trees.

[0087] The embodiments of the application can increase the exploration range of faults, let the low-frequency and easily ignored fault tree have a chance to be selected, improve the fault troubleshooting coverage, and avoid missing low-probability faults by randomly selecting a target fault tree from the multiple fault trees.

[0088] The embodiments of the present application can automatically select the target fault tree through the setting of the selection mode of the target fault tree. Through the configuration of different selection modes of the target fault tree, the target fault tree is dynamically selected, which can adapt to the fault diagnosis needs of diversified production line equipment.

[0089] In actual production process, the production line fault is often associated with historical fault state, and mainly related to the previous fault state, which conforms to the Markov chain property, that is: . Wherein, represents the probability of occurrence of a future fault state under the condition that the previous fault state is known. Therefore, the embodiments of the present application provide the following fuzzy reasoning mechanism for the emergency troubleshooting application scenario, which can give the fault mode set with the maximum probability according to the historical information, for emergency troubleshooting.

[0090] In some embodiments, as shown in Figure 7 , the method further includes S210-S240.

[0091] S210, acquiring historical fault modes.

[0092] In some embodiments, the historical fault data is acquired from the database corresponding to the factory, and the historical fault data includes historical fault modes.

[0093] In some embodiments, the historical fault mode includes a top-level event of fault occurrence and a basic event corresponding to the top-level event of fault occurrence. For example, in the case of failure of a certain function, the historical fault mode includes the fault function and the specific fault action corresponding to the function. The historical fault mode can be represented as:

[0094] Wherein, represents the top-level event of fault occurrence, represents the corresponding basic event of fault occurrence.

[0095] Based on this, the historical fault mode can also be represented as , that is, the historical fault mode is the basic event of a total of m types of faults.

[0096] S220, according to the historical fault mode, determining the second probability from each type of target fault mode to the fault mode other than the target fault mode, obtaining a fault transition matrix, and the target fault mode is any one of the historical fault modes.

[0097] In some embodiments, according to the historical fault mode, the second probability from each type of target fault mode to the fault mode other than the target fault mode is determined, and a fault transition matrix is obtained, including: The number of times of external transition of the i th fault event is determined by using the following formula (13): (13) wherein, represents the number of times of external transfer of the i-th type of failure event, represents the number of times of transfer of the i-th type of failure to the j-th type of failure. The historical failure mode is a basic event in which m types of failures occur in total.

[0098] The second probability from the i-th type of failure event to the j-th type of failure event is determined by using the following formula (14): (14) According to the above formulas (13) and (14), the failure transfer matrix can be obtained, and the failure transfer matrix is as follows:

[0099] wherein, the failure transfer matrix is a Markov state transition matrix.

[0100] S230, determining a third probability of occurrence of each type of failure mode in the highest failure level under the condition that each type of failure mode in the lowest failure level occurs.

[0101] In some embodiments, the basic event is associated with the top-level event, and a certain type of basic event failure can also involve multiple failure modes. According to the historical failure mode, the Monte Carlo algorithm is used to randomly test all basic events, and the third probability of occurrence of each top-level event failure under the condition that each basic event fails is obtained, that is, , k, q ∈ N+.

[0102] S240, determining the sequence of the probability of occurrence of each type of failure mode in the historical failure mode and the failure mode sequence corresponding to the sequence according to the failure transfer matrix and the third probability.

[0103] In some embodiments, according to the third probability , the failure transfer matrix P and the historical failure mode L, when a new top-level event occurs, the target is solved by using the Bayes formula, that is, the most likely failure basic event under the known combination of failure top-level events , as shown in the following formula (15): (15) wherein, according to the historical failure mode, can be uniquely determined . Therefore, the solution target R is changed to the following formula (16): (16) According to the Markov correlation property and the probability theory related formulas (17) and (18), the solution target R is solved: (17) (18) wherein, i.e. the steady-state probability of the Markov chain, can be allowed , each element in the sum is 1), the steady-state probability can be obtained by solving the eigenvalue; can be obtained from the fault transition matrix P, substituted into the above formula (16), and the , is the most likely fault position. In addition, while solving the with the largest R value, all obtained can be arranged in descending order according to the R value to obtain a fault mode sequence. The fault mode sequence is used to guide the user to troubleshoot the fault position, and the fault is sequentially checked according to the fault mode sequence until the fault position is confirmed.

[0104] Embodiments of the present application can give a sequence of the probability of occurrence of each type of fault mode and a fault mode sequence corresponding to the sequence according to the historical fault mode, so that the user can perform emergency troubleshooting according to the sequence of the probability of occurrence of each type of fault mode and the fault mode sequence corresponding to the sequence, and the efficiency of determining the fault position can be further improved.

[0105] Reliability is an important indicator for determining whether the product function can be normally implemented, and the optimization of the reliability of the production line equipment can reduce the failure rate and improve the reliability of the production line. Embodiments of the present application can start from the fault diagnosis model of equipment decomposition, construct a production line equipment reliability optimization model, and use a corresponding optimization algorithm for optimization, and the specific process is as follows.

[0106] In some embodiments, as shown in Figure 8 , the method further includes S310-S380.

[0107] S310, determining a minimum cut set set causing the fault mode in the highest fault level to occur, each minimum cut set including at least one type of fault mode in the lowest fault level.

[0108] In some embodiments, the determinant method or Boolean algebra method is used to logically simplify the fault diagnosis model to obtain a minimum cut set set causing the top event to occur, and the minimum cut set is a logical combination of basic events, i.e. each minimum cut set includes at least one type of fault mode in the lowest fault level.

[0109] S320, determining a first fault probability function of each type of fault mode in the highest fault level according to the minimum cut set set.

[0110] In some embodiments, according to the minimum cut set set, each fault mode in the highest fault level, i.e., the first fault probability function of a single fault tree, is determined, and the first fault probability function is shown in the following formula (19): (19) wherein, represents the overall fault probability of the hth fault tree, represents the fault occurrence probability of the ith minimum cut set in the hth fault tree, and k is the number of minimum cut sets corresponding to the single fault tree. The minimum cut set itself is a combination of logical AND gates of fault tree basic events, and the fault occurrence probability of the minimum cut set can be determined by using the above formula (3), which will not be repeated here.

[0111] S330, according to the first fault probability function, the second fault probability function of all fault modes in the highest fault level is determined.

[0112] In some embodiments, the production line equipment is a series connection of multiple fault trees, and each fault tree can be regarded as a reliability component unit of the equipment, and the occurrence of any fault mode will cause equipment failure. Define as the equipment fault probability function, the equipment fault probability function, i.e., the equipment reliability optimization model, represents the second fault probability function of all fault modes in the highest fault level, and the second fault probability function is shown in the following formula (20): (20) wherein, n is the number of fault trees. One fault mode in the highest fault level corresponds to one fault tree.

[0113] S340, the troubleshooting duration, troubleshooting cost, optimization cost, historical occurrence frequency and reliability requirement information of the production line equipment of each minimum cut set are obtained.

[0114] In some embodiments, in the actual optimization process, there are often many conditions such as cost, time, etc. Considering subjective factors and objective factors, the subjective factors include optimization cost, troubleshooting duration, troubleshooting cost, occurrence frequency, etc., and the objective factor is the reliability requirement of the equipment. The goal of optimization is to select the cut set item with low optimization cost, high occurrence frequency, long troubleshooting time and high troubleshooting cost after fault occurrence as much as possible under the condition of meeting the reliability requirement of the equipment, and the cut set item selected in this way should be optimized.

[0115] In some embodiments, the product life of the production line equipment usually obeys an exponential distribution, and the reliability requirement of the equipment is related to the failure rate and the product life, as shown in the following formula (21): (21) wherein, t represents the product life requirement, The device reliability requirement is represented.

[0116] In some embodiments, the troubleshooting duration, troubleshooting cost, optimization cost, and historical occurrence frequency of each minimal cut set are obtained as follows: Define the cut set set as G, and count the troubleshooting duration, troubleshooting cost, optimization cost, and historical occurrence frequency of each cut set:

[0117] wherein, , , , respectively represent the troubleshooting duration, troubleshooting cost, optimization cost, and historical occurrence frequency of the mth cut set (m is the total number of cut sets, m∈N+). The troubleshooting duration, troubleshooting cost, optimization cost, and historical occurrence frequency of a single cut set are the sum of the troubleshooting duration, troubleshooting cost, optimization cost, and historical occurrence frequency of the basic events in the constituent items thereof.

[0118] S350, according to the troubleshooting duration, troubleshooting cost, optimization cost, and historical occurrence frequency of each minimal cut set, determine the score of each minimal cut set using a preset scoring rule.

[0119] In some embodiments, the priority of each cut set is evaluated. The idea of ideal solution can be adopted, and each cut set is regarded as a point in space, and the optimal point and the worst point are calculated. By comparing the distance between each cut set point and the optimal point and the worst point, the score of the cut set is determined, which represents the priority.

[0120] In one example, the troubleshooting duration, troubleshooting cost, optimization cost, and historical occurrence frequency are normalized. The smaller the value, the higher the priority. The normalized processing formula (22) is as follows: (22) wherein, represents the maximum value in the 4th column of the matrix G.

[0121] Since the dimensions of the indicators in each column of the matrix G are different, it needs to be dimensionless. The dimensionless processing formula (23) is as follows: (23) wherein, i∈[1,4], m∈N+.

[0122] Then, the weight of each index of the matrix G is calculated using the entropy weight method, as shown in the following equations (24)-(28). First, column normalization is performed to calculate the probability of each event; then, the event information entropy and the information entropy of each index are calculated, and finally, the information utility is calculated and normalized, which is the weight corresponding to each index.

[0123] (24) (25) (26) (27) (28) wherein, represents the approximate probability of occurrence of each event, represents the information entropy of each event, represents the information entropy of each index, represents the information utility, represents the weight of each index.

[0124] Next, the distance of each cut set from the optimal point (defined as the maximum value of each index) and the worst point (defined as the minimum value of each index) is calculated, and a score is given, and the priority of the cut set is determined according to the score size, which is shown in the following equations (29)-(31).

[0125] (29) (30) (31) wherein, j∈[1, m], represents the Euclidean distance of the cut set from the optimal point, represents the Euclidean distance of the cut set from the worst point, represents the score.

[0126] S360, according to the second failure probability function, the score of each minimum cut set, and the relationship information of reliability requirement information and product life, a target function is constructed.

[0127] In some embodiments, according to the second failure probability function, the score of each minimum cut set, and the relationship information of reliability requirement information and product life, a target function is constructed. The target function to be solved is shown in equation (32): min (32) wherein, the constraint conditions of the target function are as follows: s.t.

[0128]

[0129]

[0130] wherein, represents the cut set after scoring. The higher the score, the more the cut set needs to be optimized.

[0131] S370, solving the target function to obtain a target minimum cut set, the target minimum cut set comprising at least one fault mode.

[0132] wherein, the target minimum cut set represents a minimum cut set with a larger failure rate.

[0133] In some embodiments, reliability optimization is to optimize by selecting a suitable cut set, reducing its failure probability, and improving the reliability of the device. The particle swarm optimization algorithm is used to solve the target function in the embodiments of the present application. In order to improve the convergence speed of the algorithm, the cut set priority evaluation is limited not to participate in the algorithm optimization process, but as the basis for selecting the best result. At the same time, in order to solve the problems of easy premature convergence to local optimum and slow convergence speed in the later iteration of the algorithm, the learning factor will change with the number of iterations, and there is a certain probability to accept a smaller value when updating the fitness.

[0134] In some embodiments, the above S370 can specifically include: using a particle swarm optimization algorithm to solve the target function to obtain a target minimum cut set; wherein, the solving process of the particle swarm optimization algorithm is as follows: obtaining the value range of the learning factor and the number of iterations; According to the value range of the learning factor and the number of iterations, the selection function of the learning factor is fitted, so as to select the learning factor of each iteration process according to the selection function; According to the learning factor, the fitness of the position reached in each iteration process is determined, and the descending sequence of the fitness is determined, and the position represents the minimum cut set; obtaining the target fitness that satisfies the preset retention condition, the preset retention condition being to retain the first preset number of target fitnesses in the descending sequence, and according to the fourth probability, randomly selecting the target fitness from the fitnesses in the second preset number of the descending sequence; determining the target position corresponding to the target fitness, the target position representing the target minimum cut set.

[0135] Specifically, in this embodiment, assuming that in an m-dimensional search space, the population is composed of N particles, then:

[0136] wherein, , and Xi(t) and Vi(t) represent the position and velocity of the i-th particle, respectively, and Pmin(t) represents the minimum value of all particles in the population, i∈[1, N], at the t-th iteration, Xi(t) and Vi(t) represent the position and velocity of the i-th particle, respectively, and Pmin(t) represents the minimum value of all particles in the population, i∈[1, N], at the t-th iteration, (33) wherein, , , C1(t) represents a learning factor, which controls the search speed of the particle in each direction; , rand represents a random number distributed in the interval [0, 1]; the particle velocity and position are limited between (0, 1); j∈[1, m].

[0137] In order to improve the problems of easy premature convergence to local optimum and slow convergence speed in the later iteration in the traditional algorithm, the learning factor will change with the number of iterations. In the early iteration, , C1(t) is larger, and the exploration direction is biased towards the individual particle, and C1(t) is smaller in the later iteration; and C1(t) is smaller in the early iteration, and gradually increases in the later iteration, and the exploration direction is biased towards the group optimum. The setting method of the learning factor is as follows: (1) setting the range of C1(t) and the number of iterations , , :

[0138] (2) according to the parameter range and the number of iterations, the function fitting is performed, linear fitting is adopted, , exponential fitting is adopted, and the fitting result is as shown in the following formula (34) (wherein x represents the number of iterations): (34) (3) in the fitness updating process, the updating condition is that: the fitness value of the new position is calculated each time, if the fitness value of the new position is higher, the fitness value of the particle is updated (the first three fitness values and the corresponding position information are retained), and in order to reduce the influence of the algorithm possibly converging to a local optimum, a certain probability p is set to retain the smaller fitness value, so as to facilitate the exploration of the surrounding. The specific implementation process of the particle swarm optimization algorithm is as shown in S1 to S5 of Figure 9 .

[0139] ​In order to improve the convergence speed of the algorithm, the cut set priority ranking restriction can not be used as a limiting condition in the actual optimization process, but as a preferred basis after optimization. The cut set priority ranking restriction condition is (m-1) in total, according to the top three fitness values from large to small, the number of the limit condition is checked in turn, and the maximum value corresponding to the position information is taken as the optimized solution.

[0140] By sorting the position information of the optimal particle after optimization, the minimum cut set with a larger failure rate and the corresponding basic event can be found, guiding the engineers to make corresponding improvements, and the overall reliability of the production line equipment can be improved, and the system failure rate can be reduced.

[0141] The embodiments of the present application improve the particle swarm optimization algorithm to solve the target function, solve the problem that the algorithm is prone to premature convergence to local optimum, slow convergence speed in the later iteration, and the problem that the target minimum cut set, that is, the minimum cut set with a larger failure rate, is not accurately determined.

[0142] S380, display the target minimum cut set for guiding the user to optimize the production line equipment.

[0143] In some embodiments, the target minimum cut set is displayed by using a target interface for guiding the user to optimize the production line equipment.

[0144] The embodiments of the present application guide the user to optimize the production line equipment through the target minimum cut set, and the reliability optimization of the production line equipment can reduce the failure rate and improve the reliability of the production line.

[0145] In some embodiments, the method further comprises: constructing a production line model corresponding to the production line equipment, the production line model comprising personnel information, the personnel information comprising personnel position information; in a case where the personnel position information represents that the personnel is located outside the preset area, triggering an abnormal alarm of the position corresponding to the personnel position information.

[0146] In some embodiments, the personnel is divided into different types of work, and the main attributes include position and behavior state, etc.; the personnel information digital model can be extended as , , wherein, pw and ps represent personnel position information and state respectively, W1 represents a normal position point set, s1 and s2 represent normal working and maintenance states respectively, and when the position point exceeds the range of W1, an abnormal alarm is triggered.

[0147] The embodiments of the present application can alarm the abnormal situation of the personnel by setting the personnel information in the production line model, and improve the safety of the production line equipment.

[0148] In some embodiments, the production line model further comprises device information; and the method further comprises: According to the device information, a production line device image corresponding to the production line model is displayed, and a position where a fault occurs in the production line device is marked in the production line device image.

[0149] In some embodiments, the device includes a mechanical arm, a clamp, a PC, a comprehensive tester, a calibration instrument, etc., and the main attributes include size, position, and behavior state, etc. The device information digital model can be extended as , , . Wherein, dl, dw, and ds respectively represent device size, device position, and device behavior state, and l, w, and h represent the length, width, and height of the device, and s11, s22, and s33 respectively represent normal working, abnormal (may exist fault), and maintenance state.

[0150] Embodiments of the present application mark the position where a fault occurs in the production line device in the production line device image corresponding to the production line model, so as to facilitate the user to learn the fault position as soon as possible, and improve the user experience.

[0151] In some embodiments, the production line model can be represented as The production line model is a digital model, and the digital model is a virtual mapping of the production process entity, which realizes the mapping from the physical world to the virtual space. Through the real-time MES system data, the visual display of the real-time production line three-dimensional scene is constructed on the PC end through the support of Web Graphics Library (WebGL), HyperText Markup Language (HTML), and other technologies, and the corresponding personnel are timely notified to check and maintain, which can indirectly improve the fault handling efficiency. Using the digital model to map the production line in real time can find the fault point in advance. At the same time, the bottom diagnosis logic uses the fault tree, which can be guided in real time through the PC to assist the engineering personnel to perform rapid fault diagnosis, and the operation is simple and has no technical threshold requirement.

[0152] The embodiments provided in the application are directed to a module and an Internet of Things terminal product production line. According to real-time production data, fault diagnosis is performed through a pre-constructed fault diagnosis model. Then, based on a digital model of each physical entity of the production line, real-time diagnosis data of the physical entity are fused, the behavior of the real physical entity is simulated by the digital model, the production line production profile is mapped, and is displayed through a three-dimensional visualization tool. Finally, fault work orders can be issued to corresponding personnel for processing according to the fault points. Meanwhile, the embodiments provided in the application can periodically optimize the reliability of the factory production line. The validity period of the reliability optimization model is checked periodically. After the validity period expires, the reliability optimization model is updated according to historical data, and reliability improvement suggestions are issued to corresponding engineering personnel according to the reliability optimization results, so as to provide the engineering personnel with an improvement direction, thereby improving the reliability level of the entire production line, reducing the failure probability, and forming a virtuous cycle.

[0153] The embodiments provided in the application are mainly based on factory production and historical fault data, do not need to additionally increase equipment, are simple and efficient, and are suitable for the fault diagnosis and optimization field of the module and the terminal product production line. The embodiments have certain efficiency improvement, quality improvement and cost saving effects.

[0154] Based on the method for determining the fault position of the production line equipment provided in the above embodiments, the application also provides a specific implementation mode of a device for determining the fault position of the production line equipment. Please refer to the following embodiments.

[0155] Firstly, referring to Figure 10 The device 400 for determining the fault position of the production line equipment provided in the embodiments of the application comprises: The acquisition module 410 is configured to acquire equipment information of the production line equipment in the production line. The determination module 420 is configured to determine, according to the equipment information, a fault diagnosis model of the production line equipment by using a target fault tree. The fault diagnosis model comprises at least two fault levels. Each fault level comprises at least one type of fault mode. There is a dependency relationship between adjacent fault levels. The determination module 420 is further configured to determine the required troubleshooting time, troubleshooting cost and troubleshooting path complexity of each type of fault mode. The determination module 420 is further configured to determine, according to the troubleshooting time, troubleshooting cost and troubleshooting path complexity, a troubleshooting path of the fault diagnosis model. The troubleshooting path is a troubleshooting path from the highest fault level to the lowest fault level. The determination module 420 is further configured to determine, according to the troubleshooting path, the position of the production line equipment where the fault occurs.

[0156] In some embodiments, the target fault tree is one of a plurality of fault trees. The device 400 can further comprise: The acquisition module 410 is configured to acquire a target selection manner of the target fault tree before determining the fault diagnosis model of the production line equipment by using the target fault tree according to the equipment information. The selection module is configured to select the target fault tree from the plurality of fault trees by using the target selection manner.

[0157] In some embodiments, the acquisition module 410 can be specifically configured to: acquire a fault locating duration of each selected fault tree for locating a fault in the preset time period; acquire the target moment of the selected fault tree, the historical average fault locating duration of the target moment, and the production line change rate, the target moment being any moment in the preset time period; determine a random probability parameter according to the target moment, the historical average fault locating duration of the target moment, and the production line change rate; determine the target selection manner of the target fault tree according to the random probability parameter.

[0158] In some embodiments, the target selection manner includes selecting the target fault tree with the shortest fault locating duration; and the selection module can be specifically configured to: acquire a fault locating duration of each selected fault tree for locating a fault in the preset time period; select the target fault tree with the shortest fault locating duration from the plurality of fault trees according to the fault locating duration.

[0159] In some embodiments, the target selection manner includes randomly selecting the target fault tree; and the selection module can be specifically configured to: randomly select the target fault tree from the plurality of fault trees.

[0160] In some embodiments, the determination module 420 can be specifically configured to: acquire an investigation duration and an investigation cost required for each type of fault mode in the lowest fault level; acquire a first probability of occurrence of each type of fault mode in all fault levels, the first probability being obtained by a frequency of historical occurrence of the fault mode; determine an investigation duration and an investigation cost required for each type of fault mode other than the fault mode in the lowest fault level according to the first probability and the investigation duration and the investigation cost of each type of fault mode in the lowest fault level.

[0161] In some embodiments, the determination module 420 can be specifically configured to determine the investigation path complexity of each type of fault mode according to a fault mode quantity associated with each type of fault mode.

[0162] In some embodiments, the determination module 420 can be specifically configured to: acquire a priority of the investigation duration, the investigation cost, and the investigation path complexity. According to the priorities of the troubleshooting duration, the troubleshooting cost and the troubleshooting path complexity, the troubleshooting path of the fault diagnosis model is determined for each type of fault mode.

[0163] In some embodiments, the apparatus 400 can further include: The acquisition module 410 is configured to acquire the historical fault modes. The determination module 420 is further configured to determine, according to the historical fault modes, a second probability of each type of fault mode in the target fault modes to a fault mode other than the target fault mode, to obtain a fault transition matrix, the target fault mode being any one of the historical fault modes. The determination module 420 is further configured to determine, in the case that each type of fault mode in the lowest fault level occurs, a third probability of each type of fault mode in the highest fault level occurring. The determination module 420 is further configured to determine, according to the fault transition matrix and the third probability, a sequence of the occurrence probability of each type of fault mode in the historical fault modes and a fault mode sequence corresponding to the sequence.

[0164] In some embodiments, the apparatus 400 can further include: The determination module 420 is further configured to determine a set of minimal cut sets that cause the fault modes in the highest fault level to occur, each minimal cut set including at least one type of fault mode in the lowest fault level. The determination module 420 is further configured to determine, according to the set of minimal cut sets, a first fault probability function of each type of fault mode in the highest fault level. The determination module 420 is further configured to determine, according to the first fault probability function, a second fault probability function of all fault modes in the highest fault level. The acquisition module 410 is configured to acquire the troubleshooting duration, the troubleshooting cost, the optimization cost, the historical occurrence frequency and the reliability requirement information of the production line equipment of each minimal cut set. The determination module 420 is further configured to determine, according to the troubleshooting duration, the troubleshooting cost, the optimization cost and the historical occurrence frequency of each minimal cut set, a score of each minimal cut set by using a preset scoring rule. The construction module is configured to construct a target function according to the second fault probability function, the score of each minimal cut set, the reliability requirement information and the relationship information between the product life. The solving module is configured to solve the target function to obtain a target minimal cut set, the target minimal cut set including at least one type of fault mode. The display module is configured to display the target minimal cut set to guide the user to optimize the production line equipment.

[0165] In some embodiments, the solving module can be specifically configured to: The particle swarm optimization algorithm is used to solve the target function, and a target minimum cut set is obtained. The solving process of the particle swarm optimization algorithm is as follows: An iteration number and a value range of the learning factor are obtained. According to the iteration number and the value range of the learning factor, a selection function of the learning factor is fitted, so as to select the learning factor in each iteration process according to the selection function. According to the learning factor, the fitness of the position reached in each iteration process is determined, and a descending sequence of the fitness is determined, and the position represents the minimum cut set. A target fitness satisfying a preset reservation condition is obtained, the preset reservation condition is to reserve the first preset number of target fitnesses in the descending sequence, and according to the fourth probability, the target fitness is randomly selected from the fitnesses in the second preset number of the descending sequence. The target position corresponding to the target fitness is determined, and the target position represents the target minimum cut set.

[0166] In some embodiments, the apparatus 400 can further include: The construction module is configured to construct a production line model corresponding to the production line equipment, and the production line model includes personnel information, and the personnel information includes personnel position information. The triggering module is configured to trigger an abnormal alarm of the position corresponding to the personnel position information in a case where the personnel position information represents that the personnel is located outside the preset area.

[0167] In some embodiments, the production line model further includes equipment information; and the apparatus 400 can further include: The display module is configured to display a production line equipment image corresponding to the production line model according to the equipment information, and the production line equipment image marks a position of the production line equipment where a fault occurs.

[0168] The various modules of the production line equipment fault position determination apparatus provided in the embodiments of the present application can realize the functions of the various steps of the production line equipment fault position determination method provided above, and can achieve the corresponding technical effects. For brevity, the description is not repeated here.

[0169] Based on the same inventive concept, the embodiments of the present application also provide an electronic device.

[0170] Figure 11 A hardware structure schematic diagram of the electronic device provided by the embodiments of the present application is shown.

[0171] The electronic device can include a processor 501 and a memory 502 having computer program instructions stored therein.

[0172] In particular, the processor 501 can include a central processing unit (CPU), or an application-specific integrated circuit (ASIC), or one or more integrated circuits configured to perform the operations of the embodiments of the application.

[0173] The memory 502 can include mass storage for data or instructions. As an example and not by way of limitation, the memory 502 can include a hard disk drive, a floppy disk drive, flash memory, a Compact Disc Read Only Memory (CD ROM), a Digital Versatile Disc (DVD), a magnetic tape, or a Universal Serial Bus (USB) drive or a combination of two or more of these. The memory 502 can include removable or non-removable (or fixed) media, where appropriate. The memory 502 can be considered a computer-readable medium, where appropriate. The memory 502 can be internal or external to the integrated gateway disaster recovery device, where appropriate. In particular embodiments, the memory 502 is non-volatile, solid-state memory.

[0174] The memory can include read-only memory (ROM), random access memory (RAM), magnetic disk storage mediums, optical storage mediums, flash memory devices, electrical, optical, or other physical / tangible memory storage devices. Thus, in general, the memory includes one or more tangible (non-transitory) computer-readable storage media (e.g., a memory device) encoded with software that, when executed (by one or more processors), is operable to perform operations described with reference to the methods according to the aspects of the present disclosure.

[0175] The processor 501 implements the production line equipment fault location determination method of any one of the above embodiments by reading and executing computer program instructions stored in the memory 502.

[0176] In one example, the electronic device can further include a communication interface 503 and a bus 510. As shown, the processor 501, the memory 502, and the communication interface 503 are connected through the bus 510 and complete communication with each other. Figure 11

[0177] The communication interface 503 is mainly used to realize the communication between the modules, devices, units and / or equipment in the embodiments of the application.

[0178] ​Bus 510 includes a hardware, software, or both that couples components of electronic device to each other. As an example and not by way of limitation, bus can include an Accelerated Graphics Port (AGP) or other graphics bus, an Extended Industry Standard Architecture (EISA) bus, a Front Side Bus (FSB), a Hyper Transport (HT) interconnect, an Industry Standard Architecture (ISA) bus, an InfiniBand interconnect, a Low Pin Count (LPC) bus, a memory bus, a Micro Channel Architecture (MCA) bus, a Peripheral Component Interconnect (PCI) bus, a PCI-Express (PCI-X) bus, a Serial Advanced Technology Attachment (SATA) bus, a Video Electronics Standards Association local (VESA) bus, or another suitable bus or a combination of two or more of these. Where appropriate, bus 510 can include one or more buses. Although this application describes and shows a particular bus, this application contemplates any suitable bus or interconnect. The electronic device can perform the method for determining the fault location of the production line equipment in the embodiments of the application, thereby realizing the method for determining the fault location of the production line equipment.

[0179] In addition, in combination with the method for determining the fault location of the production line equipment in the above embodiments, the embodiments of the application can provide a computer storage medium to realize. The computer storage medium has computer program instructions stored thereon; the computer program instructions are executed by a processor to realize any one of the methods for determining the fault location of the production line equipment in the above embodiments.

[0180] The application also provides a computer program product, instructions in the computer program product are executed by a processor of an electronic device, so that the electronic device executes each process of realizing any one of the embodiments of the method for determining the fault location of the production line equipment.

[0181] It is to be understood that the application is not limited to particular configurations and processes described herein and shown in the drawings. For simplicity, detailed descriptions of known methods and apparatuses are omitted so as not to obscure the disclosure. In the above-described embodiments, several specific steps are described as examples. However, the method processes of the present application are not limited to the specific steps described and illustrated herein, but can include any number of additional steps or changes to the described steps, or can be performed in any order, or in parallel, depending on the circumstances.

[0182] The functional blocks shown in the structural block diagrams described above can be implemented as hardware, software, firmware, or a combination thereof. When implemented in hardware, they can be, for example, electronic circuits, application specific integrated circuits (ASICs), appropriate firmware, plug-ins, function cards, and the like. When implemented in software, the elements of the present application are program or code segments used to perform the required tasks. The program or code segments can be stored in a machine-readable medium or transmitted through a data signal carried in a carrier wave over a transmission medium or communication link. The "machine-readable medium" can include any medium capable of storing or transmitting information. Examples of the machine-readable medium include electronic circuits, semiconductor memory devices, read-only memory (ROM), flash memory, erasable read-only memory (EROM), floppy disks, compact discs (CD-ROM), optical disks, hard disks, optical fiber media, radio frequency (RF) links, and the like. The code segments can be downloaded via a computer network such as the Internet, an intranet, and the like.

[0183] It is also to be understood that the exemplary embodiments mentioned in the present application describe some methods or systems based on a series of steps or devices. However, the present application is not limited to the order of the above steps, that is, the steps can be performed in the order mentioned in the embodiments, or in an order different from the embodiments, or several steps can be performed simultaneously.

[0184] The computer program instructions can also be loaded onto a computer, other programmable data processing apparatus, or other processing device to cause a series of operational steps to be performed on the computer, other programmable apparatus or other processing device to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide processes for implementing the functions / acts specified in the flowchart and / or block diagram block or blocks.

[0185] The above is merely specific implementation of the present application, and those skilled in the art can clearly understand the specific working process of the system, module and unit described above for the convenience and brevity of description, which can refer to the corresponding process in the foregoing method embodiments, and will not be described herein. It should be understood that the protection scope of the present application is not limited to this, and any person skilled in the art can easily think of various equivalent modifications or replacements within the technical range disclosed in the present application, and these modifications or replacements should be covered within the protection scope of the present application.

Claims

1. A method for determining the location of a fault in production line equipment, characterized in that, include: Obtain equipment information for production line equipment; Based on the equipment information, a fault diagnosis model for the production line equipment is determined using a target fault tree. The fault diagnosis model includes at least two fault levels, each fault level includes at least one type of fault mode, and there is a dependency relationship between adjacent fault levels. Determine the required troubleshooting time, troubleshooting cost, and troubleshooting path complexity for each type of failure mode; Based on the investigation time, the investigation cost, and the investigation path complexity, the fault investigation path of the fault diagnosis model is determined, and the fault investigation path is a fault investigation path from the highest fault level to the lowest fault level. According to the troubleshooting path, determine the location where the production line equipment malfunctioned.

2. The method according to claim 1, characterized in that, The target fault tree is one of multiple fault trees; before determining the fault diagnosis model of the production line equipment using the target fault tree based on the equipment information, the method further includes: Obtain the target selection method for the target fault tree; Using the target selection method, a target fault tree is selected from the plurality of fault trees.

3. The method according to claim 2, characterized in that, The target selection method for obtaining the target fault tree includes: Get the fault location time for each selected fault tree fault location within a preset time period; The target time for selecting the fault tree, the historical average fault location time at the target time, and the production line change rate are obtained, wherein the target time is any time within the preset time period; The random probability parameters are determined based on the target time, the historical average fault location time at the target time, and the production line change rate. The target selection method for the target fault tree is determined based on the random probability parameters.

4. The method according to claim 2 or 3, characterized in that, The target selection method includes selecting the target fault tree that minimizes the fault location time. The step of selecting a target fault tree from the plurality of fault trees using the target selection method includes: Get the fault location time for each selected fault tree fault location within a preset time period; Based on the fault location time, select the target fault tree that minimizes the fault location time from the multiple fault trees.

5. The method according to claim 2 or 3, characterized in that, The target selection method includes randomly selecting a target fault tree; The step of selecting a target fault tree from the plurality of fault trees using the target selection method includes: A target fault tree is randomly selected from the plurality of fault trees.

6. The method according to claim 1, characterized in that, Determine the required troubleshooting time and cost for each of the aforementioned failure modes, including: Obtain the troubleshooting time and cost required for each type of failure mode in the lowest failure level; Obtain the first probability of occurrence for each type of fault mode in all fault levels, the first probability being obtained by the historical frequency of occurrence of the fault mode; Based on the first probability and the troubleshooting time and cost of each type of fault mode in the lowest fault level, determine the troubleshooting time and cost required for each type of fault mode other than the fault modes in the lowest fault level.

7. The method according to claim 1, characterized in that, Determine the troubleshooting path complexity for each of the aforementioned fault modes, including: The troubleshooting path complexity for each type of fault mode is determined based on the number of fault modes associated with each type of fault mode.

8. The method according to claim 1, characterized in that, The step of determining the fault diagnosis model's fault troubleshooting path based on the troubleshooting time, the troubleshooting cost, and the troubleshooting path complexity includes: Prioritize the investigation time, the investigation cost, and the investigation path complexity; For each type of fault mode, the fault diagnosis model's fault troubleshooting path is determined based on the priority of the troubleshooting time, the troubleshooting cost, and the troubleshooting path complexity.

9. The method according to claim 1, characterized in that, The method further includes: Obtain historical fault modes; Based on the historical failure modes, a second probability is determined from each target failure mode to a failure mode other than the target failure mode, resulting in a failure transition matrix, where the target failure mode is any one of the historical failure modes. Determine the third probability of each type of failure mode occurring in the highest failure level, given that each type of failure mode occurs in the lowest failure level. Based on the fault transition matrix and the third probability, determine the sequence of occurrence probabilities of each type of fault mode in the historical fault modes, as well as the fault mode sequence corresponding to the sequence.

10. The method according to claim 1, characterized in that, The method further includes: Determine a set of minimal cut sets that cause the occurrence of the failure mode in the highest failure level, each of the minimal cut sets including at least one class of failure modes in the lowest failure level; Based on the set of minimum cut sets, determine the first fault probability function for each type of fault mode in the highest fault level; Based on the first fault probability function, determine the second fault probability function for all fault modes in the highest fault level; Obtain the investigation time, investigation cost, optimization cost, historical occurrence frequency, and reliability requirements of the production line equipment for each of the minimum cut sets; Based on the investigation time, investigation cost, optimization cost, and historical occurrence frequency of each minimum cut set, a score for each minimum cut set is determined using a preset scoring rule. Based on the second failure probability function, the score of each of the minimum cut sets, and the relationship between the reliability requirement information and product lifespan, an objective function is constructed. Solving the objective function yields the objective minimum cut set, which includes at least one type of fault mode; The target minimum cut set is displayed to guide the user in optimizing the production line equipment.

11. The method according to claim 10, characterized in that, Solving the objective function to obtain the objective minimum cut set includes: The objective function is solved using the particle swarm optimization algorithm to obtain the minimum cut set of the objective. The solution process of the particle swarm optimization algorithm is as follows: Obtain the range of values ​​for the learning factor and the number of iterations; Based on the range of values ​​of the learning factor and the number of iterations, a selection function for the learning factor is fitted, which is used to select the learning factor for each iteration process according to the selection function. Based on the learning factor, the fitness of the position reached in each iteration is determined, and a descending sequence of the fitness is determined, wherein the position represents the minimum cut set; Obtain the target fitness that satisfies the preset retention condition, wherein the preset retention condition is to retain the first preset number of target fitnesss in the descending sequence, and randomly select the target fitness from the second preset number of fitnesss in the descending sequence according to the fourth probability; Determine the target position corresponding to the target fitness, where the target position represents the target minimum cut set.

12. The method according to claim 1, characterized in that, The method further includes: Construct a production line model corresponding to the production line equipment. The production line model includes personnel information, which includes personnel location information. If the personnel location information indicates that a person is outside a preset area, an abnormal alarm is triggered at the location corresponding to the personnel location information.

13. The method according to claim 12, characterized in that, The production line model also includes equipment information; the method further includes: Based on the equipment information, the production line equipment image corresponding to the production line model is displayed, and the location where the production line equipment malfunctions is marked in the production line equipment image.

14. A device for determining the location of a production line equipment fault, characterized in that, include: The acquisition module is used to acquire equipment information of the production line equipment. The determination module is used to determine the fault diagnosis model of the production line equipment based on the equipment information and using the target fault tree. The fault diagnosis model includes at least two fault levels, each fault level includes at least one type of fault mode, and there is a dependency relationship between adjacent fault levels. The determining module is also used to determine the troubleshooting time, troubleshooting cost, and troubleshooting path complexity required for each type of fault mode; The determining module is further configured to determine the fault diagnosis model's fault troubleshooting path based on the troubleshooting time, the troubleshooting cost, and the troubleshooting path complexity, wherein the fault troubleshooting path is a fault troubleshooting path from the highest fault level to the lowest fault level. The determining module is also used to determine the location where the production line equipment malfunctions according to the fault investigation path.

15. An electronic device, characterized in that, The device includes: a processor and a memory storing computer program instructions; When the processor executes the computer program instructions, it implements the method for determining the fault location of production line equipment as described in any one of claims 1-13.

16. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer program instructions, which, when executed by a processor, implement the method for determining the location of a production line equipment fault as described in any one of claims 1-13.

17. A computer program product, characterized in that, When the instructions in the computer program product are executed by the processor of the electronic device, the electronic device is able to perform the method for determining the location of a production line equipment fault as described in any one of claims 1-13.

Images