Highway energy consumption monitoring interface-based analysis report generation method and system

Abstract

The embodiment of the application discloses a kind of based on expressway energy consumption monitoring interface analysis report generation method and system, method includes: through the energy consumption monitoring interface of target expressway road network, multiple-source data screening is carried out, and the first energy consumption monitoring data of road network infrastructure operation and the second energy consumption monitoring data of passing vehicle driving are acquired;The two groups of data are analyzed in space-time correlation, and multidimensional energy consumption correlation analysis result is generated;Energy consumption influencing factor in multidimensional energy consumption correlation analysis result is regarded as model input item, vehicle driving path planning parameter and infrastructure operation adjustment parameter are regarded as decision variable, mapping relationship is established to generate collaborative decision model, and the collaborative optimization decision result under specified road network operation scene is determined;Finally, according to the decision result, the graphic and text analysis report containing multidimensional energy consumption state characteristics and collaborative optimization suggestion is generated, and the collaborative optimization management of expressway road network energy consumption is realized.

Description

Technical Field

[0001] This application relates to the field of data processing technology, and in particular to a method and system for generating analysis reports based on a highway energy consumption monitoring interface. Background Technology

[0002] Energy consumption management is a crucial aspect of highway operation and management. Currently, the monitoring and management of highway network energy consumption has made some progress. On the one hand, in terms of energy consumption monitoring, methods are available to monitor the energy consumption of some infrastructure such as lighting and ventilation equipment, and basic energy consumption data of vehicles traveling on highways can be obtained, such as through vehicle fuel consumption records. On the other hand, in terms of data analysis and processing, there are also some analytical methods for specific energy consumption data, such as simple statistical analysis of the changes in energy consumption of a certain type of infrastructure over time.

[0003] However, existing technologies lack analysis of the spatiotemporal correlation between different energy consumption data in the presentation of energy consumption analysis reports, cannot deeply explore the factors affecting energy consumption, and do not comprehensively consider the collaborative decision-making of vehicle driving paths and infrastructure operation strategies (cannot achieve collaborative optimization of the two). Summary of the Invention

[0004] This application provides a method and system for generating analysis reports based on a highway energy consumption monitoring interface.

[0005] This application provides a method for generating analysis reports based on a highway energy consumption monitoring interface, applied to an analysis report generation system. The method includes: performing multi-source data filtering processing through the energy consumption monitoring interface of a target highway network to obtain first energy consumption monitoring data related to the operation of the road network infrastructure and second energy consumption monitoring data related to the travel process of passing vehicles; performing spatiotemporal correlation analysis processing on the first and second energy consumption monitoring data to generate a multidimensional energy consumption correlation analysis result for the target highway network; and using the energy consumption influencing factors in the multidimensional energy consumption correlation analysis result as model input items, vehicle travel path planning parameters, and basic... Facility operation adjustment parameters are used as decision variables. A collaborative decision model is generated by establishing a mapping relationship between the model inputs and the decision variables to make collaborative optimization decisions on vehicle travel paths and infrastructure operation strategies. The collaborative decision model is used to determine the collaborative optimization decision results for the target highway network under a specified network operation scenario. A graphic analysis report containing multi-dimensional energy consumption status characteristics and collaborative optimization suggestions is generated based on the collaborative optimization decision results. The multi-dimensional energy consumption status characteristics are obtained by feature extraction and merging of different types of decision results in the collaborative optimization decision results. The graphic analysis report is integrated and processed according to a preset report structure.

[0006] One embodiment of this application provides an analysis report generation system, including:

[0007] A processor; a storage device having a computer program stored thereon; a network interface for providing network communication functions; when the computer program is executed by the processor, the processor enables the processor to implement any of the aforementioned methods for generating analysis reports based on highway energy consumption monitoring interfaces.

[0008] One embodiment of this application provides a readable storage medium storing a program or instructions, which, when executed by a processor, implements the steps of the analysis report generation method based on a highway energy consumption monitoring interface. Attached Figure Description

[0009] Figure 1 This is a flowchart illustrating a method for generating analysis reports based on a highway energy consumption monitoring interface, as provided in an embodiment of this application.

[0010] Figure 2 This is a schematic diagram of the basic structure of an analysis report generation system provided in an embodiment of this application.

[0011] Figure 3 This is a functional block diagram of an analysis report generation device based on a highway energy consumption monitoring interface, provided in an embodiment of this application. Detailed Implementation

[0012] Please see Figure 1 , Figure 1 This is a flowchart of an analysis report generation method based on a highway energy consumption monitoring interface provided in this application embodiment. The method can be executed by the analysis report generation system or jointly executed by the analysis report generation system and the server. The method may include steps 110-140.

[0013] Step 110: Perform multi-source data filtering and processing through the energy consumption monitoring interface of the target highway network to obtain the first energy consumption monitoring data related to the operation of the road network infrastructure and the second energy consumption monitoring data related to the driving process of passing vehicles.

[0014] In this embodiment, the target highway network is a road network system composed of multiple interconnected highways within a certain region, and its energy consumption monitoring interface integrates access interfaces for multiple data sources. When performing multi-source data filtering processing, the time range and spatial range of data filtering are first determined, for example, selecting the past month as the time range and all road sections of the target highway network as the spatial range.

[0015] The primary energy consumption monitoring data comes from various sensors installed on road network infrastructure, such as power sensors for lighting systems, energy consumption metering devices for ventilation systems, and power consumption monitoring modules for traffic signal control systems. Through the data filtering function of the energy consumption monitoring interface, data directly related to the operation of these infrastructures are selected, such as real-time power data of lighting systems, runtime data of ventilation systems, and on / off status data of traffic signals. These data are then integrated into the primary energy consumption monitoring data set. Each data record in this set includes an infrastructure type identifier, a monitoring timestamp, an energy consumption value, and corresponding spatial location information.

[0016] The sources of the second energy consumption monitoring data include on-board energy consumption monitoring equipment installed on vehicles, vehicle identification systems at highway entrances and exits, and traffic flow monitoring devices on road sections. Data related to the vehicle's driving process is filtered out through the energy consumption monitoring interface, such as real-time fuel consumption data, driving speed data, lane information, and road section number. This data is then integrated into a second energy consumption monitoring data set. Each data record in this set includes the vehicle identifier, monitoring timestamp, energy consumption-related parameters, and corresponding driving location information.

[0017] During the screening process, it is necessary to perform preliminary validity checks on the data and remove obviously abnormal data, such as data with negative energy consumption values ​​or data that are outside the reasonable range, in order to ensure the accuracy and reliability of the first and second energy consumption monitoring data obtained.

[0018] Step 120: Perform spatiotemporal correlation analysis on the first energy consumption monitoring data and the second energy consumption monitoring data to generate multidimensional energy consumption correlation analysis results for the target highway network.

[0019] Step 121: Perform time-series decomposition processing on the first energy consumption monitoring data to extract the energy consumption change trend characteristics of infrastructure energy consumption at different time granularities, and perform vehicle flow statistics processing on the second energy consumption monitoring data to obtain the vehicle flow change characteristics at the corresponding time granularity. Then, associate and match the energy consumption change trend characteristics and vehicle flow change characteristics at the same time granularity to establish an energy consumption flow correlation curve in the time dimension.

[0020] In this embodiment, the first step is to use a time series decomposition method to process the first energy consumption monitoring data, decomposing the infrastructure energy consumption data into long-term trend components, seasonal periodic components, and random fluctuation components to extract energy consumption change trend characteristics at different time granularities (such as hour, day, week, and month). For example, for the daily time granularity, the change trend of lighting system energy consumption at different times of the day (such as morning, noon, evening, and night) is analyzed to obtain the daily change trend characteristics of lighting system energy consumption; for the weekly time granularity, the change trend of ventilation system energy consumption at different weekdays and weekends within a week is analyzed to obtain the weekly change trend characteristics of ventilation system energy consumption.

[0021] Simultaneously, the second energy consumption monitoring data undergoes vehicle flow statistics processing. Vehicle flow through each segment of the target highway network is statistically analyzed at the same time granularity, yielding vehicle flow change characteristics at the corresponding time granularity, such as hourly segment vehicle flow change characteristics and daily total network vehicle flow change characteristics. After obtaining the energy consumption change trend characteristics and vehicle flow change characteristics at the same time granularity, correlation coefficients are calculated for correlation matching. For example, the correlation coefficient between the hourly lighting system energy consumption change trend characteristics and the corresponding hourly segment vehicle flow change characteristics is calculated. A high correlation coefficient indicates a strong temporal correlation between lighting system energy consumption and vehicle flow during that time period. Based on the correlation matching results, energy consumption-flow correlation curves are established for each time granularity and each infrastructure type. The horizontal axis of the curve represents time, and the vertical axis simultaneously represents the energy consumption change trend value and the vehicle flow change characteristic value. The curve's trend visually demonstrates the correlation between the two over time.

[0022] Step 122: Perform spatial gridding processing on the first energy consumption monitoring data. Divide the target highway network into multiple spatial grid units according to the preset road segment length. Calculate the average energy consumption of infrastructure in each spatial grid unit as the spatial energy consumption distribution feature. Perform road segment matching processing on the vehicle driving trajectory in the second energy consumption monitoring data to determine the proportion of vehicle driving time in each spatial grid unit. Combine the spatial energy consumption distribution feature and the proportion of driving time to establish an energy consumption trajectory correlation matrix in the spatial dimension.

[0023] In this embodiment of the application, the preset road segment length can be set according to the actual situation of the target highway network, for example, it can be set to 1 kilometer. All road segments of the target highway network are divided into multiple spatial grid units according to the length of 1 kilometer. Each spatial grid unit has a unique grid number and a clear spatial boundary.

[0024] For the first energy consumption monitoring data, based on the spatial location information of each data record, it is assigned to the corresponding spatial grid unit. Then, the mean value of the energy consumption data of all infrastructure in each spatial grid unit is calculated. This mean value is the spatial energy consumption distribution characteristic of the spatial grid unit. For example, the sum of the mean energy consumption of the lighting system and the ventilation system in one spatial grid unit is the spatial energy consumption distribution characteristic value of that unit.

[0025] For the vehicle driving trajectories in the second energy consumption monitoring data, road segment matching processing is used to match the vehicle's driving trajectory points with road segments of the target highway network. This determines the vehicle's driving trajectory segment within each spatial grid cell, and then calculates the vehicle's driving time within each spatial grid cell. This driving time is then divided by the vehicle's total driving time within the entire target highway network to obtain the proportion of driving time in each spatial grid cell. For each vehicle's driving trajectory data, a driving time proportion vector corresponding to the number of spatial grid cells can be obtained.

[0026] Then, the spatial energy consumption distribution characteristics of each spatial grid cell are associated with the proportion of driving time of all vehicles corresponding to that cell to construct an energy consumption trajectory association matrix in the spatial dimension. The rows of this matrix represent the spatial grid cell number, the columns represent the vehicle identifier, and the element value in the matrix is ​​the product of the spatial energy consumption distribution characteristic value of the corresponding spatial grid cell and the proportion of driving time of the vehicle in that cell. This matrix can intuitively reflect the correlation strength between the energy consumption distribution and the vehicle driving trajectory at different spatial locations.

[0027] Step 123: By calculating the energy consumption impact weights at different spatiotemporal intersections, the energy consumption flow correlation curve in the time dimension and the energy consumption trajectory correlation matrix in the spatial dimension are spatiotemporally fused to generate a multidimensional energy consumption correlation analysis result containing both temporal dynamic characteristics and spatial distribution characteristics. The multidimensional energy consumption correlation analysis result is used to simultaneously reflect the comprehensive impact of temporal changes and spatial distribution on highway energy consumption.

[0028] In this embodiment, a spatiotemporal intersection point refers to a point with a specific timestamp and a spatial grid cell number. For example, a specific hourly segment and a specific spatial grid cell constitute a spatiotemporal intersection point. First, for each spatiotemporal intersection point, its energy consumption impact weight is calculated. This weight calculation considers the combined effects of temporal dynamic factors and spatial distribution factors. The weight of temporal dynamic factors can be determined based on the fluctuation amplitude of the energy consumption-flow correlation curve in the time dimension. For example, time periods with larger fluctuation amplitudes in the energy consumption-flow correlation curve correspond to higher weights for temporal dynamic factors. The weight of spatial distribution factors can be determined based on the magnitude of the element values ​​in the energy consumption trajectory correlation matrix in the spatial dimension. Spatial grid cells with larger element values ​​correspond to higher weights for spatial distribution factors. The weights of temporal dynamic factors and spatial distribution factors are weighted and averaged to obtain the energy consumption impact weight of each spatiotemporal intersection point.

[0029] Then, using the energy consumption-flow correlation curve in the time dimension as the time axis and the energy consumption trajectory correlation matrix in the spatial dimension as the spatial surface, the energy consumption influence weight is used as the fusion coefficient to perform spatiotemporal fusion processing on the two. Specifically, for each time granularity, the energy consumption-flow correlation curve data at that time granularity is allocated to each spatial grid cell according to the energy consumption influence weight. At the same time, the data in the spatial energy consumption trajectory correlation matrix is ​​combined with the corresponding time data according to the energy consumption influence weight to form a three-dimensional multidimensional energy consumption correlation analysis result data structure. The three dimensions of this data structure are time, space, and energy consumption correlation strength. Through this spatiotemporal fusion processing, the multidimensional energy consumption correlation analysis results can clearly characterize the correlation between infrastructure energy consumption and vehicle driving energy consumption at different times and different spatial locations. For example, it can be seen that during the morning rush hour on weekdays, the energy consumption correlation strength of certain specific spatial grid cells (such as road sections near city entrances and exits) is significantly higher than that of other time periods and other spatial locations, thus achieving the goal of simultaneously reflecting the comprehensive impact of time changes and spatial distribution on highway energy consumption.

[0030] Step 130: Using the energy consumption influencing factors in the multidimensional energy consumption correlation analysis results as model inputs, vehicle travel path planning parameters and infrastructure operation adjustment parameters as decision variables, a collaborative decision model is generated by establishing a mapping relationship between the model inputs and the decision variables to make collaborative optimization decisions on vehicle travel paths and infrastructure operation strategies. The collaborative optimization decision results are then determined by the collaborative decision model for the target highway network under a specified network operation scenario.

[0031] Step 131: Extract the key factors affecting the energy consumption of the highway network from the results of the multidimensional energy consumption correlation analysis as model inputs for the collaborative decision-making model and perform feature standardization processing.

[0032] In this embodiment, the multidimensional energy consumption correlation analysis results contain a large amount of energy consumption-related information, and key factors need to be extracted from them as input to the collaborative decision-making model. These key factors may involve multiple aspects such as time, space, traffic flow, and infrastructure status. Through feature standardization, the differences in dimensions and orders of magnitude between different factors can be eliminated, enabling the model to learn the relationships between factors more accurately.

[0033] Step 1311: Perform factor type identification processing on the multidimensional energy consumption correlation analysis results to distinguish between time dynamic characteristic related factors and spatial distribution characteristic related factors. The time dynamic characteristic related factors are regarded as the first type of influencing factors, and the spatial distribution characteristic related factors are regarded as the second type of influencing factors.

[0034] In this embodiment, the factor type identification process employs a classification method based on feature attributes. First, for each factor in the multidimensional energy consumption correlation analysis results, it is analyzed whether its feature attributes contain time-related descriptions or time-series data. For example, if the value of a factor shows a clear trend over time, it is identified as a time-dynamic characteristic-related factor, i.e., the first type of influencing factor. This type of factor includes hourly variation data of traffic flow, energy consumption data of lighting systems at different times, etc. For factors whose feature attributes mainly describe spatial location, regional distribution, etc., such as the average energy consumption value of different road sections, the proportion of vehicle travel time in a specific spatial grid unit, etc., they are identified as spatial distribution characteristic-related factors, i.e., the second type of influencing factor. Through this factor type identification process, the influencing factors can be clearly divided into two major categories: time and space.

[0035] Step 1312: Perform trend feature extraction processing on the first type of influencing factors, identify factors with periodic fluctuation characteristics and factors with non-periodic fluctuation characteristics, classify factors with periodic fluctuation characteristics into a time periodic factor set, and classify factors with non-periodic fluctuation characteristics into a time random factor set.

[0036] In this embodiment, time series analysis is used to extract trend features from the first category of influencing factors. For each first category of influencing factor, its historical data sequence over a longer time period is obtained, such as hourly data sequences from the past year. Then, periodic analysis is performed on the data sequence, and the autocorrelation function is calculated to detect whether the data sequence exhibits periodic fluctuation characteristics. If the autocorrelation function shows a significant peak at one of the time lag values, and this peak shows a regular repetition, it indicates that the factor has periodic fluctuation characteristics. For example, daily traffic flow data usually peaks during the morning and evening peak hours, showing a daily periodic fluctuation characteristic, and such factors are classified into the set of time-periodic factors. If the autocorrelation function does not show a significant periodic peak, and the fluctuation of the data sequence shows random changes, then the factor is classified into the set of time-random factors, such as abnormal fluctuations in traffic flow caused by sudden traffic accidents. Through this trend feature extraction process, the classification of the first category of influencing factors can be further refined.

[0037] Step 1313: Perform spatial correlation analysis on the second type of influencing factors, calculate the distribution similarity of factors within different spatial grid units, classify factors whose distribution similarity reaches the preset similarity into a spatial clustered factor set, and classify factors whose distribution similarity does not reach the preset similarity into a spatial discrete factor set.

[0038] In this embodiment, when performing spatial correlation analysis on the second type of influencing factors, spatial grid cells are used as the basic analysis unit. For each second type of influencing factor, its values ​​in all spatial grid cells are collected to form a spatial distribution vector. Then, the similarity of the factor's distribution vector between any two spatial grid cells is calculated using a cosine similarity algorithm. The cosine similarity value ranges from -1 to 1, with a higher similarity indicating a higher distribution. The preset similarity can be set according to the actual situation of the target highway network and the analysis requirements, for example, 0.7. For a second type of influencing factor, if the proportion of its distribution similarity with other spatial grid cells reaching the preset similarity exceeds a certain threshold (e.g., 50%), it indicates that the factor exhibits a clustered distribution characteristic in space and is classified as a set of spatially clustered factors. For example, the energy consumption of lighting systems in certain areas is generally high, exhibiting spatial clustering characteristics. If the distribution similarity mostly does not reach the preset similarity, it indicates that the factor is spatially dispersed and is classified as a set of spatially discrete factors. For example, the local energy consumption increase caused by random vehicle breakdowns in the road network.

[0039] Step 1314: Generate an initial set of key factors based on the set of time-periodic factors, the set of time-random factors, the set of spatially clustered factors, and the set of spatially discrete factors.

[0040] In this embodiment, all factors from the time-periodic factor set, the time-random factor set, the spatially clustered factor set, and the spatially discrete factor set are aggregated, and duplicate factors are removed to form an initial key factor set. For example, the time-periodic factor set includes daily traffic flow periodic factors, weekly lighting energy consumption periodic factors, etc.; the time-random factor set includes sudden weather impact factors, temporary traffic control factors, etc.; the spatially clustered factor set includes energy consumption clustering factors in tunnel areas, energy consumption clustering factors around service areas, etc.; and the spatially discrete factor set includes random vehicle failure energy consumption factors, energy consumption factors for maintenance of individual road sections, etc. These factors from different sets are integrated to form the initial key factor set, which covers various factors that have a significant impact on the energy consumption of the highway network identified from both time and spatial dimensions.

[0041] Step 1315: Map the feature values ​​of each factor in the initial set of key factors to a preset standard value range according to the data type of the factor, and generate model input items that match the input requirements of the collaborative decision-making model.

[0042] In this embodiment, the factors in the initial set of key factors have different data types, such as continuous data (e.g., energy consumption values, traffic flow, etc.), discrete data (e.g., road segment numbers, weather types, etc.), and Boolean data (e.g., equipment on / off status, etc.). For continuous data, the min-max standardization method is used to map its feature values ​​to the standard value range [0, 1]. The specific calculation method is as follows: for the feature value x of one of the continuous factors, its standardized value x' = (xx... min ) / (x max -x min ), where x min x is the minimum value of the characteristic value of this factor. max This represents the maximum value of the characteristic value of this factor.

[0043] For discrete data, a one-hot encoding method is used to convert it into multiple binary feature vectors, each corresponding to a discrete value. For example, if a road segment number has three values: A, B, and C, it is converted into three binary features. When the road segment is A, the first feature value is 1, and the rest are 0; when the road segment is B, the second feature value is 1, and the rest are 0, and so on. For Boolean data, it is directly mapped to 0 or 1. For example, the on state of a device is mapped to 1, and the off state is mapped to 0. Through this feature standardization process, the feature values ​​of each factor in the initial set of key factors are uniformly converted into a format and value range that meet the input requirements of the collaborative decision-making model, generating model input items and ensuring that the model can uniformly process and analyze different types of factors.

[0044] Step 132: Normalize the total driving energy consumption of vehicles within the target highway network and the total operating energy consumption of the network infrastructure. Calculate the normalization coefficient based on historical energy consumption data or a preset energy consumption benchmark value to obtain the normalized vehicle driving energy consumption and the normalized infrastructure operating energy consumption. Determine the bi-objective optimization function of the collaborative decision-making model and transform the bi-objective optimization function into a single-objective comprehensive optimization function by setting objective weight coefficients. The first optimization objective is to minimize the normalized vehicle driving energy consumption, and the second optimization objective is to minimize the normalized infrastructure operating energy consumption.

[0045] In this embodiment, the total vehicle driving energy consumption data and total infrastructure operating energy consumption data of the target highway network over a past period (e.g., the past year) are first collected as historical energy consumption data, or a preset energy consumption benchmark value is set according to the road network planning and design standards. For the total driving energy consumption of vehicles within the target highway network, its normalization coefficient is calculated. The normalization coefficient can be the maximum total vehicle driving energy consumption value in the historical energy consumption data or the preset vehicle energy consumption benchmark value. The current total vehicle driving energy consumption is divided by the normalization coefficient to obtain the normalized vehicle driving energy consumption, which ranges from 0 to 1.

[0046] Similarly, for the total operating energy consumption of road network infrastructure, the same method is used to calculate the normalization coefficient (which can be the historical maximum infrastructure energy consumption value or a preset infrastructure energy consumption benchmark value). The current total operating energy consumption of infrastructure is divided by this normalization coefficient to obtain the normalized infrastructure operating energy consumption. Then, the bi-objective optimization functions of the collaborative decision-making model are determined: the first optimization objective function f1 = normalized vehicle driving energy consumption, and the second optimization objective function f2 = normalized infrastructure operating energy consumption. To transform the bi-objective optimization problem into a single-objective optimization problem, objective weight coefficients α and β are set, where α + β = 1. The values ​​of α and β are determined according to the energy consumption optimization priority of the target highway network. For example, when more emphasis is placed on reducing vehicle driving energy consumption, α is taken as a larger value (e.g., α = 0.6, β = 0.4), and vice versa. The single-objective comprehensive optimization function F = α * f1 + β * f2 combines the two optimization objectives into one objective for subsequent optimization solutions.

[0047] Step 133: Combine the road segment selection variable, lane selection variable, and driving speed variable in the vehicle driving path planning parameters with the lighting system switch control variable, ventilation system power adjustment variable, and traffic signal cycle adjustment variable in the infrastructure operation adjustment parameters as decision variables for the collaborative decision-making model.

[0048] In this embodiment, the vehicle travel path planning parameters are key parameters affecting vehicle energy consumption. Among them, the road segment selection variable represents the sequence of highway road segments selected by the vehicle from the starting point to the destination. For example, in the target highway network, from the starting point S to the destination T, the vehicle can choose road segment A to road segment B to road segment C, or it can choose road segment D to road segment E. The value of the road segment selection variable is these possible road segment sequences. The lane selection variable represents the lane selected by the vehicle when traveling on the selected road segment, such as the inner lane, middle lane or outer lane of the highway. The value of the lane selection variable is the lane number of the corresponding road segment. The driving speed variable represents the driving speed of the vehicle on the selected road segment and lane. Its value range is limited by the speed limit standard and traffic flow conditions of the road segment. Infrastructure operation and regulation parameters are key parameters affecting infrastructure energy consumption. Among them, the lighting system switch control variable represents the on / off state of the lighting system; for different road sections and time periods, this variable can be either on or off. The ventilation system power regulation variable represents the power level of the ventilation system during operation, such as low, medium, or high power; different power levels correspond to different energy consumption levels. The traffic signal cycle regulation variable represents the signal cycle duration of the traffic signal control system; its value can be adjusted within a certain range to adapt to different traffic flow conditions. These variables are collectively used as decision variables in a collaborative decision-making model, which will adjust the values ​​of these variables to achieve the goal of energy consumption optimization.

[0049] Step 134: By analyzing the sensitivity of each decision variable to the energy consumption of the highway network under different energy consumption influencing factors, establish a mapping relationship model between energy consumption influencing factors and decision variables, generate a mapping function that reflects the energy consumption transmission mechanism, and use the mapping function to transform the feature values ​​of the model input terms into the optimal value range of the decision variables.

[0050] Step 1341: Select any energy consumption influencing factor in the model input items as the current analysis factor, keep the feature values ​​of energy consumption influencing factors other than the current analysis factor unchanged, and adjust the feature values ​​of the current analysis factor to different levels in turn.

[0051] In this embodiment, the model input includes multiple energy consumption influencing factors, such as time-periodic factors, time-random factors, spatially clustered factors, and spatially discrete factors. One factor is selected as the current analysis factor, for example, the daily traffic flow periodic factor from the time-periodic factor set. Then, the characteristic values ​​of all other energy consumption influencing factors are fixed to their current values; for example, the weather factor is fixed to sunny days, and the road maintenance factor is fixed to a no-maintenance state. Next, the characteristic value of the daily traffic flow periodic factor is sequentially adjusted to different levels, for example, from a low flow level (e.g., 500 vehicles per hour) to a medium flow level (e.g., 1000 vehicles per hour), a high flow level (e.g., 1500 vehicles per hour), etc., with each level at equal intervals, covering the possible value range of the factor.

[0052] Step 1342: At each eigenvalue level, adjust the values ​​of each decision variable, record the change in energy consumption of the highway network when the value of each decision variable changes, and calculate the ratio of the change in energy consumption to the change in the value of the decision variable as the sensitivity coefficient at the current eigenvalue level.

[0053] In this embodiment, at one of the characteristic value levels of the daily traffic flow cycle factor (such as the medium flow level), for each decision variable, such as the road segment selection variable, different road segment sequences are selected sequentially as its values. For example, first, road segment A to road segment B to road segment C are selected, and the total energy consumption of the highway network at this time is recorded; then, road segment D to road segment E are selected, and the total energy consumption of the network is recorded again. The difference between the two total energy consumptions is calculated as the energy consumption change. At the same time, the difference between the two road segment selection variable values ​​is calculated (such as using the difference in road segment length or travel time to represent the change in the decision variable value). The energy consumption change is divided by the change in the decision variable value to obtain the sensitivity coefficient of the road segment selection variable at this characteristic value level. Using the same method, the values ​​of decision variables such as lane selection (e.g., from the inner lane to the outer lane), driving speed (e.g., from 80 km / h to 100 km / h), lighting system switch control (e.g., from on to off), ventilation system power adjustment (e.g., from low power to high power), and traffic signal cycle adjustment (e.g., from 60 seconds to 90 seconds) are adjusted respectively, and the sensitivity coefficient of each decision variable at the corresponding eigenvalue level is calculated.

[0054] Step 1343: Perform trend fitting on the sensitivity coefficients at all eigenvalue levels to generate sensitivity change curves between the current analysis factor and each decision variable.

[0055] In this embodiment, for the current analysis factor (daily traffic flow cycle factor) and one of the decision variables (such as driving speed variable), the sensitivity coefficients calculated at different eigenvalue levels are used as data points. A scatter plot is drawn with the eigenvalue level as the horizontal axis and the sensitivity coefficient as the vertical axis. Then, the least squares method is used to perform trend fitting on these scatter points, and an appropriate fitting function (such as a linear function, quadratic function, or exponential function) is selected to describe the changing trend of the sensitivity coefficient with the eigenvalue level, resulting in a sensitivity change curve between the current analysis factor and the decision variable. For example, the sensitivity coefficient of the driving speed variable may show a quadratic function curve that first increases and then decreases with the increase of daily traffic flow. The above trend fitting process is performed on each decision variable to generate sensitivity change curves between the current analysis factor and each decision variable.

[0056] Step 1344: Replace the next energy consumption influencing factor as the current analysis factor, and repeat the steps of adjusting the eigenvalue level, calculating the sensitivity coefficient, and generating the sensitivity change curve until all energy consumption influencing factors have been analyzed.

[0057] In this embodiment, after analyzing the cyclical factors of daily traffic flow, the next energy consumption influencing factor, such as the tunnel area energy consumption clustering factor in the spatial clustering factor set, is selected as the current analysis factor. Keeping the characteristic values ​​of other energy consumption influencing factors unchanged, the characteristic values ​​of the tunnel area energy consumption clustering factor are sequentially adjusted to different levels (e.g., low clustering level, medium clustering level, high clustering level). At each level, the values ​​of each decision variable are adjusted, the sensitivity coefficient is calculated, and a trend fitting is performed on the sensitivity coefficient to generate a sensitivity change curve. In this manner, all energy consumption influencing factors in the model input are analyzed sequentially, generating a sensitivity change curve between each energy consumption influencing factor and each decision variable.

[0058] Step 1345: Based on the sensitivity change curves of all energy consumption influencing factors and decision variables, construct a mapping relationship model between energy consumption influencing factors and decision variables. In the mapping relationship model, each decision variable corresponds to a set of functions composed of sensitivity change curves of different energy consumption influencing factors.

[0059] In this embodiment, each decision variable (such as lane selection) corresponds to multiple sensitivity change curves, and each sensitivity change curve corresponds to an energy consumption influencing factor (such as daily traffic flow cycle factor, tunnel area energy consumption aggregation factor, etc.). These sensitivity change curves are treated as functions and combined into a function set, which describes the mapping relationship between the energy consumption influencing factor and the decision variable. For example, the function set for the lane selection variable includes the daily traffic flow cycle factor-lane selection sensitivity curve function, the tunnel area energy consumption aggregation factor-lane selection sensitivity curve function, etc. For all decision variables, the above function set is constructed to form a complete mapping relationship model between energy consumption influencing factors and decision variables. This model can reflect how different energy consumption influencing factors affect the values ​​of each decision variable through sensitivity changes.

[0060] Step 1346: Extract the sensitivity threshold range of each decision variable from the function set, take the energy consumption influencing factors whose sensitivity reaches the sensitivity threshold as the main influencing factors, and take the factors whose sensitivity does not reach the sensitivity threshold as the secondary influencing factors, and generate the set of main influencing factors of the decision variable.

[0061] In this embodiment, for each sensitivity change curve function in the function set of each decision variable, the distribution range of its function value (sensitivity coefficient) is analyzed, and a sensitivity threshold is set, for example, the average value of the sensitivity coefficient plus twice the standard deviation is used as the sensitivity threshold. For one of the sensitivity change curve functions, when the function value (sensitivity coefficient) reaches or exceeds the sensitivity threshold within a certain characteristic value interval, that interval is the sensitivity threshold interval of the energy consumption influencing factor for the decision variable. The energy consumption influencing factors whose corresponding sensitivity change curve functions have a sensitivity threshold interval are regarded as the main influencing factors of the decision variable, while the energy consumption influencing factors whose sensitivity coefficients never reach the sensitivity threshold are regarded as secondary influencing factors. For example, for the driving speed variable, in its function set, the sensitivity change curve of the daily traffic flow cycle factor exceeds the sensitivity threshold in the medium-high flow level interval, and the sensitivity change curve of the tunnel area energy consumption aggregation factor exceeds the sensitivity threshold in the high aggregation level interval. Then, the daily traffic flow cycle factor and the tunnel area energy consumption aggregation factor are regarded as the main influencing factors of the driving speed variable, generating the main influencing factor set of the driving speed variable. Perform the above processing on each decision variable to generate its own set of main influencing factors.

[0062] Step 1347: Based on the set of main influencing factors and the mapping relationship model, generate a mapping function that reflects the energy consumption transmission mechanism. The mapping function takes the characteristic values ​​of the main influencing factors as input parameters and the range of values ​​of the decision variables as output results.

[0063] In this embodiment, for each decision variable, a mapping function is constructed based on the sensitivity change curve function in its set of main influencing factors and the mapping relationship model. The input parameters of the mapping function are the current characteristic values ​​of each factor in the set of main influencing factors, and the output result is the value range of the decision variable. For example, for the driving speed variable, its main influencing factors are the daily traffic flow cycle factor (characteristic value is the current hourly traffic flow level) and the tunnel area energy consumption aggregation factor (characteristic value is the current tunnel area energy consumption aggregation degree). According to the sensitivity change curve functions of these two main influencing factors, when the input current hourly traffic flow level is medium flow and the current tunnel area energy consumption aggregation degree is high aggregation, the reasonable value range of the driving speed variable is determined by finding the corresponding sensitivity threshold interval in the sensitivity change curve function, for example, 80-90 km / h. This value range is the output result of the mapping function. In this way, a mapping function reflecting the energy consumption transmission mechanism is generated for each decision variable.

[0064] Step 1348: Substitute the feature values ​​of the model input terms into the mapping function to calculate the optimal value range of each decision variable under the current feature value. The optimal value range is the feasible interval of the decision variable under the constraint of the energy consumption transmission mechanism.

[0065] In this embodiment, the model input includes the current feature values ​​of all energy consumption influencing factors. Feature values ​​belonging to the main influencing factor set of each decision variable are extracted and substituted into the mapping function of the corresponding decision variable. For example, the current feature value of the daily traffic flow cycle factor (e.g., current hourly traffic flow of 1200 vehicles / hour) and the current feature value of the tunnel area energy consumption aggregation factor (e.g., current tunnel area energy consumption aggregation level is medium to high) are substituted into the mapping function of the driving speed variable, and the optimal range of the driving speed variable is calculated to be 85-95 km / h. Using the same method, the corresponding main influencing factor feature values ​​are substituted into the mapping function of each decision variable to calculate the optimal range of all decision variables under the current feature values. These ranges represent the feasible intervals in which the decision variables can take values ​​considering the energy consumption transmission mechanism constraint.

[0066] Step 135: Combining the optimized value range and the specified road network operation scenario of the target highway network, the single-objective comprehensive optimization function is solved using an optimization solution method under multiple constraints. Under the premise of satisfying vehicle travel time constraints, infrastructure operation safety constraints, and traffic flow control constraints, the combination of decision variables that minimizes the single-objective comprehensive optimization function is found, and this combination of decision variables is used as the collaborative optimization decision result of the collaborative decision model.

[0067] In this embodiment, the specified road network operation scenario is a specific scenario determined based on the actual operation of the target highway network, such as "traffic congestion at city entrances and exits during weekday morning rush hour". First, the multiple constraints under this scenario are clarified: vehicle travel time constraints mean that the total travel time from the origin to the destination must not exceed the preset maximum allowable travel time; for example, during the morning rush hour, the maximum allowable travel time from city A to city B is 2 hours. Infrastructure operation safety constraints include the requirement that lighting systems must be turned on at night, ventilation systems in tunnels must not be lower than safety standards, and traffic signal switching intervals must not be less than safe times. Traffic flow control constraints mean that the vehicle flow on a road segment must not exceed the maximum capacity of that segment; for example, the maximum capacity of a certain road segment is 2000 vehicles per hour. Then, the optimal value range of each decision variable is used as the boundary condition of the variable, and the single-objective comprehensive optimization function F is used as the objective function. The particle swarm optimization algorithm is used to solve this single-objective optimization problem. The initial particle swarm of the particle swarm optimization algorithm consists of combinations of decision variables randomly generated within the optimal value range of each decision variable, with each particle representing a possible combination of decision variables. During the algorithm iteration process, the position and velocity of the particles are adjusted based on their fitness values ​​(i.e., the value of the single-objective comprehensive optimization function F), continuously searching for better combinations of decision variables. In each iteration, the generated combinations of decision variables also need to be checked for constraints to ensure they meet constraints on vehicle travel time, infrastructure operation safety, and traffic flow control. When the algorithm reaches the preset maximum number of iterations or the fitness value converges, the decision variable combination corresponding to the particle with the smallest fitness value is output. This combination of decision variables represents the collaborative optimization decision result of the collaborative decision-making model under the specified road network operation scenario.

[0068] Step 140: Generate a graphic analysis report containing multi-dimensional energy consumption state characteristics and collaborative optimization suggestions based on the collaborative optimization decision results.

[0069] In this embodiment of the application, the multidimensional energy consumption state features are obtained by extracting and merging features from different types of decision results in the collaborative optimization decision results, and the graphic analysis report is obtained by integrating and processing according to a preset report structure.

[0070] Step 141: Perform decision variable analysis processing on the collaborative optimization decision results, extract the vehicle driving path planning parameter optimization results and the infrastructure operation adjustment parameter optimization results, take the vehicle driving path planning parameter optimization results as the first type of decision results, and take the infrastructure operation adjustment parameter optimization results as the second type of decision results.

[0071] In this embodiment, the collaborative optimization decision result is a combination of optimized values ​​for all decision variables. Decision variable analysis involves classifying and extracting these variables. First, the vehicle path planning parameters in the collaborative optimization decision result are identified, such as the optimized values ​​of the road segment selection variable (selected road segment sequence), the optimized values ​​of the lane selection variable (selected lane number), and the optimized values ​​of the driving speed variable (selected driving speed range). The optimization results of these parameters are extracted and integrated into the vehicle path planning parameter optimization result, which is the first type of decision result. Then, the infrastructure operation adjustment parameters in the collaborative optimization decision result are identified, such as the optimized values ​​of the lighting system switch control variable (on or off state), the optimized values ​​of the ventilation system power adjustment variable (selected power level), and the optimized values ​​of the traffic signal cycle adjustment variable (selected signal cycle duration). The optimization results of these parameters are extracted and integrated into the infrastructure operation adjustment parameter optimization result, which is the second type of decision result.

[0072] Step 142: Extract energy consumption impact features from the first type of decision results, analyze the energy consumption change features under different combinations of road segment selection variables, lane selection variables and driving speed variables, generate the first multidimensional energy consumption state features related to vehicle driving, and use the first multidimensional energy consumption state features as the vehicle energy consumption feature set.

[0073] Step 1421: Extract all possible combinations of road segment selection variables, lane selection variables, and driving speed variables from the first type of decision results, and treat each combination as a decision combination unit.

[0074] In this embodiment, the road segment selection variable in the first type of decision result may contain multiple optional road segment sequences, the lane selection variable may contain multiple optional lane numbers, and the driving speed variable may contain multiple optional speed ranges. All possible combinations of road segment selection variables, lane selection variables, and driving speed variables are generated using an exhaustive method or orthogonal experimental design. For example, if the road segment selection variable has two possible road segment sequences (S1 and S2), the lane selection variable has three possible lane numbers (L1, L2, and L3), and the driving speed variable has two possible speed ranges (V1 and V2), then there are 2 × 3 × 2 = 12 potential combinations. Each combination (such as S1+L1+V1, S1+L1+V2, S1+L2+V1, etc.) is treated as an independent decision combination unit, and each decision combination unit represents a specific vehicle driving scheme.

[0075] Step 1422: Perform energy consumption simulation analysis on each decision combination unit, simulate the vehicle's driving process under the road segment, lane and speed conditions corresponding to the decision combination unit, record the energy consumption change data during the driving process, and extract the peak energy consumption value, average energy consumption value and energy consumption fluctuation amplitude value from the energy consumption change data.

[0076] In this embodiment, for each decision-making unit, energy consumption simulation analysis is performed using a vehicle dynamics model and an energy consumption calculation model. First, the vehicle's route is determined based on the road segment selection variable in the decision-making unit, and detailed terrain data (such as slope and curvature), road surface condition data (such as friction coefficient), and traffic condition data (such as traffic flow and congestion level) are obtained. The vehicle's lane on each road segment is determined based on the lane selection variable, and the lane width and speed limit information are obtained. The target driving speed curve for the vehicle on different road segments and lanes is determined based on the driving speed variable. Then, these data are input into the vehicle dynamics model to simulate the vehicle's acceleration, deceleration, and constant speed driving processes, calculating the instantaneous energy consumption of the vehicle during driving. Energy consumption change data throughout the entire driving process is recorded, resulting in a curve showing energy consumption changes over time. From this curve, peak energy consumption (the maximum value in the curve), average energy consumption (the arithmetic mean of the curve), and energy consumption fluctuation amplitude (the difference between the maximum and minimum values ​​in the curve) are extracted. These values ​​reflect the energy consumption characteristics of the vehicle under this decision-making unit from different perspectives.

[0077] Step 1423: Use the peak energy consumption value, average energy consumption value and energy consumption fluctuation amplitude value as basic feature parameters, and normalize the basic feature parameters to generate standardized feature parameters.

[0078] In this embodiment, the basic characteristic parameters (peak energy consumption, average energy consumption, and energy consumption fluctuation amplitude) have different dimensions and orders of magnitude. To facilitate subsequent feature fusion and analysis, they need to be normalized. For the peak energy consumption, the peak energy consumption values ​​of all decision-making combination units are collected, the maximum and minimum values ​​are found, and the min-max normalization method is used to map the peak energy consumption value of each decision-making combination unit to the interval [0, 1] to obtain the standardized peak energy consumption parameter. The same method is used to normalize the average energy consumption value and the energy consumption fluctuation amplitude value to obtain the standardized average energy consumption parameter and the standardized energy consumption fluctuation amplitude parameter, respectively. For example, if the peak energy consumption value of one decision-making combination unit is 10 units, the maximum peak energy consumption value of all decision-making combination units is 15 units, and the minimum peak energy consumption value is 5 units, then its standardized peak energy consumption parameter is (10-5) / (15-5) = 0.5. Through this normalization process, the basic characteristic parameters are converted into standardized characteristic parameters with uniform dimensions and value ranges.

[0079] Step 1424: Perform feature fusion processing on the standardized feature parameters, combine the peak energy consumption standardized parameters, average energy consumption standardized parameters and energy consumption fluctuation amplitude standardized parameters of the same decision combination unit into a feature vector, and use the feature vector as the energy consumption state feature vector corresponding to the decision combination unit.

[0080] In this embodiment, the feature fusion processing employs a simple vector concatenation method. For each decision-making unit, its standardized peak energy consumption parameter, standardized average energy consumption parameter, and standardized energy consumption fluctuation amplitude parameter are arranged in a fixed order (e.g., peak, average, fluctuation amplitude) to form a three-dimensional feature vector. For example, if the standardized peak energy consumption parameter of a decision-making unit is 0.5, the standardized average energy consumption parameter is 0.4, and the standardized energy consumption fluctuation amplitude parameter is 0.6, then its corresponding energy consumption state feature vector is [0.5, 0.4, 0.6]. This feature vector integrates the standardized features of the decision-making unit in the three dimensions of peak energy consumption, average energy consumption, and energy consumption fluctuation amplitude, and can comprehensively represent the energy consumption state of the decision-making unit.

[0081] Step 1425: Perform cluster analysis on the energy consumption state feature vectors of all decision-making combination units, classify decision-making combination units with feature vector similarity values ​​higher than the preset similarity value into the same feature category, and calculate the central feature vector of each feature category as the representative energy consumption state feature of that category.

[0082] In this embodiment, the K-means clustering algorithm is used to perform cluster analysis on the energy consumption state feature vectors of all decision-making units. First, based on the number of decision-making units and the distribution of feature vectors, the number of clusters K is preset, for example, K=5. Then, K initial cluster center vectors are randomly selected, and the Euclidean distance between each energy consumption state feature vector and each initial cluster center vector is calculated as a similarity value. Each energy consumption state feature vector is assigned to the cluster containing the cluster center with the highest similarity value (closest distance). Next, the center feature vector of each cluster (the average of all feature vectors in that cluster) is recalculated, and the above assignment and center update process is repeated until the cluster center vectors no longer change significantly or the preset number of iterations is reached. The preset similarity value can be set as one of the thresholds of the Euclidean distance; when the Euclidean distance between two feature vectors is less than this threshold, their similarity value is considered to be higher than the preset similarity value. After the clustering analysis is completed, the decision-making combination units in each category have similar energy consumption state feature vectors. The central feature vector of each category (the arithmetic mean of all energy consumption state feature vectors in that category) is calculated. This central feature vector is used as the representative energy consumption state feature of that category, which can reflect the typical energy consumption state of all decision-making combination units in that category.

[0083] Step 1426: Integrate the representative performance consumption state features of all categories into a set and use it as the vehicle energy consumption feature set. Each feature in the vehicle energy consumption feature set corresponds to a typical energy consumption state of a decision combination unit.

[0084] In this embodiment, after completing cluster analysis and obtaining representative performance consumption state features for each category, these representative performance consumption state features are arranged in category order and integrated into a set, which is the vehicle energy consumption feature set. For example, if clustering yields 5 categories, each category corresponding to a representative performance consumption state feature vector, then the vehicle energy consumption feature set contains 5 of the aforementioned feature vectors, each feature vector corresponding to a typical energy consumption state of a decision-making combination unit. Through the vehicle energy consumption feature set, the typical energy consumption states corresponding to different types of vehicle driving schemes can be clearly understood.

[0085] Step 143: Extract energy consumption impact features from the second type of decision results, analyze the energy consumption change features under different combinations of lighting system switch control variables, ventilation system power adjustment variables and traffic signal cycle adjustment variables, generate second multidimensional energy consumption state features related to infrastructure operation, and use the second multidimensional energy consumption state features as the facility energy consumption feature set.

[0086] In this embodiment, the second type of decision result includes optimized values ​​for infrastructure operation adjustment parameters, such as the on / off state of the lighting system switch control variable, the low / medium / high power level of the ventilation system power adjustment variable, and the different cycle durations of the traffic signal cycle adjustment variable. When extracting energy consumption impact features from the second type of decision result, different combinations of these variables also need to be considered. For example, combinations of switch control for the lighting system in different road sections and time periods, combinations of power adjustment for the ventilation system in different tunnels and time periods, and combinations of cycle adjustment for traffic signals at different intersections. For each combination scheme, the corresponding energy consumption value is calculated using an infrastructure energy consumption simulation model. The characteristics of energy consumption changes with lighting switch status, ventilation power level, and signal cycle duration are analyzed, such as energy consumption comparisons of different lighting switch combinations, energy consumption differences of different ventilation power levels, and the pattern of energy consumption changes with the signal cycle. Based on these energy consumption change characteristics, multiple dimensional feature indicators are extracted, such as lighting energy consumption ratio, ventilation power energy consumption coefficient, and signal cycle energy consumption sensitivity. These feature indicators are combined to generate a second multi-dimensional energy consumption state feature, i.e., a set of facility energy consumption features.

[0087] Step 144: Merge the vehicle energy consumption feature set and the facility energy consumption feature set to generate a comprehensive energy consumption status feature set. Perform hierarchical division on the features in the comprehensive energy consumption status feature set, and use the core hierarchical features as the key content to be displayed in the graphic analysis report, and use the non-core hierarchical features as supplementary content.

[0088] In this embodiment, the vehicle energy consumption feature set includes multi-dimensional energy consumption state features related to vehicle operation, and the facility energy consumption feature set includes multi-dimensional energy consumption state features related to infrastructure operation. The feature indicators in these two sets are merged, and duplicate feature indicators are removed to obtain a comprehensive energy consumption state feature set. This set covers all aspects of the target highway network's energy consumption. When performing hierarchical classification of the features in the comprehensive energy consumption state feature set, the analytic hierarchy process (AHP) is used. Based on factors such as the degree of influence of each feature indicator on total energy consumption, data availability, and importance, the feature indicators are divided into core, important, and general levels. Core level features are those that have the greatest impact on total energy consumption and best reflect the energy consumption optimization effect, such as the reduction rate of total vehicle driving energy consumption and the reduction rate of total infrastructure operating energy consumption. Important level features are those that play an important auxiliary role in energy consumption analysis, such as the road segment energy consumption unevenness coefficient and the proportion of lighting system energy consumption. General level features are secondary, supplementary feature indicators. The core hierarchical features will be highlighted in the graphic analysis report and presented in a prominent chart format; important and general hierarchical features will be presented as supplementary content in the form of appendices or detailed data tables.

[0089] Step 145: Generate path optimization suggestions based on the optimization results of road segment selection variables in the vehicle energy consumption feature set, generate driving strategy optimization suggestions by combining the optimization results of lane selection variables and driving speed variables, and generate facility control optimization suggestions by using the optimization results of infrastructure operation adjustment parameters.

[0090] Step 1451: Extract the optimization results of road segment selection variables from the vehicle energy consumption feature set, analyze the energy consumption change ratio of different road segments before and after optimization, and take the road segments with energy consumption change ratio higher than the preset ratio value as key optimization road segments.

[0091] In this embodiment, the vehicle energy consumption feature set includes energy consumption state features under different combinations of road segment selection variables. The optimization results of the road segment selection variables are extracted from this set, resulting in the optimized road segment sequence. The energy consumption data of each road segment in the optimized sequence is compared with the energy consumption data of the corresponding road segment before optimization (e.g., the original route planning), and the energy consumption change ratio is calculated as: Energy consumption change ratio = (Energy consumption before optimization - Energy consumption after optimization) / Energy consumption before optimization × 100%. A preset ratio value can be set according to the energy consumption optimization target, for example, 10%. Road segments with an energy consumption change ratio higher than 10% are identified as key optimization road segments. These road segments show significant energy consumption reduction after optimization and are the focus of route optimization. For example, if road segment D has an energy consumption of 100 units before optimization and 85 units after optimization, with an energy consumption change ratio of 15%, which is higher than the preset ratio value of 10%, then road segment D is identified as a key optimization road segment.

[0092] Step 1452: Perform path correlation analysis on the key optimized road segment, identify the upstream and downstream road segments directly connected to the key optimized road segment, construct a path network including the key optimized road segment and its upstream and downstream road segments, find the path combination with the minimum total energy consumption in the path network, and generate path optimization suggestions. The path optimization suggestions include a recommended driving road segment sequence and a recommended driving direction for each road segment.

[0093] In this embodiment, when performing path correlation analysis on key optimized road segments, the upstream road segments (segments entering the key optimized road segment) and downstream road segments (segments exiting the key optimized road segment) directly connected to the key optimized road segment are identified based on the topology data of the target highway network. For example, if the key optimized road segment is segment D, its upstream road segments may be segment A and segment B, and its downstream road segments may be segment E and segment F. A path network is constructed that includes the key optimized road segment D and its upstream and downstream road segments A, B, E, and F. Nodes in this network are connection points of road segments, edges are road segments, and the optimized energy consumption value of each road segment is labeled on the edges. Then, in this path network, using the main starting and ending points of the target highway network as endpoints, Dijkstra's algorithm is used to find the path combination with the minimum total energy consumption from the starting point to the ending point. For example, given a starting point S and an ending point T, a path is found in the path network that leads from S to segment A, then to segment D, then to segment E, and finally to T. The total energy consumption of this path is the sum of the energy consumption of each segment, and it is the minimum among all possible paths. This path combination is used as a path optimization suggestion, which includes a recommended sequence of driving segments (S to segment A, then to segment D, then to segment E, and finally to T) and the recommended driving direction for each segment (e.g., the recommended driving direction for segment A is from east to west, and the recommended driving direction for segment D is from north to south, etc.).

[0094] Step 1453: Extract the optimization results of lane selection variable and driving speed variable from the vehicle energy consumption feature set, analyze the energy saving index under different lane and speed combinations, and take the lane and speed combination with the largest weight value corresponding to the energy saving index as the optimal driving combination.

[0095] In this embodiment, the vehicle energy consumption feature set includes energy consumption state features under different combinations of lane selection and driving speed variables, from which the optimization results of these two variables are extracted. The energy saving index refers to the amount of energy saved by using a certain lane and speed combination compared to a baseline combination (such as the default lane and average speed). Energy saving index = baseline combination energy consumption - current combination energy consumption. To comprehensively evaluate the energy-saving effect of different combinations, weight values ​​are assigned to the energy saving index. The magnitude of the weight value is determined based on the applicability frequency and energy-saving potential of the combination in the target highway network. For example, in road sections with high traffic volume, a certain lane and speed combination has a high applicability frequency and great energy-saving potential, thus its weight value is larger. The product of the energy saving index and the corresponding weight value for each lane and speed combination is calculated, and the combination with the largest product is determined as the optimal driving combination. For example, the combination of selecting the inner lane and driving at 85 km / h has an energy saving index of 15 units, a weight value of 0.8, and a product of 12; the combination of selecting the middle lane and driving at 90 km / h has an energy saving index of 12 units, a weight value of 0.7, and a product of 8.4. Therefore, the combination of the inner lane and 85 km / h is considered the optimal driving combination.

[0096] Step 1454: Perform scenario adaptability analysis on the optimal driving combination based on different traffic flow conditions to generate differentiated driving strategies under different traffic flow scenarios. Integrate the differentiated driving strategies into driving strategy optimization suggestions, which include recommended lanes and recommended driving speed ranges under different scenarios.

[0097] In this embodiment, different traffic flow conditions include low-flow scenarios (e.g., less than 500 vehicles per hour), medium-flow scenarios (e.g., 500-1500 vehicles per hour), and high-flow scenarios (e.g., more than 1500 vehicles per hour). The optimal driving combination (inner lane, 85 km / h) is analyzed for its adaptability under different traffic flow conditions. A traffic flow simulation model is used to simulate the driving effect and energy consumption of this combination under different flow scenarios. In low-flow scenarios, traffic congestion is less, and vehicles can maintain a higher speed. In this case, the recommended speed range can be appropriately increased, such as 85-95 km / h, and the inner lane is still recommended. In medium-flow scenarios, traffic flow is relatively stable, maintaining the recommended lane and speed range of the optimal driving combination, i.e., the inner lane, 80-90 km / h. In high-flow scenarios, the possibility of traffic congestion increases. To avoid frequent acceleration and deceleration leading to increased energy consumption, it is recommended to choose the middle lane, and the speed range should be adjusted to 75-85 km / h. These differentiated driving strategies under different traffic scenarios are integrated into driving strategy optimization suggestions, which clearly list the recommended lanes and recommended driving speed ranges for low, medium and high traffic scenarios.

[0098] Step 1455: Extract the optimization results of infrastructure operation adjustment parameters from the second type of decision results and determine the optimization parameters of the lighting system, ventilation system and traffic signal system. Use the optimization parameters of the lighting system as lighting control parameters, the optimization parameters of the ventilation system as ventilation control parameters, and the optimization parameters of the traffic signal system as signal control parameters.

[0099] In this embodiment, the second type of decision result includes optimized values ​​of infrastructure operation adjustment parameters, from which parameters related to the lighting system, ventilation system, and traffic signal system are extracted. For the lighting system, the optimized parameters may include the on / off status of different road segments and time periods, such as turning off 50% of the lighting fixtures on road segment A from 22:00 to 5:00 the next day. These parameters are determined as lighting control parameters. For the ventilation system, the optimized parameters may include the power levels of different tunnels and time periods, such as using a medium power level in tunnel B during peak traffic hours and a low power level during off-peak hours. These parameters are determined as ventilation control parameters. For the traffic signal system, the optimized parameters may include the signal cycle duration at different intersections and time periods, such as a 90-second signal cycle at intersection C during morning peak hours and a 60-second signal cycle during off-peak hours. These parameters are determined as signal control parameters.

[0100] Step 1456: Perform time-period adaptive analysis on the lighting control parameters, and generate time-period lighting system switch control suggestions based on the natural light intensity characteristics of different time periods; perform spatial adaptive analysis on the ventilation control parameters, and generate time-period ventilation system power adjustment suggestions based on the tunnel length characteristics of different road sections; perform traffic flow adaptive analysis on the signal control parameters, and generate time-period traffic signal cycle adjustment suggestions based on the traffic flow characteristics of different time periods.

[0101] In this embodiment, when performing time-period adaptive analysis on lighting control parameters, historical natural light intensity data of the target highway network area is collected. The day is divided into different time periods, such as early morning, morning, noon, afternoon, evening, and night, and the natural light intensity characteristics of each time period are analyzed. For example, natural light intensity gradually increases from 5:00 to 7:00 AM, is strongest from 12:00 to 2:00 PM, and is weakest from 8:00 PM to 4:00 AM the next day. Based on the lighting control parameters, time-period lighting system switching control suggestions are generated, such as turning on all lighting fixtures from 4:00 to 5:00 AM, gradually turning off some lighting fixtures from 5:00 to 7:00 AM, and turning off most lighting fixtures from 12:00 to 2:00 PM. When performing spatial adaptive analysis on ventilation control parameters, the tunnel length characteristics of different road sections are considered; longer tunnels require stronger ventilation capacity, while shorter tunnels have relatively lower ventilation needs. Based on ventilation control parameters, power adjustment suggestions for ventilation systems are generated for different road sections. For example, tunnels longer than 3 kilometers should use a high power setting during peak traffic hours, tunnels 1-3 kilometers long should use a medium power setting, and tunnels shorter than 1 kilometer long should use a low power setting. When performing traffic flow adaptability analysis on signal control parameters, traffic flow characteristics at different times are considered, such as high traffic flow during morning peak hours (7:00-9:00) and evening peak hours (17:00-19:00) and low traffic flow during off-peak hours. Based on signal control parameters, time-specific traffic signal cycle adjustment suggestions are generated, such as using a longer signal cycle (e.g., 90 seconds) during morning and evening peak hours, a shorter signal cycle (e.g., 60 seconds) during off-peak hours, and the shortest signal cycle (e.g., 40 seconds) at night.

[0102] Step 1457: Integrate the time-based lighting system switch control recommendations, the road segment-based ventilation system power adjustment recommendations, and the time-based traffic signal cycle adjustment recommendations into facility control optimization recommendations.

[0103] In this embodiment, the time-segmented lighting system switch control recommendations detail the on / off status of lighting fixtures on each road segment during different time periods. The road segment-segmented ventilation system power adjustment recommendations specify the power levels for tunnels of different lengths under different traffic flow conditions. The time-segmented traffic signal cycle adjustment recommendations determine the signal cycle duration for each intersection during different time periods. These three types of recommendations are categorized and integrated according to infrastructure type (lighting system, ventilation system, traffic signal system) to form facility control optimization recommendations. During the integration process, it is necessary to ensure the coordination and consistency between the recommendations. For example, the switch control recommendations for the lighting system should not conflict with the cycle adjustment recommendations for the traffic signal system in terms of energy consumption control objectives. The facility control optimization recommendations are presented in a clear tabular or textual description format for easy understanding and implementation by management personnel.

[0104] Step 146: Integrate the core-level features, non-core-level features, path optimization suggestions, driving strategy optimization suggestions, and facility control optimization suggestions according to the preset report structure to generate a graphic analysis report containing text descriptions and charts.

[0105] In this embodiment, the pre-defined report structure includes a cover, table of contents, executive summary, multi-dimensional energy consumption status feature analysis, collaborative optimization suggestions, conclusions, and outlook. During integration, core-level features are prominently displayed in the "Multi-dimensional Energy Consumption Status Feature Analysis" section, presented using charts such as bar charts and line graphs. For example, a bar chart compares the total energy consumption before and after optimization, while a line graph shows the changing trend of core-level features over time. Non-core-level features are presented as supplementary content in tables or appendices. Route optimization suggestions, driving strategy optimization suggestions, and facility control optimization suggestions are placed in the "Collaborative Optimization Suggestions" section, each with detailed textual descriptions and accompanying charts such as recommended route diagrams, driving strategy scenario comparison tables, and facility control parameter adjustment tables. The executive summary briefly summarizes the report's main conclusions and core recommendations. The cover includes the report title, generation date, and compiling unit, while the table of contents lists the titles and page numbers of each chapter. This integration process generates a comprehensive, clearly structured, and richly illustrated text and image analysis report.

[0106] Optionally, the method further includes:

[0107] Step 150: Perform real-time data acquisition and processing on the energy consumption monitoring interface of the target highway network to obtain the real-time update values ​​of the first and second energy consumption monitoring data at the current moment.

[0108] In this embodiment, the energy consumption monitoring interface has a real-time data acquisition function. Through real-time communication interfaces with various data sources, it collects the latest energy consumption monitoring data periodically (e.g., every minute) or triggered automatically. For the first real-time updated value of energy consumption monitoring data, the energy consumption monitoring interface establishes a real-time data transmission channel with infrastructure sensors to receive the latest operating data from lighting systems, ventilation systems, traffic signal systems, etc., such as the real-time power of the lighting system, the current operating power level of the ventilation system, and the current phase state of the traffic signal. This data is then timestamped at the current moment and used as the first real-time updated value of energy consumption monitoring data. For the second real-time updated value of energy consumption monitoring data, the energy consumption monitoring interface obtains the latest energy consumption data of vehicles traveling within the target highway network at the current moment through real-time data interfaces with onboard energy consumption monitoring equipment, vehicle identification systems, and traffic flow monitoring devices. This data includes information such as the vehicle's instantaneous fuel consumption, current speed, lane, and road segment. This data is also timestamped at the current moment and used as the second real-time updated value of energy consumption monitoring data.

[0109] Step 151: Compare and analyze the real-time updated value of the first energy consumption monitoring data with the historical first energy consumption monitoring data to calculate the data change rate. When the data change rate exceeds the preset change rate, trigger the first type of data update event; compare and analyze the real-time updated value of the second energy consumption monitoring data with the historical second energy consumption monitoring data to calculate the data change rate. When the data change rate exceeds the preset change rate, trigger the second type of data update event.

[0110] In this embodiment, historical first energy consumption monitoring data refers to the average or median of first energy consumption monitoring data for the same time period over a past period (e.g., the past 24 hours). For example, if the current time is 10:00 AM, historical first energy consumption monitoring data is the average of the first energy consumption monitoring data at 10:00 AM each day over the past week. The formula for calculating the data change rate is: Data change rate = (Real-time update value of first energy consumption monitoring data - Historical first energy consumption monitoring data) / Historical first energy consumption monitoring data × 100%. The preset change rate can be set according to the energy consumption characteristics of the infrastructure, for example, set to ±20%. When the calculated data change rate exceeds the preset change rate (e.g., greater than 20% or less than -20%), it indicates that the energy consumption of the infrastructure has changed significantly, triggering a first type of data update event. Similarly, for the real-time update value of second energy consumption monitoring data, it is compared with the corresponding historical second energy consumption monitoring data (e.g., the average value for the same time period over the past week), and the data change rate is calculated. When the data change rate exceeds the preset change rate (e.g., ±15%), a second type of data update event is triggered.

[0111] Step 152: When the first type of data update event or the second type of data update event is triggered, the multidimensional energy consumption correlation analysis results are dynamically updated, the spatiotemporal correlation analysis processing steps are re-executed, and the updated multidimensional energy consumption correlation analysis results are generated.

[0112] In this embodiment, when either the first or second type of data update event is triggered, it indicates a significant change in the energy consumption monitoring data. The original multidimensional energy consumption correlation analysis results may no longer accurately reflect the current energy consumption correlation situation, requiring dynamic updating. The dynamic updating process involves re-executing the spatiotemporal correlation analysis steps in step 120. This involves performing time-series decomposition, spatial gridding, and spatiotemporal fusion processing on the updated first energy consumption monitoring data (the integration of the original first energy consumption monitoring data and its real-time update value) and the second energy consumption monitoring data (the integration of the original second energy consumption monitoring data and its real-time update value) to generate updated multidimensional energy consumption correlation analysis results. The updated multidimensional energy consumption correlation analysis results include the latest energy consumption influencing factors and correlations, enabling a more accurate reflection of the current energy consumption status of the target highway network.

[0113] Step 153: Input the updated multidimensional energy consumption correlation analysis results into the collaborative decision-making model, and re-execute the steps of model input item extraction, optimization function solution and decision variable combination determination to generate updated collaborative optimization decision results.

[0114] In this embodiment, the updated multidimensional energy consumption correlation analysis results are used as new input data and input into the collaborative decision-making model. The model input extraction step in step 131 is re-executed to extract the latest energy consumption influencing factors from the updated multidimensional energy consumption correlation analysis results as model inputs and perform feature standardization. The optimization function solution and decision variable combination determination steps in steps 132-135 are re-executed. Based on the updated model inputs, the normalization coefficients are recalculated, a single-objective comprehensive optimization function is constructed, the optimal value range of the decision variables is determined, and an optimization solution method is used to obtain a new combination of decision variables under a specified road network operation scenario, i.e., the updated collaborative optimization decision result. This result reflects the optimal vehicle driving path and infrastructure operation strategy after the current energy consumption monitoring data is updated.

[0115] Step 154: Regenerate the graphic analysis report based on the updated collaborative optimization decision results, replace the original graphic analysis report, and display the report update prompt information on the energy consumption monitoring interface. The report update prompt information includes the update time and a summary of the updated content.

[0116] In this embodiment, after obtaining the updated collaborative optimization decision results, the graphical analysis report is regenerated according to the method in step 140. The multidimensional energy consumption status characteristics and collaborative optimization suggestions in the report are all based on the latest collaborative optimization decision results. After generating the new graphical analysis report, the original graphical analysis report in the storage medium is replaced with the new report by file replacement. At the same time, the energy consumption monitoring interface has a message prompt function. After the report is updated, a report update prompt message is displayed at a designated location on the interface (such as the top status bar). The prompt message includes the report update time (such as "2023-10-26, 10:30") and a summary of the update content (such as "Updated path optimization suggestions and facility control parameters"), so that managers can understand the report update status and view the latest analysis results in a timely manner.

[0117] Optionally, the method further includes:

[0118] Step 160: Perform 3D modeling preprocessing on the basic geographic data of the target highway network, extract the road segment orientation features, node connection relationships and terrain undulation parameters of the road network, determine the road segment orientation features as the first type of geometric features, determine the node connection relationships as the second type of geometric features, and determine the terrain undulation parameters as the third type of geometric features.

[0119] In this embodiment, the basic geographic data of the target highway network includes high-precision electronic map data, topographic elevation data, and road network design drawings. The 3D modeling preprocessing first cleans and converts this basic geographic data to unify the data coordinate system and accuracy. Then, it extracts the road segment orientation features. By analyzing the coordinate sequence of road segments in the electronic map, it determines the starting and ending coordinates of each road segment and calculates the azimuth angle of the road segment (e.g., 30 degrees east of north). These features describing the spatial extension direction of the road segments are identified as the first type of geometric features. When extracting node connection relationships, it identifies interchanges, entrances / exits, and other nodes in the road network, records the road segment number and connection method (e.g., T-shaped connection, cross connection) connected to each node, and identifies these features describing the connection between nodes and road segments as the second type of geometric features. When extracting terrain undulation parameters, the elevation changes along each road segment are calculated based on terrain elevation data, such as maximum elevation, minimum elevation, slope (uphill slope, downhill slope) and slope length. These features describing the terrain changes of the road segment are identified as the third type of geometric features.

[0120] Step 161: Based on the first type of geometric features, the second type of geometric features and the third type of geometric features, construct a three-dimensional spatial framework model of the target highway network, and perform road segment unit division processing on the three-dimensional spatial framework model. Divide the continuous road segments into multiple independent three-dimensional road segment units according to the actual direction of the road network. Each three-dimensional road segment unit contains the corresponding spatial coordinate range and geometric shape description.

[0121] In this embodiment, 3D modeling software (such as AutoCAD3D, SketchUp, etc.) is used to construct the 3D road segment axes of the road network based on the first type of geometric features (road segment orientation features). The road segment axes are then connected at nodes according to the second type of geometric features (node ​​connection relationships) to form the 3D topology of the road network. Then, combined with the third type of geometric features (terrain undulation parameters), terrain fitting is performed on the 3D road segment axes to make them conform to the actual terrain undulations, thereby constructing a 3D spatial framework model of the target highway network. This model can roughly reflect the spatial morphology and terrain features of the road network. When dividing the 3D spatial framework model into road segment units, the road segment management unit division standards or terrain change characteristics of the actual road network are referenced, and continuous 3D road segments are divided into multiple independent 3D road segment units according to a certain length (e.g., 1 kilometer) or terrain change nodes. Each 3D road segment unit defines its spatial coordinate range through its starting and ending spatial coordinates. For example, the spatial coordinate range of one 3D road segment unit is (X1, Y1, Z1) to (X2, Y2, Z2), where X and Y are planar coordinates and Z is an elevation coordinate. Simultaneously, the geometry of each 3D road segment unit is described, such as the road width, number of lanes, road surface type, and slope gradient, forming complete 3D road segment unit attribute information.

[0122] Step 162: Extract the energy consumption distribution characteristics of each spatial grid unit from the multidimensional energy consumption correlation analysis results, perform spatial mapping processing on the energy consumption distribution characteristics of the spatial grid unit and the three-dimensional road segment unit, and generate an energy consumption spatial correlation table by establishing the correlation between energy consumption data and three-dimensional spatial location.

[0123] In this embodiment, the multidimensional energy consumption correlation analysis results include the energy consumption distribution characteristics of each spatial grid unit, such as the average energy consumption value and peak energy consumption of each spatial grid unit. Spatial mapping processing establishes a spatial correspondence between planar spatial grid units and three-dimensional road segment units. First, the planar coordinate range of the spatial grid units is converted to the same spatial coordinate system as the three-dimensional road segment units. Then, through spatial overlay analysis, it is determined which three-dimensional road segment units each spatial grid unit overlaps with, and the area ratio of the overlapping region is calculated. When the area ratio of the overlapping region exceeds a preset threshold (e.g., 50%), the spatial grid unit is considered to have a major correlation with the corresponding three-dimensional road segment unit. The energy consumption distribution characteristics (e.g., average energy consumption value) of the spatial grid unit are assigned to the three-dimensional road segment units with which it has a major correlation, establishing a correlation between energy consumption data (energy consumption distribution characteristics of spatial grid units) and three-dimensional spatial location (three-dimensional road segment units). This relationship is recorded in a table. The rows of the table represent the three-dimensional road segment unit number, the columns represent the energy consumption distribution characteristic type (such as average energy consumption, peak energy consumption), and the cell values ​​in the table are the corresponding energy consumption distribution characteristic values. This table is the energy consumption spatial association table.

[0124] Step 163: Set different visual representation forms according to the differences in the types of energy consumption distribution characteristics, correspond energy consumption characteristics related to time dynamic characteristics to dynamic annotation elements and energy consumption characteristics related to spatial distribution characteristics to static annotation elements, and generate energy consumption feature visualization annotation rules based on the dynamic annotation elements and the static annotation elements.

[0125] In this embodiment, the types of energy consumption distribution characteristics include energy consumption characteristics related to time dynamics (such as the rate of change of energy consumption over time, the amplitude of periodic fluctuations, etc.) and energy consumption characteristics related to spatial distribution (such as the spatial distribution density of energy consumption, the difference in energy consumption between road segments, etc.). For energy consumption characteristics related to time dynamics, dynamic annotation elements are set as their visual representation, for example, using color gradient animation to represent the trend of energy consumption changing over time, with the color gradually transitioning from blue (low energy consumption) to red (high energy consumption), and the playback speed of the animation corresponding to the speed of time change; or using dynamic bar charts to update and display the energy consumption values ​​of different time periods in real time on the three-dimensional road segment unit. For energy consumption characteristics related to spatial distribution, static annotation elements are set as their visual representation, for example, using color mapping (such as heat map) to represent the spatial distribution density of energy consumption, with darker colors indicating higher energy consumption density; or using bars of different heights superimposed on the three-dimensional road segment unit to represent the difference in energy consumption between road segments, with taller bars indicating greater energy consumption. Based on the settings of these dynamic and static annotation elements, energy consumption feature visualization annotation rules are formulated. The rules specify the annotation elements, color coding scheme, size ratio, update frequency and other parameters corresponding to each type of energy consumption feature.

[0126] Step 164: Based on the energy consumption spatial association table and the energy consumption feature visualization annotation rules, perform energy consumption information annotation processing on each three-dimensional road segment unit in the three-dimensional road network model, and superimpose dynamic annotation elements and static annotation elements onto the surface of the corresponding three-dimensional road segment unit to generate a three-dimensional visualized road network model containing energy consumption information.

[0127] In this embodiment, the energy consumption spatial association table provides energy consumption distribution characteristic values ​​corresponding to each three-dimensional road segment unit, and the energy consumption characteristic visualization annotation rules specify how to display these characteristic values ​​in the form of visual elements. When performing energy consumption information annotation processing on each three-dimensional road segment unit in the three-dimensional road network model, firstly, the various energy consumption distribution characteristic values ​​of the three-dimensional road segment unit are found according to the energy consumption spatial association table, and then, according to the energy consumption characteristic visualization annotation rules, the corresponding dynamic annotation elements or static annotation elements are selected, and the attributes of the elements (such as color, size, animation parameters, etc.) are set. For example, for one three-dimensional road segment unit, its average energy consumption value related to spatial distribution characteristics is 80 units. According to the rules, the corresponding static annotation element is a color mapping, and 80 units corresponds to orange. Therefore, an orange color overlay is superimposed on the surface of the three-dimensional road segment unit; its energy consumption change rate related to time dynamic characteristics is positive 5%, and the corresponding dynamic annotation element is a color gradient animation, gradually transitioning from the current orange to red. After overlaying the dynamic and static annotation elements of all three-dimensional road segment units, a three-dimensional visualized road network model containing energy consumption information is generated. This model can intuitively display the energy consumption distribution and dynamic changes of each segment of the target highway network.

[0128] Step 165: Perform multi-view rendering processing on the three-dimensional visualized road network model to generate a visualized image.

[0129] In this embodiment, multi-view rendering refers to rendering a 3D visualized road network model from different observation angles to generate visualized images from multiple perspectives, comprehensively showcasing the energy consumption information of the road network. Selected observation perspectives include a global perspective (an aerial view of the entire target highway network, showing the overall energy consumption distribution), a regional perspective (focusing on a specific area of ​​the road network, such as the periphery of a city, showing local energy consumption details), and a road segment perspective (perspective observation along the driving direction of a specific road segment, showing the dynamic changes in energy consumption of that segment). For each perspective, appropriate rendering parameters are set, such as lighting conditions (simulating the effects of natural light at different times), background color, and fog effects, to enhance the three-dimensionality and readability of the visualized images. After rendering is complete, visualized images corresponding to each perspective are generated; these images can be static pictures or dynamic video frame sequences.

[0130] Step 166: Associate and integrate the visualized image with the multidimensional energy consumption status features in the graphic analysis report, configure the visualized image reference mark in the graphic analysis report, and generate an enhanced graphic analysis report containing three-dimensional visualization content; each visualized image reference mark corresponds to a multidimensional energy consumption status feature described in the report text.

[0131] In this embodiment, visualized images can intuitively display the spatial distribution and dynamic change characteristics of multidimensional energy consumption state features. Integrating these images with the text and image analysis report enhances the report's expressiveness and readability. During the integration process, the various multidimensional energy consumption state features described in the "Multidimensional Energy Consumption State Feature Analysis" section of the text and image analysis report are first analyzed to identify which features are suitable for visualization, such as spatial energy consumption distribution features and dynamic characteristics of energy consumption changes over time. For these features, visualized image reference markers are configured in the corresponding description positions in the report text. These reference markers can be image numbers or icons. Then, the visualized images generated by multi-view rendering are inserted into the corresponding positions in the report (such as the appendix of the text and image analysis report or a dedicated visualization section), ensuring that each visualized image reference marker establishes a clear correspondence with its corresponding visualized image; that is, each reference marker corresponds to a multidimensional energy consumption state feature described in the report text and a visualized image in the appendix. For example, when describing the multidimensional energy consumption status feature that “the energy consumption of road segment D is significantly lower than that of other road segments”, the reference tag “Pic1” is configured. This tag corresponds to the visualization image of the global energy consumption distribution heat map of the road network shown in the appendix, where road segment D is displayed in dark color (representing low energy consumption).

[0132] This application embodiment obtains energy consumption monitoring data on road network infrastructure operation and vehicle traffic by filtering multi-source data from the target highway network energy consumption monitoring interface. Then, it conducts spatiotemporal correlation analysis to generate multi-dimensional energy consumption correlation analysis results. Based on this, a collaborative decision-making model is constructed using energy consumption influencing factors as inputs and vehicle travel path planning parameters and infrastructure operation adjustment parameters as decision variables. This enables collaborative optimization decisions on vehicle travel paths and infrastructure operation strategies. Finally, a graphical analysis report is generated based on the decision results. Overall, this application embodiment realizes a complete process from multi-source energy consumption data acquisition and analysis to decision-making and report generation. It can comprehensively and deeply understand the energy consumption of the target highway network, providing scientific and accurate collaborative optimization decisions for highway energy consumption management, effectively reducing the overall energy consumption of the highway network, and improving network operation efficiency and energy utilization efficiency.

[0133] Please see Figure 2The figure is a schematic diagram of the basic structure of an analysis report generation system 200 provided in an embodiment of this application. The analysis report generation system 200 includes: a processor 201; a storage device 202 on which a computer program 2020 is stored; and a network interface 203 for providing network communication functions. When the computer program 2020 is executed by the processor 201, the processor 201 implements any of the analysis report generation methods based on the highway energy consumption monitoring interface.

[0134] Please see Figure 3 This application provides a functional block diagram of an analysis report generation device based on a highway energy consumption monitoring interface. The analysis report generation device based on a highway energy consumption monitoring interface includes:

[0135] The data filtering module is used to filter multi-source data through the energy consumption monitoring interface of the target highway network to obtain first energy consumption monitoring data related to the operation of the road network infrastructure and second energy consumption monitoring data related to the driving process of passing vehicles. The correlation analysis module is used to perform spatiotemporal correlation analysis on the first energy consumption monitoring data and the second energy consumption monitoring data to generate multidimensional energy consumption correlation analysis results of the target highway network. The model building module is used to take the energy consumption influencing factors in the multidimensional energy consumption correlation analysis results as model input items, vehicle driving path planning parameters and infrastructure operation adjustment parameters as decision variables, and generate a collaborative decision model for collaborative optimization decision-making on vehicle driving paths and infrastructure operation strategies by establishing a mapping relationship between the model input items and the decision variables. The collaborative decision model is used to determine the collaborative optimization decision results output for the target highway network under a specified road network operation scenario. The report generation module is used to generate a graphic analysis report containing multidimensional energy consumption status characteristics and collaborative optimization suggestions based on the collaborative optimization decision results.

[0136] Based on the above, a readable storage medium is provided, on which a program or instructions are stored, and when the program or instructions are executed by a processor, the steps of the above method are implemented.

[0137] Furthermore, it should be noted that this application also provides a computer program product, which may include a computer program that can be stored in a computer-readable storage medium. The processor of the analysis report generation system reads the computer program from the computer-readable storage medium, and the processor can execute the computer program, causing the analysis report generation system to perform the aforementioned... Figure 1The methods described in the corresponding embodiments are already known, and therefore will not be repeated here. Furthermore, the beneficial effects of using the same method will also not be repeated. For technical details not disclosed in the computer program product embodiments related to this application, please refer to the description of the method embodiments of this application.

Claims

1. A method for generating an analysis report based on a highway energy consumption monitoring interface, characterized by, The method comprises: Through the energy consumption monitoring interface of the target highway network, multi-source data filtering processing is performed to obtain first energy consumption monitoring data related to the operation process of the highway network infrastructure and second energy consumption monitoring data related to the driving process of the passing vehicles; Temporal and spatial correlation analysis processing is performed on the first energy consumption monitoring data and the second energy consumption monitoring data to generate a multi-dimensional energy consumption correlation analysis result of the target highway network; The energy consumption influencing factors in the multi-dimensional energy consumption correlation analysis result are taken as model input items, the vehicle driving path planning parameters and the infrastructure operation adjustment parameters are taken as decision variables, a mapping relationship between the model input items and the decision variables is established to generate a collaborative decision model for collaborative optimization and decision of the vehicle driving path and the infrastructure operation strategy, and a collaborative optimization and decision result output by the collaborative decision model under a specified network operation scenario of the target highway network is determined, comprising: Key factors influencing the energy consumption of the highway network are extracted from the multi-dimensional energy consumption correlation analysis result as model input items of the collaborative decision model and subjected to feature standardization processing; The total driving energy consumption of the vehicles in the target highway network and the total operation energy consumption of the highway network infrastructure are respectively subjected to normalization processing, a normalization coefficient is calculated based on historical energy consumption data or a preset energy consumption benchmark value to obtain normalized vehicle driving energy consumption and normalized infrastructure operation energy consumption, a double-objective optimization function of the collaborative decision model is determined, and the double-objective optimization function is converted into a single-objective comprehensive optimization function by setting a target weight coefficient, a first optimization objective is to minimize the normalized vehicle driving energy consumption, and a second optimization objective is to minimize the normalized infrastructure operation energy consumption; The road section selection variable, the lane selection variable and the driving speed variable in the vehicle driving path planning parameter, and the lighting system switch control variable, the ventilation system power adjustment variable and the traffic signal cycle adjustment variable in the infrastructure operation adjustment parameter are taken as decision variables of the collaborative decision model; A mapping relationship model between the energy consumption influencing factors and the decision variables is established by analyzing the sensitivity of each decision variable to the energy consumption of the highway network under different energy consumption influencing factors to generate a mapping function reflecting the energy consumption transmission mechanism, and the feature values of the model input items are converted into the optimization value range of the decision variables by using the mapping function; In combination with the optimization value range and the specified network operation scenario of the target highway network, an optimization solving method under multiple constraints is adopted to solve the single-objective comprehensive optimization function, and under the premise of meeting the vehicle passing time constraint, the infrastructure operation safety constraint and the traffic flow control constraint, a decision variable combination that minimizes the single-objective comprehensive optimization function is found, and the decision variable combination is taken as the collaborative optimization and decision result of the collaborative decision model. According to the collaborative optimization decision result, a graphic and text analysis report containing multi-dimensional energy consumption state features and collaborative optimization suggestions is generated; wherein, the multi-dimensional energy consumption state features are obtained by feature extraction and combination of different class decision results in the collaborative optimization decision result, and the graphic and text analysis report is obtained by integration processing according to a preset report structure.

2. The method of claim 1, wherein, The spatio-temporal correlation analysis processing is performed on the first energy consumption monitoring data and the second energy consumption monitoring data to generate a multi-dimensional energy consumption correlation analysis result of the target highway network, including: The first energy consumption monitoring data is subjected to time series decomposition processing to extract energy consumption change trend features of the infrastructure energy consumption at different time granularities, and the second energy consumption monitoring data is subjected to vehicle flow statistical processing to obtain vehicle flow change features at corresponding time granularities, and the energy consumption change trend features and the vehicle flow change features at the same time granularity are correlated and matched to establish an energy consumption flow correlation curve in the time dimension; The first energy consumption monitoring data is subjected to spatial grid division processing, and the target highway network is divided into a plurality of spatial grid units according to a preset road section length, and the average infrastructure energy consumption in each spatial grid unit is calculated as a spatial energy consumption distribution feature, and the vehicle driving track in the second energy consumption monitoring data is subjected to road section matching processing to determine the driving time proportion of the vehicle in each spatial grid unit, and the spatial energy consumption distribution feature and the driving time proportion are combined to establish an energy consumption track correlation matrix in the spatial dimension; The energy consumption flow correlation curve in the time dimension and the energy consumption track correlation matrix in the spatial dimension are subjected to spatio-temporal fusion processing by calculating the energy consumption influence weight of different spatio-temporal intersection points to generate a multi-dimensional energy consumption correlation analysis result containing time dynamic characteristics and spatial distribution characteristics, and the multi-dimensional energy consumption correlation analysis result is used to reflect the comprehensive influence of time change and spatial distribution on the highway energy consumption.

3. The method of claim 1, wherein, The key factors influencing the highway network energy consumption are extracted from the multi-dimensional energy consumption correlation analysis result as model input items of the collaborative decision model and subjected to feature standardization processing, including: The multi-dimensional energy consumption correlation analysis result is subjected to factor type identification processing to distinguish time dynamic characteristic related factors and spatial distribution characteristic related factors, the time dynamic characteristic related factors are taken as first type influence factors, and the spatial distribution characteristic related factors are taken as second type influence factors; The first type influence factors are subjected to trend feature extraction processing to identify factors with periodic fluctuation characteristics and factors with non-periodic fluctuation characteristics, the factors with periodic fluctuation characteristics are classified into a time periodicity factor set, and the factors with non-periodic fluctuation characteristics are classified into a time randomness factor set; The second type influence factors are subjected to spatial correlation analysis processing to calculate the distribution similarity of the factors in different spatial grid units, the factors with a distribution similarity reaching a preset similarity are classified into a spatial aggregation factor set, and the factors with a distribution similarity not reaching the preset similarity are classified into a spatial dispersion factor set; generating an initial key factor set based on the set of time periodic factors, the set of time random factors, the set of spatial aggregation factors and the set of spatial dispersion factors; mapping the characteristic values of each factor in the initial key factor set to a preset standard value interval according to the data type of the factor, and generating model input items matching the input requirements of the collaborative decision-making model.

4. The method of claim 1, wherein, The mapping relationship model between the energy consumption influencing factors and the decision variables is established by analyzing the sensitivity of each decision variable to the highway network energy consumption under different energy consumption influencing factors, a mapping function reflecting the energy consumption transmission mechanism is generated, and the characteristic values of the model input items are converted into the optimization value range of the decision variables by using the mapping function, which includes: selecting any energy consumption influencing factor in the model input item as the current analysis factor, keeping the characteristic values of the energy consumption influencing factors other than the current analysis factor unchanged, and adjusting the characteristic values of the current analysis factor to different levels in turn; At each characteristic value level, the values of each decision variable are adjusted respectively, the energy consumption change amount of the highway network is recorded when each decision variable value changes, and the ratio of the energy consumption change amount to the decision variable value change amount is calculated as the sensitivity coefficient at the current characteristic value level; trend fitting processing is performed on the sensitivity coefficients at all characteristic value levels to generate a sensitivity change curve between the current analysis factor and each decision variable; the next energy consumption influencing factor is replaced as the current analysis factor, and the steps of characteristic value level adjustment, sensitivity coefficient calculation and sensitivity change curve generation are repeated until all energy consumption influencing factors are analyzed; Based on the sensitivity change curves of all energy consumption influencing factors and decision variables, a mapping relationship model between energy consumption influencing factors and decision variables is constructed, and each decision variable in the mapping relationship model corresponds to a function set composed of sensitivity change curves of different energy consumption influencing factors; extracting the sensitivity threshold interval of each decision variable from the function set, taking the energy consumption influencing factors with sensitivity reaching the sensitivity threshold as the main influencing factors, and taking the energy consumption influencing factors with sensitivity not reaching the sensitivity threshold as the secondary influencing factors, to generate a main influencing factor set of the decision variables; According to the main influencing factor set and the mapping relationship model, a mapping function reflecting the energy consumption transmission mechanism is generated, and the mapping function takes the characteristic values of the main influencing factors as input parameters and takes the value range of the decision variables as output results. The characteristic values of the model input items are substituted into the mapping function, and the optimization value range of each decision variable under the current characteristic value is calculated, which is the feasible interval of the decision variable under the constraint of the energy consumption transmission mechanism.

5. The method of claim 1, wherein, The collaborative optimization decision result is analyzed and processed, the vehicle driving path planning parameter optimization result and the infrastructure operation adjustment parameter optimization result are extracted, the vehicle driving path planning parameter optimization result is taken as the first type of decision result, and the infrastructure operation adjustment parameter optimization result is taken as the second type of decision result. ​ The first type of decision result is subjected to energy consumption influence feature extraction processing, and energy consumption change characteristics under different road section selection variables, lane selection variables and driving speed variable combinations are analyzed to generate a first multi-dimensional energy consumption state feature related to vehicle driving, and the first multi-dimensional energy consumption state feature is taken as a vehicle energy consumption feature set; The second type of decision result is subjected to energy consumption influence feature extraction processing, and energy consumption change characteristics under different lighting system switch control variables, ventilation system power adjustment variables and traffic signal cycle adjustment variable combinations are analyzed to generate a second multi-dimensional energy consumption state feature related to infrastructure operation, and the second multi-dimensional energy consumption state feature is taken as a facility energy consumption feature set; The vehicle energy consumption feature set and the facility energy consumption feature set are merged to generate a comprehensive energy consumption state feature set, and the features in the comprehensive energy consumption state feature set are subjected to hierarchical division processing, the core level features are taken as the key display content of the graphic-text analysis report, and the non-core level features are taken as the supplementary display content; According to the road section selection variable optimization result in the vehicle energy consumption feature set, a path optimization suggestion is generated, a driving strategy optimization suggestion is generated in combination with the lane selection variable and driving speed variable optimization results, and a facility control optimization suggestion is generated using the infrastructure operation adjustment parameter optimization result; The core level features, the non-core level features, the path optimization suggestion, the driving strategy optimization suggestion and the facility control optimization suggestion are integrated according to the preset report structure to generate a graphic-text analysis report including a text description part and a chart display part.

6. The method of claim 5, wherein, The first type of decision result is subjected to energy consumption influence feature extraction processing, and energy consumption change characteristics under different road section selection variables, lane selection variables and driving speed variable combinations are analyzed to generate a first multi-dimensional energy consumption state feature related to vehicle driving, and the first multi-dimensional energy consumption state feature is taken as a vehicle energy consumption feature set, including: All potential combinations of road section selection variables, lane selection variables and driving speed variables are extracted from the first type of decision result, and each combination is taken as a decision combination unit; Each decision combination unit is subjected to energy consumption simulation analysis processing, the driving process of the vehicle under the road section, lane and speed conditions corresponding to the decision combination unit is simulated, the energy consumption change data in the driving process is recorded, and the peak energy consumption value, average energy consumption value and energy consumption fluctuation amplitude value in the energy consumption change data are extracted; The peak energy consumption value, average energy consumption value and energy consumption fluctuation amplitude value are taken as basic feature parameters, and the basic feature parameters are subjected to normalization processing to generate standardized feature parameters; The standardized feature parameters are subjected to feature fusion processing, the peak energy consumption standardized parameter, average energy consumption standardized parameter and energy consumption fluctuation amplitude standardized parameter of the same decision combination unit are combined into a feature vector, and the feature vector is taken as the energy consumption state feature vector corresponding to the decision combination unit; The energy consumption state feature vectors of all decision combination units are subjected to clustering analysis processing, and the decision combination units with a feature vector similarity value higher than a preset similarity value are classified into the same feature category, and the center feature vector of each feature category is calculated as the representative performance energy consumption state feature of the category; The representative performance energy consumption state features of all categories are integrated into a set and used as a vehicle energy consumption feature set, and each feature in the vehicle energy consumption feature set corresponds to a typical energy consumption state of a decision combination unit.

7. The method of claim 5, wherein, The path optimization suggestion is generated according to the road section selection variable optimization result in the vehicle energy consumption feature set, the driving strategy optimization suggestion is generated in combination with the lane selection variable and driving speed variable optimization results, and the facility control optimization suggestion is generated by using the infrastructure operation adjustment parameter optimization result, which includes: The road section selection variable optimization result is extracted from the vehicle energy consumption feature set, the energy consumption change proportion of different road sections before and after optimization is analyzed, and the road section with an energy consumption change proportion higher than a preset proportion value is used as a key optimization road section; The path correlation analysis processing is performed on the key optimization road section, the upstream and downstream road sections directly connected to the key optimization road section are identified, a path network containing the key optimization road section and its upstream and downstream road sections is constructed, and the path combination with the minimum energy consumption sum is found in the path network to generate a path optimization suggestion, which includes a recommended driving road section sequence and a recommended driving direction of each road section; The lane selection variable and driving speed variable optimization results are extracted from the vehicle energy consumption feature set, the energy consumption saving index under different lane and speed combinations is analyzed, and the lane and speed combination with the maximum weight value corresponding to the energy consumption saving index is used as the optimal driving combination; The scenario adaptability analysis processing is performed on the optimal driving combination based on different traffic flow conditions to generate a differentiated driving strategy under different traffic flow scenarios, and the differentiated driving strategy is integrated into a driving strategy optimization suggestion, which includes a recommended lane and a recommended driving speed range under different scenarios; The infrastructure operation adjustment parameter optimization result is extracted from the second type of decision result, and the optimization parameters of the lighting system, ventilation system and traffic signal system are determined, the optimization parameters of the lighting system are used as lighting control parameters, the optimization parameters of the ventilation system are used as ventilation control parameters, and the optimization parameters of the traffic signal system are used as signal control parameters; The time period adaptability analysis processing is performed on the lighting control parameters, the lighting system switch control suggestion is generated in different time periods in combination with the natural light intensity characteristics of different time periods, the space adaptability analysis processing is performed on the ventilation control parameters, the ventilation system power adjustment suggestion is generated in different road sections in combination with the tunnel length characteristics of different road sections, and the flow adaptability analysis processing is performed on the signal control parameters, the traffic signal cycle adjustment suggestion is generated in different time periods in combination with the traffic flow characteristics of different time periods; The lighting system switch control suggestion in different time periods, the ventilation system power adjustment suggestion in different road sections and the traffic signal cycle adjustment suggestion in different time periods are integrated into a facility control optimization suggestion.

8. An analytical report generation system characterized by, which includes: The processor, a storage device having a computer program stored thereon, a network interface for providing network communication function, when the computer program is executed by the processor, the processor implements the analysis report generation method based on the expressway energy consumption monitoring interface as claimed in any one of claims 1-7.

9. A readable storage medium, characterized by, The program or instruction stored on the readable storage medium is executed by the processor to implement the analysis report generation method based on the expressway energy consumption monitoring interface as claimed in any one of claims 1-7.

Images