Visual Interlocking Specification for Industrial Automation

Abstract

The present invention relates to the field of methods for industrial processes, in particular for industrial plants and / or processes, in particular for visualizing the rules of an industrial process, the method comprising the steps of providing a topology model of the industrial process, wherein the industrial process comprises at least one sensor and at least one actuator, associating rules (R) with the topology model, wherein the rules (R) comprise three parts: Cause (C), Traversal (T), Effect (E), where Cause (C) comprises a range of values ​​from at least one sensor, Effect (E) comprises an action to be performed by at least one actuator, and Traversal (T) comprises a relationship between Cause (C) and Effect (E); marking causes (C), traversals (T), and / or effects (E); and visualizing the elements of the rule (R) in the topology model.

Description

[Technical field]

[0001] The present invention relates to the field of methods for industrial processes, in particular for industrial plants and / or processes, for example in chemical, mechanical and / or other production and / or handling processes. Furthermore, the present invention relates to a computer readable storage medium, an artificial neural network (ANN) and uses. [Background technology]

[0002] To describe industrial plants and / or processes, so-called "interlocks" are often used. Algorithms may implement these interlock specifications, for example by basing them on plant topology models from process engineers. However, at least in some cases, safety engineers, service personnel, and / or others whose work involves these plants and / or processes, for example, may not trust such algorithms when they are executed automatically and may therefore revert to tedious and error-prone manual interlock specifications. Summary of the Invention

[0003] It is therefore an object of the present invention to provide an improved ability to check the behavior of industrial plants and / or processes. This object is achieved by the subject matter of the independent claims. Further embodiments are evident from the dependent claims and the following description.

[0004] One aspect relates to a method for visualizing rules of an industrial process or processing system, the method comprising: providing a topology model of an industrial process or processing system, wherein the industrial process or processing system comprises at least one sensor and at least one actuator; associating a rule with the topology model, where the rule has a triple: cause, traversal, effect; Where: The cause comprises a range of values ​​from at least one sensor; the result comprises an action performed by at least one actuator; Traversal comprises the relationship between cause and effect. marking the cause, traversal, and / or effect; and visualizing the elements of the rule in the topology model.

[0005] The industrial process may in particular be an industrial automation application and / or an industrial process in an industrial plant, for example in chemical, mechanical, and / or other production and / or handling processes. The rule, or rules, may be part of a so-called interlock specification, which describes at least some aspects of the industrial process. The rules may link two or more devices, for example based on the functions for which they are interlocked. At least some interlock specifications may comprise rules that may describe relationships of the devices, for example their functional and / or other relationships. The rules may comprise safety regulations. For example, a pump may be started only if a valve at its outlet is opened to prevent overpressure.

[0006] The visualization may be performed on a visual interface of a computing device, such as a display, a screen, a touch screen, a beamer, and / or the like. The visualization may comprise, for example, generating a visual specification of the interlock specification. The visualization may be implemented as a computer-implemented method.

[0007] A topological model of an industrial process may be represented by visualization means elements of a process or a plant, as well as their connectivity, e.g. pipes, vessels, etc. for transporting fluids and / or media, electrical connections, mechanical, logical and / or other connections. The topological model comprises at least one sensor and one actuator. These devices of the topological model naturally strongly depend on the type of process or plant in which they are used. The sensors may comprise metering devices, e.g. for pressure, temperature, light, sound, etc., and may include complex devices, e.g. for distance, flow, chemical analysis, and / or for further purposes. The actuators may comprise switches, valves, heaters, coolers, mixers, etc., configured to bring about a physical effect.

[0008] For interlocking specifications, a topology model has associated thereto a rule, or multiple rules, where each rule comprises the three parts <cause, traversal, result>. A rule may comprise further entries. The entry "cause" comprises a range of values ​​from at least one sensor. The range of values ​​may be direct from one or more sensors, pre-processed and / or a result of linking several sensors. The entry "result" comprises any action performed by the actuator (or actuators) depending on the actuator type. The entry "traversal" comprises a relationship between causes and results. One or more causes may contribute to one or more results. One or more causes may be connected logically (AND, OR, ...) and / or in other ways, e.g. via functions. One or more results may be connected logically and / or in other ways, e.g. via functions.

[0009] Marking the cause, traversal, and / or effect may be implemented, for example, by clicking and / or another input on one or more sensors, on one or more actuators, on one or more traversal paths of the topology model, and / or by marking the rules (or "lines") of the interlock specification. Visualizing the elements of the rules in the topology model may be implemented, for example, by highlighting the affected components, for example, by coloring the involved elements and / or their backgrounds. Additionally or alternatively, the involved rules may be indicated, for example, on a visual interface of a computing device. The visualization may be accompanied by further actions.

[0010] This may provide an improved ability to access the behavior of an industrial plant and / or process. It may advantageously contribute to visual interlock specifications that provide visual feedback on algorithmic interlock specifications to safety engineers and / or others working in the process or plant. As a further consequence, the method may enable faster engineering of interlocks, may reuse existing domain-specific knowledge, may be less error-prone, and may reduce or avoid costly inconsistencies in specifications, e.g., in plant and / or process interlock specifications and / or other specifications.

[0011] In various embodiments, the topology model is associated with the rules based on visualization means. This can be done, for example, by marking causes and effects, for example by "clicking" on them, describing further relationships between causes and effects, and / or visually programming. This further contributes to an intuitive handling of the model.

[0012] In various embodiments, the method further comprises the step of approving or accepting the rules, and in particular the interlocks detected by the rules. This may be done by a safety engineer and / or other personnel responsible for the process or plant. Advantageously, this may contribute to a higher level of confidence in the specification.

[0013] In various embodiments, the method further comprises the step of rejecting the rules, in particular the interlocks detected by the rules, which can advantageously be part of a fast and intuitive "debug process" of a process or plant specification.

[0014] In various embodiments, the method further comprises refining the rule, in particular the interlock, by at least one of adding further causes, adding further effects, adding further traversals, and / or adding further elements to the rule. The further elements may be for example comments, further outputs (for example acoustic signals, messages, etc.), and / or for example connections to a database or another interface. Formally speaking, the three parts of a rule may thus become an n-tuple (n>3). Advantageously, this may contribute to a more intuitive way of engineering the plant.

[0015] Approvals, rejections, and / or refinements may be tracked, for example, by a logging journal.

[0016] In various embodiments, the method further comprises generating a topology model of the industrial process from a piping and instrumentation diagram (P&ID) of the industrial process. To date, at least some safety engineers may manually input instrument references from I / O lists and / or P&IDs, often adding handwritten annotations. This manual construction may result in inconsistencies between different artifacts, e.g., misspelled instrument references resulting in inconsistent P&IDs. As a result of the above and / or below methods, plant topology information may be derived from the P&IDs, thus reducing error-prone "media-breaks."

[0017] In various embodiments, the topology model comprises a number of rules. This may bring the benefit of being able to deal with complexity. So far, P&IDs, C&E (cause and effect) matrices, and / or further means may be used, but these may be less intuitive than methods with visualization steps. The cognitive load of analyzing complex P&IDs with thousands of elements and incorporating various safety regulations and customer requirements may also lead to oversights, e.g. missing required interlocks. Although C&E matrices are generally thoroughly reviewed, errors or gaps may propagate to the automation system and cause harm to equipment or personnel. Furthermore, costly clarification and feedback loops between safety engineers and control engineers may be required. Rule-based engineering approaches using interlock specifications may offer a good compromise between a highly optimized plant and reusable plant knowledge in the form of interlock rules. However, they may suffer from a lack of intuitive user feedback on how the rules were applied and which subsets of the plant topology were included. This may be the reason why at least some safety engineers do not rely on these tools. Therefore, the methods described above and / or below may contribute to improving this situation.

[0018] In various embodiments, the method further comprises applying at least one step of the above and / or below methods to each rule of the plurality of rules. By applying these methods, a certain completeness of checking the plant behavior can be achieved. Thus, this semi-automatic method, and even a system configured to execute this method, can contribute to enabling a safety engineer to rapidly generate a C&E matrix that improves or guarantees correctness via visual feedback. Applying the method can comprise a process engineer of the process plant supplying P&I diagrams to a software tool, where a rule engine can process them to generate interlocks based on pre-specified domain-specific rules. The results can be displayed, for example, as a visual overlay on a P&I diagram or a topology model, including marking up causal equipment, traversal paths, and affected equipment. The process engineer can evaluate this visual representation and approve, reject, correct, and / or otherwise address the generated cause and effect relationships. The approval step can be advantageous due to the high heterogeneity of at least some industrial plants with multiple possibilities for assembling equipment. Furthermore, this can improve the confidence of the safety engineer in trusting the method.

[0019] In various embodiments, the method further comprises generating a Cause and Effect (C&E) matrix from the topology model of the industrial process. Advantageously, this can help to end the "media break" between several representations used by different engineers, since at least some people trust this representation more than others.

[0020] In various embodiments, the method further comprises generating a control logic according to IEC 61131-3 from the C&E matrix of the industrial process, which may advantageously make at least some parts or aspects of the interlock specifications directly applicable to the process or sub-process of the plant.

[0021] In various embodiments, the method further comprises generating a piping and instrumentation diagram (P&ID) from the topology model of the industrial process, which may contribute to better understanding and / or validating the interlock specifications and / or the topology model.

[0022] In various embodiments, the method further comprises checking the associated causes for overlapping ranges and / or gaps between the ranges when at least two rules relate to the same sensor. A "cause" may comprise a range of values ​​from at least one sensor, which are not necessarily consistent, e.g. due to typing and / or construction errors. This may contribute to reducing such consistency errors.

[0023] In various embodiments, the method further comprises checking the related results for intersections when at least two rules relate to the same actuator. For example, if two power sources drive an electric motor, the method can check whether their cooperation is non-destructive. This check can be done on a rule basis.

[0024] One aspect relates to a non-transitory computer readable storage medium having stored thereon a program which, when executed on a processor, instructs the processor to perform any of the methods described above.

[0025] One embodiment relates to an artificial neural network (ANN) configured to be trained by and / or to implement a method for checking whether at least two rules relate to the same sensor and / or whether at least two rules relate to the same actuator and evaluating the result. To that end, the ANN may be trained, for example, with inconsistent and / or destructive cooperative behavior of the actuators and / or their controls. Additionally or alternatively, a machine learning model may be trained based on an engineer's decision regarding the rule application.

[0026] The system adds the approved cause and effect relationships to a matrix, which can later be fed to a control logic generation tool. In parallel, the system uses the safety engineer's decisions to train a machine learning mechanism (e.g., a neural network). This may refine the rule base and improve the quality of interlock generation in later runs. This solution may lower the chance of errors or oversights, since device or equipment tag names may be derived directly from the original P&I diagram without manual intervention. Furthermore, the rule engine may be enabled to handle complex topologies, thus preventing cognitive overload problems. Since existing domain knowledge is encoded in the rules, it may be reused with limited effort. Furthermore, the safety engineer has full control over the process and may manually add interlocks to the generated C&E matrix, including issuing peer reviews.

[0027] Additionally or alternatively, the system may also enable engineers to visually specify new interlocking rules by marking up equipment, drawing paths, and assigning them causal, traversal, and / or effect properties.

[0028] An embodiment relates to the use of the above and / or below described method for checking safety and / or specifications of an industrial process.

[0029] For the sake of clarity, the present invention will now be described with reference to the embodiments shown in the drawings, which are merely examples and should not be considered as limiting. [Brief description of the drawings]

[0030] [Figure 1] 1 is a diagram of method steps according to one embodiment; [Diagram 2] 4 is a diagram of additional or alternative steps of a method according to one embodiment; [Diagram 3]4 is a diagram of additional or alternative steps of a method according to one embodiment; [Figure 4] 4 is a diagram of additional or alternative steps of a method according to one embodiment; [Diagram 5] FIG. 2 is an example visualization according to one embodiment. [Figure 6] FIG. 13 is a diagram of a further example visualization according to one embodiment. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0031] FIG. 1 illustrates steps of a method according to an embodiment. It is not necessary that all steps illustrated here are performed in every embodiment. FIG. 1 illustrates components, e.g., software components, as rectangles and artifacts, e.g., tables and / or other representations that may possibly be stored in memory, as rectangles with beveled corners. FIG. 1 illustrates a process engineer feeding a P&I diagram to a P&ID importer that extracts a topology model from the P&I diagram in step 1. The topology model normalizes this diagram for different types of shapes and naming conventions, and may also strip information that is not related to the interlock specification, e.g., manual valve elements, specific equipment properties, or versioning annotations. The rules engine processes the topology model in step 2 by identifying equipment that may be relevant to the interlock generation. It applies rules, e.g., in the form of pre-specified if-then rules, from a rule base in step 3 by checking whether an if conditional occurs in the topology model and whether a path can be constructed from a "cause" equipment to a "result" equipment. For example, if a high temperature alarm is specified on a tank (the "cause"), a "traversal" on the topology model from the tank to the heat exchanger (the "effect") may be attempted and, if successful, a corresponding signal to shut down the heat exchanger during runtime may be linked to this cause. Such rules may be coded, for example, by a domain expert or reverse engineered from a previous manual specification.

[0032] The rules engine then passes the identified interlock elements to an interlock topology visualizer in step 4. This component creates a custom view on the topology model that focuses on elements relevant to the identified interlocks and a limited number of context elements in step 5. It then visually marks up the causes, traversal paths, and consequences, for example by using transparent, possibly colored overlays. The safety engineer responsible for the interlock specification has this visualization displayed, for example on a screen, in step 6. The safety engineer can then approve, reject, etc. the identified interlocks via a user interface in step 7, which feeds that information back to the rules engine in step 8. Corrections can be implemented by manipulating the cause and effect markers of the visualization and adjusting the path topology.

[0033] The rule engine may train a machine learning model, e.g., an artificial neural network (ANN), that supports the interlock generation algorithm. Based on, e.g., context information, safety engineer's decisions, and history of applying certain rules, the rule engine may refine the rules or adjust the specific mechanisms that propose rules for a given topology model in step 9. Steps 3 to 9 may be applied iteratively until all rules are applied to the entire topology model. The rule engine may then generate a C&E matrix in step 10 that may be expanded, possibly manually, and later reviewed, e.g., by other engineers. In step 11, the C&E matrix may be imported into a control logic generation tool, e.g., according to IEC 61131-3.

[0034] As a variant, the safety engineer may use the interlock topology visualization to visually specify new interlock rules, for example by drawing an overlay on the topology model. Another variant, especially in brownfield scenarios, may involve reverse engineering the existing interlock control logic in the automation system into a C&E matrix and comparing it to the matrix generated by the system. This may help to keep the specification consistent and may contribute to detecting undesired violations.

[0035] Thus, a visual overlay or another type of visually generated C&E is provided on top of the customized and streamlined plant topology model cutout. This provides a mechanism to allow safety engineers and / or other personnel running the process to interact with the rules engine. Furthermore, machine learning models can be trained with human input in the context of the interlock specification. For example, the rules engine may incorporate engineer decisions as "learning", context information can be recorded for the interlock / C&E situation, and the ML algorithms can be recalibrated, for example by adjusting weights in the ANN accordingly. Furthermore, reverse engineering of IEC61131 control logic into a C&E matrix and comparing the original C&E matrix against it can be performed. According to the above and / or below methods, tools are available that allow visual specification of new interlock rules by marking up equipment, drawing paths, and assigning cause / traversal / effect properties to them. Furthermore, a suggested automatic path by calculating the shortest path between the specified C&E can be added. The methods presented may result in faster engineering of interlocks, reuse of existing knowledge encoded in the rules, may be advantageously less error-prone than, for example, manual and / or "media-breaking" approaches, may contribute to avoiding human oversight, and / or may contribute to avoiding artifact inconsistencies.

[0036] 2 shows additional or alternative steps of a method according to one embodiment. In step 1, visual feedback is provided to the engineer by the visualization tool as described above and / or below. In step 2, the engineer accepts, rejects, etc. the identified interlocks via a user interface. This may be done iteratively. In step 3, a C&E matrix may be generated from the verified interlock topology. In step 4, the C&E matrix may be imported into a control logic generation tool, for example according to IEC 61131-3.

[0037] 3 shows additional or alternative steps of a method according to an embodiment, whereby checking the control logic against customer specifications for interlocks by using a C&E matrix can be implemented. In step 1, customer input is given, for example in the form of a C&E matrix #1. In step 2, a control logic is generated, for example compliant with IEC 61131-3. In step 3, automatic reverse engineering can be performed by generating a C&E matrix #2, which can be compared to the C&E matrix #1 for consistency check. In step 4, feedback is given to an engineer, who can then approve, reject, etc. the identified interlocks via a user interface.

[0038] Figure 4 shows additional or alternative steps of a method according to one embodiment. In step 1, general rules are specified by an engineer, e.g. a domain expert, process engineer or knowledge engineer, using P&ID shapes. This is the basis for checking the interlock specifications as described above and / or below. In step 2, an automatic translation of the rules into a formal language is performed. One syntax used looks like this: Valve.Closed -> Valve / Pipe / Reactor-1 -> Reactor-1\Pipe\Valve-2 => Valve-2.Close In the example shown, closing a valve on a pipe that feeds a reactor closes another valve on another pipe that also feeds the reactor. The symbol " / " indicates traversal of the topology model in the direction of material or logical flow, while the symbol "\" indicates traversal of the topology model in the opposite direction of material or logical flow. This example interlocks all valves that feed a reactor, closing one of them closes them all. In step 3, an automatic check is performed and feedback is given to the engineer. In step 4, based on the feedback, existing rules are matched or new rules are created.

[0039] Figure 5 shows an example visualization according to one embodiment. In this example, sensor B 104 emits a "high temperature alarm" as "cause" C. Through a "traversal" T-path "information flow connection" P-4, this signal is forwarded to an "effect" E, i.e. to an actuator "Heat Exchanger E 104", which is stopped by a rule. One syntax used looks like this: Cause: High temperature alarm B104 Traversal: Information flow connections Result: Heat exchanger E104 is stopped In this tool, the "Cause" C, the "Traversal" T or traversal path, and the "Result" E may be highlighted. Additionally, a window displaying this rule may be shown and a checkbox "Approve / Reject" may be displayed. The engineer may then approve or reject the highlighted rule.

[0040] Figure 6 shows a further visualization example according to one embodiment. This example shows two actuators that are affected by a rule. The rule looks like this: Cause: High level alarm B114 Traversal: All entrances to connected vessels Result: Close valves V102 and V104 Thus, sensor "high level alarm B114" as "cause" C closes valves V102, V104 as "result" E. The invention as originally claimed in the present application is set forth below. [1] A method for visualizing the rules of an industrial process or treatment system, comprising: providing a topology model of the industrial process, wherein the industrial process comprises at least one sensor and at least one actuator; associating a rule (R) with the topology model, wherein the rule (R) comprises three parts: Cause (C), Traversal (T), Effect (E); Where: the cause (C) comprises a range of values ​​from the at least one sensor; the result (E) comprises an action performed by the at least one actuator; The traversal (T) comprises a relationship between the cause (C) and the effect (E); marking said cause (C), said traversal (T), and / or said effect (E); and visualizing elements of the rules (R) in the topology model. [2] The method according to [1], wherein associating rules (R) with the topology model is based on visualization means. [3] The method of [1], further comprising the step of accepting the rule (R) or rejecting the rule (R). [4] In the above rule (R) Adding further causes (C), Adding further results (E); Adding further traversals (T), and / or Adding more elements, The method of claim 1, further comprising the step of refining the rules (R) by implementing at least one of the following: [5] The method of [1], further comprising a step of generating the topology model of the industrial process from a piping and instrumentation diagram (P&ID) of the industrial process. [6] The method according to any one of [1] to [5], wherein the topology model comprises a plurality of rules (R). [7] The method described in [6], further comprising a step of applying at least one step described in any one of [1] to [5] to each rule (R) of the plurality of rules (R). [8] The method described in [7], further comprising the step of generating a cause and effect (C&E) matrix from the topology model of the industrial process. [9] The method described in [8], further comprising a step of generating a control logic according to IEC 61131-3 from the C&E matrix of the industrial process.

[10] The method of [6], further comprising the step of generating a piping and instrumentation diagram (P&ID) from the topology model of the industrial process.

[11] The method according to [6], further comprising the step of checking the associated causes (C) for overlapping ranges and / or gaps between the ranges if at least two rules (R) relate to the same sensor.

[12] The method according to [6], further comprising the step of checking the associated results (E) for a common portion if at least two rules (R) relate to the same actuator.

[13] A non-transitory computer-readable storage medium having a program stored thereon, the program instructing the processor to perform a method according to any one of [1] to

[12] when executed on the processor.

[14] An artificial neural network (ANN) configured to be trained by and / or to implement the methods described in

[11] or

[12] .

[15] Use of the method according to any one of [1] to

[12] for checking the safety and / or specifications of an industrial process.

Claims

1. 1. A method for visualizing rules of an industrial process or treatment system, comprising: providing a topology model of the industrial process, the industrial process comprising at least one sensor and at least one actuator; associating a rule (R) with the topology model, the rule (R) comprising three parts: Cause (C), Traversal (T), Effect (E); Where: the cause (C) comprises a range of values ​​from the at least one sensor; the result (E) comprises an action performed by the at least one actuator; The traversal (T) comprises a relationship between the cause (C) and the effect (E). marking said cause (C), said traversal (T), and / or said effect (E); and visualizing elements of the rule (R) in the topology model, the topology model comprising a plurality of rules (R), If at least two rules (R) relate to the same sensor, the method further comprises the step of checking the associated causes (C) for overlapping ranges and / or gaps between their ranges.

2. 1. A method for visualizing rules of an industrial process or treatment system, comprising: providing a topology model of the industrial process, the industrial process comprising at least one sensor and at least one actuator; associating a rule (R) with the topology model, the rule (R) comprising three parts: Cause (C), Traversal (T), Effect (E); Where: the cause (C) comprises a range of values ​​from the at least one sensor; the result (E) comprises an action performed by the at least one actuator; The traversal (T) comprises a relationship between the cause (C) and the effect (E). marking said cause (C), said traversal (T), and / or said effect (E); and visualizing elements of the rule (R) in the topology model, the topology model comprising a plurality of rules (R), If at least two rules (R) relate to the same actuator, the method further comprises the step of checking the related results (E) for an intersection.

3. The method according to claim 1 or 2, wherein the associating of rules (R) with the topology model is based on visualization means.

4. The method according to claim 1 or 2, further comprising the step of approving said rule (R) or rejecting said rule (R).

5. In the rule (R), Adding a further cause (C), adding a further result (E); Adding a further traversal (T), and / or Adding more elements, The method according to claim 1 or 2, further comprising the step of refining said rules (R) by implementing at least one of the following:

6. The method of claim 1 or 2, further comprising generating the topology model of the industrial process from a piping and instrumentation diagram (P&ID) of the industrial process.

7. The method further comprising the step of applying at least one step according to any one of claims 1 to 6 to each rule (R) of said plurality of rules (R).

8. The method of claim 7 , further comprising generating a cause and effect (C&E) matrix from the topology model of the industrial process.

9. The method of any one of claims 1 to 6, further comprising generating a piping and instrumentation diagram (P&ID) from the topology model of the industrial process.

10. A non-transitory computer readable storage medium having a program stored thereon, the program, when executed on a processor, instructing the processor to perform a method according to any one of claims 1 to 9.

11. Use of the method according to any one of claims 1 to 9 for checking the safety and / or specifications of an industrial process.

Images