State monitoring method of traffic signal lamp and traffic signal system

Abstract

The invention provides a traffic signal lamp state monitoring method and a traffic signal system, and relates to the technical field of intelligent traffic. According to the method, the current data of the traffic signal lamps in all directions are collected, the current characteristics of each traffic signal lamp are extracted, fault diagnosis is carried out, the running state of each traffic signal lamp is determined, and automatic and accurate detection of the traffic signal lamps is achieved. Compared with a manual inspection mode, the method can achieve the fault detection through the current of the traffic signal lamp, shortens the fault detection time of the traffic signal lamp, and improves the inspection efficiency of a traffic signal system.

Description

Technical Field

[0001] The present invention relates to the field of intelligent transportation technology, and in particular to a traffic signal light status monitoring method and a traffic signal system. Background Art

[0002] Traffic lights play a vital role in modern intelligent transportation systems, and their normal operation is directly related to traffic safety and efficiency. In the event of power outages, signal machine failures, signal light failures, etc., if they are not discovered and handled in time, they may cause traffic congestion, traffic accidents and other problems. There are many types of traffic lights at light-controlled intersections with different functions, such as motor vehicles, non-motor vehicles, crosswalks, direction indicators, lanes, crossings, U-turns (some intersections), flashing warning (special circumstances) lights, etc. Such a variety of types of traffic lights cooperate with each other to form a complex intersection traffic signal control system. Different types of traffic lights have different working methods and applicable scenarios, which makes the operation and management of the entire traffic signal system extremely complex.

[0003] At present, there is a method for detecting faults of traffic lights, which realizes fault detection of traffic lights by detecting the effective value of the current of the traffic lights. However, for scenarios with a wide variety of traffic lights and variable adjustment of the duration of traffic lights, only detecting the effective value of the current of the traffic lights is prone to misjudgment. For example, in the scenario of variable adjustment of the duration of traffic lights, the duration of the red and green lights changes, resulting in changes in the effective value of the current of the traffic lights. Using only the effective value of the current for threshold determination leads to misjudgment of the status of the traffic lights.

[0004] In view of the possibility of misjudgment when using the effective value of current to determine the state, the traffic light state detection process needs to adopt the method of patrol inspection by civilian and auxiliary police in the jurisdiction to cooperate in realizing the state inspection of traffic lights. However, this manual inspection method has problems such as slow feedback, poor real-time performance, and high cost. Summary of the invention

[0005] The present invention provides a traffic signal lamp status monitoring method and a traffic signal system, which can realize accurate detection of traffic signal lamps and improve the inspection efficiency of the traffic signal system.

[0006] In a first aspect, the present invention provides a method for monitoring the status of a traffic light, which is applied to a monitoring terminal, wherein the monitoring terminal is connected to a monitoring platform and current sensors of multiple traffic lights, and the method comprises: obtaining current sampling data of multiple traffic lights in a current time period, the current sampling data comprising the power supply current of the signal machine at each moment in the current time period and the main red light current in each direction; calculating the current characteristics of each traffic light in the current time period based on the current sampling data of multiple traffic lights in the current time period; performing fault diagnosis based on the current characteristics of each traffic light in the current time period to determine the operating status of each traffic light; if the operating status of any traffic light is a faulty state, then generating and sending first reporting information to the monitoring platform based on the operating status, location information and occurrence time of any traffic light.

[0007] In a possible implementation, based on the current sampling data of multiple traffic lights in the current time period, the current characteristics of each traffic light in the current time period are calculated, including: based on the current sampling data of each traffic light in the current time period, generating a current sampling sample, each sample in the current sampling sample is the current at any moment in the current time period; sorting each sample in the current sampling sample in order of size to obtain a current sorted sample; splitting the current sorted sample to obtain an open sample and a closed sample; the open sample is the first T samples when sorted from large to small, and the closed sample is the first T samples when sorted from small to large; wherein T is a positive integer; based on the open samples, calculating an open average value; the open average value is used to characterize the average current when the traffic light is turned on; based on the closed samples, calculating a closed average value; the closed average value is used to characterize the average current when the traffic light is turned off; based on the open average value and the closed average value, determining the maximum average current difference; based on the current sorted samples, performing a sliding time window calculation to obtain the average current difference of each sliding time window.

[0008] In one possible implementation, fault diagnosis is performed based on the current characteristics of each traffic light in the current time period to determine the operating status of each traffic light, including: based on the maximum average current difference and / or the average current difference of each sliding time window, a threshold judgment is performed to determine whether the traffic light is working normally; if the traffic light is working normally, the operating status of the traffic light is determined to be normal; and based on the open average value and the closed average value, as well as the average current difference, the normal opening current and the normal closing current of the traffic light are updated; if the traffic light is working abnormally, the number of times the traffic light has worked abnormally continuously is recorded; if the number of times the traffic light has worked abnormally continuously is greater than a preset number of times, the operating status of the traffic light is determined to be a faulty state.

[0009] In a possible implementation, any current data in the current sampling data is obtained in the following manner: obtaining multiple sampling values ​​of the current sensor corresponding to the any current data in the current time period; determining multiple sampling voltages based on the multiple sampling values ​​of the current sensor corresponding to the any current data in the current time period, the resolution of the acquisition device, and the reference voltage of the acquisition device; performing DC bias removal and square sum calculation based on the multiple acquisition voltages to obtain the voltage square sum; performing root mean square calculation based on the voltage square sum, the detection resistor and the turns ratio of the current sensor to obtain the any current data.

[0010] In a possible implementation, after fault diagnosis is performed based on the current characteristics of each traffic light in the current time period and the operating status of each traffic light is determined, the method further includes: if a target traffic light is in a faulty state, an input vector is generated based on the current characteristics of the target traffic light; the target traffic light is any one of a plurality of traffic lights; based on the input vector and a preset fault detection model, the fault type of the target traffic light is obtained; based on the fault type of the target traffic light and the location information of the target traffic light, the impact scope of the target traffic light when the fault occurs is determined; based on the fault type and the impact scope, a repair time limit is determined; based on the fault type and the impact scope, a second reporting information is generated; and the second reporting information is sent to operation and maintenance personnel.

[0011] In a second aspect, an embodiment of the present invention provides a method for monitoring the status of a traffic light, which is applied to a monitoring platform, wherein the monitoring platform is connected to multiple monitoring terminals, and each monitoring terminal is connected to current sensors of multiple traffic lights, the method comprising: obtaining current sampling data of multiple traffic lights in a current time period reported by multiple monitoring terminals, the current sampling data including the power supply current of the signal machine at each moment in the current time period and the main red light current in each direction; calculating the current characteristics of each traffic light in the current time period based on the current sampling data of multiple traffic lights; performing fault diagnosis based on the current characteristics of each traffic light in the current time period to determine the operating status of each traffic light; if the operating status of any traffic light is a faulty state, generating an operation and maintenance instruction based on the operating status, location information and occurrence time of any traffic light, the operation and maintenance instruction being used to instruct the operation and maintenance personnel to inspect and repair any traffic light.

[0012] In a possible implementation, the method also includes: obtaining current sampling data when multiple traffic lights fail during a historical period, and the fault type of each traffic light; performing feature extraction based on the circuit sampling data when each traffic light fails, and determining the current characteristics of each traffic light; generating multiple training samples based on the current characteristics of each traffic light and the fault type of each traffic light; and constructing a fault detection model based on the multiple training samples using a decision tree algorithm.

[0013] In a possible implementation, the current characteristics include the open average value, the closed average value, the maximum average current difference, the average current difference of each sliding time window, the current peak value, the current valley value, the current fluctuation range, the current change trend, the current change cycle and each harmonic current in a set period before and after the traffic light fails; based on multiple training samples, a decision tree algorithm is used to build a fault detection model, including: initializing the decision tree model, determining the constraints of the decision tree model, the constraints include the maximum depth of the decision tree, the maximum number of leaf nodes and the minimum number of node samples; traversing each current characteristic and calculating the Gini index of each current characteristic; determining the root node of the decision tree model based on the Gini index of each current characteristic; performing node splitting based on the root node of the decision tree model and multiple training samples to determine multiple subsets, each subset corresponding to a branch in the decision tree model; building a decision tree model based on multiple subsets and constraints; and performing pruning operations on the constructed decision tree model to determine the fault detection model.

[0014] In a possible implementation, the method also includes: receiving first reporting information and / or second reporting information returned by each monitoring terminal; extracting current sampling data of each faulty signal light based on the first reporting information and / or second reporting information returned by each monitoring terminal; extracting current characteristics of each faulty signal light based on the current sampling data of each faulty signal light; determining the fault type of each faulty signal light based on the current characteristics of each faulty signal light and a fault detection model; determining whether there is a collective fault in the traffic signal system based on the fault type, location information and occurrence time of each faulty signal light; a collective fault indicates that the fault causes of multiple faulty signal lights are the same; if there is a collective fault in the traffic signal system, a fault analysis result is generated based on the location information of the multiple faulty signal lights that have a collective fault, and the fault analysis result includes the fault type, location information and suspected fault cause of the multiple faulty signal lights that have a collective fault; generating third reporting information based on the fault analysis result; and sending the third reporting information to operation and maintenance personnel.

[0015] In a third aspect, an embodiment of the present invention provides a traffic light status monitoring device, which is applied to a monitoring terminal, and the monitoring terminal is connected to a monitoring platform and current sensors of multiple traffic lights. The device includes: a communication module and a processing module, the communication module is used to obtain current sampling data of multiple traffic lights in a current time period, and the current sampling data includes the power supply current of the signal machine at each moment in the current time period and the main red light current in each direction; the processing module is used to calculate the current characteristics of each traffic light in the current time period based on the current sampling data of multiple traffic lights in the current time period; based on the current characteristics of each traffic light in the current time period, fault diagnosis is performed to determine the operating status of each traffic light; if the operating status of any traffic light is a faulty state, then based on the operating status, location information and occurrence time of any traffic light, a first reporting information is generated and sent to the monitoring platform.

[0016] In a fourth aspect, an embodiment of the present invention provides a traffic light status monitoring device, which is applied to a monitoring platform, wherein the monitoring platform is connected to multiple monitoring terminals, and each monitoring terminal is connected to current sensors of multiple traffic lights. The device includes: a communication module and a processing module, wherein the communication module is used to obtain current sampling data of multiple traffic lights in the current time period reported by multiple monitoring terminals, wherein the current sampling data includes the power supply current of the signal machine at each moment in the current time period and the main red light current in each direction; the processing module is used to calculate the current characteristics of each traffic light in the current time period based on the current sampling data of multiple traffic lights; based on the current characteristics of each traffic light in the current time period, fault diagnosis is performed to determine the operating status of each traffic light; if the operating status of any traffic light is a faulty state, an operation and maintenance instruction is generated based on the operating status, location information and occurrence time of any traffic light, and the operation and maintenance instruction is used to instruct the operation and maintenance personnel to inspect and repair any traffic light.

[0017] In a fifth aspect, an embodiment of the present invention provides a traffic signal system, which includes an electronic device, the electronic device includes a memory and a processor, the memory stores a computer program, and the processor is used to call and run the computer program stored in the memory to execute the method described in any possible implementation method of the first aspect above, or any possible implementation method of the second aspect.

[0018] In a sixth aspect, an embodiment of the present invention provides a computer-readable storage medium, wherein the computer-readable storage medium stores a computer program, and wherein when the computer program is executed by a processor, the steps of the method described in the first aspect and any possible implementation method of the first aspect are implemented.

[0019] The present invention provides a traffic light status monitoring method and a traffic signal system. The present invention collects current data of traffic lights in all directions and extracts the current characteristics of each traffic light to perform fault diagnosis and determine the operating status of each traffic light, thereby realizing automatic and accurate detection of traffic lights. Compared with the manual inspection method, the present invention can perform fault detection through the current of traffic lights, shorten the time for finding faults of traffic lights, and improve the inspection efficiency of the traffic signal system. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the drawings required for use in the embodiments or the description of the prior art 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.

[0021] Figure 1 It is a schematic diagram of the architecture of a smart traffic alarm linkage control system provided by an embodiment of the present invention; Figure 2 It is a flow chart of a method for monitoring the state of a traffic light provided by an embodiment of the present invention; Figure 3 It is a structural schematic diagram of a state monitoring device for a traffic light provided by an embodiment of the present invention; Figure 4 It is a structural schematic diagram of an electronic device provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0022] In the following description, specific details such as specific system structures, technologies, etc. are provided for the purpose of illustration rather than limitation, so as to provide a thorough understanding of the embodiments of the present invention. However, it should be clear to those skilled in the art that the present invention may be implemented in other embodiments without these specific details. In other cases, detailed descriptions of well-known systems, devices, circuits, and methods are omitted to prevent unnecessary details from obstructing the description of the present invention.

[0023] In the description of the present invention, unless otherwise specified, " / " means "or", for example, A / B can mean A or B. "And / or" in this article is only a description of the association relationship of associated objects, indicating that there can be three relationships. For example, A and / or B can mean: A exists alone, A and B exist at the same time, and B exists alone. In addition, "at least one" and "plurality" refer to two or more. The words "first", "second", etc. do not limit the quantity and execution order, and the words "first", "second", etc. do not limit them to be different.

[0024] In the embodiments of the present application, words such as "exemplary" or "for example" are used to indicate examples, illustrations or descriptions. Any embodiment or design described as "exemplary" or "for example" in the embodiments of the present application should not be interpreted as being more preferred or more advantageous than other embodiments or designs. Specifically, the use of words such as "exemplary" or "for example" is intended to present related concepts in a concrete way for easy understanding.

[0025] In addition, the terms "including" and "having" and any variations thereof mentioned in the description of this application are intended to cover non-exclusive inclusions. For example, a process, method, system, product or device that includes a series of steps or modules is not limited to the listed steps or modules, but may optionally include other steps or modules that are not listed, or may optionally include other steps or modules that are inherent to these processes, methods, products or devices.

[0026] In order to make the purpose, technical solutions and advantages of the present invention more clear, the following will be described through specific embodiments in conjunction with the accompanying drawings of the present invention.

[0027] As described in the background technology, the current traffic signal system has problems such as slow feedback, poor real-time performance, and high cost in the state detection of traffic lights. Figure 1 As shown, an embodiment of the present invention provides a smart traffic alarm linkage control system. The system includes a monitoring terminal and a monitoring platform. By real-time monitoring of the current of multiple traffic lights in the traffic signal system, intelligent inspection of the traffic signal system is achieved.

[0028] In the embodiment of the present application, the monitoring platform is connected to the monitoring terminal via a wireless network, for example, a 4G wireless network. Each monitoring terminal is connected to the current sensors of multiple traffic lights. For example, each monitoring terminal can monitor multiple traffic lights at an intersection, such as motor vehicles, non-motor vehicles, pedestrian crossings, direction indicators, lanes, crossings, U-turns (part of intersections), flashing warning (special circumstances) lights, etc. at an intersection.

[0029] In some embodiments, the monitoring terminal is responsible for collecting the current data of the traffic light and transmitting the data to the monitoring platform through the wireless communication module. In terms of current collection, by installing a transformer on the signal light line, according to the working principle of the signal light, the signal machine power supply current and the main red light current in each direction are monitored. During normal operation, these currents are within a specific range, and there is a specific current change pattern when the yellow light flashes, so as to judge the working status of the signal light. For example, the current of a normal red light should be kept within a certain value range. If the current suddenly disappears or exceeds the range, it may indicate a red light failure.

[0030] The built-in wireless communication module of the monitoring terminal supports multiple network formats and can communicate with the monitoring platform in real time. The monitoring terminal has stable wireless communication capabilities to ensure timely and reliable data transmission. During data processing and transmission, the collected current data is first pre-processed, including filtering, amplification and other operations to improve data quality. Then the data is encapsulated and transmitted to the monitoring platform according to the set communication protocol.

[0031] In some embodiments, the monitoring platform mainly implements functions such as real-time monitoring, fault diagnosis and data analysis of the status of traffic lights. The real-time monitoring function receives data uploaded by the monitoring terminal, processes and analyzes it, and displays the status of the traffic lights in an intuitive interface. The fault diagnosis function determines the type of fault based on the preset fault diagnosis algorithm and the pattern recognition algorithm based on the current characteristics. Once a fault is detected, an alarm notification is immediately sent to the operation and maintenance personnel, and the notification methods include WeChat, pop-up windows, etc., to ensure that the operation and maintenance personnel obtain fault information in a timely manner. The data analysis and report generation function conducts in-depth analysis of historical data and generates detailed reports to provide decision-making basis for traffic management departments. It can optimize maintenance plans and resource allocation based on the frequency and time distribution of traffic light failures.

[0032] based on Figure 1 The intelligent traffic alarm linkage control system shown in FIG. Figure 2 As shown, an embodiment of the present invention provides a method for monitoring the state of a traffic light, which is applied to a monitoring terminal, the monitoring terminal is connected to a monitoring platform and current sensors of multiple traffic lights, and the method includes steps S101-S104.

[0033] S101. Obtain current sampling data of multiple traffic lights in a current period.

[0034] In the embodiment of the present application, the current sampling data includes the power supply current of the traffic light at each moment in the current period and the main red light current in each direction.

[0035] For example, the signal machine power supply current collection uses 1 current collection channel, which is realized by adding a transformer to the wire. The main red light current collection uses 4 current collection channels, and transformers are also installed to accurately obtain the main red light current in each direction. Considering the size of the equipment and the convenience of installation, for multi-signal light detection, the detection object should be reasonably selected to avoid too many transformers causing the equipment to be too large and difficult to install. The monitoring terminal also has a 485 interface, which can expand the detection channel to meet the needs of different application scenarios.

[0036] For example, the built-in wireless communication module of the monitoring device can communicate with the monitoring platform in real time and report the detected faults in time. The data upload rules are set according to different situations. For example, when there is no current in the four channels for more than 10 seconds, the alarm and the status of all channels are uploaded; when the current value of a channel is 50% lower than the normal value for 10 minutes, the alarm is uploaded, etc. These rules ensure that the monitoring platform can grasp the changes in the status of the signal lights in time and effectively provide fault warnings.

[0037] For example, after collecting the current sampling data of the traffic light, the embodiment of the present invention can directly transmit the current sampling data to the monitoring platform through a wireless network, and the monitoring platform performs the analysis of the current sampling data, such as steps S301-S304.

[0038] In another exemplary embodiment of the present invention, after collecting the current sampling data of the traffic light, the monitoring terminal can directly analyze the current sampling data and send the analysis results, i.e., the first reporting information and / or the second reporting information, to the monitoring platform. For example, steps S102-S104, steps S201-S206. After receiving the first reporting information and / or the second reporting information, the monitoring platform performs summary analysis. For example, steps S501-S508.

[0039] As a possible implementation manner, any current data in the current sampling data can be obtained through steps A1-A4.

[0040] A1. Acquire multiple sampling values ​​of the current sensor corresponding to any current data in the current period.

[0041] A2. Determine multiple acquisition voltages based on multiple sampling values ​​of the current sensor corresponding to any current data in the current period, the resolution of the acquisition device, and the reference voltage of the acquisition device.

[0042] A3. Based on the multiple collected voltages, DC offset removal and square sum calculation are performed to obtain the voltage square sum.

[0043] A4. Based on the voltage square sum, the detection resistor and the turns ratio of the current sensor, a root mean square calculation is performed to obtain any current data.

[0044] Exemplarily, in the present invention, the red light fault is mainly determined by detecting whether the current is abnormal, so the current detection stage is particularly important, and the algorithms used internally include filtering and root mean square RMS calculation.

[0045] The embodiment of the present invention samples the output signal of the current transformer through the digital-to-analog converter ADC and converts the sampled value (digital quantity) into an actual voltage value. This process involves the mapping of the reference voltage and the ADC resolution. Assuming that the sampled value at a certain moment in the current period, that is, the ADC sampled value is x, the reference voltage is Vref, and the maximum resolution of the ADC is Nmax (Nmax=4095 for a 12-bit ADC), the calculation formula for the voltage value Vout is: ; After the above process, the ADC value is converted into a voltage value. Next, the collected sine wave is calculated through RMS to obtain the current value.

[0046] In some embodiments, the present invention removes the DC bias before calculating the RMS value. Assume that the voltage value of the i-th sampling signal obtained by the above conversion is , the DC bias is , then the AC component of the i-th sampling signal for: .

[0047] Square sum calculation: The present invention squares and adds the AC components of each sample value in the current period to obtain the square sum : .

[0048] RMS calculation: , where N is the number of sampling points in the current period. The RMS value obtained is proportionally calculated to obtain the actual current value of the red light: , R is the detection resistor, K is the turns ratio of the current transformer, and A is the amplification factor of the operational amplifier in the digital-to-analog converter. Finally, the average value filter is used to obtain a stable current value.

[0049] S102: Calculate the current characteristics of each traffic light in the current period based on the current sampling data of the plurality of traffic lights in the current period.

[0050] In some embodiments, the current characteristics include an open average value, a closed average value, a maximum average current difference value, and an average current difference value in each sliding time window in the current period.

[0051] In some embodiments, the current characteristics also include the open average value, closed average value, maximum average current difference, average current difference of each sliding time window, current peak value, current valley value, current fluctuation range, current change trend, current change cycle and each harmonic current of a set time period before the fault occurs and a set time period after the fault occurs.

[0052] As a possible implementation manner, step S102 may be specifically implemented as steps S1021 - S1027 .

[0053] S1021. Generate current sampling samples based on the current sampling data of each traffic light in the current period.

[0054] In some embodiments, each sample in the current sampling samples is the current at any moment in the current period.

[0055] S1022. Sort each sample in the current sampling sample in order of size to obtain a current sorting sample.

[0056] S1023. Split the current sorting samples to obtain open samples and closed samples.

[0057] In some embodiments, the open samples are the first T samples when sorted from large to small, and the closed samples are the first T samples when sorted from small to large; wherein T is a positive integer.

[0058] S1024. Based on the open samples, calculate the open average value.

[0059] In some embodiments, the open average value is used to characterize the average current when the traffic light is turned on.

[0060] S1025. Calculate the closing average value based on the closing samples.

[0061] In some embodiments, the off average value is used to characterize the average current when the traffic light is off.

[0062] S1026. Determine a maximum average current difference based on the open average value and the closed average value.

[0063] S1027: Perform sliding time window calculation based on the current sorting samples to obtain an average current difference in each sliding time window.

[0064] S103: Perform fault diagnosis based on the current characteristics of each traffic light in the current period to determine the operating status of each traffic light.

[0065] As a possible implementation manner, step S103 may be specifically implemented as steps S1031 - S1034 .

[0066] S1031. Perform a threshold judgment based on the maximum average current difference and / or the average current difference in each sliding time window to determine whether the traffic light is working normally.

[0067] S1032: If the traffic light is working normally, determine that the operating state of the traffic light is normal; and update the normal opening current and the normal closing current of the traffic light based on the opening average value and the closing average value, and the average current difference.

[0068] S1033. If the traffic light operates abnormally, the number of times the traffic light operates abnormally continuously is recorded.

[0069] S1034: If the number of times that the traffic light continuously operates abnormally is greater than a preset number, it is determined that the operating state of the traffic light is a fault state.

[0070] For example, the embodiments of the present invention can use a dynamic signal integrity analysis algorithm to analyze the current sampling data to obtain the operating status of the traffic light. The algorithm detects current anomalies by dynamically analyzing the integrity of the signal and is suitable for real-time monitoring and fault diagnosis of multi-channel current signals. The algorithm determines whether the signal is normal through sampling, sorting, average value calculation and threshold comparison, and updates the threshold or reports the fault according to the detection result. The main process is as follows: Signal sampling: Sample the signal of each channel multiple times and store the sampled values.

[0071] Data sorting: Sort the sample values ​​and extract the maximum and minimum values.

[0072] Threshold calculation: Calculate the normal value and zero value (alarm value) based on the sorted data.

[0073] Status judgment: Based on the comparison between the calculated value and the preset threshold, it is judged whether the signal status is normal or faulty.

[0074] Fault handling: If the signal status is abnormal, the fault is recorded and the reporting logic is triggered.

[0075] Exemplarily, (1) in the sampling phase, the embodiment of the present invention may collect N samples for each channel c: ;in, is the i-th sampling current.

[0076] (2) Sorting and average calculation: sort the samples in ascending order.

[0077] and ; Calculate the average of the top and bottom T samples to get the open average and close average .

[0078] ; (3) Difference calculation, calculation of open average and close average The difference between the maximum average current . .

[0079] It should be noted that the maximum average current difference Indicates the difference between the open average and the closed average in the current period, that is, the difference in average current when the traffic light is on and off. If it is greater than or equal to the threshold, it means that the maximum average current difference is large, that is, the difference between the average current when the traffic light is in the on state and the average current when the traffic light is in the off state is large, which indicates that the traffic light is working normally.

[0080] When the maximum average current difference If it is less than the threshold, it means that the maximum average current difference is small, that is, the difference between the average current when the traffic light is in the on state and the off state is small, which means that the on state current and the off state current of the traffic light are similar, the traffic light is in the normally on state or the normally off state, and the traffic light is working abnormally.

[0081] (4) Threshold comparison: determine whether the difference exceeds the threshold.

[0082] like , then update the normal value and zero value. Otherwise, increase the fault count and report the fault.

[0083] Where R represents the judgment error. L is the judgment threshold. For example, R can be 0.7, which means the maximum average current difference. 0.7 times of is greater than the judgment threshold, indicating that the current difference between the traffic light when it is turned on and off is large, which can indicate that the traffic light is working properly. Setting the judgment error can reduce the probability of misjudgment of the traffic light and improve the accuracy of traffic light fault detection.

[0084] Exemplarily, the judgment threshold L can be determined and updated according to the normal value and the zero value. For example, the difference between the normal value and the zero value can be directly determined as the judgment threshold. It is also possible to directly determine 0.5 times or 0.6 times the difference between the normal value and the zero value as the judgment threshold.

[0085] In another exemplary embodiment, the judgment threshold may be determined and updated based on the open average value and the closed average value. For example, the difference between the open average value and the closed average value may be directly determined as the judgment threshold. Alternatively, 0.5 times or 0.6 times the difference between the open average value and the closed average value may be directly determined as the judgment threshold.

[0086] As another example, the embodiment of the present invention may update the normal value and the zero value according to the open average value and the closed average value.

[0087] For example, after a traffic light is replaced, the normal value and zero value of the new traffic light will change compared with the original traffic light. Accordingly, the detection threshold of the traffic light in the monitoring terminal or monitoring platform should also be changed accordingly to prevent misjudgment and improve detection accuracy.

[0088] For another example, after a traffic light has been used for a period of time, the normal operating current of the traffic light will change, which may increase slightly or decrease slightly. The present invention adjusts the judgment threshold according to the open average value and the closed average value, which can reduce the probability of misjudgment caused by changes in the working conditions of the traffic light and improve the probability of misjudgment of the traffic light.

[0089] For example, the embodiment of the present invention can also set an upper limit and a lower limit for the judgment threshold according to the factory setting information of the traffic light. When the average current value of the traffic light exceeds the upper limit or the lower limit, it indicates that the traffic light may be faulty and further judgment is required. This can prevent the error problem caused by the continuous adjustment of the judgment threshold. That is, when the judgment threshold exceeds the normal range, the judgment threshold is limited and regulated by the upper limit and the lower limit.

[0090] In some embodiments, if , indicating that the traffic light is working normally, updating the normal value and zero value.

[0091] In some embodiments, if , indicating that the traffic light is working abnormally, it is necessary to continuously count the faults of the traffic light and generate the first reporting information to report to the monitoring platform.

[0092] (5) Updating the normal value and the zero value. The embodiment of the present invention can update the normal value and the zero value according to the following formula.

[0093] ; Among them, the normal value Indicates the working current when the traffic light is normally turned on, zero value Indicates the operating current when the traffic light is normally closed.

[0094] (6) Fault detection, if , and it happens K times in a row, it means that the traffic light is in a faulty state and the working state of the traffic light needs to be reported.

[0095] (7) Adaptive threshold update: The embodiment of the present invention may also introduce an exponentially weighted moving average (EWMA) to adapt to changes in the working signal of a traffic light.

[0096] ; in, is the smoothing factor, . It represents the normal start-up current of the k+1th adjustment cycle, represents the normal closing current of the k+1th adjustment cycle, represents the normal start-up current of the kth adjustment cycle, Represents the normal off current of the kth adjustment cycle. To turn off averaging, To turn on averaging. Maximum average current difference, For judgment error.

[0097] (8) Sliding window fault detection: Embodiments of the present invention can use the average difference of the sliding window to detect faults.

[0098] like , it means that the traffic light is in a faulty state and needs to be reported.

[0099] in, .

[0100] represents the maximum average current difference of the i-th time window, that is, for the current sorting samples in step S1022, the current sorting samples are divided into sliding time windows to obtain multiple time windows, It represents the difference between the average current in the i-th time window and the average current in the i+1-th time window. represents the average value of the maximum average current difference of K time windows. R represents the judgment error. L represents the judgment threshold.

[0101] The embodiment of the present invention can divide the current in the current period into sliding time windows, calculate the average current difference, make a judgment, avoid the influence of traffic light fault harmonics, and improve the accuracy of traffic light fault detection.

[0102] For example, in the case of abnormal flashing of traffic lights, the maximum average current difference indicates that the traffic lights are working properly, which results in a misjudgment. However, by detecting faults through the average difference of the sliding time window, it can be determined that the traffic lights are in an abnormal state, thus improving the accuracy of fault detection.

[0103] S104: If the operating state of any traffic light is a fault state, based on the operating state, location information and occurrence time of any traffic light, generate and send first reporting information to the monitoring platform.

[0104] The present invention provides a traffic light status monitoring method, which collects the current data of traffic lights in all directions, extracts the current characteristics of each traffic light, performs fault diagnosis, determines the operating status of each traffic light, and realizes automatic and accurate detection of traffic lights. Compared with the manual inspection method, the present invention can detect faults through the current of traffic lights, shortens the time for finding faults of traffic lights, and improves the inspection efficiency of traffic signal systems.

[0105] It should be noted that the current feature extraction and analysis process of traffic lights can be completed by the monitoring terminal or by the monitoring platform. When completed by the monitoring terminal, the data processing pressure of the monitoring platform can be reduced, the response of the monitoring platform can be faster, and the processing capacity and speed of the monitoring platform can be improved. When completed by the monitoring platform, the monitoring platform can summarize the overall fault conditions of multiple traffic lights, perform an overall diagnosis of the traffic signal system, identify common faults, and improve the diagnostic efficiency and accuracy of the traffic signal system.

[0106] Optionally, the method for monitoring the state of a traffic light provided in an embodiment of the present invention further includes steps S201 to S206 after step S103.

[0107] S201: If the target signal lamp is in a fault state, generate an input vector based on the current characteristics of the target signal lamp.

[0108] In some embodiments, the target traffic light is any one of a plurality of traffic lights.

[0109] S202: Obtain the fault type of the target signal light based on the input vector and a preset fault detection model.

[0110] In some embodiments, the fault types include the signal light being always on, the signal light being always off, and abnormal flashing.

[0111] In some embodiments, the fault type also includes uncontrollable signal lights and errors in the lighting mechanism of signal lights in different directions.

[0112] Correspondingly, the present invention can also form an input vector based on the current characteristics of the target signal light and the current characteristics of the signal light in a direction different from the target signal light in real time to detect the control mode and lighting mechanism of the signal light.

[0113] S203: Determine the impact range of the target signal light when the target signal light fails based on the fault type of the target signal light and the location information of the target signal light.

[0114] Exemplarily, the location information of the target traffic light may be a traffic light for a pedestrian crossing, a traffic light for a motor vehicle lane, a traffic light for a U-turn lane, or the like.

[0115] Alternatively, the location information of the target traffic light may be divided according to the traffic volume, for example, traffic lights with heavy traffic volume, such as in the city center; and traffic lights with light traffic volume, such as in the suburbs.

[0116] The influence range of signal lights at different locations is different, and the influence range of signal lights with different fault types is different. The present invention can classify the influence range according to the fault type and location information to determine the influence level of the target signal light fault, for example, primary influence, secondary influence, and tertiary influence.

[0117] S204: Determine the repair time limit based on the fault type and impact scope.

[0118] Exemplarily, the embodiment of the present invention can determine the repair time limit according to the impact level. For example, the repair time limit for a first-level impact can be 1 day, the repair time limit for a second-level impact can be 8 hours, and the repair time limit for a third-level impact can be 2 hours.

[0119] S205: Generate second reporting information based on the fault type, impact range and repair time limit of the target traffic light.

[0120] S206: Send the second report information to the operation and maintenance personnel.

[0121] In this way, the embodiment of the present invention can analyze the current characteristics through the fault detection model and automatically determine the possible fault type of the traffic light, which is convenient for maintenance personnel to repair and improve the maintenance efficiency of the traffic signal system.

[0122] based on Figure 1 The smart traffic alarm linkage control system shown in the figure, the embodiment of the present invention also provides a method for monitoring the state of a traffic light, which is applied to a monitoring platform, the monitoring platform is connected to multiple monitoring terminals, each monitoring terminal is connected to current sensors of multiple traffic lights, and the method includes steps S301-S304.

[0123] S301. Obtain current sampling data of multiple traffic lights in a current period reported by multiple monitoring terminals.

[0124] In some embodiments, the current sampling data includes the power supply current of the traffic light at each moment in the current period and the main red light current in each direction.

[0125] Exemplarily, after the monitoring terminal detects the sampled current of each current sensor, the sampled current may be directly sent to the monitoring platform for analysis.

[0126] S302: Calculate the current characteristics of each traffic light in the current period based on the current sampling data of the plurality of traffic lights.

[0127] Exemplarily, the monitoring platform may execute the method of steps S1021-S1027 to calculate the current characteristics of each traffic light in the current period.

[0128] S303: Perform fault diagnosis based on the current characteristics of each traffic light in the current period to determine the operating status of each traffic light.

[0129] Exemplarily, the monitoring platform may execute the method of steps S1031-S1034 to determine the operating status of each traffic light.

[0130] S304: If the operating state of any traffic light is a fault state, an operation and maintenance instruction is generated based on the operating state, location information and occurrence time of the traffic light.

[0131] In the embodiment of the present application, the operation and maintenance instruction is used to instruct the operation and maintenance personnel to inspect and repair any traffic light.

[0132] The present invention provides a traffic light status monitoring method, which collects the current data of traffic lights in all directions, extracts the current characteristics of each traffic light, performs fault diagnosis, determines the operating status of each traffic light, and realizes automatic and accurate detection of traffic lights. Compared with the manual inspection method, the present invention can detect faults through the current of traffic lights, shortens the time for finding faults of traffic lights, and improves the inspection efficiency of traffic signal systems.

[0133] Optionally, the method for monitoring the state of a traffic light provided by an embodiment of the present invention further includes steps S401 - S406 .

[0134] S401, obtaining current sampling data when multiple traffic lights fail during a historical period, and the failure type of each traffic light; S402: Extract features based on circuit sampling data when each traffic light fails, and determine current features of each traffic light.

[0135] In some embodiments, the current characteristics include the open average value, closed average value, maximum average current difference, average current difference of each sliding time window, current peak value, current valley value, current fluctuation range, current change trend, current change cycle and each harmonic current of a set period before and after a traffic light fails.

[0136] S403, generating a plurality of training samples based on the current characteristics of each traffic light and the fault type of each traffic light; S404: Based on multiple training samples, a fault detection model is constructed using a decision tree algorithm.

[0137] As a possible implementation manner, step S404 can be specifically implemented as steps B1-B5.

[0138] B1. Initialize the decision tree model and determine the constraints of the decision tree model, including the maximum depth of the decision tree, the maximum number of leaf nodes, and the minimum number of node samples; traverse each current feature and calculate the Gini index of each current feature; B2. Determine the root node of the decision tree model based on the Gini index of each current feature; B3, based on the root node of the decision tree model and multiple training samples, split the node to determine multiple subsets, each subset corresponding to a branch in the decision tree model; B4. Construct a decision tree model based on multiple subsets and constraints; B5. Prune the constructed decision tree model to determine the fault detection model.

[0139] In this way, the embodiment of the present invention can detect the operating status of the traffic light by constructing a fault detection model, thereby improving the fault detection efficiency and accuracy of the traffic light.

[0140] In addition, the decision tree model can solve the problem of traffic light fault diagnosis with a small hard disk usage and memory overhead, and can be applied to scenarios such as monitoring terminals where both hard disk and memory are small. In this way, the monitoring platform can build a fault detection model, and the monitoring terminal can run the fault detection model, which can reduce the memory overhead of the monitoring platform and improve the processing capacity of the monitoring platform.

[0141] Optionally, the method for monitoring the state of a traffic light provided in an embodiment of the present invention, after step S404, further includes: the monitoring platform sending a model file of a fault detection model to the monitoring terminal to instruct the monitoring terminal to detect the traffic light based on the fault detection model.

[0142] like Figure 1 As shown, the intelligent traffic alarm linkage control system can also include a sub-control platform, which is connected to the monitoring platform to execute processing tasks assigned by the monitoring platform, such as the current monitoring task of traffic lights in a certain area, thereby reducing the pressure on the monitoring platform.

[0143] Optionally, the method for monitoring the state of a traffic light provided in an embodiment of the present invention further includes steps S501-S508.

[0144] S501: Receive first reporting information and / or second reporting information sent back by each monitoring terminal.

[0145] S502: extracting current sampling data of each fault signal lamp based on the first reporting information and / or the second reporting information returned by each monitoring terminal.

[0146] S503: extracting current characteristics of each fault signal lamp based on the current sampling data of each fault signal lamp.

[0147] S504: Determine the fault type of each fault signal lamp based on the current characteristics of each fault signal lamp and the fault detection model.

[0148] Exemplarily, the embodiment of the present invention can directly use the current characteristics of any fault signal lamp as an input vector, input it into the fault detection model, and obtain the fault type.

[0149] In another exemplary embodiment, the embodiment of the present invention can determine the correlation between the fault signal lights, and based on the correlation, determine several groups of related fault signal lights. Based on the current characteristics of each group of related fault signal lights, an input vector is constructed and input into the fault detection model to obtain the fault type, thereby realizing comprehensive analysis of multiple fault signal lights.

[0150] S505: Determine whether there is a collective fault in the traffic signal system based on the fault type, location information and occurrence time of each faulty signal light.

[0151] In some embodiments, the collective fault indicates that the fault causes of the multiple faulty signal lights are the same.

[0152] For example, if the fault type of multiple faulty signal lights is that the signal lights are normally closed, it may be a power outage. If multiple faulty signal lights at an intersection are normally open or flash abnormally, it may be a signal machine failure.

[0153] S506: If there is a collective fault in the traffic signal system, a fault analysis result is generated based on the location information of multiple faulty signal lights that have collectively failed.

[0154] In some embodiments, the fault analysis result includes the fault type, location information, and suspected fault cause of multiple faulty signal lamps that have collectively failed.

[0155] In some embodiments, suspected fault causes include power outages and signal failures.

[0156] S507: Generate third reporting information based on the fault analysis result.

[0157] In some embodiments, the third reporting information includes the fault type, location information, fault cause, etc. of the fault signal light.

[0158] S508: Send the third reporting information to the operation and maintenance personnel.

[0159] In this way, the embodiment of the present invention can conduct a comprehensive analysis of multiple faulty signal lights through a monitoring platform, facilitate maintenance personnel to overhaul, and improve the maintenance efficiency when a traffic signal system fails.

[0160] It should be understood that the order of execution of the steps in the above embodiment does not necessarily mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiment of the present invention.

[0161] The following is an embodiment of the device of the present invention. For details not described in detail, reference may be made to the corresponding method embodiments described above.

[0162] Figure 3 The schematic diagram of the structure of a traffic light status monitoring device provided by an embodiment of the present invention is shown. The status monitoring device 600 includes a communication module 601 and a processing module 602 .

[0163] When the state detection device 600 is applied to a monitoring terminal, the communication module 601 and the processing module 602 perform the following steps.

[0164] The communication module 601 is used to obtain current sampling data of multiple traffic lights in the current period, and the current sampling data includes the power supply current of the signal machine at each moment in the current period and the main red light current in each direction.

[0165] The processing module 602 is used to calculate the current characteristics of each traffic light in the current period based on the current sampling data of multiple traffic lights in the current period; perform fault diagnosis based on the current characteristics of each traffic light in the current period to determine the operating status of each traffic light; if the operating status of any traffic light is a faulty state, then based on the operating status, location information and occurrence time of any traffic light, generate and send first reporting information to the monitoring platform.

[0166] When the state detection device 600 is applied to a monitoring platform, the communication module 601 and the processing module 602 perform the following steps.

[0167] The communication module 601 is used to obtain current sampling data of multiple traffic lights in the current period reported by multiple monitoring terminals, and the current sampling data includes the power supply current of the signal machine at each moment in the current period and the main red light current in each direction.

[0168] The processing module 602 is used to calculate the current characteristics of each traffic light in the current period based on the current sampling data of multiple traffic lights; perform fault diagnosis based on the current characteristics of each traffic light in the current period to determine the operating status of each traffic light; if the operating status of any traffic light is a faulty state, then generate an operation and maintenance instruction based on the operating status, location information and occurrence time of any traffic light, and the operation and maintenance instruction is used to instruct the operation and maintenance personnel to inspect and repair any traffic light.

[0169] Figure 4 Schematic diagram of the structure of an electronic device provided by an embodiment of the present invention. Figure 4 As shown, the electronic device 700 includes: a processor 701, a memory 702, and a computer program 703 stored in the memory 702 and executable on the processor 701. When the processor 701 executes the computer program 703, the steps in the above-mentioned method embodiments are implemented, for example Figure 2Alternatively, when the processor 701 executes the computer program 703, the functions of the modules / units in the above-mentioned device embodiments are realized, for example, Figure 3 The functions of the communication module 601 and the processing module 602 are shown.

[0170] Exemplarily, the computer program 703 may be divided into one or more modules / units, which are stored in the memory 702 and executed by the processor 701 to implement the present invention. The one or more modules / units may be a series of computer program instruction segments that can implement specific functions, and the instruction segments are used to describe the execution process of the computer program 703 in the electronic device 700. For example, the computer program 703 may be divided into Figure 3 A communication module 601 and a processing module 602 are shown.

[0171] The processor 701 may be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. A general-purpose processor may be a microprocessor or any conventional processor, etc.

[0172] The memory 702 may be an internal storage unit of the electronic device 700, such as a hard disk or memory of the electronic device 700. The memory 702 may also be an external storage device of the electronic device 700, such as a plug-in hard disk, a smart media card (SMC), a secure digital (SD) card, a flash card, etc. equipped on the electronic device 700. Further, the memory 702 may also include both an internal storage unit of the electronic device 700 and an external storage device. The memory 702 is used to store the computer program and other programs and data required by the terminal. The memory 702 may also be used to temporarily store data that has been output or is to be output.

[0173] The embodiments described above are only used to illustrate the technical solutions of the present invention, rather than to limit the same. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that the technical solutions described in the aforementioned embodiments may still be modified, or some of the technical features may be replaced by equivalents. Such modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention, and should all be included in the protection scope of the present invention.

Claims

1. A method for monitoring the state of a traffic light, characterized in that: Applied to a monitoring terminal, the monitoring terminal is connected to a monitoring platform and current sensors of a plurality of traffic lights, the method comprising: Acquire current sampling data of multiple traffic lights in the current period, wherein the current sampling data includes the power supply current of the traffic lights at each moment in the current period and the main red light current in each direction; Based on the current sampling data of the multiple traffic lights in the current period, calculate the current characteristics of each traffic light in the current period; the current characteristics include the opening average value, the closing average value, the maximum average current difference and the average current difference of each sliding time window in the current period; Based on the current characteristics of each traffic light in the current period, fault diagnosis is performed to determine the operating status of each traffic light; If the operating state of any traffic light is a fault state, first reporting information is generated and sent to the monitoring platform based on the operating state, location information and occurrence time of any traffic light.

2. The method for monitoring the state of a traffic signal lamp according to claim 1, characterized in that: The calculating the current characteristics of each traffic light in the current period based on the current sampling data of the plurality of traffic lights in the current period includes: Generate current sampling samples based on the current sampling data of each traffic light in the current period, wherein each sample in the current sampling samples is the current at any moment in the current period; Sort each sample in the current sampling sample in order of size to obtain a current sorting sample; The current sorting samples are split to obtain open samples and closed samples; the open samples are the first T samples when sorted from large to small, and the closed samples are the first T samples when sorted from small to large; wherein T is a positive integer; Based on the open samples, an open average value is calculated; the open average value is used to characterize the average current when the traffic light is turned on; Based on the closing samples, a closing average value is calculated; the closing average value is used to characterize the average current when the traffic light is closed; determining a maximum average current difference based on the open average and the closed average; Based on the current sorting samples, a sliding time window calculation is performed to obtain an average current difference in each sliding time window.

3. The method for monitoring the state of a traffic signal lamp according to claim 2, characterized in that: The method of performing fault diagnosis based on the current characteristics of each traffic light in the current period and determining the operating state of each traffic light includes: Based on the maximum average current difference and / or the average current difference of each sliding time window, a threshold value judgment is performed to determine whether the traffic light is working normally; If the traffic light works normally, determine that the operation state of the traffic light is normal; and update the normal opening current and the normal closing current of the traffic light based on the opening average value and the closing average value, and the average current difference; If the traffic light works abnormally, the number of times the traffic light works abnormally continuously is recorded; If the number of times that the traffic light continuously operates abnormally is greater than a preset number, it is determined that the operating state of the traffic light is a fault state.

4. The method for monitoring the state of a traffic signal lamp according to claim 1, characterized in that: Any current data in the current sampling data is obtained by: Acquire multiple sampling values ​​of the current sensor corresponding to any current data in the current period; Determine multiple acquisition voltages based on multiple sampling values ​​of the current sensor corresponding to the any current data in the current period, the resolution of the acquisition device, and the reference voltage of the acquisition device; Based on the multiple collected voltages, performing DC offset removal and square sum calculation to obtain a voltage square sum; Based on the voltage square sum, the detection resistor and the turns ratio of the current sensor, a root mean square calculation is performed to obtain the arbitrary current data.

5. The method for monitoring the state of a traffic signal lamp according to any one of claims 1 to 4, characterized in that: After performing fault diagnosis based on the current characteristics of each traffic light in the current period and determining the operating state of each traffic light, the method further includes: If the target traffic light is in a fault state, an input vector is generated based on the current characteristics of the target traffic light; the target traffic light is any one of the multiple traffic lights; Based on the input vector and a preset fault detection model, obtaining the fault type of the target signal light; Determine the impact range of the target signal light when the target signal light fails based on the fault type of the target signal light and the location information of the target signal light; Determine a repair time limit based on the fault type and the impact scope; Generate second reporting information based on the fault type, impact scope and repair time limit of the target signal light; The second reporting information is sent to the operation and maintenance personnel.

6. A method for monitoring the state of a traffic light, characterized in that: Applied to a monitoring platform, the monitoring platform is connected to a plurality of monitoring terminals, each monitoring terminal is connected to a plurality of current sensors of traffic lights, the method comprises: Acquire current sampling data of multiple traffic lights in the current period reported by multiple monitoring terminals, wherein the current sampling data includes the power supply current of the signal machine at each moment in the current period and the main red light current in each direction; Based on the current sampling data of the plurality of traffic lights, the current characteristics of each traffic light in the current period are calculated; the current characteristics include the opening average value, the closing average value, the maximum average current difference and the average current difference of each sliding time window in the current period; Based on the current characteristics of each traffic light in the current period, fault diagnosis is performed to determine the operating status of each traffic light; If the operating state of any traffic light is a fault state, an operation and maintenance instruction is generated based on the operating state, location information and occurrence time of the traffic light, and the operation and maintenance instruction is used to instruct the operation and maintenance personnel to inspect and repair the traffic light.

7. The method for monitoring the state of a traffic signal lamp according to claim 6, characterized in that: The method further comprises: Obtain current sampling data when multiple traffic lights fail during a historical period, as well as the failure type of each traffic light; Extract features based on circuit sampling data when each traffic light fails, and determine the current features of each traffic light; Based on the current characteristics of each traffic light and the fault type of each traffic light, a plurality of training samples are generated; Based on the multiple training samples, a fault detection model is constructed using a decision tree algorithm.

8. The method for monitoring the state of a traffic signal lamp according to claim 7, characterized in that: The current characteristics also include the open average value, the closed average value, the maximum average current difference, the average current difference of each sliding time window, the current peak value, the current valley value, the current fluctuation range, the current change trend, the current change cycle and each harmonic current of the set time period before and after the traffic light fails; The method of constructing a fault detection model based on the multiple training samples and using a decision tree algorithm includes: Initialize the decision tree model, determine the constraints of the decision tree model, the constraints include the maximum depth of the decision tree, the maximum number of leaf nodes and the minimum number of node samples; traverse each current feature, and calculate the Gini index of each current feature; Based on the Gini index of each current feature, the root node of the decision tree model is determined; Based on the root node of the decision tree model and the multiple training samples, node splitting is performed to determine multiple subsets, each subset corresponding to a branch in the decision tree model; Based on the multiple subsets and the constraints, construct a decision tree model; The constructed decision tree model is pruned to determine the fault detection model.

9. The method for monitoring the state of a traffic signal lamp according to any one of claims 6 to 8, characterized in that: The method further comprises: Receiving first reporting information and / or second reporting information returned by each monitoring terminal; Extracting current sampling data of each fault signal lamp based on the first reporting information and / or the second reporting information returned by each monitoring terminal; Based on the current sampling data of each fault signal lamp, extract the current characteristics of each fault signal lamp; Determine the fault type of each fault signal lamp based on the current characteristics of each fault signal lamp and the fault detection model; Based on the fault type, location information and occurrence time of each faulty signal light, determining whether there is a collective fault in the traffic signal system; the collective fault indicates that the fault causes of multiple faulty signal lights are the same; If there is a collective fault in the traffic signal system, a fault analysis result is generated based on the location information of the multiple faulty signal lights that have the collective fault, wherein the fault analysis result includes the fault type, location information and suspected fault cause of the multiple faulty signal lights that have the collective fault; Based on the fault analysis result, generating third reporting information; The third reporting information is sent to the operation and maintenance personnel.

10. A traffic signal system, characterized in that: The traffic signal system includes an electronic device, which includes a memory and a processor. The memory stores a computer program, and the processor is used to call and run the computer program stored in the memory to execute the method as described in any one of claims 1 to 5, or any one of claims 6 to 9.

Images