CN111832881A - Method, medium and electronic device for predicting electric vehicle energy consumption based on road condition information - Google Patents
Method, medium and electronic device for predicting electric vehicle energy consumption based on road condition information Download PDFInfo
- Publication number
- CN111832881A CN111832881A CN202010361495.6A CN202010361495A CN111832881A CN 111832881 A CN111832881 A CN 111832881A CN 202010361495 A CN202010361495 A CN 202010361495A CN 111832881 A CN111832881 A CN 111832881A
- Authority
- CN
- China
- Prior art keywords
- historical
- track
- data
- electric vehicle
- driving
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
- G06Q10/063—Operations research, analysis or management
- G06Q10/0631—Resource planning, allocation, distributing or scheduling for enterprises or organisations
- G06Q10/06311—Scheduling, planning or task assignment for a person or group
- G06Q10/063114—Status monitoring or status determination for a person or group
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/04—Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"
Abstract
The invention provides a method, a medium and electronic equipment for predicting electric vehicle energy consumption based on road condition information, wherein the method comprises the following steps: acquiring multiple groups of historical battery data of the electric vehicle, and historical driving tracks and historical road condition data corresponding to each group of historical battery data as historical sample data of the electric vehicle; obtaining an incidence relation between a historical battery electric quantity change value of the electric vehicle and the historical driving track and the historical road condition data according to the historical sample data; acquiring order request information, and acquiring an order track and actual road condition data corresponding to the order track according to the order request information; and determining the energy consumption value required by the electric vehicle to complete the order by combining the incidence relation according to the order track and the actual road condition data. By the scheme, the prediction result of the energy consumption value of the electric vehicle is more accurate on the basis of combining the driving track and road condition data of the electric vehicle.
Description
Technical Field
The invention relates to the technical field of new energy vehicle application, in particular to a method, medium and electronic equipment for predicting electric vehicle energy consumption based on road condition information.
Background
At present, the battery endurance mileage is a key factor hindering the development of electric vehicles, and especially when the temperature is low in winter, the endurance mileage of electric vehicles is greatly reduced, which may affect the development and popularization of electric vehicles. On the other hand, a large number of electric vehicles are added into the network appointment vehicle row at present, if the order platform does not consider the residual electric quantity and the residual endurance mileage of the electric vehicle, the order platform can give orders blindly, and the problem that the order mileage exceeds the residual endurance mileage of the electric vehicle can be caused. Under the condition, the driver can only be forced to cancel the order, and the order dispatching efficiency of the platform is greatly influenced. In summary, energy consumption prediction of electric vehicles is very important for drivers, online booking and dispatching platforms and automobile manufacturers.
In the prior art, the energy consumption of the electric vehicle is estimated according to the driving data of the electric vehicle which drives in the historical time period, but the inventor finds that the energy consumption estimation result of the electric vehicle obtained by combining the historical driving data of the electric vehicle still has inaccurate condition in the process of implementing the invention, so that the method for estimating the energy consumption of the electric vehicle needs to be further improved.
Disclosure of Invention
The embodiment of the invention aims to provide a method, a medium and electronic equipment for predicting electric vehicle energy consumption based on road condition information, so as to solve the technical problem of inaccurate order distribution caused by low accuracy of electric vehicle electric quantity consumption estimation results in the prior art.
Therefore, the invention provides a method for predicting electric vehicle energy consumption based on road condition information, which comprises the following steps:
acquiring multiple groups of historical battery data of the electric vehicle, and historical driving tracks and historical road condition data corresponding to each group of historical battery data as historical sample data of the electric vehicle;
obtaining an incidence relation between a historical battery electric quantity change value of the electric vehicle and the historical driving track and the historical road condition data according to the historical sample data;
acquiring order request information, and acquiring an order track and actual road condition data corresponding to the order track according to the order request information;
and determining the energy consumption value required by the electric vehicle to complete the order by combining the incidence relation according to the order track and the actual road condition data.
Optionally, in the method for predicting energy consumption of an electric vehicle based on road condition information, the step of obtaining multiple sets of historical battery data of the electric vehicle, and the historical driving track and the historical road condition data corresponding to each set of historical battery data as historical sample data of the electric vehicle includes:
each historical driving track point comprises a plurality of track points, the first track point is used as a starting point, and the last track point is used as an end point;
the difference value between a first battery SOC value corresponding to the starting point and a second battery SOC value corresponding to the ending point of each historical driving track is used as a historical battery electric quantity change value corresponding to the historical driving track;
each historical driving track data comprises a driving range at a time point, a position coordinate, an altitude and/or a temperature value corresponding to the historical driving track;
the historical road condition data comprises the number of lanes, the speed limit value and the number of signal lamps of the road section to which the historical driving track belongs.
Optionally, in the method for predicting energy consumption of an electric vehicle based on road condition information, a time point, a driving distance, a position coordinate, an altitude and/or a temperature value corresponding to a historical driving track are obtained according to the following method:
and taking the average value of the time points, the average value of the driving mileage, the average value of the position coordinates, the average value of the altitude and/or the average value of the temperature values corresponding to all track points in the historical driving track as the time points, the driving mileage, the position coordinates, the altitude and/or the temperature values corresponding to the historical driving track.
Optionally, in the method for predicting energy consumption of an electric vehicle based on road condition information, the step of obtaining an association relationship between a historical battery power change of the electric vehicle and the historical driving track and the historical road condition data according to the historical sample data includes:
taking a time point, a driving mileage, a position coordinate, an altitude and/or a temperature value corresponding to each historical driving track as a static input sample; taking the number of lanes, the speed limit value and the number of signal lamps in each historical road condition data as a sequence input characteristic sample; taking a historical battery electric quantity change value corresponding to each historical driving track as an output characteristic sample;
training the historical sample data by using a global information learning model and a sequence information learning model; the static characteristic input samples are input into the global information learning model, the sequence characteristic input samples are input into the sequence information learning model, the output characteristic samples are used as the output of the global information learning model and the output of the sequence information learning model for training, and the trained global information learning model and the trained sequence information learning model are used as energy consumption estimation models for representing the incidence relation.
Optionally, in the method for predicting energy consumption of an electric vehicle based on road condition information, according to the order track and the actual road condition data, and in combination with the association relationship, determining that the electric vehicle completes the predicted energy consumption value of the order, the step of:
dividing the order track into a plurality of sections of driving tracks, wherein each driving track comprises a plurality of coordinate points;
acquiring the average value of predicted time points, the average value of traveled mileage, the average value of position coordinates, the average value of altitude and/or temperature values corresponding to all the coordinate points in each section of travel track as static input parameters; acquiring the number of lanes, the speed limit value and the number of signal lamps of a road section to which each section of driving track belongs as sequence input parameters;
inputting the static input parameters into the trained global information learning model to obtain first electric quantity estimation data of each driving track; inputting the sequence input parameters into the trained sequence information learning model to obtain second electric quantity estimation data of each driving track;
obtaining a predicted energy consumption value corresponding to each driving track according to the first electric quantity estimated data and the second electric quantity estimated data of each driving track;
and obtaining the energy consumption value required by the electric vehicle to complete the order after the predicted energy consumption values corresponding to all the driving tracks in the order tracks are summed.
Optionally, in the method for predicting energy consumption of an electric vehicle based on road condition information, in the step of obtaining the predicted energy consumption value corresponding to each driving track according to the first electric quantity predicted data and the second electric quantity predicted data of each driving track:
inputting the first electric quantity estimated data and the second electric quantity estimated data of each driving track into a regression prediction model for regression operation to obtain the predicted energy consumption value.
Optionally, in the method for predicting energy consumption of an electric vehicle based on road condition information, a time point, a driving distance, a position coordinate, an altitude and/or a temperature value corresponding to a historical driving track are obtained according to the following method:
and the time point, the driving mileage, the position coordinate, the altitude and/or the temperature value corresponding to each track point in the historical driving track are used as the time point, the driving mileage, the position coordinate, the altitude and/or the temperature value corresponding to the historical driving track.
Optionally, in the method for predicting energy consumption of an electric vehicle based on road condition information, the step of obtaining an association relationship between a historical battery power change of the electric vehicle and the historical driving track and the historical road condition data according to the historical sample data includes:
taking a time point, a driving mileage, a position coordinate, an altitude and/or a temperature value corresponding to each historical driving track, and the number of lanes, a speed limit value and the number of signal lamps in each historical road condition data as sequence input characteristic samples; taking a historical battery electric quantity change value corresponding to each historical driving track as an output characteristic sample;
training the historical sample data by using a sequence information learning model; the sequence characteristic input sample is used as the input of the sequence information learning model, the output characteristic sample is used as the output of the sequence information learning model, and the sequence information learning model is trained; and taking the trained sequence information learning model as an energy consumption estimation model for representing the incidence relation.
Optionally, in the method for predicting energy consumption of an electric vehicle based on road condition information, according to the order track and the actual road condition data, and in combination with the association relationship, determining that the electric vehicle completes the predicted energy consumption value of the order, the step of:
dividing the order track into a plurality of sections of driving tracks, wherein each driving track comprises a plurality of coordinate points;
acquiring a predicted time point, a predicted driving mileage, a position coordinate, an altitude and/or a temperature value corresponding to each coordinate point in each section of the driving track, and the number of lanes, the speed limit value and the number of signal lamps of a road section to which each section of the driving track belongs as sequence input parameters;
inputting the sequence input parameters into the trained sequence information learning model to obtain a predicted energy consumption value corresponding to each driving track;
and obtaining the energy consumption value required by the electric vehicle to complete the order after the predicted energy consumption values corresponding to all the driving tracks in the order tracks are summed.
Optionally, in the method for predicting energy consumption of an electric vehicle based on road condition information, the step of obtaining multiple sets of historical battery data of the electric vehicle and the historical driving track and the historical road condition data corresponding to each set of historical battery data as historical sample data of the electric vehicle includes:
in the running process of the electric vehicle, acquiring single-point battery data and single-point running track data of the electric vehicle at each track point according to a preset sampling moment;
obtaining road section information of a road section where the electric vehicle is located according to the single-point driving track data at the preset sampling moment, and obtaining single-point road condition data corresponding to the preset sampling moment according to the road section information;
correlating the preset sampling moment, the corresponding single-point battery data, the corresponding single-point driving data and the corresponding road condition data, and then storing to obtain single-point historical sample data of a single track point;
and forming history sample data of a section of historical driving track by the single-point history sample data of a plurality of continuous track points.
Based on the same inventive concept, the invention also provides a computer-readable storage medium, wherein the storage medium is stored with program instructions, and the computer reads the program instructions and then executes any one of the above methods for predicting the energy consumption of the electric vehicle based on the road condition information.
Based on the same inventive concept, the invention further provides electronic equipment which is characterized by comprising at least one processor and at least one memory, wherein program instructions are stored in the at least one memory, and the at least one processor executes any one of the methods for predicting the energy consumption of the electric vehicle based on the road condition information after reading the program instructions.
Compared with the prior art, the technical scheme provided by the embodiment of the invention at least has the following beneficial effects:
according to the method, the medium and the electronic equipment for predicting the energy consumption of the electric vehicle based on the road condition information, provided by the invention, the electric quantity change of the electric vehicle can be determined by analyzing the incidence relation between the historical battery data of the electric vehicle and the historical driving track and the historical road condition data. In the scheme of the invention, the change value of the electric quantity of the battery of the electric vehicle is influenced by the driving track and road condition data. And firstly, obtaining an order track and a road condition corresponding to the order track according to the order request information before dispatching the order after obtaining the order request information, and obtaining an energy consumption value required by the electric vehicle to finish the order according to the association relation. According to the scheme, the energy consumption prediction result of the electric vehicle can be obtained according to the actual road condition of the order and the historical driving data of the electric vehicle, the obtained energy consumption value prediction result of the electric vehicle is more accurate, and the energy consumption prediction result is used as the order dispatching condition, so that the situation of wrong order dispatching can be effectively avoided.
Drawings
Fig. 1 is a method for predicting energy consumption of an electric vehicle based on road condition information according to an embodiment of the invention;
FIG. 2 is a schematic diagram of a driving track of an electric vehicle according to an embodiment of the present invention;
fig. 3 is a flowchart illustrating a method for predicting energy consumption of an electric vehicle based on road condition information according to another embodiment of the present invention;
fig. 4 is a schematic diagram of a hardware connection relationship of an electronic device according to an embodiment of the present invention.
Detailed Description
The embodiments of the present invention will be further described with reference to the accompanying drawings. In the description of the present invention, it should be noted that the terms "center", "upper", "lower", "left", "right", "vertical", "horizontal", "inner", "outer", etc., indicate orientations or positional relationships based on the orientations or positional relationships shown in the drawings, and are only for convenience of description of the present invention, and do not indicate or imply that the device or assembly referred to must have a specific orientation, be constructed and operated in a specific orientation, and thus, should not be construed as limiting the present invention. Furthermore, the terms "first," "second," and "third" are used for descriptive purposes only and are not to be construed as indicating or implying relative importance. Wherein the terms "first position" and "second position" are two different positions.
In the description of the present invention, it should be noted that, unless otherwise explicitly specified or limited, the terms "mounted," "connected," and "connected" are to be construed broadly, e.g., as meaning either a fixed connection, a removable connection, or an integral connection; can be mechanically or electrically connected; the two components can be directly connected or indirectly connected through an intermediate medium, and the two components can be communicated with each other. The specific meanings of the above terms in the present invention can be understood in specific cases to those skilled in the art.
The technical schemes in the following embodiments provided by the invention can be combined with each other unless contradictory to each other, and technical features in different schemes can be replaced with each other.
Example 1
The embodiment provides a method for predicting electric vehicle energy consumption based on road condition information, which can be applied to a cloud server, wherein the server can realize data communication with an electric vehicle driving database and can also realize data communication with a vehicle-mounted control device on an electric vehicle, and meanwhile, the cloud server can access a city traffic network database based on a network, so that road condition data of a road section where a certain point is located can be determined based on a geographical position coordinate of the certain point, and the road condition of the road section where the electric vehicle is located in the driving process can be judged in real time. Specifically, as shown in fig. 1, the method includes the steps of:
s101: acquiring multiple groups of historical battery data of the electric vehicle, and historical driving tracks and historical road condition data corresponding to each group of historical battery data as historical sample data of the electric vehicle; preferably, each piece of historical travel track data comprises position coordinates of track points, time points corresponding to each track point, position coordinates, altitude and/or temperature values; the historical road condition data comprises the number of lanes, the speed limit value and the number of signal lamps of the road section to which the historical driving track belongs. As described above, the cloud server can obtain the running state data of each electric vehicle in the running process, wherein the running state data at least includes the running track, the road condition data of the running section of the electric vehicle, the remaining electric quantity value of the battery, the cruising mileage, the running speed, the time and the like.
Specifically, the historical travel data of the electric vehicle can be obtained by the following method: in the running process of the electric vehicle, acquiring single-point battery data and single-point running track data of the electric vehicle at each track point according to a preset sampling moment; obtaining road section information of a road section where the electric vehicle is located according to the single-point driving track data at the preset sampling moment, and obtaining single-point road condition data corresponding to the preset sampling moment according to the road section information; correlating the preset sampling moment, the corresponding single-point battery data, the corresponding single-point driving data and the corresponding road condition data, and then storing to obtain single-point historical sample data of a single track point; and forming history sample data of a section of historical driving track by the single-point history sample data of a plurality of continuous track points. That is to say, during the running process of the electric vehicle, the position coordinates of the position of the electric vehicle and the electric quantity value of the battery are collected according to the preset sampling time (the electric quantity value can be represented by the SOC value of the battery); the road section of which city the electric vehicle is in can be positioned according to the position coordinates, and road condition data such as the number of lanes, the speed limit value, the traffic signal lamp setting and the like of the road section can be determined according to the city road network data. The location coordinates include latitude and longitude coordinates and altitude. The longitude, latitude and altitude can be directly acquired by a positioning sensor arranged on the vehicle. The influence of the climbing process and the downhill process in the driving process of the vehicle on the electric energy consumption can be combined into the electric energy consumption estimation model by considering the altitude, so that the model is more fit with the actual situation. The time interval between two adjacent sampling times can be about ten seconds, so the running track is formed by one track point every ten seconds, and each track point is associated with the running data of all the electric vehicles collected at the sampling time of the track point.
S102: and obtaining the incidence relation between the historical battery electric quantity change value of the electric vehicle and the historical driving track and the historical road condition data according to the historical sample data. For the electric quantity change value of the electric vehicle, the road condition data is also used as a parallel reference index for analysis besides considering the influence of the driving track. For example, when the number of the signal lamps is large in the travel track with the same length, the number of the start and stop of the electric vehicle may increase, which may affect the change of the battery capacity of the electric vehicle. Each historical driving track point comprises a starting point, an end point and at least one track point between the starting point and the end point, and in the step, a difference value between a first battery SOC value corresponding to the starting point and a second battery SOC value corresponding to the end point of each historical driving track is used as a historical battery electric quantity change value corresponding to the historical driving track.
S103: acquiring order request information, and acquiring an order track and actual road condition data corresponding to the order track according to the order request information. The order request information is sent by the passenger according to the actual demand of the passenger, and the order request information can be realized by adopting APP installed on terminals such as a mobile phone. The order request information comprises a starting place and a destination place, an order track can be obtained after a proper driving path is selected, each road section on the whole order track can be determined after the path is determined, and then road condition data of each road section is obtained, so that actual road condition data are obtained.
S104: and determining the energy consumption value required by the electric vehicle to complete the order by combining the incidence relation according to the order track and the actual road condition data. In step S102, the correlation between the travel track point and the road condition data and the battery power variation value is determined, and the energy consumption value corresponding to the order track and the actual road condition data can be predicted according to the correlation.
According to the scheme, through analyzing the incidence relation between the historical battery data of the electric vehicle and the historical driving track and the historical road condition data, the fact that the electric quantity change of the electric vehicle is influenced by the driving track and the road condition data can be determined. Before the order dispatching operation is executed, the order track and the road condition corresponding to the order track are obtained according to the order request information, and the energy consumption value required by the electric vehicle to complete the order can be obtained according to the incidence relation among the order track, the road condition and the road condition obtained according to historical data. By the scheme, the energy consumption value prediction result of the electric vehicle is more accurate on the basis of considering road condition data.
In the above scheme, the electric vehicle can be a network appointment vehicle, and all historical driving data of the electric vehicle are stored in the cloud server of the network appointment vehicle platform, so that when the energy consumption of the electric vehicle is predicted, the historical driving data corresponding to the electric vehicle can be directly extracted from the cloud server of the network appointment vehicle platform. Generally, the historical driving data can reflect the performance of the electric vehicle and the driving habit of the vehicle, so that the energy consumption value prediction result of the electric vehicle can be obtained by analyzing a large amount of historical driving data and combining multiple influence factors such as the performance of the vehicle, the driving habit or style of a driver, road condition data and the like, and the method has higher accuracy compared with the conventional prediction method.
In addition, for each electric vehicle, a specific relationship exists among the travel track, the time and the electric quantity consumption value, for example, the longer the travel track is, the longer the travel time is, the larger the energy consumption is, the time interval between two adjacent sampling times is within a certain range, and the like, and the historical data of the electric vehicle can be cleaned according to the inevitable correspondence, for example, the feature filling is performed if the feature value is missing. And if the coordinate of the driving track deviates, error correction is carried out, and the obtained historical sample data can be ensured to be more in line with the actual situation, so that the accuracy of the energy consumption prediction result of the electric vehicle can be improved. Obtaining time points, position coordinates, altitude and/or temperature values corresponding to the historical driving track according to the following modes:
the first method is as follows: and taking the time point mean value, the position coordinate mean value, the altitude mean value and/or the temperature value mean value corresponding to all track points in the historical driving track as the time point, the position coordinate, the altitude and/or the temperature value corresponding to the historical driving track.
The second method comprises the following steps: and taking the time point, the position coordinate, the altitude and/or the temperature value corresponding to each track point in the historical driving track as the time point, the position coordinate, the altitude and/or the temperature value corresponding to the historical driving track.
Referring to fig. 3, a schematic diagram of a small section of driving track of an electric vehicle is shown, in the diagram, a1 is a starting point, a9 is a terminal point, and a1-a9 are 7 track points, where a data sampling interval between two adjacent track points is about ten seconds, and in an actual application scenario, the number of the track points can be adjusted according to data processing amount and processing capacity of equipment. The historical travel data of the electric vehicle may include a large number of historical travel tracks, which are divided into a plurality of track segments shown in fig. 3 according to distance or time period. Each track point in the map is associated with a geographic position coordinate, an altitude, a temperature, sampling time, a battery electric quantity value and the like, and if the running speed of the electric vehicle can be acquired, the running speed can be increased in a static input variable. In the first mode, the data of each track point in the historical driving track is subjected to operation processing, so that the data finally corresponding to one section of the historical driving track can be only one value, and the processing has the advantages of smaller operation data amount and lower requirement on a mathematical model. In the second mode, the data of each track point in the historical driving track are all reserved and used as the data corresponding to the historical driving track, so that the processing has the advantages that the data are more comprehensive, and the obtained association relationship is more accurate. Particularly, the historical sample data in the second mode reflects the actual road condition data more truly, for example, the electric vehicle firstly passes through an uphill road section and then passes through a downhill road section in the driving process (for example, the process of a4-a 7), if the historical driving data and the road condition data in the whole historical track are subjected to average processing, the data of the uphill road section and the data of the downhill road section may be mutually offset, so that the result that the electric vehicle drives on a gentle road section is obtained, and if the data of each position point in the historical driving track are subjected to retention operation, the actual driving process of the electric vehicle can be accurately reflected.
In the foregoing solution, when sample data of a historical driving trajectory is obtained in the first mode, the historical driving data may be regarded as static data, and then, as shown in fig. 2, the step of obtaining an association relationship between a historical battery power change of the electric vehicle and the historical driving trajectory and the historical road condition data according to the historical sample data includes:
s201: taking a time point, a battery electric quantity value, a driving mileage, a position coordinate, an altitude and/or a temperature value corresponding to each historical driving track as a static input sample; and taking the number of lanes, the speed limit value and the number of signal lamps in each historical road condition data as a sequence input characteristic sample.
S202: and taking the historical battery power change value corresponding to each historical driving track as an output characteristic sample. As mentioned above, the SOC value at A9 is subtracted from the SOC value at A1 to obtain the SOC value.
S203: training the historical sample data by using a global information learning model and combining a sequence information learning model; the static characteristic input samples are input into the global information learning model, the sequence characteristic input samples are input into the sequence information learning model, and the output characteristic samples are used as the output of the global information learning model and the output of the sequence information learning model for training. The history sample data includes a plurality of sets of input and output parameters of the travel locus as shown in fig. 3. And the data of each section of the driving track is correspondingly taken as input and output parameters and substituted into the global information learning model and sequence information learning model. In order to make the model training result more accurate, the historical sample data can be divided into two groups, wherein 80% of the data is used as training sample data, and 20% of the data is used as test sample data. And testing the trained model by using the number of the test samples, and stopping training after the test is passed.
S204: and taking the trained global information learning model and the trained sequence information learning model as energy consumption estimation models for representing the incidence relation.
In this embodiment, a static input sample corresponding to the historical driving trajectory is obtained in a second way, and the key point is to select a learning model: global model + recurrentmodel (i.e., the global information learning model in combination with the sequence information learning model). The global information learning model is similar to a traditional regression model in function, and learning training is carried out on the global information of the journey; the sequence information learning model is focused on learning and training local details such as sequences. Since most regression models such as XGBoost must receive input vectors of fixed length, and the variation of the distance data in the road section corresponding to a section of the driving track may be large, the conventional regression models generally omit features at the sequence level and use the overall statistics instead. In the scheme, in order to better retain the characteristics of the sequence level, the global information learning model selects a Wide & Deep model, the Wide branch of the Wide & Deep model carries out second-order intersection on the characteristic values of static input parameters such as the predicted time point, the position coordinate, the altitude and/or the temperature value, the historical driving data has a certain memory function, the Deep branch is of a traditional multi-layer perceptron structure, and the Deep model has better generalization capability on the characteristic values of the static input parameters such as the predicted time point, the position coordinate, the altitude and/or the temperature value. The two branches are strongly connected, and the length of each branch can be taken. The sequence information learning model carries out detailed modeling on the track, and can capture the characteristics of each sequence (the number of lanes, the speed limit value and the number of traffic lights) in the track. The sequence information learning model is an LSTM (time series prediction) model, and traffic data can be brought into the model in the scheme, and the last Hidden State of the model is used as output. The outputs of the Wide branch module and the Wide branch module are used as first electric quantity estimation data, the output of the LSTM model is used as second electric quantity estimation data, output vectors of the three modules are spliced together and enter a final regression model for prediction, and a final energy consumption prediction value is obtained.
In the above technical solution in this embodiment, step S104 can be implemented as follows:
s301: and dividing the order track into a plurality of sections of driving tracks, wherein each driving track comprises a plurality of coordinate points.
S302: acquiring the average value of the predicted time points, the average value of the predicted battery electric quantity value, the average value of the predicted driving mileage, the average value of the position coordinates, the average value of the altitude and/or the temperature value corresponding to all the coordinate points in each section of driving track as static input parameters; and acquiring the number of lanes, the speed limit value and the number of signal lamps of the road section to which each section of the driving track belongs as sequence input parameters. The predicted time point and the like can be obtained by combining the data obtained according to the order path and the road condition information.
S303: inputting the static input parameters into the trained global information learning model to obtain first electric quantity estimation data of each driving track; and inputting the sequence input parameters into the trained sequence information learning model to obtain second electric quantity estimation data of each driving track.
S304: and obtaining a predicted energy consumption value corresponding to each driving track according to the first electric quantity estimated data and the second electric quantity estimated data of each driving track.
S305: and obtaining the energy consumption value required by the electric vehicle to complete the order after the predicted energy consumption values corresponding to all the driving tracks in the order tracks are summed.
Preferably, the manner of acquiring the travel locus in step S301 and the manner of acquiring the history travel locus are kept consistent. For example, if the historical travel track includes 9 track points, each travel track in the order may be divided into 9 track points. The purpose of processing in this way is to ensure that the data base for model training by using historical driving data and the data base for calculating energy consumption have consistency, so as to ensure that the estimated energy consumption value of the electric vehicle according to the order is consistent with the execution process of the actual order.
In the foregoing solution, when sample data of a historical driving trajectory is obtained in the second mode, the historical driving data may be regarded as sequence data, and when all the sample data of the historical driving trajectory is sequence data, the step of obtaining the association relationship between the historical battery power change of the electric vehicle and the historical driving trajectory and the historical road condition data according to the historical sample data includes:
s401: taking a time point, a battery electric quantity value, a driving mileage, a position coordinate, an altitude and/or a temperature value corresponding to each historical driving track, and the number of lanes, a speed limit value and a signal lamp in each historical road condition data as a sequence input feature sample; and taking the historical battery power change value corresponding to each historical driving track as an output characteristic sample.
S402: training the historical sample data by using a sequence information learning model; the sequence characteristic input sample is used as the input of the sequence information learning model, the output characteristic sample is used as the output of the sequence information learning model, and the sequence information learning model is trained; and taking the trained sequence information learning model as an energy consumption estimation model for representing the incidence relation.
Correspondingly, the step of determining the predicted energy consumption value of the electric vehicle for completing the order by combining the incidence relation according to the order track and the actual road condition data comprises the following steps:
s501: dividing the order track into a plurality of sections of driving tracks, wherein each driving track comprises a plurality of coordinate points; the manner of acquiring the travel track and the manner of acquiring the historical travel track are kept consistent.
S502: and acquiring a predicted time point, a predicted battery electric quantity value, a predicted driving mileage, a position coordinate, an altitude and/or a temperature value corresponding to each coordinate point in each section of the driving track, and the number of lanes, the speed limit value and the number of signal lamps of a road section to which each section of the driving track belongs as sequence input parameters.
S503: inputting the sequence input parameters into the trained sequence information learning model to obtain a predicted energy consumption value corresponding to each driving track;
s504: and obtaining the energy consumption value required by the electric vehicle to complete the order after the predicted energy consumption values corresponding to all the driving tracks in the order tracks are summed.
The sequence information learning model can select an LSTM model (time series prediction model), and sequence input parameters can be brought into the model in the scheme, and the last Hidden State of the sequence input parameters is used as output. According to the scheme, the energy consumption prediction result of the electric vehicle is obtained by adopting the data of each track point in the actual running track of the electric vehicle, and the accuracy is higher.
Example 2
The present embodiment provides a computer-readable storage medium, where a computer program is stored in the storage medium, and the computer program is executed by a computer to implement the method for predicting energy consumption of an electric vehicle based on road condition information according to any one of the technical solutions in embodiment 1.
Example 3
The present embodiment provides an electronic device, as shown in fig. 4, including at least one processor 401 and at least one memory 402, where instruction information is stored in the at least one memory 402, and after the at least one processor 401 reads the program instruction, the method for predicting energy consumption of an electric vehicle based on road condition information according to any one of the embodiments 1 may be executed.
The above apparatus may further include: an input device 403 and an output device 404. The processor 401, memory 402, input device 403, and output device 404 may be connected by a bus or other means. The product can execute the method provided by the embodiment of the application, and has the corresponding functional modules and beneficial effects of the execution method. For technical details that are not described in detail in this example, reference is made to the method provided in example 1 of the present application.
Finally, it should be noted that: the above examples are only intended to illustrate the technical solution of the present invention, but not to limit it; although the present invention has been described in detail with reference to the foregoing embodiments, it will be understood by those of ordinary skill in the art that: the technical solutions described in the foregoing embodiments may still be modified, or some technical features may be equivalently replaced; and such modifications or substitutions do not depart from the spirit and scope of the corresponding technical solutions of the embodiments of the present invention.
Claims (12)
1. A method for predicting electric vehicle energy consumption based on road condition information is characterized by comprising the following steps:
acquiring multiple groups of historical battery data of the electric vehicle, and historical driving tracks and historical road condition data corresponding to each group of historical battery data as historical sample data of the electric vehicle;
obtaining an incidence relation between a historical battery electric quantity change value of the electric vehicle and the historical driving track and the historical road condition data according to the historical sample data;
acquiring order request information, and acquiring an order track and actual road condition data corresponding to the order track according to the order request information;
and determining the energy consumption value required by the electric vehicle to complete the order by combining the incidence relation according to the order track and the actual road condition data.
2. The method for predicting energy consumption of an electric vehicle based on road condition information according to claim 1, wherein the step of obtaining a plurality of sets of historical battery data of the electric vehicle and the historical driving track and the historical road condition data corresponding to each set of historical battery data as historical sample data of the electric vehicle comprises:
each historical driving track point comprises a plurality of track points, the first track point is used as a starting point, and the last track point is used as an end point;
the difference value between a first battery SOC value corresponding to the starting point and a second battery SOC value corresponding to the ending point of each historical driving track is used as a historical battery electric quantity change value corresponding to the historical driving track;
each historical driving track data comprises a time point, a driving mileage, a position coordinate, an altitude and/or a temperature value corresponding to the historical driving track;
the historical road condition data comprises the number of lanes, the speed limit value and the number of signal lamps of the road section to which the historical driving track belongs.
3. The method for predicting energy consumption of an electric vehicle based on road condition information as claimed in claim 2, wherein the time point, the driving distance, the position coordinates, the altitude and/or the temperature value corresponding to the historical driving track are obtained according to the following method:
and taking the average value of the time points, the average value of the driving mileage, the average value of the position coordinates, the average value of the altitude and/or the average value of the temperature values corresponding to all track points in the historical driving track as the time points, the driving mileage, the position coordinates, the altitude and/or the temperature values corresponding to the historical driving track.
4. The method according to claim 3, wherein the step of obtaining the correlation between the historical battery power change of the electric vehicle and the historical driving track and the historical road condition data according to the historical sample data comprises:
taking a time point, a driving mileage, a position coordinate, an altitude and/or a temperature value corresponding to each historical driving track as a static input sample; taking the number of lanes, the speed limit value and the number of signal lamps in each historical road condition data as a sequence input characteristic sample; taking a historical battery electric quantity change value corresponding to each historical driving track as an output characteristic sample;
training the historical sample data by using a global information learning model and a sequence information learning model; the static characteristic input samples are input into the global information learning model, the sequence characteristic input samples are input into the sequence information learning model, the output characteristic samples are used as the output of the global information learning model and the output of the sequence information learning model for training, and the trained global information learning model and the trained sequence information learning model are used as energy consumption estimation models for representing the incidence relation.
5. The method according to claim 4, wherein the step of determining the predicted energy consumption value of the electric vehicle for completing the order according to the order track and the actual road condition data in combination with the correlation comprises:
dividing the order track into a plurality of sections of driving tracks, wherein each driving track comprises a plurality of coordinate points;
acquiring the average value of predicted time points, the average value of traveled mileage, the average value of position coordinates, the average value of altitude and/or temperature values corresponding to all the coordinate points in each section of travel track as static input parameters; acquiring the number of lanes, the speed limit value and the number of signal lamps of a road section to which each section of driving track belongs as sequence input parameters;
inputting the static input parameters into the trained global information learning model to obtain first electric quantity estimation data of each driving track; inputting the sequence input parameters into the trained sequence information learning model to obtain second electric quantity estimation data of each driving track;
obtaining a predicted energy consumption value corresponding to each driving track according to the first electric quantity estimated data and the second electric quantity estimated data of each driving track;
and obtaining the energy consumption value required by the electric vehicle to complete the order after the predicted energy consumption values corresponding to all the driving tracks in the order tracks are summed.
6. The method of claim 5, wherein the step of obtaining the predicted energy consumption value corresponding to each driving track according to the first and second electric quantity predicted data of each driving track comprises:
inputting the first electric quantity estimated data and the second electric quantity estimated data of each driving track into a regression prediction model for regression operation to obtain the predicted energy consumption value.
7. The method for predicting energy consumption of an electric vehicle based on road condition information as claimed in claim 2, wherein the time point, the driving distance, the position coordinates, the altitude and/or the temperature value corresponding to the historical driving track are obtained according to the following method:
and the time point, the driving mileage, the position coordinate, the altitude and/or the temperature value corresponding to each track point in the historical driving track are used as the time point, the driving mileage, the position coordinate, the altitude and/or the temperature value corresponding to the historical driving track.
8. The method of claim 7, wherein the step of obtaining the correlation between the historical battery power change of the electric vehicle and the historical driving path and the historical road condition data according to the historical sample data comprises:
taking a time point, a driving mileage, a position coordinate, an altitude and/or a temperature value corresponding to each historical driving track, and the number of lanes, a speed limit value and the number of signal lamps in each historical road condition data as sequence input characteristic samples; taking a historical battery electric quantity change value corresponding to each historical driving track as an output characteristic sample;
training the historical sample data by using a sequence information learning model; the sequence characteristic input sample is used as the input of the sequence information learning model, the output characteristic sample is used as the output of the sequence information learning model, and the sequence information learning model is trained; and taking the trained sequence information learning model as an energy consumption estimation model for representing the incidence relation.
9. The method of claim 8, wherein the step of determining the predicted energy consumption value of the electric vehicle for completing the order according to the order track and the actual road condition data in combination with the correlation comprises:
dividing the order track into a plurality of sections of driving tracks, wherein each driving track comprises a plurality of coordinate points;
acquiring a predicted time point, a predicted driving mileage, a position coordinate, an altitude and/or a temperature value corresponding to each coordinate point in each section of the driving track, and the number of lanes, the speed limit value and the number of signal lamps of a road section to which each section of the driving track belongs as sequence input parameters;
inputting the sequence input parameters into the trained sequence information learning model to obtain a predicted energy consumption value corresponding to each driving track;
and obtaining the energy consumption value required by the electric vehicle to complete the order after the predicted energy consumption values corresponding to all the driving tracks in the order tracks are summed.
10. The method for predicting energy consumption of an electric vehicle based on road condition information according to any one of claims 1-9, wherein the step of obtaining a plurality of sets of historical battery data of the electric vehicle and the historical driving track and the historical road condition data corresponding to each of the sets of historical battery data as the historical sample data of the electric vehicle comprises:
in the running process of the electric vehicle, acquiring single-point battery data and single-point running track data of the electric vehicle at each track point according to a preset sampling moment;
obtaining road section information of a road section where the electric vehicle is located according to the single-point driving track data at the preset sampling moment, and obtaining single-point road condition data corresponding to the preset sampling moment according to the road section information;
correlating the preset sampling moment, the corresponding single-point battery data, the corresponding single-point driving data and the corresponding road condition data, and then storing to obtain single-point historical sample data of a single track point;
and forming history sample data of a section of historical driving track by the single-point history sample data of a plurality of continuous track points.
11. A computer-readable storage medium, wherein the storage medium stores program instructions, and after the program instructions are read by a computer, the computer executes the method for predicting energy consumption of an electric vehicle based on road condition information according to any one of claims 1 to 10.
12. An electronic device comprising at least one processor and at least one memory:
at least one of the memories stores program instructions, and at least one of the processors executes the method for predicting energy consumption of an electric vehicle based on road condition information according to any one of claims 1 to 10 after reading the program instructions.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010361495.6A CN111832881A (en) | 2020-04-30 | 2020-04-30 | Method, medium and electronic device for predicting electric vehicle energy consumption based on road condition information |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010361495.6A CN111832881A (en) | 2020-04-30 | 2020-04-30 | Method, medium and electronic device for predicting electric vehicle energy consumption based on road condition information |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN111832881A true CN111832881A (en) | 2020-10-27 |
Family
ID=72913443
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010361495.6A Pending CN111832881A (en) | 2020-04-30 | 2020-04-30 | Method, medium and electronic device for predicting electric vehicle energy consumption based on road condition information |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111832881A (en) |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112356822A (en) * | 2020-11-12 | 2021-02-12 | 北京车和家信息技术有限公司 | Hybrid vehicle, and prediction method, device and prediction system for power generation amount of hybrid vehicle |
| CN113112059A (en) * | 2021-03-31 | 2021-07-13 | 亿海蓝（北京）数据技术股份公司 | Ship berthing time prediction method and system |
| CN114004546A (en) * | 2021-12-30 | 2022-02-01 | 南京领行科技股份有限公司 | Network appointment vehicle distribution method and device, electronic equipment and storage medium |
| CN115808922A (en) * | 2022-01-07 | 2023-03-17 | 宁德时代新能源科技股份有限公司 | Commercial electric vehicle energy consumption prediction method and device and computer equipment |
| CN116992998A (en) * | 2023-06-03 | 2023-11-03 | 隆瑞三优新能源汽车科技有限公司 | New energy bus charging condition prediction method and device, electronic equipment and medium |
| CN117689272A (en) * | 2024-02-02 | 2024-03-12 | 杭州宇谷科技股份有限公司 | Method for measuring value of ESG (electronic service provider) power exchange system, electronic equipment and storage medium |
Citations (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105501070A (en) * | 2015-12-16 | 2016-04-20 | 中国北方车辆研究所 | Power supply and power consumption balance control method for active power-driven vehicle |
| CN106064568A (en) * | 2015-04-23 | 2016-11-02 | 苏州宝时得电动工具有限公司 | Electric vehicle energy supplement system, method and apparatus |
| CN107351840A (en) * | 2017-06-07 | 2017-11-17 | 同济大学 | A kind of vehicle energy saving path and economic speed dynamic programming method based on V2I |
| CN107813814A (en) * | 2016-09-12 | 2018-03-20 | 法乐第(北京)网络科技有限公司 | Energy hole track optimizing method, hybrid vehicle for hybrid vehicle |
| CN108177648A (en) * | 2018-01-02 | 2018-06-19 | 北京理工大学 | A kind of energy management method of the plug-in hybrid vehicle based on intelligent predicting |
| CN110126841A (en) * | 2019-05-09 | 2019-08-16 | 吉林大学 | EV Energy Consumption model prediction method based on road information and driving style |
| CN110222906A (en) * | 2019-06-17 | 2019-09-10 | 北京嘀嘀无限科技发展有限公司 | Electric vehicle energy consumption prediction technique, computer readable storage medium and electronic equipment |
| US20190392706A1 (en) * | 2018-06-21 | 2019-12-26 | Volkswagen Aktiengesellschaft | Assignment Of A Stoplight To Associated Traffic Lanes |
| WO2020001261A1 (en) * | 2018-06-25 | 2020-01-02 | Beijing Didi Infinity Technology And Development Co., Ltd. | Systems and methods for estimating an arriel time of a vehicle |
| CN110675619A (en) * | 2018-07-02 | 2020-01-10 | 上海汽车集团股份有限公司 | Vehicle travel energy consumption prediction method and device |
-
2020
- 2020-04-30 CN CN202010361495.6A patent/CN111832881A/en active Pending
Patent Citations (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106064568A (en) * | 2015-04-23 | 2016-11-02 | 苏州宝时得电动工具有限公司 | Electric vehicle energy supplement system, method and apparatus |
| CN105501070A (en) * | 2015-12-16 | 2016-04-20 | 中国北方车辆研究所 | Power supply and power consumption balance control method for active power-driven vehicle |
| CN107813814A (en) * | 2016-09-12 | 2018-03-20 | 法乐第(北京)网络科技有限公司 | Energy hole track optimizing method, hybrid vehicle for hybrid vehicle |
| CN107351840A (en) * | 2017-06-07 | 2017-11-17 | 同济大学 | A kind of vehicle energy saving path and economic speed dynamic programming method based on V2I |
| CN108177648A (en) * | 2018-01-02 | 2018-06-19 | 北京理工大学 | A kind of energy management method of the plug-in hybrid vehicle based on intelligent predicting |
| US20190392706A1 (en) * | 2018-06-21 | 2019-12-26 | Volkswagen Aktiengesellschaft | Assignment Of A Stoplight To Associated Traffic Lanes |
| WO2020001261A1 (en) * | 2018-06-25 | 2020-01-02 | Beijing Didi Infinity Technology And Development Co., Ltd. | Systems and methods for estimating an arriel time of a vehicle |
| CN110675619A (en) * | 2018-07-02 | 2020-01-10 | 上海汽车集团股份有限公司 | Vehicle travel energy consumption prediction method and device |
| CN110126841A (en) * | 2019-05-09 | 2019-08-16 | 吉林大学 | EV Energy Consumption model prediction method based on road information and driving style |
| CN110222906A (en) * | 2019-06-17 | 2019-09-10 | 北京嘀嘀无限科技发展有限公司 | Electric vehicle energy consumption prediction technique, computer readable storage medium and electronic equipment |
Cited By (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112356822A (en) * | 2020-11-12 | 2021-02-12 | 北京车和家信息技术有限公司 | Hybrid vehicle, and prediction method, device and prediction system for power generation amount of hybrid vehicle |
| CN113112059A (en) * | 2021-03-31 | 2021-07-13 | 亿海蓝（北京）数据技术股份公司 | Ship berthing time prediction method and system |
| CN114004546A (en) * | 2021-12-30 | 2022-02-01 | 南京领行科技股份有限公司 | Network appointment vehicle distribution method and device, electronic equipment and storage medium |
| CN115808922A (en) * | 2022-01-07 | 2023-03-17 | 宁德时代新能源科技股份有限公司 | Commercial electric vehicle energy consumption prediction method and device and computer equipment |
| WO2023130659A1 (en) * | 2022-01-07 | 2023-07-13 | 宁德时代新能源科技股份有限公司 | Method and apparatus for predicting energy consumption of commercial electric vehicle, and computer device |
| CN115808922B (en) * | 2022-01-07 | 2023-10-27 | 宁德时代新能源科技股份有限公司 | Commercial electric vehicle energy consumption prediction method and device and computer equipment |
| CN116992998A (en) * | 2023-06-03 | 2023-11-03 | 隆瑞三优新能源汽车科技有限公司 | New energy bus charging condition prediction method and device, electronic equipment and medium |
| CN116992998B (en) * | 2023-06-03 | 2024-03-26 | 隆瑞三优新能源汽车科技有限公司 | New energy bus charging condition prediction method and device, electronic equipment and medium |
| CN117689272A (en) * | 2024-02-02 | 2024-03-12 | 杭州宇谷科技股份有限公司 | Method for measuring value of ESG (electronic service provider) power exchange system, electronic equipment and storage medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111832881A (en) | Method, medium and electronic device for predicting electric vehicle energy consumption based on road condition information | |
| CN111497679B (en) | Pure electric vehicle energy consumption monitoring optimization method and system | |
| CN106908075B (en) | Big data acquisition and processing system and electric vehicle endurance estimation method based on big data acquisition and processing system | |
| WO2020253204A1 (en) | Electric vehicle energy consumption prediction method, computer readable storage medium and electronic equipment | |
| US8694232B2 (en) | Method of predicting energy consumption, apparatus for predicting energy consumption, and terminal apparatus | |
| US8874367B2 (en) | Method for estimating and displaying range of a vehicle | |
| CN112286206B (en) | Automatic driving simulation method, system, equipment, readable storage medium and platform | |
| AU2015352509A1 (en) | Apparatus and method for vehicle economy improvement | |
| US8989999B2 (en) | Vehicle range analysis using driving environment information with optional continuous averaging | |
| CN107247234B (en) | SOC real-time online estimation method and system for vehicle power battery | |
| CN113159435B (en) | Method and system for predicting remaining driving mileage of new energy vehicle | |
| CN113970338A (en) | Travel mode recommendation method, related method, device and system | |
| CN115164922A (en) | Path planning method, system, equipment and storage medium | |
| US20220011123A1 (en) | Method of characterizing a route travelled by a user | |
| Kim et al. | A machine learning method for ev range prediction with updates on route information and traffic conditions | |
| CN110789396A (en) | Battery pack charge state data detection method, device, medium, terminal and automobile | |
| CN110303908A (en) | A kind of recommended method and system of charge capacity | |
| CN111723835A (en) | Vehicle movement track distinguishing method and device and electronic equipment | |
| CN115985083B (en) | Smart city-based shared electric vehicle management system and method | |
| KR102472552B1 (en) | Method For Calculating Energy Consumption And Carbon Discharge Of Car By Utilizing Deep Learning | |
| CN111785012B (en) | Energy consumption road spectrum analysis method for vehicle cloud cooperative computing | |
| CN114043875A (en) | Remaining mileage estimation deviation analysis method and system based on big data | |
| CN112406874A (en) | Electric automobile remote charging auxiliary decision-making method | |
| CN114256523A (en) | Charging control method and device for charging pile, electronic equipment and storage medium | |
| CN112729330A (en) | Method and device for calculating mileage of motor vehicle |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |