A method and system for optimizing the layout of high-speed monitoring points in an intelligent transportation system

Abstract

The present invention belongs to the technical field of high-speed monitoring point optimization and provides a method and system for optimizing the layout of high-speed monitoring points in an intelligent transportation system. The method obtains information about target monitoring roads; determines the initial area for each monitoring point based on road characteristics and traffic congestion models; uses the increase in the number of identified congestion points and the change in traffic flow after the emergency lane is activated as positive indicators, and the monitoring cost as a negative indicator, and constructs a corresponding benefit evaluation function to form an objective function; solves the problem using a quantum particle swarm optimization algorithm, or obtains several optimization schemes by adjusting the number of monitoring points and evenly distributing them; evaluates the obtained optimization schemes based on a multivariate linear regression model, and selects the final optimization scheme. The present invention can ensure the rational design of monitoring points, facilitate the effective monitoring of traffic congestion, and consider whether to temporarily activate the emergency lane.

Description

Technical Field

[0001] The present invention belongs to the technical field of high-speed monitoring point optimization, and in particular relates to a method and system for optimizing the layout of high-speed monitoring points in an intelligent transportation system. Background Art

[0002] The statements in this section merely provide background information related to the present invention and do not necessarily constitute prior art.

[0003] During highway construction, emergency lanes are often added to certain sections to accommodate emergency vehicles such as engineering rescue, fire rescue, and medical assistance. Typically, emergency lanes are life-saving passages and should not be used arbitrarily. However, if used appropriately, for example, if traffic flow monitoring on upstream, midstream, and downstream sections of a road indicates a high likelihood of congestion, but no accidents have occurred, the use of emergency lanes can be used to reduce traffic density in a timely manner, potentially avoiding a major congestion.

[0004] At the same time, monitoring points on highways are generally arranged at set distances, and the distances between each monitoring point are relatively far, without considering issues such as regional convergence, effective monitoring of traffic congestion, and temporary activation of emergency lanes. Summary of the Invention

[0005] In order to solve the above problems, the present invention proposes a method and system for optimizing the layout of high-speed monitoring points in an intelligent transportation system. The present invention uses indicators such as the increase in the number of congestion point identifications, the reduction in congestion time, and the reduction rate of congestion events as positive indicators, and the monitoring cost as a negative indicator, establishes an objective function, and solves it. This can ensure that the design of monitoring points is reasonable, help to effectively monitor traffic congestion, and consider whether to temporarily activate the emergency lane.

[0006] According to some embodiments, the present invention adopts the following technical solutions:

[0007] A method for optimizing the layout of high-speed monitoring points in an intelligent transportation system comprises the following steps:

[0008] Obtain information of target monitoring roads;

[0009] Determine the preliminary areas for setting up each monitoring point based on road characteristics and traffic congestion model;

[0010] The increase in the number of congestion points identified and the change in traffic flow after the emergency lane is activated are used as positive indicators, and the monitoring cost is used as a negative indicator. A corresponding benefit evaluation function is constructed to form the objective function.

[0011] The number of monitoring points and the positional relationship of the existing monitoring points are used as constraints of the objective function;

[0012] In the preliminary region, based on the objective function and the constraints, a quantum particle swarm optimization algorithm is used to solve the problem and obtain a first optimization solution;

[0013] In the preliminary area, several optimization schemes are obtained by adjusting the number of monitoring points and distributing them evenly, and the obtained optimization schemes are evaluated based on a multiple linear regression model to obtain a second optimization scheme;

[0014] According to the monitoring optimization target, the first optimization scheme or the second optimization scheme is selected as the final optimization scheme.

[0015] As an optional implementation manner, the preliminary areas for setting up each monitoring point include traffic bottleneck areas, areas with a changed number of lanes, and sharp turns or slope areas.

[0016] As an optional implementation, the preliminary area for determining the setting of each monitoring point also includes a monitoring blind area.

[0017] As an optional implementation, the positive indicator of traffic flow changes after the emergency lane is activated includes the reduction in congestion time, the reduction rate of congestion events and the average speed increase rate, which are obtained based on the increase in the number of emergency lanes activated and the comparison of data before and after the emergency lane is activated.

[0018] As an optional implementation, the objective function is Max·ρBenifit-λ·Cost;

[0019] Among them, Benefit represents a positive indicator, ρ represents a positive indicator coefficient, Cost represents cost, and λ represents a negative indicator coefficient.

[0020] As an optional implementation, the benefit evaluation function of the positive indicator is:

[0021] Benefits=a·(Z point )+b·(Δt)+c·(Δf)+d·(ΔS);

[0022] Among them, Z point The number of congestion points identified increases, Δt is the total congestion time reduced by optimizing the distribution of points, Δf is the congestion event reduction rate, ΔS is the average speed improvement rate, and a, b, c, and d are the weight coefficients of the impact of these four benefits on the total benefit.

[0023] As an alternative implementation, the negative indicator function calculates the total cost based on the number and type of monitoring points and maintenance costs. Assuming that the installation and operation costs of each monitoring point are fixed, the total cost is expressed as:

[0024] Where n is the number of monitoring points, c is the fixed cost of each monitoring point, m iis the maintenance cost of the i-th monitoring point.

[0025] As an optional implementation, the process of using the number of monitoring points and the positional relationship of existing monitoring points as constraints of the objective function includes:

[0026] The total number of monitoring points on the target monitoring road, the maximum distance from existing monitoring points, and the setting interval limit of monitoring points are set as constraints.

[0027] As an optional implementation, the process of evaluating each optimization scheme based on the multivariate linear regression model includes: arranging all feasible optimization schemes, and calculating the positive index and negative index of each optimization scheme respectively;

[0028] Construct a multiple linear regression model, use the entropy weight method to process the indicators, and obtain the weights of the multiple linear regression model;

[0029] A multivariate linear regression model is used to perform weighting based on weights, and finally a new calculation result of the indicator is obtained. The solution decision is made based on the calculation result, and the optimal solution is selected.

[0030] An intelligent transportation system high-speed monitoring point layout optimization system, comprising:

[0031] A monitoring point preliminary determination module is configured to obtain information of a target monitoring road and determine a preliminary area for setting each monitoring point based on road characteristics and a traffic congestion model;

[0032] The objective function construction module is configured to use the increase in the number of congestion points identified and the change in traffic flow after the emergency lane is activated as positive indicators, and the monitoring cost as a negative indicator, and to construct a corresponding benefit evaluation function to form an objective function;

[0033] a constraint module configured to use the number of monitoring points and the positional relationship of the existing monitoring points as constraint conditions of the objective function;

[0034] The optimization evaluation module is configured to, within the preliminary area, use a quantum particle swarm optimization algorithm to solve based on the objective function and constraints to obtain a first optimization solution; within the preliminary area, obtain several optimization solutions by adjusting the number of monitoring points and uniformly distributing them, evaluate the obtained optimization solutions based on a multivariate linear regression model to obtain a second optimization solution; and select the first optimization solution or the second optimization solution as the final optimization solution based on the monitoring optimization target.

[0035] An electronic device includes a memory and a processor, and computer instructions stored in the memory and executed on the processor. When the computer instructions are executed by the processor, the steps in the above method are completed.

[0036] Compared with the prior art, the present invention has the following beneficial effects:

[0037] The present invention takes into account the temporary activation of the emergency lane. While controlling costs, it optimizes the layout of existing monitoring points or arranges new monitoring points to improve the scientific nature of the decision-making on the temporary activation of the emergency lane on the road section between two monitoring points.

[0038] The present invention proposes a quantum particle swarm optimization algorithm (QPSO) under multi-objective optimization conditions and a linear regression model decision-making algorithm based on the entropy weight method, which optimizes the layout of video surveillance points, controls costs, and improves the scientific nature of decision-making.

[0039] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, preferred embodiments are given below and described in detail with reference to the accompanying drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0040] The accompanying drawings, which constitute a part of the present invention, are used to provide a further understanding of the present invention. The exemplary embodiments of the present invention and their descriptions are used to explain the present invention and do not constitute improper limitations on the present invention.

[0041] Figure 1 is a flow chart of an optimization method according to an embodiment;

[0042] Figure 2 This is a schematic diagram of a positive indicator acquisition method according to an embodiment;

[0043] Figure 3 This is a schematic diagram of the solution results of the quantum particle swarm algorithm of an embodiment;

[0044] Figure 4 This is a schematic diagram of the results of a multiple linear regression model in an embodiment. DETAILED DESCRIPTION

[0045] The present invention will be further described below with reference to the accompanying drawings and embodiments.

[0046] It should be noted that the following detailed descriptions are illustrative and intended to provide further explanation of the present invention. Unless otherwise specified, all technical and scientific terms used herein have the same meaning as commonly understood by those skilled in the art to which the present invention belongs.

[0047] It should be noted that the terms used herein are only for describing specific embodiments and are not intended to limit the exemplary embodiments according to the present invention. As used herein, unless the context clearly indicates otherwise, the singular form is intended to include the plural form. In addition, it should be understood that when the terms "comprise" and / or "include" are used in this specification, they indicate the presence of features, steps, operations, devices, components and / or combinations thereof.

[0048] In the absence of conflict, the embodiments and features in the embodiments of this application can be combined with each other.

[0049] As described in the background technology, the existing highway monitoring points are far away, taking 3km as an example. It is necessary to increase video monitoring points to improve decision-making accuracy and ensure the scientific nature of decision-making while controlling costs.

[0050] This embodiment provides a method for optimizing the layout of high-speed monitoring points in an intelligent transportation system, including the following steps:

[0051] Obtain information of target monitoring roads;

[0052] Determine the preliminary areas for setting up each monitoring point based on road characteristics and traffic congestion model;

[0053] The increase in the number of congestion points identified and the change in traffic flow after the emergency lane is activated are used as positive indicators, and the monitoring cost is used as a negative indicator. A corresponding benefit evaluation function is constructed to form the objective function.

[0054] The number of monitoring points and the positional relationship of the existing monitoring points are used as constraints of the objective function;

[0055] In the preliminary region, based on the objective function and the constraints, a quantum particle swarm optimization algorithm is used to solve the problem and obtain a first optimization solution;

[0056] In the preliminary area, several optimization schemes are obtained by adjusting the number of monitoring points and distributing them evenly, and the obtained optimization schemes are evaluated based on a multiple linear regression model to obtain a second optimization scheme;

[0057] According to the monitoring optimization target, the first optimization scheme or the second optimization scheme is selected as the final optimization scheme.

[0058] First, analyze the road section characteristics and monitoring needs. It is necessary to determine whether there are service area ramps, slopes and sharp bends. Vehicles tend to converge in these areas, causing traffic congestion.

[0059] For example, a monitoring point can be arranged about 1 km near the ramp of the service area.

[0060] The monitoring deployment problem within 3 km is then transformed into a monitoring deployment problem within a 3-x distance, where x is the distance between the two monitoring points. Because there are no special road conditions elsewhere, in some cases, monitoring points can be evenly distributed within the 3-x distance range. However, this approach generally fails to consider issues such as the temporary use of emergency lanes in emergencies and the accurate monitoring of congestion.

[0061] In this embodiment, Figure 1 As shown in the figure, we use the increase in the number of congestion point identifications, the reduction in congestion time, and the reduction rate of congestion events as positive indicators, and the monitoring cost as a negative indicator to establish the objective function. At the same time, we use benefit evaluation, cost, and the number of monitoring as our constraints.

[0062] The Quantum Particle Swarm Optimization (QPSO) algorithm is used to solve the optimization goal of the objective function. In addition, a linear regression model based on the entropy weight method is used to evaluate multiple solutions, and the results obtained by different methods are compared to obtain the final optimization result.

[0063] Each step is described in detail below.

[0064] (1) Determine the location of the monitoring point

[0065] Based on road characteristics and traffic congestion models, monitoring points should be set up at the following key locations:

[0066] Traffic bottleneck areas (for example, ramp entrances and exits): Since vehicles tend to gather at ramp intersections, leading to increased traffic flow, it is recommended to set up monitoring points near the ramps or at the ramp entrances to monitor traffic conditions in real time and detect congestion in a timely manner.

[0067] Areas where the number of lanes changes: In some sections of road, the number of lanes may change, especially where emergency lanes may be converted into regular lanes. Monitoring points should be set up to monitor the impact of lane changes on traffic flow.

[0068] Sharp turns or slopes: These areas may affect vehicle speed and traffic density. Vehicles may slow down, leading to increased traffic density. Therefore, these areas are also the focus of monitoring.

[0069] (2) Layout density of monitoring points

[0070] Evenly distributed: If monitoring points are already set at the start and end of a road section, monitoring points can be appropriately set up in the middle section based on the road characteristics. For example, a monitoring point can be set up at a certain interval (approximately 500 to 1000 meters) to ensure that the entire road section is within the monitoring range.

[0071] Covering monitoring blind spots: If there are monitoring blind spots due to undulating terrain or other factors on the road (such as tunnels or mountain obstructions), additional monitoring points should be added near these blind spots to ensure the integrity of the monitoring data.

[0072] The final placement strategy of monitoring points can be determined by maximizing the positive indicators while controlling the cost. This can be achieved by optimizing the following objective function:

[0073] Max·ρBenifit-λ·Cost

[0074] Benefit represents a positive indicator, including the increase in the number of identified congestion points, the reduction in congestion time, and the reduction rate of congestion events. ρ represents the positive indicator coefficient. Cost represents the cost, and λ represents the negative indicator coefficient.

[0075] This embodiment can decide whether to open the emergency lane based on the congestion observation status. After the emergency lane is opened, it can be quantified to obtain quantitative indicators such as average speed improvement rate, traffic density improvement rate, traffic flow improvement rate, congestion time reduction, and congestion event reduction rate. These quantitative indicators are used as positive indicators, such as Figure 2 shown.

[0076] Assume that the number of monitoring points is n, which can establish the following positive indicators at a certain moment.

[0077] a. Increase the number of congestion points identified:

[0078] Z point =n·P average ·Y all

[0079] Among them, Z point Refers to the number of increased congestion points, n is the number of increased monitoring points, P average is the average probability (average accuracy) that the monitoring point can identify the congestion point, Y all The total number of actual congestion points.

[0080] b. Increase in the number of emergency lanes

[0081] if.M(KNN z ,SVM z ,LWR z )>1,F=1

[0082]

[0083] The above formula describes the voting mechanism formula of the KNN, SVM, and LWR models based on random forest.

[0084] Among them, M is the voting function of the three algorithms, that is, the results of the three algorithms are logically calculated. When the KNN model predicts congestion, KNN z Set to 1, otherwise set to 0. z ,LWR z Similarly, if the M voting function is greater than 1, F is set to 1, indicating that the emergency lane is open, otherwise it is 0 and the emergency lane is not open. z It refers to the increase in the number of emergency lanes in use.

[0085] In addition, this embodiment provides a method for determining the activation of an emergency lane, comprising the following steps:

[0086] Obtain historical surveillance videos of each monitoring point and perform video processing on the historical surveillance videos;

[0087] Extracting vehicle information from the processed video, determining the number and speed of each type of vehicle based on the extracted vehicle information, redefining and calculating traffic flow parameters, the traffic flow parameters including vehicle density, speed, and traffic volume, and determining a change trend of the traffic flow parameters over time;

[0088] Based on traffic flow parameters, the initial speed threshold for emergency lane opening is analyzed, and parameters of the congestion probability distribution are estimated. This allows the construction of a traffic flow interruption probability model, the calculation of the probability that road capacity is less than the observed flow rate, and the determination of the final threshold for emergency lane opening.

[0089] Based on traffic flow parameters and final thresholds, multiple classification models are trained;

[0090] According to traffic flow parameters, the traffic congestion warning model is trained;

[0091] Obtain surveillance video of the target monitoring point within the target time period, perform video processing and data extraction, and use the trained traffic congestion warning model based on the calculated real-time traffic flow parameters to predict traffic flow density and warning points;

[0092] Based on the prediction results, multiple pre-trained classification models are used to decide whether to open the emergency lane. The decision results are integrated to obtain the final judgment result on whether the emergency lane needs to be opened.

[0093] The process of extracting vehicle information from the processed video includes: extracting vehicle information from the processed video using a target detection algorithm, and introducing a Kalman filter algorithm to improve recognition accuracy and target tracking accuracy.

[0094] In this embodiment, the process of redefining and calculating traffic flow parameters includes: the traffic flow density is the total number of vehicles appearing in a frame;

[0095] The speed calculation process includes: calculating the speed of the same vehicle by tracking the position change in consecutive frames, that is, assigning a unique ID to each vehicle and recording its position in consecutive frames, and calculating the vehicle's position between consecutive frames divided by the time interval to obtain the vehicle's speed;

[0096] The traffic flow is the number of vehicles passing through the corresponding monitoring point per unit time.

[0097] In this embodiment, the warning process of the traffic congestion warning model includes: evaluating whether the current traffic density is approaching the critical density based on the traffic density data recorded by the relevant monitoring points; if the relevant monitoring points all reach the critical density range and the density growth exceeds the set value, then issuing a traffic congestion alarm;

[0098] By calculating the rate of change of the flow at relevant monitoring points, if the flow rate drops by more than a threshold and the density increases, the congestion risk increases and a traffic congestion alarm is issued;

[0099] Set the predicted time and, combined with the current traffic density, determine whether congestion will occur within the set predicted time. If so, issue a traffic congestion alert.

[0100] In this embodiment, the traffic congestion warning model is an LWR model. After the traffic congestion warning model is trained, it is cross-validated using the TPI-LSTM method. The specific process includes: based on dynamic vehicle position information, the operating speed of roads with different functional levels is obtained, and the weight of the road in the entire network is calculated according to the different road functions and traffic data. Finally, through people's perception and judgment of congestion, the index indicator value is converted to obtain a training sample. Based on the sliding window calculation method, the index indicator value is divided to obtain a training sample. Based on the training sample, the long short-term memory network neural network model is used in combination with historical data to predict whether a continuous warning state will occur, and multiple prediction results are obtained.

[0101] The prediction results of the LWR model can be used to predict within a certain period, and the results of the LSTM model can be used to predict beyond the set period.

[0102] In this embodiment, the process of determining the final threshold for opening the emergency lane includes: establishing a congestion index model based on traffic flow parameters, analyzing the initial speed threshold for opening and activating the emergency lane, estimating the parameters of the congestion probability distribution through the maximum likelihood estimation function, thereby constructing a traffic flow interruption probability model, calculating the probability that the road capacity is less than the observed flow rate, and visualizing the relationship between the congestion probability and traffic flow, thereby determining the traffic flow for activating the emergency lane.

[0103] In this embodiment, the process of establishing a congestion index model includes: grading traffic conditions and road operation conditions according to traffic density, traffic flow and speed, determining a traffic congestion index, calculating the index indicator value, and drawing a visual line graph of its changes over time.

[0104] In this embodiment, a traffic flow interruption probability model is constructed to calculate the probability that the road capacity is less than the observed traffic flow, and the process of visualizing the relationship between congestion probability and traffic flow includes: establishing a traffic flow interruption probability model:

[0105]

[0106] Where q is the number of vehicles passing through each lane within a set time, c is the capacity, that is, the traffic capacity; P(c, q) is the probability that the capacity is less than the observed traffic flow; q i is the traffic flow observed at interval i, which represents the traffic flow before the speed drop; ki refers to q>q i The number of time intervals; d i The traffic flow reaches q i The number of time intervals when each congestion is considered separately, d i =1; B is the set of congestion intervals B1, B2, ...};

[0107] Estimate distribution parameters by applying the maximum likelihood function;

[0108] Based on historical data, the relationship between traffic flow parameters is determined and a scatter plot is drawn. Based on the relationship, the speed threshold is preliminarily estimated, and the traffic flow interruption probability model is solved to calculate the congestion probability. A scatter plot of the relationship between congestion probability and flow rate is drawn. Based on the relationship scatter plot, the threshold for activating the emergency lane is determined.

[0109] In this embodiment, the process of training multiple classification models includes:

[0110] The final threshold determined by the traffic flow interruption probability model is used as the training set and as the input of the KNN model, SVM model and LWR traffic flow model for decision training.

[0111] The process of fusing the decision results includes: using the random forest voting idea to fuse the decision results of the KNN model, SVM model and LWR traffic flow model to complete the judgment of whether to open the emergency lane.

[0112] c. Reduction in congestion time, reduction rate of congestion incidents, and average speed increase rate

[0113] According to the established model, after obtaining the data on the increase in the number of emergency lanes activated, the corresponding indicators such as the reduction in congestion time and the reduction rate of congestion events can be obtained. The quantitative indicators are used as positive indicators of the optimization objective function.

[0114] Based on the above indicators such as the increase in the number of congestion point identifications, the increase in the number of emergency lanes activated, the reduction in congestion time, the reduction rate of congestion events, and the average speed improvement rate, a corresponding benefit evaluation function can be written as the first term of the objective function.

[0115] Benefits=a·(Z point )+b·(Δt)+c·(Δf)+d·(ΔS)

[0116] Among them, Δt is the total congestion time reduced by optimizing the layout, Δf is the reduction rate of congestion events after improvement, and a, b, c, and d are the weight coefficients of the impact of these four benefits on the total benefit.

[0117] Negative indicator model establishment

[0118] The cost function can calculate the total cost based on the number and type of monitoring points and maintenance costs. Assuming that the installation and operation costs of each monitoring point are fixed, the total cost can be expressed as:

[0119]

[0120] Where n is the number of monitoring points, c is the fixed cost of each monitoring point, m i is the maintenance cost of the i-th monitoring point.

[0121] Constraint setting:

[0122] We know that we need to install at least one camera within 1 km ± 200 m of the original fourth monitoring point, that the monitoring interval must be greater than 500 m within the remaining distance, and that the number of cameras between the original third and fourth monitoring points must be less than or equal to 4. The positive and negative indicator functions described above are also constraints on the objective function.

[0123] In order to ensure the persuasiveness and accuracy of the results, we use another idea and method to further solve and verify the results, namely the multiple linear regression scheme evaluation method.

[0124] In this approach, the problem is simplified and basic simplifications and assumptions are made:

[0125] 1. Based on terrain analysis, a monitoring point is set at approximately 1 km ± 200 m near the ramp at the fourth point. This continuity condition can be simplified to require that at least one monitoring point be located at 1.2 km, 1 km, and 0.8 km from the fourth point.

[0126] 2. Based on the above assumptions and simplifications, the remaining 3 km of road is divided into three scenarios: 1.8 km, 2 km, and 2.2 km. Since existing monitoring points are approximately 1 km apart, to control costs, we assume that there are at most two cameras in this remaining section of road.

[0127] Based on the above assumptions, several optimization schemes can be obtained, and each scheme is listed below.

[0128] Plan 1: Set up one monitoring point 0.8 km near the fourth point, and two monitoring points evenly distributed on the remaining sections of the road;

[0129] Option 2: Set up one monitoring point 0.8 km near the fourth point, and evenly distribute one monitoring point on the remaining sections of the road;

[0130] Plan 3: Set up one monitoring point 1km away from the fourth point, and two monitoring points evenly distributed on the remaining sections of the road;

[0131] Plan 4: Set up one monitoring point every 1km near the fourth point, and evenly distribute one monitoring point on the remaining sections of the road;

[0132] Plan 5: Set up one monitoring point 1.2 km near the fourth point, and two monitoring points evenly distributed on the remaining sections of the road

[0133] Plan 6: Set up one monitoring point 1.2 km near the fourth point, and evenly distribute one monitoring point on the remaining sections of the road;

[0134] Based on the above schemes, the positive and negative indicators of each scheme can be calculated.

[0135] The multiple linear regression model based on entropy weight method is used to make scheme evaluation decisions.

[0136] The expression of the multiple linear regression model is:

[0137] f(x)=k T x+b

[0138] y=k1x1+k2x2+...+k d x d +b×1

[0139] Here, x is the input vector containing multiple features (independent variables); f(x) is the output or response of the model (the predicted target variable); k is the feature weight; b is the intercept or bias of the model; our goal is to make f(x) as close as possible to the true observation y by learning k and b.

[0140] Determination of the weight of each indicator feature:

[0141] To determine the feature weight k, we use the entropy weighting method to process the indicators and determine their weights. The entropy weighting method processes known data to derive weights for influencing factors. Its fundamental principle is to determine objective weights based on the variability of the indicators. The advantage of the entropy weighting method lies in its objective weighting, which avoids bias caused by human factors by determining the weights based on the variability of each indicator's values. Compared to subjective weighting methods, it offers greater accuracy and objectivity, enabling better interpretation of the results.

[0142] During model building, the first step is to normalize the indicators. This is because different indicators have different meanings. Some indicators are better when they are larger, known as positive indicators; some are better when they are smaller, known as negative indicators; and some indicators are best at a certain point or interval, known as moderate indicators. To facilitate evaluation, all indicators should be normalized.

[0143] Among them, positive indicators:

[0144] x′ ij =x ij

[0145] Negative indicators:

[0146] x i ' j =max(x ij )-x ij

[0147] Moderation indicators:

[0148]

[0149] In this question, the increase in the number of congestion point identifications, the reduction in congestion time, and the reduction rate of congestion events are positive indicators, and the cost is a negative indicator.

[0150] Secondly, data standardization is carried out, and indicators are scaled proportionally so that they fall within a specific range, thereby removing the impact of dimensions and enabling indicators of different units or magnitudes to be compared and weighted.

[0151]

[0152] Finally, calculate the information entropy:

[0153]

[0154] Finally, a multiple linear regression model is used for weighting, and finally a new indicator is obtained for program decision-making.

[0155] In this example, it is assumed that the installation and subsequent maintenance costs of each camera total RMB 100,000. The indicators of each solution are compared in Table 1.

[0156] Table 1 Result indicators

[0157]

[0158] After using the entropy weight method and the multivariate linear regression method to calculate the overall index, the evaluation of the six traffic congestion relief options revealed that Option 1, with the highest score of 0.2269, emerged as the winner, demonstrating the best overall benefits. Despite its higher cost, it significantly reduced congestion points and duration. Options 3 and 5 also performed well, scoring 0.2150 and 0.1910, respectively, demonstrating a good balance between cost and benefit. In contrast, Option 6 scored the lowest, at 0.1136, demonstrating a suboptimal overall benefit. Options 2 and 4 also received relatively low scores of 0.1326 and 0.1205, respectively, demonstrating that a single low-cost strategy does not always yield the highest benefits.

[0159] In summary, Option 1 is the most effective option. Figure 4 These assessment results emphasize that when formulating mitigation measures, costs and benefits should be considered comprehensively to achieve optimal allocation of resources and efficient optimization of traffic management.

[0160] According to the results obtained by the above method, three new monitoring points should be installed between the third and fourth points, and their intervals from the third point are 766m, 766m, 766m (evenly distributed) and 700m respectively.

[0161] Based on the quantum particle swarm algorithm, the core of the QPSO algorithm is to simulate the behavior of quantum particles and utilize quantum superposition and quantum tunneling to enhance the search process. In QPSO, particles no longer have a fixed position and velocity, but instead exist in the search space as a probability cloud. This means that particles can exist in multiple locations simultaneously, thereby increasing the coverage of the search space.

[0162] The specific process includes:

[0163] 1. Initialization: Randomly initialize the position and probability distribution of each particle in the particle swarm.

[0164] 2. Evaluate fitness: calculate the fitness value of each particle.

[0165] 3. Update individual extreme value: If the fitness of the current particle is better than the previously recorded individual extreme value, the individual extreme value is updated.

[0166]

[0167] Where μ is the current position of the particle, σ is the standard deviation, and x is the new position.

[0168] 4. Update the global extreme value: If the fitness of the current particle is better than the global extreme value, then update the global extreme value.

[0169] 5. Update probability distribution: Update the probability distribution of each particle based on individual extreme values ​​and global extreme values.

[0170] 6. Repeat steps 2-5 until the termination condition (such as number of iterations, fitness threshold, etc.) is met.

[0171] Assume that the particle population size is N, the iterative process is the t-th step, the particle moves in D-dimensional space, the potential well of the particle in the d-th dimension is Pid(t), and the update equation of the particle x(t) is as follows:

[0172]

[0173] p id (t)=φ(t)·p id (t)+(1-φ(t))·G d (t)

[0174]

[0175] Where P(t) is the current optimal position of the particle, G(t) and C(t) are the global optimal position and average optimal position of the population. and u(t) are random numbers uniformly distributed in the interval [0, 1]: β is the contraction and expansion factor, which is the only control parameter of QPSO.

[0176] Through quantum superposition and quantum tunneling, QPSO can more efficiently explore the global search space and reduce the risk of falling into local optima. QPSO allows particles to exist in multiple locations simultaneously, thereby improving search efficiency. QPSO's quantum behavior reduces the possibility of premature convergence, allowing the algorithm to maintain good diversity during the iteration process.

[0177] Table 1 QPSO iterative process parameters display

[0178]

[0179] As shown in the table, the quantum particle swarm optimization (QPSO) algorithm demonstrated its strong performance in parameter optimization after 50 iterations. At the beginning of the algorithm, the value of parameter C was large, while the value of parameter gamma was small, likely reflecting the algorithm's extensive exploration of the parameter space in the initial stages. As the iterations progressed, parameter C gradually decreased, while gamma remained low, indicating that the algorithm was gradually narrowing its search range and seeking more precise parameter values. The algorithm's fitness (accuracy) steadily improved over the course of the iterations, reaching a peak of 84.07% at the 49th and 50th iterations. This result not only demonstrates the QPSO algorithm's effectiveness in finding the optimal parameter combination but also demonstrates its stability and convergence during the iteration process. Ultimately, the algorithm converged to parameters C = 6.557312481545432 and gamma = 0.001, representing the optimal solution to the given optimization problem. This performance of the QPSO algorithm highlights its potential and value in solving complex optimization problems.

[0180] like Figure 3 As shown in the figure, according to the solution results of the quantum ion swarm optimization algorithm (QPSO), four monitoring points should be added at the third and fourth points. Starting from the third point, their intervals are 639m, 639m, 579m, 611m, and 532m respectively.

[0181] The results obtained by the two methods are compared, as shown in Table 3:

[0182] Table 3 Method comparison

[0183]

[0184] Taking all factors into consideration, if the budget is sufficient and higher efficiency and effectiveness are sought, the quantum particle swarm optimization algorithm under multi-objective optimization may be a better choice, as it outperforms Method 2 in terms of the number of monitoring points, the increase in the number of congestion points identified, and the reduction in congestion time. However, if the budget is limited and a balance between cost and efficiency is needed, the multivariate linear regression model based on the entropy weight method provides a more cost-effective solution.

[0185] An intelligent transportation system high-speed monitoring point layout optimization system, comprising:

[0186] A monitoring point preliminary determination module is configured to obtain information of a target monitoring road and determine a preliminary area for setting each monitoring point based on road characteristics and a traffic congestion model;

[0187] The objective function construction module is configured to use the increase in the number of congestion points identified and the change in traffic flow after the emergency lane is activated as positive indicators, and the monitoring cost as a negative indicator, and to construct a corresponding benefit evaluation function to form an objective function;

[0188] a constraint module configured to use the number of monitoring points and the positional relationship of the existing monitoring points as constraint conditions of the objective function;

[0189] The optimization evaluation module is configured to, within the preliminary area, use a quantum particle swarm optimization algorithm to solve based on the objective function and constraints to obtain a first optimization solution; within the preliminary area, obtain several optimization solutions by adjusting the number of monitoring points and uniformly distributing them, evaluate the obtained optimization solutions based on a multivariate linear regression model to obtain a second optimization solution; and select the first optimization solution or the second optimization solution as the final optimization solution based on the monitoring optimization target.

[0190] It will be understood by those skilled in the art that embodiments of the present invention may be provided as methods, systems, or computer program products. Thus, the present invention may take the form of an entirely hardware embodiment, an entirely software embodiment, or an embodiment combining software and hardware. Furthermore, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to magnetic disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0191] The present invention is described with reference to flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowcharts and / or block diagrams, as well as combinations of processes and / or blocks in the flowcharts and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowcharts and / or block diagrams. Figure 1 a process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0192] These computer program instructions may also be stored in a computer readable memory that can direct a computer or other programmable data processing device to work in a specific manner, so that the instructions stored in the computer readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 a process or multiple processes and / or boxes Figure 1 The function specified in one or more boxes.

[0193] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operational steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing the instructions executed on the computer or other programmable device for implementing the process. Figure 1 a process or multiple processes and / or boxes Figure 1 A step that specifies a function in one or more boxes.

[0194] The foregoing description is merely a preferred embodiment of the present invention and is not intended to limit the present invention. Those skilled in the art will readily appreciate that various modifications and variations of the present invention are possible. Any modifications, equivalent substitutions, or improvements made by those skilled in the art that fall within the spirit and principles of the present invention and do not require creative effort are intended to be within the scope of protection of the present invention.

Claims

1. A method for optimizing the layout of high-speed monitoring points in an intelligent transportation system, characterized in that: The following steps are involved: Obtain information of target monitoring roads; Determine the preliminary areas for setting up each monitoring point based on road characteristics and traffic congestion model; The increase in the number of congestion points identified and the change in traffic flow after the emergency lane is activated are used as positive indicators, and the monitoring cost is used as a negative indicator. The corresponding benefit evaluation function is constructed to form the objective function. The objective function is: Max ( λ ); Among them, Benefit represents a positive indicator. represents the positive indicator coefficient, Cost represents the cost, represents the negative indicator coefficient; The benefit evaluation function of the positive indicator is: ； Among them, Z point Increased number of congestion point identifications, The total congestion time is reduced by optimizing the distribution of points. is the congestion incident reduction rate, is the average speed improvement rate, a, b, c, d is the weight coefficient of the impact of these four benefits on the total benefit; The function of the negative indicator calculates the total cost based on the number and type of monitoring points and maintenance costs. Assuming that the installation and operation costs of each monitoring point are fixed, the total cost is expressed as: ; in is the number of monitoring points, is the fixed cost of each monitoring point, It is Maintenance cost of each monitoring point; The number of monitoring points and the positional relationship of the existing monitoring points are used as constraints of the objective function; In the preliminary region, based on the objective function and the constraints, a quantum particle swarm optimization algorithm is used to solve the problem and obtain a first optimization solution; In the preliminary area, several optimization schemes are obtained by adjusting the number of monitoring points and distributing them evenly, and the obtained optimization schemes are evaluated based on a multiple linear regression model to obtain a second optimization scheme; According to the monitoring optimization target, the first optimization scheme or the second optimization scheme is selected as the final optimization scheme.

2. The method for optimizing the layout of high-speed monitoring points in an intelligent transportation system according to claim 1, wherein: The preliminary areas for setting up each monitoring point include traffic bottleneck areas, areas with changed number of lanes, sharp turns or slopes, and monitoring blind spots.

3. The method for optimizing the layout of high-speed monitoring points in an intelligent transportation system according to claim 1, wherein: The positive indicators of traffic flow changes after the emergency lane is activated include the reduction in congestion time, the reduction rate of congestion events and the average speed increase rate, which are obtained based on the increase in the number of emergency lanes activated and the comparison of data before and after the emergency lane is activated.

4. The method for optimizing the layout of high-speed monitoring points in an intelligent transportation system according to claim 1, wherein: The process of using the number of monitoring points and the positional relationship of the existing monitoring points as the constraint conditions of the objective function includes: The total number of monitoring points on the target monitoring road, the maximum distance from existing monitoring points, and the setting interval limit of monitoring points are set as constraints.

5. The method for optimizing the layout of high-speed monitoring points in an intelligent transportation system according to claim 1, wherein: The process of evaluating each optimization scheme based on the multivariate linear regression model includes: arranging all feasible optimization schemes and calculating the positive and negative indicators of each optimization scheme; Construct a multiple linear regression model, use the entropy weight method to process the indicators, and obtain the weights of the multiple linear regression model; A multivariate linear regression model is used to perform weighting based on weights, and finally a new calculation result of the indicator is obtained. The solution decision is made based on the calculation result, and the optimal solution is selected.

6. An intelligent transportation system high-speed monitoring point layout optimization system, characterized by: include: A monitoring point preliminary determination module is configured to obtain information of a target monitoring road and determine a preliminary area for setting each monitoring point based on road characteristics and a traffic congestion model; The objective function construction module is configured to use the increase in the number of congestion points identified and the change in traffic flow after the emergency lane is activated as positive indicators, and the monitoring cost as a negative indicator, and to construct a corresponding benefit evaluation function to form an objective function. The objective function is Max ( λ ); Among them, Benefit represents a positive indicator. represents the positive indicator coefficient, Cost represents the cost, represents the negative indicator coefficient; The benefit evaluation function of the positive indicator is: ； Among them, Z point Increased number of congestion point identifications, The total congestion time is reduced by optimizing the distribution of points. is the congestion incident reduction rate, is the average speed improvement rate, a, b, c, d is the weight coefficient of the impact of these four benefits on the total benefit; The function of the negative indicator calculates the total cost based on the number and type of monitoring points and maintenance costs. Assuming that the installation and operation costs of each monitoring point are fixed, the total cost is expressed as: ; in is the number of monitoring points, is the fixed cost of each monitoring point, It is Maintenance cost of each monitoring point; a constraint module configured to use the number of monitoring points and the positional relationship of the existing monitoring points as constraint conditions of the objective function; The optimization evaluation module is configured to, within the preliminary area, use a quantum particle swarm optimization algorithm to solve based on the objective function and constraints to obtain a first optimization solution; within the preliminary area, obtain several optimization solutions by adjusting the number of monitoring points and uniformly distributing them, evaluate the obtained optimization solutions based on a multivariate linear regression model to obtain a second optimization solution; and select the first optimization solution or the second optimization solution as the final optimization solution based on the monitoring optimization target.

7. An electronic device, characterized in that: The method comprises a memory and a processor, and computer instructions stored in the memory and executed on the processor, wherein the steps of the method according to any one of claims 1 to 5 are completed when the computer instructions are executed by the processor.

Images