Abnormal bus state early warning method and system based on intelligent electronic stop board

Abstract

The invention relates to the technical field of bus early warning, in particular to an abnormal bus state early warning method and system based on an intelligent electronic stop board. The method comprises the following steps: acquiring real-time operation data and historical operation data of a bus; determining an abnormal state early warning model according to the historical operation data; determining abnormal state early warning information of the public transport vehicle according to the real-time operation data based on the abnormal state early warning model; and sending the abnormal state early warning information to an intelligent electronic stop board for display. Accurate and practical early warning information is provided, and passengers are assisted to make decisions. And the abnormal state early warning information is sent to the intelligent electronic stop board to be displayed, so that the information is ensured to be clearly transmitted, and the perception and trust of passengers on bus services are improved.

Description

Technical Field

[0001] The present application relates to the field of public transportation warning technology, and in particular to a method and system for warning of abnormal public transportation status based on smart electronic bus stops. Background Art

[0002] In modern urban public transportation systems, buses are the core carriers of public transportation, and their operating status is directly related to passengers' travel experience and urban transportation efficiency.

[0003] However, during the operation of buses, due to various uncontrollable factors such as road congestion, traffic accidents, vehicle breakdowns, bad weather, abnormal scheduling, etc., buses often fail to arrive on time.

[0004] However, the existing bus monitoring system is unable to monitor the real-time bus status, resulting in passengers waiting for a long time in vain, which greatly wears out their patience and seriously affects the public's evaluation and trust in bus services. Summary of the Invention

[0005] This application provides a method and system for warning of abnormal bus status based on smart electronic bus stops to solve the above problems.

[0006] In a first aspect, the present application provides a method for warning abnormal bus status based on a smart electronic bus stop sign, the method comprising: Obtain real-time and historical operation data of public buses; Determining an abnormal state early warning model based on the historical operation data; Determining abnormal state warning information of the public bus based on the abnormal state warning model and the real-time operation data; The abnormal status warning information is sent to the smart electronic bus stop for display.

[0007] This solution captures real-time and historical bus operating data, ensuring data availability and timeliness, and supporting the initiation of the early warning process. Based on historical operating data, an abnormality warning model is established to identify peak periods and road sections for abnormalities, enabling a shift from passive response to active prediction. Based on the abnormality warning model and real-time operating data, abnormality warning information for buses is determined, providing accurate and practical warning information to assist passengers in their decision-making. Abnormality warning information is then sent to smart electronic bus stops for display, ensuring clear communication and enhancing passengers' perception of and trust in public transportation services.

[0008] Optionally, determining an abnormal state warning model based on the historical operation data includes: Analyzing the historical operating data to determine historical abnormal status information; Clustering the historical abnormal status information to obtain a statistical report; According to the statistical report, an abnormal state early warning model is obtained.

[0009] This solution analyzes historical operational data, identifies historical abnormal status information, eliminates invalid or noisy data, ensures data quality, and provides a standardized input dataset for clustering. Clustering historical abnormal status information generates statistical reports that reveal the distribution patterns of abnormal status across time and space. Based on these statistical reports, an abnormal status early warning model is derived, achieving model generalization and providing probabilistic output for proactive early warning.

[0010] Optionally, obtaining an abnormal state early warning model based on the statistical report includes: Analyze the statistical report to determine the occurrence pattern of abnormal conditions; Analyze the occurrence patterns of the abnormal state and determine the frequency of abnormal state occurrence in different time periods or different operating intervals; Determine the time period and road section where abnormal conditions are most likely to occur based on the frequency of occurrence of the abnormal conditions; According to the abnormal state high-incidence time period and the abnormal state high-incidence road section, a time series prediction algorithm is used to establish an abnormal state prediction model.

[0011] This solution analyzes statistical reports to identify patterns in abnormal state occurrences and reveal their concentration in different time periods and operating intervals. By analyzing patterns in abnormal state occurrences and determining the frequency of abnormal state occurrences in different time periods or operating intervals, we provide precise numerical evidence for identifying high-incidence areas. Based on the frequency of abnormal state occurrences, we identify peak abnormality time periods and road sections, focusing on the training data required for the abnormal state prediction model to ensure that the model training process is optimized for areas most prone to abnormalities. Based on peak abnormality time periods and road sections, we use a time series prediction algorithm to establish an abnormal state prediction model, giving the model predictive capabilities and providing core support for proactive early warning.

[0012] Optionally, determining the abnormal state high-incidence time period and abnormal state high-incidence road section according to the abnormal state occurrence frequency includes: Set the time period division granularity and road section division rules; Determining a time-space unit according to the time period division granularity and the road section division rule; The sliding time window algorithm is used to count the abnormal occurrence rate of each time and space unit; When the abnormality occurrence rate of any time-space unit exceeds the preset occurrence threshold, it is marked as a high-incidence period of abnormal conditions or a high-incidence section of abnormal conditions.

[0013] This solution sets the granularity of time period division and the rules for road section division, establishes the division standards for time axis and spatial area, and ensures that multiple analyses are conducted based on consistent rules. Based on the granularity of time period division and the rules for road section division, the time and space units are determined, and the time and space dimensions are converted into an operational unit structure. A sliding time window algorithm is used to count the abnormality incidence rate of each time and space unit, quantify the density of abnormal states in each time and space unit, and provide a numerical basis for comparison and labeling. When the abnormality incidence rate of any time and space unit exceeds the preset occurrence threshold, it is marked as a high-incidence time period or high-incidence road section of abnormal states to avoid repeated output and provide a basis for training data screening for the abnormal state prediction model.

[0014] Optionally, when the abnormality occurrence rate of any spatiotemporal unit exceeds a preset occurrence threshold, it is marked as a high-incidence period of abnormality or a high-incidence section of abnormality, including: When the frequency of abnormality occurrence in any time period exceeds the preset occurrence threshold, the corresponding time period is determined to be a high-incidence period of abnormality; Alternatively, when the frequency of abnormality occurrence in any operating section exceeds a preset threshold, the corresponding operating section is determined to be a high-incidence section of abnormality.

[0015] With this solution, when the frequency of anomalies within any time period exceeds a preset threshold, the corresponding time period is determined to be a high-incidence period for anomalies, and the time period unit is identified as a time period with a high probability of anomalies occurring, facilitating the early warning model's identification and utilization of high-incidence periods. Alternatively, when the frequency of anomalies within any operating interval exceeds a preset threshold, the corresponding operating interval is determined to be a high-incidence section for anomalies, and the operating interval unit is identified as a section with a high probability of anomalies occurring, facilitating the early warning model's identification and utilization of high-incidence sections.

[0016] Optionally, determining abnormal state warning information of the public bus based on the abnormal state warning model and according to the real-time operation data includes: Comparing the real-time operating data with a preset normal operating threshold range; If the real-time operation data exceeds the preset normal operation threshold range, extracting real-time abnormal features; The real-time abnormal characteristics are input into the abnormal state warning model to obtain abnormal state warning information.

[0017] This solution compares real-time operating data with pre-set normal operating thresholds to identify potential abnormalities, triggering processing logic. If the real-time operating data exceeds the pre-set normal operating threshold, real-time abnormality features are extracted to quantify and convey the type and severity of the abnormality. These features are then fed into an abnormality warning model to generate abnormality warning information, generating predictive, user-oriented warning content for display on smart electronic bus stops.

[0018] Optionally, sending the abnormal status warning information to a smart electronic bus stop for display includes: Obtain historical travel data to determine historical travel demand and historical travel modes; Analyze the historical travel demand and the historical travel mode to determine travel focus points; According to the travel focus points, the abnormal status warning information is extracted, the core notification content is determined, and the core notification content is sent to the smart electronic bus stop for display.

[0019] This solution captures historical travel data, identifies historical travel demand and travel patterns, avoids subjective assumptions, and lays the data foundation for personalized information display. This solution analyzes historical travel demand and patterns to identify travel focus points, ensuring that information extraction focuses on the areas of greatest concern to passengers, avoiding interference from irrelevant or redundant information and improving the relevance and pertinence of information. Based on travel focus points, abnormal status warning information is extracted, and core notification content is determined. This core notification content is then sent to the smart electronic bus stop for display, ensuring user-friendly and practical information display, improving passenger decision-making efficiency and experience, while also avoiding information overload and misunderstandings.

[0020] Optionally, determining abnormal state warning information of the public bus based on the abnormal state warning model and according to the real-time operation data includes: Determine the current status of the bus based on the real-time operation data; Based on the abnormal status warning model, predict the status update time; Determine the bus arrival time according to the current state of the bus and the state update time; The bus arrival time is determined as abnormal state warning information of the bus.

[0021] This solution determines the current bus status based on real-time operating data, ensuring that the early warning process begins with an accurate vehicle status assessment. Based on the abnormal status early warning model, the status update time is predicted, providing a critical time reference for calculating bus arrival times and ensuring forward-looking early warning information. The bus arrival time is determined based on the current bus status and status update time, ensuring that the information is practical and supports passenger decision-making. The bus arrival time is used as the abnormal status early warning information for the bus, ensuring that the warning information focuses on the core content of passenger concern.

[0022] Optionally, determining abnormal state warning information of the public bus based on the abnormal state warning model and according to the real-time operation data includes: Determine the current bus stop based on the real-time operation data; Obtaining bus monitoring information, analyzing the bus monitoring information, and determining the current number of passengers; Analyze the historical operation data to determine current riding patterns; Predicting the passenger density after arrival at the bus stop based on the current bus stop, the current number of passengers, and the current riding patterns; Based on the abnormal state warning model, the probability of riding after arriving at the station is determined according to the riding density; The riding probability is determined as abnormal state warning information of the public bus.

[0023] This solution determines the current bus stop based on real-time operating data, avoiding prediction errors caused by positional deviations. Obtain and analyze bus monitoring information to determine the current number of passengers, ensuring that density calculations are based on actual on-board congestion. Analyze historical operating data to determine current riding patterns, enhancing the reliability and relevance of predictions. Based on the current bus stop, the current number of passengers, and current riding patterns, predict the passenger density after arrival at the station, quantify the space occupancy within the vehicle, and lay the foundation for predicting the probability of boarding. Based on the abnormal state warning model, determine the probability of boarding after arrival based on the passenger density, quantify the risk of passengers being able to board the bus, and convert the abstract density into a practical probability indicator. Determine the boarding probability as abnormal state warning information for the bus, ensuring that the abnormal state warning information can be directly displayed on the smart electronic bus stop without any filtering or rendering processing.

[0024] In a second aspect, the present application provides an abnormal bus status warning system based on a smart electronic bus stop sign, the system comprising: Data acquisition module, used to obtain real-time operation data and historical operation data of public buses; A model determination module, configured to determine an abnormal state warning model based on the historical operation data; An abnormality determination module, configured to determine abnormality warning information of the public transport vehicle based on the abnormality warning model and the real-time operation data; The information display module is used to send the abnormal state warning information to the smart electronic bus stop for display.

[0025] Optionally, when the model determination module determines the abnormal state warning model based on the historical operation data, it is used to: Analyzing the historical operating data to determine historical abnormal status information; Clustering the historical abnormal status information to obtain a statistical report; According to the statistical report, an abnormal state early warning model is obtained.

[0026] Optionally, when the model determination module obtains the abnormal state warning model according to the statistical report, it is used to: Analyze the statistical report to determine the occurrence pattern of abnormal conditions; Analyze the occurrence patterns of the abnormal state and determine the frequency of abnormal state occurrence in different time periods or different operating intervals; Determine the time period and road section where abnormal conditions are most likely to occur based on the frequency of occurrence of the abnormal conditions; According to the abnormal state high-incidence time period and the abnormal state high-incidence road section, a time series prediction algorithm is used to establish an abnormal state prediction model.

[0027] Optionally, when the model determination module determines the abnormal state high-incidence time period and abnormal state high-incidence road section based on the abnormal state occurrence frequency, it is used to: Set the time period division granularity and road section division rules; Determining a time-space unit according to the time period division granularity and the road section division rule; The sliding time window algorithm is used to count the abnormal occurrence rate of each time and space unit; When the abnormality occurrence rate of any time-space unit exceeds the preset occurrence threshold, it is marked as a high-incidence period of abnormal conditions or a high-incidence section of abnormal conditions.

[0028] Optionally, when the abnormality occurrence rate of any spatiotemporal unit exceeds a preset occurrence threshold, the model determination module is marked as a high-incidence period of abnormal conditions or a high-incidence section of abnormal conditions, and is used to: When the frequency of abnormality occurrence in any time period exceeds the preset occurrence threshold, the corresponding time period is determined to be a high-incidence period of abnormality; Alternatively, when the frequency of abnormality occurrence in any operating section exceeds a preset threshold, the corresponding operating section is determined to be a high-incidence section of abnormal conditions.

[0029] Optionally, when the abnormality determination module determines the abnormality warning information of the public transport vehicle based on the abnormality warning model and the real-time operation data, it is configured to: Comparing the real-time operating data with a preset normal operating threshold range; If the real-time operation data exceeds the preset normal operation threshold range, extracting real-time abnormal features; The real-time abnormal characteristics are input into the abnormal state warning model to obtain abnormal state warning information.

[0030] Optionally, when the information display module sends the abnormal state warning information to the smart electronic bus stop for display, it is used to: Obtain historical travel data to determine historical travel demand and historical travel modes; Analyze the historical travel demand and the historical travel mode to determine travel focus points; According to the travel focus points, the abnormal status warning information is extracted, the core notification content is determined, and the core notification content is sent to the smart electronic bus stop for display.

[0031] Optionally, when the abnormality determination module determines the abnormality warning information of the public transport vehicle based on the abnormality warning model and the real-time operation data, it is configured to: Determine the current status of the bus based on the real-time operation data; Based on the abnormal status warning model, predict the status update time; Determine the bus arrival time according to the current state of the bus and the state update time; The bus arrival time is determined as abnormal state warning information of the bus.

[0032] Optionally, when the abnormality determination module determines the abnormality warning information of the public transport vehicle based on the abnormality warning model and the real-time operation data, it is configured to: Determine the current bus stop based on the real-time operation data; Obtaining bus monitoring information, analyzing the bus monitoring information, and determining the current number of passengers; Analyze the historical operation data to determine current riding patterns; Predicting the passenger density after arrival at the bus stop based on the current bus stop, the current number of passengers, and the current riding patterns; Based on the abnormal state warning model, the probability of riding after arriving at the station is determined according to the riding density; The riding probability is determined as abnormal state warning information of the public bus. BRIEF DESCRIPTION OF THE DRAWINGS

[0033] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, a brief introduction will be given below to the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative labor.

[0034] Figure 1 A schematic diagram of an application scenario provided in one embodiment of the present application; Figure 2 A flowchart of a method for warning abnormal bus status based on a smart electronic bus stop provided in one embodiment of the present application; Figure 3 A schematic diagram of the structure of an abnormal bus status warning system based on a smart electronic bus stop sign provided in one embodiment of the present application. DETAILED DESCRIPTION

[0035] To make the purpose, technical solutions, and advantages of the embodiments of the present application clearer, the technical solutions in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of this application.

[0036] In this document, the term "and / or" simply describes a relationship between related objects, indicating that three possible relationships exist. For example, "A and / or B" can represent: A exists alone, A and B exist simultaneously, or B exists alone. Furthermore, the character " / " in this document, unless otherwise specified, generally indicates an "or" relationship between the related objects.

[0037] The embodiments of the present application are described in further detail below with reference to the accompanying drawings.

[0038] However, during the operation of buses, due to various uncontrollable factors such as road congestion, traffic accidents, vehicle breakdowns, bad weather, abnormal scheduling, etc., buses often fail to arrive on time.

[0039] However, the existing bus monitoring system is unable to monitor the real-time bus status, resulting in passengers waiting for a long time in vain, which greatly wears out their patience and seriously affects the public's evaluation and trust in bus services.

[0040] Based on this, the present application provides an abnormal bus status warning method and system based on a smart electronic bus stop, which obtains the real-time operation data and historical operation data of the bus; determines the abnormal state warning model based on the historical operation data; determines the abnormal state warning information of the bus based on the abnormal state warning model and the real-time operation data; and sends the abnormal state warning information to the smart electronic bus stop for display. The real-time operation data and historical operation data of the bus are obtained to ensure the availability and timeliness of the data and support the initiation of the warning process. Based on the historical operation data, the abnormal state warning model is determined to identify the high-incidence time periods and high-incidence road sections of the abnormal state, thereby realizing the transition from passive response to active prediction. Based on the abnormal state warning model, the abnormal state warning information of the bus is determined based on the real-time operation data, thereby providing accurate and practical warning information to assist passengers in decision-making. The abnormal state warning information is sent to the smart electronic bus stop for display to ensure that the information is clearly conveyed and enhance passengers' perception and trust in bus services.

[0041] Figure 1 This is a schematic diagram of an application scenario provided by this application. When warning of abnormal public transportation status is issued, the method provided by this application is applied.

[0042] Specifically, the method provided in this application is applied to any server, and the server interacts with the bus dispatching equipment, and obtains the real-time operation data and historical operation data of the bus in real time through the bus dispatching equipment. Based on the historical operation data, the abnormal state warning model is determined, and the high-incidence time periods and high-incidence road sections of abnormal states are identified, thereby realizing the transition from passive response to active prediction. Based on the abnormal state warning model, the abnormal state warning information of the bus is determined according to the real-time operation data, thereby providing accurate and practical warning information to assist passengers in decision-making. The abnormal state warning information is sent to the smart electronic bus stop for display to ensure that the information is clearly conveyed and enhance passengers' perception and trust in bus services.

[0043] For specific implementation methods, please refer to the following embodiments.

[0044] Figure 2 This is a flowchart of an abnormal bus status warning method and system based on a smart electronic bus stop provided in one embodiment of this application. The method of this embodiment can be applied to the server in the above scenario. Figure 2 As shown, the method includes: S201, obtaining real-time operation data and historical operation data of public buses; Real-time operation data can be dynamic data collected in real time during the operation of a public transportation vehicle.

[0045] The historical operation data can be the bus operation records accumulated in the past time period. This past time period can be taken based on experience or manually determined.

[0046] Specifically, the real-time operation data of buses is obtained in real time through the data interface of the bus dispatching equipment; then, based on the long-term collection and storage of the real-time operation data of the vehicles by the bus dispatching equipment, a historical database is formed according to the statistical regularity of abnormal states in different time and space units; then, historical operation data is extracted from the historical database.

[0047] S202. Determine an abnormal state warning model based on historical operation data; The abnormal state warning model may be a prediction model for outputting the probability of occurrence of abnormal states in different time and space units in the future.

[0048] Specifically, the historical operation data is cleaned and abnormal state labels such as periodicity, trend and suddenness are extracted; then, the time series prediction algorithm is used to train the abnormal state warning model based on the abnormal state labels in the historical operation data.

[0049] S203, based on the abnormal state warning model, according to the real-time operation data, determining abnormal state warning information of the public transport vehicle; The abnormal state warning information may be a prediction result generated by combining an abnormal state warning model with real-time operation data.

[0050] Specifically, the real-time operation data is input into the abnormal state warning model for inference calculation; then, the inference results of the abnormal state warning model are integrated to generate abnormal state warning information of the bus.

[0051] S204: Send abnormal status warning information to the smart electronic bus stop for display.

[0052] Smart electronic bus stops can be electronic display devices deployed at bus stops.

[0053] Specifically, according to the real-time location of the bus and the associated stations, the abnormal status warning information is routed to the corresponding smart electronic bus stop; when the smart electronic bus stop receives the information, it is visually displayed on the LCD screen.

[0054] This solution captures real-time and historical bus operating data, ensuring data availability and timeliness, and supporting the initiation of the early warning process. Based on historical operating data, an abnormality warning model is established to identify peak periods and road sections for abnormalities, enabling a shift from passive response to active prediction. Based on the abnormality warning model and real-time operating data, abnormality warning information for buses is determined, providing accurate and practical warning information to assist passengers in their decision-making. Abnormality warning information is then sent to smart electronic bus stops for display, ensuring clear communication and enhancing passengers' perception of and trust in public transportation services.

[0055] In some embodiments, historical operating data is analyzed to determine historical abnormal state information; the historical abnormal state information is clustered to obtain statistical reports; and an abnormal state early warning model is obtained based on the statistical reports.

[0056] The historical abnormal state information may be a set of abnormal event records determined by analyzing historical operation data.

[0057] The statistical report may be a structured data table generated by clustering historical abnormal status information.

[0058] Specifically, based on historical operating data, a rule matching algorithm is applied to check whether the data field meets the anomaly determination rules; when the historical operating data matches the anomaly conditions, the corresponding anomaly type and time-space information are marked to determine the historical anomaly status information. Then, a spatiotemporal clustering algorithm is applied to the historical anomaly status information, and the data is divided into time period units according to the occurrence timestamp; at the same time, the data is divided into road section units according to the occurrence location; then, the time period units and road section units are integrated to form a spatiotemporal unit combination; then, within each spatiotemporal unit, events of the same anomaly type are clustered and grouped, and the number of anomaly occurrences, frequency of occurrence, and average duration of each group are calculated; then, the grouped data is summarized and a statistical report is generated. Then, for each anomaly type, statistical indicators such as the frequency of anomaly occurrence and average duration in different spatiotemporal units are extracted from the statistical report; finally, based on the extracted statistical indicators, an abnormal state warning model is constructed. This solution analyzes historical operational data, identifies historical abnormal status information, eliminates invalid or noisy data, ensures data quality, and provides a standardized input dataset for clustering. Clustering historical abnormal status information generates statistical reports that reveal the distribution patterns of abnormal status across time and space. Based on these statistical reports, an abnormal status early warning model is derived, achieving model generalization and providing probabilistic output for proactive early warning.

[0059] In some embodiments, statistical reports are analyzed to determine the pattern of abnormal state occurrence; the pattern of abnormal state occurrence is analyzed to determine the frequency of abnormal state occurrence in different time periods or different operating intervals; based on the frequency of abnormal state occurrence, the time periods and road sections with high incidence of abnormal states are determined; based on the time periods and road sections with high incidence of abnormal states, a time series prediction algorithm is used to establish an abnormal state prediction model.

[0060] The law of occurrence of abnormal states can be the distribution pattern of abnormal states in the time and space dimensions.

[0061] Different time periods may be preset time division units.

[0062] Different operating intervals may be preset road segment division units.

[0063] The frequency of abnormal state occurrence may be the number of times the abnormal state occurs in different time periods or different operating intervals.

[0064] The abnormal state high-incidence period may be different time periods in which the frequency of occurrence of the abnormal state is higher than a preset frequency threshold.

[0065] The abnormal state high-incidence road section may be a different operating range where the frequency of abnormal state occurrence is higher than a preset frequency threshold.

[0066] Time series prediction algorithms can be prediction model algorithms based on sequence data.

[0067] The abnormal state prediction model may be a model for predicting the probability of an abnormal state occurring in the future.

[0068] Specifically, the report data is sorted and grouped by time and spatial sequence. The cumulative number of occurrences or average duration of abnormal states is then calculated to identify periodic or concentrated patterns, thereby determining the patterns of abnormal state occurrence. Subsequently, based on the patterns of abnormal state occurrence, the time dimension is divided into pre-set time periods, and the frequency of abnormal state occurrence within each time period is calculated. Simultaneously, the spatial dimension is divided into pre-set operating intervals, and the frequency of abnormal state occurrence within each operating interval is calculated. A frequency threshold is then set based on a pre-set rule theory. Several time periods are traversed to identify periods with abnormal state frequency exceeding the pre-set threshold as high-incidence periods. Several operating intervals are traversed to identify road sections with abnormal state frequency exceeding the pre-set threshold as high-incidence sections. Furthermore, historical abnormal state frequencies corresponding to high-incidence periods and sections are extracted as a training dataset. A time series prediction algorithm is applied, and the historical abnormal state frequencies are input to perform model training. First, the historical abnormal state frequencies are divided into a training set and a validation set. Then, the model parameters are adjusted to minimize prediction error. Finally, based on the training results, an abnormal state prediction model is constructed.

[0069] This solution analyzes statistical reports to identify patterns in abnormal state occurrences and reveal their concentration in different time periods and operating intervals. By analyzing patterns in abnormal state occurrences and determining the frequency of abnormal state occurrences in different time periods or operating intervals, we provide precise numerical evidence for identifying high-incidence areas. Based on the frequency of abnormal state occurrences, we identify peak abnormality time periods and road sections, focusing on the training data required for the abnormal state prediction model to ensure that the model training process is optimized for areas most prone to abnormalities. Based on peak abnormality time periods and road sections, we use a time series prediction algorithm to establish an abnormal state prediction model, giving the model predictive capabilities and providing core support for proactive early warning.

[0070] In some embodiments, the time period division granularity and the road section division rules are set; the time space unit is determined according to the time period division granularity and the road section division rules; the sliding time window algorithm is used to count the abnormality occurrence rate of each time space unit; when the abnormality occurrence rate of any time space unit exceeds the preset occurrence threshold, it is marked as a high-incidence time period of abnormal state or a high-incidence road section of abnormal state.

[0071] The time period division granularity may be a division unit of the time dimension.

[0072] The road segment division rule may be a division method of a spatial dimension.

[0073] The spatiotemporal unit can be an analysis unit composed of a time period and a road section.

[0074] A sliding time window algorithm may be an algorithm that moves a fixed-size window over a time series to calculate a statistical value.

[0075] The abnormal occurrence rate can be the frequency at which abnormal states occur in different time and space units.

[0076] The preset occurrence threshold value may be a critical value for determining whether the abnormality occurrence rate is high, and is pre-stored in the server and called when used.

[0077] Specifically, the user input sets the time dimension's division unit, known as the time period granularity. Simultaneously, the spatial dimension's division method, known as the segment division rule, is set using a pre-set route division strategy. Subsequently, the time axis is segmented into several time period units based on the time period granularity, while the spatial axis is segmented into several operating interval units based on the segment division rule. Each time period unit and each operating interval unit are then combined to form a spatiotemporal unit. Furthermore, a sliding time window algorithm is employed: first, the sliding time window is initialized, and the window size and sliding step are set. The window is then moved to traverse the frequency of abnormal state occurrences. For each spatiotemporal unit, the number of abnormal state occurrences within the windowed time period is counted. These occurrences are then integrated to generate an abnormality incidence rate for each spatiotemporal unit. Subsequently, based on a preset occurrence threshold read from a configuration file, the abnormality incidence rate of any spatiotemporal unit is compared with the preset threshold. If the abnormality incidence rate exceeds the preset threshold, the corresponding time period unit is marked as a high-incidence period for abnormal states, and the corresponding operating interval unit is also marked as a high-incidence section for abnormal states.

[0078] This solution sets the granularity of time period division and the rules for road section division, establishes the division standards for time axis and spatial area, and ensures that multiple analyses are conducted based on consistent rules. Based on the granularity of time period division and the rules for road section division, the time and space units are determined, and the time and space dimensions are converted into an operational unit structure. A sliding time window algorithm is used to count the abnormality incidence rate of each time and space unit, quantify the density of abnormal states in each time and space unit, and provide a numerical basis for comparison and labeling. When the abnormality incidence rate of any time and space unit exceeds the preset occurrence threshold, it is marked as a high-incidence time period or high-incidence road section of abnormal states to avoid repeated output and provide a basis for training data screening for the abnormal state prediction model.

[0079] In some embodiments, when the frequency of abnormal occurrence in any time period exceeds a preset occurrence threshold, the corresponding time period is determined to be a high-incidence period of abnormal conditions; or, when the frequency of abnormal occurrence in any operating interval exceeds a preset occurrence threshold, the corresponding operating interval is determined to be a high-incidence section of abnormal conditions.

[0080] The abnormal occurrence frequency may be the frequency at which abnormal states occur in different time and space units.

[0081] The corresponding time period may be a time period unit associated in a marking operation.

[0082] The operating range can be a spatial unit defined according to the road segment division rules.

[0083] The corresponding operating interval may be an operating interval unit associated in the marking operation.

[0084] Specifically, the abnormality frequency of the spatiotemporal unit is read, and a comparison function is called to compare the abnormality frequency with a preset threshold. If the comparison result shows that the abnormality frequency within any time period exceeds the preset threshold, the associated time period unit identifier is extracted from the current spatiotemporal unit. Then, the time period unit corresponding to the identifier is searched in the time period unit dataset, and the time period unit's flag attribute is set to a high-incidence abnormality period. Alternatively, if the comparison result shows that the abnormality frequency within any operating interval exceeds the preset threshold, the associated operating interval unit identifier is extracted from the current spatiotemporal unit. Then, the operating interval unit corresponding to the identifier is searched in the operating interval unit dataset, and the operating interval unit's flag attribute is set to a high-incidence abnormality section.

[0085] With this solution, when the frequency of anomalies within any time period exceeds a preset threshold, the corresponding time period is determined to be a high-incidence period for anomalies, and the time period unit is identified as a time period with a high probability of anomalies occurring, facilitating the early warning model's identification and utilization of high-incidence periods. Alternatively, when the frequency of anomalies within any operating interval exceeds a preset threshold, the corresponding operating interval is determined to be a high-incidence section for anomalies, and the operating interval unit is identified as a section with a high probability of anomalies occurring, facilitating the early warning model's identification and utilization of high-incidence sections.

[0086] In some embodiments, the real-time operating data is compared with a preset normal operating threshold range; if the real-time operating data exceeds the preset normal operating threshold range, the real-time abnormal features are extracted; the real-time abnormal features are input into the abnormal state warning model to obtain abnormal state warning information.

[0087] The preset normal operation threshold range can be a preset threshold range for determining whether the real-time operation data of a public transportation vehicle is normal, and is pre-stored in the server and called when used.

[0088] The real-time abnormal feature may be a feature extracted from real-time operation data that exceeds a preset normal operation threshold range.

[0089] Specifically, a preset normal operating threshold range is established based on statistical analysis of historical operating data. The upper and lower limits of the preset normal operating threshold range are determined based on the calculated data distribution under normal conditions. Then, a numerical comparison is performed between the real-time operating data and the preset normal operating threshold range. If the real-time operating data exceeds the preset normal operating threshold range, the type and degree of abnormality are identified. Then, relevant feature values ​​are calculated based on the real-time operating data. Furthermore, the relevant feature values ​​are combined to form real-time abnormality features. The real-time abnormality features are input into a pre-trained abnormality warning model. Then, a predictive calculation is performed to output abnormality warning information.

[0090] This solution compares real-time operating data with pre-set normal operating thresholds to identify potential abnormalities, triggering processing logic. If the real-time operating data exceeds the pre-set normal operating threshold, real-time abnormality features are extracted to quantify and convey the type and severity of the abnormality. These features are then fed into an abnormality warning model to generate abnormality warning information, generating predictive, user-oriented warning content for display on smart electronic bus stops.

[0091] In some embodiments, historical travel data is obtained to determine historical travel demands and historical travel methods; historical travel demands and historical travel methods are analyzed to determine travel focus points; based on travel focus points, abnormal status warning information is extracted, core notification content is determined, and the core notification content is sent to the smart electronic bus stop for display.

[0092] Historical travel data may be passenger's past travel record data.

[0093] The historical travel demand may be the type of travel demand of the passenger in the past.

[0094] Historical travel modes can be the types of transportation modes that passengers have used in the past.

[0095] Travel concerns may be travel-related aspects that passengers are concerned about.

[0096] The core notification content can be targeted information extracted from abnormal status warning information for passenger decision-making.

[0097] Specifically, historical passenger travel data is retrieved through a database query interface. A data classification algorithm is then used to process this historical travel data, using pattern recognition to categorize travel behavior into different demand and mode categories, and to determine historical travel demand and historical travel modes. Furthermore, a rule engine constructed through expert knowledge or historical data analysis matches historical travel demand and historical travel modes to corresponding focus types, generating travel focus points. Subsequently, based on abnormal status warning information, a filtering algorithm is used to select a subset of warning information related to travel focus points. This subset of warning information is then integrated to generate concise, targeted core notification content. This core notification content is then transmitted to the receiving module of the smart electronic bus stop using a wireless / wired network transmission protocol. The core notification content is then parsed by the smart electronic bus stop's display controller. Finally, the core notification content is rendered into a readable interface and displayed on the smart electronic bus stop's screen.

[0098] This solution captures historical travel data, identifies historical travel demand and travel patterns, avoids subjective assumptions, and lays the data foundation for personalized information display. This solution analyzes historical travel demand and patterns to identify travel focus points, ensuring that information extraction focuses on the areas of greatest concern to passengers, avoiding interference from irrelevant or redundant information and improving the relevance and pertinence of information. Based on travel focus points, abnormal status warning information is extracted, and core notification content is determined. This core notification content is then sent to the smart electronic bus stop for display, ensuring user-friendly and practical information display, improving passenger decision-making efficiency and experience, while also avoiding information overload and misunderstandings.

[0099] In some embodiments, the current state of the bus is determined based on real-time operating data; the state update time is predicted based on an abnormal state warning model; the bus arrival time is determined based on the current state of the bus and the state update time; and the bus arrival time is determined as abnormal state warning information of the bus vehicle.

[0100] The current state of the bus may be the current position, current speed and current running direction of the bus.

[0101] The status update moment may be a future time point when the status of the bus changes.

[0102] The bus arrival time may be the estimated time when the bus arrives at the target station.

[0103] Specifically, a data processing module analyzes real-time operating data to calculate the bus's current position, speed, and direction of travel, determining its current status. This status is then fed into a pre-established abnormal status warning model. A time series prediction algorithm is then used to output the predicted status update time. Furthermore, the remaining distance to the target stop is calculated based on the bus's current status, and the base arrival time is derived based on the current speed. This base arrival time is then adjusted based on the status update time to ultimately determine the bus's arrival time. Finally, the bus's arrival time is stored as a warning information entity through data assignment and marked as abnormal status warning information for the bus.

[0104] This solution determines the current bus status based on real-time operating data, ensuring that the early warning process begins with an accurate vehicle status assessment. Based on the abnormal status early warning model, the status update time is predicted, providing a critical time reference for calculating bus arrival times and ensuring forward-looking early warning information. The bus arrival time is determined based on the current bus status and status update time, ensuring that the information is practical and supports passenger decision-making. The bus arrival time is used as the abnormal status early warning information for the bus, ensuring that the warning information focuses on the core content of passenger concern.

[0105] In some embodiments, the current bus stop is determined based on real-time operation data; bus monitoring information is obtained and analyzed to determine the current number of passengers; historical operation data is analyzed to determine the current riding pattern; based on the current bus stop, the current number of passengers and the current riding pattern, the passenger density after arrival is predicted; based on the abnormal state warning model, the passenger probability after arrival is determined based on the passenger density; and the passenger probability is determined as abnormal state warning information of the bus.

[0106] The current bus stop may be the current location of the bus stop.

[0107] The bus monitoring information may be real-time data obtained through on-board monitoring equipment.

[0108] The current number of passengers may be the number of passengers currently in the bus.

[0109] The current riding pattern may be a typical riding pattern at the current time period and station.

[0110] Ride density can be the expected level of crowding when a bus arrives at a target stop.

[0111] The boarding probability can be the probability value of a passenger successfully getting on the bus at the target station.

[0112] Specifically, using GIS technology, bus route and stop coordinates are digitized and stored in a pre-set route topology database. GPS location coordinates in real-time operational data are then matched against the pre-set route topology database to calculate the distance from the current location to each stop. Based on the principle of closest distance, the current bus stop corresponding to the bus's current location is determined. Real-time bus monitoring information is then acquired from onboard monitoring equipment. Image processing algorithms are used to detect people and count the number of passengers currently on the bus. Furthermore, based on the current time and bus stop extracted from the real-time operational data, historical operational data is retrieved for passenger counts at the same time period and stop. Statistical analysis is then performed to extract typical passenger patterns at the bus stop during the current time period, i.e., the current passenger pattern. Subsequently, a linear regression model is constructed using the linear relationship between variables in the historical operational data, combining the current bus stop, the current number of passengers, and the current passenger pattern, to predict the passenger density of the bus upon arrival at the target stop. The predicted passenger density is then fed into the abnormal state warning model, where the probability of passengers arriving at the stop is determined through feature mapping and probability calculation. Then, the riding probability value is stored as a warning information entity through data assignment; finally, the warning information entity is marked as abnormal state warning information of the public bus.

[0113] This solution determines the current bus stop based on real-time operating data, avoiding prediction errors caused by positional deviations. Obtain and analyze bus monitoring information to determine the current number of passengers, ensuring that density calculations are based on actual on-board congestion. Analyze historical operating data to determine current riding patterns, enhancing the reliability and relevance of predictions. Based on the current bus stop, the current number of passengers, and current riding patterns, predict the passenger density after arrival at the station, quantify the space occupancy within the vehicle, and lay the foundation for predicting the probability of boarding. Based on the abnormal state warning model, determine the probability of boarding after arrival based on the passenger density, quantify the risk of passengers being able to board the bus, and convert the abstract density into a practical probability indicator. Determine the boarding probability as abnormal state warning information for the bus, ensuring that the abnormal state warning information can be directly displayed on the smart electronic bus stop without any filtering or rendering processing.

[0114] Figure 3 A schematic diagram of the structure of an abnormal bus status warning system based on a smart electronic bus stop provided in one embodiment of the present application is shown as follows: Figure 3 As shown, the abnormal public transportation status warning system 300 based on the smart electronic bus stop in this embodiment includes: a data acquisition module 301, a model determination module 302, an abnormality determination module 303, and an information display module 304.

[0115] The data acquisition module 301 is used to obtain the real-time operation data and historical operation data of the public transportation vehicle; A model determination module 302 is configured to determine an abnormal state warning model based on the historical operation data; An abnormality determination module 303 is used to determine abnormality warning information of the public transport vehicle based on the abnormality warning model and the real-time operation data; The information display module 304 is used to send the abnormal state warning information to the smart electronic bus stop for display.

[0116] Optionally, when the model determination module 302 determines the abnormal state warning model based on the historical operation data, it is used to: Analyzing the historical operating data to determine historical abnormal status information; Clustering the historical abnormal status information to obtain a statistical report; According to the statistical report, an abnormal state early warning model is obtained.

[0117] Optionally, when the model determination module 302 obtains the abnormal state warning model according to the statistical report, it is used to: Analyze the statistical report to determine the occurrence pattern of abnormal conditions; Analyze the occurrence patterns of the abnormal state and determine the frequency of abnormal state occurrence in different time periods or different operating intervals; Determine the time period and road section where abnormal conditions are most likely to occur based on the frequency of occurrence of the abnormal conditions; According to the abnormal state high-incidence time period and the abnormal state high-incidence road section, a time series prediction algorithm is used to establish an abnormal state prediction model.

[0118] Optionally, when the model determination module 302 determines the abnormal state high-incidence time period and abnormal state high-incidence road section based on the abnormal state occurrence frequency, it is configured to: Set the time period division granularity and road section division rules; Determining a time-space unit according to the time period division granularity and the road section division rule; The sliding time window algorithm is used to count the abnormal occurrence rate of each time and space unit; When the abnormality occurrence rate of any time-space unit exceeds the preset occurrence threshold, it is marked as a high-incidence period of abnormal conditions or a high-incidence section of abnormal conditions.

[0119] Optionally, when the abnormality occurrence rate of any spatiotemporal unit exceeds a preset occurrence threshold, the model determination module 302 is marked as a high-incidence period of abnormality or a high-incidence section of abnormality, and is configured to: When the frequency of abnormality occurrence in any time period exceeds the preset occurrence threshold, the corresponding time period is determined to be a high-incidence period of abnormality; Alternatively, when the frequency of abnormality occurrence in any operating section exceeds a preset threshold, the corresponding operating section is determined to be a high-incidence section of abnormality.

[0120] Optionally, when the abnormality determination module 303 determines the abnormality warning information of the public transport vehicle based on the abnormality warning model and the real-time operation data, it is configured to: Comparing the real-time operating data with a preset normal operating threshold range; If the real-time operation data exceeds the preset normal operation threshold range, extracting real-time abnormal features; The real-time abnormal characteristics are input into the abnormal state warning model to obtain abnormal state warning information.

[0121] Optionally, when the information display module 304 sends the abnormal state warning information to the smart electronic bus stop for display, it is used to: Obtain historical travel data to determine historical travel demand and historical travel modes; Analyze the historical travel demand and the historical travel mode to determine travel focus points; According to the travel focus points, the abnormal status warning information is extracted, the core notification content is determined, and the core notification content is sent to the smart electronic bus stop for display.

[0122] Optionally, when the abnormality determination module 303 determines the abnormality warning information of the public transport vehicle based on the abnormality warning model and the real-time operation data, it is configured to: Determine the current status of the bus based on the real-time operation data; Based on the abnormal status warning model, predict the status update time; Determine the bus arrival time according to the current state of the bus and the state update time; The bus arrival time is determined as abnormal state warning information of the bus.

[0123] Optionally, when the abnormality determination module 303 determines the abnormality warning information of the public transport vehicle based on the abnormality warning model and the real-time operation data, it is configured to: Determine the current bus stop based on the real-time operation data; Obtaining bus monitoring information, analyzing the bus monitoring information, and determining the current number of passengers; Analyze the historical operation data to determine current riding patterns; Predicting the passenger density after arrival at the bus stop based on the current bus stop, the current number of passengers, and the current riding patterns; Based on the abnormal state warning model, the probability of riding after arriving at the station is determined according to the riding density; The riding probability is determined as abnormal state warning information of the public bus.

[0124] The system of this embodiment can be used to execute the method of any of the above embodiments. Its implementation principles and technical effects are similar and will not be described in detail here.

Claims

1. A method for warning abnormal bus status based on smart electronic bus stop signs, characterized in that: include: Obtain real-time and historical operation data of public buses; Determining an abnormal state early warning model based on the historical operation data; Determining abnormal state warning information of the public bus based on the abnormal state warning model and the real-time operation data; The abnormal status warning information is sent to the smart electronic bus stop for display.

2. The method according to claim 1, characterized in that Determining an abnormal state warning model based on the historical operation data includes: Analyzing the historical operating data to determine historical abnormal status information; Clustering the historical abnormal status information to obtain a statistical report; According to the statistical report, an abnormal state early warning model is obtained.

3. The method according to claim 2, characterized in that The abnormal state early warning model is obtained according to the statistical report, including: Analyze the statistical report to determine the occurrence pattern of abnormal conditions; Analyze the occurrence patterns of the abnormal state and determine the frequency of abnormal state occurrence in different time periods or different operating intervals; Determine the time period and road section where abnormal conditions are most likely to occur based on the frequency of occurrence of the abnormal conditions; According to the abnormal state high-incidence time period and the abnormal state high-incidence road section, a time series prediction algorithm is used to establish an abnormal state prediction model.

4. The method according to claim 3, characterized in that The step of determining a high-incidence time period and a high-incidence road section of the abnormal state according to the frequency of occurrence of the abnormal state includes: Set the time period division granularity and road section division rules; Determining a time-space unit according to the time period division granularity and the road section division rule; The sliding time window algorithm is used to count the abnormal occurrence rate of each time and space unit; When the abnormality occurrence rate of any time-space unit exceeds the preset occurrence threshold, it is marked as a high-incidence period of abnormal conditions or a high-incidence section of abnormal conditions.

5. The method according to claim 4, characterized in that When the abnormality occurrence rate of any spatiotemporal unit exceeds a preset occurrence threshold, it is marked as a high-incidence period of abnormality or a high-incidence section of abnormality, including: When the frequency of abnormality occurrence in any time period exceeds the preset occurrence threshold, the corresponding time period is determined to be a high-incidence period of abnormality; Alternatively, when the frequency of abnormality occurrence in any operating section exceeds a preset threshold, the corresponding operating section is determined to be a high-incidence section of abnormal conditions.

6. The method according to claim 1, characterized in that The determining of abnormal state warning information of the public transport vehicle based on the abnormal state warning model and the real-time operation data includes: Comparing the real-time operating data with a preset normal operating threshold range; If the real-time operation data exceeds the preset normal operation threshold range, extracting real-time abnormal features; The real-time abnormal characteristics are input into the abnormal state warning model to obtain abnormal state warning information.

7. The method according to claim 1, characterized in that The step of sending the abnormal status warning information to the smart electronic bus stop for display includes: Obtain historical travel data to determine historical travel demand and historical travel modes; Analyze the historical travel demand and the historical travel mode to determine travel focus points; According to the travel focus points, the abnormal status warning information is extracted, the core notification content is determined, and the core notification content is sent to the smart electronic bus stop for display.

8. The method according to claim 1, characterized in that The determining of abnormal state warning information of the public transport vehicle based on the abnormal state warning model and the real-time operation data includes: Determine the current status of the bus based on the real-time operation data; Based on the abnormal status warning model, predict the status update time; Determine the bus arrival time according to the current state of the bus and the state update time; The bus arrival time is determined as abnormal state warning information of the bus.

9. The method according to claim 1, characterized in that The determining of abnormal state warning information of the public transport vehicle based on the abnormal state warning model and the real-time operation data includes: Determine the current bus stop based on the real-time operation data; Obtaining bus monitoring information, analyzing the bus monitoring information, and determining the current number of passengers; Analyze the historical operation data to determine current riding patterns; Predicting the passenger density after arrival at the bus stop based on the current bus stop, the current number of passengers, and the current riding patterns; Based on the abnormal state warning model, the probability of riding after arriving at the station is determined according to the riding density; The riding probability is determined as abnormal state warning information of the public bus.

10. An abnormal bus status warning system based on a smart electronic bus stop, applied to the method according to any one of claims 1 to 9, characterized in that: include: Data acquisition module, used to obtain real-time operation data and historical operation data of public buses; A model determination module, configured to determine an abnormal state warning model based on the historical operation data; An abnormality determination module, configured to determine abnormality warning information of the public transport vehicle based on the abnormality warning model and the real-time operation data; The information display module is used to send the abnormal state warning information to the smart electronic bus stop for display.

Images