Method, computing device, storage medium and program product for determining fault conditions of industrial systems

Abstract

The present invention relates to a method, a computing device, a storage medium and a program product for determining a fault condition of an industrial system. The industrial system includes a controller, a component and a plurality of functions corresponding to the component. The method includes obtaining fault information in the industrial system, the fault information including an activated fault type. The method also includes determining the number of functions associated with a key fault type among the multiple activated fault types, the association between the function and the fault type being represented by the edge of a knowledge graph. The method also includes, in response to the number of functions associated with the key fault type being multiple, determining a target fault type among the multiple fault types whose correlation with the associated function is greater than a preset correlation threshold. The method also includes determining the fault condition of the function corresponding to the key fault type and the fault condition of the target component corresponding to the function based on the number and activation condition of the target fault type.

Description

Technical Field

[0001] The present invention relates to the field of industrial control, and more particularly, to a method, a computing device, a storage medium, and a program product for determining a fault condition of an industrial system. Background Art

[0002] An industrial system is a system based on mechanical, electrical, instrumentation and other components, which cooperates with various units (such as controllers, equipment, etc.) to complete various operations. For example, an industrial system can perform production, manufacturing, processing, and transportation operations. Troubleshooting industrial systems is a complex and meticulous process involving multiple links and multiple components. As the degree of automation of industrial systems increases, industrial systems become more and more complex, which increases the correlation between controllers and equipment. This makes it more difficult to identify and locate problems when industrial systems fail.

[0003] Traditional methods for determining industrial system fault conditions mainly include using fault tree analysis and judging fault conditions. By using the fault tree to predict faults, the optimal prediction model can be selected and used to predict faults.

[0004] In summary, the traditional method for determining the fault conditions of industrial systems has the following shortcomings: in complex industrial systems, multiple different combinations of events may lead to the same fault, which makes the structure of the fault tree more complicated and difficult to cover all possible fault paths, making the predicted fault conditions inaccurate. Summary of the invention

[0005] The embodiments of the present invention propose a method, a computing device, a computer storage medium and a computer program product for determining the fault condition of an industrial system, which uses a knowledge graph to characterize the association between functions or components and fault types, thereby improving the efficiency and accuracy of determining the fault condition.

[0006] In a first aspect of the present invention, a method for determining a fault condition of an industrial system is provided. The industrial system includes a controller, a component, and a plurality of functions corresponding to the component, and the relationship between the controller, the component, and the plurality of functions is characterized by a knowledge graph. The method includes obtaining fault information in the industrial system, and the fault information includes an activated fault type. The method also includes determining the number of functions associated with a key fault type among the plurality of activated fault types, and the association between the function and the fault type is represented by an edge of a knowledge graph. The method also includes, in response to the number of functions associated with the key fault type being multiple, determining a target fault type among the plurality of fault types whose correlation with the associated function is greater than a preset correlation threshold. The method also includes determining the fault condition of the function corresponding to the key fault type and the fault condition of the target component corresponding to the function based on the number and activation condition of the target fault type.

[0007] In a second aspect of the present invention, an electronic device is provided, which includes one or more processors and a storage device for storing one or more programs, and when the one or more programs are executed by the one or more processors, the one or more processors implement the method provided according to the first aspect of the present invention.

[0008] In a third aspect of the present invention, a computer-readable storage medium is provided. The computer-readable storage medium stores computer-executable instructions, wherein the computer-executable instructions are executed by a processor to implement the method provided according to the first aspect of the present invention.

[0009] According to a fourth aspect of the present invention, there is provided a computer program product, comprising machine executable instructions, which, when executed, enable a machine to perform the method provided by the first aspect of the present invention.

[0010] It should be understood that the contents described in the summary of the invention are not intended to limit the key or important features of the embodiments of the present invention, nor are they intended to limit the scope of the present invention. Other features of the present invention will become easily understood through the following description. BRIEF DESCRIPTION OF THE DRAWINGS

[0011] The above and other features, advantages and aspects of the embodiments of the present invention will become more apparent with reference to the following detailed description in conjunction with the accompanying drawings. In the accompanying drawings, the same or similar reference numerals represent the same or similar elements.

[0012] Figure 1 A schematic diagram is shown of an example environment in which various embodiments of the invention may be implemented.

[0013] Figure 2 A flow chart of a method for determining a fault condition of an industrial system according to some embodiments of the present invention is shown.

[0014] Figure 3 A schematic diagram showing the association between fault types and functions according to some embodiments of the present invention.

[0015] Figure 4A A schematic diagram is shown of determining a fault condition of a function using inference rules according to some embodiments of the present invention.

[0016] Figure 4B A schematic diagram showing a method of determining a fault condition of a function by using inference rules according to some other embodiments of the present invention is shown.

[0017] Figure 5 A schematic diagram is shown of determining a fault condition of a function based on fault information according to some embodiments of the present invention.

[0018] Figure 6 A schematic diagram is shown of determining a fault condition of an industrial system based on fault information according to some other embodiments of the present invention.

[0019] Figure 7 A schematic diagram of obtaining information on a target fault type according to some embodiments of the present invention is shown.

[0020] Figure 8 A schematic diagram illustrating a failure scenario of a determination function according to some embodiments of the present invention.

[0021] Fig. 9 A block diagram of an apparatus for determining a fault condition of an industrial system according to some embodiments of the present invention is shown.

[0022] Fig.10 A block diagram is shown of a device in which various embodiments of the present invention may be implemented. DETAILED DESCRIPTION

[0023] Embodiments of the present invention will be described in more detail below with reference to the accompanying drawings. Although certain embodiments of the present invention are shown in the accompanying drawings, it should be understood that the present invention can be implemented in various forms and should not be construed as being limited to the embodiments described herein, which are instead provided for a more thorough and complete understanding of the present invention. It should be understood that the drawings and embodiments of the present invention are only for exemplary purposes and are not intended to limit the scope of protection of the present invention.

[0024] In the description of the embodiments of the present invention, the term "including" and similar terms should be understood as open inclusion, that is, "including but not limited to". The term "based on" should be understood as "based at least in part on". The term "one embodiment" or "the embodiment" should be understood as "at least one embodiment". The terms "first", "second", etc. may refer to different or the same objects. Other explicit and implicit definitions may also be included below.

[0025] As mentioned above, for industrial systems, fault trees usually assume that the cause-effect relationships in industrial systems are linear and deterministic. However, in actual industrial systems, the relationships between faults may be highly nonlinear and affected by multiple variables, which makes it difficult for fault trees to accurately reflect the actual situation. In addition, constructing and interpreting fault trees requires professional knowledge. For users who are not familiar with the fault tree analysis method, it may be difficult to understand the logical relationships and analysis results in the tree, thereby increasing labor costs and construction costs.

[0026] To this end, an embodiment of the present invention proposes a method for determining a fault condition of an industrial system. The industrial system includes a controller, a component, and multiple functions corresponding to the component, and the relationship between the controller, the component, and the multiple functions is represented by a knowledge graph. The method includes obtaining fault information in the industrial system, and the fault information includes an activated fault type. The method also includes determining the fault condition of the corresponding function and the fault condition of the target component corresponding to the function based at least on the fault information and the association relationship between the fault type and the function corresponding to the component, and the association relationship is represented by the edge of the knowledge graph.

[0027] In this way, the knowledge graph can clearly and directly represent the relationship between components, functions and fault types in the industrial system, providing a simple and direct reference for determining the fault conditions in the system, thereby improving the efficiency and accuracy of determining the fault conditions. In addition, the knowledge graph is not limited by the scale and complexity of the industrial system, which ensures that fault analysis can be effectively performed regardless of the size of the system, thereby improving the accuracy of fault analysis and the stability of the system.

[0028] Figure 1 1 is a schematic diagram of an example environment 100 in which various embodiments of the present invention may be implemented. Figure 1As shown, in the environment 100, an industrial system 102, a control system 104 connected to the industrial system for communication, and a terminal device 106 are included. Among them, the industrial system 102 may include components and controllers for controlling or deploying components. The components may be sensors or actuators, or other devices. For example, the industrial system 102 may include multiple subsystems such as a network system, a power supply system, and a motion control system, and different subsystems may include different modules or components. For example, the motion control system may include a robot motion controller and a visual recognition unit. Different modules or devices have different functions. For example, the function corresponding to the robot motion controller is to control the robot joint movement and realize trajectory planning and motion control. The function corresponding to the visual recognition unit may be to identify the position and shape of the workpiece in real time and provide workpiece recognition data to the motion control system.

[0029] It should be noted that the number of components or controllers included in the industrial system 102 is large, and the relationship is relatively complex. In order to more clearly display the architecture of the industrial system 102 and facilitate subsequent fault analysis, a tree-shaped knowledge graph 108 can be constructed for the industrial system 102. For example, the various components and controllers in the industrial system 102 can be organized into a tree structure to reflect the hierarchical relationship and structure within the industrial system 102. It can be understood that each node in the knowledge graph 108 can represent a specific system function or controller, and each edge can be used to represent the association relationship or interaction relationship between them. For example, the industrial system can be used as a root node, the multiple subsystems included in the industrial system can be used as child nodes, the multiple components included in the subsystem can be used as child nodes of the subsystem, and the fault type corresponding to the component can be used as the child node of the component. The edge between nodes can be used to characterize the association relationship between nodes. For example, the function "data communication between the device and the cloud" is associated with the fault type "communication module failure" and "monitoring data transmission abnormality". In some embodiments, the function and the fault type can be a many-to-many association relationship, or a one-to-many or many-to-one association relationship. In this way, using knowledge graphs to represent the hierarchical relationships and structures of industrial systems can make the architecture of industrial systems clearer, and make the integrity of industrial systems and the interactions between modules more explicit and intuitive, making it easier for users to understand and maintain.

[0030] In some embodiments of the present invention, after receiving the data collection instruction from the control system 104, the industrial system 102 may collect fault information in response to the instruction and send the fault information to the control system 104. The control system 104 may determine the components or functions in the industrial system 102 that are in a "faulty" state based on the fault information and the association relationships shown in the knowledge graph 108, and may also determine the probability that other functions or components in the industrial system 102 are in a "faulty" state.

[0031] like Figure 1 As shown, after determining the fault condition of the industrial system 102, the control system 104 may send the fault condition to the terminal device 106, and the terminal device 106 may display the fault condition. Among them, the terminal device 106 may include a mobile smart phone, a computer (including a laptop computer, a desktop computer), a tablet electronic device, a personal digital assistant (PDA), etc., and the present invention is not limited here. The terminal device 106 may include a visualization interface, and the visualization interface may include multiple display components, and different display components are used to display different information. For example, display component A can be used to display the fault condition of function A, and display component B can be used to display the fault condition of function B. Of course, the visualization interface may also include multiple controls, and the multiple controls may include objects for users to interact with the visualization interface to input or operate data, such as trigger buttons such as "stop running" and "run". For example, a user (administrator or maintenance personnel) can comprehensively manage and optimize the industrial system 102 according to the fault condition. In this way, the fault analysis process can be made transparent and repeatable.

[0032] In this way, the knowledge graph can clearly and directly represent the relationship between components, functions and fault types in the industrial system, providing a simple and direct reference for determining the fault conditions in the system, thereby improving the efficiency and accuracy of determining the fault conditions. In addition, the knowledge graph is not limited by the scale and complexity of the industrial system, which ensures that fault analysis can be effectively performed regardless of the size of the system, thereby improving the accuracy of fault analysis and the stability of the system.

[0033] Figure 2 FIG. 2 is a flow chart showing a method 200 for determining a fault condition of an industrial system according to some embodiments of the present invention. The method 200 may be Figure 1 The control system 104 shown in FIG. Figure 2 As shown, the industrial system includes controllers, components, and multiple functions corresponding to the components. The relationship between the controller, components, and multiple functions is represented by a knowledge graph. Figure 1 The industrial system 102 shown in the figure may be an automation system composed of a controller and a plurality of industrial components. The industrial components may be used to collect and monitor data, and the controller may be used to control and coordinate the operation of the industrial components. Through the coordinated work of the controller and the plurality of industrial components, the automation operation of the entire industrial system may be realized.

[0034] In some embodiments, the function of an industrial component refers to the tasks that the industrial component can perform or the services that it can provide. Depending on the industrial component, the corresponding functions are also different. For example, when the industrial component is a sensor, the function can be to measure various physical, chemical, and mechanical parameters (such as temperature, pressure, pH value, flow rate) in the industrial process; when the industrial component is an actuator, the function can be to adjust the control object, such as a motor, valve, pump, etc. In one example, when the industrial component is a power supply management unit, the corresponding function of the component can be to monitor the power supply status, control power distribution, etc.

[0035] In some embodiments of the present invention, a knowledge graph (e.g. Figure 1 The knowledge graph 108 shown in the figure can organize the various components and controllers of the industrial system into a tree structure to reflect the hierarchical relationship and logical structure within the industrial system. Each node in the knowledge graph can represent a controller, a component or a specific function. The edges in the knowledge graph can be used to represent the association and interaction between them. In some embodiments, the industrial system can be used as the root node, the component as the child node, and the function corresponding to the component as the child node of the component. This tree structure intuitively reflects the controller's deployment relationship to the component, as well as the subordinate relationship between the component and the function. In other embodiments, the fault type is further used as a child node of the function. It can be understood that the knowledge graph provides an intuitive visual display method for the internal structure of the industrial system, making the structure and relationship of the industrial system easier to understand and analyze. Moreover, through a graphical interface, it is easy to view the system's hierarchical structure, logical relationship, and the association or interaction relationship between the components.

[0036] In box 202, method 200 may include obtaining fault information in the industrial system, the fault information including the activated fault type. Fault information refers to the fault type activated in the industrial system and the specific information of the fault type. A fault may be caused by a problem in one or more functions, or it may affect the normal operation of a module corresponding to one or more functions. In some embodiments, during the operation of the industrial system, fault information may be collected in real time by sensors or other monitoring means, and the activated fault type may be determined based on the fault information. Among them, the fault information may include abnormal signals in the industrial system, performance degradation indicators of components, etc. The association relationship may be represented and stored by means of a knowledge graph, a fault tree, a dependency matrix, etc. In the case of using a knowledge graph to represent the association relationship, the edge of the knowledge graph may be used to represent the association relationship between the function and the fault type.

[0037] It should be noted that all fault information can be obtained at one time, for example, all fault type information can be received at one time. In this way, the fault condition of the industrial system can be comprehensively analyzed based on all fault information. In other embodiments, key fault type information can also be obtained first, and more fault type information can be requested based on the fault type information. In this way, by sequentially obtaining multiple fault type information, the root cause analysis of the fault can be gradually and deeply performed, unnecessary data transmission can be reduced, and the analysis efficiency can be improved.

[0038] In box 204, method 200 may include determining the number of functions associated with the key fault type among the multiple fault types that are activated, and the association between the function and the fault type is represented by the edge of the knowledge graph. It can be understood that, using the association relationship in the knowledge graph (such as the edges of the knowledge graph), starting from the activated fault type, the corresponding function and target component are traced along the edge of the knowledge graph. Through this process, it can be determined which functions are affected and which components may be the source of the fault. In some embodiments, the fault condition can be the state of the function or component, such as being in a "faulty" state or in a "non-faulty" state. In other embodiments, the fault condition may also include the probability that the component or function is in a "faulty" state or the probability of a fault. For example, the fault condition of function S may be that the probability of function S failing is 40%.

[0039] It can be understood that there may be multiple activated fault types. If the fault condition of the industrial system is determined based on multiple fault types at the same time, resources may be wasted, and the efficiency and accuracy of fault diagnosis analysis may be low. On this basis, according to the actual situation, one of the fault types can be selected from multiple fault types as the key fault type, and the fault type of the industrial system can be gradually inferred based on the key fault type. Among them, the key fault type can be set by the user according to the actual industrial scenario and industrial needs.

[0040] In some embodiments of the present invention, the association relationship between the fault type and the function can be a one-to-one, one-to-many or many-to-many association relationship. The complexity of the fault diagnosis analysis varies according to the different association relationships. On this basis, the reasoning path and reasoning method of the fault can be determined according to the type of association relationship between the key fault type and the function or according to the number of functions corresponding to the key fault type. For example, in the case where there are multiple functions corresponding to the key fault type, in order to more accurately determine the fault conditions of all the corresponding functions, it can be determined in combination with whether other fault types are activated and the functions corresponding to other fault types.

[0041] In box 206, method 200 may include, in response to the number of functions associated with the critical fault type being multiple, determining a target fault type among the multiple fault types whose correlation with the associated function is greater than a preset correlation threshold. For example, in the case where the number of functions associated with the critical fault type is 3, such as function 1, function 2, and function 3, the target fault types whose correlation with function 1, function 2, and function 3 is greater than the preset correlation threshold may be determined respectively. For example, there are multiple fault types associated with function 1, namely, fault type 1, fault type 2, and fault type 3. If the correlation between fault type 3 and function 1 is greater than the correlation between fault type 2 and fault type 1 and function 1, fault type 3 may be used as the target fault type. It is understood that all fault types or some fault types associated with function 1 may also be used as the target fault type.

[0042] In box 208, method 200 may include determining the fault condition of the function corresponding to the critical fault type and the fault condition of the target component corresponding to the function based on the number and activation of the target fault type. In some embodiments, it can be determined whether the target fault type is activated according to the fault type. For example, the fault condition of function 1 can be determined according to the activation of fault type 3. For example, when the target fault type (fault type 3) corresponding to function 1 is not activated, it can be determined that function 1 is in a "non-fault" state. Or, when the target fault types corresponding to function 1 are fault type 3 and fault type 2. If both fault type 3 and fault type 2 are activated, it can be determined that the function 1 fault is in a "fault" state. In this way, the fault conditions of other functions such as function 2 and function 3 corresponding to the critical fault type can be determined. In this way, the mode of step-by-step reasoning and step-by-step verification can reduce the possibility of missed judgment or misjudgment, thereby helping to improve the accuracy of fault analysis and the stability of the system. In addition, through a clear exclusion mechanism, repeated inspections of known normal functions are avoided, thereby saving diagnostic time and resources.

[0043] In some embodiments of the present invention, the fault condition of a component can be determined based on the fault conditions of multiple functions included in the component. For example, if a key function of a component is in a "fault" state, it can be determined that the component is in a "fault" state or that the probability of failure of the component is high (for example, greater than a preset fault probability threshold). Among them, the key functions can be set by the user according to actual needs. In some embodiments, if the number of functions in a "fault" state among the multiple functions included in the component is greater than a preset number threshold, it can also be determined that the component is in a "fault" state. In other embodiments, the failure probability of a component can also be determined based on the statistical value or derived value of the failure probability of multiple functions.

[0044] In this way, the knowledge graph can clearly and directly represent the relationship between components, functions and fault types in the industrial system, provide a simple and direct reference for determining the fault conditions in the system, help to quickly locate the source of the fault, improve the accuracy and reliability of fault diagnosis, and reduce downtime and maintenance costs. In addition, the knowledge graph is not limited by the scale and complexity of the industrial system, which ensures that fault analysis can be effectively performed regardless of the size of the system, thereby improving the accuracy of fault analysis and the stability of the system. In addition, the fault analysis process can be expanded and adjusted according to actual needs to adapt to different industrial environments and fault modes, thereby increasing the scope of application of fault analysis.

[0045] Figure 3 A schematic diagram showing the association between fault types and functions according to some embodiments of the present invention is shown. Figure 3 As shown, in the example environment 300, an abnormality in function 1 will activate fault a and fault b; an abnormality in function 2 will activate fault c; an abnormality in function 3 will activate fault c and fault d. On this basis, the association relationship between all fault types and all functions can be: function 1 is associated with fault a and fault b; function 2 is associated with fault c, function 3 is associated with fault c, and function 3 is associated with fault d. In some embodiments, the association relationship between functions and fault types can be represented in the form of an association relationship table or an association relationship diagram. It can be understood that the association relationship between functions and fault types can be a one-to-many, many-to-one, or many-to-many association relationship.

[0046] In some embodiments of the present invention, when a component or device in an industrial system is damaged, other components or devices related to it may also be affected. For example, if a sensor fails, it may cause the related actuator or controller to fail (that is, the fault type associated with it will be activated). In addition, a functional abnormality can activate all fault types related to it. On this basis, the activation of a fault type may be caused by a related functional failure, or it may be caused by the failure of multiple related functions. According to the above-mentioned situation, a variety of inference rules can be pre-set by the user. On this basis, combined with the inference rule and the activated fault type, the function, component that causes the fault type to be activated can be inferred, or the failure probability of the component corresponding to the activated fault type or the failure probability of the function can be inferred.

[0047] In some embodiments, the inference rules may include reverse reasoning rules, propagation rules, combination rules, selective rules, and exclusion rules. Among them, the reverse reasoning rule refers to inferring possible causes based on the activated fault type, such as inferring the components or functions that cause the fault type to be activated. The propagation rule refers to the failure of related components when a certain component fails. For example, when a bus or network fails, all devices or components linked to it may be affected. In one example, the CAN bus of the industrial networking module fails, and all sensors and actuators connected to the bus may report fault types related to bus communication.

[0048] In some embodiments, a combination rule refers to multiple functions or multiple components jointly causing the fault type to be activated. A selective rule refers to multiple factors such as one or some factors in multiple components or multiple functions causing a certain fault type to be activated. For example, among multiple possible factors, the occurrence of one factor causes the fault type to be activated. Combining the activation of the fault type and the combination rule, it can be determined which one or several factors jointly cause the fault type to be activated. For example, in the case where a power supply fault is activated, and there are multiple power supplies in the industrial system, the selective rule can select one or several power supplies as the most likely source of the fault based on factors such as the usage status and load conditions of the power supply. In some embodiments, an exclusion rule refers to the occurrence of certain factors that excludes the possibility that other factors cause the fault type to be activated. For example, in the case where a communication-related fault type is activated, if it has been determined that it is caused by a network hardware failure, then the exclusion rule will exclude other communication-related software or configuration errors as the possibility of the fault cause.

[0049] Figure 4A FIG. 2 is a schematic diagram showing a method of determining a function failure condition using inference rules according to some embodiments of the present invention. Figure 4A As shown, in the example environment 400A, the fault types associated with function 1 may be fault a and fault b. When fault a is activated and fault b is not activated, the fault condition of function 1 cannot be determined, that is, it cannot be determined that function 1 is in a "fault" state.

[0050] Figure 4B FIG. 2 is a schematic diagram showing a method of determining a function failure using inference rules according to some other embodiments of the present invention. Figure 4BAs shown, in the example environment 400B, the fault types associated with function 1 are fault a, fault b, and fault c; the fault type associated with function 2 is fault c; and the fault types associated with function 3 are fault c and fault d. According to the reverse reasoning rule, when both fault c and fault d are activated, it can be determined that there is a possibility that both function 2 and function 3 have faults. Furthermore, since fault d is activated, and the functions associated with fault d only include function 3, it can be determined that the probability that function 3 is in a "fault" state is 100%. In addition, according to the exclusion rule, since there are multiple functions associated with fault c (function 1, function 2, function 3), and the other fault types associated with function 1 are in an inactive state (i.e., the probability that function 1 fails is 0) and the probability that function 3 fails is 100%, the probability that function 2 fails is 50%.

[0051] In this way, by comprehensively applying multiple reasoning rules, we can more comprehensively consider the correlation between fault types and functions, as well as the mutual influence between fault types, avoiding the process of blind troubleshooting and trial and error, thereby further improving the accuracy and efficiency of determining industrial system fault conditions.

[0052] Figure 5 FIG. 1 is a schematic diagram showing a fault condition of a function determined based on fault information according to some embodiments of the present invention. Figure 5 As shown, in the example environment 500, the industrial system may include a networked system, a power system, and a motion control system. The networked system, the power system, and the motion control system may be controllers of the industrial system for controlling the components contained therein. The networked system may include an industrial Internet of Things module and a data acquisition unit; the power system may include a power supply management unit and a battery management system; the motion control system may include a robot motion controller and a visual recognition unit. Different components or units may have different functions. For example, the functions corresponding to the industrial Internet of Things module may include data communication between the device and the cloud and remote monitoring of the device status. The functions corresponding to the data acquisition unit may include collecting device sensor data and data preprocessing and storage.

[0053] like Figure 5As shown, there can be multiple types of faults in an industrial system, and one type of fault can correspond to one function or multiple functions. Of course, one function can also correspond to one type of fault or multiple types of faults. The association between functions and fault types can be a one-to-many or many-to-many relationship. In some embodiments, a global fault reporting mode can be used to determine the fault condition of an industrial system. For example, the control system corresponding to the industrial system can provide all fault information at one time, that is, the control system corresponding to the industrial system can return relevant information of all activated fault types at one time. It should be noted that functions and fault types can be coded with numbers, and the use of digital coding to express functions and fault types can improve information security to a certain extent. For example, the digital code corresponding to the function "data communication between the device and the cloud" and the fault type "communication module failure" is (1,1,1,1), and the digital code corresponding to the function "control battery charging and discharging" and the fault type "charging logic error" is (2,2,2,8).

[0054] like Figure 5 As shown, when it is determined that the activated fault types are "communication module failure", "power input abnormality" and "joint drive abnormality", the function corresponding to "communication module failure" and the number of corresponding functions, the function corresponding to "power input abnormality" and the number of corresponding functions, and the function corresponding to "joint drive abnormality" and the number of corresponding functions can be determined respectively. For example, the function corresponding to the activated fault type of "communication module failure" is "data communication between device and cloud". According to the reverse reasoning rule and the exclusion rule, since "communication module failure" only corresponds to one function, when this fault type is activated, it can be determined that the function of "data communication between device and cloud" is in an "abnormal state" or "faulty state". In other words, the probability of failure of the "data communication between device and cloud" function is 100%. In summary, in this case, when the number of functions associated with or corresponding to the activated fault type is one, it can be simply and directly determined that the function is in a faulty state; and when the number of functions associated with or corresponding to the activated fault type is multiple, it is necessary to combine other activated fault types to comprehensively judge the fault states corresponding to each function. In other examples, if the "collect device sensor data" function is in the "fault" state, the "sensor connection failure" and "joint drive failure" fault types will be activated. However, if the "sensor connection failure" fault type is not currently activated, it can be determined that the "collect device sensor data" function is in the "non-fault" state. Similarly, if the "distribution logic error" fault type is not activated, it can be determined that the corresponding "control power distribution" function is in the "non-fault" state.

[0055] like Figure 5As shown, the functions corresponding to the activated fault type of "power input abnormality" are "data communication between device and cloud" and "monitoring power supply status". When the "data communication between device and cloud" function is determined to be in the "fault" state, the possibility that the "monitoring power supply status" function is in the "non-fault" state cannot be ruled out, so it can be determined that the probability of the "monitoring power supply status" function being in the "fault" state is 50%. In other embodiments, if the fault type of "power input abnormality" also corresponds to other functions such as function A and function B, then function A and function B as well as the "data communication between device and cloud" function may cause "power input abnormality" to be activated, then the probability of function A, function B and the "data communication between device and cloud" function being in the "fault" state is 33.3% each. In this way, considering that multiple functions may cause the same fault type to be activated, the probability of failure of each function can be more accurately evaluated by the probability averaging method, thereby improving the accuracy of fault condition judgment.

[0056] like Figure 5 As shown, in some embodiments, according to the exclusion rules and the selectivity rules, the other two functions (the "collecting device sensor data" function and the "controlling power distribution" function) of activating the "joint drive failure" are in the "non-fault" state, then it can be determined that the probability of failure of another function "controlling robot joint drive" corresponding to the "joint drive failure" is 100%, that is, the function is in the "fault" state.

[0057] In this way, the global fault reporting model can ensure that all fault types of the industrial system are reported and analyzed in a timely manner, which helps to quickly and comprehensively discover and solve problems, thereby improving the reliability of the system. In addition, by collecting all fault type information at one time, the fault conditions of the industrial system can be fully inferred, so that the root causes of complex faults can be quickly located and the efficiency of fault diagnosis can be improved.

[0058] Figure 6 FIG. 2 is a schematic diagram showing a method of determining a fault condition of an industrial system according to fault information according to some other embodiments of the present invention. Figure 6As shown, in the example environment 600, the progressive fault analysis mode can be used to determine the fault condition of the industrial system. For example, the control system of the industrial system can first report the critical fault type. Preliminary reasoning is performed based on the received critical fault type, and based on the reasoning result, the control system of the industrial system can be requested to report further fault information, so as to gradually determine the fault condition of the industrial system. Among them, the critical fault type can be set by the user according to the actual industrial scenario and industrial needs. For example, when the joint drive function of the industrial system is more important, "joint drive failure" can be used as a critical fault type. On this basis, the information that the "joint drive failure" is activated returned by the control system can be received first.

[0059] Since there are multiple functions associated with the "joint drive failure" such as "collecting device sensor data", "controlling power distribution" and "controlling robot joint movement", it is impossible to determine the fault conditions of the "collecting device sensor data", "controlling power distribution" and "controlling robot joint movement" functions based only on the information that the "joint drive failure" is activated. On this basis, the fault conditions of the three can be determined in combination with the situations in which other fault types are activated. It can be understood that the information of the target fault type that needs to be obtained in the next step can be determined based on the degree of association between multiple fault types and the functions associated with the "joint drive failure". For example, the fault type with the highest degree of association with the function "collecting device sensor data" associated with the "joint drive failure" is "sensor connection failure", and the target fault type that needs to be obtained in the next step is "sensor connection failure". Similarly, the fault type with the highest degree of association with the function "controlling power distribution" associated with the "joint drive failure" is "allocation logic error", and the other target fault type that needs to be obtained in the next step is "allocation logic error".

[0060] Figure 7 FIG. 1 is a schematic diagram showing obtaining information about a target fault type according to some embodiments of the present invention. Figure 7As shown, in the example environment 700, it can be determined that the information of the target fault type is "sensor connection failure" is activated while "allocation logic error" is not activated. Since whether the two fault types are activated is only caused by a single functional abnormality, when the "sensor connection failure" fault type is activated, it can be determined that the probability of the function "collecting device sensor data" corresponding to this fault type failing is 100%, that is, the "collecting device sensor data" function is in a "fault" state. It can be understood that through a clear single function fault mapping, the fault state of certain functions can be determined quickly and accurately, reducing unnecessary inspections and troubleshooting, thereby saving processing resources and processing costs. When the "allocation logic error" is not activated, it can be determined that the probability of the function "control power distribution" corresponding to this fault type failing is 0, that is, the "control power distribution" function is in a "non-fault" state. On this basis, the probability of the "control robot joint movement" function failing is 50%.

[0061] In this way, using the progressive fault analysis mode to determine the fault condition of the industrial system can request more fault code information on demand, avoiding unnecessary data transmission and optimizing the use of bandwidth and computing resources. On the other hand, by gradually verifying the inference results, the possibility of misjudgment can be reduced, which helps to improve the accuracy of fault analysis and the stability of the system. In addition, through a clear exclusion mechanism, repeated checks of known normal functions are avoided, saving diagnostic time and resources.

[0062] It is understandable that the appropriate reasoning mode can be selected according to different application scenarios. For example, the global fault reporting mode can be selected for resource-rich environments, while the progressive fault analysis mode can be selected in environments with limited bandwidth or computing power.

[0063] Figure 8 Schematic diagram showing a failure situation of a determination function according to some embodiments of the present invention. Figure 8As shown, in the example environment 800, component A includes three functions such as function a, function b and function c, wherein the fault type associated with function b is multiple including fault 1, fault 2 and fault 3. When the fault rate corresponding to fault 1 is 30%, the fault rate corresponding to fault 2 is 40% and the fault rate corresponding to fault 3 is 50%, the following formula can be used: 1-(1-0.3)×(1-0.4)×(1-0.5)=0.79 to determine that the probability of function b failing is 0.79. It can be understood that a similar method can be used to determine the probability of component A failing. For example, when the probability of function a failing is 0.2, the probability of function b failing is 0.79 and the probability of function c failing is 0.45, the following formula can be used: 1-(1-0. 2)×(1-0.79)×(1-0.45)=0.9244 to determine that the failure rate of component A is 92.44%. In other embodiments, the failure rate of component A may also be determined in other ways, and the present invention is not limited thereto.

[0064] In this way, the failure probability of each function within the component and its failure type can be comprehensively considered, so that the overall failure condition of the component can be predicted more accurately, avoiding the limitations of prediction based on only a single fault or function, thereby improving the accuracy of determining the failure condition of the industrial system.

[0065] Fig. 9 A block diagram of an apparatus 900 for determining a fault condition of an industrial system according to some embodiments of the present invention is shown. The industrial system includes a controller, a component, and multiple functions corresponding to the component, and the relationship between the controller, the component, and the multiple functions is represented by a knowledge graph. Fig. 9 As shown, the device 900 includes a fault information acquisition unit 902, which is configured to acquire fault information in the industrial system, and the fault information includes an activated fault type. The device 900 also includes a function quantity determination unit 904, which is configured to determine the number of functions associated with the key fault type among the multiple activated fault types, and the association relationship between the function and the fault type is represented by the edge of the knowledge graph. The device 900 also includes a target fault type determination unit 906, which is configured to determine the target fault type among the multiple fault types whose correlation with the associated function is greater than a preset correlation threshold in response to the number of functions associated with the key fault type being multiple. The device 900 also includes a fault condition determination unit 908, which is configured to determine the fault condition of the function corresponding to the key fault type and the fault condition of the target component corresponding to the function based on the number of target fault types and the activation condition.

[0066] It can be understood that by using the device 900 of the present invention, at least one of the advantages that can be achieved by the method or process described above can be achieved.

[0067] In some embodiments of the present invention, the fault condition determination unit 908 is further configured to: determine multiple functions associated with the fault type based on the association relationship between the fault type and the function of the component; and determine the fault conditions corresponding to the multiple functions associated with the fault type based on preset inference rules and the association relationship.

[0068] In some embodiments of the present invention, the fault condition determination unit 908 is further configured to: in response to the number of functions associated with the first fault type being one, determine a corresponding functional fault.

[0069] In some embodiments of the present invention, the fault condition determination unit 908 is further configured to: in response to the number of functions associated with the critical fault type being multiple, determine the target fault type associated with the first function among the corresponding multiple functions; based on the fault information, determine the activation condition corresponding to the target fault type; and based on the activation condition corresponding to the target fault type, determine the fault condition of the first function and the fault condition of the target component corresponding to the first function.

[0070] In some embodiments of the present invention, the fault condition determination unit 908 is also configured to: determine whether there is an unactivated target fault type in the second fault type associated with the first function; and in response to the presence of an unactivated target fault type in the target fault type, determine that the fault condition of the first function is "not faulty".

[0071] In some embodiments of the present invention, the fault condition determination unit 908 is further configured to: obtain a second function associated with the target fault type based on the association relationship; and determine the fault condition of the second function based on the fault condition of the first function.

[0072] In some embodiments of the present invention, the target component includes corresponding third and fourth functions, and the fault condition determination unit 908 is further configured to determine the fault condition of the target component based on the fault condition of the third function and the fault condition of the fourth function.

[0073] In some embodiments of the present invention, the industrial system is a parent node, the controller is a first child node, the component is a second child node, the function is a third child node, and the fault type is a fourth child node; the association relationship between the function and the fault type is a many-to-many relationship.

[0074] Fig.10A schematic block diagram of an example device 1000 that can be used to implement an embodiment of the present invention is shown. As shown, the device 1000 includes a computing unit (i.e., CPU 1001), which can perform various appropriate actions and processes according to computer program instructions stored in a read-only memory (i.e., ROM 1002) or loaded from a storage unit 1008 to a random access memory (i.e., RAM 1003). In RAM 1003, various programs and data required for the operation of the device 1000 can also be stored. CPU 1001, ROM 1002, and RAM 1003 are connected to each other via a bus 1004. An input / output interface (i.e., I / O interface 1005) is also connected to the bus 1004.

[0075] A number of components in the device 1000 are connected to the I / O interface 1005, including: an input unit 1006, such as a keyboard, a mouse, etc.; an output unit 1007, such as various types of displays, speakers, etc.; a storage unit 1008, such as a disk, an optical disk, etc.; and a communication unit 1009, such as a network card, a modem, a wireless communication transceiver, etc. The communication unit 1009 allows the device 1000 to exchange information / data with other devices through a computer network such as the Internet and / or various telecommunication networks.

[0076] CPU 1001 may be a variety of general and / or special processing components with processing and computing capabilities. Some examples of CPU 1001 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various dedicated artificial intelligence (AI) computing chips, various computing units running machine learning model algorithms, digital signal processors (DSPs), and any appropriate processors, controllers, microcontrollers, etc. The computing unit 1001 performs the various methods and processes described above, such as method 200. For example, in some embodiments, method 200 may be implemented as a computer software program, which is tangibly contained in a machine-readable medium, such as a storage unit 1008. In some embodiments, part or all of the computer program may be loaded and / or installed on the device 1000 via ROM 1002 and / or communication unit 1009. When the computer program is loaded into RAM 1003 and executed by CPU 1001, one or more steps of method 200 described above may be performed. Alternatively, in other embodiments, CPU 1001 may be configured to execute method 200 in any other appropriate manner (e.g., by means of firmware).

[0077] The functions described above herein may be performed at least in part by one or more hardware logic components. For example, without limitation, exemplary types of hardware logic components that may be used include: field programmable gate arrays (FPGAs), application specific integrated circuits (ASICs), application specific standard products (ASSPs), systems on chip systems (SOCs), load programmable logic devices (CPLDs), and the like.

[0078] The program code for implementing the method of the present invention can be written in any combination of one or more programming languages. These program codes can be provided to a processor or controller of a general-purpose computer, a special-purpose computer or other programmable data processing device, so that the program code, when executed by the processor or controller, enables the functions / operations specified in the flow chart and / or block diagram to be implemented. The program code can be executed entirely on the machine, partially on the machine, partially on the machine as a stand-alone software package and partially on a remote machine, or entirely on a remote machine or server.

[0079] In the context of the present invention, a machine-readable medium may be a tangible medium that may contain or store a program for use by or in conjunction with an instruction execution system, device or equipment. A machine-readable medium may be a machine-readable signal medium or a machine-readable storage medium. A machine-readable medium may include, but is not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, device or equipment, or any suitable combination of the foregoing. A more specific example of a machine-readable storage medium may include an electrical connection based on one or more lines, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing. In addition, although each operation is depicted in a specific order, this should be understood as requiring such operations to be performed in the specific order shown or in a sequential order, or requiring that all illustrated operations should be performed to obtain the desired result. In certain environments, multitasking and parallel processing may be advantageous. Similarly, although several specific implementation details are included in the above discussion, these should not be construed as limiting the scope of the present invention. Certain features described in the context of a separate embodiment may also be implemented in a single implementation in combination. On the contrary, the various features described in the context of a single implementation may also be implemented in multiple implementations individually or in any suitable sub-combination.

[0080] Although the subject matter has been described in language specific to structural features and / or methodological logical actions, it should be understood that the subject matter defined in the appended claims is not necessarily limited to the specific features or actions described above. On the contrary, the specific features and actions described above are merely example forms of implementing the claims.

Claims

1. A method for determining a fault condition of an industrial system, characterized in that The industrial system includes a controller, a component, and multiple functions corresponding to the component, the relationship between the controller, the component, and the multiple functions is represented by a knowledge graph, and the method includes: Acquiring fault information in the industrial system, the fault information including multiple activated fault types; Determine the number of functions associated with the key fault types among the activated multiple fault types, wherein the association between the functions and the fault types is represented by the edges of the knowledge graph; In response to the number of functions associated with the critical fault type being multiple, determining a target fault type among the multiple fault types whose correlation with the associated functions is greater than a preset correlation threshold; and Based on the number and activation status of the target fault types, a progressive fault analysis mode is used to determine the fault status of the functions corresponding to the key fault types and the fault status of the target components corresponding to the functions. The activation status includes whether all the target fault types are activated.

2. The method according to claim 1, characterized in that Determine the corresponding function fault conditions include: Based on the association relationship between the fault type and the function of the component, determining a plurality of functions associated with the fault type; and Based on the preset inference rule and the association relationship, the fault conditions respectively corresponding to the multiple functions associated with the fault type are determined.

3. The method according to claim 1, characterized in that Also includes: In response to the number of functions associated with the critical fault type being one, a functional fault corresponding to the critical fault type is determined.

4. The method according to claim 1, characterized in that: Determining the failure conditions of the functions corresponding to the key failure types includes: In response to the number of functions associated with the critical fault type being multiple, determining a target fault type associated with a first function among the corresponding multiple functions; Based on the fault information, determining an activation status corresponding to the target fault type; and Based on the activation status corresponding to the target fault type, a fault status of the first function and a fault status of a target component corresponding to the first function are determined.

5. The method according to claim 4, characterized in that Determining the fault condition of the target component corresponding to the first function includes: determining whether there is an inactivated target fault type among the target fault types associated with the first function; and In response to the existence of an inactivated target fault type among the target fault types, determining the fault condition of the first function as “not faulted”.

6. The method according to claim 5, characterized in that Determining the failure condition of the corresponding function also includes: Based on the association relationship, obtaining a second function associated with the target fault type; and Based on the failure condition of the first function, a failure condition of the second function is determined.

7. The method according to claim 1, characterized in that The target component includes a corresponding third function and a fourth function, and determining a fault condition of the corresponding function and a fault condition of the target component corresponding to the function includes: Based on the fault condition of the third function and the fault condition of the fourth function, a fault condition of the target component is determined.

8. The method according to claim 1, characterized in that The industrial system is a parent node, the controller is a first child node, the component is a second child node, the function is a third child node, and the fault type is a fourth child node; the association relationship between the function and the fault type is a many-to-many relationship.

9. A computing device, characterized in that: include: at least one processor; as well as A memory coupled to the at least one processor and having instructions stored thereon, the instructions, when executed by the at least one processor, causing the apparatus to perform the method according to any one of claims 1-8.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, and when the computer program is executed by a machine, the method according to any one of claims 1 to 8 is performed.

11. A computer program product, characterized in that The method comprises machine executable instructions which, when executed, cause a machine to perform the method according to any one of claims 1 to 8.

Images