Vehicle fuel consumption prediction method and device and computer program product

Abstract

The invention discloses a vehicle fuel consumption prediction method and device and a computer program product, and the method comprises the steps: obtaining a vehicle inherent attribute parameter set, an environment parameter set and navigation path information containing altitude elevation; when the altitude of the destination is higher than the altitude of the current position of the vehicle, the maximum altitude difference is calculated based on the navigation path information, and the final climbing height is determined in combination with the altitude of the destination; according to the air density change corresponding to the final climbing height and the gradient feature of the navigation path, the incremental sliding resistance in the vehicle climbing process is calculated; inputting the vehicle inherent attribute parameter set and the environment parameter set into a pre-trained neural network model, and calculating the compensation oil consumption of the vehicle power system to the altitude change; calculating extra fuel consumption of the vehicle according to the incremental sliding resistance, the average speed of the vehicle in the navigation path, the compensation fuel consumption and the fuel consumption rate of an engine; and according to the current fuel consumption of the vehicle and the additional fuel consumption of the vehicle, corrected fuel consumption of the vehicle is calculated.

Description

Technical Field

[0001] The present invention relates to the technical field of intelligent networked vehicles, and in particular to a vehicle fuel consumption prediction method, device and computer program product. Background Art

[0002] The fuel consumption estimation technology of traditional fuel and hybrid vehicles is mainly based on the fusion feedback mechanism of historical driving data and real-time fuel level monitoring. Its core algorithm is reflected in the actual measurement calculation method of fuel consumption per 100 kilometers (fuel consumption ÷ mileage × 100) and the theoretical model method (engine power × fuel efficiency ÷ driving speed × time). The fuel level change data is obtained in real time through capacitive, resistive or electromagnetic fuel level sensors, and dynamically corrected in combination with odometer information. Among them:

[0003] (1) Capacitive sensors change the capacitance between the plates through changes in the dielectric constant of the fuel;

[0004] (2) The resistive sensor uses a float to drive a variable resistor to achieve voltage division detection;

[0005] (3) The electromagnetic sensor drives the rotor to deflect by changing the magnetic field of the dual coils.

[0006] The existing technology has three obvious defects: first, fuel consumption estimation is highly dependent on operating conditions, and its displayed value is actually a lagging mapping of the previous driving conditions; second, the cruising range prediction is only of reference value under similar conditions; third, it lacks forward-looking dynamic prediction capabilities.

[0007] Although some car companies have introduced navigation mileage matching technology (by memorizing historical fuel consumption, planning route mileage, calculating fuel demand and comparing current fuel volume), this solution still has the following technical limitations:

[0008] (1) A static altitude model is used, ignoring the direct impact of terrain elevation changes on engine load;

[0009] (2) The dynamic effects of aerodynamic parameters (such as changes in air density caused by altitude and oxygen content attenuation) on combustion efficiency are not considered;

[0010] (3) The lack of predictive compensation algorithms for special working conditions such as continuous climbing leads to the following technical risks:

[0011] Fuel compensation is not timely when the engine intake oxygen content decreases;

[0012] The surge in power demand caused by altitude gradients is not factored into the calculations;

[0013] There is a nonlinear sudden change in fuel consumption rate on the continuous climbing section;

[0014] The predicted range is seriously different from the actual value.

[0015] The above-mentioned technical defects are particularly prominent in complex terrain areas such as plateaus and mountains, and may cause safety hazards such as vehicle breakdown during travel. Summary of the invention

[0016] The technical problem to be solved by the embodiments of the present invention is to provide a vehicle fuel consumption prediction method, device and computer program product to integrate multi-dimensional environmental parameters for adaptive fuel consumption prediction, thereby improving the fuel estimation accuracy and warning reliability under complex working conditions.

[0017] In order to solve the above technical problems, the present invention provides a vehicle fuel consumption prediction method, comprising the following steps:

[0018] Obtain vehicle inherent attribute parameter set, environmental parameter set and navigation path information including altitude;

[0019] When the altitude of the destination is higher than the altitude of the vehicle's current location, the maximum altitude difference is calculated based on the navigation path information, and the final climbing altitude is determined in combination with the altitude of the destination;

[0020] Calculating the incremental sliding resistance of the vehicle during the climbing process according to the air density change corresponding to the final climbing height and the slope characteristics of the navigation path;

[0021] Inputting the vehicle inherent attribute parameter set and the environmental parameter set into a pre-trained neural network model to calculate the compensatory fuel consumption of the vehicle power system for altitude changes;

[0022] Calculating the additional fuel consumption of the vehicle according to the incremental sliding resistance, the average speed of the vehicle on the navigation path, the compensated fuel consumption and the engine fuel consumption rate;

[0023] The vehicle's corrected fuel consumption is calculated based on the vehicle's current fuel consumption and the vehicle's additional fuel consumption.

[0024] Preferably, the vehicle-specific attribute parameter set includes the vehicle's transmission type T v 、Displacement E m 、Engine thermal efficiency η e and the main reduction ratio i0, the environmental parameters include the oxygen content O p The step of inputting the vehicle inherent attribute parameter set and the environmental parameter set into a pre-trained neural network model to calculate the compensatory fuel consumption of the vehicle power system for altitude changes specifically includes:

[0025] Set the current vehicle's transmission type to T v 、Displacement E m 、Engine thermal efficiency η e , main reduction ratio i0 and oxygen content O p Input into the pre-trained neural network model to obtain the compensated fuel consumption Qp :

[0026] Q p =w1T v +w2E m +w3η e +w4i0+w5O p

[0027] Among them, w1, w2, w3, w4, and w5 are the weights corresponding to each input feature value.

[0028] Preferably, the training process of the neural network model is as follows:

[0029] Collect the transmission types, engine displacement, engine thermal efficiency, final reduction ratio, oxygen content in high and low altitude areas, and actual measured values ​​of compensated fuel consumption under corresponding working conditions of different vehicles, and form a multi-dimensional training data set after cleaning, normalization, and annotation;

[0030] Use the training set data to train the neural network, use the mean square error as the loss function, iteratively update the network weight matrix and bias vector through the back propagation algorithm, and dynamically adjust the learning rate during the training process. Use the validation set data for early stopping monitoring;

[0031] The network parameter combination with the smallest validation set loss is saved as the parameters of the neural network model.

[0032] Preferably, the calculating the vehicle's additional fuel consumption according to the incremental sliding resistance, the average speed of the vehicle on the navigation path, the compensated fuel consumption and the engine fuel consumption rate specifically includes:

[0033] Calculating the power of the vehicle to overcome the incremental sliding resistance according to the incremental sliding resistance and the average vehicle speed;

[0034] The power is integrated and calculated to obtain the energy consumption E required to climb from the first waypoint to the second waypoint via the navigation path. i , wherein the first waypoint and the second waypoint are waypoints with the largest altitude difference determined by pairwise altitude difference calculation in the navigation path;

[0035] According to the energy consumption E i The additional fuel consumption Q' is calculated by the following formula: i :

[0036]

[0037] Among them, Q p To compensate for fuel consumption, S i It is the mileage from the first waypoint to the second waypoint via the navigation path.

[0038] Preferably, the vehicle corrected fuel consumption is calculated based on the vehicle's current fuel consumption and the vehicle's additional fuel consumption, specifically: the vehicle corrected fuel consumption is obtained by summing the vehicle's current fuel consumption and the vehicle's additional fuel consumption.

[0039] Preferably, the calculation of the incremental sliding resistance during the climbing process of the vehicle is specifically:

[0040] Calculating the sliding resistance of the vehicle at the first waypoint and the second waypoint respectively according to the rolling resistance, air resistance and slope resistance of the vehicle at the first waypoint and the second waypoint;

[0041] The sliding resistance of the vehicle at the first waypoint is subtracted from the sliding resistance of the vehicle at the second waypoint to obtain a net increase in the sliding resistance of the vehicle from the first waypoint to the second waypoint, which is used as the incremental sliding resistance.

[0042] Preferably, after calculating the vehicle corrected fuel consumption, the method further includes:

[0043] Recalculating the vehicle's mileage based on the vehicle's corrected fuel consumption and the current remaining fuel volume;

[0044] The driver is informed of the vehicle's possible cruising range.

[0045] The present invention also provides a vehicle fuel consumption prediction device, comprising:

[0046] An acquisition module, used to acquire a vehicle inherent attribute parameter set, an environmental parameter set, and navigation path information including altitude;

[0047] A climbing height determination module, used to calculate the maximum altitude difference based on the navigation path information when the altitude of the destination is higher than the altitude of the current position of the vehicle, and determine the final climbing height in combination with the altitude of the destination;

[0048] An incremental sliding resistance calculation module, used to calculate the incremental sliding resistance of the vehicle during the climbing process according to the air density change corresponding to the final climbing height and the slope characteristics of the navigation path;

[0049] A fuel consumption compensation calculation module, used for inputting the vehicle inherent attribute parameter set and the environmental parameter set into a pre-trained neural network model to calculate the fuel consumption compensation of the vehicle power system for altitude changes;

[0050] an additional fuel consumption calculation module, used for calculating the additional fuel consumption of the vehicle according to the incremental sliding resistance, the average speed of the vehicle on the navigation path, the compensated fuel consumption and the engine fuel consumption rate;

[0051] The corrected fuel consumption calculation module is used to calculate the corrected fuel consumption of the vehicle based on the current fuel consumption of the vehicle and the additional fuel consumption of the vehicle.

[0052] The present invention also provides a vehicle fuel consumption prediction device, comprising:

[0053] one or more processors;

[0054] Memory;

[0055] One or more applications, wherein the one or more applications are stored in the memory and are configured to be executed by the one or more processors, and the one or more applications are configured to execute the vehicle fuel consumption prediction method.

[0056] The present invention also provides a computer program product, comprising computer instructions, wherein the computer instructions instruct a computer device to execute operations corresponding to the method.

[0057] The implementation of the present invention has the following beneficial effects: by introducing altitude climbing requirements and multi-dimensional environmental parameters, the present invention significantly improves the accuracy of fuel consumption prediction under complex terrain, overcoming the limitation of traditional methods that only rely on absolute mileage; by pre-training the neural network model, the intelligent coupling of vehicle dynamic characteristics and environmental parameters is realized, accurately reflecting the nonlinear effect of altitude changes on engine combustion efficiency; using path elevation analysis technology, the slope characteristics and aerodynamic parameters of key climbing sections are dynamically extracted to provide accurate input for fuel consumption compensation. The present invention effectively solves the problem of sudden changes in fuel consumption under complex terrains such as plateaus and mountains, significantly reduces the risk of vehicle breakdown midway, and provides drivers with reliable mileage prediction and fuel supply decision support, which has important safety value and economic significance. BRIEF DESCRIPTION OF THE DRAWINGS

[0058] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0059] Figure 1 It is a flow chart of a vehicle fuel consumption prediction method according to an embodiment of the present invention. DETAILED DESCRIPTION

[0060] The following descriptions of the embodiments refer to the accompanying drawings to illustrate specific embodiments in which the present invention may be implemented.

[0061] Please refer to Figure 1 As shown, the first embodiment of the present invention provides a vehicle fuel consumption prediction method, comprising the following steps:

[0062] Obtain vehicle inherent attribute parameter set, environmental parameter set and navigation path information including altitude;

[0063] When the altitude of the destination is higher than the altitude of the current location of the vehicle, the maximum altitude difference is calculated based on the navigation path information, and the final climbing altitude is determined in combination with the altitude of the destination;

[0064] Calculating the incremental sliding resistance of the vehicle during the climbing process according to the air density change corresponding to the final climbing height and the slope characteristics of the navigation path;

[0065] Inputting the vehicle inherent attribute parameter set and the environmental parameter set into a pre-trained neural network model to calculate the compensatory fuel consumption of the vehicle power system for altitude changes;

[0066] Calculating the additional fuel consumption of the vehicle according to the incremental sliding resistance, the average speed of the vehicle on the navigation path, the compensated fuel consumption and the engine fuel consumption rate;

[0067] The vehicle's corrected fuel consumption is calculated based on the vehicle's current fuel consumption and the vehicle's additional fuel consumption.

[0068] Through the above steps, it can be seen that the embodiment of the present invention integrates the inherent properties of the vehicle, environmental parameters and path elevation characteristics to achieve accurate fuel consumption prediction under complex terrain. The embodiment of the present invention starts the compensation calculation when there is a climbing demand, effectively reducing the energy consumption of the system; by establishing a dual correction mechanism combining the sliding resistance model with the pre-trained neural network, it accurately reflects the nonlinear effects of air density changes and power system response characteristics on fuel consumption; and uses a dynamic integral algorithm to calculate additional fuel consumption to ensure that the prediction results are accurately matched with the real-time working conditions. The present invention significantly improves the accuracy of mileage prediction in complex terrains such as plateaus and mountains, effectively prevents vehicles from breaking down midway, and provides drivers with reliable fuel supply decision support.

[0069] Specifically, in the embodiment of the present invention, the acquired vehicle inherent attribute parameters include the vehicle model maximum load M, vehicle transmission type T v 、Displacement E m , engine fuel consumption rate G, engine thermal efficiency η e , main reduction ratio i0, etc.; environmental parameters include atmospheric pressure, atmospheric temperature, wind speed, wind direction, oxygen content O p wait.

[0070] According to the vehicle's current fuel level information, it is converted into the remaining fuel volume L (L / 100km) according to the vehicle model characteristics, so as to obtain the vehicle's current cruising range B and current fuel consumption Q o .

[0071] The navigation path information including the altitude includes the destination information and each navigation path generated according to the destination information and the current position of the vehicle. Assume that the driver finally chooses the navigation path P0, and the total mileage is S kilometers. According to the positioning information returned by the satellite, the altitude of the destination is D D , the current altitude is D0, and the required altitude increase is D=D D -D0.

[0072] Determine whether D is greater than 0. If so, continue to execute subsequent steps. If not, the process of the present invention terminates.

[0073] Next, select 10 waypoints X in the navigation path P0. Each waypoint X i The altitude is D i , compare the altitudes of any two waypoints and take the maximum difference, then the altitude climb requirement is corrected to D′=max 1≤i,j≤10 {|D i -D j |,D}, D′ is fed back to the vehicle computer, and the final climbing height is replaced by H i , calculate the waypoint X with the largest altitude difference i , X j The mileage between the navigation path P0 is S i , the estimated travel time is T i The average speed is

[0074] It is understandable that when the altitude of the destination is higher than the altitude of the current location of the vehicle, the vehicle needs to climb to an altitude. The embodiment of the present invention not only calculates the sliding resistance of the vehicle during the climbing process, but also calculates the additional sliding resistance that needs to be overcome during the climbing, that is, the incremental sliding resistance. The following is a detailed description.

[0075] It is known that vehicle sliding resistance can be divided into vehicle rolling resistance, vehicle air resistance, and slope resistance:

[0076] F=F 滚动 +F 空气 +F 坡度

[0077] According to relevant automobile theories, rolling resistance is mainly related to the tire rolling resistance coefficient, air temperature, vehicle speed, wind speed, etc., and air resistance is related to air density, windward area, vehicle speed, and wind speed. Among them, air density and temperature are affected by changes in altitude.

[0078] First, calculate the effect of air temperature on air resistance. It is known that atmospheric pressure decreases with increasing altitude. As the surrounding air pressure gradually decreases, rising air currents are mostly in the form of large air masses. According to the principles of thermodynamics, the heat loss of an air mass is mainly manifested as the loss of internal energy of the air mass, which is specifically manifested as the work done between the entire air mass and the outside world. As the air pressure gradually decreases with increasing altitude, the air mass is always doing work on the outside world. Near the ground, the temperature decreases by 6 to 7 degrees Celsius for every kilometer increase in altitude. The temperature decrease value is usually set to 6.5 degrees Celsius.

[0079] Air density is affected by atmospheric temperature and atmospheric pressure. According to atmospheric science principles and test experience, atmospheric pressure is related to altitude, atmospheric temperature and weather conditions. According to atmospheric science principles, within the actual temperature change range of the atmosphere, temperature changes have little effect on atmospheric pressure. Atmospheric pressure measured at the same altitude increases slightly with increasing temperature, but the change is very small, and the effect of temperature on pressure can be ignored. Here we mainly consider the effect of altitude on atmospheric pressure.

[0080] The relevant standards in the ICAO standard "1964, ICAO Standard Atmosphere" stipulate that within the range of ground to 11,000 meters above sea level, the ground temperature is 15°C, the atmospheric pressure is 1013.25hPa, and the relationship between atmospheric pressure and altitude satisfies the following formula, P H is the atmospheric pressure at altitude H (hPa), H is the altitude (m):

[0081]

[0082] According to the gas state equation, the relationship between air density and atmospheric pressure and temperature is as follows: ρ (g / L) is the air density, M is the molar mass of air (28.98 g / mol), R is the gas constant (8.314 J / (mol·K)), and T is the air temperature (℃):

[0083]

[0084] The vehicle rolling resistance can be directly calculated using the internationally accepted algorithm:

[0085]

[0086] Among them, M is the maximum load of the vehicle, g is the acceleration of gravity, μ is the rolling resistance coefficient, which is usually between 0.015 and 0.02, and V is the current vehicle speed.

[0087] The vehicle slope resistance can also be directly calculated using the internationally accepted algorithm:

[0088] F 坡度 =M×g×sinθ

[0089] Among them, θ is the slope angle, which is considered to be the waypoint X with the largest altitude difference among the waypoints. i , X j Altitude and distance S i The pitch angle formed is the highest climbing height above sea level H i , D i , D j The pitch angle formed between them. Usually θ∈[0,π / 2).

[0090] Taking the influence of the slope into account, the positive pressure of the vehicle on the road surface on the slope changes, and the rolling resistance of the vehicle is corrected to:

[0091]

[0092] The vehicle air resistance is calculated as follows:

[0093]

[0094] Among them C D is the air resistance coefficient, and A is the windward area.

[0095] The sliding resistance F is finally calculated as follows:

[0096]

[0097] Among them, v x 、v y are the wind speed components parallel and perpendicular to the direction of the car's travel, respectively. x When it is in the same direction as V, S=-1, otherwise it is 1.

[0098] According to the above sliding resistance calculation formula during the climbing process, the vehicle can calculate the sliding resistance when climbing to X i The sliding resistance F i for:

[0099]

[0100] Assuming the vehicle is moving at a constant speed, the lowest point X j For a flat road, X j The sliding resistance F j for:

[0101]

[0102] The vehicle climbs due to altitude (from the lowest point X j Climb to the highest point X i , climbing height H i ) brings about a net increase in sliding resistance, i.e., the incremental sliding resistance δ F for:

[0103]

[0104] It should be noted that when there is an altitude difference, the additional traction requirement caused by the vehicle's climbing demand is different from the numerical estimate on flat ground. During the climbing process, the change in sliding resistance F relative to flat ground conditions is affected by the altitude, which can be equivalent to the additional traction requirement for climbing.

[0105] The embodiment of the present invention obtains the vehicle transmission type T according to the pre-trained neural network model. v 、Displacement E m 、Engine thermal efficiency η e , main reduction ratio i0 and oxygen content O p The fuel consumption relationship, combined with the sliding resistance δ F , engine fuel consumption rate G, and get additional fuel consumption Q' i .

[0106] Specifically, in the design and construction of the neural network model of the embodiment of the present invention, the overall network structure uses a multi-layer perceptron model (MLP), the input layer is the transmission type (which can be converted into a numerical value through encoding), displacement, engine thermal efficiency, main reduction ratio and air oxygen content and other variables, the output layer is the corresponding compensation fuel consumption, the activation function uses Leaky ReLU to improve the calculation efficiency and avoid gradient disappearance, and the optimization algorithm uses Adam to adapt the learning rate. It should be noted that the input vehicle inherent attribute parameters and environmental parameters are not limited to the above five variables.

[0107] Firstly, a large amount of data on different vehicle transmission types (including manual, automatic, continuously variable transmission, etc.), displacement (engine data of different displacements), engine thermal efficiency (thermal efficiency data of engines under different working conditions), main reduction ratio (main reduction ratio data of different vehicles), oxygen content (air oxygen content data in high altitude and low altitude areas) and their corresponding compensated fuel consumption data are collected, and a training set is formed after cleaning, normalization and labeling; secondly, the training set data is used to train the neural network, with the mean square error as the loss function, and the network weight matrix and bias vector are iteratively updated through the back propagation algorithm, and the learning rate is dynamically adjusted during the training process, and the validation set data is used for early stopping monitoring to timely adjust the model parameters; finally, the network parameter combination with the smallest validation set loss is saved as the parameters of the neural network model.

[0108] After training is completed, the transmission type T of the current vehicle is changed v 、Displacement E m 、Engine thermal efficiency η e , main reduction ratio i0 and oxygen content O pInput into the trained neural network model to obtain the corresponding compensated fuel consumption Q p :

[0109] Q p =w1T v +w2E m +w3η e +w4i0+w5O p

[0110] It should be noted that in the field of vehicle fuel consumption estimation, the prior art has long relied on empirical formulas or simple calculation models based on physical principles. Although these methods can achieve basic fuel consumption estimation, it is difficult to make accurate comprehensive considerations when faced with many interrelated and nonlinear factors that affect fuel consumption, such as vehicle transmission type, displacement, engine thermal efficiency, main reduction ratio, oxygen content, etc. They focus too much on formula derivation and simple parameter adjustment, are limited by traditional thinking patterns, and lack attempts to solve fuel consumption calculation problems from a data-driven perspective. Neural network models have powerful nonlinear mapping capabilities, can automatically learn the inherent relationships between complex data, and model and predict highly complex systems. However, building an effective neural network model requires a large amount of data collection, organization, annotation, and complex model training and tuning processes, which cannot be achieved by simple technical attempts. Before the present invention, there was no prior art that input the above-mentioned complex vehicle parameters and environmental factors into the neural network model for additional fuel consumption calculation, and those skilled in the art were not aware of its possibility and advantages. Therefore, the present invention inputs these parameters into the pre-trained neural network model to calculate the additional fuel consumption, which can accurately reflect the nonlinear effects of air density changes and power system response characteristics on fuel consumption.

[0111] From the above, we can know that the navigation path P0 reaches the waypoint X i With X j The mileage is S i (Km), the estimated driving time is T i (h), the average vehicle speed is Then overcome the above incremental sliding resistance δ F The power P is:

[0112]

[0113] in, is the component of the average vehicle speed parallel to the vehicle's forward direction, and in the embodiment of the present invention, the coefficient of this component is set to 1;

[0114] According to the incremental sliding resistance δ F With average speed The integral calculation can be obtained from the navigation path P0 from the passing point X j Climb to waypoint X iThe energy consumption required E i (Kwh) is:

[0115]

[0116] The energy consumption E calculated above i , and the engine fuel consumption rate G in the vehicle's inherent parameter set, the required additional fuel consumption Q' is calculated as follows i :

[0117]

[0118] Corrected fuel consumption Q i for:

[0119] Q i =Q0+Q′ i

[0120] The car computer corrects the fuel consumption Q i , and regain the vehicle's range B′; and remind the owner through pop-up windows, voice broadcasts, etc.

[0121] Corresponding to the vehicle fuel consumption prediction method described in the first embodiment of the present invention, the second embodiment of the present invention further provides a vehicle fuel consumption prediction device, including:

[0122] An acquisition module, used to acquire a vehicle inherent attribute parameter set, an environmental parameter set, and navigation path information including altitude;

[0123] A climbing height determination module, used to calculate the maximum altitude difference based on the navigation path information when the altitude of the destination is higher than the altitude of the current position of the vehicle, and determine the final climbing height in combination with the altitude of the destination;

[0124] An incremental sliding resistance calculation module, used to calculate the incremental sliding resistance of the vehicle during the climbing process according to the air density change corresponding to the final climbing height and the slope characteristics of the navigation path;

[0125] A fuel consumption compensation calculation module, used for inputting the vehicle inherent attribute parameter set and the environmental parameter set into a pre-trained neural network model to calculate the fuel consumption compensation of the vehicle power system for altitude changes;

[0126] an additional fuel consumption calculation module, used for calculating the additional fuel consumption of the vehicle according to the incremental sliding resistance, the average speed of the vehicle on the navigation path, the compensated fuel consumption and the engine fuel consumption rate;

[0127] The corrected fuel consumption calculation module is used to calculate the corrected fuel consumption of the vehicle based on the current fuel consumption of the vehicle and the additional fuel consumption of the vehicle.

[0128] Corresponding to the vehicle fuel consumption prediction method described in the first embodiment of the present invention, the third embodiment of the present invention further provides a vehicle fuel consumption prediction device, including:

[0129] one or more processors;

[0130] Memory;

[0131] One or more applications, wherein the one or more applications are stored in the memory and configured to be executed by the one or more processors, and the one or more applications are configured to execute the vehicle fuel consumption prediction method described in the aforementioned embodiment 1 of the present invention.

[0132] Corresponding to the vehicle fuel consumption prediction method described in the aforementioned embodiment 1 of the present invention, embodiment 4 of the present invention also provides a computer program product, including computer instructions, and the computer instructions instruct a computer device to perform operations corresponding to the vehicle fuel consumption prediction method described in the aforementioned embodiment 1 of the present invention.

[0133] Preferably, the processor may be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor may be a microprocessor, or the processor may be any conventional processor. The processor is the control center of the device, and various parts of the device are connected using various interfaces and lines.

[0134] The memory mainly includes a program storage area and a data storage area, wherein the program storage area can store an operating system, an application required for at least one function, etc., and the data storage area can store related data, etc. In addition, the memory can be a high-speed random access memory, or a non-volatile memory, such as a plug-in hard disk, a smart memory card (SmartMedia Card, SMC), a secure digital (Secure Digital, SD) card, and a flash card (Flash Card), etc., or the memory can also be other volatile solid-state storage devices.

[0135] It should be noted that the above-mentioned device may include but is not limited to a processor and a memory, which can be understood by those skilled in the art.

[0136] For the working principle and process of the above embodiment, please refer to the description of the above embodiment of the present invention, which will not be repeated here.

[0137] It can be seen from the above description that compared with the prior art, the beneficial effects of the present invention are as follows: the present invention significantly improves the fuel consumption prediction accuracy under complex terrain by introducing altitude climbing requirements and multi-dimensional environmental parameters, overcoming the limitation of the traditional method that only relies on absolute mileage; through the pre-trained neural network model, the intelligent coupling of vehicle dynamic characteristics and environmental parameters is realized, accurately reflecting the nonlinear effect of altitude changes on engine combustion efficiency; using path elevation analysis technology, the slope characteristics and aerodynamic parameters of key climbing sections are dynamically extracted to provide accurate input for fuel consumption compensation. The present invention effectively solves the problem of sudden changes in fuel consumption under complex terrains such as plateaus and mountains, significantly reduces the risk of vehicle breakdown midway, and provides drivers with reliable mileage prediction and fuel supply decision support, which has important safety value and economic significance.

[0138] The above disclosure is only the preferred embodiment of the present invention, which certainly cannot be used to limit the scope of the present invention. Therefore, equivalent changes made according to the claims of the present invention are still within the scope of the present invention.

Claims

1. A vehicle fuel consumption prediction method, characterized in that: The following steps are involved: Obtain vehicle inherent attribute parameter set, environmental parameter set and navigation path information including altitude; When the altitude of the destination is higher than the altitude of the current location of the vehicle, the maximum altitude difference is calculated based on the navigation path information, and the final climbing altitude is determined in combination with the altitude of the destination; Calculating the incremental sliding resistance of the vehicle during the climbing process according to the air density change corresponding to the final climbing height and the slope characteristics of the navigation path; Inputting the vehicle inherent attribute parameter set and the environmental parameter set into a pre-trained neural network model to calculate the compensatory fuel consumption of the vehicle power system for altitude changes; Calculating the additional fuel consumption of the vehicle according to the incremental sliding resistance, the average speed of the vehicle on the navigation path, the compensated fuel consumption and the engine fuel consumption rate; The vehicle's corrected fuel consumption is calculated based on the vehicle's current fuel consumption and the vehicle's additional fuel consumption.

2. The method according to claim 1, characterized in that The vehicle inherent attribute parameter set includes the vehicle's transmission type T v 、Displacement E m 、Engine thermal efficiency η e and the main reduction ratio i0, the environmental parameters include the oxygen content O p The step of inputting the vehicle inherent attribute parameter set and the environmental parameter set into a pre-trained neural network model to calculate the compensatory fuel consumption of the vehicle power system for altitude changes specifically includes: Set the current vehicle's transmission type to T p 、Displacement E m 、Engine thermal efficiency η e , main reduction ratio i0 and oxygen content O p Input into the pre-trained neural network model to obtain the compensated fuel consumption Q p : Q p w1T v +w2E m +w3η e +w4i0+w5O p Among them, w1, w2, w3, w4, and w5 are the weights corresponding to each input feature value.

3. The method according to claim 2, characterized in that The training process of the neural network model is as follows: Collect the transmission types, engine displacement, engine thermal efficiency, final reduction ratio, oxygen content in high and low altitude areas, and actual measured values ​​of compensated fuel consumption under corresponding working conditions of different vehicles, and form a multi-dimensional training data set after cleaning, normalization, and annotation; Use the training set data to train the neural network, use the mean square error as the loss function, iteratively update the network weight matrix and bias vector through the back propagation algorithm, and dynamically adjust the learning rate during the training process. Use the validation set data for early stopping monitoring; The network parameter combination with the smallest validation set loss is saved as the parameters of the neural network model.

4. The method according to claim 2, characterized in that: The calculating the vehicle's additional fuel consumption according to the incremental sliding resistance, the average speed of the vehicle on the navigation path, the compensated fuel consumption and the engine fuel consumption rate specifically includes: Calculating the power of the vehicle to overcome the incremental sliding resistance according to the incremental sliding resistance and the average vehicle speed; The power is integrated and calculated to obtain the energy consumption E required to climb from the first waypoint to the second waypoint via the navigation path. i , wherein the first waypoint and the second waypoint are waypoints with the largest altitude difference determined by pairwise altitude difference calculation in the navigation path; According to the energy consumption E i The additional fuel consumption Q is calculated by the following formula: i ′ : Among them, Q p To compensate for fuel consumption, S i It is the mileage from the first waypoint to the second waypoint via the navigation path.

5. The method according to claim 1, characterized in that The vehicle corrected fuel consumption is calculated based on the vehicle's current fuel consumption and the vehicle's additional fuel consumption. Specifically, the vehicle's current fuel consumption and the vehicle's additional fuel consumption are summed to obtain the vehicle corrected fuel consumption.

6. The method according to claim 4, characterized in that The calculation of the incremental sliding resistance during the vehicle climbing process is specifically: Calculating the sliding resistance of the vehicle at the first waypoint and the second waypoint respectively according to the rolling resistance, air resistance and slope resistance of the vehicle at the first waypoint and the second waypoint; The sliding resistance of the vehicle at the first waypoint is subtracted from the sliding resistance of the vehicle at the second waypoint to obtain a net increase in the sliding resistance of the vehicle from the first waypoint to the second waypoint, which is used as the incremental sliding resistance.

7. The method according to claim 1, characterized in that The vehicle's corrected fuel consumption is also calculated to include: Recalculating the vehicle's mileage based on the vehicle's corrected fuel consumption and the current remaining fuel volume; The driver is informed of the vehicle's possible cruising range.

8. A vehicle fuel consumption prediction device, characterized in that: include: An acquisition module, used to acquire a vehicle inherent attribute parameter set, an environmental parameter set, and navigation path information including altitude; A climbing height determination module, used to calculate the maximum altitude difference based on the navigation path information when the altitude of the destination is higher than the altitude of the current position of the vehicle, and determine the final climbing height in combination with the altitude of the destination; An incremental sliding resistance calculation module, used to calculate the incremental sliding resistance of the vehicle during the climbing process according to the air density change corresponding to the final climbing height and the slope characteristics of the navigation path; A fuel consumption compensation calculation module, used for inputting the vehicle inherent attribute parameter set and the environmental parameter set into a pre-trained neural network model to calculate the fuel consumption compensation of the vehicle power system for altitude changes; an additional fuel consumption calculation module, used for calculating the additional fuel consumption of the vehicle according to the incremental sliding resistance, the average speed of the vehicle on the navigation path, the compensated fuel consumption and the engine fuel consumption rate; The corrected fuel consumption calculation module is used to calculate the corrected fuel consumption of the vehicle based on the current fuel consumption of the vehicle and the additional fuel consumption of the vehicle.

9. A vehicle fuel consumption prediction device, characterized in that: include: one or more processors; Memory; One or more applications, wherein the one or more applications are stored in the memory and are configured to be executed by the one or more processors, and the one or more applications are configured to execute the vehicle fuel consumption prediction method as described in any one of claims 1 to 7.

10. A computer program product, characterized in that The method comprises computer instructions, wherein the computer instructions instruct a computer device to execute operations corresponding to the method according to any one of claims 1 to 7.

Images