Event data's nuclear power interface procedure coupling risk analysis method and system

Abstract

The application provides a method and system for analyzing the coupling risk of nuclear power interface procedures based on event data. The method comprises: obtaining nuclear power plant operation event data, and constructing a structured event data set through cleaning, standardization and semantic unification processing; performing multi-dimensional labeling on each event record based on a preset human factor event labeling rule to obtain multi-dimensional labeling results; performing mechanism decomposition on the multi-dimensional labeling results to obtain interface factor decomposition results; and determining the change degree of procedure deviation risk under the intervention condition of interface factors based on the multi-dimensional labeling results and the interface factor decomposition results. The application can identify the coupling failure mode between the interface and the procedure based on the real event data system, quantitatively evaluate the influence of the coupling failure mode on the reliability of the procedure execution, and provide objective and traceable technical basis for human-machine interface design, operation procedure improvement and human factor risk control.

Description

Technical Field

[0001] This application relates to the field of nuclear power human factors engineering and safety analysis technology, and in particular to a method and system for risk analysis of nuclear power interface procedures coupled with event data. Background Technology

[0002] With the continuous development of digitalization and integration technologies in instrumentation and control systems and the main control room of nuclear power plants, the traditional human-machine interaction methods, mainly based on analog instruments and hardware switches, are gradually being replaced by digital human-machine interfaces, soft control operations, and computerized procedural systems. Digital main control rooms, through multi-screen display architecture, dynamic interface switching logic, and highly integrated information presentation mechanisms, significantly improve the breadth of information access and operational flexibility for operators. However, this also brings a significant increase in the semantic complexity of the interface and cognitive load. In this technological context, when operators perform procedural tasks, their cognitive decision-making and operational execution processes are highly dependent on the accuracy, consistency, and identifiability of the information presented by the digital interface. The intrinsic relationship between interface design and procedural logic becomes a key factor affecting human-factor reliability.

[0003] In related technologies, various systematic assessment methods have been developed for the analysis of human factor reliability, such as human error rate prediction technology, standardized power plant risk human factor reliability analysis, human factor event analysis technology, and the latest analysis framework based on cognitive process models. These methods mainly model and quantify the human factor failure process from dimensions such as the logical correctness of operating procedures, the cognitive stage transition of operators, and external performance influencing factors.

[0004] Feedback from actual nuclear power plant operation experience and incident investigations have revealed that failures, even when operators strictly adhere to established procedures, are not solely due to violations of procedural clauses or accidental operational errors. A significant portion of these incidents stem from discrepancies or inconsistencies between the digital interface's display semantics, layout, visual coding methods, or naming conventions and the implicit expectations or assumptions about the operational objects implied in the operating procedures. This leads operators to develop incorrect situational perceptions or pattern confusion regarding the current system state, ultimately resulting in erroneous operational actions or deviations from the procedural execution path.

[0005] In this regard, the relevant technologies are insufficient to provide direct and objective decision-making basis and data support for the precise optimization design of digital human-machine interfaces and the targeted revision of operating procedures when dealing with such problems. Summary of the Invention

[0006] This application provides a method and system for risk analysis of nuclear power interface procedures coupled with event data, providing decision-making basis and data support for digital human-machine interfaces.

[0007] Firstly, this application provides a method for nuclear power interface procedure coupling risk analysis based on event data, including: Acquire event data during nuclear power plant operation events, clean, standardize, and semantically unify the event data to construct a structured event dataset for human factors analysis; Based on preset human factor event annotation rules, each event in the structured event dataset is annotated in multiple dimensions to obtain multidimensional annotation results; Mechanistic decomposition of the multidimensional annotation results yields interface factor decomposition results; Based on the multidimensional annotation results and the interface factor decomposition results, the risk of deviation from the procedure is determined.

[0008] Optionally, the event data includes multiple events, each of which includes at least event description information, operation procedure execution information, human-machine interface feature information, and event result information, wherein the human-machine interface feature information includes interface layout information, interface display semantic information, and interface identification information.

[0009] Optionally, the step of performing multidimensional annotation on each event in the structured event dataset based on preset human factor event annotation rules to obtain multidimensional annotation results includes: Based on preset human factor event annotation rules, each event in the structured event dataset is labeled with procedure deviation to characterize whether there are any omissions, errors, or improper execution of procedures in the event, thus obtaining the procedure deviation annotation results. Each event is labeled with an interface problem to characterize whether there are human-computer interface related defects in the event, and the interface problem labeling results are obtained. Each event is labeled with interface and procedure coupling to characterize whether the interface problem in the event constitutes a direct cause of inducing or amplifying the deviation of the procedure, and the interface and procedure coupling labeling results are obtained. The deviation annotation results, interface problem annotation results, and interface-procedure coupling annotation results are determined as multi-dimensional annotation results.

[0010] Optionally, the step of performing mechanistic decomposition on the multidimensional annotation results to obtain the interface factor decomposition results includes: The events with human-computer interface-related defects in the interface problem annotation results are analyzed to obtain the interface factors that cause the human-computer interface-related defects, and the interface factors that cause the human-computer interface-related defects and their corresponding values ​​are determined as the interface factor decomposition results.

[0011] Optionally, the interface factors include: layout and operation area factors, semantic and status indication factors, interface and procedure semantic mismatch factors, and identification and naming factors.

[0012] Optionally, determining the procedure deviation risk based on the multidimensional annotation results and the interface factor decomposition results includes: Using the deviation from the standard procedure annotation results, the deviation rate of the benchmark procedure is calculated; Weights are assigned to each interface factor using the interface problem annotation results and the interface and procedure coupling annotation results; The risk of deviation from the procedure is calculated by using the weights assigned to each interface factor and the deviation rate of the benchmark procedure.

[0013] Optionally, calculating the procedure deviation risk using the weights assigned to each interface factor and the baseline procedure deviation rate includes: The coupling amplification factor is calculated using the deviation rate of the aforementioned reference procedure; The interface risk factor is calculated by using the weights and corresponding values ​​of each interface factor. The deviation risk of the procedure is calculated using the coupling amplification factor and the interface risk factor.

[0014] Optionally, after determining the procedure deviation risk based on the multidimensional annotation results and the interface factor decomposition results, the nuclear power interface procedure coupling risk analysis method for event data further includes: For each event, the event is determined to be a high-risk coupling event based on the deviation risk of the procedure. If it is, interface optimization suggestions and procedure revision suggestions are generated for the high-risk coupling event.

[0015] Optionally, the interface optimization suggestions include at least display semantic optimization suggestions and operation area layout optimization suggestions; the procedure revision suggestions include at least procedure text and interface identifier consistency revision suggestions.

[0016] Secondly, this application provides a nuclear power interface procedure coupling risk analysis system for event data, including: The data acquisition and preprocessing module is used to acquire event data during nuclear power plant operation events, clean, standardize, and semantically unify the event data, and construct a structured event dataset for human factors analysis; the structured event dataset contains multiple events. The multidimensional annotation module is used to perform multidimensional annotation on each event in the structured event dataset based on preset human factor event annotation rules, and obtain multidimensional annotation results; The mechanism decomposition module is used to perform mechanism decomposition on the multidimensional annotation results to obtain the interface factor decomposition results. The risk quantification calculation module is used to determine the deviation risk of the procedure based on the multidimensional annotation results and the interface factor decomposition results.

[0017] This application provides a method and system for nuclear power plant interface-procedure coupling risk analysis based on event data. The method includes: acquiring nuclear power plant operation event data; constructing a structured event dataset through cleaning, standardization, and semantic unification processing; performing multi-dimensional annotation on each event record based on preset human factor event annotation rules to obtain multi-dimensional annotation results; performing mechanistic decomposition on the multi-dimensional annotation results to obtain interface factor decomposition results; and calculating the degree of change in procedure deviation risk under interface factor intervention conditions based on the multi-dimensional annotation results and interface factor decomposition results. This application can systematically identify the coupling failure modes between the interface and procedures based on real event data, quantitatively assess their impact on the reliability of procedure execution, and provide objective and traceable technical basis for human-machine interface design optimization, operating procedure improvement, and human factor risk management. Attached Figure Description

[0018] Figure 1 The diagram shown is a flowchart illustrating a nuclear power plant interface procedure coupling risk analysis method for event data provided in an embodiment of this application. Figure 2 The diagram shown is a flowchart of multidimensional annotation provided in an embodiment of this application; Figure 3 The diagram shown is a schematic representation of the overall process for determining the risk of deviation from the procedure according to an embodiment of this application. Figure 4 The diagram shown is a flowchart illustrating the deviation risk of the calculation procedure provided in an embodiment of this application. Figure 5 The diagram shown is a structural diagram of the nuclear power interface procedure coupling risk analysis system for event data provided in an embodiment of this application. Detailed Implementation

[0019] Exemplary embodiments will now be described in detail, examples of which are illustrated in the accompanying drawings. When the following description relates to the drawings, unless otherwise indicated, the same numerals in different drawings denote the same or similar elements. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with one or more embodiments of this specification. Rather, they are merely examples of systems and methods consistent with some aspects of one or more embodiments of this specification as detailed in the appended claims.

[0020] It should be noted that the steps of the corresponding methods are not necessarily performed in the order shown and described in this specification in other embodiments. In some other embodiments, the methods may include more or fewer steps than described in this specification. Furthermore, a single step described in this specification may be broken down into multiple steps in other embodiments; and multiple steps described in this specification may be combined into a single step in other embodiments.

[0021] In related technologies, various systematic assessment methods have been developed for the analysis of human factor reliability, such as human error rate prediction technology, standardized power plant risk human factor reliability analysis, human factor event analysis technology, and the latest analysis framework based on cognitive process models. These methods mainly model and quantify the human factor failure process from dimensions such as the logical correctness of operating procedures, the cognitive stage transition of operators, and external performance influencing factors. However, under existing analysis frameworks, the quality and effectiveness of operating procedures and the design quality of human-machine interfaces are usually evaluated and treated separately as two independent influencing factors. Such analysis frameworks rarely systematically characterize and evaluate the dynamic coupling relationship between the display semantic features, visual layout features, interaction logic, and implicit state assumptions and operational expectations of digital interfaces, and their impact mechanisms on human factor failure.

[0022] Feedback from actual nuclear power plant operation experience and incident investigations have revealed that failures, even when operators strictly adhere to established procedures, are not solely due to violations of procedure clauses or accidental operational errors. A significant number of incidents stem from discrepancies or inconsistencies between the digital interface's display semantics, layout, visual coding methods, or naming conventions and the implicit expected states or operational assumptions of the operating procedures. This leads operators to develop incorrect situational perceptions or pattern confusion regarding the current system state, ultimately resulting in incorrect operational actions or deviations from the procedural execution path. Specific manifestations of these problems include: inconsistencies between the implicit meanings of interface indicators and the textual descriptions in the procedures; mis-touching or mis-selection of controls with similar functions or closely spaced positions on the interface without clear visual distinguishing features; insufficient or visually inadequate presentation of status feedback information after critical operations, causing operators to incorrectly confirm operation results; and unclear labeling or inconsistent naming rules for similar equipment or components on the interface, leading to operator confusion regarding the operational objects. The above phenomenon indicates a human-caused failure mechanism coupled with the interface, where the risk level is not determined by a single interface defect or procedure flaw, but rather by the amplifying effect of interface design defects and cognitive expectations during procedure execution. However, related technologies generally suffer from the following shortcomings when dealing with such problems: There is a lack of a data-driven analysis method based on real-world operational events to systematically identify and classify the coupling risks between interfaces and procedures; there is a lack of fine-grained mechanisms for decomposing the coupling failures of interface factors, making it difficult to effectively distinguish the specific impact paths and contributions of different dimensions of interface features, such as layout design, semantic expression, and labeling specifications, on procedure deviation behavior; and there is a lack of quantifiable models that can quantitatively assess the degree of change in procedure execution deviation risks under the intervention of interface factors. Therefore, it is difficult to provide direct and objective decision-making basis and data support for the precise optimization design of digital human-machine interfaces and the targeted revision of operating procedures.

[0023] To address the technical challenge of providing direct and objective decision-making basis and data support for the precise optimization design of digital human-machine interfaces and the targeted revision of operating procedures in the aforementioned related technologies, this application aims to provide a method and system for nuclear power plant interface-procedure coupling risk analysis based on event data. By performing structured modeling of nuclear power plant operation events or operational process data, the system identifies coupled human-cause failure mechanisms between interface factors and operating procedures, quantitatively assesses the impact of interface and procedure coupling on human-cause reliability, thereby optimizing the nuclear power plant interface design in the digital control room and providing a more intuitive display of operating procedures. Furthermore, it provides objective and traceable technical evidence for the optimization of human-machine interface design, improvement of operating procedures, and human-cause risk management in the digital control room.

[0024] Combination Figures 1 to 5 , Figure 1 The diagram shown is a flowchart illustrating a nuclear power plant interface procedure coupling risk analysis method for event data provided in an embodiment of this application. Figure 2 The diagram shown is a flowchart of multidimensional annotation provided in an embodiment of this application; Figure 3 The diagram shown is a schematic representation of the overall process for determining the risk of deviation from the procedure according to an embodiment of this application. Figure 4 The diagram shown is a flowchart illustrating the deviation risk of the calculation procedure provided in an embodiment of this application. Figure 5 The diagram shown is a structural diagram of the nuclear power interface procedure coupling risk analysis system for event data provided in an embodiment of this application.

[0025] like Figure 1 As shown, this application provides a method for nuclear power interface procedure coupling risk analysis based on event data, including: S100: Acquire event data during nuclear power plant operation events, clean, standardize, and semantically unify the event data to construct a structured event dataset for human factors analysis; the event data includes multiple events, each event including at least event description information, operation procedure execution information, human-machine interface feature information, and event result information, wherein the human-machine interface feature information includes interface layout information, interface display semantic information, and interface identification information.

[0026] This application can acquire event data or operation log data from the operation of the digital nuclear power plant main control room. This data records multiple events and their detailed information, including event descriptions, operating procedure execution information, human-machine interface (HMI) feature information, and event outcome information. The HMI feature information includes interface layout information, interface display semantic information, and interface identification information. This application cleans, standardizes, and semantically unifies the event data to form a structured event dataset that can be used for human factors analysis.

[0027] S200, based on preset human factor event annotation rules, perform multidimensional annotation on each event in the structured event dataset to obtain multidimensional annotation results; In one specific embodiment of this application, such as Figure 2 As shown, S200 includes: S210, Based on the preset human factor event annotation rules, perform procedure deviation annotation on each event in the structured event dataset to characterize whether there are any procedures omissions, procedures errors, or improper procedures in the event, and obtain the procedure deviation annotation results; S220, each event is labeled with an interface problem to characterize whether there are human-computer interface related defects in the event, and the interface problem labeling results are obtained; S230, Perform interface and procedure coupling annotation on each event to characterize whether the interface problem in the event constitutes a direct cause of inducing or amplifying the deviation of the procedure, and obtain the interface and procedure coupling annotation result. S240, the deviation annotation results of the procedure, the interface problem annotation results, and the interface and procedure coupling annotation results are determined as multi-dimensional annotation results.

[0028] S300, perform mechanism decomposition on the multidimensional annotation results to obtain interface factor decomposition results; This step analyzes the events with human-computer interface (HCI) related defects in the interface problem annotation results, identifies the interface factors causing these HCI-related defects, and determines the interface factors causing these HCI-related defects and their corresponding values ​​as the interface factor decomposition results. The interface factors include: The factors considered include layout and operation area factors, semantic and status indication factors, interface and procedure semantic mismatch factors, and identification and naming factors. Specifically, the layout and operation area factors characterize whether the spatial arrangement, positional distribution, and operation area planning of controls in the digital human-machine interface have design flaws that could induce operational errors. The semantic and status indication factors characterize whether the presentation method, symbol meaning, and visual encoding of system status information in the digital human-machine interface are clear, intuitive, and accurately understandable by the operator. The interface and procedure semantic mismatch factors characterize whether there are inconsistencies or contradictions between the information content, expression method, or status definition presented by the digital human-machine interface and the descriptive assumptions in the operation procedure text. The identification and naming factors characterize whether the identification names and label text of devices, controls, or functional areas in the digital human-machine interface are missing, unclear, inconsistently named, or easily confused, indicating design flaws.

[0029] S400, based on the multidimensional annotation results and the interface factor decomposition results, determine the procedure deviation risk.

[0030] In one specific embodiment of this application, such as Figure 3 As shown, S400 includes: S410, using the deviation annotation results of the procedure, calculate the deviation rate of the benchmark procedure, expressed as: , This indicates the number of times the procedure was marked as a deviation. This indicates the number of times that a human-computer interface-related defect was identified.

[0031] S420, using the interface problem annotation results and the interface-procedure coupling annotation results, assign weights to each interface factor, expressed as follows: In the formula, Indicates the first Frequency of each interface factor Indicates the first The co-frequency of deviations between interface factors and procedures This indicates the total number of interface elements.

[0032] S430, using the weights assigned to each interface factor and the baseline procedure deviation rate, calculate the procedure deviation risk.

[0033] In one specific embodiment of this application, such as Figure 4 As shown, S430 includes: S431, the coupling amplification factor is calculated using the deviation rate of the aforementioned reference procedure, and is expressed as follows: ; S432, using the weights and corresponding values ​​of each interface factor, calculate the interface risk factor, expressed as: ; In the formula, Indicates assignment to the first The weight of each interface factor, Indicates the first The values ​​of each interface factor.

[0034] S433, using the coupling amplification factor and the interface risk factor, the procedure deviation risk is calculated as follows: .

[0035] In one specific embodiment of this application, after S400, the nuclear power plant interface procedure coupling risk analysis method for event data provided in this application further includes: S500: For each event, determine whether the event is a high-risk coupling event based on the deviation risk of the procedure. If so, generate interface optimization suggestions and procedure revision suggestions for the high-risk coupling event. The interface optimization suggestions include at least display semantic optimization suggestions and operation area layout optimization suggestions. The procedure revision suggestions include at least procedure text and interface identifier consistency revision suggestions.

[0036] like Figure 5 As shown, this application also provides a nuclear power plant interface procedure coupling risk analysis system for event data, including: The data acquisition and preprocessing module 10 is used to acquire event data during nuclear power plant operation events, clean, standardize, and semantically unify the event data, and construct a structured event dataset for human factors analysis; the structured event dataset contains multiple events. The multidimensional annotation module 20 is used to perform multidimensional annotation on each event in the structured event dataset based on preset human factor event annotation rules, and obtain multidimensional annotation results. Mechanism decomposition module 30 is used to perform mechanism decomposition on the multidimensional annotation results to obtain interface factor decomposition results; The risk quantification calculation module 40 is used to determine the deviation risk of the procedure based on the multidimensional annotation results and the interface factor decomposition results.

[0037] This application proposes a method and system for risk analysis of nuclear power plant interface-procedure coupling based on event data. First, event data generated during nuclear power plant operation is acquired. This data, including event descriptions, procedure execution, human-machine interface characteristics, and event results, is cleaned, standardized, and semantically unified to construct a structured event dataset suitable for human factors analysis. Based on this, events are annotated in multiple dimensions according to pre-defined human factors event annotation rules, including procedure deviation, interface problems, and interface-procedure coupling, to characterize the correlation between interface problems and procedure execution deviations. Subsequently, events with interface problems are further analyzed mechanistically, decomposing them into interface factors such as layout and operating area design, semantic and state expression, interface-procedure semantic mismatch, and identification and naming design, and establishing a correspondence between these factors and procedure execution results. Then, based on the multi-dimensional annotation and interface factor decomposition results, the procedure deviation risk under interface factor intervention conditions is calculated to quantify the impact of interface-procedure coupling on human factors reliability. Finally, high-risk interface-procedure coupling patterns are identified, generating corresponding risk assessment results and improvement suggestions to support human-machine interface optimization design, procedure revision, and operation management decisions.

[0038] To illustrate the specific implementation process of this application, the following description is provided by way of example: Example 1 In a primary equipment isolation operation in the main control room of a digital nuclear power plant (corresponding to the acquired Class B operational event data), operators must confirm that a certain isolation valve is in the closed state according to the operating procedures before proceeding to the next step. The valve's status is displayed on the digital human-machine interface using the symbol 'O', while the operating procedure text describes the valve's status in text form ('open / closed'), without explicitly explaining the meaning of the interface symbols.

[0039] The operator mistakenly interpreted the symbol 'O' as 'closed', while in the actual interface logic 'O' represents 'open', causing subsequent operations to continue even though the valve was not fully closed, resulting in a procedure deviation event.

[0040] S100: Acquire event data during nuclear power plant operation events, clean, standardize, and semantically unify the event data, and construct a structured event dataset for human factors analysis; The structured information of the event is extracted from the formal operation event report, including the event process description, procedure execution record, interface status characteristics and corrective measures. These are uniformly encoded into structured entries that can be used for human factors analysis, forming a structured event dataset.

[0041] S200: Based on preset human factor event annotation rules, multidimensional annotation is performed on each event in the structured event dataset to obtain the multidimensional annotation results, as shown in Table 1: Table 1. Multidimensional Annotation Results S300, Perform mechanistic decomposition on the multidimensional annotation results to obtain the interface factor decomposition results: A secondary mechanism decomposition was performed on the event with UI=1. Based on the event description, root cause analysis, and corrective measures (the corrective measures pointed to 'changing the symbol to a clear text label'), triangulation was conducted to identify the interface factors as shown in Table 2: Table 2 Decomposition Results of Interface Factors S400, based on the multidimensional annotation results and the interface factor decomposition results, determine the procedure deviation risk.

[0042] (1) Deviation rate of benchmark procedure Calculation: This represents the base probability of an event being labeled as a procedure deviation when there are no UI issues; it is directly derived from the multidimensional annotation statistics. Of the 59 events in the full sample, 37 were events with UI=0, and 27 of these were labeled with PROC=1. Therefore: ; (2) Weight of interface factors Derivation: The risk weights of each interface factor are determined based on the frequency statistics of the 22 UI=1 events in S300 and their co-occurrence rates with deviations from the procedure. The frequency-co-occurrence rate product normalization method is used, and the weights are expressed as follows: ; In the formula, Indicates the first Frequency of each interface factor Indicates the first The co-frequency of deviations between interface factors and procedures This indicates the total number of interface elements.

[0043] See Table 3 for the weight allocation of interface factors: Table 3 Weighting of Interface Factors Normalized results of the weights of each factor: .

[0044] (3) Coupling amplification factor Calculation: Coupling amplification factor The overall amplification factor reflecting the risk of procedural deviation after the intervention of interface factors is derived from the results of binary logistic regression analysis of the entire sample. The proportion of procedural deviation in the UI=1 event was 86.36% (19 / 22), and in the UI=0 event it was 72.97% (27 / 37). Based on this, the amplification magnitude is estimated as follows: After converting to linear amplification parameters, and considering the range constraint of the logistic regression odds ratio (OR=2.35, 95%CI: 0.57–9.68), this example takes... (The value is taken in the middle, which is reasonable). The specific value can be determined based on historical event statistics, expert rules, or data-driven models.

[0045] (4) Calculation of interface risk factors: Substitute the values ​​of the interface factors for this event ( and normalized weights: (5) Calculation of deviation risk from procedures: The calculation results show that, due to semantic factors ( ) and Interface-Procedure Mismatch Factors ( When both are activated, the risk of deviation from the standard increases from 0.730 to 0.597, meaning that interface defects amplify the risk by approximately 18.2%.

[0046] S500: For each event, determine whether the event is a high-risk coupling event based on the deviation risk of the procedure. If so, generate interface optimization suggestions and procedure revision suggestions for the high-risk coupling event. If the value exceeds a preset high-risk threshold (e.g., 0.55), the event is identified as a high-risk interface-procedure coupling event, and the following suggestions are output: 1. Interface rectification: Change the valve status indicator symbol 'O' to the explicit text 'OPEN / CLOSED' (this has been verified in the corrective actions); 2. Procedure Improvement: Add explanations of the interface display logic to the procedure steps involving valve status confirmation; 3. HRA parameter recommendations: , The combined identification is a 'semantic coupling trap' composite PSF, corresponding to the cognitive failure mode of comprehension error.

[0047] Example 2 In another actual operational event (a type B event collected), the operator needed to use a soft button to switch systems according to the procedure. This button was closely arranged with adjacent function buttons, and its color and shape were highly similar. During the operation, the operator accidentally pressed an adjacent button, causing the system to enter an unexpected state and deviating from the procedure.

[0048] S100: Acquire event data during nuclear power plant operation events, clean, standardize, and semantically unify the event data, and construct a structured event dataset for human factors analysis; The structured information of the event is extracted from the formal operation event report, including the event process description, procedure execution record, interface status characteristics and corrective measures. These are uniformly encoded into structured entries that can be used for human factors analysis, forming a structured event dataset.

[0049] S200: Based on preset human factor event annotation rules, multidimensional annotation is performed on each event in the structured event dataset to obtain the multidimensional annotation results, as shown in Table 4: Table 4 Multidimensional Annotation Results S300, Perform mechanistic decomposition on the multidimensional annotation results to obtain the interface factor decomposition results: A secondary mechanism decomposition was performed on the event with UI=1. Based on the event description, root cause analysis, and corrective measures (the corrective measures pointed to 'changing the symbol to a clear text label'), triangulation was conducted to identify the interface factors as shown in Table 5: Table 5. Decomposition Results of Interface Factors S400, based on the multidimensional annotation results and the interface factor decomposition results, determine the procedure deviation risk.

[0050] Using the weights derived in Example 1 (from statistical results of the same event dataset, ensuring consistency within the method): .

[0051] Interface risk factor calculation: Calculation of deviation risk from procedures: S500: For each event, determine whether the event is a high-risk coupling event based on the deviation risk of the procedure. If so, generate interface optimization suggestions and procedure revision suggestions for the high-risk coupling event. Exceeding the high-risk threshold, this is identified as a high-risk layout trap-type coupling event. Layout factors ( The largest weight ) is the main driving factor, Label factor ( This constitutes a minor contribution; it is recommended to prioritize improving the spacing between control areas and the salience of labels.

[0052] The two examples share the same set of weights (both from the dataset statistics in step three). and All data are derived from full sample analysis, demonstrating the consistency of the methodology.

[0053] The only difference between the two examples lies in the activation factors (semantic / mismatch vs. layout / label), resulting in different U-values ​​(0.409 vs. 0.591). There are also differences (0.597 vs. 0.593).

[0054] This demonstrates that the method can distinguish between different types of interface-procedure coupling risk mechanisms, the data flow between S100 and S400 is complete, and the formula parameters all have clear statistical sources.

[0055] The above description is merely a preferred embodiment of this specification and is not intended to limit this specification. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this specification should be included within the scope of protection of this specification.

[0056] It should also be noted that the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such process, method, article, or apparatus. Without further limitation, an element qualified by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.

Claims

1. A method for risk analysis of nuclear power plant interface protocol coupling based on event data, characterized in that, include: Acquire event data during nuclear power plant operation events, clean, standardize, and semantically unify the event data to construct a structured event dataset for human factors analysis; Based on preset human factor event annotation rules, each event in the structured event dataset is annotated in multiple dimensions to obtain multidimensional annotation results; Mechanistic decomposition of the multidimensional annotation results yields interface factor decomposition results; Based on the multidimensional annotation results and the interface factor decomposition results, the risk of deviation from the procedure is determined.

2. The nuclear power interface procedure coupling risk analysis method for event data according to claim 1, characterized in that, The event data includes multiple events, each of which includes at least event description information, operation procedure execution information, human-machine interface feature information, and event result information. The human-machine interface feature information includes interface layout information, interface display semantic information, and interface identification information.

3. The nuclear power plant interface procedure coupling risk analysis method for event data according to claim 1, characterized in that, The multidimensional annotation of each event in the structured event dataset based on preset human factor event annotation rules yields the following results: Based on preset human factor event annotation rules, each event in the structured event dataset is labeled with procedure deviation to characterize whether there are any omissions, errors, or improper execution of procedures in the event, thus obtaining the procedure deviation annotation results. Each event is labeled with an interface problem to characterize whether there are human-computer interface related defects in the event, and the interface problem labeling results are obtained. Each event is labeled with interface and procedure coupling to characterize whether the interface problem in the event constitutes a direct cause of inducing or amplifying the deviation of the procedure, and the interface and procedure coupling labeling results are obtained. The deviation annotation results, interface problem annotation results, and interface-procedure coupling annotation results are determined as multi-dimensional annotation results.

4. The nuclear power plant interface procedure coupling risk analysis method for event data according to claim 3, characterized in that, The mechanism decomposition of the multidimensional annotation results to obtain the interface factor decomposition results includes: The events with human-computer interface-related defects in the interface problem annotation results are analyzed to obtain the interface factors that cause the human-computer interface-related defects, and the interface factors that cause the human-computer interface-related defects and their corresponding values ​​are determined as the interface factor decomposition results.

5. The nuclear power plant interface procedure coupling risk analysis method for event data according to claim 4, characterized in that, The interface factors include: layout and operation area factors, semantic and status indication factors, interface and procedure semantic mismatch factors, and identification and naming factors.

6. The nuclear power plant interface procedure coupling risk analysis method for event data according to claim 4, characterized in that, The determination of procedure deviation risk based on the multidimensional annotation results and the interface factor decomposition results includes: The deviation rate of the benchmark procedure is determined using the deviation annotation results of the aforementioned procedure. Weights are assigned to each interface factor using the interface problem annotation results and the interface and procedure coupling annotation results; The risk of deviation from the procedure is determined by using the weights assigned to each interface factor and the deviation rate of the benchmark procedure.

7. The nuclear power plant interface procedure coupling risk analysis method for event data according to claim 6, characterized in that, The determination of procedure deviation risk using the weights assigned to each interface factor and the baseline procedure deviation rate includes: The coupling amplification factor is determined using the deviation rate of the aforementioned reference procedure; By utilizing the weights and corresponding values ​​of each interface factor, the interface risk factor is determined. The risk of deviation from the procedure is determined using the coupling amplification factor and the interface risk factor.

8. The method for nuclear power interface procedure coupling risk analysis of event data according to claim 1, characterized in that, After determining the procedure deviation risk based on the multidimensional annotation results and the interface factor decomposition results, the nuclear power interface procedure coupling risk analysis method for event data further includes: For each event, the event is determined to be a high-risk coupling event based on the deviation risk of the procedure. If it is, interface optimization suggestions and procedure revision suggestions are generated for the high-risk coupling event.

9. The method for nuclear power plant interface procedure coupling risk analysis of event data according to claim 8, characterized in that, The interface optimization suggestions include at least display semantic optimization suggestions and operation area layout optimization suggestions; The proposed revisions to the procedures should include at least the proposed revisions to ensure consistency between the procedure text and the interface identifiers.

10. A nuclear power plant interface procedure-coupled risk analysis system for event data, characterized in that, include: The data acquisition and preprocessing module is used to acquire event data during nuclear power plant operation events, clean, standardize, and semantically unify the event data, and construct a structured event dataset for human factors analysis; the structured event dataset contains multiple events. The multidimensional annotation module is used to perform multidimensional annotation on each event in the structured event dataset based on preset human factor event annotation rules, and obtain multidimensional annotation results; The mechanism decomposition module is used to perform mechanism decomposition on the multidimensional annotation results to obtain the interface factor decomposition results. The risk quantification calculation module is used to determine the deviation risk of the procedure based on the multidimensional annotation results and the interface factor decomposition results.

Images