A new energy charging management method for smart cities

Abstract

The present application relates to the field of new energy charging, and in particular to a new energy charging management method for smart cities. The method includes: obtaining a municipal vehicle operation feature data set, and on this basis, dividing municipal vehicles into types through a vehicle type classification strategy to determine a municipal vehicle drive type set; obtaining a municipal vehicle task set, and on this basis, tracking the dynamic task variable set of the task execution vehicle in the municipal vehicle drive type set during the task execution process; according to the dynamic task variable set, analyzing the correlation between the dynamic task variable set and the real-time power consumption data through a multimodal fusion strategy to determine the dynamic power change information; obtaining a charging station information set, combining the municipal vehicle task set and the dynamic power change information, dynamically optimizing the charging decision, and determining and outputting the preferred charging strategy. In the municipal vehicle charging decision-making process, the present application improves the utilization rate of charging resources and ensures the continuity of municipal task execution.

Description

Technical Field

[0001] The present application relates to the field of new energy charging, and in particular to a new energy charging management method for smart cities. Background Art

[0002] Under the construction of smart cities, the use of new energy for municipal vehicles has become a key path to improving urban operation efficiency, reducing carbon emissions and improving environmental quality. As an important carrier of urban public services, its efficient and reliable operation directly affects the achievement of urban governance efficiency goals.

[0003] However, the existing charging management methods for municipal new energy vehicles have significant limitations and are unable to predict and match the dynamic power consumption generated by municipal vehicles during their missions, resulting in inefficient use of power resources and increased operating costs. Summary of the Invention

[0004] This application provides a new energy charging management method for a smart city to solve the above technical problems.

[0005] This application provides a new energy charging management method for a smart city, the method comprising:

[0006] Obtain a municipal vehicle operation feature data set, and based on the municipal vehicle operation feature data set, classify the municipal vehicles into types through a vehicle type classification strategy to determine a municipal vehicle drive type set; obtain a municipal vehicle task set, and based on the municipal vehicle drive type set, track a dynamic task variable set of the task-performing vehicles in the municipal vehicle drive type set during the task execution process; based on the dynamic task variable set, analyze the correlation between the dynamic task variable set and the real-time power consumption data through a multimodal fusion strategy to determine dynamic power change information; obtain a charging station information set, and based on the municipal vehicle task set, the dynamic power change information and the charging station information set, dynamically optimize the charging decision, and determine and output the preferred charging strategy.

[0007] This solution collects operational characteristic data for municipal vehicles, classifies them into different types, identifies their drive types, and obtains a list of municipal vehicle tasks. Dynamic parameters during the execution of task vehicles are monitored based on their drive types. Multimodal fusion technology is used to analyze the correlation between these dynamic parameters and real-time power consumption data, deriving patterns of dynamic power changes. Charging station information is collected, and the optimal charging strategy is dynamically generated by integrating the task list, power change patterns, and charging station information. This solution accurately predicts dynamic power change information through a multimodal fusion strategy, significantly reducing the error rate in power prediction for task vehicles and significantly improving task assurance. By further combining the municipal vehicle drive type set and the charging station information set, triple data collaborative optimization is achieved in the dynamic optimization of charging decisions, ensuring minimal conflict in charging periods at the task level, driving increased utilization of charging resources and charging stations, and effectively reducing duplicate investment in municipal infrastructure.

[0008] Optionally, the vehicle type classification strategy includes:

[0009] Extracting the main causes of energy consumption fluctuations in the municipal vehicle operation characteristic dataset based on the vehicle type classification strategy;

[0010] The main causes of energy consumption fluctuations include changes in physical carriers, environmental factors, and event factors;

[0011] If the energy consumption fluctuation of the current municipal vehicle is mainly caused by the change in the physical carrier, then the current municipal vehicle is determined to be a physical-driven vehicle;

[0012] If the energy consumption fluctuation of the current municipal vehicle is mainly caused by the change in the environmental factors, then the current municipal vehicle is determined to be an environmentally driven vehicle;

[0013] If the energy consumption fluctuation of the current municipal vehicle is mainly caused by the change in the event factor, then the current municipal vehicle is determined to be an event-driven vehicle;

[0014] The municipal vehicle driving type set is generated according to the entity-driven vehicle, the environment-driven vehicle and the event-driven vehicle.

[0015] Through this solution, by accurately identifying the dominant factors of energy consumption, we can achieve accurate classification of municipal vehicles, laying an important foundation for subsequent charging strategies; based on entity-driven vehicles, environment-driven vehicles and event-driven vehicles, we generate a set of municipal vehicle drive types, which improves the accuracy of charging strategies, significantly reduces the error in charging volume estimation, and reduces the interruption rate of municipal task execution, ensuring the reliability of municipal services.

[0016] Optionally, the acquiring of the municipal vehicle task set and tracking of the dynamic task variable set of the task-performing vehicle in the municipal vehicle drive type set during the task execution process according to the municipal vehicle drive type set include:

[0017] The municipal vehicle task set includes operation characteristics, path characteristics and vehicle status characteristics;

[0018] Tracking changes in the main causes of energy consumption fluctuations during task execution of the current municipal vehicle according to the municipal vehicle drive type set, and determining information on changes in the main causes of energy consumption fluctuations;

[0019] According to the information on the main causes of energy consumption fluctuations, feature mapping and dynamic processing are performed on the operation characteristics, the path characteristics, and the vehicle status characteristics to generate a dynamic load operation feature set, an environmental response situation feature set, and an event-driven task feature set;

[0020] The dynamic load operation feature set includes real-time entity task load and load state change rate;

[0021] The environmental response situation feature set includes a real-time natural environment data set and a real-time man-made environment data set;

[0022] The event-driven task feature set includes task urgency, task realization intensity and task target location;

[0023] The dynamic task variable set is generated according to the dynamic load operation feature set, the environmental response situation feature set and the event-driven task feature set.

[0024] Through this solution, the energy consumption management of municipal vehicles is significantly improved by constructing three types of collaborative dynamic feature sets. The dynamic load feature set accurately quantifies the changes in physical carriers, greatly improving the energy consumption prediction accuracy of physical vehicles when the load suddenly changes; the environmental response feature set integrates natural and man-made environmental data, effectively optimizing the response speed of environmental vehicles in complex scenarios; the event task feature set quickly analyzes key task indicators, significantly enhancing the reliability of event vehicles in responding to emergency dispatch.

[0025] Optionally, the multimodal fusion strategy includes:

[0026] If the vehicle type is the entity-driven vehicle, extract the dynamic load operation feature set from the dynamic task variable set, combine it with the real-time power consumption data, analyze the gain / loss effect of the real-time entity task load and the load state change rate on the power consumption rate, and generate an entity change impact sub-strategy;

[0027] If the vehicle type is the environment-driven vehicle, extracting the environmental response situation feature set from the dynamic task variable set, combining it with the real-time power consumption data, analyzing the gain / loss effect of the real-time natural environment dataset and the real-time man-made environment dataset on the power consumption rate, and generating an environmental change impact sub-strategy;

[0028] If the vehicle type is the event-driven vehicle, extract the event-driven task feature set from the dynamic task variable set, combine it with the real-time power consumption data, analyze the gain / loss effect of the task urgency, the task implementation intensity, and the task target location on the power consumption rate, and generate an event-focused impact sub-strategy;

[0029] The multimodal fusion strategy is generated according to the entity change impact sub-strategy, the environment change impact sub-strategy and the event focus impact sub-strategy.

[0030] This solution generates dynamic sub-policies by type through a multimodal fusion mechanism, and can directly call preset strategy modules to achieve rapid response in emergency task scenarios, avoiding the time delay of global model retraining. At the same time, relying on precise energy consumption gain / loss analysis, it optimizes vehicle charging priority allocation under the constraint of limited charging resources, effectively avoiding charging station conflicts and grid overload risks, thereby systematically supporting the economy and operational reliability of the municipal charging network.

[0031] Optionally, the entity change affects the sub-strategy, including:

[0032] Based on the real-time entity task load and the load state change rate, identifying gain nodes and loss nodes on the path of the task execution vehicle;

[0033] The gain node represents a node where the real-time entity task load generates a gain on the power consumption rate;

[0034] The depreciation node represents a node where the real-time entity task load depreciates the power consumption rate;

[0035] According to the gain node and the degauss node, combined with the load state change rate, the change trend of the power consumption rate is predicted to determine the dynamic power change information of the physical drive vehicle.

[0036] This solution utilizes real-time physical task load and load state change rate to identify gain and loss nodes on the task execution vehicle's path. Through load-path position linkage, change rate-driven modeling, and dynamic charging decision-making fulcrum generation, the "physical load dynamics → power consumption response → charging strategy generation" closed loop is established. Through precise identification of gain / loss nodes and dynamic fusion of load change rate, the prediction error of the power consumption rate of physical-driven vehicles is reduced, thereby reducing the risk of power exhaustion mid-task.

[0037] Optionally, the environmental change affects the sub-strategy, including:

[0038] For the environment-driven vehicle, the environment change impact sub-strategy is adopted:

[0039] Analyze the real-time natural environment data set, and use the correlation factors in the natural environment that generate a gain on the power consumption rate as natural environment positive correlation factors, and use the correlation factors in the environment that generate a loss on the power consumption rate as natural environment negative correlation factors;

[0040] Analyzing the plurality of positively correlated natural environment factors, performing gain superposition on the plurality of positively correlated natural environment factors, and determining a first natural environment impact coefficient;

[0041] Analyze the plurality of negatively correlated factors of the natural environment, perform a reduction and superposition on the plurality of negatively correlated factors of the natural environment, and determine a second natural environment impact coefficient;

[0042] performing normalization processing on the first natural environment impact coefficient and the second natural environment impact coefficient according to the first natural environment impact coefficient and the second natural environment impact coefficient to determine a natural environment response intensity coefficient;

[0043] Analyze the real-time human environment data set, and use the correlation factors in the human environment that increase the power consumption rate as positive human environment correlation factors, and use the correlation factors in the environment that decrease the power consumption rate as negative human environment correlation factors;

[0044] Analyzing the plurality of positively correlated factors of the human-made environment, performing gain superposition on the plurality of positively correlated factors of the human-made environment, and determining a first human-made environment impact coefficient;

[0045] Analyzing the plurality of negatively correlated factors of the human-made environment, performing a deduction and superposition on the plurality of negatively correlated factors of the human-made environment, and determining a second human-made environment impact coefficient;

[0046] performing normalization processing on the first man-made environmental impact coefficient and the second man-made environmental impact coefficient according to the first man-made environmental impact coefficient and the second man-made environmental impact coefficient to determine a man-made environmental response intensity coefficient;

[0047] Analyzing the real-time human environment data set to determine infrastructure characteristics, and setting a number of infrastructure verification time intervals with different time spans based on the infrastructure characteristics;

[0048] Based on the real-time power consumption data, the impact of the superposition of the natural environment response intensity coefficient and the man-made environment response intensity coefficient on the power consumption rate in each infrastructure approved time interval is analyzed, the changing trend of the power consumption rate is predicted, and the dynamic power change information of the environmentally driven vehicle is determined.

[0049] This solution utilizes multi-factor collaborative analysis to classify positive and negative correlation factors in the natural and man-made environments, superimposes coefficients to quantify coupling effects, resolves prediction biases caused by environmental coupling effects, and reduces power consumption rate prediction errors. Through spatiotemporal dynamic calibration, regional adaptability is achieved by binding infrastructure to approved time intervals, and charging opportunities are dynamically recommended, enhancing the forward-looking nature of charging strategies. Relying on normalized decision support, multi-source data is transformed into a unified dimension, and a dynamic mapping relationship between the environment and energy consumption is constructed, providing necessary support for high-precision power prediction and forward-looking charging strategies.

[0050] Optionally, the event focuses on affecting sub-strategies, including:

[0051] For the event-driven vehicle, the event-focused impact sub-strategy is adopted:

[0052] Determining a task urgency index according to the task urgency, and using the task urgency index as a first impact coefficient;

[0053] Determining current task information based on the municipal vehicle task set, and determining task duration and task operation complexity of the corresponding task based on the current task information;

[0054] Normalizing the task duration and the task operation complexity according to a preset numerical normalization strategy to determine a task duration index and a task complexity index;

[0055] The product of the task duration index and the task complexity index is used as the second impact coefficient;

[0056] Determine the mission arrival route based on the mission target location;

[0057] Based on preset traffic remote sensing information and according to the task arrival route, determining the real-time traffic complexity and driving distance corresponding to the task arrival route;

[0058] Normalizing the real-time traffic complexity and the driving distance according to a preset numerical normalization strategy to determine a traffic complexity index and a driving distance index;

[0059] taking the product of the road condition complexity index and the driving distance index as a third influence coefficient;

[0060] taking the sum of the first influence coefficient, the second influence coefficient, and the third influence coefficient as the comprehensive event influence factor;

[0061] According to the comprehensive event impact factor and the real-time power consumption data, the change trend of the power consumption rate is predicted, and the dynamic power change information of the event-driven vehicle is determined.

[0062] Through this solution, through quantitative analysis of multi-dimensional event factors, dynamic power change information can accurately reflect the power consumption patterns of event-driven vehicles in different mission scenarios, avoiding power shortages or waste caused by the disconnection between charging strategies and mission requirements; at the same time, the comprehensive event influencing factors provide a quantitative basis for the dynamic optimization of charging strategies, and can adjust the charging priority and charging amount in real time according to factors such as task urgency and road conditions, thereby improving the utilization efficiency of charging resources.

[0063] Optionally, acquiring the charging station information set, dynamically optimizing charging decisions based on the municipal vehicle task set, the dynamic power change information, and the charging station information set, and determining and outputting a preferred charging strategy may include:

[0064] Based on the dynamic power change information and the municipal vehicle task set, predict the remaining power change trend of the task execution vehicle during the subsequent task execution process, and determine whether the remaining power will fall below a preset safety threshold before completing the subsequent task;

[0065] If insufficient power is predicted, the charging strategy planning is executed:

[0066] Extracting the path topology and task time nodes of the current and subsequent tasks according to the municipal vehicle task set;

[0067] Determine the charging time window and the corresponding geographical location range of the vehicle during the task interval according to the path topology;

[0068] Within the geographical location range, screening a set of candidate charging stations that meet the distance constraint according to the charging station information set;

[0069] Based on the candidate charging station set, charging decisions are dynamically optimized to generate the preferred charging strategy.

[0070] Through this solution, a four-layer collaborative optimization mechanism is utilized to ensure early intervention based on risk prediction based on dynamic power forecasting, generate charging windows that can be embedded in task processes through the fusion of time and space constraints, screen candidate charging stations with distance constraints to avoid detour delays, and make multi-objective decisions to prioritize tasks with the shortest time while taking into account charging efficiency, thus achieving seamless connection between charging behavior and municipal tasks. This dynamic optimization solution is an important technical hub for balancing the dual rigid demands of zero tolerance for public safety and efficient use of municipal resources.

[0071] Optionally, dynamically optimizing charging decisions based on the set of candidate charging stations to generate the preferred charging strategy includes:

[0072] Based on the set of candidate charging stations, charging feasibility verification is performed on each candidate charging station:

[0073] Predict the arrival time at the charging station based on the vehicle's current location, driving speed, and the location of candidate charging stations;

[0074] Determining the remaining power based on the dynamic power change information;

[0075] Obtaining the real-time available number of charging piles and the charging rate of each pile at the charging station according to the charging station information set;

[0076] Determine a maximum charging duration based on the charging time window and the arrival time;

[0077] Determining a maximum amount of charge that can be replenished based on the single-pile charging rate and the maximum charging time;

[0078] Verify whether the total power after charging is sufficient for the entire subsequent task;

[0079] The total amount of electricity after the charging is completed is the sum of the remaining amount and the maximum amount of electricity that can be replenished;

[0080] Adding the verified candidate charging station to a set of feasible charging stations;

[0081] From the set of feasible charging stations, preferentially selecting the charging station that minimizes the total time taken for the vehicle mission as the comprehensive optimal target charging station;

[0082] If the time consumption is the same, the charging station with the highest power redundancy after charging is selected as the comprehensive optimal target charging station;

[0083] According to the comprehensive optimal target charging station, a charging route is determined and the preferred charging strategy is generated.

[0084] This solution precisely matches charging times with task intervals to avoid task delays due to charging, ensuring that municipal vehicles complete public service tasks as planned, such as timely garbage removal and rapid emergency vehicle response. Dynamic calculations can also replenish the maximum amount of power, avoiding energy waste caused by overcharging while ensuring sufficient power, reducing idle driving or secondary charging requirements due to insufficient power, and lowering overall energy consumption. Finally, decisions are made based on the real-time status of charging stations (number of available charging piles, charging rate), enabling dynamic allocation of charging resources, alleviating congestion at charging stations during peak hours, and improving the utilization efficiency of urban energy infrastructure.

[0085] Optionally, the method further includes:

[0086] After the task execution vehicle completes charging at the target charging station, the V2G reverse power supply strategy is executed:

[0087] Monitor whether the vehicle's current battery level has reached the preset full charge threshold and detect whether there is a surplus charging time window;

[0088] If the vehicle's power reaches the full charge threshold and there is a surplus charging time window, the grid load data is obtained in real time;

[0089] When the grid load data is higher than a preset threshold, the amount of reverse power supply is determined based on the remaining vehicle stay time and the maximum discharge power of the battery;

[0090] According to the reverse power supply amount, a discharge request is sent to the power grid dispatching system, and the vehicle is controlled to reversely supply power to the power grid through the charging pile.

[0091] Through this solution, idle electricity after recycling and charging can be used to achieve the secondary utilization of on-board energy, reduce carbon dioxide emissions, and improve the utilization rate of clean energy in the city; municipal vehicles can obtain social benefits through reverse power supply and reduce charging costs. The power grid and charging stations can reduce the investment in energy storage facility construction by dispatching vehicle energy storage resources, and improve the operational efficiency of the energy system; strengthen the vehicle-grid interaction capability, provide dynamic energy storage support for smart cities, promote the transformation of new energy vehicles from "electricity terminals" to "energy nodes", and help improve the level of digital energy management in cities. BRIEF DESCRIPTION OF THE DRAWINGS

[0092] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, a brief introduction will be given below to the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative labor.

[0093] Figure 1A schematic diagram of an application scenario provided in one embodiment of the present application;

[0094] Figure 2 This is a flowchart of a new energy charging management method for a smart city provided in one embodiment of the present application. DETAILED DESCRIPTION

[0095] To make the purpose, technical solutions, and advantages of the embodiments of the present application clearer, the technical solutions in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of this application.

[0096] In this document, the term "and / or" simply describes a relationship between related objects, indicating that three possible relationships exist. For example, "A and / or B" can represent: A exists alone, A and B exist simultaneously, or B exists alone. Furthermore, the character " / " in this document, unless otherwise specified, generally indicates an "or" relationship between the related objects.

[0097] The embodiments of the present application are described in further detail below with reference to the accompanying drawings.

[0098] However, the existing charging management methods for municipal new energy vehicles have significant limitations and are unable to predict and match the power consumption generated by municipal vehicles during their missions, resulting in inefficient use of power resources and increased operating costs.

[0099] Based on this, the present application provides a new energy charging management method for a smart city. Collect the operating characteristic data of municipal vehicles, divide the vehicle types and identify their drive types accordingly, obtain the municipal vehicle task list, monitor the dynamic parameters of the task vehicle during execution according to the drive type, use multimodal fusion technology to analyze the correlation between these dynamic parameters and real-time power consumption data, derive the dynamic change law of power, collect charging station information, and dynamically generate the optimal charging strategy based on the task list, power change law and charging station information, and output it to the vehicle driver. This solution accurately predicts dynamic power change information through a multimodal fusion strategy, greatly reduces the power prediction error rate of the task vehicle, and significantly improves the task security. By further combining the municipal vehicle drive type set and the charging station information set, triple data collaborative optimization is achieved in the dynamic optimization charging decision, so that the task level ensures minimal conflict in the charging period, promotes the utilization rate of charging resources and charging station utilization, and effectively reduces duplicate investment in municipal infrastructure.

[0100] Figure 1This is a schematic diagram of an application scenario provided by this application. In the process of municipal vehicle charging management, the method provided by this application can dynamically match the optimal charging station according to the changes in vehicle power during the execution of municipal vehicles, thereby improving the utilization rate of charging resources.

[0101] Specifically, the method of the present application is applied to any server, which communicates with the municipal vehicle management platform, the municipal task scheduling system and the charging pile data monitoring platform, and obtains the municipal vehicle operation feature data set provided by the municipal vehicle management platform, the municipal vehicle task set provided by the municipal task scheduling system and the charging pile information set provided by the charging pile data monitoring platform through the server, collects municipal vehicle operation feature data, divides vehicle types and identifies their drive types based on this, obtains the municipal vehicle task list, monitors the dynamic parameters of the task vehicle during execution according to the drive type, uses multimodal fusion technology to analyze the correlation between these dynamic parameters and real-time power consumption data, derives the dynamic change law of power, collects charging station information, and dynamically generates the optimal charging strategy based on the task list, power change law and charging station information, and outputs it to the vehicle driver.

[0102] For specific implementation methods, please refer to the following embodiments.

[0103] Figure 2 This is a flow chart of a new energy charging management method for a smart city provided by an embodiment of this application. The method of this embodiment can be applied to the server in the above scenario. Figure 2 As shown, the method includes:

[0104] S201. Obtain a municipal vehicle operation characteristic data set, classify municipal vehicles into types based on the municipal vehicle operation characteristic data set using a vehicle type classification strategy, and determine a municipal vehicle drive type set.

[0105] The municipal vehicle operation characteristic dataset refers to a multi-dimensional data set generated by municipal vehicles (such as sanitation vehicles, patrol vehicles, and buses) during operation, including but not limited to the vehicle's real-time location, driving speed, load status, and battery health status. The data comes from the municipal vehicle management platform.

[0106] The vehicle type classification strategy may refer to a hierarchical classification rule based on factors that determine the vehicle's power consumption rate and mission characteristics.

[0107] The municipal vehicle driving type set may be an output result of a vehicle set label grouped by type based on factors that determine the vehicle power consumption rate.

[0108] Specifically, traditional smart city charging management often ignores the differentiated operating characteristics of municipal vehicles and adopts a "one-size-fits-all" charging management model based on a single dimension of current remaining power. For example, garbage trucks need to start and stop frequently and have large load fluctuations. Their power consumption patterns are completely different from those of road sweepers. If the operating feature dataset is not obtained and classified according to the vehicle power consumption rate, it will be impossible to accurately identify the charging needs of high-energy-consuming vehicles. Vehicles may stop midway due to insufficient power, affecting the continuity of municipal services. Using the same charging strategy for municipal vehicles with different power consumption rate factors will lead to overcharging of municipal vehicle batteries, shortening battery life, and municipal vehicles may delay their tasks due to insufficient charging power. This step collects and analyzes raw operating data from the municipal vehicle management platform database through the classification of the operating feature dataset, extracts key features such as real-time power and load weight, and based on the pre-processed data, uses an algorithm to classify vehicles according to the power consumption rate factor. The generated municipal vehicle drive type set is generated by accurately matching vehicle tasks with charging needs, reducing task interruptions caused by insufficient power, and solving the "supply and demand mismatch" problem under the traditional management model.

[0109] S202 , obtaining a municipal vehicle task set, and tracking a dynamic task variable set of a task execution vehicle in the municipal vehicle drive type set during task execution according to the municipal vehicle drive type set.

[0110] The municipal vehicle task set can refer to a list of tasks assigned by the municipal department (such as garbage collection routes and police patrol areas), and the data comes from the municipal task scheduling system.

[0111] A dynamic task variable set may refer to a set of parameters that change in real time during task execution.

[0112] Specifically, the power consumption rate of municipal vehicles during task execution is dynamic. For example, the operating speed and spraying volume of sprinkler trucks in different sections of the road will be adjusted according to weather conditions and road conditions, which directly affects the power consumption rate. If dynamic task variables are not tracked, the current power consumption status of the vehicle will not be grasped in real time. When charging station resources are tight during the peak task period, charging resources cannot be reserved for high-priority task vehicles. In addition, static task planning cannot adapt to emergencies. For example, if the task variables are not updated for temporarily added emergency task vehicles, their charging needs may be ignored, resulting in task delays. This step tracks dynamic task variables in real time, performs data labeling and normalization on the dynamic variables, and generates a dynamic task variable set. When the power consumption rate of municipal vehicles of different drive types changes dynamically, it can predict their power consumption trends, avoid interruptions to operations due to insufficient power, and respond to municipal task adjustments and charging resource changes in real time to ensure the rapid execution of emergency tasks and enhance the flexibility and emergency response capabilities of smart city management.

[0113] S203 : Analyze the correlation between the dynamic task variable set and the real-time power consumption data according to the dynamic task variable set through a multimodal fusion strategy to determine dynamic power change information.

[0114] The multimodal fusion strategy can refer to a joint analysis algorithm that integrates spatiotemporal modalities (driving distance, driving direction), environmental modalities (temperature, wind speed), and mission modalities (start-stop frequency, load changes).

[0115] The real-time power consumption data may be the real-time power consumption of municipal vehicles while performing tasks.

[0116] The dynamic power change information may be output as a power consumption prediction curve based on task evolution.

[0117] Specifically, traditional charging management makes charging decisions solely based on the remaining battery charge, without analyzing the dynamic changes in factors affecting the vehicle's battery consumption rate during task execution. For example, a maintenance vehicle with the same remaining battery charge may experience significant differences in its actual range when driving on congested roads during peak hours versus on unobstructed roads during off-peak hours. Failure to analyze dynamic battery changes can lead to vehicle breakdowns during task execution due to misjudgment of range. For example, if a vehicle is misjudged as able to complete the remaining task in a congested section, but in reality loses power midway, it will not be able to optimize charging timing. For example, a road sweeper has brief idle periods between tasks. Without considering its dynamic battery change prediction, it may miss charging during off-peak electricity price periods, increasing charging costs. This step uses a multimodal fusion strategy to simultaneously acquire a dynamic task variable set and real-time battery consumption data. Task variables (such as driving speed) within the same time interval are then associated with battery consumption data to generate dynamic battery change information, including battery consumption prediction curves for different task scenarios. This accurately predicts the changes in factors affecting the vehicle's battery consumption rate during different task executions, providing dynamic data support for charging decisions.

[0118] S204: Obtain a charging station information set, dynamically optimize charging decisions based on the municipal vehicle task set, the dynamic power change information, and the charging station information set, and determine and output an optimal charging strategy.

[0119] The charging station information set may include the location of the charging pile, idle status, rate period, maximum output power, etc. The data comes from the charging pile data monitoring platform.

[0120] The optimal charging strategy can output the optimal matching solution of vehicle-charging pile-time period (such as patrol car A uses the No. 7 charging pile of station B for fast charging from 14:00 to 14:30).

[0121] Specifically, urban charging resources are unevenly distributed and change in real time. Without combining charging station information sets with dynamic vehicle task optimization strategies, vehicles may travel long distances to charging stations only to find no idle charging piles. This increases ineffective driving energy consumption and time costs, and fails to balance the load on charging stations, causing some charging stations to overload and shorten equipment lifespan, while other charging station resources remain idle. This solution achieves a three-dimensional dynamic matching of "task-power-charging resources" by integrating task, power, and charging station information in real time. For example, for vehicles about to complete a garbage collection task, idle fast-charging stations near the task endpoint are planned in advance, and the optimal charging strategy is output to the vehicle driver, reducing waiting time and driving energy consumption. Dynamic optimization of charging decisions enables intelligent matching of vehicles and charging stations, improves the utilization rate of idle charging piles at charging stations, reduces ineffective driving mileage of vehicles searching for charging piles, and reduces urban transportation carbon emissions.

[0122] This solution collects operational characteristic data for municipal vehicles, classifies them into different types, identifies their drive types, and obtains a list of municipal vehicle tasks. Dynamic parameters during the execution of task vehicles are monitored based on their drive types. Multimodal fusion technology is used to analyze the correlation between these dynamic parameters and real-time power consumption data, deriving patterns of dynamic power changes. Charging station information is collected, and the optimal charging strategy is dynamically generated by integrating the task list, power change patterns, and charging station information. This solution accurately predicts dynamic power change information through a multimodal fusion strategy, significantly reducing the error rate in power prediction for task vehicles and significantly improving task assurance. By further combining the municipal vehicle drive type set and the charging station information set, triple data collaborative optimization is achieved in the dynamic optimization of charging decisions, ensuring minimal conflict in charging periods at the task level, driving increased utilization of charging resources and charging stations, and effectively reducing duplicate investment in municipal infrastructure.

[0123] In some embodiments, based on the vehicle type classification strategy, the main causes of energy consumption fluctuations in the municipal vehicle operation characteristic data set are extracted; the main causes of energy consumption fluctuations include changes in physical carriers, changes in environmental factors, and changes in event factors; if the main cause of the energy consumption fluctuation of the current municipal vehicle is the change in physical carriers, then the current municipal vehicle is determined to be an entity-driven vehicle; if the main cause of the energy consumption fluctuation of the current municipal vehicle is the change in environmental factors, then the current municipal vehicle is determined to be an environment-driven vehicle; if the main cause of the energy consumption fluctuation of the current municipal vehicle is the change in event factors, then the current municipal vehicle is determined to be an event-driven vehicle; based on entity-driven vehicles, environment-driven vehicles, and event-driven vehicles, a municipal vehicle driving type set is generated.

[0124] The main cause of energy consumption fluctuation can refer to the core factors that lead to changes in municipal vehicle energy consumption.

[0125] The change in physical carrier can be the energy consumption fluctuation caused by the change in the vehicle's own physical state (such as the increase or decrease in weight other than the vehicle body).

[0126] The variation of environmental factors can be the impact of the external environment on energy consumption (such as weather and temperature).

[0127] The change in event factors can be sudden changes in energy consumption caused by sudden task events (such as emergency dispatch and health emergencies).

[0128] Physically driven vehicles can be vehicles whose energy consumption fluctuations are mainly dominated by changes in the physical carrier (such as garbage trucks).

[0129] Environmentally driven vehicles can be vehicles whose energy consumption fluctuations are mainly dominated by environmental factors (such as new energy sprinkler trucks).

[0130] An event-driven vehicle can be a vehicle whose energy consumption fluctuations are mainly driven by mission events (such as a fire patrol vehicle).

[0131] Specifically, this vehicle type classification strategy is based on three core contradictions in smart city municipal vehicle energy management: First, the contradiction of vehicle heterogeneity leads to essential differences in the energy consumption mechanisms of dozens of vehicle models. If a unified charging and discharging strategy is adopted, it will cause resource mismatch (such as event-driven fire trucks interrupting emergency tasks due to fixed charging cycles) and energy waste (such as environmentally driven buses failing to predict the efficiency drop in low temperature environments, resulting in 40% charging redundancy); second, traditional static classification cannot solve the problem of task-energy consumption decoupling, and it is even more difficult to cope with scenarios where multiple factors are superimposed (such as the combined influence of environmental and event factors will trigger a surge in energy consumption); finally, the contradiction in charging facility coordination causes load spikes (the centralized charging of event-driven vehicles increases the risk of grid overload) and resource idleness (low-priority vehicles occupy high-demand resources) when classification is missing. In response to the above problems, this step first extracts physical carrier indicators (such as real-time load), environmental indicators (such as temperature and humidity sensor data) and event indicators (such as the number of task plan changes) from the municipal vehicle operation feature data set; then dynamically classifies each municipal vehicle according to the dominant factor judgment rule: if the variance of the physical carrier indicator change is greater than the variance of the environmental and event indicators, and the correlation coefficient between the load change and the energy consumption fluctuation is greater than the set value (such as ≥0.7) (for example, when a garbage truck is fully loaded, the energy consumption increase is 3 times the environmental impact), then it is judged to be entity-driven; if the variance of the environmental indicator change exceeds If the variance of physical and event indicators, and the correlation coefficient between environmental indicators and energy consumption fluctuations is greater than or equal to a set value (such as ≥0.6) (for example, the power consumption per unit distance of a snowplow increases by 45% at -5°C), it is judged to be environment-driven; if the synchronization rate of the suddenness of event indicators (such as the frequency of task changes) and energy consumption fluctuations is greater than a set ratio (such as >80%) (for example, the power consumption of an emergency command vehicle surges due to temporary dispatch), it is judged to be event-driven; finally, a structured municipal vehicle drive type set is generated based on the classification results, and vehicle IDs are aggregated into three categories: entity-driven, environment-driven, and event-driven.

[0132] Through this solution, by accurately identifying the dominant factors of energy consumption, we can achieve accurate classification of municipal vehicles, laying an important foundation for subsequent charging strategies; based on entity-driven vehicles, environment-driven vehicles and event-driven vehicles, we generate a set of municipal vehicle drive types, which improves the accuracy of charging strategies, significantly reduces the error in charging volume estimation, and reduces the interruption rate of municipal task execution, ensuring the reliability of municipal services.

[0133] In some embodiments, the municipal vehicle task set includes operation characteristics, path characteristics and vehicle status characteristics; according to the municipal vehicle drive type set, the changes in the main causes of energy consumption fluctuations of the current municipal vehicle during the task execution are tracked, and the change information of the main causes of energy consumption fluctuations is determined; according to the change information of the main causes of energy consumption fluctuations, the operation characteristics, path characteristics and vehicle status characteristics are feature mapped and dynamically processed to generate a dynamic load operation feature set, an environmental response situation feature set and an event-driven task feature set; the dynamic load operation feature set includes real-time entity task load and load state change rate; the environmental response situation feature set includes real-time natural environment data set and real-time artificial environment data set; the event-driven task feature set includes task urgency, task implementation intensity and task target location; based on the dynamic load operation feature set, the environmental response situation feature set and the event-driven task feature set, a dynamic task variable set is generated.

[0134] The municipal vehicle task set may refer to a multi-dimensional feature set involved when a municipal vehicle performs a task, including operation content characteristics, driving path characteristics, and vehicle operating status characteristics.

[0135] Job characteristics can be parameters that describe the inherent properties of the task itself (such as task intensity).

[0136] Path features can be route-related parameters (such as travel distance) during task execution.

[0137] The vehicle status characteristics may be real-time operating parameters of the vehicle (such as real-time power consumption data, availability status).

[0138] The information on changes in the main causes of energy consumption fluctuations can be dynamic change data that characterizes the core factors (physical carrier / environmental / event factors) that cause energy consumption fluctuations of municipal vehicles during task execution.

[0139] The dynamic load operation feature set can be a set that describes the dynamic characteristics of a physical-driven vehicle caused by changes in physical load, including the real-time physical task load (such as the real-time loading weight of a garbage truck) and the load state change rate (the rate of increase or decrease of the load per unit time).

[0140] The environmental response situation feature set can be the dynamic response characteristics of the environment-driven vehicle to the external environment, including real-time natural environment data sets (such as temperature, weather, road adhesion coefficient) and real-time artificial environment data sets (such as traffic congestion index, road surface material).

[0141] The event-driven task feature set can be the dynamic features of an event-driven vehicle triggered by an emergency task, including the task urgency (such as the fire truck's alarm level), the task implementation intensity (the operating indicators to be achieved) and the task target location (destination coordinates).

[0142] The real-time entity task load may be the actual task load of current municipal vehicles.

[0143] The load state change rate may be a municipal vehicle load change rate per unit time.

[0144] The real-time natural environment dataset can be the natural environment information (such as temperature, rainfall intensity, and road icing index) when municipal vehicles are performing tasks.

[0145] Real-time human environment datasets can be human environment information (such as traffic congestion index and waiting time at traffic lights) during the execution of tasks by municipal vehicles.

[0146] Task urgency can be a dynamic indicator to quantify the timeliness requirements of event-driven tasks. The higher the value, the lower the tolerance for response delay.

[0147] Task implementation intensity can be a technical indicator of the work intensity required to complete a specific event-driven task, reflecting the energy consumption benchmark.

[0148] The mission target location can be the spatial coordinates of the event-driven mission execution terminal, which determines the vehicle path planning and mileage energy consumption.

[0149] The dynamic task variable set can be a spatiotemporal matrix that integrates three types of dynamic characteristics: physical carrier, environment, and event, and represents the evolution process of energy consumption driving factors when municipal vehicles perform tasks.

[0150] Specifically, given that municipal vehicles face three core causes of energy consumption fluctuations during the execution of dynamic tasks, namely changes in physical carriers (such as a sudden increase in the load of garbage trucks), sudden changes in the environment (such as snowplows encountering blizzards), and event triggering (such as fire trucks responding to emergency calls), traditional static management solutions are unable to capture the synergistic impact of multi-source heterogeneous data in real time (manifested as increased energy consumption prediction deviation when the environment and events interact), lack a dynamic mapping mechanism of the characteristics of operations, paths, and vehicle states (such as the failure of the strategy when a sprinkler truck is affected by the combined effects of weather and temporary tasks), and face the risk of scheduling interruption due to decision-making lags (a surge in load change rate does not trigger charging compensation). To address the above issues, this step tracks the main factors causing energy consumption fluctuations in three types of vehicles in real time. For physical-driven vehicles (such as garbage trucks), cargo compartment weight sensor data is monitored. For environmental-driven vehicles (such as water trucks), real-time natural environmental conditions are obtained through onboard weather stations, and real-time human environmental information is obtained through onboard GPS. For event-driven vehicles (such as ambulances), task priority codes are received from the command center. When the priority reaches a certain level, a response to changes in event factors is immediately triggered. Subsequently, multi-dimensional features are dynamically reconstructed. Based on the load data stream, the real-time physical task load (current weight / maximum load) and load state change rate (percentage of weight change per minute) are calculated. Rainfall intensity and incline meter data from the meteorological API are integrated to generate a real-time natural environment dataset. Real-time vehicle speed and intersection waiting time from the traffic platform are integrated to form a real-time human environment dataset. The KPI fields in the task instructions are parsed to quantify the task urgency (time countdown ratio), task achievement intensity (the required equipment power value), and task target location (coordinate hash code). Finally, the dynamic load operation feature set, environmental response status feature set, and event-driven task feature set are aligned by timestamp to construct a dynamic task variable set.

[0151] Through this solution, the energy consumption management of municipal vehicles is significantly improved by constructing three types of collaborative dynamic feature sets. The dynamic load feature set accurately quantifies the changes in physical carriers, greatly improving the energy consumption prediction accuracy of physical vehicles when the load suddenly changes; the environmental response feature set integrates natural and man-made environmental data, effectively optimizing the response speed of environmental vehicles in complex scenarios; the event task feature set quickly analyzes key task indicators, significantly enhancing the reliability of event vehicles in responding to emergency dispatch.

[0152] In some embodiments, if the vehicle type is an entity-driven vehicle, the dynamic load operation feature set in the dynamic task variable set is extracted, and combined with the real-time power consumption data, the gain / decrease effect of the real-time entity task load and load state change rate on the power consumption rate is analyzed to generate an entity change impact sub-strategy; if the vehicle type is an environment-driven vehicle, the environmental response situation feature set in the dynamic task variable set is extracted, and combined with the real-time power consumption data, the gain / decrease effect of the real-time natural environment data set and the real-time artificial environment data set on the power consumption rate is analyzed to generate an environment change impact sub-strategy; if the vehicle type is an event-driven vehicle, the event-driven task feature set in the dynamic task variable set is extracted, and combined with the real-time power consumption data, the gain / decrease effect of the task urgency, task implementation intensity and task target position on the power consumption rate is analyzed to generate an event-focused impact sub-strategy; based on the entity change impact sub-strategy, the environment change impact sub-strategy and the event-focused impact sub-strategy, a multimodal fusion strategy is generated.

[0153] The power consumption rate can be the value of electric energy consumed by the vehicle per unit time, reflecting the real-time energy consumption intensity of the vehicle. It is the core quantitative indicator of dynamic power change information.

[0154] The gain / loss effect can refer to the positive promotion (gain) or negative inhibition (loss) of the dynamic task variable on the power consumption rate, such as an increase in load leading to an increase in the energy consumption rate (gain) and a downhill section reducing energy consumption (loss).

[0155] The entity change impact sub-strategy can be an energy consumption prediction strategy specifically for entity-driven vehicles. By establishing a mathematical mapping relationship between the real-time entity task load, load state change rate and power consumption rate, the load-energy consumption gain / loss coefficient is output.

[0156] The environmental change impact sub-strategy can be an energy consumption compensation strategy specifically for environment-driven vehicles. By analyzing the coupling effect of natural environment data and artificial environment data, it outputs a correction value of the power consumption rate under the comprehensive influence of the environment.

[0157] The event-focused impact sub-strategy can be an energy consumption weighting strategy specifically for event-driven vehicles. It outputs an event-driven power consumption rate adjustment factor by quantifying the combined impact of task urgency, implementation intensity, and target location on energy consumption.

[0158] The multimodal fusion strategy can be a collaborative decision-making framework that integrates three types of sub-strategies: entity changes, environmental changes, and event focus. It dynamically calls the corresponding sub-strategy according to the vehicle type and outputs unified dynamic power change information.

[0159] Specifically, given the significant fundamental differences in the energy consumption mechanisms of municipal vehicles—the energy consumption peaks of physical-driven vehicles are dominated by physical loads, while those of environmentally driven vehicles are sensitive to the natural and man-made environments, and those of event-driven vehicles dynamically jump with the urgency of the mission—traditional single energy consumption models are unable to adapt to this heterogeneity, which can easily lead to charging strategies seriously deviating from actual needs (such as insufficient power reserve for high-load vehicles or overcharging of idle vehicles). In addition, sudden changes in municipal missions (such as emergency rescue) require charging strategies to have real-time reconstruction capabilities. To address the above issues, this step first determines the vehicle type based on the following: if it is a physical-driven vehicle, a dynamic load operation feature set (including real-time physical task load and load state change rate) is extracted; if it is an environmentally driven vehicle, an environmental response situation feature set (including real-time natural and man-made environment datasets) is extracted; if it is an event-driven vehicle, an event-driven task feature set (including task urgency, implementation intensity, and target location) is extracted. Then, a typified power consumption impact analysis is performed. For physical-driven vehicles, the impact of task load on power consumption rate is analyzed (e.g., a 10% increase in load increases energy consumption by 8%). The sub-strategy for the impact of entity changes is generated by analyzing the detrimental effect of the load change rate (such as a 5% decrease in energy consumption when the load is stable). For environment-driven vehicles, the superimposed effects of the natural environment (such as a 12% increase in energy consumption when the temperature is >30°C) and the man-made environment (such as a 15% increase in energy consumption when the congestion index is >0.7) are analyzed to generate a sub-strategy for the impact of environmental changes. For event-driven vehicles, the comprehensive impact of the mission urgency (such as a 20% increase in energy consumption for emergency missions), implementation intensity (such as a 10% increase in energy consumption for high-intensity missions), and target location (such as a 5% increase in energy consumption for every 1 km increase in distance) is quantified to generate a sub-strategy focusing on event impacts.

[0160] This solution generates dynamic sub-policies by type through a multimodal fusion mechanism, and can directly call preset strategy modules to achieve rapid response in emergency task scenarios, avoiding the time delay of global model retraining. At the same time, relying on precise energy consumption gain / loss analysis, it optimizes vehicle charging priority allocation under the constraint of limited charging resources, effectively avoiding charging station conflicts and grid overload risks, thereby systematically supporting the economy and operational reliability of the municipal charging network.

[0161] In some embodiments, based on the real-time entity task load and the load state change rate, the gain nodes and deduction nodes on the path of the task execution vehicle are identified; the gain node represents the node where the real-time entity task load generates a gain on the power consumption rate; the deduction node represents the node where the real-time entity task load generates a deduction on the power consumption rate; based on the gain node and the deduction node, combined with the load state change rate, the changing trend of the power consumption rate is predicted, and the dynamic power change information of the entity-driven vehicle is determined.

[0162] Gain nodes are path locations (such as garbage disposal stations and material distribution points) where the power consumption rate decreases due to a reduction in load (such as garbage dumping and cargo unloading).

[0163] Debt nodes are path locations where the rate of power consumption increases due to increased load (such as garbage loading, cargo loading) (such as residential garbage collection points and warehouse loading areas).

[0164] Specifically, during the execution of tasks of entity-driven vehicles (such as garbage trucks and logistics distribution vehicles), the dynamic changes in the load of the entity task are the core variables that affect power consumption (such as the amount of garbage carried by the garbage truck in real time). Traditional charging management strategies ignore the temporal and spatial heterogeneity of load (such as the load difference of the same vehicle at different stages leading to errors in endurance), cannot respond to sudden task fluctuations (such as the temporary addition of task points), and have blind spots affected by path nodes (such as the unidentified unloading point leading to a reduction in power consumption rate), resulting in distorted power prediction and causing task interruption or charging redundancy. To address the above problems, this step first uses a dynamic path scanning algorithm to input the real-time position and task path of the vehicle and call the historical task database to match the location of typical load events. Based on the real-time entity task load curve, the load drop points are detected and marked as gain node candidates, and the load rise points are detected and marked as loss node candidates. Then, the GIS map information is integrated to screen the candidate points within a certain range of the municipal facilities to confirm the final node coordinates; then the change trend prediction stage is entered, the current load state change rate is extracted (such as the load change slope within 30 seconds) and the path distance to the next node is calculated to determine the expected load change time window, and the power consumption rate is established. The test model is used (for example, the increase in power consumption rate at the gain node is calculated by loading amount × 0.15kWh / kg, and the decrease in power consumption rate at the depreciation node is calculated by unloading amount × 0.18kWh / kg). After combining with the vehicle dynamics parameter correction, a power consumption rate change trend graph with path distance as the horizontal axis is generated. Finally, in the dynamic power information generation stage, the trend graph is converted into dynamic power change information in the form of an energy consumption curve, and the power consumption rate value and direction at the key nodes are marked (for example, "coordinate Y point: the rate drops from 1.2kWh / min to 0.7kWh / min"), and the vehicle's remaining power and the remaining mission mileage are superimposed to generate an endurance attenuation map.

[0165] This solution utilizes real-time physical task load and load state change rate to identify gain and loss nodes on the task execution vehicle's path. Through load-path position linkage, change rate-driven modeling, and dynamic charging decision-making fulcrum generation, the "physical load dynamics → power consumption response → charging strategy generation" closed loop is established. Through precise identification of gain / loss nodes and dynamic fusion of load change rate, the prediction error of the power consumption rate of physical-driven vehicles is reduced, thereby reducing the risk of power exhaustion mid-task.

[0166] In some embodiments, for environment-driven vehicles, an environmental change impact sub-strategy is adopted: analyze the real-time natural environment data set, take the relevant factors that produce gains on the power consumption rate in the natural environment as natural environment positive correlation factors, and take the relevant factors that produce reductions on the power consumption rate in the environment as natural environment negative correlation factors; analyze several natural environment positive correlation factors, perform gain superposition on several natural environment positive correlation factors, and determine a first natural environment impact coefficient; analyze several natural environment negative correlation factors, perform reduction superposition on several natural environment negative correlation factors, and determine a second natural environment impact coefficient; normalize the first natural environment impact coefficient and the second natural environment impact coefficient according to the first natural environment impact coefficient and the second natural environment impact coefficient to determine a natural environment response intensity coefficient; analyze the real-time artificial environment data set, take the relevant factors that produce gains on the power consumption rate in the artificial environment as artificial environment positive correlation factors, and take the relevant factors that produce reductions on the power consumption rate in the environment as artificial environment negative correlation factors. The related factors with reduced benefits are taken as negative related factors of the human environment; several positive related factors of the human environment are analyzed, and gains of several positive related factors of the human environment are superimposed to determine the first human environment impact coefficient; several negative related factors of the human environment are analyzed, and reduction of gains of several negative related factors of the human environment are superimposed to determine the second human environment impact coefficient; according to the first human environment impact coefficient and the second human environment impact coefficient, the first human environment impact coefficient and the second human environment impact coefficient are normalized to determine the human environment response intensity coefficient; the real-time human environment data set is analyzed to determine the characteristics of the infrastructure, and according to the characteristics of the infrastructure, several infrastructure approval time intervals with different time spans are set; according to the real-time power consumption data, the influence of the superposition of the natural environment response intensity coefficient and the human environment response intensity coefficient on the power consumption rate in each infrastructure approval time interval is analyzed, the changing trend of the power consumption rate is predicted, and the dynamic power change information of the environment-driven vehicle is determined.

[0167] Positively correlated factors of the natural environment can be physical environmental parameters that lead to an increase in the rate of power consumption (such as high slopes and strong headwinds).

[0168] Negatively correlated factors of the natural environment can be physical environmental parameters that reduce the rate of power consumption (such as tailwind, downhill).

[0169] The first / second natural environment impact coefficient may be a quantified superimposed gain and loss value of all positive / negative correlation factors on the power consumption rate.

[0170] The natural environment response intensity coefficient can be a normalized comprehensive environmental impact index in the range of [-1,1]. A positive value indicates that the natural environment accelerates the rate of power consumption, and a negative value indicates that the natural environment slows down the rate of power consumption.

[0171] Positively correlated factors of the human environment can be human factors that increase energy consumption (such as congestion index > 0.7, waiting time at a red light > 60 seconds); human factors that increase energy consumption (such as congestion index > 0.7, waiting time at a red light > 60 seconds).

[0172] Negatively correlated factors of the human environment can be human factors that reduce energy consumption (such as green wave sections and low traffic periods).

[0173] The first / second human environmental impact coefficient may be a coefficient that quantifies the gain and loss strength of the positive / negative human correlation factor, respectively.

[0174] The human-induced environmental response intensity coefficient may be a normalized human-induced environmental impact value in a range of [-1, 1], where a positive value indicates that the human-induced environment accelerates the rate of power consumption, and a negative value indicates that the human-induced environment slows down the rate of power consumption.

[0175] Infrastructure characteristics can be static properties of roads and surrounding facilities (such as road grade and road material).

[0176] The infrastructure approval time interval can be a typical time segment divided according to the characteristics of the infrastructure (such as a tarmac road that takes ten minutes to pass in five minutes).

[0177] Specifically, in the new energy charging management of smart cities, the power consumption of municipal vehicles is significantly affected by the natural environment (such as temperature and wind speed) and the man-made environment (such as traffic congestion and construction sections), and these factors have dynamic changing characteristics. If these environmental factors are ignored, the static charging strategy will lead to energy consumption prediction deviations, resulting in insufficient charging affecting the continuity of municipal tasks (such as emergency vehicle power outages delaying rescue) or overcharging reducing energy utilization efficiency. In addition, natural and man-made environmental factors superimpose on each other, and a single analysis is difficult to accurately reflect the actual situation of power consumption. Therefore, a multimodal fusion strategy must be used to comprehensively analyze the impact of various environmental factors on power consumption in order to achieve accurate charging decisions, ensure vehicle mission execution and optimize energy allocation. To address these issues, this step first acquires real-time natural and anthropogenic environmental data through onboard sensors and traffic platforms. Based on preset rules, these data are classified into positively correlated factors that exacerbate energy consumption (e.g., high temperatures and steep slopes) and negatively correlated factors that inhibit energy consumption (e.g., tailwinds and downhill slopes). Impact coefficients are then calculated for the natural environmental data—the gain strength of positively correlated factors is aggregated to generate a first natural environmental impact coefficient, and the loss strength of negatively correlated factors is aggregated to generate a second natural environmental impact coefficient. These coefficients are then normalized to obtain the natural environmental response intensity coefficient. Similar processing is performed on the anthropogenic environmental data, calculating the first and second anthropogenic environmental impact coefficients and normalizing them to output the anthropogenic response intensity coefficient. Infrastructure features (e.g., road grade and charging station density) are then retrieved from the GIS based on the vehicle's location and matched to a preset infrastructure approval time interval (e.g., morning rush hour corresponds to high congestion). Finally, within the selected time interval, the combined effects of the natural and anthropogenic environmental response intensity coefficients are analyzed (e.g., double positive coefficients indicate a significant increase in energy consumption). Based on this combined effect, the power consumption rate trend is predicted and dynamic power consumption information is output, including the expected energy consumption curve and inflection point location for the future time period.

[0178] This solution utilizes multi-factor collaborative analysis to classify positive and negative correlation factors in the natural and man-made environments, superimposes coefficients to quantify coupling effects, resolves prediction biases caused by environmental coupling effects, and reduces power consumption rate prediction errors. Through spatiotemporal dynamic calibration, regional adaptability is achieved by binding infrastructure to approved time intervals, and charging opportunities are dynamically recommended, enhancing the forward-looking nature of charging strategies. Relying on normalized decision support, multi-source data is transformed into a unified dimension, and a dynamic mapping relationship between the environment and energy consumption is constructed, providing necessary support for high-precision power prediction and forward-looking charging strategies.

[0179] In some embodiments, for event-driven vehicles, an event-focused impact sub-strategy is adopted: according to the urgency of the task, the task urgency index is determined, and the task urgency index is used as the first impact coefficient; according to the municipal vehicle task set, the current task information is determined, and according to the current task information, the task duration and task operation complexity of the corresponding task are determined; according to the preset numerical normalization strategy, the task duration and task operation complexity are normalized to determine the task duration index and task complexity index; the product of the task duration index and the task complexity index is used as the second impact coefficient; according to the task target location, the task arrival route is determined ; Based on the preset traffic remote sensing information, according to the task arrival route, determine the real-time road condition complexity and driving distance of the corresponding task arrival route; according to the preset numerical normalization strategy, normalize the real-time road condition complexity and driving distance to determine the road condition complexity index and driving distance index; take the product of the road condition complexity index and the driving distance index as the third influence coefficient; take the sum of the first influence coefficient, the second influence coefficient and the third influence coefficient as the comprehensive event influence factor; based on the comprehensive event influence factor and the real-time power consumption data, predict the changing trend of the power consumption rate and determine the dynamic power change information of the event-driven vehicle.

[0180] The task urgency index (first impact coefficient) may be a quantitative mapping value of the urgency of the task.

[0181] The current mission information may be a dataset describing the mission that the vehicle is currently performing.

[0182] The task duration can be the estimated time (in hours) from the start to the completion of the task.

[0183] Task operation complexity can be a grading indicator of the difficulty of task execution.

[0184] The second impact coefficient can be the impact value of the task's own characteristics on energy consumption, reflecting the coupling effect of duration and complexity.

[0185] The preset traffic remote sensing information may be a data set for dynamic monitoring of urban traffic, including a road congestion index, accident locations, etc.

[0186] The real-time traffic complexity can be a path difficulty score generated based on preset traffic remote sensing information.

[0187] The driving distance can be the path length between the mission target location and the vehicle's current location.

[0188] The preset numerical normalization strategy may be a standardization method for mapping parameters of different dimensions to the (0, 1) interval.

[0189] The third influence coefficient may be an influence value of the path characteristics on the energy consumption, reflecting the combined effect of the road condition and the distance.

[0190] The comprehensive event impact factor can be the synthetic impact value of event characteristics on the power consumption rate. A larger value indicates a higher risk of energy consumption surge.

[0191] Specifically, the operating characteristics of event-driven vehicles are essentially different from those of entity-driven and environment-driven vehicles. Their power consumption patterns are mainly determined by the attributes of the task. For example, when emergency rescue vehicles and emergency operation vehicles are performing tasks, the urgency of the task directly affects the vehicle's driving speed and equipment usage intensity, which in turn significantly changes the power consumption rate. If the influence of the task urgency index is ignored, the charging strategy may not be able to meet the power demand of the vehicle's sudden tasks, resulting in insufficient power and delaying the task; the coupling effect of task duration and operation complexity is also crucial. Long-term and highly complex tasks (such as continuous road construction) will cause vehicle equipment to continue to run at full load, and the power consumption rate will increase nonlinearly. If quantitative analysis is not performed through the task duration index and complexity index, it will be difficult to accurately predict the power changes in different task stages, which may result in waste or insufficient charging resources. To address the above problems, this step first extracts the urgency information of the current task from the municipal vehicle task set and converts it into a task urgency index as the first influence coefficient through normalization; then obtains the specific content of the current task from the task set to determine the task duration and task operation complexity, and then uses a preset normalization strategy to process the two separately to obtain the task duration index and task complexity index and multiply them to obtain the second influence coefficient; then uses the navigation system to generate the task arrival route according to the task target location, and at the same time obtains the real-time road condition complexity and driving distance of the route based on traffic remote sensing information, and then uses the normalization strategy to obtain the road condition complexity index and driving distance index and multiply them to obtain the third influence coefficient, and then adds the first, second and third influence coefficients to obtain a comprehensive event influence factor; finally, combines this factor with real-time power consumption data to analyze and predict the power consumption rate change trend of event-driven vehicles (for example, the higher the factor, the power consumption rate increases by 20%), thereby determining the dynamic power change information including the vehicle's subsequent task power energy consumption curve.

[0192] Through this solution, through quantitative analysis of multi-dimensional event factors, dynamic power change information can accurately reflect the power consumption patterns of event-driven vehicles in different mission scenarios, avoiding power shortages or waste caused by the disconnection between charging strategies and mission requirements; at the same time, the comprehensive event influencing factors provide a quantitative basis for the dynamic optimization of charging strategies, and can adjust the charging priority and charging amount in real time according to factors such as task urgency and road conditions, thereby improving the utilization efficiency of charging resources.

[0193] In some embodiments, based on dynamic power change information and the municipal vehicle task set, the remaining power change trend of the task-executing vehicle during the execution of subsequent tasks is predicted to determine whether the remaining power will be lower than a preset safety threshold before completing the subsequent tasks; if insufficient power is predicted, charging strategy planning is executed: based on the municipal vehicle task set, the path topology and task time nodes of the current and subsequent tasks are extracted; based on the path topology, the charging time window and the corresponding geographical location range of the vehicle in the task gap are determined; within the geographical location range, the candidate charging station set that meets the distance constraint is screened based on the charging station information set; based on the candidate charging station set, the charging decision is dynamically optimized to generate the optimal charging strategy.

[0194] The remaining power change trend may refer to a predicted change trajectory of the remaining power of the mission execution vehicle during the subsequent mission execution process.

[0195] The preset safety threshold may be the minimum power critical value at which the vehicle can safely perform a task.

[0196] The path topology can be the route network structure of the mission execution vehicle from the current location to the mission target location, including road hierarchy, intersections and connectivity relationships.

[0197] Task time nodes can be the planned start time, end time, and key milestone time points of task execution.

[0198] The rechargeable time window can be the time period during which charging is allowed between tasks, and must meet the task time node constraints.

[0199] The geographic location range may be a spatial area centered on the vehicle mission path and formed by a maximum detour distance (eg, 5 kilometers).

[0200] The distance constraint can be a threshold of the shortest achievable distance between the charging station and the task path (e.g., ≤ 3 kilometers).

[0201] The candidate charging station set may be a collection of available charging stations that meet geographic location range and distance constraints.

[0202] Specifically, municipal vehicles (such as fire trucks and ambulances) undertake key public services. If the power planning is inaccurate, it is easy to cause task interruption, which in turn causes the decline in the quality of municipal vehicle services. The traditional fixed charging strategy cannot adapt to the three dynamic characteristics of task suddenness (temporary tasks of event-driven vehicles), path uncertainty (real-time road conditions of environment-driven vehicles) and power consumption fluctuations (sudden changes in load of entity-driven vehicles). Therefore, there are inherent defects such as resource mismatch, redundant charging and lack of emergency response capabilities. In response to the above problems, this step first triggers the charging strategy planning based on the vehicle's subsequent task power consumption curve in the dynamic power change information when the predicted remaining power is lower than the minimum dynamic power threshold required to ensure that various municipal vehicles can complete the current task; then parse the municipal vehicle task set, extract the start and end positions / times of the current and next tasks, estimate the movement time based on the path topology and vehicle speed, deduct the movement time from the task gap time as the charging time window, and define the geographical location range of the dynamic radius that the current power can support the vehicle to reach with the path as the center; then filter out the charging station information set that meets the requirements of being located in the area. The system generates a set of candidate charging stations by sorting out the location range, the shortest distance to the path ≤ the distance constraint (such as 3 kilometers), and the number of real-time available charging piles > 0. Finally, the charging feasibility of each candidate station is verified one by one: the vehicle arrival time and the remaining power at arrival are predicted, the maximum amount of power that can be replenished is calculated based on the single-pile charging rate and the charging window, and it is verified that "the remaining power at arrival + the amount of power that can be replenished > the power required for the subsequent task". The passing stations are added to the feasible set, and the target charging station is selected based on the rule that the shortest total round-trip time is given priority, and the maximum power redundancy after charging is the same when the time is the same. The optimal charging strategy including the location of the charging station, charging time and the best route is output.

[0203] Through this solution, a four-layer collaborative optimization mechanism is utilized to ensure early intervention based on risk prediction based on dynamic power forecasting, generate charging windows that can be embedded in task processes through the fusion of time and space constraints, screen candidate charging stations with distance constraints to avoid detour delays, and make multi-objective decisions to prioritize tasks with the shortest time while taking into account charging efficiency, thus achieving seamless connection between charging behavior and municipal tasks. This dynamic optimization solution is an important technical hub for balancing the dual rigid demands of zero tolerance for public safety and efficient use of municipal resources.

[0204] In some embodiments, charging feasibility verification is performed on each candidate charging station based on the set of candidate charging stations: the arrival time at the charging station is predicted based on the current position of the vehicle, the driving speed and the position of the candidate charging station; the remaining amount of power at arrival is determined based on the dynamic power change information; the real-time available number of charging piles and the charging rate of a single pile of the charging station are obtained based on the charging station information set; the maximum charging time is determined in combination with the charging time window and the arrival time; the maximum amount of power that can be replenished is determined based on the charging rate of a single pile and the maximum charging time; it is verified whether the total amount of power after charging is completed meets the full requirements of subsequent tasks; the total amount of power after charging is completed is the sum of the remaining amount of power at arrival and the maximum amount of power that can be replenished; the candidate charging stations that pass the verification are added to the set of feasible charging stations; from the set of feasible charging stations, the charging station that makes the total time of the vehicle task the shortest is preferentially selected as the comprehensive optimal target charging station; if the time is the same, the charging station with the highest power redundancy after charging is selected as the comprehensive optimal target charging station; based on the comprehensive optimal target charging station, the charging route is determined and the optimal charging strategy is generated.

[0205] Charging feasibility verification can be a quantitative evaluation process of whether a candidate charging station can meet the vehicle charging needs.

[0206] The arrival time can be the estimated driving time from the vehicle's current location to the charging station, calculated based on the vehicle's driving speed and path topology.

[0207] The arrived remaining power may be the predicted remaining power when the vehicle arrives at the charging station, which is derived based on modeling of dynamic power change information.

[0208] The real-time available number of charging piles may be the number of charging piles currently available at the charging station, which is derived from real-time updated data of the charging station information set.

[0209] The single-pile charging rate can be the amount of charge per unit time of a single charging pile, which is determined by the hardware parameters of the charging station (charging station information set).

[0210] The maximum charging time may be the net charging time after deducting the arrival time from the charging time window.

[0211] The maximum amount of electricity that can be replenished can be the product of the maximum charging time and the single-pile charging rate.

[0212] The feasible charging station set can be a set of candidate stations that have passed the charging feasibility verification and is the basic input for dynamic optimization decision-making.

[0213] The comprehensive optimal target charging station can be the final selection result based on the dual-objective optimization of task time and power redundancy.

[0214] Specifically, in the smart city new energy charging management scenario, the task execution of municipal vehicles is highly time-sensitive and has fixed routes. If the charging strategy is unreasonable, the vehicle may stagnate due to insufficient power during the task, affecting the efficiency of urban public services. Traditional charging strategies do not combine the dynamic characteristics of the task with the real-time status of the charging station, and are prone to wasting charging time or insufficient power replenishment. To address the above issues, this step, when screening the candidate charging station set, first determines the geographical location range of the task gap based on the path topology of the municipal vehicle's current and subsequent tasks. Then, the charging stations within this range are screened from the charging station information set to form a candidate set. Charging feasibility is then verified for each candidate charging station. This involves predicting the arrival time based on the vehicle's current location, driving speed, and real-time road conditions. The remaining battery capacity upon arrival is calculated using dynamic battery capacity change information. The available number of charging piles and the charging rate of each pile at the arrival time are obtained from the charging station information set. The maximum available charging time is calculated based on the available charging time window and the arrival time. The maximum amount of battery capacity that can be replenished is calculated based on the charging rate and the maximum available charging time. The total battery capacity after charging is verified to meet the requirements of subsequent tasks. Candidate charging stations that pass the verification are added to the feasible set. The charging station that minimizes the total vehicle mission time is then prioritized from the feasible set. If the time is the same, the charging station with the highest post-charge battery capacity redundancy is selected as the optimal target charging station. Finally, a charging route is planned based on the location of the charging station, specifying the departure time, driving path, and charging start / end times, to generate an optimal charging strategy.

[0215] This solution precisely matches charging times with task intervals to avoid task delays due to charging, ensuring that municipal vehicles complete public service tasks as planned, such as timely garbage removal and rapid emergency vehicle response. Dynamic calculations can also replenish the maximum amount of power, avoiding energy waste caused by overcharging while ensuring sufficient power, reducing idle driving or secondary charging requirements due to insufficient power, and lowering overall energy consumption. Finally, decisions are made based on the real-time status of charging stations (number of available charging piles, charging rate), enabling dynamic allocation of charging resources, alleviating congestion at charging stations during peak hours, and improving the utilization efficiency of urban energy infrastructure.

[0216] In some embodiments, after the task execution vehicle completes charging at the target charging station, the V2G reverse power supply strategy is executed: monitor whether the current power level of the vehicle has reached a preset full charge threshold, and detect whether there is a surplus charging time window; if the vehicle power level reaches the full charge threshold and there is a surplus charging time window, the grid load data is obtained in real time; when the grid load data is higher than the preset threshold, the amount of reverse power supply is determined based on the remaining stay time of the vehicle and the maximum discharge power of the battery; based on the amount of reverse power supply, a discharge request is sent to the grid dispatching system, and the vehicle is controlled to reversely supply power to the grid through the charging pile.

[0217] The V2G reverse power supply strategy can refer to an energy interaction strategy in which the vehicle transmits the on-board electric energy back to the power grid through the charging pile after charging is completed, which can realize the two-way flow of energy between the vehicle and the grid.

[0218] The preset full charge threshold can be a critical value of power used to determine whether the vehicle has reached a fully charged state. The specific value can be adjusted according to the vehicle battery characteristics and user-defined requirements.

[0219] The surplus charging time window can be the remaining idle time between the time the vehicle is completed charging and the time the staff plans to pick up the vehicle.

[0220] The grid load data may be real-time data reflecting the degree of power load on the grid at a certain moment, including information such as real-time load power, peak and valley time division, etc.

[0221] The amount of reverse power supply is the maximum amount of energy a vehicle can export to the grid while still meeting its own power requirements for subsequent missions. This value is determined by the vehicle's remaining dwell time, the battery's maximum discharge power, and its real-time charge level.

[0222] Specifically, staff will charge municipal vehicles before leaving get off work, but when the vehicles are fully charged, the staff have already left work, causing the vehicles to be idle for a long time after being fully charged, forming a huge idle charging time window, resulting in some electric energy waste. If not utilized, the on-board electric energy will be in a static storage state, causing energy waste. Through reverse power supply, this part of idle energy can be reconnected to the power grid, improving the overall energy utilization rate; at the same time, the urban power grid often faces power supply pressure during peak power periods (such as morning and evening peaks), and there is a problem of excess electricity during valley power periods. Municipal vehicles are stable mobile energy storage units in the city, and their centralized reverse power supply can form a "virtual energy storage power station". When the grid load is higher than the preset threshold, vehicle discharge can effectively alleviate the pressure on the grid. To address the above issues, after the vehicle completes charging in this step, the system continuously monitors whether the battery power reaches the preset full charge threshold (such as ≥95%) and whether there is a surplus charging time window before the next task starts (such as ≥15 minutes); when both are met at the same time, the system obtains grid load data in real time through the grid dispatch interface. If the load value exceeds the preset threshold (such as regional load rate >85%), the system dynamically calculates the reverse power supply based on the remaining stay time and the maximum discharge power of the battery (the calculation logic is to take the minimum value of the remaining stay time × discharge power and the battery's safe discharge upper limit), and then sends a structured discharge request containing the vehicle ID, available power and time window to the grid dispatch center; after obtaining authorization, the vehicle switches to discharge mode through the bidirectional charging pile and feeds power to the grid according to the authorized power. During the process, if it is monitored that the task time window is nearing the end (remaining <5 minutes) or the battery power is lower than the safety redundancy value (such as <80%), the discharge is immediately terminated and switched to standby state.

[0223] Through this solution, idle electricity after recycling and charging can be used to achieve the secondary utilization of on-board energy, reduce carbon dioxide emissions, and improve the utilization rate of clean energy in the city; municipal vehicles can obtain social benefits through reverse power supply and reduce charging costs. The power grid and charging stations can reduce the investment in energy storage facility construction by dispatching vehicle energy storage resources, and improve the operational efficiency of the energy system; strengthen the vehicle-grid interaction capability, provide dynamic energy storage support for smart cities, promote the transformation of new energy vehicles from "electricity terminals" to "energy nodes", and help improve the level of digital energy management in cities.

Claims

1. A new energy charging management method for a smart city, characterized in that: include: Acquire a municipal vehicle operation characteristic data set, and classify municipal vehicles into types using a vehicle type classification strategy based on the municipal vehicle operation characteristic data set to determine a municipal vehicle drive type set; Acquire a municipal vehicle task set, and track a dynamic task variable set of a task-performing vehicle in the municipal vehicle drive type set during task execution according to the municipal vehicle drive type set; According to the dynamic task variable set, analyzing the correlation between the dynamic task variable set and the real-time power consumption data through a multimodal fusion strategy to determine dynamic power change information; Obtain a charging station information set, dynamically optimize charging decisions based on the municipal vehicle task set, the dynamic power change information, and the charging station information set, and determine and output an optimal charging strategy.

2. The method according to claim 1, characterized in that The method of classifying municipal vehicles by vehicle type classification strategy and determining a municipal vehicle driving type set includes: Extracting the main causes of energy consumption fluctuations in the municipal vehicle operation characteristic dataset based on the vehicle type classification strategy; The main causes of energy consumption fluctuations include changes in physical carriers, environmental factors, and event factors; If the energy consumption fluctuation of the current municipal vehicle is mainly caused by the change in the physical carrier, then the current municipal vehicle is determined to be a physical-driven vehicle; If the energy consumption fluctuation of the current municipal vehicle is mainly caused by the change in the environmental factors, then the current municipal vehicle is determined to be an environmentally driven vehicle; If the energy consumption fluctuation of the current municipal vehicle is mainly caused by the change in the event factor, then the current municipal vehicle is determined to be an event-driven vehicle; The municipal vehicle driving type set is generated according to the entity-driven vehicle, the environment-driven vehicle and the event-driven vehicle.

3. The method according to claim 2, characterized in that The acquiring of the municipal vehicle task set and tracking of the dynamic task variable set of the task execution vehicle in the municipal vehicle drive type set during the task execution process according to the municipal vehicle drive type set include: The municipal vehicle task set includes operation characteristics, path characteristics and vehicle status characteristics; Tracking the changes of the main causes of energy consumption fluctuations of the current municipal vehicles during task execution according to the municipal vehicle drive type set, and determining information on changes in the main causes of energy consumption fluctuations; According to the information on the main causes of energy consumption fluctuations, feature mapping and dynamic processing are performed on the operation characteristics, the path characteristics, and the vehicle status characteristics to generate a dynamic load operation feature set, an environmental response situation feature set, and an event-driven task feature set; The dynamic load operation feature set includes real-time entity task load and load state change rate; The environmental response situation feature set includes a real-time natural environment data set and a real-time man-made environment data set; The event-driven task feature set includes task urgency, task realization intensity and task target location; The dynamic task variable set is generated according to the dynamic load operation feature set, the environmental response situation feature set and the event-driven task feature set.

4. The method according to claim 3, characterized in that include: If the vehicle type is the entity-driven vehicle, extract the dynamic load operation feature set from the dynamic task variable set, combine it with the real-time power consumption data, analyze the gain / loss effect of the real-time entity task load and the load state change rate on the power consumption rate, and generate an entity change impact sub-strategy; If the vehicle type is the environment-driven vehicle, extracting the environmental response situation feature set from the dynamic task variable set, combining it with the real-time power consumption data, analyzing the gain / loss effect of the real-time natural environment dataset and the real-time man-made environment dataset on the power consumption rate, and generating an environmental change impact sub-strategy; If the vehicle type is the event-driven vehicle, extract the event-driven task feature set from the dynamic task variable set, combine it with the real-time power consumption data, analyze the gain / loss effect of the task urgency, the task implementation intensity, and the task target location on the power consumption rate, and generate an event-focused impact sub-strategy; The multimodal fusion strategy is generated according to the entity change impact sub-strategy, the environment change impact sub-strategy and the event focus impact sub-strategy.

5. The method according to claim 4, characterized in that The entity changes affect sub-strategies, including: Based on the real-time entity task load and the load state change rate, identifying gain nodes and loss nodes on the path of the task execution vehicle; The gain node represents a node where the real-time entity task load generates a gain on the power consumption rate; The depreciation node represents a node where the real-time entity task load depreciates the power consumption rate; According to the gain node and the degauss node, combined with the load state change rate, the change trend of the power consumption rate is predicted to determine the dynamic power change information of the physical drive vehicle.

6. The method according to claim 4, characterized in that include: For the environment-driven vehicle, the environment change impact sub-strategy is adopted: Analyze the real-time natural environment data set, and use the correlation factors in the natural environment that generate a gain on the power consumption rate as natural environment positive correlation factors, and use the correlation factors in the environment that generate a loss on the power consumption rate as natural environment negative correlation factors; Analyzing the plurality of positively correlated natural environment factors, performing gain superposition on the plurality of positively correlated natural environment factors, and determining a first natural environment impact coefficient; Analyze the plurality of negatively correlated factors of the natural environment, perform a reduction and superposition on the plurality of negatively correlated factors of the natural environment, and determine a second natural environment impact coefficient; performing normalization processing on the first natural environment impact coefficient and the second natural environment impact coefficient according to the first natural environment impact coefficient and the second natural environment impact coefficient to determine a natural environment response intensity coefficient; Analyze the real-time human environment data set, and use the correlation factors in the human environment that increase the power consumption rate as positive human environment correlation factors, and use the correlation factors in the environment that decrease the power consumption rate as negative human environment correlation factors; Analyzing the plurality of positively correlated factors of the human-made environment, performing gain superposition on the plurality of positively correlated factors of the human-made environment, and determining a first human-made environment impact coefficient; Analyzing the plurality of negatively correlated factors of the human-made environment, performing a deduction and superposition on the plurality of negatively correlated factors of the human-made environment, and determining a second human-made environment impact coefficient; performing normalization processing on the first man-made environmental impact coefficient and the second man-made environmental impact coefficient according to the first man-made environmental impact coefficient and the second man-made environmental impact coefficient to determine a man-made environmental response intensity coefficient; Analyzing the real-time human environment data set to determine infrastructure characteristics, and setting a number of infrastructure verification time intervals with different time spans based on the infrastructure characteristics; Based on the real-time power consumption data, the impact of the superposition of the natural environment response intensity coefficient and the man-made environment response intensity coefficient on the power consumption rate in each infrastructure approved time interval is analyzed, the changing trend of the power consumption rate is predicted, and the dynamic power change information of the environmentally driven vehicle is determined.

7. The method according to claim 4, characterized in that include: For the event-driven vehicle, the event-focused impact sub-strategy is adopted: Determining a task urgency index according to the task urgency, and using the task urgency index as a first impact coefficient; Determining current task information based on the municipal vehicle task set, and determining task duration and task operation complexity of the corresponding task based on the current task information; Normalizing the task duration and the task operation complexity according to a preset numerical normalization strategy to determine a task duration index and a task complexity index; The product of the task duration index and the task complexity index is used as the second impact coefficient; Determine the mission arrival route based on the mission target location; Based on the preset traffic remote sensing information, according to the task arrival route, determine the real-time traffic complexity and driving distance corresponding to the task arrival route; Normalizing the real-time traffic complexity and the driving distance according to a preset numerical normalization strategy to determine a traffic complexity index and a driving distance index; taking the product of the road condition complexity index and the driving distance index as a third influence coefficient; taking the sum of the first influence coefficient, the second influence coefficient, and the third influence coefficient as the comprehensive event influence factor; According to the comprehensive event impact factor and the real-time power consumption data, the change trend of the power consumption rate is predicted, and the dynamic power change information of the event-driven vehicle is determined.

8. The method according to claim 4, characterized in that The acquiring of the charging station information set, dynamically optimizing the charging decision based on the municipal vehicle task set, the dynamic power change information, and the charging station information set, and determining and outputting the optimal charging strategy includes: Based on the dynamic power change information and the municipal vehicle task set, predict the remaining power change trend of the task execution vehicle during the subsequent task execution process, and determine whether the remaining power will fall below a preset safety threshold before completing the subsequent task; If insufficient power is predicted, the charging strategy planning is executed: Extracting the path topology and task time nodes of the current and subsequent tasks according to the municipal vehicle task set; Determine the charging time window and the corresponding geographical location range of the vehicle during the task interval according to the path topology; Within the geographical location range, screening a set of candidate charging stations that meet the distance constraint according to the charging station information set; Based on the candidate charging station set, charging decisions are dynamically optimized to generate the preferred charging strategy.

9. The method according to claim 8, characterized in that The dynamically optimizing charging decisions based on the candidate charging station set to generate the preferred charging strategy includes: Based on the set of candidate charging stations, charging feasibility verification is performed on each candidate charging station: Predict the arrival time at the charging station based on the vehicle's current location, driving speed, and the location of candidate charging stations; Determining the remaining power based on the dynamic power change information; Obtaining the real-time available number of charging piles and the charging rate of each pile at the charging station according to the charging station information set; Determine a maximum charging duration based on the charging time window and the arrival time; Determining a maximum amount of charge that can be replenished based on the single-pile charging rate and the maximum charging time; Verify whether the total power after charging is sufficient for the entire subsequent task; The total amount of electricity after the charging is completed is the sum of the remaining amount and the maximum amount of electricity that can be replenished; Adding the verified candidate charging station to a set of feasible charging stations; From the set of feasible charging stations, preferentially selecting the charging station that minimizes the total time taken for the vehicle mission as the comprehensive optimal target charging station; If the time consumption is the same, the charging station with the highest power redundancy after charging is selected as the comprehensive optimal target charging station; According to the comprehensive optimal target charging station, a charging route is determined and the preferred charging strategy is generated.

10. The method according to claim 9, characterized in that The method further comprises: After the task execution vehicle completes charging at the target charging station, the V2G reverse power supply strategy is executed: Monitor whether the vehicle's current battery level has reached the preset full charge threshold and detect whether there is a surplus charging time window; If the vehicle's power reaches the full charge threshold and there is a surplus charging time window, the grid load data is obtained in real time; When the grid load data is higher than a preset threshold, the amount of reverse power supply is determined based on the remaining vehicle stay time and the maximum discharge power of the battery; According to the reverse power supply amount, a discharge request is sent to the power grid dispatching system, and the vehicle is controlled to reversely supply power to the power grid through the charging pile.

Images