ETC (Electronic Toll Collection) equipment state monitoring and early warning system and method based on swan monk system

Abstract

The invention provides an ETC (Electronic Toll Collection) equipment state monitoring and early warning system and method based on a swan gap system, and the system architecture is a multi-layer system which comprises a sensing layer, a transmission layer, a cloud computing layer and an application layer. Wherein the sensing layer is connected with sensors of a plurality of ETC devices. The transmission layer is provided with a distributed device network and is connected with the sensing layer and the cloud computing layer. The cloud computing layer is provided with a cloud server and is provided with a machine learning model. The application layer is provided with multiple types of terminals which are all connected with the cloud computing layer and used for receiving and transmitting data. According to the invention, by constructing an ETC equipment Internet of Things multilayer architecture system based on the swan Mongolia system and integrating related equipment resources, abnormal conditions of equipment can be quickly and accurately found, possible fault types are predicted based on the application of a machine learning model, information is automatically published through multiple channels, the accuracy of problem judgment and the comprehensive processing efficiency are greatly improved, and the system is suitable for large-scale popularization and application. Therefore, the safety, the stability and the passing efficiency of traffic are guaranteed.

Description

Technical Field

[0001] The present invention relates to the field of transportation technology, and specifically to an ETC equipment status monitoring and early warning system and method based on the Hongmeng system. Background Art

[0002] ETC (Electronic Toll Collection) is an electronic toll collection system, and its equipment mainly consists of an on-board unit (OBU) and a roadside unit (RSU). The on-board unit is installed on the vehicle and communicates wirelessly with the roadside unit to enable the vehicle to automatically complete the toll collection without stopping.

[0003] Among them, since the on-board unit is located inside the vehicle, it is less affected by the external environment and generally focuses on its communication and safety issues. As for the roadside unit and other ETC equipment, since their main working scenes are outdoor work, they are easily affected by climate and emergencies, such as strong winds, rain, hail, and exposure to the sun. In addition, some areas also have extreme conditions such as high and low temperature cycles, and emergencies such as flying rocks, falling goods, or collisions with hard objects such as out-of-control vehicles; therefore, various ETC equipment on the road side is easily damaged, and once the ETC equipment is damaged, it will lead to a serious decline in traffic efficiency, especially during busy traffic.

[0004] Therefore, in order to ensure the stable operation of the system, if the status of the ETC equipment can be monitored in real time and early warnings can be issued for some potential faults, potential safety hazards can be discovered in time and resolved through maintenance, ensuring the stable operation of the system, thereby providing users with a more convenient and efficient travel experience.

[0005] Although the current ETC equipment has networking capabilities and corresponding detection methods, there are still some problems, as follows:

[0006] Single monitoring method: Existing ETC equipment monitoring technology mainly relies on the equipment's own status indicator light or simple communication protocol feedback. This monitoring method can only provide limited information and cannot fully understand the equipment's operating status.

[0007] Early warning is not timely: When the existing technology finds equipment failure, it often requires manual intervention to issue early warning information. This method has a certain delay, which may cause the failure to worsen and affect the normal operation of the system.

[0008] Lack of data analysis capabilities: Existing ETC equipment monitoring systems can usually only collect and store equipment status data, but lack the ability to conduct in-depth analysis and mining of these data. This makes it impossible for the system to detect potential failure risks in advance, and it is also unable to provide effective decision support for equipment maintenance and management.

[0009] Poor compatibility: ETC devices produced by different manufacturers have differences in communication protocols, interface standards, etc., which makes it difficult for the existing monitoring system to achieve compatibility with different devices. This increases the construction and maintenance costs of the system and also limits the promotion and application of the system.

[0010] In summary, it is also necessary to integrate the existing ETC equipment at the system level and establish a unified monitoring and early warning system to quickly and efficiently detect abnormal situations and potential problems, so as to maintain the equipment in advance and improve the overall operational stability and traffic efficiency. Summary of the invention

[0011] In view of the problems existing in the prior art, the present invention provides an ETC equipment status monitoring and early warning system.

[0012] To achieve the above purpose, the technical solution adopted by the present invention is as follows:

[0013] On the one hand, the present invention provides an ETC equipment status monitoring and early warning system based on the Hongmeng system, and the system architecture is a multi-layer system, including a perception layer, a transmission layer, a cloud computing layer, and an application layer;

[0014] The perception layer is provided with a plurality of data receiving nodes, each of which is connected to a plurality of sensors for real-time or periodic collection of working parameters and environmental parameters of the ETC equipment;

[0015] The transport layer is constructed with a distributed device network based on Hongmeng distributed soft bus technology; the transport layer is connected to the perception layer;

[0016] The cloud computing layer is provided with a cloud server and is connected to at least one transport layer; the cloud server is deployed with a machine learning model for ETC equipment failure warning;

[0017] The application layer is provided with multiple terminals of multiple types, and all of them are connected to the cloud computing layer for sending and receiving data.

[0018] Optionally, at least some sensors are installed on the ETC device, including temperature sensors, humidity sensors, vibration sensors, posture sensors and electromagnetic sensors.

[0019] Optionally, the sensor includes an environmental sensor, and the environmental sensor is installed on or around the ETC device;

[0020] The environmental sensors include meteorological sensors and light sensors.

[0021] Optionally, the distributed device network includes a gateway device, toll station supporting equipment and a roadside unit;

[0022] The toll station supporting equipment and the roadside unit are interconnected through the gateway device to form the distributed device network.

[0023] Optionally, the multiple types of terminals include dedicated communication terminals, fixed terminals and portable mobile terminals.

[0024] On the other hand, the present invention also provides an ETC equipment status monitoring and early warning method based on the Hongmeng system, which is characterized by:

[0025] The steps include:

[0026] Collect the working parameters and environmental parameters of ETC equipment through the perception layer;

[0027] Packing the working parameters and environmental parameters to obtain a monitoring data packet of the ETC device, and uploading the monitoring data packet to the cloud computing layer through the transport layer;

[0028] In the cloud computing layer, the monitoring data packet is input into a machine learning model and a prediction result is output;

[0029] When the prediction result is an abnormal situation, the prediction result and the corresponding monitoring data packet are sent to the application layer.

[0030] Optionally, the working parameters of the ETC device include temperature, humidity, vibration, attitude angle and electromagnetic intensity inside the device;

[0031] The environmental parameters of the ETC device include temperature, humidity, wind speed and light intensity outside the device.

[0032] Optionally, at least one distributed device network is constructed in the transport layer, including a gateway device, toll station supporting equipment and a roadside unit;

[0033] The monitoring data packet is configured with a first timestamp in the perception layer;

[0034] Filtering and compressing data of a distributed device network in a time period adjacent to the first timestamp in the transport layer, and obtaining a supplementary data packet;

[0035] Embedding the supplementary data packet into the monitoring data packet, and obtaining an encapsulated monitoring data packet;

[0036] The transport layer transmits the encapsulated monitoring data packet to the cloud computing layer.

[0037] Optionally, the method for constructing the machine learning model includes the following steps:

[0038] Collecting fault information of the ETC device, the fault information including the fault type and the working parameters and environmental parameters of the ETC device in the corresponding period, and data of each device in the distributed device network;

[0039] Clustering and classifying the fault information, and then forming a training data set with fault information of the same type;

[0040] Training the corresponding neural network model using the training data set;

[0041] The machine learning model is obtained by combining multiple neural network models.

[0042] Optionally, the cluster classification includes the following steps:

[0043] Performing data cleaning on the fault information and assigning extreme values ​​to missing values ​​or removed values;

[0044] The cleaned fault information is normalized and classified based on a clustering algorithm to obtain a classification label for the fault information.

[0045] Compared with the prior art, the present invention has the following beneficial effects:

[0046] The present invention constructs a multi-layer architecture system of the Internet of Things for ETC equipment based on the Hongmeng system and integrates relevant equipment resources. It can quickly and accurately discover abnormal conditions of equipment, predict possible fault types based on the application of machine learning models, and automatically publish information through multiple channels to notify personnel to promptly perform necessary abnormal condition confirmation and preventive maintenance, greatly improving the accuracy of problem judgment and the overall processing efficiency, thereby ensuring traffic safety, stability and traffic efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0047] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative labor.

[0048] Figure 1 It is a system diagram in a specific embodiment of the present invention;

[0049] Figure 2 It is a flow chart of the method in a specific embodiment of the present invention. DETAILED DESCRIPTION

[0050] In order to make the purpose, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0051] It should be noted that similar reference numerals and letters denote similar items in the following drawings, and therefore, once an item is defined in one drawing, further definition and explanation thereof is not required in subsequent drawings.

[0052] In the description of the present invention, “plurality” means two or more than two, unless otherwise clearly and specifically defined.

[0053] It is worth noting that the methods used in the present invention are all conventional methods unless otherwise specified; the raw materials and devices used are all conventional commercially available products, and their sources are not specifically limited unless otherwise specified.

[0054] On the one hand, this embodiment provides an ETC equipment status monitoring and early warning system based on the Hongmeng system, such as Figure 1 As shown, its system architecture is a multi-layer system based on the Hongmeng system, including the perception layer, transmission layer, cloud computing layer, and application layer.

[0055] Among them, each perception layer for a single ETC device is equipped with multiple data receiving nodes, which are responsible for connecting one or more sensors, such as sensor 111-sensor 11n in perception layer 1, sensor 211-sensor 21m in perception layer 2, and n and m are natural numbers greater than 1. The sensors collect the working parameters and environmental parameters of the ETC equipment (such as ETC equipment 11 and ETC equipment 21 corresponding to each sensor) in real time or periodically. Optionally, at least some sensors are installed on the ETC equipment, including temperature sensors, humidity sensors, vibration sensors, posture sensors and electromagnetic sensors. Furthermore, the sensor includes an environmental sensor, and the environmental sensor is installed on or around the ETC equipment; the environmental sensor includes a meteorological sensor and a light sensor.

[0056] Each transport layer is connected to at least one perception layer, the core of which is the gateway device. A distributed device network is built based on Hongmeng distributed soft bus technology, which includes the aforementioned gateway devices, toll station supporting equipment and adjacent roadside units. The toll station supporting equipment and the roadside unit are interconnected through the gateway device. Among them, the toll station supporting equipment includes surveillance cameras, lane controllers, vehicle detectors, etc.

[0057] Among them, the data communication schemes of the transport layer are divided into three categories, specifically:

[0058] The perception layer is connected to the transport layer: The sensor transmits the collected data to the gateway device of the transport layer through the data receiving node. HarmonyOS's short-range communication technology, such as Bluetooth Mesh or Wi-Fi Direct, is used to ensure the stability and low power consumption of data transmission.

[0059] Internal connection of the transport layer: Toll station supporting equipment and roadside units form a distributed device network through gateway devices. Hongmeng's distributed soft bus technology is based on a distributed communication protocol to achieve self-discovery, self-organizing networking and efficient communication between devices. For example, the IPv6-based low-power wireless personal area network technology (6LoWPAN) is combined with Hongmeng's distributed data management to achieve low-power, long-distance data transmission and sharing.

[0060] The transport layer is connected to the cloud computing layer: The gateway device transmits the processed data packets to the cloud server of the cloud computing layer through a wired or wireless network (such as 4G, 5G). Data transmission uses a secure encrypted communication protocol, such as the secure transmission protocol based on Hongmeng, to ensure the security and integrity of data during transmission.

[0061] The cloud computing layer is equipped with a cloud server and is connected to at least one transport layer. The cloud server is deployed with a machine learning model for ETC equipment failure warning. The cloud server is a Hongmeng cloud server, which makes full use of the distributed capabilities of the Hongmeng system to achieve efficient data storage and processing, and maintain consistency with other layer systems, thereby improving data communication and processing efficiency.

[0062] The application layer is provided with multiple types of multiple terminals, all of which are connected to the cloud computing layer for sending and receiving data. Optionally, multiple types of multiple terminals (such as terminal 1-terminal z, z is a natural number greater than 1) include dedicated communication terminals (such as customized communicators based on the Hongmeng operating system), fixed terminals (such as desktop computers) and portable mobile terminals (such as mobile phones, tablets). Since the Hongmeng system has good compatibility, especially high compatibility with the Android system, the above-mentioned terminals can be based on the Hongmeng operating system, or they can be developed by other systems and installed with supporting APPs to run, and they can achieve seamless connection and collaborative work with the system of this embodiment.

[0063] Connection between the cloud computing layer and the application layer: The cloud server is connected to multiple types of terminals through the Internet. A distributed service framework based on Hongmeng can be used to achieve two-way data transmission. The terminal can obtain the status information and warning information of the ETC device by sending requests or passively receiving notifications, and can also send instructions and feedback information to the cloud server.

[0064] On the other hand, Figure 2 As shown, this embodiment further provides an ETC equipment status monitoring and early warning method based on the Hongmeng system based on the above system, which includes the following steps:

[0065] The working parameters and environmental parameters of the ETC equipment are collected through the perception layer. Optionally, based on the configuration of the aforementioned sensors, the working parameters of the ETC equipment include the temperature, humidity, vibration, attitude angle and electromagnetic intensity inside the equipment; the environmental parameters of the ETC equipment include the temperature, humidity, wind speed and light intensity outside the equipment. Sensor data collection: Data collection is performed using a timed trigger. With the help of Hongmeng's lightweight device management framework, efficient control and data collection of sensors can be achieved. For example, parameters such as temperature, humidity, vibration (amplitude and frequency), attitude angle and electromagnetic intensity inside the device are collected every 5 minutes.

[0066] Environmental sensors collect data based on actual needs, either by event triggering or timing triggering. HarmonyOS's sensor fusion technology is used to fuse data from different types of environmental sensors to improve data accuracy and reliability. For example, meteorological sensors trigger data collection when changes in ambient temperature, humidity, wind speed and other parameters exceed a certain threshold, and light sensors collect ambient light intensity every 10 minutes.

[0067] Finally, in the perception layer, the data collected by the sensor is converted by A / D and transmitted to the data receiving node in the form of a digital signal. The stable communication between the sensor and the data receiving node is ensured through the Hongmeng device driver framework. The data receiving node pre-processes the data collected by the sensor, including data filtering, denoising, calibration and other operations. The Hongmeng data analysis framework is used to achieve rapid processing and analysis of data. For example, for the data collected by the vibration sensor, the median filtering algorithm is used to remove high-frequency noise, and then the data is calibrated through the calibration algorithm to ensure the accuracy of the vibration data.

[0068] The preprocessed data is packaged according to the preset data format to obtain the monitoring data packet of the ETC device, and then the first timestamp and other information are added for subsequent data processing and analysis. This step uses Hongmeng's data packaging framework to ensure the uniformity and compatibility of the data format.

[0069] After the data is transmitted to the transport layer, the data of the distributed device network in the time period adjacent to the first timestamp in the transport layer is screened and compressed, and a supplementary data packet is obtained; the supplementary data packet is embedded in the monitoring data packet, and an encapsulated monitoring data packet is obtained.

[0070] Specifically, the data of the distributed device network in the adjacent time period of the first timestamp in the transport layer is screened. Among them, the screening is based on factors such as the importance and timeliness of the data. With the help of Hongmeng's distributed data management framework, the data can be quickly screened and classified. For example, only data directly related to the status of the ETC equipment is selected, such as toll station monitoring images, signal strength of the roadside unit, and communication traffic of the toll station supporting equipment. After that, the screened data is compressed using a lossless compression algorithm to reduce the amount of data transmission. Using Hongmeng's efficient compression algorithm library, the data can be quickly compressed and decompressed. For example, for data that changes slowly, such as temperature and humidity, Huffman coding can be used for compression. The supplementary data packet is embedded in the monitoring data packet, and the encapsulated monitoring data packet is obtained. The encapsulated data packet is transmitted in a fixed format and protocol, such as JSON format and Hongmeng-based MQTT protocol to ensure data integrity and compatibility. The reason why the supplementary data packet is introduced in the above steps is that only a single factor is monitored in traditional technology. Generally, when it exceeds the threshold, a substantial failure has occurred and it does not play a warning role. Setting a low threshold will lead to frequent alarms, affecting normal work. Therefore, in order to achieve more accurate fault warning capabilities and provide better reference data for subsequent predictions, it is necessary to consider the problem of multi-data linkage, especially the linkage between internal sensors and external sensors and other equipment operating conditions. For example, when a roadside unit encounters a low temperature scene after rainfall, water vapor is easily accumulated inside it, causing corrosion of electronic components and short circuits or open circuits. On this basis, if the data of the meteorological sensor in adjacent time periods is not used for joint prediction, it is difficult to make accurate or timely predictions only through internal current and voltage monitoring of the equipment; for example, if the attitude sensor collects data to support the judgment of the tipping state, the general practice is that the system alarms and a dedicated person carries special equipment to perform corrective maintenance, but it is impossible to determine whether it is a sensor abnormality. This embodiment can combine the monitoring data (or recent vibration data) to confirm whether it has tipped over. If not, it can be normally predicted that it is a sensor abnormality. In this way, the dispatched personnel only need to carry the tools and spare parts required for sensor replacement.

[0071] The transport layer transmits the encapsulated monitoring data packets to the cloud computing layer. The gateway device transmits the encapsulated monitoring data packets to the cloud server of the cloud computing layer through a wired or wireless network. During the data transmission process, encrypted communication protocols, such as the secure transmission protocol based on Hongmeng, are used to ensure the security of the data. At the same time, the gateway device is also responsible for managing and coordinating the devices in the distributed device network to ensure smooth communication between devices. Hongmeng's distributed device management framework is used to achieve efficient management and coordination of gateway devices and other devices. For example, when a device fails, the gateway device can automatically adjust the communication path to ensure normal data transmission.

[0072] In the cloud computing layer, the monitoring data packets are input into the machine learning model, and the prediction results are output; optionally, the method for constructing the machine learning model includes the following steps:

[0073] First, based on the historical database of ETC equipment, collect the fault information of past ETC equipment, including the fault type and the working parameters and environmental parameters of the ETC equipment in the corresponding period, and the data of other devices in the distributed device network with the same ETC equipment (and the time period is required to be the same as the fault time period or at least include the fault time period). Use Hongmeng's distributed data storage and processing framework to achieve efficient storage and processing of large-scale data.

[0074] Afterwards, the fault information is clustered and classified, and similar fault information is combined into a training data set (the target is the fault type); further, clustering classification includes the following steps:

[0075] Clean the fault information, remove noise, outliers and duplicate data, and use interpolation to assign extreme values ​​to missing or removed values ​​to ensure data integrity. In particular, the early performance of some faults depends on a single monitoring value, such as the internal temperature exceeding the threshold. The step of removing outliers removes the temperature value, which will lead to the lack of feedback in the subsequent training process. Therefore, in order to ensure the integrity of the data and explore the feedback function of the data, it is necessary to fine-tune according to the actual situation and reassign the empty values. Use Hongmeng's data analysis framework and algorithm library to achieve rapid cleaning and processing of data.

[0076] Normalize the cleaned fault information and use the maximum and minimum normalization method to unify the data into the [0,1] interval. Then classify based on the clustering algorithm to obtain the classification label of the fault information. For example, use the K-Means clustering algorithm to classify the fault information into different categories. Use Hongmeng's machine learning framework and algorithm library to achieve rapid clustering and classification of data.

[0077] The corresponding neural network model is trained by using a training data set with classification labels, such as using a deep neural network (DNN) for training, thereby forming multiple neural network models.

[0078] Finally, multiple neural network models are combined to obtain the required machine learning model. Specifically, multiple DNN models are combined using an ensemble learning method to improve the model's predictive power and robustness. Hongmeng's deep learning framework and algorithm library are used to achieve efficient training and combination of neural network models.

[0079] The encapsulated monitoring data packets sent from the transport layer to the cloud computing layer and the classification labels obtained by clustering and classification of the encapsulated monitoring data packets are input into the above-mentioned machine learning model for real-time analysis and prediction, and the prediction results are output. At the same time, the cloud server can also analyze historical data, explore potential failure modes and trends, and provide decision support for equipment maintenance and management. HarmonyOS's distributed computing framework and machine learning framework are used to achieve real-time analysis and prediction of data.

[0080] When the prediction result is abnormal, the prediction result and the corresponding monitoring data packet are sent to the application layer, as follows:

[0081] Data display and interaction:

[0082] Multi-type and multi-terminals display the status information, warning information, and historical data of ETC equipment through an intuitive interface. Data is displayed in various forms such as charts, reports, and maps, allowing users to quickly understand the operation of the equipment. For example, a bar chart is used to display the changing trend of parameters such as temperature and humidity, and a map is used to display the distribution and status of ETC equipment. HarmonyOS's UI framework and visualization components are used to achieve intuitive display and interaction of data.

[0083] Multi-terminals support interactive operations between users and the system. Users can set warning thresholds, send instructions, etc. Users can set warning thresholds for parameters such as temperature and humidity according to actual conditions. When the threshold is exceeded, the system automatically issues a warning message. Users can also send instructions to the cloud server through the terminal, such as remote control of ETC equipment restart, parameter adjustment, etc. Using Hongmeng's interaction framework and communication framework, efficient interaction between users and the system is achieved.

[0084] Data processing and feedback:

[0085] After the terminal receives the warning information sent by the cloud server, the user can perform corresponding processing and feedback. The user can view the details of the warning information, determine the type and location of the fault, and then arrange maintenance personnel to handle it on site. After the processing is completed, the user can feedback the processing results to the cloud server through the terminal so that the system can record and analyze it. HarmonyOS's communication framework and data management framework can be used to achieve rapid data processing and feedback.

[0086] This embodiment fully demonstrates the application of Hongmeng system in the real-time monitoring and early warning system of ETC equipment status; the system's efficient operation and intelligent early warning are achieved by using Hongmeng's distributed soft bus technology, equipment management framework, data analysis framework, machine learning framework and other advanced technologies. At the same time, the multi-type and multi-terminal developed based on the Hongmeng operating system provides users with convenient interaction methods and data display, improving the practicality and ease of use of the system.

[0087] Finally, it should be noted that the above content is only used to illustrate the technical solution of the present invention, rather than to limit the scope of protection of the present invention. Simple modifications or equivalent substitutions of the technical solution of the present invention by ordinary technicians in this field do not deviate from the essence and scope of the technical solution of the present invention.

Claims

1. An ETC equipment status monitoring and early warning system based on Hongmeng system, characterized by: The system architecture is a multi-layer system, including perception layer, transmission layer, cloud computing layer, and application layer; The perception layer is provided with a plurality of data receiving nodes, each of which is connected to a plurality of sensors, for collecting the working parameters and environmental parameters of the ETC equipment in real time or periodically; The transport layer is constructed with a distributed device network based on Hongmeng distributed soft bus technology; the transport layer is connected to the perception layer; The cloud computing layer is provided with a cloud server and is connected to at least one transport layer; the cloud server is deployed with a machine learning model for ETC equipment failure warning; The application layer is provided with multiple terminals of multiple types, and all of them are connected to the cloud computing layer for sending and receiving data.

2. According to claim 1, the ETC equipment status monitoring and early warning system based on the Hongmeng system is characterized by: At least some sensors are installed on the ETC device, including temperature sensors, humidity sensors, vibration sensors, posture sensors and electromagnetic sensors.

3. The ETC equipment status monitoring and early warning system based on the Hongmeng system according to claim 1 or 2 is characterized in that: The sensor includes an environmental sensor, and the environmental sensor is installed on or around the ETC device; The environmental sensors include meteorological sensors and light sensors.

4. The ETC equipment status monitoring and early warning system based on the Hongmeng system according to claim 1 is characterized by: The distributed device network includes gateway devices, toll station supporting equipment and roadside units; The toll station supporting equipment and the roadside unit are interconnected through the gateway device to form the distributed device network.

5. The ETC equipment status monitoring and early warning system based on the Hongmeng system according to claim 1 is characterized by: The multiple types of terminals include dedicated communication terminals, fixed terminals and portable mobile terminals.

6. A method for monitoring and early warning of ETC equipment status based on Hongmeng system, characterized by: The steps include: Collect the working parameters and environmental parameters of ETC equipment through the perception layer; Packing the working parameters and environmental parameters to obtain a monitoring data packet of the ETC device, and uploading the monitoring data packet to the cloud computing layer through the transport layer; In the cloud computing layer, the monitoring data packet is input into a machine learning model and a prediction result is output; When the prediction result is an abnormal situation, the prediction result and the corresponding monitoring data packet are sent to the application layer.

7. The ETC equipment status monitoring and early warning method based on the Hongmeng system according to claim 6 is characterized by: The working parameters of the ETC device include the temperature, humidity, vibration, attitude angle and electromagnetic intensity inside the device; The environmental parameters of the ETC device include temperature, humidity, wind speed and light intensity outside the device.

8. The ETC equipment status monitoring and early warning method based on the Hongmeng system according to claim 6 is characterized by: At least one distributed device network is constructed in the transmission layer, including gateway devices, toll station supporting equipment and roadside units; The monitoring data packet is configured with a first timestamp in the perception layer; Filtering and compressing data of a distributed device network in a time period adjacent to the first timestamp in the transport layer, and obtaining a supplementary data packet; Embedding the supplementary data packet into the monitoring data packet, and obtaining an encapsulated monitoring data packet; The transport layer transmits the encapsulated monitoring data packet to the cloud computing layer.

9. The ETC equipment status monitoring and early warning method based on the Hongmeng system according to claim 8 is characterized by: The method for constructing the machine learning model comprises the following steps: Collecting fault information of the ETC device, the fault information including the fault type and the working parameters and environmental parameters of the ETC device in the corresponding period, and data of each device in the distributed device network; Clustering and classifying the fault information, and then forming a training data set with fault information of the same type; Training the corresponding neural network model using the training data set; The machine learning model is obtained by combining multiple neural network models.

10. The ETC equipment status monitoring and early warning method based on the Hongmeng system according to claim 9 is characterized in that: The cluster classification comprises the following steps: Performing data cleaning on the fault information and assigning extreme values ​​to missing values ​​or removed values; The cleaned fault information is normalized and classified based on a clustering algorithm to obtain a classification label for the fault information.

Images