Method for constructing typical road profile working conditions based on genetic algorithm combined with new energy platform data

Abstract

The application relates to the technical field of quantifiable test indexes of new energy vehicles, and particularly discloses a method for constructing a typical road spectrum working condition based on a genetic algorithm and new energy platform data, data acquisition, invalid data section elimination rules including static data section elimination and abnormal continuous duration elimination; data abnormal value identification rules: speed abnormal values are defined and a correction method is set; data smoothing and denoising rules: a median filter is used to preliminarily denoise a speed sequence; characteristic parameter calculation: including kinematic characteristic parameters and new energy light truck characteristic parameters; construction of an initial section library; multi-objective genetic algorithm optimization; road spectrum synthesis and post-processing: optimal sections filtered by the algorithm are spliced, and smoothing processing is carried out to ensure the continuity of a speed curve; multi-objective genetic algorithm selection: selecting an NSGA-III algorithm as a multi-genetic objective algorithm condition; the application realizes rapid construction and iteration of the road spectrum, and is high in efficiency.

Description

Technical Field

[0001] This invention belongs to the technical field of quantifiable test indicators for new energy vehicles, specifically involving a method for constructing typical road spectrum operating conditions based on genetic algorithms combined with new energy platform data. Background Technology

[0002] In recent years, my country's urban express logistics industry has developed rapidly, and the demand for short-distance delivery capacity within cities and intercity areas has continued to increase. This presents a huge market opportunity for new energy electric light trucks, which feature zero emissions and suitability for short- and medium-distance delivery. However, due to limitations in technologies such as power batteries, driving range has become a bottleneck restricting the development of electric vehicles. The accuracy of driving range calculations based on typical operating conditions is crucial for the promotion and development of electric vehicles, as well as for providing a good user experience and building trust among new energy vehicle users. Currently, commonly used road spectrum construction methods, such as those based on real-vehicle road condition data collected from a small number of actual vehicles, suffer from drawbacks such as small data volume, high cost, and insufficient representativeness. Methods based on standard operating conditions (such as NEDC, WLTC, CLTC, etc.) are limited by their simplistic approach and difficulty in reflecting the true operating characteristics of specific regions and vehicle models. The operating characteristics of new energy vehicles (especially light trucks) differ significantly from those of traditional fuel vehicles. Their energy flow, regenerative braking, torque response, and accessory power consumption—all crucial economic decomposition points—are highly sensitive to the transient characteristics of road profiles. Traditional road profile construction methods based on fuel vehicles cannot accurately be used for energy consumption assessment, range prediction, powertrain matching optimization, and durability testing of new energy light trucks. To more accurately reflect the typical road profile characteristics of different regions and vehicle models, this paper leverages operational data from OEM big data platforms and vehicle networking technology. Overcoming the shortcomings of existing road profile fitting techniques, it provides a highly efficient and representative method specifically for constructing typical road profile operating conditions for new energy light trucks. Summary of the Invention

[0003] The purpose of this invention is to provide a method for constructing typical road spectrum operating conditions based on genetic algorithms and new energy platform data, so as to achieve rapid construction and iteration of road spectrum.

[0004] To address the aforementioned problems in the existing technology, the technical solution adopted in this invention is: a method for constructing typical road spectrum operating conditions based on genetic algorithms combined with new energy platform data, comprising the following steps: S1. Data Collection: Collect operational data from the vehicle networking platform as the data source; collect fixed route information of the fleet and speed, acceleration, and time information of each operating vehicle over n months. S2. Invalid data segment removal rules: including removal of static data segments and removal based on abnormal duration; S3. Data outlier identification rules: Define speed outliers and set correction methods; S4. Data smoothing and denoising rules: Use a median filter to perform preliminary denoising on the velocity sequence, and use a Savitzky-Golay filter to smooth the denoised velocity data. S5. Characteristic parameter calculation: including kinematic characteristic parameters and new energy light truck characteristic parameters; S6. Construct the initial fragment library; S7. Multi-objective genetic algorithm optimization; S8. Road spectrum synthesis and post-processing: The optimal segments selected by the algorithm through filtering are spliced ​​together and smoothed to ensure the continuity of the speed curve. S9. Selection of multi-objective genetic algorithm: NSGA-Ⅲ algorithm is selected as the condition for multi-objective genetic algorithm.

[0005] Preferably, the data saved in step S1 is of the type of road map, document, and table processing.

[0006] Preferably, in step S2, the static data segment removal is as follows: Static data segment removal: segments with continuous speed values ​​≤ S1km / h and duration > 180 seconds are considered as long-term parking segments and are removed, while short red light parking times ≤ 180s are retained; Abnormal duration removal: Records with a total trip duration of <60s or >10 hours are considered invalid trips and are removed entirely.

[0007] Preferably, in step S3, the velocity anomaly value is defined as follows: a physical upper limit of velocity Vmax is defined, and any data point where Vr > Vmax is considered an anomaly; Correction method: If the data before and after the Vr anomaly point is valid, then use linear interpolation of the data before and after it to replace it; if the anomaly is continuous, then mark the segment as invalid.

[0008] Preferably, the kinematic characteristic parameters in step S5 include average speed, average driving speed, idling time ratio, average acceleration, average deceleration, acceleration standard deviation, speed standard deviation, maximum speed, and acceleration ratio distribution.

[0009] Preferably, the characteristic parameters of the new energy light truck in step S5 include driving energy intensity, recovered energy intensity, energy recovery rate, proportion of motor operating in high-efficiency zone, average battery discharge power, and SOC change.

[0010] Preferably, in step S6, the initial segment library is constructed by dividing the driving segments from the time the light truck stops charging to the next round of charging, using time as the arrangement information, and heuristically generating the detection sorting string and the detection time string based on the process information to construct the initial population P(t).

[0011] Preferably, the multi-objective genetic algorithm optimization in step S7 uses the NSGA-Ⅲ algorithm, with fragment combinations as chromosomes, minimizing feature error as the objective, and introducing the driving time before and after charging as a reference point, iterating until convergence or reaching the maximum number of iterations.

[0012] The beneficial effects of this invention are as follows: This invention presents a method for constructing typical road spectrum operating conditions based on genetic algorithms combined with new energy platform data. By using the genetic firefly algorithm, a typical road spectrum construction tool is constructed that efficiently extracts representative segments from massive amounts of data, quickly generates and iterates with customer operating characteristics, and achieves rapid construction and iteration of road spectra with high efficiency. Attached Figure Description

[0013] Figure 1 The overall flowchart is shown for a method to construct typical road spectrum operating conditions based on genetic algorithms and new energy platform data.

[0014] Figure 2 This is a schematic diagram of the optimization process of a multi-objective genetic algorithm.

[0015] Figure 3 This is a comparison chart of the overall characteristic parameters of the generated typical road spectrum conditions and the original data.

[0016] Figure 4 This is the final typical road spectrum velocity-time curve after synthesis. Detailed Implementation

[0017] The present invention will be further described below with reference to the accompanying drawings and reference numerals.

[0018] To better understand the above-mentioned objectives, features, and advantages of the present invention, the present invention will be described in detail below with reference to the accompanying drawings and specific embodiments. It should be noted that, unless otherwise specified, the embodiments and features described in these embodiments can be combined with each other.

[0019] The terms “first,” “second,” “third,” etc., are used only to distinguish descriptions and should not be interpreted as indicating or implying relative importance.

[0020] In the description of this invention, it should also be noted that, unless otherwise explicitly specified and limited, the terms "set," "install," "connect," and "link" should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral connection; they can refer to a mechanical connection or an electrical connection; they can refer to a direct connection or an indirect connection through an intermediate medium; and they can refer to the internal connection of two components. Those skilled in the art can understand the specific meaning of the above terms in this invention based on the specific circumstances.

[0021] The specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings. It should be understood that the specific embodiments described herein are for illustration and explanation only and are not intended to limit the present invention.

[0022] like Figure 1 As shown, a method for constructing typical road spectrum operating conditions based on genetic algorithms combined with new energy platform data includes the following steps: S1: Data Acquisition The system collects operational data from the vehicle networking platform as a data source, as well as fixed route information for the fleet and speed, acceleration, and time information of each operating vehicle over n months. Data storage types include route information, documents, and spreadsheets.

[0023] S2: Invalid data segment removal rules: 1. Removal of stationary data segments: Segments with continuous speed values ​​≤ S1km / h and duration > 180 seconds are considered long-term stops and are removed (such as loading / unloading, charging). Short red light stops (≤ 180s) are retained. 2. Removal of abnormal duration: Records with a total trip duration of <60s or >10 hours are considered invalid trips and are removed entirely.

[0024] S3: Outlier Identification Rules: 1. Speed ​​outliers: Define a physical speed limit Vmax (e.g., the maximum speed of a light truck is 120km / h). Any data point where Vr > Vmax is considered an outlier. 2. Correction method: If the data before and after the Vr anomaly point is valid, then use linear interpolation of the data before and after it to replace it; if the anomaly is continuous, then mark the segment as invalid.

[0025] S4: Data Smoothing and Denoising Rules: 1. Use a median filter (5-7 data points) to perform preliminary denoising on the velocity sequence to eliminate small fluctuations and spikes; 2. The Savitzky-Golay filter (polynomial order 2, 11 data points) is used to smooth the denoised velocity data. This filter can better preserve the local distribution characteristics and is superior to the traditional average filter.

[0026] S5: Feature parameter calculation formula: 1. Kinematic characteristic parameters: 2. Specific characteristic parameters of new energy light trucks: S6: Constructing the initial segment library: Based on the light truck vehicle's parking and charging to the next parking and charging cycle, driving segments are divided. Time is used as the arrangement information. Based on the process information, detection sorting strings and detection time strings are heuristically generated to construct the initial population P(t).

[0027] S7: Multi-objective genetic algorithm optimization: 1. Chromosome coding: How to use gene sequences to represent a candidate road profile consisting of multiple operating condition segments.

[0028] 2. Fitness function design: Detail how to quantify and combine multiple objectives (such as velocity error, acceleration distribution error, SOC consumption error, etc.).

[0029] 3. Genetic operations: specific strategies for selection, crossover, and mutation (such as roulette wheel selection, single-point crossover, and basic position mutation).

[0030] 4. Algorithm termination conditions: such as maximum number of iterations or convergence of fitness.

[0031] S8: Road spectrum synthesis and post-processing: The optimal segments selected by the algorithm through filtering are spliced ​​together and smoothed to ensure the continuity of the speed curve.

[0032] S9: Selection of Multi-Objective Genetic Algorithm: The NSGA-III algorithm is selected as the condition for the multi-objective genetic algorithm, such as... Figure 2 As shown, a reference point is introduced to uniformly distribute the total driving time of light trucks before and after charging. The road spectrum conditions can guide the selection of the population (with longer driving time as the ideal reference point and intercept point). In the selection process, the NSGA-III algorithm not only considers the non-dominated level of the individual, but also the reference point correlation degree of the individual, that is, the distance between the individual and the reference point and the crowding degree of the reference point. This can better balance the diversity and convergence of the population, and at the same time avoid over-selection or ignoring certain targets.

[0033] Example 1: S1: Using the operational data of 100 electric light trucks from a logistics fleet operation platform as the data source, collect the fixed route map information of the fleet and the speed, acceleration and time information of each operating vehicle for 6 months; data storage types: route map, document, and table processing.

[0034] S2: Static data segment elimination: Segments with continuous speed values ​​≤2km / h and duration >180 seconds are considered long-term parking segments and are eliminated (such as loading / unloading, charging), while short red light parking segments (≤180s) are retained.

[0035] S3: Abnormal Duration Removal: Records with a total trip duration of <60s or >10 hours are considered invalid trips and are removed entirely.

[0036] S4: Ten feature parameters were selected, including average speed, idle speed ratio, and acceleration standard deviation. The NSGA-III algorithm was selected as the condition for the multi-objective genetic algorithm. The total driving time of the light truck before and after charging was introduced as a reference point for uniform distribution. The population size of the genetic algorithm was set to 200, and 500 iterations were performed to finally generate a typical road spectrum with a length of 3600 seconds. Figures 3-4 As shown, the average error between the various characteristic parameters of this road spectrum and the overall data is less than 5%.

[0037] Example 2: S1: Testing and analysis were conducted on a semi-physical test bench for new energy motors to generate road profiles for vehicle electric efficiency performance tests. Deviation analysis of electric efficiency between the generated road profiles and big data sets was performed, and the measured driving range deviated from the user's actual average driving range by less than 20%, proving the effectiveness of this method.

[0038] S2: Road spectrum representativeness verification: The generated road spectrum was compared with the data characteristics of 2,000 vehicles operating in the plain area over 2 years, and the data fit was >80%.

[0039] This invention is not limited to the above-described optional embodiments. Anyone can derive other various forms of products under the guidance of this invention. However, regardless of any changes made in their shape or structure, any technical solution that falls within the scope of the claims of this invention shall be protected by this invention.

Claims

1. A method for constructing typical road spectrum operating conditions based on genetic algorithms combined with new energy platform data, characterized in that, Includes the following steps: S1. Data Collection: Collect operational data from the vehicle networking platform as the data source; collect fixed route information of the fleet and speed, acceleration, and time information of each operating vehicle over n months. S2. Invalid data segment removal rules: including removal of static data segments and removal based on abnormal duration; S3. Data outlier identification rules: Define speed outliers and set correction methods; S4. Data smoothing and denoising rules: Use a median filter to perform preliminary denoising on the velocity sequence, and use a Savitzky-Golay filter to smooth the denoised velocity data. S5. Characteristic parameter calculation: including kinematic characteristic parameters and new energy light truck characteristic parameters; S6. Construct the initial fragment library; S7. Multi-objective genetic algorithm optimization; S8. Road spectrum synthesis and post-processing: The optimal segments selected by the algorithm through filtering are spliced ​​together and smoothed to ensure the continuity of the speed curve. S9. Selection of multi-objective genetic algorithm: NSGA-Ⅲ algorithm is selected as the condition for multi-objective genetic algorithm.

2. The method for constructing typical road spectrum operating conditions based on genetic algorithm combined with new energy platform data according to claim 1, characterized in that, The data saved in step S1 is of the following types: road map, document, and table processing.

3. The method for constructing typical road spectrum operating conditions based on genetic algorithm combined with new energy platform data according to claim 1, characterized in that, In step S2, static data segment removal: Static data segment removal: segments with continuous speed values ​​≤ S1km / h and duration > 180 seconds are considered long-term parking segments and are removed, while short red light parking times ≤ 180s are retained; Abnormal duration removal: Records with a total trip duration of <60s or >10 hours are considered invalid trips and are removed entirely.

4. The method for constructing typical road spectrum operating conditions based on genetic algorithm combined with new energy platform data according to claim 1, characterized in that, In step S3, the velocity anomaly is defined as follows: a physical upper limit for velocity Vmax is defined, and any data point where Vr > Vmax is considered an anomaly. Correction method: If the data before and after the Vr anomaly point is valid, then use linear interpolation of the data before and after it to replace it; if the anomaly is continuous, then mark the segment as invalid.

5. The method for constructing typical road spectrum operating conditions based on genetic algorithm combined with new energy platform data according to claim 1, characterized in that, The kinematic characteristic parameters in step S5 include average speed, average driving speed, idling time ratio, average acceleration, average deceleration, acceleration standard deviation, speed standard deviation, maximum speed, and acceleration ratio distribution.

6. The method for constructing typical road spectrum operating conditions based on genetic algorithm combined with new energy platform data according to claim 5, characterized in that, The characteristic parameters of the new energy light truck in step S5 include driving energy intensity, recovered energy intensity, energy recovery rate, proportion of motor operating in high-efficiency zone, average battery discharge power, and SOC change.

7. The method for constructing typical road spectrum operating conditions based on genetic algorithm combined with new energy platform data according to claim 1, characterized in that, In step S6, the initial segment library is constructed by dividing the driving segments from the time a light truck stops charging to the next round of charging, using time as the arrangement information, and heuristically generating detection sorting strings and detection time strings based on process information to construct the initial population P(t).

8. The method for constructing typical road spectrum operating conditions based on genetic algorithm combined with new energy platform data according to claim 1, characterized in that, In step S7, the multi-objective genetic algorithm optimization is performed using the NSGA-Ⅲ algorithm. The fragment combination is used as chromosomes, the feature error is minimized, the driving time before and after charging is used as a reference point, and the iteration is performed until convergence or the maximum number of iterations is reached.

Images