New Energy Heavy Truck Energy Consumption Map Construction Method and Electronic Device

Abstract

The present invention relates to the field of new energy technologies, and specifically discloses a method for constructing an energy consumption map of a new energy heavy truck and an electronic device. The method for constructing an energy consumption map of a new energy heavy truck includes: cleaning invalid data in vehicle driving data; clustering the above-mentioned valid data according to the driving vehicle weight to form a data structure of {vehicle weight interval: {set of driving data segments}}, where each vehicle weight interval corresponds to a set of clustered driving data segments; associating the obtained data structure with the geographical road network to match the vehicle energy consumption information with the road sections actually traveled by the vehicle; establishing and training a road section-level energy consumption model based on the data after the above-mentioned matching, and integrating the output results of the model into a global energy consumption map. The method for constructing an energy consumption map of a new energy heavy truck implemented by the present invention constructs an energy consumption map of a new energy heavy truck by combining geographical information with vehicle energy consumption data, improves driving safety and efficiency; and accurately plans low-energy consumption routes.

Description

Technical Field

[0001] The present invention relates to the field of new energy technologies, and particularly to a method for constructing an energy consumption map of a new energy heavy truck and an electronic device. Background Art

[0002] With the development of new energy vehicles, how to reduce the energy consumption of new energy vehicles and increase the cruising range has become a key challenge. During the operation of new energy heavy trucks, due to significant changes in vehicle weight, high dependence on energy replenishment facilities, and the fact that existing research has not fully considered their driving characteristics, the energy utilization efficiency is not high, the energy consumption remains high, and thus the cruising range is affected.

[0003] Current research on energy consumption maps fails to fully reflect the specific operating characteristics of new energy heavy trucks. Existing solutions mostly focus on improving battery technology and drive system efficiency, ignoring the use of geographical information to optimize route selection and unable to provide users with the most economical and efficient driving routes. Summary of the Invention

[0004] The present invention aims to at least partly solve one of the technical problems in the related technologies. To this end, the object of the present invention is to propose a method for constructing an energy consumption map of a new energy heavy truck and an electronic device, which combines geographical information with vehicle energy consumption data to reduce vehicle energy consumption.

[0005] To achieve the above object, a first aspect embodiment of the present invention proposes a method for constructing an energy consumption map of a new energy heavy truck, the method comprising:

[0006] Cleaning invalid data in the vehicle driving data to obtain valid data;

[0007] Clustering the above valid data according to the driving vehicle weight to form a data structure of {vehicle weight interval: {set of driving data segments}}, where each vehicle weight interval corresponds to a set of clustered driving data segments;

[0008] Associating the obtained data structure with the geographical road network, matching the vehicle energy consumption information with the actual driving sections of the vehicle, and establishing a mapping relationship of "section - energy consumption";

[0009] Based on the above - matched data, establishing a section - level energy consumption model and training it, and integrating the output results of the model into a global energy consumption map.

[0010] In some embodiments of the present invention, the cleaning of invalid data in the vehicle driving data includes the following:

[0011] For any vehicle driving trajectory point, if it satisfies the following conditions, it is invalid driving data:

[0012] ① Data marked as invalid itself;

[0013] ② Abnormal position;

[0014] ③ Abnormal speed;

[0015] ④ Abnormal current or voltage;

[0016] The clearing condition of the trajectory points is as follows:

[0017]

[0018] In the above formula, is the data validity flag, is the lower left longitude of the geographical scope of China, is the upper right longitude of the geographical scope of China, is the lower left latitude of the geographical scope of China, is the upper right latitude of the geographical scope of China, is the theoretical maximum vehicle speed, is the maximum current threshold, is the minimum current threshold, is the maximum voltage threshold, is the minimum voltage threshold.

[0019] In some embodiments of the present invention, the clustering of the above-mentioned valid data according to the driving vehicle weight includes the following:

[0020] First, the driving trajectory of the vehicle is segmented into several driving data segments according to the actual vehicle weight during driving, and each driving data segment represents the continuous driving process of the vehicle under a certain driving vehicle weight;

[0021] Then, for each defined driving vehicle weight interval, all driving data segments belonging to this interval are collected to form a data structure of {vehicle weight interval: {set of driving data segments}}.

[0022] In some embodiments of the present invention, the matching of the vehicle energy consumption information with the actual driving section of the vehicle includes the following:

[0023] Obtain the matching section, and select different obtaining methods according to different situations of the vehicle trajectory points;

[0024] If the previous trajectory point of the vehicle matches successfully with a certain section in the geographical road network, and the current trajectory point is less than the perpendicular distance from the matching point corresponding to the previous successful match, the matching section is obtained by using the road network topology method;

[0025] If the previous trajectory point of the vehicle fails to match successfully with a certain section in the geographical road network, a geographical rectangle frame is constructed based on the current trajectory point, and the matching section is obtained by using the geographical frame selection method.

[0026] In some embodiments of the present invention, the method for obtaining matching road segments using the road network topology is expressed as follows:

[0027]

[0028] In the above formula, is the road network topology road segment set, is the first-level topology road segment, is the second-level topology road segment.

[0029] In some embodiments of the present invention, when obtaining matching road segments using the geographic selection method, the selection area should be dynamically adjusted according to the vehicle motion state, including the following:

[0030] First, dynamically calculate the side length of the selection:

[0031] Based on the current vehicle speed v of the trajectory point and the direction change rate adjust the side length L of the selection. The side length L of the selection is expressed as follows:

[0032]

[0033] In the above formula, is the reference side length, and are adjustment coefficients, is the maximum direction change angle;

[0034] Then, constrain the fan-shaped area in the selection direction:

[0035]

[0036] Define the angular range of the fan-shaped search area, and only match candidate road segments with the same direction.

[0037] In some embodiments of the present invention, the matching road segments obtained by the road network topology method and the geographic selection method are used as alternatives, and the optimal matching road segment is found from the set of alternative road segments, specifically including the following:

[0038] Set as the angular distance between the trajectory point and the road segment, and set as the perpendicular distance between the trajectory point and the road segment;

[0039] can be expressed as:

[0040]

[0041] In the above formula, is the angular influence coefficient threshold, is the included angle between the trajectory point direction and the road segment direction;

[0042] For and , first perform normalization processing on both of them. The perpendicular distance after normalization processing can be expressed as:

[0043]

[0044] In the above formula, is the maximum value between the trajectory point and the road segment, is the minimum value between the trajectory point and the road segment;

[0045] The angular distance after normalization processing can be expressed as:

[0046]

[0047] For each road segment, its comprehensive score The calculation formula is:

[0048]

[0049] In the above formula, is the weight of the direction factor, is the weight of the distance factor, and + = 1, 0 ≤ ≤ 1, 0 ≤ ≤ 1;

[0050] By comparing the comprehensive scores of adjacent road segments of the trajectory point, the optimal matching road segment is obtained.

[0051] In some embodiments of the present invention, when establishing the road segment-level energy consumption model, the weighted trimmed mean algorithm is used to remove extreme values in the energy consumption data, and the interval index storage structure is used to store the energy consumption data classified by vehicle weight interval, specifically including:

[0052] For a certain vehicle weight interval on the road segment, there are n observed energy consumption values ; first, sort these energy consumption values, let k = 0.05n, remove the highest k values and the lowest k values, and then calculate the weighted average of the remaining data as the road segment energy consumption of this vehicle weight interval. The formula is expressed as:

[0053]

[0054] In the above formula, is the weight of each energy consumption value, and p is the index in the sorted energy consumption value sequence.

[0055] In some embodiments of the present invention, when constructing the global energy consumption map, the road segments in the geographical road network are stratified, including low-level road segments and high-level road segments. Among them, the relationship between the high-level road segments and the low-level road segments is a one-to-many relationship. The energy consumption corresponding to the high-level road segments can be expressed as:

[0056]

[0057]

[0058] In the above formula, is the average energy consumption value per kilometer of the i-th road segment in the low-level road segment sequence, is the weighted value of the road segment attribute, is the length of the i-th road segment in the low-level road segment sequence, is the number of low-level road segments corresponding to the high-level road segment, is the length of the high-level road segment.

[0059] To achieve the above object, an embodiment of the second aspect of the present invention proposes an electronic device, including a memory, a processor, and a computer program stored on the memory. When the computer program is executed by the processor, the above-mentioned new energy heavy truck energy consumption map construction method is implemented.

[0060] The new energy heavy truck energy consumption map construction method and the electronic device according to the embodiments of the present invention can construct a new energy heavy truck energy consumption map by combining geographical information with vehicle energy consumption data, which can cultivate good driving habits of drivers in real time, improve driving safety and efficiency; accurately plan low-energy consumption routes, reduce the cost of inducing drivers; relieve traffic congestion, improve road use efficiency; and accurately reflect vehicle operation characteristics, optimize route selection, increase the cruising range, and reduce the dependence on energy replenishment facilities. BRIEF DESCRIPTION OF THE DRAWINGS

[0061] Figure 1 is a schematic flowchart of a new energy heavy truck energy consumption map construction method according to an embodiment of the present invention;

[0062] Figure 2 is a schematic diagram of obtaining matching road segments by using a road network topology method according to an embodiment of the present invention;

[0063] Figure 3 is a schematic diagram of obtaining matching road segments by using a box selection method according to an embodiment of the present invention;

[0064] Figure 4 is a schematic diagram of the corresponding relationship between high-level road segments and low-level road segments according to an embodiment of the present invention;

[0065] Figure 5 is a schematic structural diagram of an electronic device according to another embodiment of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0066] Embodiments of the present invention will be described in detail below. Examples of the embodiments are shown in the accompanying drawings, where the same or similar reference numerals denote the same or similar elements or elements having the same or similar functions throughout. The embodiments described below by referring to the accompanying drawings are exemplary and are intended to explain the present invention, and should not be construed as limiting the present invention.

[0067] The new energy heavy truck energy consumption map construction method and electronic device according to an embodiment of the present invention will be described below with reference to the accompanying drawings.

[0068] Figure 1 It is a flowchart of the new energy heavy truck energy consumption map construction method according to an embodiment of the present invention.

[0069] As shown in Figure 1 the new energy heavy truck energy consumption map construction method includes:

[0070] S1. Clean the invalid data in the vehicle driving data to obtain valid data.

[0071] This step aims to eliminate interference data and ensure the accuracy of subsequent analysis. First, according to the data itself mark, remove the data marked as invalid. Then, judge the abnormal position through the longitude and latitude range. Finally, compare with the maximum theoretical vehicle speed and the maximum current threshold, minimum current threshold, maximum voltage threshold, and minimum voltage threshold in turn. If it exceeds the normal range, it is regarded as abnormal.

[0072] As an example, for any vehicle driving trajectory point, if it meets the following conditions, it is invalid driving data:

[0073] ① The data marked as invalid itself;

[0074] ② Abnormal position;

[0075] ③ Abnormal speed;

[0076] ④ Abnormal current or voltage;

[0077] The clearing conditions of the trajectory point are as follows:

[0078]

[0079] In the above formula, is the data validity flag, is the lower left longitude of the Chinese geographical range, is the upper right longitude of the Chinese geographical range, is the lower left latitude of the Chinese geographical range, and is the upper right latitude of the Chinese geographical range, is the theoretical maximum vehicle speed, is the maximum current threshold, is the minimum current threshold, is the maximum voltage threshold, is the minimum voltage threshold.

[0080] S2. Cluster the above effective data according to the driving vehicle weight to form a data structure of {vehicle weight interval: {set of driving data segments}}, where each vehicle weight interval corresponds to a set of clustered driving data segments.

[0081] Since the vehicle weight significantly affects the energy consumption of new energy heavy trucks, this step first divides the vehicle driving trajectory according to the actual driving vehicle weight. For example, during the process of loading and unloading goods, the vehicle weight changes, and the driving data segments are divided by the vehicle weight change nodes. Each data segment represents a continuous driving process under a specific vehicle weight. Then, different vehicle weight intervals are set, such as [10−15]t, [15−20]t, etc., and each data segment is classified into the corresponding interval to form a data structure of {vehicle weight interval: {set of driving data segments}}, which is convenient for subsequent analysis of energy consumption characteristics for different vehicle weight intervals.

[0082] S3. Associate the obtained data structure with the geographical road network, match the vehicle energy consumption information with the actual driving sections of the vehicle, and establish a mapping relationship of "section - energy consumption".

[0083] As an example, matching the vehicle energy consumption information with the actual driving sections of the vehicle includes the following content:

[0084] Obtain the matching section, and select different obtaining methods according to different situations of vehicle trajectory points;

[0085] If the previous trajectory point of the vehicle successfully matches a certain section in the geographical road network, and the current trajectory point is less than the perpendicular distance from the matching point corresponding to the previous successful match, then the road network topology method is used to obtain the matching section;

[0086] If the previous trajectory point of the vehicle fails to match a certain section in the geographical road network, then a geographical rectangle is constructed based on the current trajectory point, and the geographical box selection method is used to obtain the matching section.

[0087] S4. Based on the above - matched data, establish a section - level energy consumption model and train it, and integrate the output results of the model into a global energy consumption map.

[0088] This solution deeply combines geographical information with vehicle energy consumption data, and constructs a global energy consumption map according to the energy consumption conditions of different sections. With the help of this map, drivers can intuitively compare the energy consumption differences of different routes, accurately select the driving route with the lowest energy consumption. In urban scenarios, new energy heavy trucks can avoid congested sections with high energy consumption or sections with large slopes, and choose flat and less - trafficked routes with lower energy consumption, reducing the vehicle energy consumption and the cost of inducing drivers to choose high - energy - consumption routes.

[0089] In some embodiments of the present invention, the geographical range is dynamically adjusted according to the actual operation area of the vehicle, and the adjustment formula can be expressed as:

[0090] = min(trajectory longitude) - x, = max(trajectory longitude) + x

[0091] Where, x is the extended buffer value (such as 0.1 degree).

[0092] As an example, assume that the longitude of a certain trajectory point is 116.40°, the latitude is 39.90°, the vehicle speed is 150 km / h (exceeding the theoretical maximum value of 120 km / h), and the current is 300 A (exceeding the threshold [50 A, 250 A]). Then this point is marked as invalid data.

[0093] In some embodiments of the present invention, clustering the above valid data according to the driving vehicle weight includes the following:

[0094] First, divide the driving trajectory of the vehicle into several driving data segments according to the actual vehicle weight during driving. Each driving data segment represents the continuous driving process of the vehicle under a certain driving vehicle weight;

[0095] Then, for each defined driving vehicle weight interval, collect all the driving data segments belonging to this interval to form a data structure of {vehicle weight interval: {set of driving data segments}}.

[0096] As an example, during the process of loading and unloading goods, the vehicle weight changes. Divide the driving data segments with the vehicle weight change nodes as the boundaries. Each data segment represents the continuous driving process under a specific vehicle weight. Then, set different vehicle weight intervals, such as [10 - 15] t, [15 - 20] t, etc., and classify each data segment into the corresponding interval to form a data structure of {vehicle weight interval: {set of driving data segments}}, which is convenient for subsequent analysis of energy consumption characteristics for different vehicle weight intervals.

[0097] Example: A new energy heavy truck transports goods. When it departs, the vehicle weight is 13 t. After driving a certain distance, it arrives at the unloading point and the vehicle weight becomes 8 t. Then the driving data segment from departure to before unloading is classified into the vehicle weight interval of [10 - 15] t, and the driving data segment after unloading is reclassified according to the subsequent vehicle weight situation. If the vehicle weight stabilizes at 8 t after unloading, this data segment can be classified into the interval of [5 - 10] t (assuming such an interval exists).

[0098] It should be noted that considering the non - linear relationship between vehicle weight and energy consumption, on the basis of the existing clustering method, improvements can also be made in aspects such as vehicle weight interval division, data processing, and model construction. For example:

[0099] The traditional way of dividing the fixed vehicle weight range cannot adapt to the non-linear relationship between vehicle weight and energy consumption. The density-based spatial clustering algorithm (DBSCAN) is used to dynamically determine the vehicle weight range. This algorithm can automatically identify the core area and boundary area of the data according to the distribution density of the data points, and divide the data points with connected density into the same class. In the driving data of new energy heavy trucks, the distribution of vehicle weight and energy consumption data points can reflect different vehicle weight - energy consumption relationship clusters. Through the DBSCAN algorithm, a more reasonable vehicle weight range can be automatically divided in the area where the vehicle weight data is dense. If in a certain specific transportation scenario, the vehicle weight difference is large when the vehicle is fully loaded and unloaded, and the energy consumption performance is different, the algorithm can accurately identify the vehicle weight ranges corresponding to different states such as full load, half load, and no load, avoiding the problem of excessive data differences within the range caused by fixed range division.

[0100] In addition, if the data volume in a certain range is insufficient (such as < 100 records), it is merged with the adjacent range to ensure statistical significance.

[0101] For example: In a certain section, the dense area of vehicle weight data is 18 - 22 tons, and DBSCAN automatically divides the range [18, 22), and the remaining data is merged into the adjacent range.

[0102] As Figure 2 shown, in some embodiments of the present invention, the matching section is obtained by using the road network topology method as follows:

[0103]

[0104] In the above formula, is the road network topology section set, is the primary topology section, is the secondary topology section.

[0105] It should be noted that in the actual complex road network, such as urban roads including main roads, secondary roads and branch roads, the primary topology section can be understood as the section directly connected to the current matching point, and the secondary topology section is the surrounding section connected to the primary topology section. By comprehensively analyzing these sections, the matching section set that most conforms to the vehicle driving path is determined.

[0106] In practical applications, the elevated road and the ground road can be further distinguished to avoid cross-layer mis-matching, and the matching condition is set as:

[0107] Section level ( ) = current track point level

[0108] As an example, if the vehicle is driving on an elevated road section, only match the elevated topology section , and exclude the ground road.

[0109] As Figure 3As shown, in some embodiments of the present invention, when obtaining matching road segments by means of geographical selection, the selected area should be dynamically adjusted according to the vehicle's motion state, including the following:

[0110] First, dynamically calculate the side length of the selection:

[0111] Based on the current vehicle speed v of the trajectory point and the direction change rate Adjust the side length L of the selection. The side length L of the selection is expressed as follows:

[0112]

[0113] In the above formula, is the reference side length, and are adjustment coefficients used to adjust the influence degrees of the vehicle speed v and the direction change rate on the side length L of the selection, is the maximum direction change angle;

[0114] It should be noted that the faster the vehicle speed, the larger v / , and the larger the side length of the selection, so as to cover the possible driving range of the vehicle; the greater the direction change rate, the larger ∣ ∣ / , and the smaller the side length of the selection, ensuring that the selected area is more in line with the driving direction of the vehicle;

[0115] Then, constrain the fan-shaped area in the selection direction:

[0116]

[0117] Define the angle range of the fan-shaped search area, and only match candidate road segments with the same direction; this formula calculates the angle range of the fan-shaped search area through the proportional relationship between the vehicle speed and the maximum vehicle speed, combined with a fixed angle, and only matches candidate road segments with the same direction within this range, reducing invalid matches and improving the matching efficiency.

[0118] Here, the significance of dynamically adjusting the selected area according to the vehicle's motion state is to significantly reduce the false matching rate through a hybrid search strategy combining dynamic selection and direction fan shape, and the algorithm steps can be summarized as:

[0119] First, calculate L and according to the vehicle's real-time speed and direction change rate;

[0120] Then, construct a rectangular frame with the current trajectory point as the center and side length L, and superimpose the fan-shaped direction filtering;

[0121] Finally, extract candidate road segments from the selected range and perform secondary filtering in combination with road attributes (height limit, weight limit).

[0122] It should be noted here that when obtaining the matching road section, in a certain urban road scenario, the vehicle is driving along the main road. The previous trajectory point is successfully matched with a certain road section of the main road, and the current trajectory point is relatively close to the previous matching point. At this time, the road network topology method is used to obtain the matching road section. Suppose there are secondary roads and branch roads around the main road. The first-level topological road section includes the part of the main road directly connected to the current matching point, and the second-level topological road section includes the adjacent secondary roads and branch road parts. The final matching road section is determined by screening these road sections. If the vehicle enters a new area and the previous trajectory point fails to be matched, the geographical selection method is used.

[0123] As an example, let the current vehicle speed v = 60 km / h, the reference side length = 200 m, = 0.6, = 0.4, = 100 km / h, = 20°, = 80°. Calculate the selected side length as follows:

[0124] L = 200×(1 + 0.6× )×(1 - 0.4× ) = 200×1.36×0.9 = 244.8 m;

[0125] Then calculate the angular range of the fan-shaped search area = max(30°, 90° - ×45°) = 63°;

[0126] Therefore, it is concluded that within a rectangular frame with the current trajectory point as the center and a side length of 244.8 m, and within a 63° fan-shaped area, candidate road sections with the same direction are searched for matching.

[0127] For further refinement of the above solution, the method of obtaining the matching road section by the selection method can be optimized. It is set that when the speed is low, the angular range is reduced to improve the matching accuracy. For example:

[0128] When v = 10 m / s, = 75°; when v = 30 m / s, = 45°.

[0129] In some embodiments of the present invention, the matching road sections obtained by the road network topology method and the geographical selection method are used as alternatives, and the optimal matching road section is found from the set of alternative road sections. Specifically, it includes the following contents:

[0130] Set as the angular distance between the trajectory point and the road section, and set is the perpendicular distance from the trajectory point to the road segment;

[0131] It can be expressed as:

[0132]

[0133] In the above formula, is the threshold of the angle influence coefficient, is the included angle between the direction of the trajectory point and the direction of the road segment;

[0134] For and , first perform normalization processing on the two, and the perpendicular distance after normalization processing can be expressed as:

[0135]

[0136] In the above formula, is the maximum value between the trajectory point and the road segment, is the minimum value between the trajectory point and the road segment;

[0137] The angle distance after normalization processing can be expressed as:

[0138]

[0139] For each road segment, its comprehensive score The calculation formula is:

[0140]

[0141] In the above formula, is the weight of the direction factor, is the weight of the distance factor, and + = 1, 0 ≤ ≤ 1, 0 ≤ ≤ 1;

[0142] By comparing the comprehensive scores of adjacent road segments of the trajectory point, the optimal matching road segment is obtained.

[0143] As an example, step one, screen out three candidate road segments through the road network topology and geographical bounding box method;

[0144] Road segment A: perpendicular distance 10 meters, direction included angle 30°;

[0145] Road segment B: perpendicular distance 20 meters, direction included angle 90°;

[0146] Road segment C: perpendicular distance 5 meters, direction included angle 15°.

[0147] Step 2: Calculate the perpendicular distance and the angular distance;

[0148] Perpendicular distance : Section A: 10 m; Section B: 20 m; Section C: 5 m;

[0149] The range of the perpendicular distance of all candidate sections: = 5 m, = 20 m.

[0150] Angular distance : Let = 0.5, then for Section A: (A) = 0.5 × 30° = 15°;

[0151] For Section B: The included angle exceeds the threshold ( ≈ 75°), (B) = ∞;

[0152] For Section C: (C) = 0.5 × 15° = 7.5°.

[0153] Step 3: Perform normalization;

[0154] Normalization of the perpendicular distance:

[0155] = 0.333; 1.0; 0.0;

[0156] Normalization of the angular distance:

[0157] = 0.083; ∞) (set as 1); = 0.042.

[0158] Step 4: Calculate the comprehensive score;

[0159] Assume the weights = 0.6 (direction), = 0.4 (distance);

[0160] 0.6 × (1 - 0.083) + 0.4 × (1 - 0.333) ≈ 0.817;

[0161] = 0.6 × (1 - 1) + 0.4 × (1 - 1) = 0.0 (eliminated);

[0162] = 0.6 × (1 - 0.042) + 0.4 × (1 - 0.0) ≈ 0.975.

[0163] Step Five, select the optimal matching section

[0164] Section C is selected with the highest score Score = 0.975 because its perpendicular distance is the shortest and the direction deviation is the smallest.

[0165] It should be noted that if the adjusted weights are = 0.4, = 0.6, then Section C still has the highest score (0.983), but the score of Section A increases to (0.767), which reflects the influence of the distance weight.

[0166] If the direction angle of Section B is 70° (not exceeding the threshold), then its score needs to be recalculated:[[]]

[0167] = 0.5×70° = 35°, ≈ 0.194;

[0168] = 0.6×(1 - 0.194)+0.4×(1 - 1.0)≈ 0.484+0.0 = 0.484

[0169] At this time, Section C is still the best, but the tendency of the selection of the best section can be seen:[[]]

[0170] Section C has the highest score because its perpendicular distance is the shortest (5 meters) and the direction deviation is the smallest (15°); Section B is eliminated due to excessive direction deviation; Section A has a medium score due to the second-best balance between distance and direction.

[0171] In some embodiments of the present invention, when establishing a section-level energy consumption model, the weighted trimmed mean algorithm is used to remove extreme values from the energy consumption data, and the interval index storage structure is used to store the energy consumption data classified by vehicle weight interval, specifically including:[[]]

[0172] For a certain vehicle weight interval on the section, there are n observed energy consumption values ; First, sort these energy consumption values, let k = 0.05n, remove the highest k values and the lowest k values, and then calculate the weighted average of the remaining data as the section energy consumption of this vehicle weight interval. The formula is expressed as:[[]]

[0173]

[0174] In the above formula, is the weight of each energy consumption value, and p is the index in the sorted energy consumption value sequence.

[0175] The above The energy consumption of a road section in a specific vehicle weight range is the result obtained by weighted average calculation of the processed energy consumption values. It comprehensively considers the energy consumption in various driving states within the specific vehicle weight range of this road section and is a key indicator for measuring the energy consumption level of this road section under this specific condition.

[0176] In summary, in the process of constructing the global energy consumption map, through the weighted truncated mean algorithm (removing the extreme values of the highest and lowest 5%), abnormal data (such as sudden increases in energy consumption caused by sudden acceleration or sensor failures) is effectively filtered, ensuring that the training data input into the model has a high signal-to-noise ratio. And using the interval index storage structure to classify and store the energy consumption data according to the vehicle weight range facilitates subsequent rapid query and analysis of the energy consumption of road sections under different vehicle weight ranges. Different vehicle weights have a significant impact on the energy consumption of new energy heavy trucks. This classification storage method combined with the actual situation provides strong support for establishing an accurate road section-level energy consumption model. The energy consumption value calculated by removing extreme values within a specific vehicle weight range of a certain road section can more truly reflect the energy consumption level of this road section under this vehicle weight condition and provide a reliable data basis for the model.

[0177] Such as Figure 4 As shown, in some embodiments of the present invention, when constructing the global energy consumption map, the road sections in the geographical road network are stratified, including low-level road sections and high-level road sections. Among them, the relationship between the high-level road sections and the low-level road sections is a one-to-many relationship. For example, the high-level road section The corresponding next-level road sections are , , ;

[0178] The energy consumption corresponding to the high-level road section can be expressed as:

[0179]

[0180]

[0181] In the above formula, is the average kilometer energy consumption value of the i-th road section in the low-level road section sequence, is the weighted value of the road section attribute, is the length of the i-th road section in the low-level road section sequence, is the number of low-level road sections corresponding to the high-level road section, is the length of the high-level road section. This way of calculating energy consumption by stratification can more comprehensively and accurately reflect the energy consumption of different road sections, provide richer information for the global energy consumption map, and help users more intuitively understand the energy consumption distribution of the entire road network.

[0182] To further refine the above solution, let 0.6× +0.4×

[0183] wherein, is the slope (%) and is the curvature (1 / km); this clarifies the specific weight distribution between the slope and the curvature.

[0184] In addition, a congestion index (0 - 1) can be introduced to dynamically adjust the weights :

[0185] 0.5× +0.3× +0.2×

[0186] The above-mentioned road segment stratification provides input features for the road segment-level model. The high-level road segments (such as 5-kilometer arterial roads) optimize the global path planning efficiency by weighted aggregation of low-level data. For establishing the road segment-level energy consumption model, not only the energy consumption data of individual road segments should be concerned, but also the hierarchical relationship and comprehensive influence between road segments should be considered. By this way of calculating energy consumption in a stratified manner, the energy consumption situation of different road segments can be reflected more comprehensively and accurately, providing richer information for the road segment-level energy consumption model and making it more in line with the actual road network energy consumption distribution.

[0187] As an example, when training the road segment-level energy consumption model, in addition to removing extreme values and classifying and storing data according to vehicle weight intervals as mentioned above, the data also needs to be cleaned and preprocessed, checking the integrity of the data and filling in missing values, which can be done by interpolation method or estimation based on similar data. Subsequently, a suitable model algorithm needs to be selected according to the data characteristics and problem requirements. In addition to the weighted trimmed mean algorithm for preliminary data processing, machine learning algorithms such as linear regression, decision tree, neural network, etc. can also be used, and finally the prepared data is used to train the model.

[0188] As an example, when integrating the model output results into the global energy consumption map, it should be noted that since the road segments in the geographical road network are divided into low-level road segments and high-level road segments, when integrating the model output results, it is necessary to ensure the accurate correspondence of the energy consumption data of different-level road segments; and because the vehicle driving data and road condition information are dynamically changing, when integrating the model output results to construct the global energy consumption map, the data update frequency should be considered. Regularly collect new vehicle driving data, retrain the road segment-level energy consumption model, and update the global energy consumption map.

[0189] Further refining the above solution and finally presenting the model output results to users in a visual way is an important part of constructing a global energy consumption map. At this time, a suitable map drawing tool should be selected, such as the Amap API, Baidu Map API, etc., to visually display the energy consumption information of road sections on the map. Different colors can be used to represent different energy consumption levels, with green indicating low-energy consumption road sections and red indicating high-energy consumption road sections. Provide user interaction functions. Users can query the energy consumption of different routes according to their own needs, such as the current vehicle weight and driving destination of the vehicle. The map can plan low-energy consumption routes in real time according to the user input and display them on the map.

[0190] As an example, given the limited number of current new energy heavy trucks and the barriers to data sharing among manufacturers, vehicle driving data often cannot cover all road sections that heavy trucks can pass through (road width, upper height limit, weight limit, and slope will affect whether heavy trucks can pass). If these uncovered roads are not considered for route planning, it will be difficult to ensure that the planned energy consumption route is optimal. To solve this problem, we put forward an idea: using the energy consumption attributes of the road sections covered by the existing driving data to estimate the energy consumption characteristics of the uncovered road sections that heavy trucks can pass through. This method not only makes up for the problem of insufficient data but also improves the accuracy of route planning and the effect of energy consumption optimization, thus more effectively supporting the development of green transportation and sustainable transportation. The specific contents are as follows:

[0191] Mine features closely related to energy consumption from the existing data. In addition to road type, slope, curvature, and road width, it includes analyzing traffic environment characteristics and geospatial and environmental factors. Traffic environment characteristics such as the traffic flow change law of road sections and the distribution of surrounding traffic facilities; geospatial and environmental factors such as road sections in the same mountain range may have similar slope and curvature characteristics; road sections in the same urban area may have similar traffic flow patterns and road infrastructure characteristics, ensuring that various factors that may affect energy consumption are covered;

[0192] Then standardize and preprocess the above-obtained data, and establish a feature correlation coefficient. Given the complexity and non-linearity of the energy consumption prediction problem, use multiple machine learning models and compare their performance, such as the random forest algorithm, neural network models (such as multi-layer perceptrons), etc. Use the existing driving data (including various characteristics of the road sections with driving data and the corresponding energy consumption data) to train the selected model. Divide the data into a training set, a validation set, and a test set, for example, according to the ratio of 70%, 20%, and 10%. During the training process, use the training set data to learn the parameters of the model so that the model can predict the energy consumption value according to the input features, and finally obtain the estimated energy consumption data of new energy heavy trucks on the uncovered road sections.

[0193] It should be noted that a small amount of actual driving data for un-covered sections can also be collected for fine-tuning the above-mentioned model to optimize the results of energy consumption prediction again.

[0194] In addition, the technical solution of the present invention can be further refined. For example, during the driving of the vehicle, the system will record the actual energy consumption values in real time and upload this data to the cloud. After the cloud platform receives the new driving data from different vehicles, it will regularly update and optimize the energy consumption values of each section. This process not only includes the cleaning of new trajectory data, but also involves the fusion and matching with the existing road network information to ensure the accuracy and consistency of the data. For details, please refer to the foregoing text.

[0195] For the energy consumption update method for a certain vehicle weight range on a specific section, the energy consumption model of this section is adjusted and optimized by continuously adding the latest driving data.

[0196]

[0197] Specifically, represents the latest energy consumption value of the corresponding vehicle weight range of the updated section; refers to the energy consumption value within the same vehicle weight range calculated according to the newly added vehicle data; and represents the energy consumption value of the corresponding vehicle weight range originally recorded for the section. Where λ is a threshold value, which can be changed according to road changes and data sample conditions.

[0198] By comparing and integrating and , we can obtain a more accurate , and improve the accuracy through self-update. With the accumulation of more driving data, the energy consumption model can gradually exclude outliers, making the prediction results closer to the actual situation. And external factors such as adaptation changes, traffic conditions, and section conditions may change over time. The self-update mechanism can help the model reflect these changes in a timely manner and maintain its timeliness.

[0199] Corresponding to the above embodiments, the present invention also proposes an electronic device.

[0200] As Figure 5 shown is a schematic structural diagram of an electronic device in the present invention. The electronic device 200 includes: a processor 201 and a memory 203. Among them, the processor 201 and the memory 203 are connected, such as through a bus 202. Optionally, the electronic device 200 may further include a transceiver 204. It should be noted that in actual applications, the transceiver 204 is not limited to one, and the structure of the electronic device 200 does not constitute a limitation to the embodiments of the present invention.

[0201] The processor 201 may be a CPU (Central Processing Unit), a general-purpose processor, a DSP (Digital Signal Processor), an ASIC (Application Specific Integrated Circuit), an FPGA (Field Programmable Gate Array), or other programmable logic devices, transistor logic devices, hardware components, or any combination thereof. It can implement or execute various exemplary logic blocks, modules, and circuits described in connection with the disclosure of the present invention. The processor 201 may also be a combination that implements computing functions, such as a combination of one or more microprocessors, a combination of a DSP and a microprocessor, etc.

[0202] The bus 202 may include a path for transmitting information between the above components. The bus 202 may be a PCI (Peripheral Component Interconnect) bus, an EISA (Extended Industry Standard Architecture) bus, or the like. The bus 202 may be divided into an address bus, a data bus, a control bus, etc. For the sake of simplicity of representation, Figure 5 only a thick line is shown herein, but it does not mean that there is only one bus or one type of bus.

[0203] The memory 203 is used to store a computer program corresponding to the new energy heavy truck energy consumption map construction method of the foregoing embodiments of the present invention, and the computer program is controlled and executed by the processor 201. The processor 201 is used to execute the computer program stored in the memory 203 to implement the content shown in the foregoing method embodiments.

[0204] Among them, the electronic device 200 includes, but is not limited to: mobile terminals such as laptop computers, PADs (tablet computers), etc., and fixed terminals such as desktop computers, etc. Figure 5 The illustrated electronic device 200 is only an example and should not impose any limitation on the functions and usage scope of the embodiments of the present invention.

[0205] The electronic device 200 according to the embodiments of the present invention deeply combines geographical information with vehicle energy consumption data, constructs a global energy consumption map based on the energy consumption conditions of different road sections, can intuitively compare the energy consumption differences of different routes, accurately select the driving route with the lowest energy consumption, and accurate low-energy route planning can not only reduce the vehicle's own energy consumption, but also guide the vehicles to be reasonably distributed in the road network. When many drivers can select appropriate routes based on the energy consumption map, it can effectively avoid the over-concentration of vehicles on some high-energy consumption or congested road sections, make more reasonable use of road resources, relieve traffic congestion, and improve the utilization efficiency of the entire road network;

[0206] Considering the significant change in the vehicle weight of new energy heavy trucks, through clustering analysis of driving data according to vehicle weight, the energy consumption characteristics of vehicles under different vehicle weights can be more accurately reflected. The energy consumption model established based on this and the constructed energy consumption map can provide drivers with route planning suggestions that are more in line with the actual operation of the vehicle. Drivers can select a route with lower energy consumption according to the current vehicle weight of the vehicle, reduce energy consumption, and thus increase the vehicle's cruising range. For new energy heavy trucks that highly rely on energy replenishment facilities, this can reduce the dependence on energy replenishment facilities and reduce the operation risks caused by inconvenient energy replenishment.

[0207] It should be noted that the logic and / or steps represented in the flowchart or described in other ways herein, for example, can be considered as a definite sequence list of executable instructions for implementing logical functions, and can be specifically implemented in any computer-readable medium for use by an instruction execution system, apparatus, or device (such as a computer-based system, a system including a processor, or other systems that can fetch instructions from the instruction execution system, apparatus, or device and execute the instructions), or in combination with these instruction execution systems, apparatus, or devices. For the purposes of this specification, a "computer-readable medium" can be any device that can contain, store, new energy, propagate, or transmit a program for use by an instruction execution system, apparatus, or device or in combination with these instruction execution systems, apparatus, or devices. More specific examples (non-exhaustive list) of computer-readable media include the following: an electrical connection portion (electronic device) having one or more wirings, a portable computer diskette (magnetic device), a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber device, and a portable compact disc read-only memory (CDROM). Additionally, a computer-readable medium can even be paper or other suitable media on which a program can be printed, because the program can be obtained electronically, for example, by optically scanning the paper or other media, then editing, interpreting, or processing it in other suitable ways if necessary, and then storing it in a computer memory.

[0208] It should be understood that various parts of the present invention can be implemented by hardware, software, firmware or a combination thereof. In the above embodiments, multiple steps or methods can be implemented by software or firmware stored in a memory and executed by a suitable instruction execution system. For example, if implemented by hardware, as in another embodiment, any one of the following techniques known in the art or a combination thereof can be used: discrete logic circuits having logic gate circuits for implementing logic functions on data signals, application specific integrated circuits having appropriate combinational logic gate circuits, programmable gate arrays (PGAs), field programmable gate arrays (FPGAs), etc.

[0209] In the description of this specification, the description with reference to terms such as "one embodiment", "some embodiments", "example", "specific example", or "some examples" means that the specific features, structures, materials, or characteristics described in connection with the embodiment or example are included in at least one embodiment or example of the present invention. In this specification, the schematic representations of the above terms do not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials, or characteristics described can be combined in a suitable manner in any one or more embodiments or examples.

[0210] In addition, the terms "first" and "second" are used only for descriptive purposes and cannot be construed as indicating or implying relative importance or implicitly specifying the quantity of the indicated technical features. Thus, the features defined with "first" and "second" may explicitly or implicitly include at least one of the features. In the description of the present invention, the meaning of "a plurality" is at least two, such as two, three, etc., unless otherwise specifically defined.

[0211] Although the embodiments of the present invention have been shown and described above, it can be understood that the above embodiments are exemplary and should not be construed as limiting the present invention. Those of ordinary skill in the art can make changes, modifications, substitutions, and variations to the above embodiments within the scope of the present invention.

Claims

1. A method for constructing an energy consumption map of a new energy heavy truck, characterized in that: The method comprises: Clean invalid data in vehicle driving data to obtain valid data; Clustering the above valid data according to the vehicle weight to form a data structure of {vehicle weight interval:{driving data segment set}}, wherein each vehicle weight interval corresponds to a set of clustered driving data segments; According to the above-obtained data structure, the data is associated with the geographical road network, the vehicle energy consumption information is matched with the road section where the vehicle actually travels, and a "road section-energy consumption" mapping relationship is established; Based on the above matched data, a road section-level energy consumption model is established and trained, and the output results of the model are integrated into a global energy consumption map; The invalid data in the cleaning vehicle driving data includes the following: For any vehicle driving trajectory point, if the following conditions are met, it is invalid driving data: ① Data that is marked as invalid; ② Abnormal position; ③ Abnormal speed; ④Abnormal current or voltage; The clearing condition of the track point is as follows: In the above formula, It is the data valid flag. is the lower left longitude of China’s geographical area, is the upper right longitude of China’s geographical area, is the lower left latitude of China’s geographical area, is the upper right latitude of China’s geographical area, is the theoretical maximum speed, is the maximum current threshold, is the minimum current threshold, is the maximum voltage threshold, is the minimum voltage threshold; The matching of the vehicle energy consumption information with the road section where the vehicle actually travels includes the following contents: Obtain matching road sections and select different acquisition methods according to different situations of vehicle trajectory points; If the last track point of the vehicle successfully matches a road segment in the geographic road network, and the distance between the current track point and the matching point corresponding to the last successful match is less than the vertical distance between the track point and the road segment, the matching road segment is obtained by using the road network topology method, which is: the matching road segment is obtained by forming a road network topology road segment set through the primary topology road segment directly connected to the current matching point and the secondary topology road segment connected to the primary topology road segment; If the vehicle's last trajectory point fails to match a certain road section in the geographic road network, a geographic rectangular box is constructed based on the current trajectory point, and a matching road section is obtained by a geographic box selection method. The geographic box selection method is: by dynamically adjusting the side length of the geographic rectangular box and the fan-shaped search area to screen the candidate road sections, and then obtain the matching road section.

2. The method for constructing a new energy heavy truck energy consumption map according to claim 1, characterized in that: The clustering of the valid data according to the vehicle weight includes the following contents: Firstly, the vehicle's driving trajectory is divided into several driving data segments according to the actual vehicle weight during driving, and each driving data segment represents the continuous driving process of the vehicle under a certain driving weight; Then, for each defined driving vehicle weight interval, all driving data segments belonging to the interval are collected to form a data structure of {vehicle weight interval: {driving data segment set}}.

3. The method for constructing an energy consumption map of a new energy heavy truck according to claim 1, characterized in that: The method of obtaining the matching road segments by using the road network topology is expressed as follows: In the above formula, is the road network topology segment set, A primary topological section is a section directly connected to the current matching point; It is a secondary topological section, which is a peripheral section connected to the primary topological section.

4. The method for constructing a new energy heavy truck energy consumption map according to claim 1, characterized in that: When the geographic selection method is used to obtain the matching road section, the selection area should be dynamically adjusted according to the vehicle movement state, including the following contents: First, dynamically calculate the length of the frame selection: Based on the current vehicle speed v and direction change rate at the trajectory point Adjust the selection side length L. The selection side length L is expressed as follows: In the above formula, is the base side length, and is the adjustment coefficient, is the maximum direction change angle; Then constrain the fan-shaped area in the direction of the box selection: Defines the angle range of the fan-shaped search area , only matching candidate road segments with the same direction.

5. The method for constructing an energy consumption map of a new energy heavy truck according to claim 1, characterized in that: The matching sections obtained by the road network topology method and the geographic frame selection method are selected as candidates, and the best matching section is found from the candidate section set. Includes the following: set up is the angular distance between the trajectory point and the road segment, set is the vertical distance between the trajectory point and the road segment; It is expressed as: In the above formula, is the angle influence coefficient threshold, is the angle between the trajectory point direction and the road section direction; for and , first normalize the two, and the normalized vertical distance is expressed as: In the above formula, is the maximum value between the trajectory point and the road segment, is the minimum value between the trajectory point and the road segment; The normalized angular distance is expressed as: For each road segment, its comprehensive score The calculation formula is: In the above formula, is the weight of the directional factor, is the weight of the distance factor, and + =1,0≤ ≤1,0≤ ≤1; By comparing the comprehensive scores of adjacent sections of the trajectory point, the optimal matching section is obtained.

6. The method for constructing a new energy heavy truck energy consumption map according to claim 1, characterized in that: When establishing the road section level energy consumption model, a weighted truncated mean algorithm is used to remove extreme values ​​in the energy consumption data, and an interval index storage structure is used to store the energy consumption data according to the vehicle weight interval classification, specifically including: For a certain vehicle weight interval on a road section, there are n observed energy consumption values ; First, sort these energy consumption values, set k = 0.05n, remove the highest k values ​​and the lowest k values, and then calculate the weighted average of the remaining data as the road section energy consumption of the vehicle weight range. The formula is expressed as: In the above formula, is the weight of each energy consumption value, and p is the index in the sorted energy consumption value sequence.

7. The method for constructing an energy consumption map of a new energy heavy truck according to claim 1, characterized in that: When the above integration is performed into a global energy consumption map, the road sections in the geographical road network are layered, including low-level road sections and high-level road sections, wherein the relationship between high-level road sections and low-level road sections is a one-to-many relationship, and the energy consumption corresponding to the high-level road sections is expressed as: In the above formula, is the average energy consumption per kilometer of the ith road section in the low-level road section sequence, is the weighted value of the road segment attribute, is the length of the i-th road segment in the low-level road segment sequence, is the number of high-level sections corresponding to low-level sections, is the length of the high-rise section, is the weighting function.

8. An electronic device, characterized in that: It comprises a memory, a processor and a computer program stored in the memory. When the computer program is executed by the processor, it implements the method for constructing an energy consumption map of a new energy heavy truck as described in any one of claims 1 to 7.

Images