A traffic entropy-based bus priority traffic scheduling method, device and equipment

Abstract

The application discloses a kind of public transport priority traffic scheduling method, device and equipment based on traffic entropy, generates time information to station by obtaining bus position data, establishes time credit limit mechanism based on punctuality rate and historical application frequency to carry out differentiated resource allocation for each line;Determine the operating load index by executing multi-line resource competition analysis, formulate priority execution strategy based on quota constraint;Introduce traffic entropy theory, calculate the confusion degree caused by signal adjustment through speed variance and queue length change, generate green light timing scheme with entropy constraint;Analyze the road network ripple effect formed by traffic disturbance propagation, build entropy decreasing control mechanism through game equilibrium analysis and entropy propagation model;Formulate road network coordinated control strategy, realize unified control through hierarchical coordination architecture and time load clearing, finally form global scheduling strategy based on entropy optimization, realize the intelligent evolution control of traffic system from disorder to order.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of intelligent traffic control, in particular to a bus priority traffic scheduling method, device and equipment based on traffic entropy. BACKGROUND

[0002] With the continuous expansion of city size and the continuous growth of motor vehicle ownership, traffic congestion has become an important factor restricting the development of the city. Public transportation is considered as a key way to solve the urban traffic problem due to its high carrying efficiency and low per capita emission characteristics. As an important technical means to improve the level of public transport service, bus priority signal control can reduce the number of stops and delay time of public transport vehicles by providing priority at signalized intersections.

[0003] However, in the urban center area where public transport lines are dense, multiple lines often share the same road resources and signalized intersections. When multiple buses arrive at the same intersection at the same time or successively, how to reasonably allocate limited priority resources becomes a complex technical problem. The existing technology usually adopts a simple strategy of first-come-first-served or fixed priority, ignoring the actual operating conditions and service demand differences of different lines. In addition, signal priority measures at a single intersection will have a chain effect on downstream intersections, which may lead to uneven distribution of traffic flow and formation of new congestion points. Traditional methods lack evaluation and control means for such systematic effects, often resulting in the phenomenon of "pressing down the gourd to float up the dipper". At the same time, there is a natural contradiction between bus priority and the demand for social vehicle traffic, and how to ensure bus priority while maintaining the stability of overall traffic order requires more sophisticated and intelligent control strategies. SUMMARY

[0004] The present application provides a bus priority traffic scheduling method, device and equipment based on traffic entropy, aiming to solve the key technical problems of uneven resource allocation among multiple lines, difficulty in quantifying signal adjustment impact, and lack of coordinated optimization among intersections in traditional bus priority control. Through innovative technical means such as time credit management, traffic entropy calculation, game equilibrium analysis, and entropy decreasing control, the present application comprehensively and accurately processes key links such as bus arrival time prediction, dynamic allocation of time resources, entropy-constrained signal control, and coordinated regulation of road network disturbances, thereby realizing dynamic balance optimization of bus priority and traffic order based on entropy theory, and ultimately establishing an intelligent bus priority control system with entropy value prediction capability and self-adaptive regulation function.

[0005] The present application provides a bus priority traffic scheduling method, device and equipment based on traffic entropy, aiming to solve the key technical problems of uneven resource allocation among multiple lines, difficulty in quantifying signal adjustment impact, and lack of coordinated optimization among intersections in traditional bus priority control. Through innovative technical means such as time credit management, traffic entropy calculation, game equilibrium analysis, and entropy decreasing control, the present application comprehensively and accurately processes key links such as bus arrival time prediction, dynamic allocation of time resources, entropy-constrained signal control, and coordinated regulation of road network disturbances, thereby realizing dynamic balance optimization of bus priority and traffic order based on entropy theory, and ultimately establishing an intelligent bus priority control system with entropy value prediction capability and self-adaptive regulation function.

[0006] Collecting the trajectory coordinate sequence output by the bus positioning terminal, and generating arrival time information by performing time prediction processing on the trajectory coordinate sequence;

[0007] The arrival time information is used to perform time load assessment to generate time credit limits for each line. A time resource pool is established based on the time credit limits, and the time resource pool is dynamically allocated to obtain the time allocation weight for each line.

[0008] Based on the arrival time information, perform multi-line resource competition analysis to determine the operating load index of each line, generate time allocation requirements according to the time allocation weight and the operating load index of each line, and generate a time resource allocation scheme based on the time allocation requirements and time load balancing processing.

[0009] Based on the time resource allocation scheme, a quota-based priority execution strategy is formulated. The priority execution strategy is applied to the phase adjustment to generate signal changes. Traffic entropy value is calculated on the signal changes to obtain the current traffic congestion data. Phase extension analysis is performed on the traffic congestion data to generate a green light timing scheme.

[0010] Traffic disturbance propagation analysis is performed on the green light timing scheme to generate road network ripple effect. Inter-intersection game analysis is performed on the road network ripple effect to generate game equilibrium point. Entropy propagation analysis is performed on the road network ripple effect to obtain disorder diffusion parameters. Entropy reduction control mechanism is constructed using the disorder diffusion parameters and the game equilibrium point.

[0011] The entropy reduction control mechanism and the time resource allocation scheme are combined to generate full-path impact assessment data. Based on the full-path impact assessment data, a chain adjustment strategy is formulated, and based on the chain adjustment strategy, an inter-intersection resource balance scheme is generated.

[0012] Based on the resource balancing scheme between intersections, a hierarchical coordination mechanism is constructed to generate a comprehensive control scheme. The comprehensive control scheme is then verified by time load clearing to obtain the load balance status. Based on the load balance status, a comprehensive optimization strategy is generated to complete the bus priority traffic scheduling.

[0013] A second aspect of the present invention provides a public transport priority dispatching device based on traffic entropy, comprising:

[0014] The data acquisition module is used to collect the trajectory coordinate sequence output by the bus positioning terminal, and perform time prediction processing on the trajectory coordinate sequence to generate arrival time information.

[0015] The load assessment module is used to perform time load assessment on the arrival time information to generate time credit limits for each line, establish a time resource pool based on the time credit limits, and dynamically allocate the time resource pool to obtain the time allocation weight for each line.

[0016] The allocation processing module is used to perform multi-line resource competition analysis based on the arrival time information to determine the operating load index of each line, generate time allocation requirements based on the time allocation weight and the operating load index of each line, and generate a time resource allocation scheme based on the time allocation requirements by performing time load balancing processing.

[0017] The signal control module is used to formulate a quota-based priority execution strategy according to the time resource allocation scheme, apply the priority execution strategy to the phase adjustment to generate signal changes, calculate the traffic entropy value of the signal changes to obtain the current traffic chaos data, and perform phase extension analysis on the traffic chaos data to generate a green light timing scheme.

[0018] The game analysis module is used to perform traffic disturbance propagation analysis on the green light timing scheme to generate road network ripple effects, perform inter-intersection game analysis on the road network ripple effects to generate game equilibrium points, perform entropy propagation analysis based on the road network ripple effects to obtain disorder diffusion parameters, and use the disorder diffusion parameters and the game equilibrium points to construct an entropy reduction control mechanism.

[0019] The collaborative control module is used to generate full-path impact assessment data by combining the entropy reduction control mechanism and the time resource allocation scheme, formulate a chain adjustment strategy based on the full-path impact assessment data, and generate an inter-intersection resource balance scheme based on the chain adjustment strategy.

[0020] The integrated optimization module is used to construct a hierarchical coordination mechanism based on the resource balancing scheme between intersections to generate an integrated control scheme, perform time load clearing and verification on the integrated control scheme to obtain the load balance status, generate an integrated optimization strategy based on the load balance status, and complete the bus priority traffic scheduling.

[0021] A third aspect of the present invention provides a computer device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the program to implement the steps of a traffic entropy-based public transport priority scheduling method disclosed in the first aspect.

[0022] The beneficial effects of this invention are reflected in the following points: 1. By establishing a time credit quota mechanism and a dynamic time resource pool, differentiated quota allocation is carried out based on the punctuality rate and historical priority application frequency of each route. Real-time weight adjustment enables dynamic optimization of resource allocation, changing the traditional fixed allocation model. This allows routes with poor operating conditions and heavy service loads to receive more priority guarantees, while avoiding excessive resource occupation by individual routes, thus improving the overall operational efficiency of the public transportation system. 2. The concept of traffic entropy is introduced to quantitatively assess the degree of traffic chaos caused by signal changes. Chaos data is obtained by calculating the entropy value of speed variance and queue length changes, and game theory methods are used to analyze the interest relationships between intersections. This quantitative assessment system allows the impact of priority measures to be accurately predicted and controlled. Managers can anticipate potential negative effects before implementing priority and control the chaos within an acceptable range through an entropy reduction control mechanism, achieving a balance between priority effects and traffic order. 3. A complete technical system from single-point control to road network coordination has been constructed. By analyzing the propagation path of traffic disturbances, the propagation path of priority measures is identified, road network coordinated control strategies and resource balancing schemes between intersections are formulated, and unified management and control of the entire network is achieved through a hierarchical coordination mechanism. This avoids the transfer of problems caused by local optimization and ensures that the overall operation and stability of the road network are maintained while improving public transportation services.

[0023] It should be understood that the above general description and the following detailed description are exemplary and explanatory only, and do not limit this application. Attached Figure Description

[0024] The accompanying drawings illustrate specific examples of the technical solutions described in this invention and, together with the detailed embodiments, form part of the specification, serving to explain the technical solutions, principles, and effects of this invention.

[0025] Unless otherwise specified or defined, the same reference numerals in different figures represent the same or similar technical features, and different reference numerals may be used to represent the same or similar technical features.

[0026] Figure 1 This is a flowchart illustrating a public transport priority scheduling method based on traffic entropy according to the present invention.

[0027] Figure 2 This is a structural block diagram of a public transport priority dispatching device based on traffic entropy according to the present invention.

[0028] Figure 3 This is a schematic diagram of the structure of a computer device according to the present invention. Detailed Implementation

[0029] In the following description, specific details such as particular system architectures and techniques are set forth for illustrative purposes and not for limitation, in order to provide a thorough understanding of the embodiments of this application. However, those skilled in the art will understand that this application may also be implemented in other embodiments without these specific details. In other instances, detailed descriptions of well-known systems, apparatuses, circuits, and methods have been omitted so as not to obscure the description of this application with unnecessary detail.

[0030] It should be understood that, when used in this application specification and the appended claims, the term "comprising" indicates the presence of the described features, integrals, steps, operations, elements and / or components, but does not exclude the presence or addition of one or more other features, integrals, steps, operations, elements, components and / or a collection thereof.

[0031] As used in this application specification and the appended claims, the term "if" may be interpreted, depending on the context, as "when," "once," "in response to determination," or "in response to detection." Similarly, the phrase "if determined" or "if detected [the described condition or event]" may be interpreted, depending on the context, as meaning "once determined," "in response to determination," "once detected [the described condition or event]," or "in response to detection [the described condition or event]."

[0032] The technical solutions of the embodiments of this application will be described below.

[0033] like Figure 1 As shown, this embodiment of the invention provides a public transport priority scheduling method based on traffic entropy, including the following steps S110-S170:

[0034] Step S110: Collect the trajectory coordinate sequence output by the bus positioning terminal, and perform time prediction processing on the trajectory coordinate sequence to generate arrival time information.

[0035] Specifically, the bus positioning terminal outputs a real-time trajectory coordinate sequence. Each coordinate point includes four basic elements: longitude, latitude, elevation, and timestamp. The sampling frequency is set to 1Hz to ensure trajectory continuity. The positioning terminal integrates a GPS / BeiDou dual-mode receiver, an inertial measurement unit, and a vehicle speed pulse interface, generating high-precision trajectory coordinates through a multi-sensor fusion algorithm. The trajectory coordinate sequence is stored chronologically, forming a structured spatiotemporal data stream. Each coordinate point is also supplemented with auxiliary information such as positioning mode identifier, horizontal accuracy factor, and number of satellites. When the vehicle is traveling in an area with signal obstruction, the system automatically switches to inertial navigation mode, calculating the current position from the previous trajectory to ensure the integrity of the coordinate sequence. A real-time caching mechanism for the coordinate sequence is established, retaining the trajectory data of the most recent 10 minutes for predictive calculations, while historical data exceeding the time limit is automatically archived. An anomaly detection algorithm for the trajectory coordinates identifies and marks situations such as position jumps and speed anomalies, providing data quality indicators for subsequent processing.

[0036] The collected trajectory coordinate sequence undergoes time prediction processing. First, discrete coordinate points are interpolated using cubic spline interpolation to construct a continuous driving trajectory curve. The spherical distance and time interval between adjacent coordinate points are calculated to obtain the instantaneous speed sequence for each road segment. Kalman filtering is used to smooth the speed sequence, eliminating measurement noise and extracting the true speed change trend. Based on the processed speed data and remaining distance, an autoregressive moving average model is used to predict the arrival time at each station. The prediction model considers the influence of three types of factors: road segment characteristics, historical operating patterns, and current traffic conditions, and improves prediction accuracy through weighted combination. The generated arrival time information includes three core elements: predicted arrival time, upper and lower limits of the confidence interval, and prediction confidence score. An information update mechanism ensures that the arrival time is recalculated immediately upon receiving new coordinate points, achieving dynamic updating of arrival times and ultimately generating the final arrival time information.

[0037] Step S120: Perform time load assessment on arrival time information to generate time credit limits for each line, establish a time resource pool based on the time credit limits, and dynamically allocate time resources to obtain time allocation weights for each line.

[0038] In some embodiments, the step of performing time load assessment on the arrival time information to generate time credit limits for each line includes: analyzing the arrival time information to obtain arrival time deviation data for each line; generating on-time rate indicators for each line based on the arrival time deviation data; performing time load assessment based on the on-time rate indicators and historical priority application frequency to obtain load assessment results; and generating time credit limits for each line based on the load assessment results.

[0039] Arrival time information was analyzed to obtain arrival time deviation data for each route. Taking a major urban route as an example, the standard arrival time is 8:00 AM, and the actual estimated arrival time is 8:03 AM, resulting in a calculated time deviation of +3 minutes. Through seven consecutive days of data collection and statistical analysis, the time deviation data sequence for this route during the morning peak hours was [+2, +4, +1, +5, +3, +2, +4] minutes, with an average deviation of +3 minutes and a standard deviation of 1.4 minutes. A classification and statistical mechanism for deviation data was established, precisely classifying time deviations into four levels: more than 2 minutes ahead, on-time within ±2 minutes, slight delay of 2-5 minutes, and severe delay of more than 5 minutes. An outlier detection algorithm was used to identify and filter extreme deviation data caused by unforeseen events. When the deviation value exceeded three times the standard deviation of the normal range (±4.2 minutes), it was marked as an outlier and removed from the statistics. Different deviation tolerance standards are set for different types of routes: the deviation tolerance for Bus Rapid Transit (BRT) routes is set at ±1 minute, for regular bus routes at ±2 minutes, for community shuttle routes at ±3 minutes, and for night bus routes at ±5 minutes.

[0040] The basic on-time rate index for each route is calculated by comparing the number of arrivals within the tolerance range to the total number of arrivals. Using the deviation data of urban trunk routes [+2, +4, +1, +5, +3, +2, +4] minutes as a base, the number of arrivals within the ±2 minute tolerance range is 2 (+1 and +2 minutes). The total number of arrivals over 7 days is 168, of which 134 are on-time, resulting in an on-time rate of 79.8%. A multi-time-period weighted on-time rate calculation method is established, dividing the day into four periods: morning peak (7:00-9:00), off-peak (9:00-17:00), evening peak (17:00-19:00), and nighttime (19:00-7:00), with weighting coefficients of 1.5, 1.0, 1.5, and 0.8 respectively. A weighted calculation method is used to comprehensively evaluate the punctuality rate for different time periods. The punctuality performance during morning and evening peak hours has a greater impact on the overall evaluation, highlighting the operational importance of key periods. A punctuality rate rating standard is established, classifying routes into four levels based on their punctuality performance: Excellent (above 95%), Good (90%-95%), Satisfactory (80%-90%), and Needs Improvement (below 80%).

[0041] Using the generated on-time rate data as the core input for load assessment, a load assessment model is established. On-time rate serves as a measure of line operational quality, while historical priority application frequency reflects the line's demand for signal priority resources. Taking a city trunk line with an on-time rate of 79.8% as an example, combined with its historical priority application frequency of 3.5 times per hour, a weighted calculation is performed using the load assessment algorithm. The on-time rate weight coefficient is set to 0.6, and the application frequency weight coefficient is set to 0.4, resulting in a load index of 12.2 for this line. The load assessment algorithm adopts a weighted scoring model: L = 0.4×Ps + 0.3×Fs + 0.3×Cs, where Ps is the on-time rate score (calculated as the number of on-time arrivals divided by the total number of arrivals multiplied by 100), Fs is the application frequency score (calculated as the ratio of historical application frequency to standard application frequency multiplied by 100), and Cs is the passenger load score (calculated as the actual passenger load divided by the standard passenger load multiplied by 100). A load level classification standard was established, dividing routes into four levels based on load index values: light load (0-5), medium load (5-15), heavy load (15-25), and overload (over 25). This particular route, with a load index of 12.2, falls within the medium load range. A load trend analysis algorithm was used to calculate the load index changes over the past 30 days, identifying rising, falling, or stable trends in load status through comparative analysis. A load early warning mechanism was established: a heavy load warning is triggered when the route's load index exceeds 20 for three consecutive days, and an overload warning is triggered when it exceeds 25 for seven consecutive days, automatically sending warning information to the dispatch center. Differentiated load assessment parameters were set for different route types: a weighting coefficient of 0.7 was set for BRT routes emphasizing punctuality, 0.6 for regular routes, and 0.5 for feeder routes, reflecting the operational characteristics of different route types. Based on comprehensive evaluation and trend analysis, the final load assessment results were determined.

[0042] The load level and load index from the load assessment results are used as the basis for calculating credit limit allocation, and a credit limit allocation algorithm is established. Taking a line with a load index of 12.2 and a load level of medium load as an example, a base credit limit of 100 units is used as the starting value. The standard allocation ratio of 1.0 is maintained for medium load lines, an increase coefficient of 1.2 is set for heavily loaded lines, a decrease coefficient of 0.8 is set for lightly loaded lines, and an increase coefficient of 1.5 is set for overloaded lines. A secondary adjustment is made based on the line importance coefficient: 1.3 for trunk lines, 1.0 for secondary trunk lines, and 0.7 for branch lines. The final credit limit for this trunk line is calculated as 100 × 1.0 × 1.3 = 130 units. A credit limit tier system is established, dividing the credit limit into four levels: high limit (above 130), standard limit (100-130), base limit (70-100), and restricted limit (below 70). Different levels enjoy different priority application rights and usage restrictions. A dynamic credit limit adjustment mechanism was implemented, updating credit allocation weekly based on the latest load assessment results. Lines with consistently excellent performance received a 10% increase in credit limit, while lines with consistently poor performance received a 10% decrease. Credit usage rules were established, stipulating that each priority signal request consumes 5 credit units, and restricting non-urgent requests when the remaining credit limit falls below 20 units, ensuring the rational use of resources. Through the combined application of the credit limit allocation algorithm and the dynamic adjustment mechanism, precise allocation of time-based credit limits for each line was achieved.

[0043] All generated time-based credit limit data for each route is aggregated into a central resource pool, enabling unified scheduling and allocation of resources across the entire network. Taking a city's bus routes as an example, the credit limits for each route are different values ​​such as 130, 110, 90, and 120, resulting in a total resource pool capacity of 5500 units. A hierarchical management mechanism for the resource pool is established, allocating the total capacity to three levels in a 6:3:1 ratio: a core resource layer of 3300 units for daily priority signal allocation, a buffer resource layer of 1650 units for peak-hour demand increases, and an emergency resource layer of 550 units for handling sudden events and emergencies. Real-time monitoring parameters for the resource pool are set. When the resource utilization rate exceeds 80% of the total capacity (4400 units), a yellow alert is activated; when it exceeds 90% (4950 units), a red alert is activated, automatically initiating a resource conservation mode to restrict non-emergency requests. A load balancing algorithm for the resource pool is established to monitor resource usage in each area. When resource demand in a certain area surges beyond its allocated capacity, resources are temporarily borrowed from other areas with lower utilization rates using a resource allocation algorithm. An automatic resource recycling mechanism is set up so that when a line ends its operation or resources are not used for two consecutive hours, they are automatically recycled to the resource pool for other lines to apply for.

[0044] Dynamic weight allocation calculation is performed based on the established time resource pool. A weight calculation model is established, extracting the current available quota, real-time application frequency, and historical usage efficiency data of each route from the resource pool as the three main factors for weight calculation. Taking a certain route as an example, the resource pool shows that its current available quota is 130 units, the real-time application frequency is twice every 15 minutes, and the historical usage efficiency is 85%. Weighted calculation is performed according to a weight ratio of 4:4:2, resulting in a comprehensive weight value of 0.142 for this route. The comprehensive weight value is calculated using the entropy weight method: Wi = (1-Ei) / Σ(1-Ej), where Ei is the information entropy of the i-th route, and Ej is the information entropy of the j-th route. A weight normalization mechanism is established, summing and normalizing the weight values ​​of all routes to ensure that the sum of all weights equals 1.0, achieving complete allocation of resources from the resource pool. A dynamic weight update frequency is set: under normal circumstances, data is reread from the resource pool and the weight allocation is updated every 15 minutes; during peak traffic periods, this is adjusted to once every 5 minutes to ensure real-time responsiveness of weight allocation. Weight allocation constraints are established to ensure that each line receives a minimum weight of 0.01 and a maximum weight of 0.15 from the resource pool, preventing excessive resource concentration on a few lines or leaving some lines without resources. In the event of emergencies such as traffic accidents or severe weather, additional resources are automatically extracted from the emergency resource layer of the resource pool, temporarily increasing the weight of affected lines by 20%. The duration of this increase is dynamically adjusted based on the resolution of the event. Through the coordinated operation of intelligent weight allocation and dynamic update mechanisms, the final time allocation weight for each line is obtained.

[0045] Step S130: Perform multi-line resource competition analysis based on arrival time information to determine the operating load index of each line, generate time allocation requirements based on time allocation weights and operating load indices of each line, and generate a time resource allocation scheme based on time load balancing processing of time allocation requirements.

[0046] Specifically, a multi-route resource competition analysis model is constructed, using arrival time information as the input parameter to quantitatively evaluate the resource competition intensity of different routes at the same spatiotemporal node. Taking a major intersection as an example, analysis reveals that the bus time windows of three routes overlap around 8:05: the east-west Line 1 is expected to arrive at 8:04, the north-south Line 2 at 8:05, and the east-west Line 3 at 8:06, forming a typical resource competition scenario. The resource competition analysis algorithm adopts a multi-dimensional evaluation system, using arrival time interval, passenger density, cumulative delay time, and route level as quantitative indicators of competition intensity, assigning weight coefficients of 30%, 25%, 25%, and 20% respectively for comprehensive calculation. The calculation of the operational load index introduces a dynamic parameter model, with core parameters including real-time passenger load factor, departure interval coefficient, operating period weight, and delay accumulation factor. Taking Route 1 as an example, with a real-time passenger load factor of 85% (standard capacity of 85 / 100 passengers), a departure interval of 8 minutes, and 3 cumulative delays on the day, the operational load index is calculated to be 7.2 using a weighted algorithm (passenger load factor weight 40%, departure interval weight 30%, delay factor weight 30%). A load index grading system is established, dividing the operational load index into four levels: extremely high load (above 8.0), high load (5.0-8.0), medium load (3.0-5.0), and low load (below 3.0). Different levels have different priority weights in resource competition. The resource competition resolution rules clearly define the priority sequence for conflict handling, with extremely high load routes having the highest priority. For routes with the same load level, resource allocation is determined in the order of first-come, first-served, delay compensation, and passenger load priority.

[0047] The time allocation weight is coupled with the operational load index to generate the time allocation requirements for each route. Taking a certain route as an example, the time allocation weight is 0.142, the operational load index is 7.2, and the demand calculation model N=W×L×T is used, where W is the time allocation weight, L is the operational load index, and T is the standardized time base of 10. The calculated time allocation requirement for this route is 10.2 units. The operational load index is calculated as: L = Σ(wi×fi), where wi is the weight of the i-th factor and satisfies Σwi=1, fi is the standardized function equal to (xi-xmin) / (xmax-xmin), xi is the actual value, xmin is the minimum value, and xmax is the maximum value. The main factors include load factor (w1=0.35), departure interval deviation (w2=0.25), delay duration (w3=0.25), and application frequency (w4=0.15). Demand tier management categorizes time-based allocation demands into four levels: urgent (15 units or more), important (10-15 units), routine (5-10 units), and low priority (less than 5 units), ensuring hierarchical management of resource allocation. A conflict identification algorithm scans the demand distribution within the same spatiotemporal range, identifying potential resource conflict points by calculating temporal overlap and spatial intersection coefficients. The demand optimization engine performs global optimization under resource constraints; when the total demand exceeds available resources, intelligent peak shaving and off-peak scheduling are executed according to demand level and line priority. Demand boundary constraints ensure system stability, with a single-line demand upper limit set at 120% of maximum service capacity and a lower limit no less than 80% of basic guarantee capacity. Predictive demand analysis, based on historical data and real-time trends, anticipates demand fluctuations 30 minutes in advance, providing decision support for dynamic resource allocation.

[0048] In some embodiments, the step of generating a time resource allocation scheme by performing time load balancing processing based on the time allocation requirements includes: performing structured processing on the time allocation requirements to generate an allocation requirement matrix; constructing a load relationship map between lines based on the allocation requirement matrix; performing balancing processing on the load relationship map to obtain time allocation coefficients; and generating a time resource allocation scheme using the time allocation coefficients.

[0049] The time allocation requirements are structured to generate an allocation requirement matrix. Taking six routes in a certain area during the 8:00-9:00 AM time slot as an example, the allocation requirements for each route are 10.2, 8.5, 12.3, 6.8, 15.1, and 9.4 units respectively. A 6×12 requirement matrix is ​​constructed, with rows representing routes and columns representing time slots, where each of the 12 columns represents a 5-minute time segment. A rule is established for assigning values ​​to matrix elements: each element represents the allocation requirement for a specific route during a specific time slot; null elements are assigned a value of 0, and outliers exceeding the normal range are corrected using the average of adjacent time slots. Matrix standardization is employed to normalize the requirement values, using a max-min normalization method to map all requirement values ​​to the 0-1 range, ensuring comparability between requirement values ​​of different magnitudes. Matrix sparsity optimization is implemented: when zero elements exceed 70% of the matrix, a compressed storage format is used to reduce memory usage and improve computational efficiency. A matrix symmetry check is set up to assess the balance of demand distribution by calculating the symmetry of the matrix. When the symmetry is less than 0.6, demand redistribution is triggered.

[0050] A load relationship graph between lines is constructed based on the allocation demand matrix. A graph theory algorithm is used, treating lines as graph nodes and load relationships as connecting edges. The weight of an edge is determined by the demand difference and time overlap between the two lines, calculated as W = |D1 - D2| × O, where D1 and D2 are the demand values ​​of the two lines, and O is the time overlap coefficient. For example, in the demand matrix, line 1 has a demand of 10.2 units, and line 2 has a demand of 8.5 units. The demand difference between the two lines is 1.7, and the time overlap is 80% (0.8), resulting in a calculated edge weight of 1.36. A relationship strength grading standard is established, classifying line relationships into four levels: strong association (weight above 1.5), medium association (weight 0.5-1.5), weak association (weight 0.1-0.5), and no association (weight below 0.1). Different levels of relationships correspond to different scheduling strategies and coordination mechanisms. A graph clustering algorithm is used to identify groups of lines with similar load characteristics, forming a clustered distribution of load relationships by calculating the shortest path and connectivity coefficient between nodes. Establish a hierarchical structure for the relationship graph, dividing it into three layers: core layer, buffer layer, and outer layer. The core layer routes undertake the main transportation tasks, while the outer layer routes provide supplementary services.

[0051] The load relationship graph is balanced to obtain time allocation coefficients. A balancing objective function is established, with the dual optimization objectives of minimizing the sum of squared load differences between lines Σ(Li-Lavg)² and maximizing the overall system efficiency ΣLi / Ti, where Li is the line load, Lavg is the average load, and Ti is the line time consumption. A genetic algorithm is used for iterative optimization, with a population size of 50, a crossover probability of 0.8, and a mutation probability of 0.1. The resource allocation ratio of each line is gradually adjusted through multiple generations of evolution. Taking the connection edge with a weight of 1.36 in the graph as an example, after 35 generations of genetic algorithm optimization, the allocation coefficient of line 1 is calculated to be 0.52, and the allocation coefficient of line 2 is 0.48. The sum of the two equals 1.0, achieving balanced resource allocation. Balance constraints are established to ensure that the minimum allocation coefficient of each line is not less than 0.1 and the maximum allocation coefficient does not exceed 0.6, preventing excessive resource skew and unfair allocation. A balance convergence criterion is set: when the improvement of the fitness function for 10 consecutive generations is less than 0.001, the optimal balance state is considered to be reached. Establish a balancing effect evaluation mechanism, calculate the load variance and Gini coefficient before and after balancing treatment, and quantitatively evaluate the degree of improvement in balancing effect.

[0052] A time resource allocation scheme is generated using time allocation coefficients. A time-segmented allocation mechanism is established, dividing the 24 hours of a day into six periods: morning peak (7:00-9:00), morning off-peak (9:00-12:00), afternoon peak (12:00-14:00), afternoon off-peak (14:00-17:00), evening peak (17:00-19:00), and nighttime peak (19:00-7:00). For each period, the priority signal time quota for each line is determined based on the allocation coefficient. Taking the morning peak period as an example, line 1 with an allocation coefficient of 0.52 corresponds to a 52-second priority green light extension time; line 2 with an allocation coefficient of 0.48 corresponds to 48 seconds; line 3 with an allocation coefficient of 0.35 corresponds to 35 seconds; line 4 with an allocation coefficient of 0.42 corresponds to 42 seconds; line 5 with an allocation coefficient of 0.58 corresponds to 58 seconds; and line 6 with an allocation coefficient of 0.33 corresponds to 33 seconds. A hierarchical execution mechanism for the allocation strategy is established. High-coefficient lines have priority in requesting priority signals and resource guarantees; medium-coefficient lines are allocated signals when resources are sufficient; and low-coefficient lines are allocated signals during off-peak hours to avoid peak-hour conflicts. An allocation conflict resolution algorithm is employed. When multiple lines simultaneously request priority signals at the same intersection, the service order is determined according to the allocation coefficient from highest to lowest. Lines with a coefficient difference greater than 0.1 have absolute priority, while lines with a coefficient difference less than 0.1 are allocated signals using a time-alternating mechanism. Execution rules for the allocation scheme are set, with the allocation time for each line quantified in 5-second increments. The duration of a single priority signal cannot exceed 150% of the time corresponding to the allocation coefficient and cannot be less than 50%.

[0053] Step S140: Formulate a quota-based priority execution strategy according to the time resource allocation scheme, apply the priority execution strategy to the signal changes generated by phase adjustment, calculate the traffic entropy value of the signal changes to obtain the current traffic confusion data, and perform phase extension analysis on the traffic confusion data to generate a green light timing scheme.

[0054] Specifically, a quota-based priority execution strategy is formulated based on the time resource allocation scheme. A quota-driven algorithm is used, prioritizing and resolving conflicts by using the time-segmented quotas of each route as constraints for the execution strategy. Taking an intersection during the morning rush hour as an example, route 1 is allocated 52 seconds of priority time, route 2 48 seconds, and route 3 35 seconds. When all three routes simultaneously request priority signals, the service order is determined from highest to lowest allocated time as route 1 → route 2 → route 3. A classification system for priority execution strategies is established, including four basic types: green light extension strategy, red light shortening strategy, phase insertion strategy, and phase skipping strategy. The green light extension strategy is applicable when buses are about to arrive and the current light phase is green. The extension time is determined according to the allocated quota; route 1 can be extended by 15-52 seconds, and route 2 by 12-48 seconds. The red light shortening strategy is applicable when buses are waiting and the current light phase is red; the shortening time does not exceed 80% of the allocated quota. Establish execution constraints for priority execution strategies. The maximum execution time for a single priority strategy shall not exceed 150% of the allocated quota, and the minimum execution time shall not be less than 5 seconds. The interval between two consecutive priority strategy executions shall not be less than 30 seconds. Set up a conflict resolution mechanism for priority execution strategies. When priority requests from multiple lines conflict in time, a weighted queuing algorithm W_queue = T_allocation × U_urgency shall be used, where W_queue is the weighted queuing priority value, T_allocation is the allocation time, and U_urgency is the urgency coefficient.

[0055] Priority execution strategies are applied to signal changes generated by phase adjustment. A phase adjustment algorithm is used to calculate the adjustment range of phase parameters based on the priority execution strategy type. Taking the green light extension strategy as an example, when Line 1 requests a 25-second extension of the green light, the system calculates the remaining green light time to be 8 seconds, resulting in an actual extension of 25 seconds and a total adjusted green light time of 33 seconds. Timing control for phase adjustment is established to ensure smoothness and safety of phase switching. The execution delay of phase adjustment commands is controlled within 200 milliseconds, and a 3-second yellow light transition time is set during phase switching. Limitations are set for the range of phase adjustment: a single green light extension cannot exceed 60 seconds, a red light shortening cannot exceed 40 seconds, and the total cycle time variation is controlled within ±30%. A real-time execution mechanism for phase adjustment is established. Upon receiving the adjustment command, the signal controller immediately executes the phase change and records the changes in phase parameters before and after the adjustment. Phase status monitoring technology is used to track the phase change process of the traffic lights in real time, including key parameters such as green light duration change ΔG, red light duration change ΔR, and cycle time change ΔC.

[0056] In some embodiments, the step of calculating traffic entropy values ​​to obtain current traffic congestion data based on the signal changes includes: sampling the vehicle operating state caused by the signal changes to obtain vehicle state sampling data; generating speed variance parameters and queue length change parameters based on the vehicle state sampling data; performing entropy value processing on the speed variance parameters and queue length change parameters to obtain entropy value processing results; and generating current traffic congestion data based on the entropy value processing results.

[0057] A vehicle status sampling system was established, deploying video detectors and geomagnetic sensors at intersection entrances and exits to collect real-time vehicle status data within the influence range of signal changes. A vehicle trajectory tracking algorithm was employed to identify and record the position, speed, acceleration, and parking status changes of each vehicle. Taking a 13-second extension of the east-west green light as an example, the sampling system detected 15 vehicles within a 200-meter upstream radius in that direction. Of these, 12 vehicles transitioned from a stopped state to a moving state, with an average starting acceleration of 1.2 m / s², while 3 vehicles maintained a constant speed of 35 km / h. A classification and recording mechanism for the sampling data was established, categorizing vehicle states into four basic states: acceleration, constant speed, deceleration, and stopping. The number of vehicles, duration, and speed change amplitude for each state were recorded. The sampling frequency was set to 10 times per second to ensure the capture of rapid changes in vehicle status. A dynamic adjustment mechanism for the sampling range was established, determining the sampling range based on the degree of influence of the signal change. The influence range of a green light extension was 300 meters upstream and 100 meters downstream, while the influence range of a red light shortening was 150 meters upstream and 200 meters downstream.

[0058] Speed ​​variance and queue length variation parameters were generated based on vehicle state sampling data. Statistical analysis was used to calculate the variance of state sampling data from 15 vehicles, with stationary vehicles recorded as having a speed of 0 km / h and moving vehicles ranging from 28 to 37 km / h. The variance analysis yielded a speed variance of 285.6 and a standard deviation of 16.9 km / h. A grading standard for speed variance was established, classifying it into four levels: low variance (0-100), medium variance (100-300), high variance (300-500), and extremely high variance (above 500). The speed variance at this intersection falls into the medium variance category. A queue length detection algorithm was used to statistically analyze the changes in the number of vehicles queuing at each approach lane before and after a signal change. Taking the east-west approach lane as an example, before the signal change, there were 12 vehicles queuing; after the green light was extended by 13 seconds, the number of vehicles queuing decreased to 3, with a queue length change of ΔQ = -9 vehicles, and a queue dissipation rate of 75%. Quantitative indicators for queue length changes are established, including queue growth rate R_growth = ΔQ_positive / Q_initial (where ΔQ_positive is the increase in queue length and Q_initial is the initial queue length), queue dissipation rate R_dissipation = ΔQ_negative / Q_initial (where ΔQ_negative is the absolute value of the decrease in queue length and Q_initial is the initial queue length), and queue fluctuation rate R_fluctuation = |ΔQ| / Q_max (where |ΔQ| is the absolute value of the change in queue length and Q_max is the maximum queue length during the observation period). The calculation period for queue length changes is set to the signal phase period, and the queue length change parameters are recalculated after each period.

[0059] Entropy processing is performed on the speed variance parameter and queue length variation parameter to obtain the entropy value results. Entropy processing is based on the application of the second law of thermodynamics to traffic systems, transforming the randomness and uncertainty of vehicle motion into a measurable entropy index. The speed entropy is calculated using the differential entropy method for continuous random variables. Let the probability density function of the speed distribution be f(v), then the speed entropy H_v = -∫f(v)log2f(v)dv, with the integration interval covering the observed speed range. In actual calculations, the continuous distribution is discretized, divided into k speed intervals [v_{i-1}, v_i], and the proportion of vehicles p_i in each interval is statistically analyzed. The discrete entropy formula H_v = -Σ(p_i×log2p_i) is applied, where i ranges from 1 to k. A larger speed variance indicates a more dispersed probability distribution, resulting in a higher calculated entropy value, indicating increased variability and unpredictability of vehicle speeds. The entropy value of queue length change is handled using a state transition entropy model. The queue state space is defined as S = {rapid growth, slow growth, stable, slow dissipation, rapid dissipation}, and a state transition probability matrix P = [p_{ij}] is constructed, where p_{ij} represents the probability of transitioning from state i to state j. The queue entropy is calculated as H_q = -ΣΣ(π_i × p_{ij} × log2p_{ij}), where π_i is the steady-state probability of state i, solved using the stationary distribution of a Markov chain. The queue length change parameter determines the state transition probability; the more drastic the change, the greater the uncertainty of the state transition and the higher the entropy value. The comprehensive entropy value is calculated using a weighted entropy fusion method, defined as total entropy H_total = α × H_v + β × H_q + γ × H_v × H_q, where α and β are linear weight coefficients, and γ is the interaction term coefficient, reflecting the coupling effect between speed and queueing. The weight coefficients are determined using the entropy weighting method, adaptively adjusted according to the degree of variation of each component's entropy, with components with greater variation receiving higher weights. Normalization is performed using the maximum entropy normalization method, H_norm = H_total / H_max, where H_max = log₂N is the maximum possible entropy value of the system, and N is the number of system states. The temporal evolution characteristics of the entropy value are captured using the sliding window method, calculating the average entropy value and the rate of change of entropy value within the window to reflect the dynamic trend of traffic disorder.

[0060] Traffic disorder data is generated based on entropy processing results. The disorder conversion formula is D = a × H_norm + b, where D is the disorder index, H_norm is the normalized traffic state entropy value, a is the conversion coefficient, and b is the baseline adjustment value. The conversion process considers the nonlinear characteristics of entropy and the practical significance of disorder. The traffic entropy calculation formula is: H = α × Hs + β × Hq + γ × Hd, where speed entropy... PV represents the proportion of vehicles in each speed range. Queue entropy Hq and density entropy Hd are calculated similarly, with weighting coefficients α=0.4, β=0.3, and γ=0.3. After normalization, H_norm = H / Hmax. A classification standard for traffic disorder levels is established, dividing traffic disorder into four levels: low disorder (0-80), medium disorder (80-120), high disorder (120-160), and extremely high disorder (above 160). Different levels correspond to different traffic management strategies. A multi-dimensional disorder assessment method is adopted, combining multiple factors such as speed variance, queue length variation, and traffic density for comprehensive evaluation. Each factor is assigned a corresponding weighting coefficient according to its impact on traffic disorder. A time-weighted mechanism for disorder is established, with higher weights for recent disorder data and gradually decreasing weights for historical disorder data. A dynamic update mechanism for disorder is set up, recalculating the disorder value after each signal cycle to maintain the real-time nature of the disorder data.

[0061] Phase duration analysis technology is employed, using current traffic disorder data as the foundation for timing optimization. A timing demand assessment method is used to determine the intensity of timing adjustments based on the disorder level, with different adjustment coefficients corresponding to different disorder levels. The timing demand time for each direction is calculated using the formula T=k×D×w, where T is the timing adjustment time, k is the adjustment coefficient, D is the disorder index, and w is the direction weighting coefficient. Different direction weighting coefficients are used for east-west and north-south directions. An optimization objective for traffic light timing is established, aiming to minimize total disorder and maximize traffic efficiency. A timing allocation method is used to distribute the total timing demand to each phase according to traffic flow proportions. Constraints are set for the green light timing scheme: a single green light duration cannot exceed 45 seconds and cannot be less than 5 seconds, and the total cycle duration cannot exceed 25% of the original cycle. A tiered execution strategy for the green light timing scheme is established, employing different timing intensities based on the disorder level: conservative timing for low disorder and aggressive timing for high disorder. Taking a major intersection as an example, when the chaos index reaches a moderate level, the generated green light timing scheme adjusts the green light duration for east-west traffic to 70 seconds and for north-south traffic to 50 seconds. The timing execution order is determined according to traffic flow ratio, prioritizing east-west traffic when east-west flow is higher, thus forming a dynamic timing adjustment strategy based on time periods and directions. The green light timing scheme is generated through chaos analysis and timing optimization.

[0062] Step S150: Perform traffic disturbance propagation analysis on the green light timing scheme to generate road network spillover effect; perform inter-intersection game analysis on the road network spillover effect to generate game equilibrium point; perform entropy propagation analysis based on the road network spillover effect to obtain disorder diffusion parameters; and use the disorder diffusion parameters and game equilibrium point to construct an entropy reduction control mechanism.

[0063] Specifically, traffic disturbance propagation analysis is conducted based on the generated green light timing scheme to identify the propagation mechanism and impact range of signal adjustments in the road network. Traffic disturbances originate from the continuous nature of traffic flow. When a green light delay at an intersection releases more traffic, this traffic flow moves downstream as a whole, altering the traffic load at downstream intersections. A disturbance propagation tracking mechanism is established, determining the transmission path of the disturbance through the topological relationships between intersections and the traffic flow turning ratio. The calculation of disturbance intensity considers factors such as the initial released traffic volume, propagation distance, and road capacity, resulting in a decreasing intensity distribution. The propagation time is determined by the distance between intersections and real-time vehicle speed; in urban road environments, the propagation speed typically fluctuates within the range of 20-40 km / h. The time window characteristic of the disturbance is that it reaches its peak within the initial 2-3 signal cycles, then gradually decays to a baseline level. The disturbance propagation characteristics of roads of different grades differ significantly. On arterial roads, due to high traffic volume and strong continuity, traffic disturbances can propagate to 3-5 intersections, while on secondary roads, due to significant turning and diversion, the impact is limited to 1-2 intersections. Taking a specific green light timing scheme as an example, extending the east-west direction by 25 seconds resulted in the release of an additional 30 vehicles in that direction. These vehicles arrived at the downstream intersection 500 meters away 75 seconds later, increasing the east-west traffic flow at that intersection by 35%, and raising the saturation level from 0.75 to 0.88. Through systematic propagation tracing and disturbance quantification analysis, the road network spillover effect, encompassing spatial scope, intensity distribution, and temporal evolution, was obtained.

[0064] In some embodiments, the step of performing inter-intersection game analysis on the road network spillover effect to generate a game equilibrium point includes: analyzing the road network spillover effect to obtain the interest relationships between each intersection; constructing an inter-intersection game relationship matrix based on the interest relationships; performing payoff analysis on the game relationship matrix to obtain payoff analysis results; and performing equilibrium solution based on the payoff analysis results to generate a game equilibrium point.

[0065] By analyzing the impact range, intensity distribution, and temporal evolution characteristics of the road network spillover effect, we can identify the interest relationships between intersections. The impact range determines which intersections have direct or indirect interest relationships. An impact range of 3-5 intersections on a main road indicates that these intersections form a close-knit community of interests, and their decisions will have a chain reaction. The intensity distribution reflects the closeness of the interest relationship. Calculated at a 15% decrease rate, directly adjacent intersections bear the strongest impact, with an impact strength of 0.85. The impact strength of second-level intersections decreases to 0.72, and that of third-level intersections further decreases to 0.61. With each additional intersection, the interest relationship gradually weakens. The temporal evolution characteristics reveal the dynamic process of the interest impact. The impact begins to appear in the first cycle after the initial adjustment, reaches its peak in the second-to-third cycle, at which point the interest conflicts among intersections are most intense, requiring focused coordination. Subsequently, the impact gradually diminishes. The essence of interest relationships is the redistribution of traffic resources in time and space. Upstream intersections gain additional travel time resources by extending green lights, correspondingly transferring queuing and delays to downstream intersections, forming a zero-sum game of interests. The identification of the type of relationship is based on the positiveness and directionality of the impact. When the main traffic flow of upstream and downstream intersections is consistent, a positive collaborative relationship is formed, and both parties can achieve a win-win situation through coordination. When there is a directional conflict in traffic demand, a negative competitive relationship is formed, and the improvement of one party will inevitably come at the expense of the deterioration of the other party. When there is little traffic flow exchange between intersections, a neutral relationship is formed, and the mutual impact can be ignored.

[0066] Based on the obtained interest relationships at each intersection, a game theory relationship matrix is ​​constructed, transforming the qualitative competitive, cooperative, and neutral relationships into quantitative influence coefficients. The matrix uses an N×N square matrix to represent the mutual influence relationships between N intersections. Competitive relationships correspond to negative elements indicating negative impact, cooperative relationships correspond to positive elements indicating positive promotion, and neutral relationships correspond to elements close to zero indicating weak impact. Based on the identified association strength levels in the interest relationships, corresponding influence coefficient ranges are set for intersections at different distances. Directly adjacent intersections have the most direct impact, with a coefficient range of [-0.85, 0.85]. The impact of second-level intersections is attenuated after one transmission, with a coefficient range of [-0.72, 0.72]. The impact of third-level intersections is further weakened, with a coefficient range of [-0.61, 0.61]. The specific value of matrix element G[i,j] is determined through comprehensive calculation. The calculation formula is G[i,j]=Sign(Type)×Strength×Distance_Factor, where G[i,j] is the influence coefficient of intersection i on intersection j, Sign(Type) is the relation type sign function, which takes -1 to make the element negative for competitive relations, +1 to make the element positive for cooperative relations, and 0 to make the element close to zero for neutral relations; Strength is the basic association strength; and Distance_Factor is the distance decay factor. The diagonal element G[i,i] is uniformly set to 1, indicating that each intersection has complete control and decision-making power over itself.

[0067] Payoff analysis is performed on the game relationship matrix to obtain the payoff analysis results. The total payoff function of intersection i is defined as R[i]=B[i]+Σ(G[j,i]×S[j]), where R[i] is the total payoff of intersection i, B[i] is the basic payoff reflecting the direct traffic improvement brought about by intersection i implementing its own strategy, Σ(G[j,i]×S[j]) is the interaction payoff calculated by summing the influence of all other intersections, G[j,i] is the influence coefficient of intersection j on intersection i, and S[j] is the strategy strength of intersection j. When the element G[j,i] in the game relationship matrix is ​​negative, it indicates that the strategy S[j] of intersection j has a negative impact on intersection i, and the larger the negative value, the more serious the impact; when G[j,i] is positive, it indicates a positive promoting effect, which helps to improve the traffic conditions of intersection i. Payoff calculation is performed by traversing the i-th column of the game relationship matrix, extracting the influence coefficients of all intersections on intersection i, multiplying them by the corresponding strategy strengths, and then summing the results to obtain the overall impact of the intersection under a specific strategy combination. The sparsity of the matrix plays a role in the calculation, with only non-zero elements participating in the operation and zero elements being skipped, thus improving computational efficiency. Different strategy strengths S[j] reflect the priority measures taken at intersection j, and multiplying them by the corresponding matrix element G[j,i] reflects the linear superposition characteristic of the impact.

[0068] The equilibrium point of the game is generated by solving the equilibrium problem based on the payoff analysis results. The optimal response function for each intersection is extracted from the payoff analysis results; that is, given the strategies of other intersections, the strategy that maximizes the payoff R[i] of the current intersection is found from the payoff analysis results. In the initialization phase, each intersection queries the individual action portion of the payoff analysis results and selects the strategy that maximizes its own basic payoff B[i] as the starting point for iteration, forming the initial strategy combination. In each iteration, intersection i keeps the current strategies of other intersections unchanged and searches for all possible values ​​of S[i] in the payoff analysis results, finding the strategy S[i]* that maximizes the total payoff R[i] as the new choice. The iteration process continues, with all intersections updating their strategies in turn, until the strategy adjustment magnitude of all intersections in two consecutive iterations is less than a preset convergence threshold, usually set to a 5% change rate. At this point, a candidate equilibrium state is considered to have been reached. The equilibrium point is determined using the Nash equilibrium iterative algorithm: the payoff function R[i] = B[i] - C[i] × S[i] + Σ(G[j,i] × S[j]), where B[i] is the base payoff, C[i] is the cost coefficient, S[i] is the strategy variable, and G[j,i] is the influence coefficient of intersection j on intersection i. Candidate equilibrium points are rigorously verified. Payoff analysis confirms that under this strategy combination, any unilateral adjustment of the strategy at any intersection by ±5 seconds will lead to a decrease in payoff, satisfying the Nash equilibrium definition. When multiple strategy combinations satisfying the equilibrium condition are found during the search process, further comparison is needed. The total payoff of each equilibrium point, R_total = ΣR[i], is calculated, where R_total is the total payoff of the road network. The strategy combination that maximizes the overall payoff of the road network is selected as the final solution.

[0069] In some embodiments, obtaining the disorder diffusion parameters based on the entropy propagation analysis of the road network ripple effect includes: tracking the propagation path of the road network ripple effect to obtain propagation path data; generating a traffic disorder diffusion speed based on the propagation path data; analyzing the attenuation law of the traffic disorder diffusion speed to obtain attenuation law parameters; and generating disorder diffusion parameters according to the traffic disorder diffusion speed and the attenuation law parameters.

[0070] Propagation path data was obtained by tracing the propagation path of the road network spillover effect. The recorded impact range of 3-5 intersections in the road network spillover effect provided the spatial boundary for path tracing, and the 15% decrease rate of the intensity distribution indicated the main direction of propagation. Starting from the initial disturbance intersection, the transmission process of disorder was traced along the actual movement direction of traffic flow. When priority measures were implemented at intersection A, generating initial disorder, this disorder state propagated downstream with the affected traffic flow. Propagation path identification was based on the actual driving trajectories of vehicles, and the main propagation direction and secondary branches were determined by analyzing the turning ratio of traffic flow. The main propagation path extended along the arterial road direction, covering the sequence of intersections where straight-through traffic flow was dominant, and the location coordinates and intersection number of each node were recorded. Secondary paths branched at major intersections, spreading to adjacent roads following the turning traffic flow, and the branch point locations and diversion ratios were marked. The temporal characteristics of the path were obtained by recording the arrival time of disorder at each node. The first downstream intersection B received the propagated disorder after 75 seconds, the second intersection C after 152 seconds, and the third intersection D after 234 seconds, forming a complete time series. The spatial distance between nodes is determined by the road network topology: AB is 500 meters apart, BC is 650 meters apart, and CD is 550 meters apart.

[0071] The diffusion speed of confusion is generated based on the distance between nodes and arrival time in the propagation path data. The diffusion speed is calculated using the formula V=D / T, where V is the diffusion speed, D is the spatial distance between adjacent nodes on the path, and T is the propagation time of confusion between nodes. The diffusion speed of the first segment AB is calculated as V1=500 / 75=6.67 m / s, equivalent to 24 km / h; the diffusion speed of the second segment BC is V2=650 / 77=8.44 m / s, approximately 30.4 km / h; and the diffusion speed of the third segment CD is V3=550 / 82=6.71 m / s, approximately 24.1 km / h. The changes in diffusion speed reflect the differences in traffic conditions across different road segments. The higher speed in segment BC indicates relatively smooth traffic flow, while the similar speeds in segments AB and CD indicate similar traffic conditions. The diffusion speed of branch paths is calculated separately. Due to traffic diversion and turning delays, the diffusion speed of branch paths is usually lower than that of the main path. Time-varying characteristic analysis compares the diffusion speeds at different times to form a dynamic speed distribution. The acceleration and deceleration characteristics during the diffusion process are calculated using the speed difference between adjacent road segments, with the initial acceleration being approximately 0.5 m / s².

[0072] The decay law of traffic disorder diffusion speed was analyzed to obtain decay law parameters. The differentiated distribution of diffusion speed from 24 km / h to 30.4 km / h revealed the non-uniformity of disorder propagation, requiring in-depth analysis of the decay law under different diffusion speed conditions. A decay analysis framework based on diffusion speed was established, dividing the road network into high-speed diffusion zone, medium-speed diffusion zone, and low-speed diffusion zone, and the decay characteristics of each zone were studied separately. The decay of disorder in the high-speed diffusion zone showed a rapid decreasing trend, because the traffic flow is more likely to disperse during rapid movement, and the concentration of disorder decreases rapidly; the medium-speed diffusion zone showed a stable exponential decay characteristic, which is consistent with the classical diffusion decay theory; the decay process in the low-speed diffusion zone was relatively slow, and the spatial persistence of disorder was stronger. By comparing and analyzing the spatial distribution data of disorder under different diffusion speeds, a positive correlation was found between diffusion speed and decay intensity. A speed-dependent decay model λ(v)=λ0+α·(v-v_avg) was established, where λ0 is the baseline decay coefficient, α is the speed sensitivity coefficient, and v_avg is the average diffusion speed. Road environment factors moderate the attenuation process. The open space of main roads promotes the rapid dissipation of disorder, while the narrow environment of side roads slows down the attenuation process. This difference is reflected by the road width correction coefficient. After systematic analysis of attenuation law and calibration of model parameters, the attenuation law parameters Λ={λ0,α,γ_w,γ_t} were finally obtained, where λ0 is the baseline attenuation coefficient, α is the speed sensitivity coefficient, γ_w is the road width correction coefficient, and γ_t is the time period adjustment coefficient.

[0073] Based on the traffic disorder diffusion speed V and the attenuation law parameter Λ, a complete disorder diffusion parameter Ψ is generated using a parameter synthesis model: Ψ=G(V,Λ)={V_char,λ_eff,K(d,t),d_c}, where V_char is the set of speed characteristic parameters, λ_eff is the effective attenuation coefficient, K(d,t) is the diffusion kernel function, and d_c is the critical propagation distance. The disorder diffusion parameter is a comprehensive index system describing the propagation behavior of traffic disorder in the road network, requiring the organic integration of speed and attenuation characteristics. The diffusion speed data V provides the dynamic characteristics of disorder propagation, including the propagation speed, acceleration changes, and speed distribution range, among other time-dimensional information. The attenuation law parameter Λ characterizes the spatial evolution of disorder intensity, reflecting the quantitative relationship that disorder weakens with increasing distance. The parameter generation process employs feature extraction and parameter synthesis methods, extracting steady-state, transient, and statistical features from the original speed time series to form the speed characteristic parameter set V_char={v_min,v_max,v_mean,σ_v}. The diffusion kernel function K(d,t)=exp(-λ_eff·d)·Θ(v·td) comprehensively describes the spatiotemporal propagation characteristics of disorder, where the effective attenuation coefficient λ_eff=λ0+α·(v-v_avg) is determined by both the attenuation law parameter and the real-time velocity. Considering practical application requirements, a directional parameter is introduced to distinguish the diffusion differences between the main propagation path and the secondary propagation path; the main path typically has a faster propagation speed and a stronger influence range. A dynamic adaptation mechanism allows the diffusion parameters to be adjusted according to real-time traffic conditions, employing appropriate parameter configurations for different traffic scenarios.

[0074] An entropy reduction control mechanism is constructed using the disorder diffusion parameter Ψ and the game equilibrium point. The design goal of the control mechanism is to continuously reduce the overall disorder of the road network while maintaining stable operation of each intersection near the game equilibrium point. The disorder diffusion parameter Ψ provides a time window for prediction and intervention. The time it takes for disorder to reach downstream intersections can be calculated using the velocity characteristic parameter V_char in Ψ, and the intensity of the impact can be assessed using the effective attenuation coefficient λ_eff. The game equilibrium point determines the target strategy for each intersection, providing a stable reference benchmark for control. The control strategy adopts a hierarchical architecture. The upper layer formulates the target state of each intersection based on the game equilibrium point, and the lower layer implements predictive control based on the diffusion parameter Ψ. When high disorder is detected at an intersection, the entropy reduction control process is immediately initiated: first, the strategy value of the intersection at the game equilibrium point is queried to ensure that the current control does not deviate from the equilibrium; then, the time it takes for disorder to propagate to each downstream intersection is calculated using the diffusion velocity in Ψ, and pre-adjustments are implemented before the disorder arrives. The timing of pre-adjustment is determined by the diffusion kernel function K(d,t) in the diffusion parameter Ψ. If the diffusion speed is 24 km / h, preparations need to begin 75 seconds in advance for the downstream intersection 500 meters away. The control intensity is dynamically adjusted according to the attenuation law in Ψ. Closer intersections require stronger control measures, while farther intersections can use weaker interventions due to natural attenuation. The calculation of the disorder diffusion parameters is based on the diffusion equation: diffusion coefficient D = 0.1×Va², propagation speed v = 0.8×Vt, attenuation coefficient λ = 1.2 / d², where Va is the average vehicle speed, Vt is the actual traffic speed, and d is the propagation distance.

[0075] Step S160: Combine the entropy reduction control mechanism and the time resource allocation scheme to generate full-path impact assessment data, formulate a chain adjustment strategy based on the full-path impact assessment data, and generate a resource balance scheme between intersections based on the chain adjustment strategy.

[0076] Specifically, a full-path impact assessment is generated by combining an entropy reduction control mechanism and a time resource allocation scheme. The entropy reduction control mechanism provides dynamic prediction and control strategies for disorder, enabling the prediction of traffic state changes at intersections at different times. The time resource allocation scheme clarifies the priority signal time quotas for each route at different times, such as the specific allocation of 52 seconds for Route 1 and 48 seconds for Route 2 during the morning peak hours. The full-path impact assessment starts from the starting station of the bus route and analyzes the impact at each intersection along the operating route. When Route 1 requests 52 seconds of priority time at intersection A, the entropy reduction control mechanism predicts that this measure will generate an increase in disorder at intersection B after 75 seconds, and simultaneously calculates the remaining time quota for the route at intersection B. The spatial dimension of the impact assessment covers all intersections traversed by the route, and the temporal dimension extends to the entire operating period. Assessment indicators include the change in disorder at each intersection, the saving of transit time for buses, the increase in delays for private vehicles, and the fluctuation range of intersection saturation. Cumulative effect analysis reveals the cumulative impact of multiple priority applications. When the same route applies for priority at consecutive intersections, the pressure on downstream intersections gradually increases.

[0077] Based on the changes in disorder at each intersection, resource consumption, and cumulative effects recorded in the full-path impact assessment data, targeted adjustment strategies were formulated. Assessment data showed that the disorder level at intersection B reached 156, exceeding the permissible threshold of 140, and the resource utilization rate at intersection C reached 85%, approaching saturation. These issues require resolution through adjustment strategies. The core idea of ​​cascading adjustments is to transform local adjustments at a single point into coordinated responses at multiple points, preventing the transfer and accumulation of problems within the road network. The formulation of adjustment strategies follows the principles of "source control, process optimization, and end-point compensation." Priority measures are controlled at the source intersections where disorder arises; signal timing is optimized during propagation to mitigate the impact; and residual effects are eliminated at the end intersections through compensation measures. Strategy types include time adjustment strategies, spatial diversion strategies, and intensity control strategies. The timing adjustment strategy utilizes priority access during off-peak hours, delaying priority requests at intersection B by 10 minutes to avoid peak upstream impact. The spatial diversion strategy opens auxiliary lanes at intersection C, guiding 30% of traffic flow to use alternative routes. The intensity control strategy limits priority time at intersection A from 25 seconds to 18 seconds, reducing initial disturbance intensity. This coordinated adjustment mechanism ensures coordinated responses from upstream and downstream intersections, resulting in a coordinated adjustment scheme where priority time is reduced by 7 seconds at intersection A, delayed by 10 minutes at intersection B, and traffic is diverted by 30% at intersection C.

[0078] In some embodiments, generating an inter-intersection resource balancing scheme based on the interlocking adjustment strategy includes: decomposing the interlocking adjustment strategy into intersections to obtain the adjustment requirements of each intersection; processing the adjustment requirements of each intersection to obtain compensation data; performing inter-intersection collaborative configuration on the compensation data to obtain collaborative configuration results; and generating an inter-intersection resource balancing scheme based on the collaborative configuration results.

[0079] The overall adjustment requirements of the chain-link adjustment strategy are broken down into specific intersections, clarifying the adjustment tasks to be performed at each intersection. The chain-link adjustment strategy defines the overall plan as reducing the priority time by 7 seconds at intersection A, delaying it by 10 minutes at intersection B, and diverting 30% of traffic at intersection C. These requirements need to be translated into specific execution tasks for each intersection. The adjustment requirements for intersection A include: shortening the east-west green light extension time from 25 seconds to 18 seconds between 8:15 and 8:30, with the 7-second reduction needing to be compensated for through other measures; and monitoring changes in north-south traffic flow to prevent new congestion points. The adjustment requirements for intersection B are: delaying the priority application originally scheduled for 8:20 to 8:30, maintaining standard signal timing during the waiting period; and preparing for potential traffic overlap after 8:30. The adjustment requirements for intersection C are: opening the eastern auxiliary lane, setting up guiding signs to direct 30% of the straight-through traffic to the auxiliary lane; and adjusting lane traffic lights to ensure the safe passage of diverted vehicles. The quantitative parameters for regulating demand include time parameters (start and end times, duration), spatial parameters (lanes involved, direction of impact), and intensity parameters (regulation magnitude, target flow rate). Constraints limit the regulation range of each intersection to ensure that regulation does not cause new problems. After the intersections are decomposed, the regulation demand for each intersection is obtained.

[0080] Compensation data is obtained by processing the compensation amount based on the adjustment needs of each intersection. The impact of reducing the priority time by 7 seconds at intersection A on public transportation needs to be quantitatively assessed. Based on the flow-time relationship, the loss of 7 seconds of green light is equivalent to the capacity of 12 vehicles, and this loss needs to be compensated for through other measures. The compensation amount is calculated using an equivalent conversion method. Different compensation measures have different effects. Optimizing the signal cycle by 1 second is approximately equivalent to extending the green light by 0.5 seconds. Opening auxiliary lanes can compensate for a capacity of 15 vehicles per hour. The compensation amount for delaying traffic by 10 minutes at intersection B is reflected in the time dimension. Additional passage guarantees need to be provided for affected public transportation vehicles during the delay period. It is calculated that an additional 15 seconds of priority time is needed as compensation. The compensation amount for diverting 30% of the traffic flow at intersection C is determined through capacity analysis. The auxiliary lanes need to provide at least 300 vehicles per hour of additional capacity to accommodate the diverted vehicles. The time distribution of compensation is determined based on the impact characteristics. Immediate compensation is implemented simultaneously with adjustment, while delayed compensation is implemented after the peak. The compensation intensity is set in stages according to the degree of adjustment: 20% compensation is required for mild adjustment, 50% for moderate adjustment, and more than 80% for severe adjustment.

[0081] The compensation data is used to perform inter-intersection collaborative configuration to obtain the results. The compensation data shows that intersection A requires 12 vehicles' worth of capacity compensation, intersection B requires 15 seconds of time compensation, and intersection C requires 300 vehicles / hour of capacity compensation. These needs need to be coordinated across the road network. Collaborative configuration first identifies the source of compensation resources. Intersection D has surplus capacity during the same period and can absorb part of intersection A's compensation needs; intersection E has room for signal cycle optimization and can provide time support for intersection B. Spatial coordination is achieved through traffic transfer, transferring some of the reduced capacity from intersection A to intersection D. The two intersections are 800 meters apart, and traffic flow can be naturally diverted through guidance. Temporal coordination arranges the execution order of different compensation measures. Capacity compensation at intersection D begins at 8:15, synchronized with the adjustment at intersection A; cycle optimization at intersection E starts at 8:25, preparing for the postponement request from intersection B. Compensation transfer allows surplus resources to flow between intersections; intersection F transfers its unused 5 seconds of priority time to intersection B, enhancing the compensation effect. The coordination priority is determined based on the intersection association strength. Intersections with strong association (AD) receive priority coordination, followed by those with medium association (BE). Conflict resolution ensures that compensation measures do not create new conflicts. When increased traffic at intersection D may affect intersection G, the compensation strength is appropriately reduced.

[0082] Based on the compensation measures combination and implementation sequence determined in the collaborative configuration results, an executable resource balancing scheme between intersections is generated. The scheme clearly stipulates that: Intersection A will implement intensity control from 8:15 to 8:30, reducing the east-west green light time by 7 seconds; simultaneously, Intersection D will add 5 seconds of east-west green light time during the same period to accommodate diverted traffic; Intersection B will prioritize requesting a delay from 8:20 to 8:30, during which Intersection E will save 10 seconds through signal cycle optimization as a time reserve for Intersection B; Intersection C will open its auxiliary lane at 8:15, coordinating with the guidance measures at Intersection F to jointly complete the 30% traffic diversion task. The execution details of the compensation scheme include specific signal parameter adjustment values, lane opening times, traffic guidance measures, and other operational requirements. The scheme's hierarchical structure is divided into core compensation and auxiliary compensation. Core compensation directly corresponds to the adjustment needs, while auxiliary compensation provides additional support. The timing arrangement ensures that the compensation effect is synchronized with the adjustment impact, avoiding imbalances caused by time differences. Monitoring indicators are set with thresholds for traffic flow, delay, and queue length at each intersection to track the execution effect in real time.

[0083] Step S170: Based on the resource balancing scheme between intersections, a hierarchical coordination mechanism is constructed to generate a comprehensive control scheme. The comprehensive control scheme is verified by time load clearing to obtain the load balance status. Based on the load balance status, a comprehensive optimization strategy is generated to complete the bus priority traffic scheduling based on traffic entropy.

[0084] Specifically, a hierarchical coordination mechanism is constructed based on the resource balancing scheme between intersections to generate a comprehensive control scheme. The collaborative compensation scheme specifies measures such as reducing the time at intersection A by 7 seconds in coordination with increasing it by 5 seconds at intersection D, delaying traffic at intersection B by 10 minutes in coordination with cycle optimization at intersection E, and diverting traffic at intersection C by 30% in coordination with guidance at intersection F. These dispersed measures need to be coordinated through hierarchical layers to form an overall control system. The hierarchical architecture sets up three control levels: the regional coordination layer is responsible for the coordinated control of 3-5 adjacent intersections to ensure local coordination of compensation measures; the trunk line coordination layer manages all intersections along the main roads to maintain traffic flow continuity; and the network optimization layer balances the resource allocation of each region and trunk line from a global perspective. The regional coordination layer classifies intersections A, D, and G as a control unit. When intersection A implements a time reduction measure, this level coordinates intersection D to synchronously increase its time and monitors the response of intersection G. The trunk line coordination layer ensures that intersections A, B, and C on the east-west main roads maintain a reasonable signal phase difference, so that even adjustments at individual intersections do not disrupt green wave coordination. When the network optimization layer detects a congestion trend in the northern region due to compensation measures, it allocates surplus resources from the southern region to provide support. Control commands are issued following a hierarchical transmission principle: the network layer sets overall goals, the trunk layer breaks them down into road segment tasks, and the regional layer translates them into intersection commands. Real-time information is uploaded via a reverse path: intersection execution status is summarized to the regional layer, regional status is reported to the trunk layer, and trunk line operation is fed back to the network layer.

[0085] In some embodiments, the step of performing time-based load clearing verification on the integrated control scheme to obtain the load balance state includes: performing a comprehensive impact assessment on the integrated control scheme to obtain the comprehensive benefits of the transportation system; constructing a load optimization mechanism based on the comprehensive benefits of the transportation system; performing load clearing processing using the load optimization mechanism to obtain clearing processing results; and generating a load balance state based on the clearing processing results.

[0086] A comprehensive impact assessment of the integrated control scheme is conducted to obtain the overall benefits of the transportation system. The integrated control scheme encompasses all control measures under hierarchical coordination, requiring an evaluation of the overall effectiveness of these measures rather than individual benefits. The assessment dimensions cover four aspects: improvement in public transport services, impact of private vehicles, road network operational efficiency, and resource utilization. Improvement in public transport services is measured by calculating the increase in punctuality, reduction in travel time, and decrease in passenger waiting time for each route. For example, the punctuality rate of Route 1 increased from 79.8% to 85.2%, and the average travel time was reduced by 3.5 minutes. The impact of private vehicles is assessed by evaluating the additional delays experienced by private cars, taxis, and other modes of transportation. The average delay for private vehicles increased by 18 seconds per vehicle in east-west directions and by 12 seconds per vehicle in north-south directions. Road network operational efficiency is reflected by changes in saturation at key intersections, changes in average vehicle speed, and changes in congestion mileage. The average saturation at major intersections decreased from 0.82 to 0.78, indicating an overall improvement in efficiency. Resource utilization is assessed by evaluating the consumption and remaining amount of time credits for each route. Route 1 used 78% of its quota, and Route 2 used 65%. Taking into account both positive and negative impacts, a weighted summation method was used to calculate the overall benefit value, with a weight of 0.4 for public transportation improvement, 0.3 for private vehicles, 0.2 for road network efficiency, and 0.1 for resource efficiency.

[0087] A dynamic load optimization mechanism is constructed based on the comprehensive benefits of the transportation system obtained from the assessment. The comprehensive benefit analysis reveals an imbalance in load across different intersections and routes, with some intersections overloaded while others are underutilized, necessitating an optimization mechanism for rebalancing. The objective function of load optimization is set to maximize overall benefits while minimizing load differences, solved using a bi-objective optimization method. Optimization variables include adjustable elements such as time allocation at each intersection, resource quotas for each route, and the strength parameters of compensation measures. Constraints ensure that the optimization process does not violate basic traffic operation principles, such as signal cycle length limits, minimum green light time requirements, and minimum service levels for routes. The optimization algorithm employs gradient descent, starting from the current load state and adjusting parameters along the direction of fastest benefit improvement. During iteration, the marginal benefit of each adjustment step is dynamically evaluated, and optimization stops when the marginal benefit falls below a set threshold. Load transfer rules allow high-load nodes to transfer some tasks to low-load nodes, but spatial proximity and functional similarity requirements must be met.

[0088] Load optimization mechanisms are used to perform load clearing and obtain the results. The core of the clearing is to verify the resource usage of each intersection and route, identify overspending and surplus, and redistribute them through optimization mechanisms. The clearing process first compiles the time resource usage details for each route. Route 1, allocated 130 units, has used 101.4 units, leaving 28.6 units remaining; Route 2, allocated 110 units, has used 71.5 units, leaving 38.5 units remaining. Intersection load clearing checks the deviation between the actual load and the planned load at each intersection. Intersection A's planned load index was 12.2, but the actual load reached 14.8, an overload of 21%; Intersection D's planned load was 8.5, but the actual load was only 6.9, indicating surplus capacity. Overspending items are handled by allocating resources from surplus nodes through the load optimization mechanism, transferring some capacity from Intersection D to Intersection A to alleviate overload pressure. The redistribution of surplus resources follows the principles of proximity and demand priority, with the remaining time of Route 2 being prioritized for allocation to other routes within the same area.

[0089] The load balance status is generated based on the resource usage and load distribution characteristics reflected in the settlement results. The settlement results show that after optimized allocation, the load index of intersection A decreased from 14.8 to 13.1, while that of intersection D increased from 6.9 to 8.2, significantly reducing the load difference between intersections. The balance of line resources is achieved through the redistribution of surplus resources; line 1 receives 10 units of quota transferred from line 2, and line 3 receives 5 units, resulting in a more balanced resource guarantee for each line. Quantitative indicators of load balance include statistics such as load variance, Gini coefficient, and range. After optimization, the load variance decreased from 18.5 to 12.3, and the Gini coefficient decreased from 0.28 to 0.21. The quantitative indicator of load balance status is: BI = 1 - σ² / σ²max, where σ² = Σ(Li - Lavg)² / n is the load variance, Li is the load of each intersection, Lavg is the average load, n is the number of intersections, and σ²max is the theoretical maximum variance. BI ∈ [0,1], and the closer to 1, the more balanced the system. The stability of the equilibrium state was verified through disturbance testing. Within the normal fluctuation range, the load distribution remained relatively stable, without any accumulation of imbalance. The dynamic equilibrium characteristics indicate that the load state adjusts with changes in traffic demand. The equilibrium point during the morning peak differs from that during the off-peak period, but both remain within an acceptable range. The equilibrium margin setting provides adjustment space for emergencies, with each node maintaining a 15-20% load reserve to avoid full-load operation.

[0090] Based on the obtained load balance status, a comprehensive optimization strategy for future operation is formulated. The load balance status provides the optimal benchmark for current resource allocation, with the load index at each intersection stable in the range of 10-14, and the resource utilization rate of each line maintained within a reasonable range of 70-85%. The optimization strategy aims to maintain and improve this balance while enhancing overall operational efficiency. The strategy framework includes three levels: preventative measures, responsive measures, and improvement measures. Preventative measures avoid imbalances through load forecasting and advance allocation; when a high load is predicted at an intersection in 2 hours, traffic diversion preparation begins 15 minutes in advance. Responsive measures quickly adjust to real-time load fluctuations, setting load warning thresholds, and immediately initiating emergency allocation procedures when the threshold is exceeded. The dynamic strategy library contains optimization schemes for different scenarios, with corresponding strategies for special situations such as weekday morning rush hours, holidays, and large-scale events. Multi-objective collaboration ensures a balanced achievement of goals such as public transport priority, traffic efficiency, and resource conservation. The analysis of the load balance status and the formulation of multi-level optimization measures form a comprehensive optimization strategy, ultimately achieving public transport priority traffic scheduling.

[0091] To implement the traffic entropy-based public transport priority scheduling method corresponding to the above method embodiments, and to achieve the corresponding functions and technical effects, see [link to relevant documentation]. Figure 2 , Figure 2 This diagram illustrates a structural block diagram of a traffic entropy-based bus priority traffic dispatching device 200 according to an embodiment of this application. For ease of explanation, only the parts relevant to this embodiment are shown. The traffic entropy-based bus priority traffic dispatching device 200 provided in this embodiment includes:

[0092] Data acquisition module 201 is used to collect the trajectory coordinate sequence output by the bus positioning terminal and perform time prediction processing on the trajectory coordinate sequence to generate arrival time information.

[0093] The load assessment module 202 is used to perform time load assessment on the arrival time information to generate time credit limits for each line, establish a time resource pool based on the time credit limits, and dynamically allocate the time resource pool to obtain the time allocation weight for each line.

[0094] The allocation processing module 203 is used to perform multi-line resource competition analysis based on the arrival time information to determine the operating load index of each line, generate time allocation requirements according to the time allocation weight and the operating load index of each line, and generate a time resource allocation scheme based on the time allocation requirements by performing time load balancing processing.

[0095] The signal control module 204 is used to formulate a quota-based priority execution strategy according to the time resource allocation scheme, apply the priority execution strategy to the phase adjustment to generate signal changes, calculate the traffic entropy value of the signal changes to obtain the current traffic confusion data, and perform phase extension analysis on the traffic confusion data to generate a green light timing scheme.

[0096] Game analysis module 205 is used to perform traffic disturbance propagation analysis on the green light timing scheme to generate road network ripple effect, perform inter-intersection game analysis on the road network ripple effect to generate game equilibrium point, perform entropy propagation analysis based on the road network ripple effect to obtain disorder diffusion parameters, and use the disorder diffusion parameters and the game equilibrium point to construct an entropy reduction control mechanism.

[0097] The collaborative control module 206 is used to generate full-path impact assessment data by combining the entropy reduction control mechanism and the time resource allocation scheme, formulate a chain adjustment strategy based on the full-path impact assessment data, and generate an inter-intersection resource balance scheme based on the chain adjustment strategy.

[0098] The comprehensive optimization module 207 is used to construct a hierarchical coordination mechanism to generate a comprehensive control scheme based on the resource balancing scheme between intersections, perform time load clearing and verification on the comprehensive control scheme to obtain the load balance status, generate a comprehensive optimization strategy based on the load balance status, and complete the bus priority traffic scheduling.

[0099] The aforementioned traffic entropy-based bus priority traffic dispatching device 200 can implement the traffic entropy-based bus priority traffic dispatching method of the above-described method embodiments. The options in the above method embodiments are also applicable to this embodiment, and will not be detailed here. The remaining content of this application's embodiments can be referred to the content of the above method embodiments, and will not be repeated in this embodiment.

[0100] like Figure 3 As shown, the third embodiment of the present invention also provides a computer device, including a memory 301, a processor 302, and a computer program stored in the memory 301 and executable on the processor 302, characterized in that the processor 302 executes the program to implement the steps of the public transport priority traffic scheduling method based on traffic entropy described in the first embodiment of the present invention.

[0101] The above embodiments are not an exhaustive list based on the present invention, and there may be many other embodiments not listed. Any substitutions and improvements made without departing from the concept of the present invention are within the protection scope of the present invention.

Claims

1. A public transport priority traffic scheduling method based on traffic entropy, characterized in that, include: Collect the trajectory coordinate sequence output by the bus vehicle positioning terminal, and perform time prediction processing on the trajectory coordinate sequence to generate arrival time information; The arrival time information is used to perform time load assessment to generate time credit quotas for each line. The time load assessment refers to a comprehensive weighted assessment process based on the on-time rate index calculated from the arrival time deviation data and the historical priority application frequency. The time credit quota refers to the number of resource quota units that a line can use to apply for priority signals, calculated based on the load level and load index in the load assessment results and combined with the line importance coefficient. A time resource pool is established based on the time credit quota, and the time resource pool is dynamically allocated to obtain the time allocation weights for each line. Based on the arrival time information, a multi-line resource competition analysis is performed to determine the operational load index of each line. The line operational load index is a quantitative indicator that represents the operational pressure and urgency of resource demand of the line, calculated by a weighted algorithm by comprehensively considering real-time passenger load factor, departure interval coefficient, operating period weight, and delay accumulation factor. Time allocation requirements are generated according to the time allocation weight and the operational load index of each line. Time load balancing is performed based on the time allocation requirements to generate a time resource allocation scheme. Based on the time resource allocation scheme, a quota-based priority execution strategy is formulated. The priority execution strategy is applied to the phase adjustment to generate signal changes. Traffic entropy value is calculated on the signal changes to obtain the current traffic congestion data. Phase extension analysis is performed on the traffic congestion data to generate a green light timing scheme. Traffic disturbance propagation analysis is performed on the green light timing scheme to generate road network ripple effect. Inter-intersection game analysis is performed on the road network ripple effect to generate game equilibrium point. Entropy propagation analysis is performed on the road network ripple effect to obtain disorder diffusion parameters. Entropy reduction control mechanism is constructed using the disorder diffusion parameters and the game equilibrium point. The entropy reduction control mechanism and the time resource allocation scheme are combined to generate full-path impact assessment data. Based on the full-path impact assessment data, a chain adjustment strategy is formulated, and based on the chain adjustment strategy, an inter-intersection resource balance scheme is generated. Based on the resource balancing scheme between intersections, a hierarchical coordination mechanism is constructed to generate a comprehensive control scheme. The comprehensive control scheme is then verified by time load clearing to obtain the load balance status. Based on the load balance status, a comprehensive optimization strategy is generated to complete the bus priority traffic scheduling.

2. The method according to claim 1, characterized in that, The step of generating time credit limits for each route by evaluating the arrival time information includes: The arrival time information is analyzed to obtain arrival time deviation data for each line; On-time performance indicators for each route are generated based on the arrival time deviation data. The time load assessment results are obtained by performing a time load assessment based on the on-time rate index and the historical priority application frequency. The time credit limit for each line is generated based on the load assessment results.

3. The method according to claim 1, characterized in that, The process of generating a time resource allocation scheme based on the time allocation requirements includes: The time allocation requirements are structured to generate an allocation requirement matrix; Based on the aforementioned allocation demand matrix, a load relationship map between lines is constructed. The time allocation coefficient is obtained by balancing the load relationship graph; A time resource allocation scheme is generated using the time allocation coefficient.

4. The method according to claim 1, characterized in that, The step of calculating traffic entropy values ​​from the signal changes to obtain current traffic congestion data includes: The vehicle operating status caused by the signal change is sampled to obtain vehicle status sampling data; Based on the vehicle state sampling data, generate speed variance parameters and queue length variation parameters; The velocity variance parameter and the queue length variation parameter are subjected to entropy processing to obtain the entropy processing result. The current traffic disorder data is generated based on the entropy value processing result.

5. The method according to claim 1, characterized in that, The step of generating a game equilibrium point through inter-intersection game analysis of the road network spillover effect includes: The road network spillover effect is analyzed to obtain the interest correlation of each intersection; Construct a game relationship matrix between intersections based on the aforementioned interest relationships; Perform payoff analysis on the game relationship matrix to obtain the payoff analysis results; The equilibrium point of the game is generated by solving the equilibrium problem based on the results of the aforementioned benefit analysis.

6. The method according to claim 1, characterized in that, The method of obtaining disorder diffusion parameters through entropy propagation analysis based on the road network ripple effect includes: The propagation path of the road network ripple effect is traced to obtain propagation path data; The speed at which traffic chaos spreads is generated based on the propagation path data; The decay law parameters of the traffic disorder diffusion rate are obtained by analyzing the decay law. The disorder diffusion parameters are generated based on the traffic disorder diffusion rate and the attenuation law parameters.

7. The method according to claim 1, characterized in that, The step of generating an inter-intersection resource balancing scheme based on the interlocking adjustment strategy includes: The interlocking adjustment strategy is decomposed at intersections to obtain the adjustment requirements of each intersection. Compensation amount data is obtained by processing the compensation amount based on the adjustment requirements of each intersection. The compensation data is used to perform inter-intersection collaborative configuration to obtain the collaborative configuration results; Based on the collaborative configuration results, a resource balancing scheme between intersections is generated.

8. The method according to claim 1, characterized in that, The step of performing time-based load clearing verification on the integrated control scheme to obtain the load balance status includes: A comprehensive impact assessment of the aforementioned integrated control scheme is conducted to obtain the overall benefits of the transportation system. A load optimization mechanism is constructed based on the comprehensive benefits of the aforementioned transportation system; The load optimization mechanism is used to perform load clearing processing to obtain the clearing results; A load balance state is generated based on the liquidation process results.

9. A public transport priority dispatching device based on traffic entropy, characterized in that, include: The data acquisition module is used to collect the trajectory coordinate sequence output by the bus positioning terminal, and perform time prediction processing on the trajectory coordinate sequence to generate arrival time information. The load assessment module is used to perform time load assessment on the arrival time information to generate time credit quotas for each line. The time load assessment refers to a comprehensive weighted assessment process based on the on-time rate index calculated from the arrival time deviation data and the historical priority application frequency. The time credit quota refers to the number of resource quota units that the line can use to apply for priority signals, calculated based on the load level and load index in the load assessment results and combined with the line importance coefficient. A time resource pool is established based on the time credit quota, and the time resource pool is dynamically allocated to obtain the time allocation weights for each line. The allocation processing module is used to perform multi-line resource competition analysis based on the arrival time information to determine the operating load index of each line. The line operating load index is a quantitative indicator that represents the line's operating pressure and the urgency of resource demand, calculated by a weighted algorithm by comprehensively considering real-time passenger load factor, departure interval coefficient, operating period weight, and delay accumulation factor. The module generates time allocation requirements based on the time allocation weight and the operating load index of each line, and performs time load balancing processing based on the time allocation requirements to generate a time resource allocation scheme. The signal control module is used to formulate a quota-based priority execution strategy according to the time resource allocation scheme, apply the priority execution strategy to the phase adjustment to generate signal changes, calculate the traffic entropy value of the signal changes to obtain the current traffic chaos data, and perform phase extension analysis on the traffic chaos data to generate a green light timing scheme. The game analysis module is used to perform traffic disturbance propagation analysis on the green light timing scheme to generate road network ripple effects, perform inter-intersection game analysis on the road network ripple effects to generate game equilibrium points, perform entropy propagation analysis based on the road network ripple effects to obtain disorder diffusion parameters, and use the disorder diffusion parameters and the game equilibrium points to construct an entropy reduction control mechanism. The collaborative control module is used to generate full-path impact assessment data by combining the entropy reduction control mechanism and the time resource allocation scheme, formulate a chain adjustment strategy based on the full-path impact assessment data, and generate an inter-intersection resource balance scheme based on the chain adjustment strategy. The integrated optimization module is used to construct a hierarchical coordination mechanism based on the resource balancing scheme between intersections to generate an integrated control scheme, perform time load clearing and verification on the integrated control scheme to obtain the load balance status, generate an integrated optimization strategy based on the load balance status, and complete the bus priority traffic scheduling.

10. A computer device, characterized in that, It includes a memory and a processor; the memory is used to store a computer program; the processor is used to execute the computer program and, when executing the computer program, implement the method as described in any one of claims 1 to 8.

Images