Vehicle dispatching method and system based on parking and charging integrated stereo garage

Abstract

The application provides a vehicle scheduling method and system based on a parking and charging integrated stereo garage, the method being: obtaining real-time vehicle information and historical charging records of a vehicle to be charged; the real-time vehicle information including residual power, estimated driving distance, estimated external power consumption and estimated charging time; constructing a fuzzy evaluation model based on the vehicle information and a charging mode, inputting the real-time vehicle information into the fuzzy evaluation model to obtain a preliminary charging mode; searching for a historical similar scene based on the real-time vehicle information, calling a historical charging mode of the historical similar scene to correct the preliminary charging mode to obtain a corrected charging mode; obtaining a driver preference and a time-of-use electricity price, and optimizing a charging duration based on multi-objective linear programming; calculating a comprehensive score based on the corrected charging mode, determining a charging order according to the score, and scheduling the vehicle to charge. The fuzzy evaluation is combined with historical data to correct the charging mode, the double-layer programming is used to optimize the duration, the charging rationality and resource efficiency are improved, the cost is reduced, and the parking and charging problems are solved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of electric vehicle charging, and in particular to a vehicle scheduling method and system based on parking and charging integrated multi-level garage. BACKGROUND

[0002] In the modern urban transportation and energy management system, the number of electric vehicles is growing rapidly, and parking resources are becoming increasingly scarce. The problems of parking difficulty and charging difficulty are prominent, which seriously affects the efficient operation of the city. The traditional multi-level garage has low efficiency in the vehicle access link, and is effectively integrated with autonomous driving technology. The driver needs to wait in line for a long time, causing time waste. In terms of electric vehicle charging management, the traditional mode lacks an effective scheduling mechanism and cannot reasonably plan the charging method and order according to the actual situation of the vehicle. At the same time, the garage has deficiencies in energy utilization, such as the application of clean energy such as photovoltaic power generation in the garage and vehicle charging has not been fully valued.

[0003] Therefore, the problems of parking and charging have become a technical problem that needs to be solved urgently, and new solutions are needed to realize the efficient cooperation of parking and charging and promote the intelligent development of urban transportation and energy management. SUMMARY

[0004] To solve the above problems, the present application provides a vehicle scheduling method and system based on parking and charging integrated multi-level garage, which corrects the preliminary charging method through historical similar scenarios, making the charging level more in line with the actual demand; based on a multi-objective linear programming double-layer model, the upper layer meets the driver's power and cost demand, and the lower layer optimizes the time length combined with time-of-use electricity price, realizes dynamic adaptation of charging strategy, and effectively improves the parking and charging efficiency and user satisfaction.

[0005] To achieve the above purpose, the present application adopts the following technical solutions:

[0006] In a first aspect, the present application provides a vehicle scheduling method based on parking and charging integrated multi-level garage, comprising:

[0007] Obtaining real-time vehicle information and historical charging records of the vehicle to be charged; the real-time vehicle information includes residual power, estimated driving distance, estimated external power consumption and estimated charging time; the historical charging records include historical vehicle information and corresponding historical charging method;

[0008] Based on the vehicle information and the charging method, a fuzzy evaluation model is constructed, the real-time vehicle information is input into the fuzzy evaluation model, and the preliminary charging method is obtained;

[0009] Comparing the real-time vehicle information with the historical vehicle information to find the historical similar scenario; calling the historical charging method of the historical similar scenario to correct the preliminary charging method and obtain the corrected charging method;

[0010] The driver preference and the time-of-use electricity price are acquired, and the charging time under the modified charging mode is optimized based on multi-objective linear programming;

[0011] The comprehensive score is calculated based on the modified charging mode, the charging sequence is determined according to the score, and the vehicle to be charged is scheduled to be charged according to the charging sequence, the modified charging mode and the charging time.

[0012] In a second aspect, the present application provides a vehicle scheduling system based on a parking and charging integrated stereo garage, comprising:

[0013] The basic information acquisition module is configured to acquire real-time vehicle information and historical charging records of the vehicle to be charged; the real-time vehicle information comprises residual power, estimated driving distance, estimated external power consumption and estimated charging time; and the historical charging records comprise historical vehicle information and corresponding historical charging modes;

[0014] The preliminary strategy acquisition module is configured to construct a fuzzy evaluation model based on the vehicle information and the charging mode, input the real-time vehicle information into the fuzzy evaluation model, and obtain a preliminary charging mode;

[0015] The modified strategy acquisition module is configured to compare the real-time vehicle information with the historical vehicle information, find a historical similar scenario, and modify the preliminary charging mode by calling a historical charging mode of the historical similar scenario to obtain a modified charging mode;

[0016] The charging time acquisition module is configured to acquire a driver preference and a time-of-use electricity price, and optimize the charging time under the modified charging mode based on multi-objective linear programming;

[0017] The scheduling execution module is configured to calculate a comprehensive score based on the modified charging mode, determine a charging sequence according to the score, and schedule the vehicle to be charged to be charged according to the charging sequence, the modified charging mode and the charging time.

[0018] In a third aspect, the present application provides a computer readable storage medium having a computer program stored thereon, the program being executed by a processor to implement the steps of the vehicle scheduling method based on the parking and charging integrated stereo garage according to the first aspect.

[0019] In a fourth aspect, the present application provides a computer device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, wherein the processor executes the program to implement the steps of the vehicle scheduling method based on the parking and charging integrated stereo garage according to the first aspect.

[0020] Compared with the prior art, the present application has the following advantages:

[0021] (1) In order to solve the problems of low scheduling efficiency and lack of dynamic optimization of charging strategy in the traditional stereo garage, the present application fuses real-time vehicle information and historical charging data to construct a fuzzy evaluation model to generate a preliminary charging mode, and corrects the strategy based on the historical similar scene to avoid the blindness of charging. At the same time, the double-layer linear programming model of driver preference and time-of-use electricity price is introduced to solve the defects of high charging cost and uneven power grid load in the traditional mode, realize the fine allocation of charging resources, and improve the overall operation efficiency of the garage.

[0022] (2) The present application determines the charging grade of electric vehicles by fuzzy system comprehensive evaluation method, calculates the total score of the charging grade of each vehicle in the buffer photovoltaic greenhouse area, optimizes the scheduling vehicles and reasonably selects the charging mode according to the score, and the higher the score, the higher the scheduling priority. This ensures that the vehicles in urgent need of charging are handled in time, reasonably arranges the charging sequence, avoids the waste of charging resources, improves the utilization efficiency of charging resources, and further improves the overall work efficiency of the garage.

[0023] (3) According to the historical scheduling and charging conditions and the feedback of the driver, when the vehicle battery SOC, distance and other information are similar, the evaluation matrix and factor weight vector in the fuzzy comprehensive evaluation algorithm are optimized, so that the scheduling and charging mode selection are more reasonable. Considering the personal driving habits and individual needs of the driver, the algorithm is optimized to accurately serve and reduce the operation and waiting time. In terms of charging time optimization, the charging time is determined by linear programming according to the priority scheduling order and time-of-use price factors, so as to realize the cost optimization and improve the energy utilization and work efficiency.

[0024] The advantages of the additional aspects of the present application will be partially given in the following description, partially will become obvious from the following description, or will be known by the practice of the present application. BRIEF DESCRIPTION OF DRAWINGS

[0025] The drawings accompanying the specification of the present application form a part thereof, serve to provide further understanding of the present application, and together with the description of the exemplary embodiments of the present application and their description serve to explain the present application, and do not constitute limitations on the present application.

[0026] Figure 1 A main flowchart of a vehicle scheduling method based on a parking and charging integrated stereo garage is provided for the embodiments of the present application.

[0027] Figure 2 A detailed flowchart of a vehicle scheduling method based on a parking and charging integrated stereo garage is provided for the embodiments of the present application.

[0028] Figure 3 A flowchart of fuzzy comprehensive evaluation method for determining and optimizing scheduling and charging mode is provided for the embodiments of the present application. DETAILED DESCRIPTION

[0029] The present invention will be further described below with reference to the accompanying drawings and embodiments.

[0030] Example 1

[0031] like Figure 1 As shown, this embodiment discloses a vehicle scheduling method based on an integrated parking and charging multi-level parking garage, including the following steps:

[0032] S1: Obtain real-time vehicle information and historical charging records of the vehicle to be charged; the real-time vehicle information includes remaining battery power, estimated driving distance, estimated external power consumption, and estimated charging time; the historical charging records include historical vehicle information and corresponding historical charging methods;

[0033] S2: Construct a fuzzy evaluation model based on vehicle information and charging method. Input real-time vehicle information into the fuzzy evaluation model to obtain the preliminary charging method.

[0034] S3: Compare real-time vehicle information with historical vehicle information to find similar historical scenarios; call the historical charging methods of similar historical scenarios to correct the initial charging method and obtain the corrected charging method;

[0035] S4: Obtain driver preferences and time-of-use electricity prices, and optimize the charging time under different charging methods based on multi-objective linear programming;

[0036] S5: Calculate the comprehensive score based on the corrected charging method, determine the charging order according to the score, and schedule the vehicles to be charged according to the charging order, corrected charging method and charging time.

[0037] Before dispatching vehicles to the charging location in the multi-level parking garage, it is necessary to first construct an integrated parking and charging garage equipped with a photovoltaic buffer zone.

[0038] Specifically, the garage is connected to the autonomous vehicle's communication module: high-performance communication equipment, such as modules supporting 5G communication, is installed in both the garage and the autonomous vehicle to ensure fast and stable data transmission between the vehicle and the garage, and between the garage and the cloud. The garage-side communication equipment is connected to the control system, while the vehicle-side communication equipment is integrated into the onboard computer system to achieve real-time information interaction, including the transmission of vehicle information data such as vehicle location and battery SOC.

[0039] Photovoltaic buffer zone construction: In the stereo garage, a photovoltaic buffer zone is built in a suitable position, such as the top or the peripheral open area. High-efficiency photovoltaic panels are selected, and the number and layout of the photovoltaic panels are determined according to the power demand of the garage and the estimated power demand for charging the vehicles. An inverter is installed to convert the direct current generated by the photovoltaic panels into alternating current, which is connected to the garage power grid system. An energy storage device (such as a lithium battery pack) is also installed to store excess power and provide power to the garage and vehicles during periods of insufficient light or high power demand.

[0040] The photovoltaic buffer zone is equipped with an energy storage device to store redundant photovoltaic power and release power during periods of insufficient light or high power demand, ensuring charging stability.

[0041] A photovoltaic buffer zone is set up in the stereo garage, and its photovoltaic power generation system is connected to the conventional power grid to provide power to the garage. The photovoltaic buffer zone is equipped with an energy storage device to store redundant photovoltaic power and release power during periods of insufficient light or high power demand, ensuring charging stability.

[0042] Sensor and data acquisition equipment setup: Vehicle detection sensors such as infrared sensors and geomagnetic sensors are installed at key locations in the stereo garage to monitor the entry and exit of vehicles and their parking positions. High-precision power monitoring sensors are installed in the charging area and directly connected to the garage integrated information system to obtain real-time charging status and power changes of electric vehicles. Traffic information collection equipment such as cameras and traffic flow sensors are installed at the entrance of the garage and on the surrounding roads to collect road information and transmit it to the cloud.

[0043] As a specific implementation, the stereo garage integrated information system provides a charging scheme recommendation interface for vehicle owners, allowing them to manually adjust charging priority, SOC target, or cost budget according to their preferences.

[0044] After the driver arrives near the destination, the vehicle automatically searches for the garage, queues up, and is automatically stored and retrieved by the stereo garage based on the vehicle dispatching method proposed in this embodiment.

[0045] In this embodiment, the stereo garage equipped with a photovoltaic buffer zone can convert light energy into electrical energy and store it through the photovoltaic power generation system and energy storage device, forming a synergistic effect with vehicle dispatching: on the one hand, photovoltaic power is preferentially used for vehicle charging, especially during valley periods or periods of sufficient light, which can reduce dependence on traditional power grids and further reduce charging costs through time-of-use pricing, such as using valley photovoltaic power for fast charging of high-priority vehicles to reduce peak power grid pressure; on the other hand, the energy storage state of the photovoltaic buffer zone is used as a dispatching parameter to affect charging mode allocation, such as increasing the slow charging ratio to protect the battery when the energy storage is sufficient, and calling valley power from the power grid to meet fast charging demands when the energy storage is insufficient. In addition, its clean energy characteristics can reduce carbon footprint and improve energy utilization efficiency and sustainability of vehicle dispatching.

[0046] Next, in combination with Figure 2 , a vehicle dispatching method based on a parking and charging integrated stereo garage is disclosed in detail.

[0047] In S1, vehicle information is obtained from a vehicle terminal or the cloud, including real-time residual power (SOC) of an electric vehicle, estimated driving distance (distance between departure and destination), estimated external power consumption, estimated charging time, and historical charging records, which are stored in a cloud database.

[0048] The historical charging records include historical vehicle information and corresponding historical charging methods. The charging methods are divided into four levels: first-level fast charging, second-level fast charging, first-level slow charging, and second-level slow charging.

[0049] The estimated external power consumption is represented by road information, based on the corresponding relationship between road congestion and power consumption, to obtain the additional power consumption required for a predetermined road passage.

[0050] It should be understood that the specific charging power, speed level definition of the "first-level fast charging, second-level fast charging, first-level slow charging, and second-level slow charging" can be determined by those skilled in the art according to actual application scenarios. Here, the naming is to clearly distinguish different charging levels. At the same time, the corresponding relationship between road congestion and power consumption can also be obtained or calculated by those skilled in the art.

[0051] As a specific implementation, the cloud database is linked with the city intelligent transportation system to obtain real-time road congestion information and dynamically adjust the external power consumption parameter, improving the robustness of dispatching.

[0052] Further, by constructing a fuzzy evaluation model, the optimal charging method is determined based on vehicle information.

[0053] This embodiment collects multi-dimensional real-time data such as residual power and estimated driving distance to build an information base for dynamic scheduling. Unlike the traditional "only rely on current power" extensive mode, the new "estimated external power consumption" (such as additional energy consumption caused by road congestion) and "estimated charging time" dimensions enable the system to more accurately assess the actual needs of vehicles. At the same time, by accessing historical charging records (including charging methods and corresponding scenarios), data training samples are provided for subsequent fuzzy evaluation and historical correction, avoiding the one-sidedness of relying solely on real-time data, thereby improving the perception of users' implicit needs.

[0054] In S2, the factor set and evaluation set are defined, and a fuzzy evaluation model is established, as shown in Figure 3 , specifically including:

[0055] S201: taking the vehicle information as a factor set and the charging mode as an evaluation set; the charging mode includes first-level fast charging, second-level fast charging, first-level slow charging and second-level slow charging;

[0056] S202: taking the elements in the factor set as rows and the elements in the evaluation set as columns to construct a single-factor fuzzy evaluation matrix;

[0057] S203: determining the importance weight of each vehicle information by using the analytic hierarchy process;

[0058] S204: obtaining the membership degree of the vehicle to be charged to each charging mode based on the single-factor fuzzy evaluation matrix and the importance weight; determining the preliminary charging mode based on the highest membership degree.

[0059] In S201, the factor set is a general set composed of various factors affecting the evaluation object, usually denoted by U, wherein the element represents the ith factor affecting the evaluation object. These factors usually have different degrees of fuzziness.

[0060] Here, the charging level of the electric vehicle is set as , represents the SOC of the electric vehicle battery, represents the distance between the departure place and the destination, represents the power consumption of the electric vehicle caused by external factors such as traffic jams when returning to the departure place, represents the charging time of the electric vehicle (from the start of charging the electric vehicle to the time when the vehicle returns to the departure place).

[0061] The evaluation set is a set composed of various results that the evaluator can make on the evaluation object, usually denoted by V, wherein the element represents the jth evaluation result, which can be represented by different levels, comments or numbers according to the actual needs (note that m and n appear in the following text, m represents m factors set, and n represents n evaluation set).

[0062] Here, the evaluation set for evaluating the electric vehicle level is set as , which respectively represents first-level fast charging, second-level fast charging, first-level slow charging and second-level slow charging.

[0063] As an embodiment, in this embodiment, the preferred first-level fast charging is direct current fast charging with a charging power of about 40kw; the second-level fast charging is direct current fast charging with a charging power of about 30kw; the first-level slow charging is alternating current slow charging with a charging power of about 3.5kw; and the second-level slow charging is alternating current slow charging with a charging power of about 7kw.

[0064] In S202, a single-factor fuzzy evaluation matrix R is constructed based on the factor set and the evaluation set. The rows of the matrix R correspond to the factor elements, and the columns correspond to the charging methods.

[0065] If the membership degree of the i-th element in the factor set U to the 1st element in the evaluation set V is then the result of the single-factor evaluation for the i-th element is represented by a fuzzy set as follows: The single-factor evaluation set of the m-th factor is denoted as The matrix composed of the m single-factor evaluation sets as rows is denoted as , which is called a fuzzy comprehensive evaluation matrix. Table 1 shows an example of the fuzzy comprehensive evaluation matrix.

[0066] Table 1 Fuzzy comprehensive evaluation matrix

[0067]

[0068] In the charging level decision of the electric vehicle, the stereo garage comprehensive information system can determine the preliminary membership degree of each ordered pair in U x V by saving the historical charging data of the autonomous vehicle in the L4 autonomous driving cloud, such as the charging method and the charging time.

[0069] The single-factor evaluation matrix is obtained as follows:

[0070]

[0071] where the assignment logic is based on historical data or expert experience, for example, the higher the SOC, the higher the membership degree to fast charging.

[0072] In S203, the importance of each factor is different in the evaluation process. Therefore, a weight is given to each factor based on the current vehicle information, which is a factor weight vector. The fuzzy set of the weight set of each factor is denoted as A.

[0073] To improve the reliability of the model, the pair-wise comparison matrix of the analytic hierarchy process (AHP) is used to construct this weight vector. It should be understood that the specific method of using the analytic hierarchy process can be implemented by those skilled in the art.

[0074] In S204, the membership degree B of the vehicle to be charged to each charging method is obtained through fuzzy transformation .

[0075] As can be seen, based on the current vehicle information, the membership degree of the first-level charging method is the highest, so the first-level fast charging is selected as the preliminary charging method.

[0076] ​​​​As a specific embodiment, the membership of the single-factor fuzzy evaluation matrix R is dynamically updated by a cloud deep learning model, and the membership assignment logic is optimized in combination with real-time charging feedback data.

[0077] In this embodiment, multi-factor comprehensive decision is realized by a fuzzy evaluation model. Vehicle information such as remaining power and driving distance is taken as a factor set, and charging modes are taken as an evaluation set. Dynamic weights are given to each factor by using the analytic hierarchy process (AHP), so as to avoid the mechanicalness of determining the charging level by a single power threshold in the traditional way. In addition, the introduction of the single-factor fuzzy evaluation matrix (based on historical data statistics) makes the strategy generation more in line with the habits of the user group, and improves the scientificity and resource adaptability of the initial strategy.

[0078] In S3, the real-time vehicle information is compared with the historical vehicle information to find a similar historical scenario, and the preliminary charging mode is corrected based on the similar historical scenario. The correction is to adjust the distribution of the membership degrees of different charging modes for the charging mode selected for the similar historical scenario, so that the fuzzy evaluation result (the final score of each charging mode) is more in line with the actual selection of the similar historical scenario, and a more reasonable charging mode is finally derived through the optimized weight vector. Specifically:

[0079] S301: Scene similarity determination. The remaining power, the predicted driving distance and the predicted external power consumption in the current vehicle information are compared with the historical remaining power, the predicted driving distance and the predicted external power consumption stored in the historical charging records of the historical charging events in the historical charging records, respectively. If the deviation is less than a set threshold, it is determined that the scenes are similar.

[0080] The threshold can be set by a person skilled in the art according to the actual application scenario and demand through experimental tests, simulation analysis and other reasonable methods; as for the similarity comparison, the Euclidean distance calculation method can be used, which will not be described here.

[0081] The historical charging records include historical vehicle information and corresponding historical charging modes. The charging modes are divided into four levels: first-level fast charging, second-level fast charging, first-level slow charging and second-level slow charging.

[0082] S302: Evaluation matrix optimization. Based on the actual charging mode selection results of the similar scenarios in the historical data, the least squares method is used to adjust the current single-factor fuzzy evaluation matrix R, so that the mean square error between the fuzzy evaluation result and the actual selection result is minimized.

[0083] ;

[0084] wherein, is a learning rate (default 0.1), is the statistical distribution of the actual charging mode in the similar historical scenario, is the current matrix.

[0085] S303: Weight vector optimization. The analytic hierarchy process (AHP) is combined with the driver feedback score (1-5 points) in the historical scene to dynamically update the factor weight vector A, and the specific formula is:

[0086] ;

[0087] wherein a is a historical weight decay coefficient (default 0.7), is the weight vector recalculated according to the historical satisfaction feedback of the driver; is the current weight vector.

[0088] Based on the adjusted and update the membership , and get the modified charging mode according to the membership.

[0089] In this embodiment, by using the least squares method, the proportion of the supplementary charging mode is adjusted, so that the error between the adjusted matrix evaluation result and the historical selection result is as small as possible, and the factor weight vector is updated according to the historical user satisfaction, so as to consider the current scene and the historical scene at the same time, provide a more referential charging scheme for the user, and improve the charging efficiency and the user satisfaction.

[0090] Further, rationality can be judged. The charging completion rate of the vehicle after scheduling (≥95%), the driver satisfaction score (≥4 points), and the charging resource utilization rate (≥85%) are taken as quantitative indexes of optimization rationality.

[0091] This embodiment realizes dynamic calibration of the strategy through comparison of historical similar scenes. Specifically, similar scenes are matched by the Euclidean distance, and the evaluation matrix and the weight vector are adjusted based on the historical actual selection result, solving the defects of the traditional fuzzy model relying on fixed parameters and being unable to adapt to individual differences, and avoiding experience deviation caused by unified strategy. By continuously learning historical data, the scheduling logic is dynamically optimized, and the user satisfaction and the resource use efficiency are improved.

[0092] In S4, after getting the modified charging mode, the charging duration is further determined.

[0093] Under the premise of meeting the driving demand, the optimal charging duration is determined by using linear programming combined with time-of-use electricity price, so as to reduce the charging cost.

[0094] The charging duration optimization is realized by a double-layer planning model:

[0095] Upper model: taking the individualized demand of the driver as the target, taking the maximum SOC after charging and the minimum total charging cost as the objective function, taking the time satisfying the acceptable time and being within the rated charging capacity as the constraint condition, and constructing a multi-objective linear programming model:

[0096] ;

[0097] wherein, is the remaining power after charging, represents the current remaining power, is the charging efficiency, is the charging power, t is the preliminary charging duration, and represent the minimum and maximum dwell time allowed by the driver, represents the total charging cost, represents the time-of-use electricity price of the kth period, represents the charging duration of the kth period, is the number of periods, represents the charging capacity constraint.

[0098] Lower model: guided by the time-of-use electricity price, correct the preliminary charging duration output by the upper model:

[0099] ;

[0100] wherein, represents the total charging cost adjusted according to the time-of-use electricity price, represents the preliminary charging duration output by the upper model (unit: minutes); represents the time-of-use electricity price of the kth period (unit: yuan / kWh), which is usually divided into peak segment (such as 0.8 yuan / kWh), flat segment (0.6 yuan / kWh), and valley segment (0.4 yuan / kWh); represents the total number of electricity price periods divided in a day (such as T=3, corresponding to peak, flat, and valley three segments); represents the charging duration constraint in the kth period.

[0101] In this embodiment, the upper model focuses on the driver's demand and realizes personalized solutions; the lower model links the electricity price signal to ensure that the solution fits the real-time market mechanism. This hierarchical design makes the system not only adapt to individual differences, but also respond to macro grid regulation, improving the overall charging scheduling efficiency.

[0102] As an implementation manner, the double-layer model is converted into a single-layer mixed integer linear programming problem for solving through KKT condition, and the specific implementation manner can be realized by those skilled in the art.

[0103] Based on the above multi-objective function and constraint conditions, a multi-objective linear programming method is used for solving. In the solving process, according to the two objectives of the driver to maximize the remaining power after charging and minimize the total charging cost, the preliminary charging duration is determined; further, the time-of-use electricity price optimization charging cost is considered. Through the linear programming algorithm, under the premise of meeting the driving demand (such as ensuring that the vehicle can travel to the destination) and the garage resource limit (such as the number of charging piles, power, etc.), the optimal charging duration under the modified charging mode is determined by fusing with the time-of-use electricity price.

[0104] In S5, based on the membership degree of the modified charging mode and the evaluation grade score vector , the comprehensive score F is calculated by the formula , wherein and respectively represent the membership degree and the evaluation grade score vector corresponding to the charging mode j. The evaluation grade score vector is pre-set, and exemplary, the first-class fast charging corresponds to 100 points, the second-class fast charging corresponds to 80 points, etc.

[0105] The comprehensive score F directly reflects the emergency degree of vehicle charging: the higher the score, the higher the priority (such as vehicles with low remaining power and long-distance travel are preferentially dispatched). The garage sorts according to the score, combines the photovoltaic buffer zone state and the charging pile occupation condition, and dynamically arranges the vehicle to enter the fast charging / slow charging area.

[0106] As a specific embodiment, when the vehicle battery SOC reaches a pre-set threshold (such as 85%) or the destination distance is less than a set value (such as 7.5km), the vehicle is preferentially dispatched to the slow charging area, thereby prolonging the battery life and reducing the energy consumption.

[0107] Embodiment 1

[0108] Taking a certain electric vehicle (remaining power before charging =30%, target power after charging =80%, and stay time 2 hours) as an example:

[0109] (1) Fuzzy evaluation generates preliminary charging mode

[0110] The real-time vehicle information (remaining power 30%, predicted driving distance 100km, predicted external power consumption 15kWh, and predicted charging time 120min) is input to the fuzzy evaluation model.

[0111] Based on the initial single-factor fuzzy evaluation matrix and the factor weight vector =[0.4,0.3,0.2,0.1], the membership degree is calculated, and the output is "first-class fast charging" (comprehensive score 85 points), corresponding to the charging power 40kW.

[0112] (2) Historical scenario correction charging strategy

[0113] The system retrieves historical charging records, matches to similar scenarios (remaining power 28%, estimated driving distance 98km, etc.), optimizes the evaluation matrix, selects the result based on the actual charging method corresponding to the similar scenario in the historical data, adjusts the current single-factor fuzzy evaluation matrix R using the least squares method to minimize the mean square error between the fuzzy evaluation result and the actual selection result, and obtains the updated evaluation matrix; At the same time, the weight vector is optimized, and it is found that the driver's satisfaction score for "first-level fast charging" in this scenario is only 3.2 points (full score 5 points), based on this, the weight vector is dynamically adjusted by the analytic hierarchy process, and = [0.3, 0.4, 0.2, 0.1].

[0114] Based on the updated evaluation matrix and weight vector, the updated membership is obtained, and based on the membership, "second-level fast charging" is obtained as the final charging method.

[0115] (3) Double-layer planning optimization of charging time

[0116] The upper model generates = 45min (cost 30yuan, SOC increased to 78%);

[0117] The lower model adjusts to the valley segment charging combined with the time-of-use electricity price (valley segment 0.4 yuan / kWh, peak segment 0.8 yuan / kWh) = 60min, cost reduced to 23yuan, SOC reached 80%.

[0118] (4) Comprehensive score and charging execution

[0119] Based on the corrected "second-level fast charging" method, combined with the remaining power, driving distance and other parameters, the comprehensive score (82 points) is calculated, the garage is preferentially dispatched to the valley segment charging pile, and after completing the charging, the driver is notified to take the car.

[0120] It should be understood that the above embodiment 1 is only an exemplary scenario set for the purpose of understanding the technical solution, and the vehicle parameters, scores, electricity prices and calculation results involved are all hypothetical demonstrations, which are intended to explain the scheduling logic based on fuzzy evaluation, historical correction and double-layer planning. In actual application, the parameter values, weight vector adjustment rules and linear programming solution results can be determined by those skilled in the art according to the specific scene data, equipment characteristics and scheduling targets, which all belong to the reasonable implementation scope of the technical solution of the present application.

[0121] The embodiment fuses the fuzzy evaluation model and the historical scene correction to construct a dynamic charging mode decision mechanism, optimizes the charging time based on the driver preference and the multi-objective linear programming, improves the rationality of the charging strategy and the user adaptability, introduces the photovoltaic buffer zone and the energy storage equipment to realize the efficient use of clean energy and guarantee the charging stability, builds a real-time interaction system of vehicle-garage-cloud based on the 5G communication and the sensor network, optimizes the vehicle scheduling and the charging sequence, and reduces the waiting time of the driver. The problems of parking difficulty, slow charging and energy waste are effectively solved, and the intelligent development of urban traffic and energy management is promoted.

[0122] Embodiment two

[0123] The embodiment provides a vehicle scheduling system based on a parking and charging integrated stereo garage, which comprises:

[0124] A basic information acquisition module is configured to acquire real-time vehicle information and historical charging records of a vehicle to be charged; the real-time vehicle information comprises residual power, expected driving distance, expected external power consumption and expected charging time; and the historical charging records comprise historical vehicle information and corresponding historical charging modes;

[0125] A preliminary strategy acquisition module is configured to construct a fuzzy evaluation model based on the vehicle information and the charging mode, input the real-time vehicle information into the fuzzy evaluation model, and obtain a preliminary charging mode;

[0126] A correction strategy acquisition module is configured to compare the real-time vehicle information with the historical vehicle information, find a historical similar scene, call the historical charging mode of the historical similar scene to correct the preliminary charging mode, and obtain a corrected charging mode;

[0127] A charging time acquisition module is configured to acquire a driver preference and a time-of-use electricity price, and optimize the charging time under the corrected charging mode based on multi-objective linear programming;

[0128] A scheduling execution module is configured to calculate a comprehensive score based on the corrected charging mode, determine a charging sequence according to the score, and schedule the vehicle to be charged to charge according to the charging sequence, the corrected charging mode and the charging time.

[0129] Embodiment three

[0130] The embodiment provides a computer readable storage medium, which stores a computer program, and the program is executed by a processor to realize the steps in the vehicle scheduling method based on the parking and charging integrated stereo garage according to the above embodiment one.

[0131] Embodiment four

[0132] The embodiment provides a computer device, which comprises a memory, a processor and a computer program stored in the memory and capable of running on the processor, and the processor implements the steps in a vehicle scheduling method based on a parking and charging integrated stereo garage according to the embodiment one when executing the program.

[0133] The steps or modules involved in the above embodiments two to four correspond to the embodiment one, and the specific implementation can be referred to the related description part of the embodiment one. The term "computer readable storage medium" should be understood as including a single medium or multiple media of one or more instruction sets; and should also be understood as including any medium capable of storing, encoding or carrying the instruction set for execution by the processor and causing the processor to execute any method in the present application.

[0134] The above merely describes the preferred embodiments of the present application and is not used to limit the present application. For those skilled in the art, the present application can have various modifications and changes. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application should be included in the protection scope of the present application.

Claims

1. A vehicle dispatching method based on a parking and charging integrated multi-level garage, characterized in that, The method comprises the following steps: acquiring real-time vehicle information and historical charging records of a vehicle to be charged; the real-time vehicle information comprises residual power, estimated driving distance, estimated external power consumption and estimated charging time; the historical charging records comprise historical vehicle information and corresponding historical charging modes; constructing a fuzzy evaluation model based on the vehicle information and the charging modes, inputting the real-time vehicle information into the fuzzy evaluation model, and obtaining a preliminary charging mode; specifically, taking the vehicle information as a factor set and the charging modes as an evaluation set; the charging modes comprise first fast charging, second fast charging, first slow charging and second slow charging; elements in the factor set are taken as rows, and elements in the evaluation set are taken as columns to construct a single-factor fuzzy evaluation matrix; the analytic hierarchy process is used to determine a factor weight vector of each vehicle information; based on the single-factor fuzzy evaluation matrix and the factor weight vector, obtaining a membership degree of the vehicle to be charged to each charging mode; determining the preliminary charging mode based on the highest membership degree; comparing the real-time vehicle information with the historical vehicle information to find a historical similar scenario; calling a historical charging mode of the historical similar scenario to correct the preliminary charging mode, and obtaining a corrected charging mode; acquiring a driver preference and a time-of-use electricity price, and optimizing a charging time length under the corrected charging mode based on multi-objective linear programming; calculating a comprehensive score based on the corrected charging mode, and determining a charging order according to the score; scheduling the vehicle to be charged to charge according to the charging order, the corrected charging mode and the charging time length.

2. The vehicle dispatching method based on the parking and charging integrated vertical garage according to claim 1, wherein The estimated external power consumption is determined according to road information of an estimated driving road.

3. The vehicle dispatching method based on the parking and charging integrated vertical garage according to claim 1, characterized in that, The comparison of the real-time vehicle information with the historical vehicle information to find the historical similar scenario specifically comprises: extracting residual power, estimated driving distance and estimated external power consumption in the current vehicle information, and comparing them with historical residual power, estimated driving distance and estimated external power consumption stored in historical charging events in the historical charging records respectively; if a deviation is less than a set threshold, the scenario is determined as a similar scenario.

4. The vehicle dispatching method based on the parking and charging integrated vertical garage according to claim 1, characterized in that, The calling of the historical charging mode of the historical similar scenario to correct the preliminary charging mode to obtain the corrected charging mode specifically comprises: adjusting a current single-factor fuzzy evaluation matrix based on a selection result of an actual charging mode corresponding to the historical similar scenario by using the least square method; dynamically updating a factor weight vector by the analytic hierarchy process combined with a driver feedback score in the historical scenario; updating the membership degree, reconfirming the charging mode and obtaining the corrected charging mode based on the adjusted single-factor fuzzy evaluation matrix and the updated factor weight vector.

5. The vehicle dispatching method based on the parking and charging integrated vertical garage according to claim 1, characterized in that, The acquisition of the driver preference and the time-of-use electricity price and the optimization of the charging time length under the corrected charging mode based on the multi-objective linear programming specifically comprises: acquiring the driver preference and the time-of-use electricity price, and constructing a bi-level programming model; wherein, an upper model is constructed based on individualized needs of the driver, and a lower model is constructed based on the time-of-use electricity price; the lower model is used to optimize a preliminary charging time length output by the upper model; solving the bi-level programming model by using linear programming to determine the charging time length under the corrected charging mode.

6. The vehicle dispatching method based on the parking and charging integrated vertical garage according to claim 1, wherein The calculation of the comprehensive score based on the corrected charging mode specifically comprises: obtaining a comprehensive score based on an updated membership degree and an evaluation level score vector; the evaluation level score vector is a preset vector.

7. A vehicle dispatching system based on a parking and charging integrated multi-level garage, characterized in that, The method comprises the following steps: The basic information acquisition module is configured to acquire real-time vehicle information and historical charging records of the vehicle to be charged; the real-time vehicle information includes residual power, estimated driving distance, estimated external power consumption, and estimated charging time; and the historical charging records include historical vehicle information and corresponding historical charging methods. The preliminary strategy acquisition module is configured to construct a fuzzy evaluation model based on the vehicle information and the charging methods, input the real-time vehicle information into the fuzzy evaluation model, and obtain a preliminary charging method; specifically, the vehicle information is taken as a factor set, and the charging methods are taken as an evaluation set; the charging methods include first-level fast charging, second-level fast charging, first-level slow charging, and second-level slow charging; elements in the factor set are taken as rows, and elements in the evaluation set are taken as columns to construct a single-factor fuzzy evaluation matrix; and an analytic hierarchy process is used to determine a factor weight vector of each vehicle information. Based on the single-factor fuzzy evaluation matrix and the factor weight vector, a membership degree of the vehicle to be charged to each charging method is obtained. The preliminary charging method is determined based on the highest membership degree. The correction strategy acquisition module is configured to compare the real-time vehicle information with the historical vehicle information, and find a historical similar scenario. The historical charging method of the historical similar scenario is called to correct the preliminary charging method, and a corrected charging method is obtained. The charging duration acquisition module is configured to acquire a driver preference and a time-of-use electricity price, and optimize a charging duration under the corrected charging method based on multi-objective linear programming. The scheduling execution module is configured to calculate a comprehensive score based on the corrected charging method, and determine a charging order according to the score. The vehicle to be charged is scheduled to be charged according to the charging order, the corrected charging method, and the charging duration.

8. A computer-readable storage medium having stored thereon a computer program, characterized in that, The program is executed by the processor to implement the steps in the vehicle scheduling method based on the parking and charging integrated stereo garage according to any one of claims 1-6.

9. A computer device comprising a memory, a processor, and a computer program stored on the memory and executable on the processor, characterized in that, The processor executes the program to implement the steps in the vehicle scheduling method based on the parking and charging integrated stereo garage according to any one of claims 1-6.

Images