PID (Proportion Integration Differentiation)-based dynamic step size signal control timing scheme generation method

Abstract

The invention relates to the field of traffic signal control, and discloses a PID (Proportion Integration Differentiation)-based dynamic step size signal control timing scheme generation method, which comprises the following steps of: dynamically adjusting signal timing by collecting data such as traffic flow, vehicle speed and queuing length in real time and combining a PID control algorithm; specifically, the system calculates the control deviation and takes the control deviation as the input of a PID algorithm, and adjusts the control deviation of each signal steering according to the weighted synthesis of proportion, integration and differential terms; and finally realizing optimization of a signal control scheme by aggregating the adjustment step length of each steering to a stage and generating a new signal period and a new green signal ratio. The method has high adaptability and stability to various traffic detection data sources, signal timing can be accurately adjusted according to real-time traffic demands, the traffic efficiency of traffic flow is remarkably improved, traffic congestion is reduced, and the intelligent level of urban traffic management is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of traffic signal control, and specifically to a method for generating a dynamic step signal control timing plan based on PID. Background Art

[0002] Currently, with the continuous growth of urban traffic flow, traditional fixed detectors are difficult to meet the needs of rapid large-scale intelligent signal control intelligent information upgrade. Such detection devices not only have high thresholds for hardware layout and expensive installation costs, but also are accompanied by complex maintenance problems, such as network interruption, device offline, and data loss, which affect the stability of the signal control system. At the same time, since many existing signal control systems are still based on fixed cycles and preset green light times and lack the ability to dynamically respond to real-time traffic flow, they cannot be optimized and adjusted during traffic peaks or emergencies, resulting in waste of traffic resources and a decline in road capacity. In addition, relying on the method of hardware sensors makes it challenging to improve the synchronization of the overall urban signal control level on a large scale.

[0003] In contrast, the popularity of Internet mobile travel platforms and navigation software has brought rich intelligent connected sampling vehicle trajectory data. Such data can flexibly cover most of the urban road network and can be used to detect key indicators such as traffic delays, parking times, and queue lengths. Since these data sources come from navigation software or shared travel platforms and do not require additional hardware installation, they have the advantages of low cost and flexible maintenance, providing potential low-cost high-quality data support for signal control within the city. However, due to the uncertainty of the distribution of sampling vehicles, the stability and coverage of their detection indicators are affected by the operation of sampling vehicles, limiting the direct use of connected sampling vehicle data.

[0004] Facing these problems, although some signal control algorithms have been proposed in the prior art, they often only optimize for a single traffic indicator, lack a comprehensive analysis of multi-source data, or need to rely on complex data fusion methods to obtain detection information, resulting in greater implementation and deployment difficulties. Therefore, there is an urgent need for a signal control algorithm architecture that can integrate multi-source data and is simple and efficient in current urban traffic signal control to reduce the dependence on the completeness of a single data source, make full use of existing multiple traffic data resources, and achieve flexible and efficient optimization of urban signal control. Summary of the Invention

[0005] In view of the deficiencies of the prior art, the present invention provides a method for generating a signal control timing scheme with a dynamic step based on PID, which can dynamically adjust the signal timing according to the real-time traffic flow, improve the intelligent signal of traffic passing efficiency, has strong algorithm adaptability and high fault tolerance, reduces the threshold of multi-source traffic flow data participating in signal control, uses non-signal control service data sources such as intelligent connected vehicles for signal control, has a simple algorithm and is easy to deploy, greatly reduces the signal control detection cost, and gives play to the social benefits of intelligent connected vehicle data resources.

[0006] To achieve the above objectives, the present invention is realized through the following technical solutions: A method for generating a signal control timing scheme with a dynamic step based on PID, including the following steps: Obtain real-time traffic flow data and historical signal control schemes; Calculate the control deviation of each turning, and the control deviation is the difference between the actual green light time and the required green light time; Determine the dynamic adjustment step of each turning based on the PID control algorithm; Aggregate the dynamic adjustment steps of each turning to the corresponding signal control stage; Adjust the stage time according to the aggregated steps to generate a new signal control scheme.

[0007] Preferably, the real-time traffic flow data includes at least one of coil detection flow, radar-vision detection flow, and vehicle networking data.

[0008] Preferably, the required green light time is calculated by a required flow rate model: Required flow rate model: Calculate the number of vehicles expected to pass during the cycle based on the required flow rate, and the required green light time.

[0009] Preferably, the input of the PID control algorithm is the control deviation after confidence limit, and the calculation method of the control deviation is as follows: raw error (t)=capacity greentime (t)-demand greentime (t); error(t)=max(min(raw error (t),confidence(t)×base greenstep ),-confidence(t)× base greenstep ); Among them, raw error (t) represents the initial control deviation at the current moment t; capacity greentime (t) represents the green light time set by the signal control scheme at the current moment t; demand greentime(t) represents the required green light time calculated at the current moment t; error(t) represents the control deviation after confidence limit at the current moment t; confidence(t) represents the confidence range of the control deviation calculated at the current moment t; base greenstep is the basic amplitude of single-step green light adjustment, which is the amplitude above the median of the single-step green light adjustment amplitude of this intersection statistically from historical data.

[0010] Preferably, the calculation of the confidence level satisfies: where base confidence represents the basic confidence level statistically based on historical data, the number of sampled vehicles represents the number of effective sampled vehicles at the current moment, and N represents a preset normalization coefficient.

[0011] Preferably, if there are I data sources of multiple sources for cross-validation, the control deviations corresponding to each data source caliber i∈I are aggregated to obtain the final control deviation input to the PID model. The calculation formula is: error(t) = MEAN i∈I (error i (t)) / (1 + STD i∈I (sign(error i (t)))) where error i (t) is the control deviation of the current single data source caliber, i∈I represents the index range of all data source calibers, MEAN(·) is the mean function, STD(·) is the standard deviation function, and sign(·) is the positive and negative value sign function.

[0012] Preferably, the dynamic adjustment step size is calculated by the following formula: error predict (t) = Kp·error(t) + Ki·(error(t - 1) + error(t - 2)) + Kd·sign(error(t) - error(t - 1))·error(t); control step (t) = -error predict (t); where error predict (t) represents the error prediction value at the current moment t; control step(t) represents the dynamically adjusted step size calculated at the current moment t; error(t) represents the control deviation after confidence limit at the current moment t; error(t-1) and error(t-2) represent the control deviations at the previous two moments; sign(error(t)-error(t-1)) is the sign function of the deviation change direction; Kp, Ki, and Kd represent the proportional, integral, and differential coefficients respectively.

[0013] Preferably, aggregate the steering control deviations within the phase, select the maximum steering control deviation as the phase control deviation, and adjust the phase time according to the aggregated step size to satisfy: capacity greentime (t + 1) = capacity greentime (t) + control step (t); Among them, capacity greentime (t + 1) represents the phase green light time after step size adjustment; capacity greentime (t) represents the phase green light time actually allocated by the current signal timing plan; control step (t) represents the adjustment step size of the phase green light time.

[0014] Preferably, the step of adjusting the phase time according to the aggregated step size, calculating the new cycle and green ratio, and generating a new signal control plan further includes: Cycle = ∑ each phase time; Green ratio = phase green light time / cycle; Among them, the cycle represents the total duration of the signal plan; the green ratio represents the proportion of the green light time of each phase in the cycle.

[0015] The present invention also provides a device for generating a dynamic step size signal control timing plan based on PID, including: A data acquisition module for collecting real-time traffic flow data and historical signal control plans; A control deviation calculation module for calculating the steering control deviation based on the supply-demand time difference; A PID control module for generating a dynamically adjusted step size according to the PID algorithm; A phase aggregation module for aggregating the steering step sizes to the corresponding signal control phases; A plan generation module for adjusting the phase time and generating a new signal control plan.

[0016] The present invention provides a method for generating a dynamic step size signal control timing plan based on PID. It has the following beneficial effects: 1. By collecting traffic flow, vehicle speed, and queue length data in real time and dynamically adjusting the green light time in combination with the proportional, integral, and differential terms of the PID algorithm, the system can quickly respond to the instantaneous fluctuations of traffic flow and achieve precise matching of intersection timing and demand.

[0017] 2. The integral term of the PID algorithm of the present invention effectively eliminates long-term cumulative errors, and the confidence mechanism suppresses sudden noise interference. The two cooperate to reduce the deviation of green light time allocation, thereby shortening the vehicle waiting time and alleviating the congestion problem during peak hours.

[0018] 3. The present invention introduces a confidence threshold to limit the adjustment step size amplitude and combines the sign function to predict the change trend of errors, avoiding over-adjustment caused by detection errors or temporary abnormal traffic flows, and ensuring a smooth transition of signal cycle switching.

[0019] 4. The present invention adopts the maximum step size aggregation strategy within a stage, preferentially meets the traffic flow demands of key turns, and dynamically calculates the green signal ratio and the total cycle duration at the same time, realizing the balanced distribution of green light time in each phase and improving the overall traffic capacity of the intersection.

[0020] 5. This method deeply integrates PID dynamic control, multi-source data fusion, and real-time optimization algorithms, breaks through the limitations of traditional fixed timing, provides an intelligent signal control solution with strong adaptability, high stability, and scalability for urban traffic networks, and promotes the development of intelligent transportation systems towards high-efficiency intensification. BRIEF DESCRIPTION OF THE DRAWINGS

[0021] Figure 1 is one of the schematic diagrams of the method flow of the present invention; Figure 2 is the second schematic diagram of the method flow of the present invention; Figure 3 is the schematic diagram of the device structure of the present invention.

[0022] Among them, 10. Data acquisition module; 20. Control deviation calculation module; 30. PID control module; 40. Stage aggregation module; 50. Scheme generation module. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0023] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the drawings of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all of the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.

[0024] Please refer to the attached Figure 1 - attached Figure 2, the present invention provides a method for generating a signal timing plan based on PID dynamic step size control, which can dynamically adjust the green light time according to real-time traffic data, enabling the signal timing plan to adapt to changes in traffic demand and thus improving road traffic efficiency.

[0025] The core idea of the present invention is: using the PID control algorithm to calculate the dynamic adjustment step size for each turn of the signal lamp, and forming a new signal control plan through aggregation optimization to ensure that the signal timing plan has a good response ability to traffic flow changes.

[0026] As Figure 1 shown, the method for generating a signal timing plan based on PID dynamic step size control may include the following steps: S1. Obtain real-time traffic flow data and historical signal control plans; S2. Calculate the control deviation for each turn; S3. Determine the dynamic adjustment step size for each turn based on the PID control algorithm; S4. Aggregate the dynamic adjustment step sizes for each turn into the corresponding signal control phases; S5. Adjust the phase time according to the aggregated step sizes to generate a new signal control plan.

[0027] The following is a detailed description of each step in the method of the present invention, comprehensively elaborating on the specific implementation principles, technical details, and processes for each step.

[0028] For step S1, in this embodiment, real-time traffic flow data of the intersection is obtained through various sensors and data acquisition devices. The sources of these data include fixed traffic detectors, video surveillance systems, and vehicle-to-everything (V2X) devices.

[0029] Specifically, the fixed traffic detectors include geomagnetics, induction loops, and microwave radars, which are used to monitor traffic parameters such as road surface flow, vehicle speed, and occupancy rate in real time. The video surveillance system estimates the queue length and flow information of the vehicle flow through image processing technology. The vehicle-to-everything (V2X) system transmits the position information and driving status of vehicles in real time through wireless communication technology to provide traffic data.

[0030] The specific data acquisition methods are as follows: Traffic flow data: The traffic flow at the intersection or section is obtained through devices such as geomagnetic sensors and induction loop detectors. The flow data can be expressed as the number of vehicles passing through a certain section per unit time.

[0031] Vehicle speed data: The average vehicle speed in the traffic flow is obtained through microwave radars, video surveillance, and V2X data. The vehicle speed can reflect the smoothness of the traffic flow and thus affect the estimation of the green light time requirement.

[0032] Occupancy data: Occupancy refers to the proportion of vehicles occupying a road section and is used to estimate traffic demand intensity. It is measured by devices such as geomagnetism and coils.

[0033] Queue length data: The real-time queue length of the vehicle fleet is obtained by using a video surveillance system and used as the basis for evaluating traffic signal demand.

[0034] The system performs preliminary processing on the raw data collected by the above devices and outputs real-time statistical values of traffic flow, vehicle speed, occupancy, and queue length. These data are further input into the real-time calculation module to generate a preliminary estimate of the demand green time. The core basis of the demand green time calculation model is the comparison between real-time traffic flow and historical signal control schemes.

[0035] The historical signal control scheme data (including past signal timing schemes and actual green times) obtained by the system are used as background information. These historical data can provide reference for the adjustment of the current signal timing scheme. The data of historical signal control schemes are sourced from the signal timing schemes stored in the traffic signal control system, including green times and cycle information for each phase. Historical schemes are used to reflect the signal control rules at intersections and the changes in time-period traffic demands.

[0036] All the obtained data need to be passed into the signal control calculation module for subsequent calculations. Through the comprehensive analysis of the real-time traffic flow model and historical signal control scheme data, the calculation module outputs the demand green time for each phase as the input for the subsequent PID control algorithm.

[0037] In this embodiment, the association method between real-time data and historical data is as follows: By comparing the differences between the green times in the historical signal control scheme and real-time traffic flow data, a demand-based green time estimate is generated. This process specifically depends on the changes in real-time traffic flow data and the historical backtracking of past data.

[0038] The data flow process is as follows: 1. Data collection: Real-time traffic data, including traffic flow, vehicle speed, occupancy, etc., are collected through multiple sensors.

[0039] 2. Data processing: The raw data are analyzed and converted by the data processing module to calculate a preliminary estimate of the demand green time.

[0040] 3. Historical data integration: The signal timing data in the historical signal control scheme are combined with real-time data for analysis and comparison to obtain the final estimated value of the demand signal timing.

[0041] Through the above process, the traffic flow data at intersections and historical signal control schemes can be obtained in real time, and based on these data, input can be provided for PID control to ensure the flexibility and accuracy of signal timing.

[0042] For step S2, in this embodiment, the system needs to calculate the control deviation for each turn, and adjusts the signal timing plan by comparing the difference between the actual green time and the required green time to ensure that the traffic signal can flexibly respond to the real-time traffic demand. The control deviation provides the necessary input for the subsequent PID control.

[0043] The calculation of the required green time is based on the demand flow rate model. The required green time is calculated by the following formula: demand greentime (t) = T headway × demand flowrate (t) × cycle time Where: demand greentime (t) is the required green time for the turn within a single cycle at the t-th moment, which mainly depends on the demand flow rate. The statistical method and confidence level of the required green time within a single cycle calculated by the common demand flow rate caliber of the connected sampled vehicles are shown in Table 1; T headway is the headway of the saturation flow rate at the intersection; demand flowrate (t) is the traffic demand flow rate at the t-th moment. Here, the flow rate needs to be converted to the flow rate per 1 s according to the unit time; cycle time is the signal cycle at the t-th moment.

[0044] This model is based on the principle of traffic demand-driven adaptive signal control. By real-time monitoring the traffic flow changes, it dynamically adjusts the green time to optimize the road traffic efficiency. Its core idea is to determine the allocation ratio of the green time according to the number of vehicles expected to pass through per unit time (i.e., the demand flow rate), ensuring that the traffic demand in all directions is reasonably met, thereby reducing vehicle delays and queue lengths.

[0045] The calculation of the control deviation is based on the difference between the required green time and the actual green time. The specific calculation formula is as follows: raw error (t) = capacity greentime (t) - demand greentime (t) Where: capacity greentime (t) is the green time set by the current signal control plan, that is, the actual green time allocated to each turn in the signal cycle; raw error (t) is the calculated raw control deviation, indicating the gap between the actual green time and the required green time of the signal.

[0046] The control deviation reflects the difference between the actual signal timing and the traffic demand. If the control deviation is positive, it indicates that the current green light time is relatively long; if it is negative, it means the green light time is relatively short. The system dynamically adjusts the signal timing plan based on this deviation.

[0047] Since data such as traffic flow, queue length, and vehicle speed may be noisy or unstable, the present invention introduces a confidence level to limit the control deviation, ensuring that the signal control system does not generate excessive errors when processing unstable data. The calculation formula for the confidence level is: confidence(t) = base confidence + context confidence (t) Where: base confidence is the basic confidence level obtained based on historical data and long-term accumulation; it is the average correlation of the demand flow rate index data of this caliber and the demand flow rate caliber index data recognized by the business.

[0048] context confidence (t) is the confidence level fine-tuned according to the current sampled number of vehicles, usually adjusted based on the sampled number of vehicles.

[0049] The restricted control deviation directly affects the subsequent PID control steps. The specific calculation formula for the control deviation of the restricted demand green light caliber is as follows: error(t) = max(min(raw error (t), confidence(t) × base greenstep ), -confidence(t) × base greenstep ) Where, base greenstep is the basic adjustment range of the green light, which is related to the change range of the intersection traffic flow. It is the range above the median of the single-step green light adjustment range of this intersection under historical statistical data, and the default value is 5s.

[0050] Through this formula, the amplitude of the control error is limited within the range of confidence(t) × base greenstep , thus ensuring the stable response of the system.

[0051] To further clarify how to set base confidence and context confidence (t) in different situations, this embodiment provides reference values for the confidence level of the intelligent connected vehicle data source indicators that vary greatly with the sampled vehicle penetration rate, as shown in Table 1, which reflects the reference values of the confidence level under different demand green light time models: Table 1 Schematic of the calculation method for the green light time related to the connected sampling vehicle and the reference value of confidence level The data in the table shows that according to different calculation models for the green light time requirements, the values of base confidence and context confidence (t) will be different. In the case of having multiple calculation models for the green light time from multiple data sources, the control deviations of different data sources can be cross-validated and aggregated, so as to more accurately reflect the confidence level adjustment under different traffic patterns. The formula for aggregating the control deviations of all detection calibers is as follows: error(t) = MEAN i∈I (error i (t)) / (1 + STD i∈I (sign(error i (t)))) Among them, error i (t) is the control deviation of the current single data source caliber, i ∈ I represents the index range of all data source calibers, MEAN(·) is the mean function, STD(·) is the standard deviation function, and sign(·) is the positive and negative value sign function.

[0052] The physical implementation of this process obtains real-time traffic data (traffic flow, vehicle speed, queue length, etc.) through traffic monitoring devices (such as geomagnetic sensors, video monitoring, vehicle networking devices, etc.), and calculates the required green light time through the data processing module. Then, the actual green light time is calculated through historical signal control data, and finally the control deviation is calculated and restricted through the confidence level mechanism.

[0053] By introducing the flow-density-speed model and the queue length model to calculate the required green light time, and then restricting the control deviation through the confidence level mechanism, the present invention can dynamically adjust the signal timing to ensure that the change of the green light time matches the traffic flow demand, thereby effectively reducing traffic delays and congestion and improving the road traffic efficiency.

[0054] For step S3, in this embodiment, the system uses the PID control algorithm to calculate the dynamic adjustment step length for each turn to adjust the signal timing to ensure the effective diversion of traffic flow and the adaptive ability of the system. The PID control algorithm generates an adjustment signal through real-time calculation to control the dynamic change of the green light time.

[0055] First, the system obtains the control deviation error(t) through the calculation in step S2. This deviation is used as the input of the PID control algorithm for dynamic adjustment calculation. The PID control algorithm consists of three parts: the proportional term, the integral term, and the derivative term, thereby adjusting the signal timing to ensure the stability and accuracy of the system.

[0056] Specifically, the calculation formula for the prediction error error predict (t) is: error predict (t) = Kp·error(t) + Ki·(error(t - 1) + error(t - 2)) + Kd·sign(error(t) - error(t - 1))·error(t) Where: error(t) is the control deviation after confidence limit at the current moment, reflecting the gap between the current green light time and the required green light time; Kp is the proportional coefficient, controlling the influence of the current error on the step size adjustment, and the recommended value is between 0.5 and 1; Ki is the integral coefficient, controlling the influence of error accumulation on the step size adjustment, and the recommended value is about half of Kp; Kd is the derivative coefficient, controlling the influence of the error change rate on the step size adjustment, and the recommended value is about one-tenth of Kp; sign(error(t) - error(t - 1)) is the sign function, used to indicate the change trend of the error to assist the derivative term calculation; error(t - 1) and error(t - 2) are the error values at the previous moment and the moment before the previous moment respectively, reflecting the accumulation and trend of the error.

[0057] This formula comprehensively considers the current error, error history, and error change rate through the three terms of proportional, integral, and derivative to accurately calculate the prediction error error predict (t).

[0058] Then, the calculation formula for the control step size control step (t) is: control step (t) = -error predict (t) This formula obtains the adjustment step size of the signal timing at the current moment through the reverse operation of the prediction error. The negative sign indicates that the adjustment direction is opposite to the error direction, ensuring that the green light time is shortened when the error is positive and the green light time is extended when the error is negative.

[0059] During implementation, each coefficient (Kp, Ki, Kd) of the PID control algorithm is adjusted according to the actual traffic conditions. The setting of the proportional coefficient Kp controls the sensitivity of the signal response; the integral coefficient Ki is used to eliminate the errors accumulated over a long time to ensure the stability of the system; the derivative coefficient Kd is mainly used to suppress the error fluctuations and improve the robustness of the system.

[0060] In practical applications, the system calculates and updates the green light duration of the signal every second. By controlling the control step (t), the signal control system can flexibly adjust the green light time of each turn according to the real-time traffic flow changes to ensure that the traffic signal is consistent with the traffic flow demand.

[0061] Through the above steps, the system can dynamically adjust the signal control plan according to the real-time traffic flow and queue conditions, effectively reduce traffic congestion, improve the road traffic efficiency, and achieve intelligent traffic signal control.

[0062] In a preferred embodiment, in order to aggregate the adjustment steps of multiple turns into the same signal control stage, the system needs to compare the control steps of all turns within the same signal stage and select the maximum value as the final adjustment step of this signal stage.

[0063] For a specific signal control stage k, assume that the turns involved in this stage are T k ={t1,t2,...,t n},where t1,t2,...,t n are the turns of this stage. Then the adjustment step calculation formula is: Where: is the final adjustment step of signal control stage k, representing the maximum value of the adjustment steps of all turns within this stage; control step (t i ) is the dynamic adjustment step of turn t i , reflecting the adjustment amount of the green light time of this turn.

[0064] Through this aggregation method, the system ensures that in each signal control stage, the green light time adjustments of all turns can uniformly respond to the most urgent traffic demands. Even if the adjustment demand of a certain turn is large, the adjustment step of the final signal control stage is still based on the maximum adjustment amount of this turn, so as to ensure that the timing adjustment of this stage fully meets the most urgent needs of traffic flow changes.

[0065] The implementation of this aggregation method can effectively avoid the overall adjustment within the signal control phase being too small due to minor adjustments in individual turns, ensuring that the adjustment step sizes for each signal phase have high responsiveness and adaptability. In situations where traffic flow is relatively complex or variable, the maximum value aggregation method can avoid insufficient local adjustments, thereby enhancing the optimization effect of the traffic signal control system.

[0066] For step S4, in this embodiment, the green light time for each phase is corrected based on the previously calculated dynamic adjustment step size control step (t), thereby optimizing the signal timing of the traffic signal.

[0067] The system calculates the control step size control step (t) through the PID control algorithm in step S3, and this step size is used to adjust the green light time for each phase. According to the magnitude of this control step size, the system corrects the green light time for each phase within the current signal cycle. Specifically, the system uses the following formula to calculate the corrected green light time capacity greentime (t + 1): capacity greentime (t + 1) = capacity greentime (t) + control step (t) Where: capacity greentime (t + 1) is the green light time for the corrected phase (unit: seconds), representing the final value of the green light time for the current turn after PID adjustment; capacity greentime (t) is the actual allocated green light time for the current phase, from the setting of the previous signal control scheme; control step (t) is the dynamic adjustment step size calculated in step S3, reflecting the adjustment amount of the signal timing.

[0068] Finally, the system applies capacity greentime (t + 1) to the correction of the green light time for each phase of this signal phase, thereby ensuring that the signal timing within the entire signal cycle can accurately reflect the traffic demand for each phase, improving the traffic capacity of the traffic flow and reducing traffic congestion.

[0069] For step S5, in this embodiment, by adjusting the time for each signal phase according to the corrected green light time capacity greentime (t + 1), a new signal control scheme is generated. This process further optimizes the signal timing to adapt to real-time traffic demand changes.

[0070] Specifically, correct the green light time capacity greentime (t + 1) is the correction amount of the green light time for each signal phase in the next cycle, and it is applied to the current signal control scheme.

[0071] Next, the system calculates the new signal cycle and green signal ratio. The signal cycle period(t) is the total duration of all signal phases within a complete signal cycle, and the calculation formula is: Where: period(t) is the new signal cycle, the total duration; is the adjusted green light time for each signal phase k; N is the number of signal phases within the signal cycle.

[0072] On this basis, the system calculates the green signal ratio for each signal phase That is, the proportion of the green light time of each signal phase in the entire signal cycle. The calculation formula for the green signal ratio is: Where: is the green signal ratio of signal phase k, indicating the proportion of the green light time of this phase in the entire signal cycle; is the adjusted green light time of signal phase k; period(t) is the new signal cycle, the total duration.

[0073] Through the above calculations, the system obtains the new signal cycle and green signal ratio, thereby generating a new signal control scheme. This new control scheme can more accurately reflect the real-time traffic demand, ensure that the green light time of each signal phase matches the traffic flow demand, thereby optimizing the road traffic efficiency and reducing traffic congestion.

[0074] Finally, the generated new signal control scheme is applied to the next signal cycle, and the system will arrange the corresponding signal light cycle according to the new green light time to achieve continuous optimization and adjustment.

[0075] In summary, step S5 generates a new signal control scheme by adjusting the aggregated step size and combining the calculations of the cycle and green signal ratio. This scheme can timely adjust the signal cycle and green light time according to traffic flow changes, further enhancing the self-adaptability and optimization effect of traffic signals, thereby achieving more efficient traffic management.

[0076] The PID-based dynamic step signal control timing scheme generation device described below can be correspondingly referred to the PID-based dynamic step signal control timing scheme generation method described above.

[0077] Please refer to the appendix Figure 3 , the present invention also provides a PID-based dynamic step signal control timing scheme generation device, including: A data acquisition module 10 for collecting real-time traffic flow data and historical signal control schemes; A control deviation calculation module 20 for calculating the control deviation of each steering based on the supply-demand time difference; A PID control module 30 for generating a dynamically adjusted step size according to the PID algorithm; A phase aggregation module 40 for aggregating the steering step sizes to the corresponding signal control phases; A scheme generation module 50 for adjusting the phase time and generating a new signal control scheme.

[0078] The device of this embodiment can be used to execute the above method embodiment, and its principle and technical effect are similar, which will not be elaborated here.

[0079] Although the embodiments of the present invention have been shown and described, for those of ordinary skill in the art, it can be understood that various changes, modifications, substitutions, and variations can be made to these embodiments without departing from the principle and spirit of the present invention. The scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. A method for generating a dynamic step signal control timing scheme based on PID, characterized in that: The following steps are involved: Obtain real-time traffic flow data and historical signal control solutions; Calculate the control deviation of each turn, where the control deviation is the difference between the actual green light time and the required green light time; Determine the dynamic adjustment step size of each steering based on the PID control algorithm; Aggregate each steering dynamic adjustment step into the corresponding signal control stage; The phase time is adjusted according to the aggregated step size to generate a new signal control solution.

2. The method for generating a dynamic step signal control timing scheme based on PID according to claim 1, characterized in that: The real-time traffic flow data includes at least one of coil detection flow, radar detection flow, and Internet of Vehicles data.

3. The method for generating a dynamic step signal control timing scheme based on PID according to claim 1, characterized in that: The demand green time is calculated by the demand flow rate model: Demand flow rate model: The number of vehicles expected to pass through the cycle and the required green light time are calculated based on the demand flow rate.

4. The method for generating a dynamic step signal control timing scheme based on PID according to claim 1, characterized in that: The steering control deviation is calculated as follows: raw error (t)=capacity greentime (t)-demand greentime (t); error(t)=max(min(raw error (t),confidence(t)×base greenstep ),-confidence(t)× base greenstep ); Among them, raw error (t) represents the initial control deviation at the current time t; capacity greentime (t) represents the green light time set by the signal control scheme at the current time t; demand greentime (t) represents the demand green light time calculated at the current time t; error(t) represents the control deviation after confidence limit at the current time t; confidence(t) represents the demand green light confidence calculated at the current time t; base greenstep It is the green light to adjust the basic amplitude in single step.

5. The method for generating a dynamic step signal control timing scheme based on PID according to claim 4, characterized in that: The calculation of the confidence level satisfies: Among them, base confidence It represents the basic confidence based on historical data statistics, the number of sampled vehicles represents the number of valid sampled vehicles at the current moment, and N represents the normalization coefficient preset after historical data analysis.

6. The method for generating a dynamic step signal control timing scheme based on PID according to claim 4, characterized in that: The green light single-step adjusts the base amplitude base greenstep It is the amplitude above the median of the single-step adjustment amplitude in the historical data statistics of the intersection.

7. The method for generating a dynamic step signal control timing scheme based on PID according to claim 4, characterized in that: If there are multiple sources of data available for cross-validation, the control deviations of different data sources are calculated using the following formula to obtain the control deviations that are finally input into the PID model: error(t)=MEAN i∈I (error i (t)) / (1+STD i∈I (sign(error i (t)))); Among them, error i (t) is the control deviation of the current single data source caliber, i∈I represents the index range of all data source calibers, MEAN(·) is the mean function, STD(·) is the standard deviation function, and sign(·) is the sign function of positive and negative values.

8. The method for generating a dynamic step signal control timing scheme based on PID according to claim 1, characterized in that: The dynamic adjustment step size is calculated by the following formula: error predict (t)=Kp·error(t)+Ki·(error(t-1)+error(t-2))+Kd·sign(error(t)-error(t-1))·error(t); control step (t)=-error predict (t); Among them, error predict (t) represents the error prediction value at the current time t; control step (t) represents the dynamic adjustment step calculated at the current moment t; error(t) represents the control deviation after confidence limit at the current moment t; error(t-1) and error(t-2) represent the control deviations at the previous two moments; sign(error(t)-error(t-1)) is the sign function of the direction of deviation change; Kp, Ki, and Kd represent the proportional, integral, and differential coefficients, respectively.

9. The method for generating a dynamic step signal control timing scheme based on PID according to claim 1, characterized in that: Aggregate the steering control deviations within each stage, select the largest steering control deviation as the stage control deviation, and adjust the stage time according to the aggregated step size to meet the following requirements: capacity greentime (t+1)=capacity greentime (t)+control step (t); Among them, capacity greentime (t+1) represents the green light time after the step size is adjusted; capacity greentime (t) represents the green light time actually allocated by the current signal timing scheme; control step (t) represents the step length of green light time adjustment; The steps of adjusting the stage time according to the aggregated step length, calculating the new cycle and green-to-signal ratio, and generating a new signal control scheme include: cycle = ∑ each stage time = ∑ (each stage green light time + stage loss time); Green-to-signal ratio = stage green light time / cycle; Among them, the cycle represents the total duration of the signal scheme; the green-to-signal ratio represents the proportion of the green light time in each stage to the cycle.

10. A PID-based dynamic step signal control timing scheme generation device, used to execute the method according to any one of claims 1 to 9, characterized in that: include: Data acquisition module, used to collect real-time traffic flow data and historical signal control solutions; A control deviation calculation module calculates each steering control deviation based on the supply and demand time difference; PID control module, which generates dynamic adjustment step size according to PID algorithm; The stage aggregation module aggregates the steering step length to the corresponding signal control stage; The scheme generation module adjusts the stage time and generates a new signal control scheme.

Images