Electric vehicle energy consumption prediction method and system using gps data

Abstract

The application is suitable for the technical field of electric vehicle energy consumption management, and provides an electric vehicle energy consumption prediction method and system using GPS data, which comprises the following steps: acquiring a GPS original data stream in real time, preprocessing and fusing a digital elevation model to generate an altitude sequence; calculating an instantaneous slope value and a slope change rate based on the altitude sequence to quantify the degree of terrain undulation; calculating additional acceleration and additional energy consumption loss in combination with the slope change rate; predicting recoverable electric energy based on the altitude sequence, speed data and real-time state of a kinetic energy recovery system; calculating predicted net energy consumption in combination with reference energy consumption, additional energy consumption and recoverable electric energy, and dynamically calibrating a model through continuous error triggering. The application improves the energy consumption prediction accuracy and provides reliable support for electric vehicle energy management.

Description

Technical Field

[0001] This invention belongs to the field of electric vehicle energy consumption management technology, and particularly relates to a method and system for predicting the energy consumption of electric vehicles using GPS data. Background Technology

[0002] As the penetration rate of electric vehicles gradually increases, users' demand for accurate perception of driving range is becoming increasingly urgent. Energy consumption prediction has become one of the core technologies for optimizing vehicle energy management and alleviating range anxiety. Currently, GPS technology, which can provide real-time multi-dimensional data such as vehicle location, speed, and altitude, has been widely used in the field of electric vehicle route planning and status monitoring. Energy consumption prediction technology based on GPS data has also become a research hotspot in the industry. By capturing environmental and vehicle motion characteristics during driving using GPS data, dynamic estimation of energy consumption can be achieved, providing support for user travel planning and vehicle energy scheduling.

[0003] Existing energy consumption prediction schemes for electric vehicles that utilize GPS data cannot effectively eliminate random fluctuations in GPS altitude data, resulting in insufficient continuity and accuracy of altitude sequences. Furthermore, it is difficult to quantify the dynamic impact of terrain undulations on energy consumption. Relying solely on static slope values ​​to calculate additional energy consumption deviates from the actual effect of terrain changes on energy consumption during actual driving, and cannot stably output net energy consumption results that match actual routes and driving conditions. Summary of the Invention

[0004] The purpose of this invention is to provide a method and system for predicting the energy consumption of electric vehicles using GPS data, in order to solve the technical problems existing in the prior art as identified in the background art.

[0005] This invention is implemented as follows: a method for predicting the energy consumption of electric vehicles using GPS data, the method comprising:

[0006] Real-time acquisition of raw vehicle GPS data stream, preprocessing, and extraction of altitude sequence;

[0007] Based on the altitude sequence, the instantaneous slope value of the altitude at adjacent time points is calculated, and the slope change rate reflecting the degree of terrain undulation is calculated.

[0008] Based on the slope change rate, the additional acceleration requirement to overcome the slope change is calculated in real time, and then the additional energy consumption is calculated.

[0009] Based on altitude sequence and speed data, the current kinetic energy recovery mode of the vehicle is identified, and the vehicle kinetic energy recovery efficiency model is invoked to predict the electrical energy that can be effectively recovered on the next downhill section.

[0010] Based on the vehicle's baseline energy consumption, combined with additional energy loss and recoverable electrical energy, the predicted net energy consumption of the vehicle under the current route and driving conditions is analyzed.

[0011] As a further aspect of the present invention, the real-time acquisition of the vehicle GPS raw data stream specifically includes:

[0012] Continuously acquire raw GPS data streams containing timestamps, latitude and longitude, speed and altitude from the vehicle's GPS receiver;

[0013] Smoothing is performed on the altitude data in the raw GPS data stream to eliminate random fluctuations caused by signal multipath effects and atmospheric delay.

[0014] The smoothed altitude data is fused and corrected with the digital elevation model to generate a continuous and smooth altitude sequence.

[0015] As a further aspect of the present invention, the calculation of the slope change rate specifically includes:

[0016] The altitude sequence is sampled at equal time intervals, the difference in altitude between adjacent sampling points is calculated, and the instantaneous slope value of each sampling point is obtained by combining the time intervals.

[0017] Construct a sliding time window for the instantaneous slope value sequence, fit the slope change trend within the sliding time window, and obtain the instantaneous slope change rate sequence;

[0018] The root mean square value of the slope change rate sequence is statistically analyzed using a sliding time window, and is used as an indicator of the slope change rate to quantify the severity of terrain undulation.

[0019] As a further aspect of the present invention, the calculation of additional energy consumption specifically includes:

[0020] Establish a longitudinal dynamics equation for the vehicle that includes the slope change rate as input, and calculate the additional acceleration required to maintain the vehicle's constant speed based on the real-time acquired slope change rate.

[0021] Calculate the additional force required to overcome terrain changes, taking into account the vehicle's total mass and additional acceleration requirements.

[0022] The additional force is multiplied by the instantaneous velocity obtained through the GPS data stream to calculate the additional power loss caused by terrain fluctuations in real time. The additional power loss is then numerically integrated along the time axis to obtain the additional energy loss for the entire predicted journey.

[0023] As a further aspect of the present invention, the prediction of the electrical energy that can be effectively recovered on the next downhill section specifically includes:

[0024] Continuously collect the status parameters of the vehicle's kinetic energy recovery system, including the recovery intensity level setting, battery charging acceptance capacity, and motor operating mode, to identify the vehicle's current actual kinetic energy recovery mode;

[0025] Road profile analysis based on elevation sequence was performed to identify downhill sections and extract parameters of downhill sections, including slope angle, length and elevation difference.

[0026] Calculate the corresponding energy recovery efficiency coefficient based on the identified actual kinetic energy recovery mode and downhill road parameters;

[0027] Based on the downhill road parameters and the vehicle's raw GPS data stream, the theoretical recoverable energy is calculated, and then multiplied by the energy recovery efficiency coefficient to obtain the actual recoverable electrical energy.

[0028] As a further aspect of the present invention, the analysis of the predicted net energy consumption of the vehicle under the current path and driving conditions specifically includes:

[0029] Identify the vehicle's baseline energy consumption, align the additional energy loss obtained from the vehicle dynamics model with the baseline energy consumption over time, and calculate the total positive energy consumption.

[0030] Calculate the complete predicted net energy consumption of the vehicle based on the total positive energy consumption and the predicted recoverable electrical energy.

[0031] The current journey is divided into several independent evaluation units according to a preset distance interval. When the average error between the predicted net energy consumption and the actual energy consumption in three consecutive evaluation units exceeds 5%, the additional force calculation coefficient in the vehicle dynamics model is dynamically updated based on the average error.

[0032] Another object of the present invention is to provide an electric vehicle energy consumption prediction system utilizing GPS data, the system comprising:

[0033] The GPS data stream acquisition module is used to acquire the vehicle's raw GPS data stream in real time, perform preprocessing, and extract the altitude sequence.

[0034] The slope calculation module is used to calculate the instantaneous slope value of adjacent time points based on the altitude sequence, and to calculate the slope change rate that reflects the severity of terrain undulation.

[0035] The additional energy consumption calculation module is used to calculate in real time the additional acceleration required to overcome the slope change based on the slope change rate, and then calculate the additional energy consumption loss.

[0036] The electric energy prediction module is used to identify the vehicle's current kinetic energy recovery mode based on altitude sequence and speed data, and call the vehicle's kinetic energy recovery efficiency model to predict the electric energy that can be effectively recovered on the next downhill section.

[0037] The net energy consumption analysis module is used to analyze the vehicle's predicted net energy consumption under the current route and driving conditions, based on the vehicle's baseline energy consumption, combined with additional energy loss and recoverable electrical energy.

[0038] As a further embodiment of the present invention, the slope calculation module includes:

[0039] The interval sampling module is used to sample the altitude sequence at equal time intervals, calculate the difference in altitude between adjacent sampling points, and combine the time intervals to obtain the instantaneous slope value of each sampling point.

[0040] The sliding window construction module is used to construct a sliding time window for the instantaneous slope value sequence, fit the slope change trend within the sliding time window, and obtain the instantaneous slope change rate sequence.

[0041] The rate of change index calculation module is used to calculate the root mean square value of the slope change rate sequence using a sliding time window, which serves as a slope change rate index to quantify the severity of terrain undulation.

[0042] As a further embodiment of the present invention, the additional energy consumption calculation module includes:

[0043] The equation building module is used to build the vehicle longitudinal dynamics equations that include the slope change rate input, and calculate the additional acceleration required to maintain the vehicle's constant speed based on the real-time acquired slope change rate.

[0044] The additional force calculation module is used to calculate the additional force required to overcome terrain changes by combining the total mass of the vehicle and the additional acceleration requirements.

[0045] The energy loss calculation module multiplies the additional force with the instantaneous velocity obtained through the GPS data stream to calculate the additional power loss caused by terrain fluctuations in real time. It then performs numerical integration of the additional power loss along the time axis to obtain the additional energy loss for the entire predicted journey.

[0046] As a further embodiment of the present invention, the electrical energy prediction module includes:

[0047] The kinetic energy recovery parameter acquisition module is used to continuously collect the status parameters of the vehicle's kinetic energy recovery system, including the recovery intensity level setting, battery charging acceptance capacity and motor operating mode, and to identify the vehicle's current actual kinetic energy recovery mode.

[0048] The downhill section identification module is used to perform road profile analysis based on the altitude sequence, identify downhill sections, and extract downhill section parameters, including slope angle, length, and elevation difference parameters.

[0049] The energy recovery efficiency calculation module is used to calculate the corresponding energy recovery efficiency coefficient based on the identified actual kinetic energy recovery mode and downhill road parameters.

[0050] The actual recovery analysis module is used to calculate the theoretical recoverable energy based on downhill road parameters and vehicle GPS raw data stream, and multiply it by the energy recovery efficiency coefficient to obtain the actual recoverable electrical energy.

[0051] The beneficial effects of this invention are:

[0052] This invention generates a continuous and accurate altitude sequence by smoothing the raw GPS altitude data and fusing it with a digital elevation model for correction, thus laying a reliable data foundation for subsequent terrain analysis and energy consumption calculation.

[0053] By calculating instantaneous slope values, constructing a sliding window to fit the slope change trend, and statistically analyzing the root mean square value, the intensity of terrain undulation is quantified. Combined with the vehicle longitudinal dynamics equation, additional acceleration and extra energy consumption are dynamically calculated, so that the impact of terrain on energy consumption can be accurately reflected.

[0054] By combining the real-time status parameters of the vehicle's kinetic energy recovery system with the specific characteristics of downhill sections, the energy recovery efficiency coefficient is dynamically adjusted to improve the accuracy of recoverable energy prediction. By aligning time series, the baseline energy consumption and additional energy consumption are accurately superimposed. Combined with recoverable energy, a complete net energy consumption calculation logic is formed. Through a segmented evaluation and dynamic calibration mechanism triggered by continuous errors, the vehicle dynamics model parameters are continuously optimized to ensure stable long-term prediction accuracy.

[0055] The overall solution achieves deep integration of GPS data with all aspects of energy consumption prediction, significantly improving the accuracy and reliability of electric vehicle energy consumption prediction. It can provide users with more realistic decision-making basis for range planning and vehicle energy management, effectively alleviate range anxiety, and provide technical support for the formulation of electric vehicle energy optimization strategies. Attached Figure Description

[0056] Figure 1 A flowchart of an electric vehicle energy consumption prediction method using GPS data provided in an embodiment of the present invention;

[0057] Figure 2 A flowchart for real-time acquisition of raw vehicle GPS data stream provided in an embodiment of the present invention;

[0058] Figure 3 A flowchart for calculating the rate of change of slope provided in an embodiment of the present invention;

[0059] Figure 4 A flowchart for calculating additional energy consumption loss is provided for embodiments of the present invention;

[0060] Figure 5 A flowchart for predicting the electrical energy that can be effectively recovered on the next downhill section, provided as an embodiment of the present invention;

[0061] Figure 6 A flowchart for analyzing the predicted net energy consumption of a vehicle under current path and driving conditions, provided for embodiments of the present invention;

[0062] Figure 7 This is a structural block diagram of an electric vehicle energy consumption prediction system utilizing GPS data provided in an embodiment of the present invention;

[0063] Figure 8 This is a structural block diagram of the slope calculation module provided in an embodiment of the present invention;

[0064] Figure 9 This is a structural block diagram of the additional energy consumption calculation module provided in an embodiment of the present invention;

[0065] Figure 10 This is a structural block diagram of the electrical energy prediction module provided in an embodiment of the present invention. Detailed Implementation

[0066] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the invention.

[0067] Figure 1 A flowchart of the electric vehicle energy consumption prediction method using GPS data provided in an embodiment of the present invention is shown below. Figure 1 As shown, the method includes:

[0068] S100 acquires the vehicle's raw GPS data stream in real time, performs preprocessing, and extracts the altitude sequence;

[0069] Relying on the vehicle's GPS receiver, it continuously captures a full-dimensional raw data stream covering timestamps, latitude and longitude, speed and altitude. This continuous acquisition mode can completely preserve the spatiotemporal and motion status information of the vehicle during its driving process.

[0070] To address the issue of raw GPS elevation data being susceptible to external environmental interference, a smoothing process is employed to filter out random fluctuations caused by signal multipath effects and atmospheric delays. This ensures that the elevation data's variation trend closely matches the actual terrain undulations, avoiding meaningless numerical jumps that could interfere with subsequent analysis. A digital elevation model (DEM) is introduced for fusion correction. Leveraging its extensive terrain elevation data coverage, this compensates for potential accuracy deficiencies or data discontinuities in localized areas of single GPS data. The resulting elevation sequence not only possesses continuous and smooth characteristics but also accurately recreates the details of elevation changes along the vehicle's travel path.

[0071] Whether it's GPS signal instability caused by building obstruction in urban roads, or signal fluctuations in complex terrains such as mountains and hills, this data processing mode can effectively ensure the reliability of altitude data, ensuring that subsequent terrain analysis and energy consumption calculations based on this sequence can accurately reflect actual driving conditions.

[0072] S200, based on the altitude sequence, calculates the instantaneous slope value of the altitude at adjacent time points, and calculates the slope change rate that reflects the severity of terrain undulation.

[0073] Sampling the altitude sequence at equal time intervals adapts to the temporal characteristics of GPS data. Simultaneously, combining this with the vehicle's real-time speed to calculate the horizontal distance between adjacent sampling points ensures that the instantaneous slope value calculation is not only based on altitude differences but also closely reflects the vehicle's actual driving trajectory. At the same altitude difference, different vehicle speeds lead to variations in horizontal distance, thus affecting the perceived actual slope. Only by combining the instantaneous slope value obtained through time intervals and speed can the true terrain slope encountered by the vehicle during driving be reflected.

[0074] After obtaining the instantaneous slope value sequence, a sliding time window is constructed for trend fitting. This is because the instantaneous slope value may be affected by local data fluctuations, and directly calculating the rate of change will result in frequent jumps, which cannot reflect the overall change pattern of the terrain. By covering multiple consecutive instantaneous slope values ​​through a sliding window, linear fitting is used to extract the slope change trend within the window, smoothing local noise, and obtaining a continuous and stable instantaneous slope change rate sequence. This sequence can clearly reflect whether the slope is gradually increasing, decreasing, or remaining stable.

[0075] By using a sliding time window to statistically analyze the root mean square value of the slope change rate series, the dynamic fluctuation of the slope change rate is transformed into a single quantitative indicator. Since the slope change rate can be positive or negative, i.e. the slope increases or decreases, directly taking the average value will cause the positive and negative changes to cancel each other out, and cannot reflect the fluctuation range. The root mean square value can effectively integrate the amplitude information of positive and negative changes, and the resulting slope change rate indicator can intuitively distinguish the degree of undulation of different terrains.

[0076] The S300 calculates the additional acceleration required to overcome the slope change based on the slope change rate in real time, and then calculates the additional energy consumption loss.

[0077] A longitudinal dynamics equation for the vehicle, incorporating the gradient change rate as input, is established. This equation, based on the physical nature of vehicle movement, correlates the gradient change rate with the force and acceleration required for the vehicle to maintain a constant speed. By acquiring the gradient change rate in real time, the corresponding additional acceleration requirement is accurately calculated. Through dynamic response to the current gradient change trend, the calculated acceleration is ensured to be highly matched with the actual terrain's resistance requirements on the vehicle. For example, when the gradient change rate increases, the vehicle requires greater additional acceleration to counteract terrain resistance and maintain the preset constant speed.

[0078] Next, the additional force required to overcome terrain changes is calculated by combining the vehicle's total mass and additional acceleration requirements. The vehicle's total mass, as the core physical quantity of influence, determines the magnitude of the force required under the same acceleration, transforming the abstract acceleration requirement into a specific additional force, thus making the actual resistance effect of terrain on the vehicle concrete.

[0079] The additional force is multiplied by the instantaneous speed obtained through GPS data stream to obtain the additional power loss caused by terrain fluctuations. Power is the instantaneous manifestation of energy consumption. Combined with real-time speed, it can accurately reflect the energy consumption rate caused by terrain resistance at different driving speeds. The higher the speed, the greater the power loss under the same additional force, ensuring the synchronization of energy consumption rate with the actual driving state of the vehicle.

[0080] The additional power loss is numerically integrated along the time axis to obtain the additional energy loss for the entire predicted journey. The integration operation accumulates the instantaneous power consumption into the total additional energy consumption for the entire journey, covering every time period in the journey, and avoiding the overall deviation caused by calculating only the energy consumption at a single point.

[0081] The gradient change rate of continuous uphill and downhill road sections fluctuates frequently, and the additional acceleration will be adjusted in real time accordingly. The additional force and additional power loss will also change accordingly. By integrating over time, the final additional energy loss can accurately reflect the actual energy consumption of the road section due to the frequent fluctuations in terrain. On the other hand, on road sections with gentle gradient changes, the gradient change rate is small, and the additional acceleration, additional force and additional power loss are all at a low level. The integration result will also be reduced accordingly, which is highly consistent with the actual energy consumption of the road section.

[0082] The S400, based on altitude sequence and speed data, identifies the vehicle's current kinetic energy recovery mode and calls the vehicle's kinetic energy recovery efficiency model to predict the amount of electrical energy that can be effectively recovered on the next downhill section.

[0083] Continuously collecting the vehicle's kinetic energy recovery system status parameters, the recovery intensity level setting is directly related to the system's ability to capture braking energy, the battery charging acceptance capacity reflects the upper limit of the energy that the battery can absorb and recover, and the motor operating mode affects the efficiency of the motor in converting kinetic energy into electrical energy. These three factors together determine the basic ability of the vehicle to actually recover energy. Real-time collection of these parameters can avoid the deviation caused by using a fixed recovery efficiency and ensure that the judgment of the vehicle's current recovery potential is consistent with the actual operating conditions.

[0084] Road profile analysis based on altitude sequence identifies road sections with continuously decreasing altitude by continuously tracking the trend of altitude change and defining the downhill range. Among the extracted parameters of slope angle, length and elevation difference, the slope angle reflects the steepness of the downhill, the length is related to the driving time of vehicles on the slope, and the elevation difference directly determines the total amount of gravitational potential energy released. These parameters provide key terrain dimension data for subsequent energy calculations, and the continuity of GPS data ensures the completeness of downhill section identification, without missing short-distance or gentle downhill sections.

[0085] The energy recovery efficiency coefficient is calculated and dynamically generated by combining the actual kinetic energy recovery mode and downhill road parameters. For example, in steep slope environments, the vehicle braking demand is higher, and the recovery system can capture energy more fully, resulting in a higher efficiency coefficient. In gentle slope environments, the recovery efficiency is relatively lower. This dynamic calculation method makes the efficiency coefficient more consistent with the actual energy conversion law.

[0086] When calculating the theoretical recoverable energy, the elevation difference in the downhill section parameters and the velocity data in the GPS raw data stream are combined. This incorporates both the change in gravitational potential energy (determined by the elevation difference) and the increase or decrease in kinetic energy (determined by the difference between the initial and final velocities), ensuring that the theoretical value covers all energy sources that can be converted into electrical energy during the downhill process. The actual recoverable electrical energy is then obtained by multiplying it by the efficiency coefficient, so that the prediction results take into account both theoretical potential and actual conversion limitations.

[0087] In real-world scenarios, mountain roads often feature long, steep slopes with significant elevation differences and steep angles. If the energy recovery intensity is set to a high level and the battery is in a half-charged state (with strong charging acceptance), the motor will be in a high-efficiency feedback mode, resulting in a significant increase in the actual recoverable energy. On the other hand, urban roads often have short, gentle downhill slopes with small elevation differences and gentle angles. Even if the energy recovery intensity is set to medium, the total energy release is limited due to the shorter slope length, resulting in a relatively low actual recoverable energy. The calculation logic in this step can accurately adapt to these different scenarios.

[0088] S500 analyzes the vehicle's predicted net energy consumption under current route and driving conditions, based on the vehicle's baseline energy consumption and incorporating additional energy losses and recoverable energy.

[0089] Baseline energy consumption is the basic energy consumption of a vehicle under standard operating conditions. Its identification requires combining the vehicle's own parameters with speed information from GPS data. Speed ​​is a key variable affecting baseline energy consumption, and the baseline energy consumption varies under different constant speed conditions. The baseline energy consumption value under the corresponding operating condition can be matched by real-time GPS speed data. Subsequently, the additional energy loss and baseline energy consumption are aligned in time series. Because additional energy loss is generated in real time with changes in terrain, and baseline energy consumption also changes dynamically with time (and speed), only by matching the energy consumption data of the two within the same time period based on the timestamps provided by GPS can the deviation in total energy consumption calculation caused by time misalignment be avoided, thus accurately obtaining the total positive energy consumption, reflecting the total energy input required by the vehicle to overcome basic resistance and terrain resistance during driving.

[0090] The predicted net energy consumption of a vehicle can be obtained by subtracting the predicted recoverable electrical energy from the total positive energy consumption. This takes into account the kinetic energy recovery process unique to electric vehicles, making the prediction results more consistent with the actual energy income and expenditure of the vehicle.

[0091] The current journey is divided into several independent evaluation units at preset distance intervals. The distance intervals are calculated based on GPS location data, and the unit boundaries are determined by the distance differences between consecutive location points. This avoids the problem of significant local deviations being masked by overall data when making a single error judgment for the entire long journey. When the average error between predicted net energy consumption and actual energy consumption in three consecutive evaluation units exceeds a threshold, the additional force calculation coefficients in the vehicle dynamics model are dynamically updated based on the average error. The error in a single evaluation unit may stem from random factors, but the excessive errors in multiple consecutive units indicate a decrease in the adaptability of the additional force calculation in the model to the actual terrain and vehicle conditions. For example, an increase in vehicle load leads to a change in total mass, or the vehicle's resistance parameters change after long-term driving. In this case, updating the additional force calculation coefficients allows the model to readjust to the current actual conditions and maintain the accuracy of subsequent predictions.

[0092] In real-world applications, vehicles may travel from flat plains to mountainous areas with multiple slopes during long-distance travel. The additional force coefficient calibrated based on plain road conditions may have significant errors in mountainous areas. When the error of the evaluation unit exceeds the standard for three consecutive mountain road sections, the system automatically updates the coefficient to make the calculation of additional energy consumption for subsequent mountain road sections more accurate. In urban roads, if a vehicle temporarily increases its load, the original additional force coefficient cannot match the new total mass. The error of three consecutive short-distance evaluation units will trigger a coefficient update to ensure that the net energy consumption prediction under subsequent urban road conditions still matches reality.

[0093] like Figure 2 As shown, the real-time acquisition of the vehicle GPS raw data stream specifically includes:

[0094] S110 continuously acquires a raw GPS data stream containing timestamps, latitude and longitude, speed and altitude from the vehicle's GPS receiver;

[0095] S120 smooths the altitude data in the raw GPS data stream, eliminating random fluctuations caused by signal multipath effects and atmospheric delay.

[0096] S130 integrates and corrects the smoothed altitude data with the digital elevation model to generate a continuous and smooth altitude sequence.

[0097] This altitude sequence is a collection of altitude data arranged continuously in chronological order, with each time point corresponding to an accurate altitude value. The data shows no significant jumps and exhibits a continuous and smooth trend. As the foundational data for subsequent terrain analysis, it provides reliable input for calculating instantaneous slope, identifying uphill and downhill sections, and assessing the degree of terrain undulation, eliminating random errors in the original GPS data and ensuring the accuracy of energy consumption prediction-related calculations.

[0098] like Figure 3 As shown, the calculation of the slope change rate specifically includes:

[0099] S210, sample the altitude sequence at equal time intervals, calculate the difference in altitude between adjacent sampling points, and combine the time intervals to obtain the instantaneous slope value of each sampling point;

[0100] For instantaneous slope values:

[0101] ;

[0102] in:

[0103] : No. The instantaneous slope value at each sampling point reflects the elevation change per unit horizontal distance;

[0104] : No. The sampling point and the first The horizontal distance between each sampling point;

[0105] : No. The instantaneous velocity of each sampling point is derived from GPS data;

[0106] The time interval for sampling at equal time intervals is a fixed value.

[0107] S220, construct a sliding time window for the instantaneous slope value sequence, fit the slope change trend within the sliding time window, and obtain the instantaneous slope change rate sequence;

[0108] Instantaneous slope value sequences may contain noise, and directly calculating the rate of change will result in excessively fluctuating results that fail to reflect the true trend of terrain change. Sliding windows can smooth out noise by fitting local data, and more accurately extract the pattern of slope change over time.

[0109] Specifically:

[0110] 1. Set up a sliding window: Define the window size, including... A series of instantaneous slope values, Set according to road condition complexity;

[0111] 2. Window sliding: The window slides from left to right along the instantaneous slope value sequence, moving one sampling point at a time to cover the entire sequence;

[0112] 3. Trend Fitting: For the instantaneous slope value sequence within each window The linear least squares method is used to fit the straight line. ,in This represents the rate of change of slope within the window (i.e., the trend slope).

[0113] 4. Generate a sequence: Calculate the slope change rate for each window. Corresponding to the center or starting time of the window, a continuous sequence of instantaneous slope change rates is formed.

[0114] Through linear fitting within the sliding window, each window outputs a slope change rate, ultimately forming an instantaneous slope change rate sequence with the same length as the original sequence (or slightly shorter, depending on the window size), reflecting the continuous change trend of slope over time.

[0115] S230 uses the root mean square value of the slope change rate sequence statistically obtained by a sliding time window as an indicator of the slope change rate to quantify the severity of terrain undulation.

[0116] The root mean square value can effectively quantify the fluctuation range of a sequence (i.e., the severity of slope changes) and avoid the cancellation of positive and negative fluctuations; the sliding window can continuously evaluate over time, balance local fluctuations with the overall trend, reduce the impact of single outliers, and make the results more stable.

[0117] The slope change rate index is used to quantify the severity of terrain undulations and is a core input for subsequent calculations of additional acceleration requirements. The larger the index value, the more severe the terrain undulations, and the more additional energy the vehicle needs to overcome the slope changes. Therefore, this index is directly related to the accuracy of the calculation of additional energy consumption.

[0118] like Figure 4 As shown, the calculation of additional energy loss specifically includes:

[0119] S310: Establish the vehicle longitudinal dynamics equation including the slope change rate input, and calculate the additional acceleration required to maintain the vehicle's constant speed based on the real-time acquired slope change rate.

[0120] ;

[0121] in:

[0122] The additional acceleration required to maintain a constant vehicle speed;

[0123] Gravitational acceleration;

[0124] : No. The rate of change of slope at any given time, i.e. the value in the instantaneous rate of change of slope sequence, reflects the amount of change in slope per unit time.

[0125] S320, taking into account the total vehicle mass and additional acceleration requirements, calculates the additional force required to overcome terrain changes;

[0126] ;

[0127] in:

[0128] The additional forces required to overcome changes in terrain;

[0129] Total vehicle mass.

[0130] The S330 multiplies the additional force with the instantaneous velocity obtained through the GPS data stream to calculate the additional power loss caused by terrain fluctuations in real time, and then performs numerical integration of the additional power loss along the time axis to obtain the additional energy loss for the entire predicted journey.

[0131] 1. Regarding additional power loss:

[0132] ;

[0133] : Additional power loss due to terrain fluctuations at all times;

[0134] : Constantly overcoming the additional forces required by changes in terrain;

[0135] : Instantaneous velocity at a given moment.

[0136] 2. Regarding additional energy consumption and loss:

[0137] ;

[0138] Additional energy loss during the entire predicted journey;

[0139] Trip start time;

[0140] End time of the trip.

[0141] like Figure 5 As shown, the predicted electrical energy that can be effectively recovered on the next downhill section specifically includes:

[0142] S410 continuously collects the status parameters of the vehicle's kinetic energy recovery system, including the recovery intensity level setting, battery charging acceptance capacity, and motor operating mode, and identifies the vehicle's current actual kinetic energy recovery mode.

[0143] S420 uses road profile analysis based on elevation sequence to identify downhill sections and extract parameters of downhill sections, including slope angle, length and elevation difference.

[0144] The identification of downhill sections is as follows:

[0145] 1. Sort the altitude sequence by time to ensure continuity;

[0146] 2. Calculate the elevation difference between adjacent sampling points. ;

[0147] 3. When When the altitude decreases, it is marked as a potential downhill point;

[0148] 4. Multiple consecutive The sampling points are divided into road segments, and the total horizontal distance of the road segment is calculated. ) and total elevation difference ( );

[0149] 5. Validity Verification: If the total horizontal distance of the road segment exceeds the preset threshold and the average slope (total elevation difference / total horizontal distance) is... If so, it is determined to be a valid downhill section.

[0150] S430 calculates the corresponding energy recovery efficiency coefficient based on the identified actual kinetic energy recovery mode and downhill road parameters.

[0151] ;

[0152] in:

[0153] Energy recovery efficiency coefficient ( );

[0154] : Recycling intensity level coefficient, based on the recycling intensity set for the vehicle;

[0155] The battery charge acceptance factor, provided by the battery management system, reflects the proportion of charging power that the battery can currently accept.

[0156] Motor operating mode coefficient, reflecting the regenerative braking efficiency of the motor.

[0157] S440 calculates the theoretical recoverable energy based on downhill road parameters and vehicle GPS raw data stream, and multiplies it by the energy recovery efficiency coefficient to obtain the actual recoverable electrical energy.

[0158] ;

[0159] in:

[0160] The theoretical recoverable energy on a downhill section reflects the sum of changes in the vehicle's gravitational potential energy and kinetic energy during the downhill process;

[0161] The total elevation difference of a downhill section is a parameter of the downhill section, namely the difference in altitude between the starting point and the ending point of the section.

[0162] : The initial instantaneous speed of the vehicle entering the downhill section, speed parameters from the GPS raw data stream;

[0163] The final instantaneous speed of the vehicle as it exits the downhill section is a speed parameter derived from the raw GPS data stream. If the vehicle is descending at a constant speed... At this point, the change in kinetic energy is 0.

[0164] like Figure 6 As shown, the analysis of the vehicle's predicted net energy consumption under the current path and driving conditions specifically includes:

[0165] S510 identifies the vehicle's baseline energy consumption, aligns the additional energy loss obtained from the vehicle dynamics model with the baseline energy consumption over time, and calculates the total positive energy consumption.

[0166] ;

[0167] Parameter description:

[0168] Total positive energy consumption;

[0169] Vehicle baseline energy consumption refers to the energy consumption of driving at a constant speed on a flat road surface, which is obtained from the vehicle dynamics model.

[0170] S520 calculates the complete predicted net energy consumption of the vehicle based on total positive energy consumption and predicted recoverable electrical energy.

[0171] ;

[0172] Predicted net energy consumption of vehicles;

[0173] Recoverable electrical energy is determined by the theoretical recoverable energy and efficiency coefficient. Multiply them to get the result.

[0174] S530 divides the current journey into several independent evaluation units according to a preset distance interval. When the average error between the predicted net energy consumption and the actual energy consumption in three consecutive evaluation units exceeds 5%, the additional force calculation coefficient in the vehicle dynamics model is dynamically updated based on the average error.

[0175] Errors in individual evaluation units may be caused by random factors and do not reflect biases in the model itself; if the errors of three consecutive evaluation units exceed the threshold, it indicates a systematic bias in the model, requiring calibration. Dynamically updating the additional force calculation coefficients can make the vehicle dynamics model more closely match actual road conditions, reduce long-term prediction errors, and improve the accuracy and robustness of energy consumption prediction.

[0176] Figure 7 This is a structural block diagram of an electric vehicle energy consumption prediction system utilizing GPS data provided in an embodiment of the present invention, as shown below. Figure 7 As shown, the system includes:

[0177] The GPS data stream acquisition module 100 is used to acquire the vehicle's raw GPS data stream in real time, perform preprocessing, and extract the altitude sequence.

[0178] The slope calculation module 200 is used to calculate the instantaneous slope value of the altitude at adjacent time points based on the altitude sequence, and to calculate the slope change rate that reflects the severity of the terrain undulation.

[0179] The additional energy consumption calculation module 300 is used to calculate in real time the additional acceleration required to overcome the slope change based on the slope change rate, and then calculate the additional energy consumption loss.

[0180] The electric energy prediction module 400 is used to identify the vehicle's current kinetic energy recovery mode based on altitude sequence and speed data, and call the vehicle's kinetic energy recovery efficiency model to predict the electric energy that can be effectively recovered on the next downhill section.

[0181] The Net Energy Analysis Module 500 is used to analyze the vehicle's predicted net energy consumption under the current route and driving conditions, based on the vehicle's baseline energy consumption and incorporating additional energy losses and recoverable electrical energy.

[0182] like Figure 8 As shown, the slope calculation module 200 includes:

[0183] The interval sampling module 210 is used to sample the altitude sequence at equal time intervals, calculate the difference in altitude between adjacent sampling points, and obtain the instantaneous slope value of each sampling point by combining the time interval.

[0184] The sliding window construction module 220 is used to construct a sliding time window for the instantaneous slope value sequence, fit the slope change trend within the sliding time window, and obtain the instantaneous slope change rate sequence.

[0185] The rate of change index calculation module 230 is used to calculate the root mean square value of the slope change rate sequence using a sliding time window, which serves as a slope change rate index to quantify the severity of terrain undulation.

[0186] like Figure 9 As shown, the additional energy consumption calculation module 300 includes:

[0187] The equation building module 310 is used to build the vehicle longitudinal dynamics equations including the slope change rate input, and calculate the additional acceleration required to maintain the vehicle's constant speed based on the real-time acquired slope change rate.

[0188] The additional force calculation module 320 is used to calculate the additional force required to overcome terrain changes by combining the total mass of the vehicle and the additional acceleration requirements.

[0189] The energy loss calculation module 330 is used to multiply the additional force by the instantaneous speed obtained through the GPS data stream, calculate the additional power loss caused by terrain fluctuations in real time, and perform numerical integration of the additional power loss along the time axis to obtain the additional energy loss for the entire predicted journey.

[0190] like Figure 10 As shown, the electrical energy prediction module 400 includes:

[0191] The kinetic energy recovery parameter acquisition module 410 is used to continuously acquire the status parameters of the vehicle's kinetic energy recovery system, including the recovery intensity level setting, battery charging acceptance capacity and motor operating mode, and to identify the vehicle's current actual kinetic energy recovery mode.

[0192] The downhill section identification module 420 is used to perform road profile analysis based on the altitude sequence, identify downhill sections, and extract downhill section parameters, including slope angle, length, and elevation difference parameters.

[0193] The energy recovery efficiency calculation module 430 is used to calculate the corresponding energy recovery efficiency coefficient based on the identified actual kinetic energy recovery mode and downhill road parameters.

[0194] The actual recovery analysis module 440 is used to calculate the theoretical recoverable energy based on the downhill road parameters and the vehicle's raw GPS data stream, and multiply it by the energy recovery efficiency coefficient to obtain the actual recoverable electrical energy.

[0195] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0196] The embodiments described above are merely illustrative of several implementations of the present invention, and while the descriptions are specific and detailed, they should not be construed as limiting the scope of the present invention. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of the present invention, and these modifications and improvements all fall within the scope of protection of the present invention. Therefore, the scope of protection of this patent should be determined by the appended claims.

[0197] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A method for predicting energy consumption of an electric vehicle using GPS data, characterized by, The method comprises: Real-time acquisition of vehicle GPS raw data stream, and preprocessing, extraction of altitude sequence; Based on the altitude sequence, the instantaneous slope value of the adjacent time point altitude is calculated, and the slope change rate reflecting the degree of terrain undulation is calculated; According to the slope change rate, the additional acceleration requirement for overcoming the slope change is calculated in real time, and then the additional energy consumption is calculated; Based on the altitude sequence and speed data, the current kinetic energy recovery mode of the vehicle is identified, and the vehicle kinetic energy recovery efficiency model is called to predict the electric energy that can be effectively recovered in the next downhill section; Based on the vehicle reference energy consumption, combined with the additional energy consumption and the recoverable electric energy, the vehicle predicted net energy consumption under the current path and driving conditions is analyzed; The real-time acquisition of vehicle GPS raw data stream specifically includes: Continuously acquiring GPS raw data stream containing time stamp, latitude, longitude, speed and altitude from vehicle GPS receiver; Smooth the altitude data in the GPS raw data stream to eliminate random fluctuations caused by signal multipath effect and atmospheric delay; Fuse and correct the smoothed altitude data with digital elevation model to generate continuous and smooth altitude sequence; The calculation of slope change rate specifically includes: Sampling the altitude sequence at equal time intervals, calculating the altitude difference between adjacent sampling points, and combining the time interval to obtain the instantaneous slope value of each sampling point; Construct a sliding time window of the instantaneous slope value sequence, fit the slope change trend in the sliding time window, and get the instantaneous slope change rate sequence; Adopting sliding time window to calculate the root mean square value of the slope change rate sequence as the slope change rate index quantifying the degree of terrain undulation; The analysis of the vehicle predicted net energy consumption under the current path and driving conditions specifically includes: Identify the vehicle reference energy consumption, align the additional energy consumption obtained from the vehicle dynamics model with the reference energy consumption in time sequence, and calculate the total forward energy consumption; Based on the total forward energy consumption and the predicted recoverable electric energy, the complete vehicle predicted net energy consumption is calculated; Divide the current trip into several independent evaluation units at a preset distance interval, when the average error of the predicted net energy consumption and the actual energy consumption in the continuous 3 evaluation units is more than 5%, dynamically update the additional force calculation coefficient in the vehicle dynamics model according to the average error.

2. The method of claim 1, wherein, The calculation of additional energy consumption specifically includes: Establish a vehicle longitudinal dynamics equation containing slope change rate input, calculate the additional acceleration requirement required to maintain uniform speed of the vehicle according to the real-time acquired slope change rate; Combined with the total mass of the vehicle and the additional acceleration requirement, the additional force required to overcome the terrain change is calculated; Multiply the additional force by the instantaneous speed obtained through the GPS data stream to calculate the additional power loss caused by terrain fluctuation in real time, and integrate the additional power loss along the time axis to get the additional energy consumption of the entire predicted trip.

3. The method of claim 2, wherein, The prediction of the electric energy that can be effectively recovered in the next downhill section specifically includes: Continuously collect vehicle kinetic energy recovery system state parameters, including recovery intensity level setting, battery charging acceptance ability and motor working mode, identify the current actual kinetic energy recovery mode of the vehicle; Based on the altitude sequence, analyze the road profile, identify the downhill section, and extract the downhill section parameters, including slope angle, length and elevation difference parameters; According to the identified actual kinetic energy recovery mode and downhill section parameters, calculate the corresponding energy recovery efficiency coefficient; According to the downhill section parameters and vehicle GPS original data stream, calculate the theoretical recoverable energy, and multiply it by the energy recovery efficiency coefficient to obtain the actual recoverable electric energy.

4. The method according to any of claims 1 to 3, characterized in that, The system for implementing the electric vehicle energy consumption prediction method using GPS data comprises: A GPS data stream acquisition module for acquiring vehicle GPS original data stream in real time, preprocessing, and extracting altitude sequence; A slope calculation module for calculating the instantaneous slope value of adjacent time points based on the altitude sequence, and calculating the slope change rate reflecting the degree of terrain fluctuation; An additional energy consumption calculation module for calculating the additional acceleration demand generated by overcoming the slope change based on the slope change rate, and then calculating the additional energy loss; An electric energy prediction module for identifying the current kinetic energy recovery mode of the vehicle based on the altitude sequence and speed data, and calling the vehicle kinetic energy recovery efficiency model to predict the electric energy that can be effectively recovered in the next downhill section; A net energy consumption analysis module for analyzing the predicted net energy consumption of the vehicle under the current path and driving conditions based on the vehicle reference energy consumption, combined with the additional energy loss and recoverable electric energy.

5. The system of claim 4, wherein, The slope calculation module comprises: An interval sampling module for sampling the altitude sequence at equal time intervals, calculating the difference in altitude between adjacent sampling points, and combining the time interval to obtain the instantaneous slope value of each sampling point; A sliding window construction module for constructing a sliding time window of the instantaneous slope value sequence, fitting the slope change trend in the sliding time window, and obtaining the instantaneous slope change rate sequence; A change rate index calculation module for calculating the root mean square value of the slope change rate sequence using the sliding time window as the slope change rate index quantifying the degree of terrain fluctuation.

6. The system of claim 5, wherein, The additional energy consumption calculation module comprises: An equation establishment module for establishing a vehicle longitudinal dynamics equation containing slope change rate input, and calculating the additional acceleration demand required to maintain uniform speed of the vehicle according to the real-time acquired slope change rate; An additional force calculation module for calculating the additional force required to overcome terrain changes by combining the total mass of the vehicle with the additional acceleration demand; An energy loss calculation module for multiplying the additional force with the instantaneous speed obtained through the GPS data stream to calculate the additional power loss caused by terrain fluctuations in real time, and numerically integrating the additional power loss along the time axis to obtain the additional energy loss of the entire predicted journey.

7. The system of claim 6, wherein, The electric energy prediction module comprises: A kinetic energy recovery parameter acquisition module for continuously collecting vehicle kinetic energy recovery system state parameters, including recovery intensity level setting, battery charging acceptance ability and motor working mode, identifying the current actual kinetic energy recovery mode of the vehicle; The downhill section identification module is configured to perform road profile analysis based on the altitude sequence, identify a downhill section, and extract downhill section parameters including a slope angle, a length, and an elevation difference parameter; The recovery efficiency calculation module is configured to calculate a current corresponding energy recovery efficiency coefficient according to the identified actual kinetic energy recovery mode and the downhill section parameters; The actual recovery analysis module is configured to calculate theoretical recoverable energy according to the downhill section parameters and the vehicle GPS original data stream, and multiply the theoretical recoverable energy by the energy recovery efficiency coefficient to obtain actual recoverable electric energy.

Images