A train man-machine interface display intelligent analysis method based on artificial intelligence

Abstract

The application relates to the technical field of intelligent monitoring of rail transit trains, and discloses a train man-machine interface display intelligent analysis method based on artificial intelligence. The method analyzes the dynamic evolution track of the interface state in a historical period, evaluates the stability of display elements, and predicts potential fault paths. Simulated operations are applied to logic conflict nodes, underlying service features and user operation behavior features are separated from mixed responses, and independent tracing and positioning are respectively performed. The method can realize early warning of abnormal behaviors of interface elements, accurately distinguish and position different levels of faults caused by system service defects and interactive logic errors, and provides a core technical means for intelligent monitoring and efficient maintenance of a train man-machine interface.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of intelligent monitoring of rail transit trains, in particular to a train man-machine interface display intelligent analysis method based on artificial intelligence. BACKGROUND

[0002] The existing train man-machine interface display state monitoring and fault diagnosis technology mainly relies on the static compliance check of preset rules and the threshold judgment of response time. The static compliance check judges whether the display content is correct and complete by comparing the current interface image with the standard template. This method can only capture static and explicit errors. Response time monitoring focuses on the overall length of time from the issuance of the operation instruction to the completion of the interface update. Once the time is exceeded, it is determined to be abnormal. These two methods are the mainstream means to ensure the availability of the train man-machine interface.

[0003] The above-mentioned prior art solution has defects. Static checking cannot deal with logical conflicts generated in dynamic interaction related to time or operation sequence. Response time monitoring can sense performance degradation, but cannot distinguish whether the abnormal source is caused by the delay of underlying data service, the failure of interface rendering engine, or the triggering of inefficient processing logic by specific operation combination. After the fault occurs, the maintenance personnel often need to rely on experience to perform a large amount of manual testing and log analysis, which is low in positioning efficiency and easy to miss deep related problems.

[0004] The present application aims to solve the problems of how to predict potential faults from dynamic interaction and how to accurately separate and locate different levels of root causes when an abnormality occurs. This requires a technical solution that can understand the long-term behavior patterns of interface elements and has the ability to decouple the source of mixed response signals. SUMMARY

[0005] The present application aims to provide a train man-machine interface display intelligent analysis method based on artificial intelligence to solve the problems raised in the background art.

[0006] To achieve the above-mentioned purpose, the present application provides a train man-machine interface display intelligent analysis method based on artificial intelligence, which comprises:

[0007] Obtaining the interface image stream of the target man-machine interface from the train operation monitoring system, outputting the interface description information containing semantic labels, and obtaining the verified real-time interface state;

[0008] Extracting the evolution track of the real-time interface state in multiple historical periods, identifying the pattern deviation interval in the evolution track, and calculating the stability quantitative value of the interface display element based on the pattern deviation interval;

[0009] The stability quantitative value is input into a fault knowledge graph, a potential fault evolution path is matched out, and a logic conflict node in the target human-computer interface is reversely deduced according to the potential fault evolution path;

[0010] An analog operation sequence is applied to the logic conflict node, a response feature of the target human-computer interface is captured, and a service feature generated by an underlying display service and a behavior feature generated by a user operation behavior are separated from the response feature;

[0011] The service feature and the behavior feature are independently traced, a service abnormal source and a behavior abnormal source are located, a multi-dimensional analysis task is constructed, a parallel diagnosis scheme is sequentially arranged according to a safety analysis procedure, an analysis instruction list is generated, and is loaded into a simulation test environment to drive full-process simulation, and a final output result is collected;

[0012] The final output result is subjected to conformity determination, and a complete intelligent analysis conclusion of the target human-computer interface is formed in combination with the multi-dimensional analysis task.

[0013] Preferably, the interface image stream of the target human-computer interface is acquired from the train operation monitoring system, and interface description information containing semantic labels is output, and a verified real-time interface state is obtained, including:

[0014] The structured display elements in the interface image stream are identified, the structured display elements are compared with entries in a preset interface knowledge base, and the interface description information containing semantic labels is output;

[0015] The interface description information is subjected to dynamic intention mapping, an interface state hypothesis under a current operation scenario is generated, and the interface state hypothesis is verified by calling a historical log, and a verified real-time interface state is obtained;

[0016] The interface description information is subjected to dynamic intention mapping, an interface state hypothesis under a current operation scenario is generated, and the interface state hypothesis is verified by calling a historical log, and a verified real-time interface state is obtained;

[0017] The semantic labels in the interface description information are parsed, and the functional component types corresponding to the semantic labels are identified;

[0018] The running mode parameters of the current train are acquired, the running mode parameters are associated and matched with the functional component types, and the current operation scenario is determined;

[0019] According to the current operation scenario, a preset scene-state mapping rule is queried, and an interface state hypothesis corresponding to the interface description information is generated;

[0020] The assumption conditions contained in the interface state hypothesis are traversed, and the assumption conditions are checked one by one for realizability in the target human-computer interface;

[0021] screening the hypothesis condition whose realizability exceeds a preset threshold, and generating an interface state hypothesis in the current operation scenario based on the hypothesis condition.

[0022] Preferably, the call history log verifies the interface state hypothesis to obtain a verified real-time interface state, including:

[0023] calling a historical operation log associated with the target human-computer interface from a central data server of the train;

[0024] finding a record segment adjacent in time to the interface state hypothesis in the historical operation log;

[0025] extracting interface response data recorded in the record segment;

[0026] comparing the interface response data with expected data of the interface state hypothesis one by one, and recording consistent items and inconsistent items;

[0027] counting the total number of the consistent items, and determining that the interface state hypothesis is established when the total number reaches a preset verification pass standard;

[0028] fusing the interface state hypothesis with data at the latest time point in the interface response data to form the verified real-time interface state.

[0029] Preferably, the extraction of the evolution track of the real-time interface state in multiple historical periods and the identification of a mode deviation interval in the evolution track include:

[0030] setting a fixed time window length, and according to the time window length, cutting continuous multiple historical period data from a historical database corresponding to the real-time interface state;

[0031] extracting key attribute values of the real-time interface state in each historical period data;

[0032] arranging the key attribute values in chronological order to form an evolution track of the real-time interface state in multiple historical periods;

[0033] setting a reference behavior mode curve on the evolution track;

[0034] calculating the difference between data at each time point on the evolution track and a corresponding point on the reference behavior mode curve, and marking a continuous time region whose difference value exceeds a preset fluctuation range as the mode deviation interval.

[0035] Preferably, the calculation of the stability quantitative value of the interface display element based on the mode deviation interval includes:

[0036] counting the cumulative duration and frequency of the mode deviation interval in the total analysis period;

[0037] measuring the numerical variation amplitude of the interface display element in each mode deviation interval;

[0038] obtaining the standard variation range of the interface display element in the normal working mode;

[0039] constructing a stability evaluation function based on the cumulative duration, the frequency, the numerical variation amplitude, and the standard variation range;

[0040] inputting the cumulative duration, the frequency, and the numerical variation amplitude into the stability evaluation function to obtain a stability quantitative value of the interface display element.

[0041] Preferably, the inputting of the stability quantitative value into the fault knowledge graph to match out a potential fault evolution path comprises:

[0042] constructing a fault knowledge graph with the interface display element as a node and the logical dependency and signal transmission relationship between elements as an edge;

[0043] annotating a historical stability quantitative value range and a corresponding typical fault mode for each node in the fault knowledge graph;

[0044] comparing the calculated stability quantitative value of the interface display element with the historical stability quantitative value range of the corresponding node in the fault knowledge graph;

[0045] if the stability quantitative value falls within the historical stability quantitative value range corresponding to a typical fault mode, activating the path associated with the typical fault mode in the fault knowledge graph;

[0046] selecting a path that meets the current train operation constraint condition from the activated path as the potential fault evolution path.

[0047] Preferably, the reverse deduction of the logical conflict node in the target human-machine interface according to the potential fault evolution path comprises:

[0048] starting from the end fault phenomenon of the potential fault evolution path and performing reverse tracing along the edges in the fault knowledge graph;

[0049] recording all the passed nodes and the causal relationship between the nodes in the reverse tracing process;

[0050] analyzing the causal relationship to identify a causal chain with contradictions or unsatisfied conditions;

[0051] locating a node to which the contradictory or unsatisfied conditions point;

[0052] determining the node as a logic conflict node in the target human-machine interface.

[0053] Preferably, the logic conflict node is subjected to a simulation operation sequence, and response characteristics of the target human-machine interface are captured, including:

[0054] According to the function definition of the logic conflict node, a simulation operation sequence covering normal operation, boundary operation and abnormal operation is designed;

[0055] In a simulation environment, the simulation operation sequence is applied to the target human-machine interface containing the logic conflict node as input;

[0056] The internal state variables and external output signals of the target human-machine interface during processing of the simulation operation sequence are monitored in real time;

[0057] The change process of the internal state variables and the timing waveform of the external output signals are recorded;

[0058] From the recorded change process and timing waveform, feature data capable of representing the dynamic behavior of the target human-machine interface are extracted as the response characteristics.

[0059] Preferably, the service characteristics and the behavior characteristics are independently traced back to locate service abnormal sources and behavior abnormal sources, multi-dimensional analysis tasks are constructed, parallel diagnosis schemes are sequentially arranged according to safety analysis procedures, an analysis instruction list is generated, loaded into a simulation test environment to drive full-process simulation, and final output results are collected, including:

[0060] locating the service abnormal sources corresponding to the service characteristics and the behavior abnormal sources corresponding to the behavior characteristics;

[0061] Based on the service abnormal sources and the behavior abnormal sources, multi-dimensional analysis tasks are constructed, which contain parallel diagnosis schemes for the target human-machine interface;

[0062] According to the preset safety analysis procedures, the parallel diagnosis schemes are sequentially arranged to generate an executable analysis instruction list;

[0063] The analysis instruction list is sequentially loaded into a simulation test environment where the target human-machine interface is located, to drive the target human-machine interface to execute full-process simulation, and after the full-process simulation is completed, the final output results of the target human-machine interface are collected.

[0064] Preferably, the compliance of the final output result is judged in combination with the multi-dimensional analysis task to form a complete intelligent analysis conclusion of the target human-computer interface, including:

[0065] Interface display content, response time and data accuracy indicators are parsed from the final output result;

[0066] The interface display content, the response time and the data accuracy indicators are compared one by one with the requirements in the predefined system specification;

[0067] According to the comparison result, the compliance, non-compliance or deviation of each indicator item is given;

[0068] The judgment conclusions of all indicator items are summarized, and the service abnormal source and the behavior abnormal source information found in the execution of the multi-dimensional analysis task are combined;

[0069] The judgment conclusions and the service abnormal source and behavior abnormal source information are integrated to generate a structured diagnostic report, which is the complete intelligent analysis conclusion of the target human-computer interface.

[0070] Compared with the prior art, the beneficial effects of the present application are:

[0071] The evolution trajectory of the interface state in multiple historical periods is extracted, the mode deviation interval is identified, and the stability quantitative value of the interface display element is calculated based on this. The analysis dimension is expanded from a static snapshot to a dynamic time series, which can quantify the predictability and consistency of each interface element behavior. The system can identify those interface elements that have not violated the static rules, but the behavior pattern has deviated slightly and presents an unstable trend, realizing early warning and trend judgment of potential faults. Based on the quantitative stability value, the fault path matching changes the prediction basis from a single rule to a comprehensive evaluation combined with historical behavior patterns, and the pertinence and foresight of the warning are enhanced.

[0072] From the overall interface response characteristics triggered by the simulation operation, the service characteristics generated by the underlying display service and the behavior characteristics generated by the user operation logic are separated, and the two are independently traced to locate the service abnormal source and the behavior abnormal source. Decoupling analysis of complex mixed signals is realized, breaking the traditional limitation of diagnosing the interface response as a whole black box. Through separation and tracing, it can be clearly distinguished whether a one-time interface exception is caused by problems in the back-end data service, communication link or rendering engine, or defects in the processing of a specific operation sequence by the front-end interface logic. Precise source location enables maintenance measures to directly target the problem core, avoiding blind troubleshooting between different levels of the system, improving diagnosis efficiency and repair accuracy. BRIEF DESCRIPTION OF DRAWINGS

[0073] Figure 1 The working principle diagram of the intelligent analysis method of the train man-machine interface based on artificial intelligence is described in the present application;

[0074] Figure 2 The flowchart for verifying the interface state hypothesis is described in the present application;

[0075] Figure 3 The flowchart for identifying the mode deviation interval is described in the present application;

[0076] Figure 4 The train man-machine interface logic conflict node path confidence analysis diagram is described in the present application;

[0077] Figure 5 The train man-machine interface element stability score comparison diagram is described in the present application. DETAILED DESCRIPTION

[0078] The technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor fall within the scope of protection of the present application.

[0079] Please refer to Figure 1 The present application provides an intelligent analysis method of a train man-machine interface based on artificial intelligence, which comprises the following steps: obtaining an interface image stream of a target man-machine interface from a train operation monitoring system, outputting interface description information containing semantic labels, and obtaining a verified real-time interface state; extracting an evolution track of the real-time interface state in multiple historical periods, identifying a mode deviation interval in the evolution track, calculating a stability quantitative value of an interface display element based on the mode deviation interval; inputting the stability quantitative value into a fault knowledge graph, matching out a potential fault evolution path, and reversely deducing a logic conflict node in the target man-machine interface according to the potential fault evolution path; applying a simulation operation sequence to the logic conflict node, capturing response characteristics of the target man-machine interface, separating service characteristics generated by a bottom display service and behavior characteristics generated by user operation behaviors from the response characteristics; independently tracing the service characteristics and the behavior characteristics respectively, locating service abnormal sources and behavior abnormal sources, constructing a multi-dimensional analysis task, sequentially arranging parallel diagnosis schemes according to safety analysis procedures, generating an analysis instruction list, loading to a simulation test environment to drive full-process simulation, and collecting final output results; performing compliance determination on the final output results, combining the multi-dimensional analysis task, and forming a complete intelligent analysis conclusion of the target man-machine interface.

[0080] In one embodiment of the present application, please refer toFigure 2 , identifying a structured display element in the interface image stream, comparing the structured display element with entries in a preset interface knowledge base, and outputting interface description information containing semantic labels. The interface description information is dynamically intent-mapped to generate an interface state hypothesis in the current operation scenario, and the interface state hypothesis is verified by calling a historical log to obtain a verified real-time interface state. The specific process of dynamically intent-mapping the interface description information to generate an interface state hypothesis in the current operation scenario includes: analyzing semantic labels in the interface description information, identifying a functional component type corresponding to the semantic labels; obtaining a running mode parameter of the current train, associating and matching the running mode parameter with the functional component type to determine the current operation scenario; according to the current operation scenario, querying a preset scene-state mapping rule to generate an interface state hypothesis corresponding to the interface description information; traversing the hypothesis conditions contained in the interface state hypothesis, and checking the realizability of the hypothesis conditions in the target human-computer interface one by one; filtering out hypothesis conditions with realizability exceeding a preset threshold, and generating an interface state hypothesis in the current operation scenario based on the hypothesis conditions. The specific process of verifying the interface state hypothesis by calling the historical log to obtain a verified real-time interface state includes: calling a historical operation log associated with the target human-computer interface from a central data server of the train; finding a record segment adjacent in time to the interface state hypothesis in the historical operation log; extracting interface response data recorded in the record segment; comparing the interface response data with expected data of the interface state hypothesis one by one, recording consistent items and inconsistent items; counting the total number of consistent items, and determining that the interface state hypothesis is established when the total number reaches a preset verification pass standard; fusing the interface state hypothesis with data at the latest time point in the interface response data to form the verified real-time interface state.

[0081] In a specific implementation, after obtaining the interface image stream of the target human-computer interface from the train operation monitoring system, the structured display elements in the interface image stream are identified, including digital instrument readings, state indicator icons, text prompt boxes, and operation button areas. The identified structured display elements are compared with entries in the preset interface knowledge base, and the preset interface knowledge base entries contain expected positions, color coding specifications, text templates, and associated signal names of the display elements. Through the comparison, the functional semantics and data sources of each matched structured display element are labeled, and interface description information containing these semantic labels is output. The interface description information is a structured data set that records the semantics and values of "current speed: 180 km / h", "braking state: applied", and "front signal: green light" in the interface.

[0082] In some embodiments, the interface description information is dynamically intention-mapped to generate an interface state hypothesis under the current operation scenario, semantic markers in the interface description information are parsed, a functional component type corresponding to the semantic markers is identified, such as "speed display", "braking state indication", and "signal state". An operation mode parameter of the current train is obtained, the operation mode parameter including a train control mode, a line condition, and a dispatching command. The operation mode parameter is associated and matched with the functional component type to determine that the current operation scenario is a "high-speed cruising braking preparation" scenario. According to the "high-speed cruising braking preparation" current operation scenario, a preset scenario-state mapping rule is queried, the mapping rule defining a state set that the interface should present under this scenario, so as to generate an interface state hypothesis corresponding to the interface description information, the interface state hypothesis including "traction force output should be zero", "electric braking icon should be lit", and "safety distance warning line should be displayed".

[0083] Optionally, the interface state hypothesis contains a hypothesis condition, and the hypothesis condition is checked one by one for realizability in the target human-machine interface. The checking process involves checking whether the underlying software components of the target human-machine interface support generating the state described in the hypothesis condition, and whether the current hardware resources are sufficient. Hypothesis conditions with realizability exceeding a preset threshold are screened out, for example, "traction force output should be zero" and "electric braking icon should be lit" are confirmed as realizable, while "safety distance warning line should be displayed" is marked as low realizability due to unavailability of current sensor data. Based on the screened hypothesis conditions with realizability exceeding the preset threshold, an interface state hypothesis under the current operation scenario is generated.

[0084] It can be understood that the history log is called to verify the interface state hypothesis to obtain a verified real-time interface state, and a history operation log associated with the target human-machine interface is called from a central data server of the train, the history operation log recording, by time stamp, previous operation instructions and responses of the interface system. A record segment adjacent in time to the interface state hypothesis is searched for in the history operation log, and a time span of the record segment covers a period of time before and after the time corresponding to the interface state hypothesis. Interface response data recorded in the record segment is extracted, the interface response data including a value actually displayed by the interface at a historical time, a state change sequence, and a system message. The interface response data is compared with expected data of the interface state hypothesis one by one, for example, "traction force actual output value" recorded in the history log is compared with "traction force output should be zero" of the hypothesis, and consistent items and inconsistent items are recorded. The total number of consistent items is counted, and when the total number reaches a preset verification pass standard, it is determined that the interface state hypothesis is established. The preset verification pass standard is defined by a proportion formula:

[0085]

[0086] wherein V represents a verification pass value, represents the total number of consistent items, represents the total number of items participating in the comparison, when V is greater than or equal to a preset threshold When the interface state hypothesis is determined to be true, the true interface state hypothesis is fused with the data of the latest time point in the interface response data to form a verified real-time interface state, which is a higher confidence interface state snapshot fused with the hypothesis logic and the historical actual data.

[0087] In one embodiment of the present application, referring to Figure 3 , a fixed time window length is set, and according to the time window length, a plurality of continuous historical period data are intercepted from the historical database corresponding to the real-time interface state. In each of the historical period data, the key attribute value of the real-time interface state is extracted; the key attribute values are arranged in time sequence to form the evolution track of the real-time interface state in a plurality of historical periods. A reference behavior mode curve is set on the evolution track; the difference between the data at each time point on the evolution track and the corresponding point on the reference behavior mode curve is calculated, and the continuous time region whose difference exceeds the preset fluctuation range is marked as the mode deviation interval. The specific process of calculating the stability quantitative value of the interface display element based on the mode deviation interval includes: counting the cumulative duration and occurrence frequency of the mode deviation interval in the total analysis period; measuring the numerical change amplitude of the interface display element in each mode deviation interval; obtaining the standard change range of the interface display element in the normal working mode; based on the cumulative duration, the occurrence frequency, the numerical change amplitude and the standard change range, a stability evaluation function is constructed; the cumulative duration, the occurrence frequency and the numerical change amplitude are input into the stability evaluation function to calculate the stability quantitative value of the interface display element.

[0088] In a specific implementation, a fixed time window length is set, which is configured as twenty-four hours, and according to the time window length, a plurality of continuous historical period data are intercepted from the historical database corresponding to the real-time interface state, which are data blocks containing interface state historical records divided according to the time window length. In each of the historical period data, the key attribute value of the real-time interface state is extracted, which can be the flicker frequency of an indicator light, the numerical reading sequence of an instrument or the pixel color mean value of a specific region. The key attribute values extracted from the continuous plurality of historical periods are arranged in time sequence to form the evolution track of the real-time interface state in a plurality of historical periods, which is a discrete sequence with time as the horizontal axis and the key attribute value as the vertical axis.

[0089] In some embodiments, a reference behavior pattern curve is set on the evolution trajectory, the reference behavior pattern curve is generated by statistically fitting the sequence of the same key attribute values under historical normal operating conditions, and represents the expected standard behavior. The difference between the data at each time point on the evolution trajectory and the corresponding point on the reference behavior pattern curve is calculated, and the difference reflects the deviation between the actual observation value and the expected standard value. The continuous time region with a difference value exceeding a preset fluctuation range is marked as a pattern deviation interval, and the preset fluctuation range is determined according to the normal operating tolerance of the key attribute value. The pattern deviation interval indicates the period of abnormality or uncertainty of the interface display behavior. Optionally, based on the pattern deviation interval, the specific process of calculating the stability quantitative value of the interface display element is as follows: the cumulative duration and occurrence frequency of the pattern deviation interval in the total analysis period are calculated, and the total analysis period is the total time span covered by the selected historical period data. The numerical change amplitude of the interface display element in each pattern deviation interval is measured, and the numerical change amplitude is the difference between the maximum value and the minimum value of the key attribute value in the pattern deviation interval. The standard change range of the interface display element under the normal working mode is obtained, and the standard change range defines the upper and lower limits of the numerical value change of the element allowed under the fault-free condition. Based on the cumulative duration, the occurrence frequency, the numerical change amplitude and the standard change range, a stability evaluation function is constructed.

[0090] It can be understood that the cumulative duration, the occurrence frequency and the numerical change amplitude are input into the stability evaluation function to calculate the stability quantitative value of the interface display element. One specific example of the stability evaluation function is as follows:

[0091]

[0092] Wherein: S represents the stability quantitative value to be solved, the value range of which is between 0 and 1, and the higher the value is, the more stable it is; T represents the cumulative duration of the pattern deviation interval; represents the length of the total analysis period; F represents the occurrence frequency of the pattern deviation interval; A represents the numerical change amplitude in the pattern deviation interval; represents the maximum allowed change amplitude of the standard change range; , , is a pre-set weight coefficient, which satisfies .

[0093] In one embodiment of the present application, a fault knowledge graph is constructed, taking interface display elements as nodes and logical dependency and signal transmission relationship between elements as edges. In the fault knowledge graph, each node is labeled with a range of historical stability quantitative values and corresponding typical fault modes. The calculated stability quantitative value of the interface display element is compared with the range of historical stability quantitative values of the corresponding node in the fault knowledge graph. If the stability quantitative value falls within the range of historical stability quantitative values corresponding to a typical fault mode, the path associated with the typical fault mode in the fault knowledge graph is activated. From the activated path, a path that meets the current train operation constraint condition is selected as the potential fault evolution path. The specific process of reverse deduction of the logical conflict node in the target human-machine interface according to the potential fault evolution path includes: starting from the end fault phenomenon of the potential fault evolution path, reverse tracing along the edges in the fault knowledge graph; recording all passed nodes and causal relationships between nodes during reverse tracing; analyzing the causal relationships to identify causal chains that have contradictions or do not meet conditions; locating the node pointed to by the causal chains that have contradictions or do not meet conditions; and determining the node as the logical conflict node in the target human-machine interface.

[0094] In a specific implementation, a fault knowledge graph is constructed, taking interface display elements as nodes and logical dependency and signal transmission relationship between elements as edges. The interface display elements include a speed gauge, a brake status indicator light, and a signal display icon. The logical dependency relationship is embodied as a prerequisite for the normal display of one interface display element, which requires another interface display element to output a specific signal. The signal transmission relationship records the flow path of data from sensors to display units through processors. In the fault knowledge graph, each node is labeled with a range of historical stability quantitative values and corresponding typical fault modes. The range of historical stability quantitative values is derived from the statistical induction of the stability quantitative values of the same interface display element in historical normal data and fault cases. The typical fault modes include "display refresh delay", "data packet loss", and "logical deadlock".

[0095] In some embodiments, the calculated stability quantification value of the interface display element is compared with the range of historical stability quantification values of the corresponding node in the fault knowledge graph, and the comparison operation checks in which historical statistical interval the current stability quantification value falls. If the stability quantification value falls within the range of historical stability quantification values corresponding to a typical fault mode, the path associated with the typical fault mode in the fault knowledge graph is activated, and the activation operation marks the nodes and edges on the path as to-be-analyzed states. From the activated path, a path that meets the current train operation constraint conditions is selected as a potential fault evolution path, and the current train operation constraint conditions include the current train speed interval, the line slope information, and the vehicle-mounted device model matching rule. Only the path that meets all these constraint conditions will be retained.

[0096] Optionally, according to the potential fault evolution path, a logical conflict node in the target human-machine interface is reversely deduced, starting from the end fault phenomenon of the potential fault evolution path, and tracing back along the edges in the fault knowledge graph. The end fault phenomenon is the fault state described by the path endpoint. In the reverse tracing process, all the passed nodes and the causal relationships between the nodes are recorded, and the causal relationships describe how the upstream node state affects the downstream node state. The recorded causal relationships are analyzed to identify the causal chains that have contradictions or conditions that are not met. The contradiction is manifested as a conflict between the input conditions of two causal chains for the same node, and the condition that is not met is manifested as an output state of a node that cannot meet the activation condition of its downstream node. The node pointed to by the causal chains with contradictions or conditions that are not met is located, and the common pointing means that the reasoning endpoints or key impact points of multiple abnormal causal chains converge on the same interface display element node.

[0097] It can be understood that the node pointed to by the causal chains is determined as the logical conflict node in the target human-machine interface, and the process of determining the logical conflict node involves evaluating the centrality of the node in the fault knowledge graph and its role in the abnormal causal chain. One way to calculate the path confidence matching degree of a path for assisting in screening the logical conflict node is:

[0098]

[0099] wherein C represents the path confidence matching degree of the logical conflict node, M represents the number of abnormal causal chains with the node as the common pointing, N represents the total number of causal chains involving the node in the fault knowledge graph, is an indicator function, which takes the value of 1 when the ith causal chain is identified as abnormal, and 0 otherwise, represents the preset weight of the ith causal chain, and the weight is set according to the signal transmission criticality of the link.

[0100] In one embodiment of the present application, a simulation operation sequence covering normal operation, boundary operation and abnormal operation is designed according to the functional definition of the logic conflict node. The simulation operation sequence is applied to the target human-machine interface containing the logic conflict node as input in a simulation environment. The internal state variables and external output signals of the target human-machine interface during processing of the simulation operation sequence are monitored in real time. The change process of the internal state variables and the timing waveform of the external output signals are recorded. The feature data representing the dynamic behavior of the target human-machine interface is extracted from the recorded change process and timing waveform as the response feature.

[0101] In a specific implementation, a simulation operation sequence covering normal operation, boundary operation and abnormal operation is designed according to the functional definition of the logic conflict node, which is derived from the system design document describing the specific data display and update logic of the node responsible for processing in the train human-machine interface. The simulation operation sequence is a set of excitation signals applied to the input interface associated with the logic conflict node in a predetermined order. The normal operation sequence simulates the operation input of the driver under standard procedures, the boundary operation sequence simulates the operation input when the input parameter is at the limit of the allowed range, and the abnormal operation sequence simulates the operation input when the input parameter exceeds the allowed range or violates the operation logic. The design process constructs systematic test cases to stimulate the behavior of the logic conflict node under different working conditions according to the input and output specifications of the logic conflict node.

[0102] In some embodiments, the simulation operation sequence is applied as input to the target human-machine interface containing the logic conflict node in a simulation environment, which is an exact copy of the target human-machine interface software and its dependent train control model on an isolated hardware or virtualization platform. The implementation of the train control model involves integrating key parameters in the train operation monitoring system, including train control mode, line condition, and dispatch command, which together form the basis for simulating train behavior; the model is implemented as a software module in the simulation environment, which accurately replicates the control logic and dynamic response of the actual train, with the integration of train current speed interval, line slope information, and on-board equipment model matching rule operating constraint conditions to ensure that the model can truly reflect the operating characteristics of the train under specific working conditions; by being deployed on an isolated hardware or virtualization platform, the train control model interacts safely with the target human-machine interface software, thereby supporting the simulation operation sequence test of the logic conflict node and capturing reliable response characteristics for subsequent analysis. The application process accurately sends each excitation signal in the simulation operation sequence to the corresponding input port of the target human-machine interface according to the preset timing through the interface injection tool of the simulation environment. The internal state variables of the target human-machine interface during processing of the simulation operation sequence, including intermediate results of calculations within the logic conflict node, buffer queue length, and thread state flag, and external output signals, such as graphics, numerical values updated on the display screen, or control instructions sent to other systems, are monitored in real time.

[0103] Optionally, the change process of internal state variables and the timing waveform of external output signals are recorded, the change process is saved in the form of a log of time stamp-variable value, and the timing waveform is saved in the form of sampling data of time-signal amplitude. Feature data representing the dynamic behavior of the target human-machine interface is extracted from the recorded change process and timing waveform as response characteristics, and the extraction operation involves filtering, calculating statistics, and identifying specific patterns on the original data. The response characteristics are a multi-dimensional vector, each dimension of which corresponds to a behavior attribute quantified from the monitoring data.

[0104] It can be understood that, to systematize the design of simulation operation sequences, one practical way is to divide the input space of the logic conflict node. Referring to Table 1, a simulation operation sequence design for a logic conflict node named "speed display controller" is shown.

[0105] Table 1: Simulation operation sequence design table

[0106]

[0107] Based on the multiple feature values extracted from the response characteristics, a comprehensive response deviation index can be calculated to quantify the degree of abnormal behavior under simulation operation. The calculation formula of the index is:

[0108]

[0109] wherein: D represents the comprehensive response deviation degree, N represents the total number of extracted features, represents the actual observation value of the i-th feature, represents the expected value of the i-th feature under the reference normal operation, represents the standard deviation of the i-th feature under the reference normal operation. The greater the value of the comprehensive response deviation degree D, the greater the difference between the response characteristics of the target human-machine interface under the simulated operation sequence and the normal reference behavior.

[0110] Referring to Figure 4 This is a train human-machine interface logic conflict node path confidence analysis chart for showing the correlation between the path confidence matching degree, the number of abnormal causal chains and the total number of causal chains of different logic conflict nodes, which belongs to the core index chart of the fault knowledge graph analysis stage. The confidence matching degree and the number of abnormal causal chains of the speed display controller are in the first place, which should be the primary object of fault troubleshooting; the path confidence matching degree and the number of abnormal causal chains of the node are positively correlated, which embodies the law of "the more the causal chain abnormalities, the more the core of the node in the fault path"; this chart can assist in accurate positioning of train human-machine interface faults, clarify the risk level of different logic conflict nodes, and support resource allocation and priority decision-making for fault repair.

[0111] In an embodiment of the present application, the service abnormal source corresponding to the service feature and the behavior abnormal source corresponding to the behavior feature are located. Based on the service abnormal source and the behavior abnormal source, a multi-dimensional analysis task is constructed, which contains a parallel diagnosis scheme for the target human-computer interface. According to a preset safety analysis procedure, the parallel diagnosis scheme is sequentially arranged to generate an executable analysis instruction list. The analysis instruction list is sequentially loaded into a simulation test environment where the target human-computer interface is located, to drive the target human-computer interface to perform full-process simulation. After the full-process simulation is completed, the final output result of the target human-computer interface is collected. The final output result is subjected to conformity determination, and combined with the multi-dimensional analysis task to form a specific process of a complete intelligent analysis conclusion of the target human-computer interface, which includes: analyzing interface display content, response time and data accuracy indicators from the final output result; comparing the interface display content, the response time and the data accuracy indicators with requirements in a predefined system specification one by one; according to the comparison result, giving a determination conclusion of conformity, non-conformity or existence of deviation for each indicator item; summarizing the determination conclusions of all indicator items, and combining the service abnormal source and the behavior abnormal source information found in the process of executing the multi-dimensional analysis task; comprehensively generating a structured diagnosis report from the determination conclusions and the service abnormal source and behavior abnormal source information, which is the complete intelligent analysis conclusion of the target human-computer interface.

[0112] In specific implementations, the service abnormal source corresponding to the positioning service feature and the behavior abnormal source corresponding to the behavior feature, the service abnormal source refers to the bottom software module, communication link or hardware component that causes the service feature to deviate from the normal mode, and the behavior abnormal source refers to the operation instruction sequence, user input mode or external interaction event that causes the behavior feature to present an abnormal mode. Based on the service abnormal source and the behavior abnormal source, a multi-dimensional analysis task is constructed, the multi-dimensional analysis task includes a parallel diagnosis scheme for the target human-computer interface, and the parallel diagnosis scheme is designed to simultaneously perform root cause exploration and impact evaluation on the identified abnormal source from the software service layer and the user interaction layer. In some embodiments, the parallel diagnosis scheme is sequentially arranged according to a preset safety analysis procedure and an executable analysis instruction list is generated, and the safety analysis procedure defines the execution priority, resource isolation requirement and safety boundary of fault injection that the diagnosis operation must follow. The sequential arrangement process arranges the diagnosis steps related to the core service security in front according to the priority rules in the safety analysis procedure, arranges the diagnosis steps that generate system disturbance in the resource isolated period, and generates an analysis instruction list arranged in sequence, including specific control commands and parameters. The analysis instruction list is loaded into the simulation test environment where the target human-computer interface is located in sequence, drives the target human-computer interface to perform full-process simulation, and collects the final output result of the target human-computer interface after the full-process simulation is completed. The final output result includes the interface static screenshot at the end of the simulation process, the system event log and the performance monitoring data of the entire simulation period.

[0113] Optionally, the final output result is subjected to conformity determination, and a complete intelligent analysis conclusion of the target human-computer interface is formed in combination with the multi-dimensional analysis task. The interface display content, response time and data accuracy index are analyzed from the final output result, the interface display content is the final state and value of each graphical element on the screen, the response time is the time delay experienced from triggering the diagnosis operation to the interface producing the expected change, and the data accuracy index is the error statistics between the interface display value and the reference true value injected by the simulation test environment. The analyzed interface display content, response time and data accuracy index are compared with the requirements in the pre-defined system specification one by one, and the system specification specifies the function and performance standards that the target human-computer interface must meet in all test scenarios in the form of a document.

[0114] Understandably, based on the comparison results, a judgment conclusion is given for each indicator item as compliant, non-compliant, or having deviations. This judgment conclusion is derived from a direct comparison logic between the measured value of the indicator and the threshold in the specification. The judgment conclusions for all indicator items are summarized, and combined with service anomaly source and behavior anomaly source information discovered during the multi-dimensional analysis task. Service anomaly source information includes the identifier and error code of the abnormal module, while behavior anomaly source information includes the sequence and context of abnormal operations. By combining the judgment conclusions and service and behavior anomaly source information, a structured diagnostic report is generated. This structured diagnostic report represents the complete intelligent analysis conclusion for the target human-machine interface. The generation process of the complete intelligent analysis conclusion introduces a compliance quantification evaluation function, the expression of which is:

[0115]

[0116] Where: Q represents the comprehensive compliance quantitative assessment value, and K represents the total number of indicators compared. This represents the measured value of the j-th indicator item. This represents the standard value or allowable range of the j-th indicator item in the system specification, and the function... It is a comparison function, when Fully compatible Output 1 if the requirement is met, 0 if the requirement is not met, and a value between 0 and 1 if there is a deviation. It is the preset weight coefficient of the j-th indicator item and satisfies .

[0117] See Figure 5 This is a comparison chart of the stability scores of train human-machine interface elements. It is used to show the differences between the measured stability values ​​and standard values ​​of different interface elements and is the core analytical chart for interface stability assessment. Data processing and safety monitoring show good stability, and the display panel is close to meeting the standards. The stability of the communication module and user interaction does not meet the standards and is a key area for optimization. The measured values ​​of the communication module and user interaction differ significantly from the standard values, and their operational anomalies should be investigated first. This chart can be used for stability assessment and optimization decisions of train human-machine interfaces, clearly identifying the performance shortcomings of each functional module and supporting continuous improvement of interface reliability.

[0118] It should be noted that, in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, 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.

[0119] While embodiments of the application have been shown and described, it is to be understood that the application is not limited to the details of the embodiments described, since numerous changes, modifications, substitutions and variations can be made thereto without departing from the spirit and scope of the application as defined by the appended claims and their equivalents.

Claims

1. An intelligent analysis method for train human-machine interface display based on artificial intelligence, characterized in that, The method includes: The system acquires the interface image stream of the target human-machine interface from the train operation monitoring system, outputs interface description information containing semantic tags, and obtains the verified real-time interface status. Extract the evolution trajectory of the real-time interface state in multiple historical periods, identify the pattern deviation interval in the evolution trajectory, and calculate the stability quantification value of the interface display elements based on the pattern deviation interval. The stability quantification value is input into the fault knowledge graph to match potential fault evolution paths. Based on the potential fault evolution paths, the logical conflict nodes in the target human-machine interface are deduced in reverse. A simulated operation sequence is applied to the logical conflict node to capture the response features of the target human-machine interface, and service features generated by the underlying display service and behavioral features generated by user operation behavior are separated from the response features. The service features and the behavior features are traced independently to locate the service anomaly source and the behavior anomaly source. Multi-dimensional analysis tasks are constructed, and parallel diagnostic schemes are arranged sequentially according to the security analysis procedure. An analysis instruction list is generated, loaded into the simulation test environment to drive the full process simulation, and the final output results are collected. The final output result is evaluated for compliance, and combined with the multi-dimensional analysis task, a complete intelligent analysis conclusion of the target human-computer interface is formed. The step of extracting the evolution trajectory of the real-time interface state over multiple historical periods and identifying the pattern deviation intervals in the evolution trajectory includes: Set a fixed time window length, and extract multiple consecutive historical period data from the historical database corresponding to the real-time interface status based on the time window length; Extract the key attribute values ​​of the real-time interface state from each of the historical period data; The key attribute values ​​are arranged in chronological order to form the evolution trajectory of the real-time interface state in multiple historical periods. A baseline behavior pattern curve is established on the evolution trajectory; Calculate the difference between the data at each time point on the evolution trajectory and the corresponding point on the baseline behavior pattern curve, and mark the continuous time area where the difference exceeds the preset fluctuation range as the pattern deviation interval.

2. The intelligent analysis method for train human-machine interface display based on artificial intelligence as described in claim 1, characterized in that, The process of acquiring the interface image stream of the target human-machine interface from the train operation monitoring system, outputting interface description information containing semantic tags, and obtaining the verified real-time interface state includes: Identify the structured display elements in the interface image stream, compare the structured display elements with entries in a preset interface knowledge base, and output interface description information containing semantic tags; Dynamic intent mapping is performed on the interface description information to generate an interface state hypothesis under the current operation scenario, and the interface state hypothesis is verified by calling historical logs to obtain the verified real-time interface state. The step of dynamically mapping the interface description information to generate interface state assumptions for the current operation scenario includes: Parse the semantic tags in the interface description information and identify the functional component types corresponding to the semantic tags; Obtain the current train's operating mode parameters, associate and match the operating mode parameters with the functional component types, and determine the current operating scenario; Based on the current operation scenario, a preset scenario-state mapping rule is queried to generate an interface state hypothesis corresponding to the interface description information; Iterate through the assumptions contained in the interface state assumptions and check the feasibility of each assumption in the target human-computer interface. Hypothetical conditions whose feasibility exceeds a preset threshold are selected, and based on the hypothetical conditions, interface state hypotheses for the current operation scenario are generated.

3. The intelligent analysis method for train human-machine interface display based on artificial intelligence as described in claim 2, characterized in that, The process of calling historical logs to verify the interface state assumption and obtain a verified real-time interface state includes: Retrieve the historical operation logs associated with the target human-machine interface from the train's central data server; Search the historical operation log for record segments that are temporally adjacent to the assumed interface state. Extract the interface response data recorded in the recorded segment; The interface response data is compared one by one with the expected data of the interface state assumption, and the consistent items and inconsistent items are recorded. The total number of the consistent items is counted. When the total number reaches the preset verification pass standard, the interface state assumption is determined to be valid. The interface state assumption is fused with the latest time data in the interface response data to form the verified real-time interface state.

4. The intelligent analysis method for train human-machine interface display based on artificial intelligence as described in claim 1, characterized in that, The step of calculating the stability quantification value of the elements displayed on the interface based on the pattern deviation interval includes: The cumulative duration and frequency of the deviation intervals of the described pattern within the total analysis period are statistically analyzed. Measure the magnitude of the numerical change of the displayed element within each deviation interval of the stated pattern; Obtain the standard variation range of the interface display elements in normal working mode; Based on the cumulative duration, the frequency of occurrence, the magnitude of numerical change, and the standard range of change, a stability evaluation function is constructed. The cumulative duration, the frequency of occurrence, and the magnitude of numerical change are input into the stability evaluation function to calculate the stability quantification value of the interface display element.

5. The intelligent analysis method for train human-machine interface display based on artificial intelligence as described in claim 1, characterized in that, The step of inputting the stability quantification value into the fault knowledge graph to match potential fault evolution paths includes: Construct a fault knowledge graph with interface display elements as nodes and logical dependencies and signal transmission relationships between elements as edges; In the fault knowledge graph, each node is labeled with its historical stability quantification value range and corresponding typical fault mode; The calculated stability quantification value of the interface display element is compared with the historical stability quantification value range of the corresponding node in the fault knowledge graph. If the stability quantification value falls within the range of historical stability quantification values ​​corresponding to a typical failure mode, then the path associated with the typical failure mode in the failure knowledge graph is activated. From the activated paths, paths that meet the current train operation constraints are selected as the potential fault evolution paths.

6. The intelligent analysis method for train human-machine interface display based on artificial intelligence as described in claim 1, characterized in that, The step of reversely deriving the logical conflict nodes in the target human-machine interface based on the potential fault evolution path includes: Starting from the terminal fault phenomenon of the potential fault evolution path, trace back in reverse along the edges in the fault knowledge graph; During the reverse tracing process, all nodes visited and the causal relationships between nodes are recorded; Analyze the causal relationships to identify causal chains that contain contradictions or whose conditions are not met; Locate the node that the contradictory or unmet causal chains all point to; The node is identified as a logical conflict node in the target human-machine interface.

7. The intelligent analysis method for train human-machine interface display based on artificial intelligence as described in claim 1, characterized in that, The step of applying a simulated operation sequence to the logical conflict node and capturing the response characteristics of the target human-machine interface includes: Based on the functional definition of the logical conflict nodes, design a simulated operation sequence that covers normal operation, boundary operation, and abnormal operation; In a simulation environment, the simulated operation sequence is used as input and applied to the target human-machine interface containing the logical conflict node; Real-time monitoring of the target human-machine interface's internal state variables and external output signals during the processing of the simulated operation sequence; Record the change process of the internal state variables and the timing waveform of the external output signal; Feature data that characterizes the dynamic behavior of the target human-machine interface is extracted from the recorded change process and the time-series waveform, and used as the response features.

8. The intelligent analysis method for train human-machine interface display based on artificial intelligence as described in claim 1, characterized in that, The process involves independently tracing the service characteristics and behavioral characteristics to pinpoint the sources of service and behavioral anomalies, constructing multi-dimensional analysis tasks, sequentially arranging parallel diagnostic schemes according to security analysis procedures, generating an analysis instruction list, loading it into a simulation test environment to drive the entire process simulation, and collecting the final output results, including: Locate the service anomaly source corresponding to the service feature, and the behavior anomaly source corresponding to the behavior feature; Based on the service anomaly source and the behavior anomaly source, a multi-dimensional analysis task is constructed, which includes a parallel diagnostic scheme for the target human-machine interface. Based on the preset security analysis procedure, the parallel diagnostic scheme is sequentially arranged to generate an executable analysis instruction list; The analysis instruction list is sequentially loaded into the simulation test environment where the target human-machine interface is located, driving the target human-machine interface to perform a full-process simulation. After the full-process simulation is completed, the final output result of the target human-machine interface is collected.

9. The intelligent analysis method for train human-machine interface display based on artificial intelligence as described in claim 8, characterized in that, The process of determining the conformity of the final output result, combined with the multi-dimensional analysis task, forms a complete intelligent analysis conclusion for the target human-computer interface, including: The final output results are used to parse the interface display content, response time, and data accuracy indicators. The interface display content, response time, and data accuracy indicators are compared one by one with the requirements in the predefined system specifications. Based on the comparison results, a judgment conclusion is given for each indicator item: it meets the requirements, it does not meet the requirements, or there is a deviation. Summarize the judgment conclusions of all indicator items, and combine them with the information of the service anomaly source and the behavior anomaly source found during the execution of the multi-dimensional analysis task; Based on the combined judgment conclusions and the information on the service anomaly sources and behavior anomalies, a structured diagnostic report is generated. This structured diagnostic report represents the complete intelligent analysis conclusions of the target human-machine interface.

Images