Hybrid train energy management method

Abstract

The application provides a kind of hybrid train energy management method, comprising: running endpoint positioning, fixed route modeling, SOC data acquisition, round trip SOC variation calculation, multi-cycle SOC average calculation, equivalent factor λ dynamic adjustment based on SOC change, obtain the mileage length of current running route and the rated capacity of vehicle power battery, dynamically set SOC change threshold ΔSOC according to mileage length and rated capacity b ;When ΔSOC b , trigger equivalent factor λ adjustment;The application is based on historical path identification and SOC quantitative analysis, realizes the precise dynamic adjustment of equivalent factor, solves the SOC imbalance problem under the fixed route round trip working condition, significantly improves fuel economy;Adopt quantitative adjustment formula and experience coefficient calibration, taking into account the adaptability of different SOC change scenarios, ensure that electric energy is fully utilized while not sacrificing power performance.

Description

Technical Field

[0001] This invention relates to a hybrid power train energy management method, and pertains to the field of vehicle control technology. Background Technology

[0002] Energy management strategies for hybrid electric vehicles (HEVs / PHEVs) directly impact fuel economy and power performance. Among these, the Equivalent Energy Minimization Management (ECMS) strategy, due to its strong real-time performance and simple control logic, has become one of the mainstream control strategies. The core of ECMS is to equate electrical energy consumption to fuel consumption through an equivalence factor λ. The value of this equivalence factor directly determines the power distribution ratio between the engine and the electric motor. A smaller λ value indicates that the system prioritizes electric drive, improving fuel economy; a larger λ value indicates that the system focuses more on engine drive, ensuring power performance. However, existing ECMS technologies often use fixed values ​​for the equivalence factor or segmented fixed values ​​based on a single operating condition (such as urban congestion or highway cruising), lacking dynamic adaptability to actual vehicle driving conditions. In practical applications, vehicle trains often face repetitive operating conditions such as switching between multiple fixed routes, and may also encounter new operating conditions that temporarily deviate from the fixed routes. Existing technologies have the following problems: For trucks operating on fixed routes, the equivalent factor is not adjusted based on historical travel paths and their corresponding SOC (State of Charge) changes. This leads to an imbalance where the SOC continuously rises or falls after multiple trips on the same route. If the SOC continues to rise, it indicates that the electric energy is not being fully utilized, and fuel economy is limited. If the SOC continues to fall, it may be due to insufficient battery power, affecting the overall power performance of the vehicle.

[0003] The adjustment of the equivalent factor λ lacks quantitative basis and relies solely on empirical values ​​or real-time vehicle parameters, such as the real-time value of SOC, to make judgments. It is impossible to accurately match the adjustment strategy according to the overall change of SOC throughout the entire operating route, making it difficult to continuously balance fuel economy and power performance under the complex and ever-changing operating conditions of truck trains.

[0004] When the vehicle's operating route changes, the original equivalent factor value cannot adapt to the new operating conditions, which can easily lead to the motor's drive participation ratio not conforming to the operating conditions, resulting in insufficient power or poor economy. Summary of the Invention

[0005] To address the aforementioned practical problems, this invention provides an energy management method for hybrid trains. The specific technical solution is as follows: A hybrid power train energy management method includes: S1, driving endpoint positioning, by collecting GPS data in real time during vehicle driving through the vehicle GPS module, the GPS data includes at least GPS coordinates and driving heading angle; when the included angle between two consecutively collected driving heading angles is greater than or equal to a preset heading angle change threshold, and the vehicle is in a start-stop condition within the time period corresponding to the two driving heading angle collection points, the location of the latter collection point in the consecutive collection points is marked as the driving endpoint. GPS coordinates include longitude and latitude coordinates; S2, fixed route modeling: obtain the GPS trajectory of multiple running segments based on the GPS coordinates in N continuously collected GPS data segments, construct a sparse matrix representing the spatial distribution of the trajectory of each running segment according to the preset grid size, and use an adaptive resonance theory neural network model to cluster the sparse matrix corresponding to each running segment according to the preset similarity threshold to identify similar paths as historical fixed routes. S3, SOC data acquisition, collects the SOC value of the vehicle in real time through the battery management system (BMS) during the driving process, and records the SOC status when reaching each driving endpoint and destination each time a fixed historical route is traveled. S4, Calculation of SOC change for round trip: Based on the SOC at the end of a single round trip and the SOC at the starting point, the change in SOC for a single round trip, ΔSOC, is calculated. S5, Multi-period SOC mean calculation: The mean value ΔSOC is calculated by averaging the change in SOC per round trip ΔSOC for multiple consecutive round trips on a historical fixed route. avg The mean ΔSOC avg As a quantitative basis for adjusting the equivalent factor λ; S6. Based on the SOC change, dynamically adjust the equivalent factor λ, obtain the current driving route mileage and the vehicle's power battery rated capacity, and dynamically set the SOC change threshold ΔSOC according to the mileage and rated capacity. b ; When ΔSOC is greater than ΔSOC b When this occurs, the equivalent factor λ is adjusted. Preferably, S6 specifically includes: ΔSOC is less than or equal to the stated ΔSOC b At this time, the equivalent factor λ adjustment is not triggered; ΔSOC is greater than the stated ΔSOC b When, the equivalent factor λ is adjusted to λ n1 : λ n1 =λ×(1-k1×(ΔSOC-ΔSOC b ) / 100); In the formula, k1 is the empirical coefficient for adjusting SOC increase; When ΔSOC continues to decrease, the equivalent factor λ is adjusted to λ. n2 : λ n2 =λ×(1+k2×(ΔSOC-ΔSOC b ) / 100); In the formula, k2 is the empirical coefficient for adjusting the SOC decrease.

[0006] Preferably, the constraint condition for adjusting the equivalent factor λ is: λ min ≤λ n1 ≤λ max ; λ min ≤λ n2 ≤λ max ; In the formula, λ min The minimum equivalent factor set at the factory; λ max The maximum equivalent factor set at the factory.

[0007] Preferably, it also includes route deviation identification and adaptive processing, wherein the route deviation identification is as follows: The system compares the current GPS coordinates with the GPS coordinate set of the historical fixed route in real time, calculates the shortest distance d between the current location and the historical fixed route, uses the shortest distance d as the coordinate offset, sets a distance threshold d0, and sets a deviation duration threshold t0. When the shortest distance d is continuously greater than d0 for a duration exceeding t0, it is determined to be a deviation from the historical fixed route. The adaptive processing is as follows: when it is determined that the route deviates from the historical fixed route, the equivalent factor λ is initialized to the original equivalent factor λ. init ; When a vehicle returns to a historical fixed route and the duration is ≥t0, the dynamic adjustment of the equivalent factor based on the change of SOC is restored, and the equivalent factor λ corresponding to the historical route is used as the initial value. To ensure the validity of the initial values ​​when the vehicle returns to the historical route, the GPS coordinate sets of multiple historical fixed routes and their corresponding equivalent factors λ are recorded.

[0008] The beneficial effects of this invention compared to the prior art are as follows: Based on historical route identification and SOC quantitative analysis, the equivalent factor can be accurately and dynamically adjusted, which solves the SOC imbalance problem under fixed route round-trip conditions and significantly improves fuel economy. The quantitative adjustment formula and empirical coefficient calibration are adopted to take into account the adaptability to different SOC change scenarios, ensuring that the power is fully utilized without sacrificing power performance. An adaptive route deviation handling mechanism has been added to avoid increased fuel consumption caused by the equivalent factor not adapting to new operating conditions, thereby improving the technology's adaptability to all operating conditions.

[0009] The control logic is simple and can be implemented based on existing vehicle hardware, without the need for additional sensors, making it low-cost and easy to industrialize and promote. Attached Figure Description

[0010] Figure 1 This is a flowchart of the present invention. Detailed Implementation

[0011] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0012] Example 1 like Figure 1 As shown, a hybrid train energy management method includes: S1, driving endpoint positioning, real-time collection of GPS data during vehicle driving through the vehicle-mounted GPS module, the GPS data includes at least GPS coordinates and driving heading angle; when the angle between two consecutively collected driving heading angles is greater than or equal to the preset heading angle change threshold, and the vehicle is in a start-stop condition within the time period corresponding to the two driving heading angle collection points, the location of the latter collection point in the consecutive collection points is marked as the driving endpoint. GPS coordinates include longitude and latitude coordinates; S2, fixed route modeling: obtain the GPS trajectory of multiple running segments based on the GPS coordinates in N continuously collected GPS data segments, construct a sparse matrix representing the spatial distribution of the trajectory of each running segment according to the preset grid size, and use an adaptive resonance theory neural network model to cluster the sparse matrix corresponding to each running segment according to the preset similarity threshold to identify similar paths as historical fixed routes. S3, SOC data acquisition, collects the SOC value of the vehicle in real time through the battery management system (BMS) during the driving process, and records the SOC status when reaching each driving endpoint and destination each time a fixed historical route is traveled. S4, Calculation of SOC change for round trip: Based on the SOC at the end of a single round trip and the SOC at the starting point, the change in SOC for a single round trip, ΔSOC, is calculated. S5, Multi-period SOC mean calculation: The mean value ΔSOC is calculated by averaging the change in SOC per round trip ΔSOC for multiple consecutive round trips on a historical fixed route. avg The mean ΔSOC avg As a quantitative basis for adjusting the equivalent factor λ; S6. Based on the SOC change, the equivalent factor λ is dynamically adjusted. The mileage of the current driving route and the rated capacity of the vehicle's power battery are obtained. The SOC change threshold ΔSOC is dynamically set according to the mileage and rated capacity. b ; When ΔSOC is greater than ΔSOC b When this occurs, the equivalent factor λ is adjusted. S6 specifically refers to: ΔSOC is less than or equal to ΔSOC b At this time, the equivalent factor λ adjustment is not triggered; ΔSOC is greater than ΔSOC b When, the equivalent factor λ is adjusted to λ n1 : λ n1 =λ×(1-k1×(ΔSOC-ΔSOC b ) / 100); In the formula, k1 is the empirical coefficient for adjusting SOC increase; When ΔSOC continues to decrease, the equivalent factor λ is adjusted to λ. n2 : λ n2 =λ×(1+k2×(ΔSOC-ΔSOC b ) / 100); In the formula, k2 is the empirical coefficient for adjusting the SOC decrease.

[0013] The constraint condition for adjusting the equivalent factor λ is: λ min ≤λ n1 ≤λ max ; λ min ≤λ n2 ≤λ max ; In the formula, λ min The minimum equivalent factor set at the factory; λ max The maximum equivalent factor set at the factory.

[0014] It also includes route deviation identification and adaptive processing. Route deviation identification is as follows: The system compares the current GPS coordinates with the GPS coordinate set of the historical fixed route in real time, calculates the shortest distance d between the current location and the historical fixed route, uses the shortest distance d as the coordinate offset, sets a distance threshold d0, and sets a deviation duration threshold t0. When the shortest distance d is continuously greater than d0 for a duration exceeding t0, it is determined to be a deviation from the historical fixed route. The adaptive processing is as follows: when it is determined that the route deviates from the historical fixed route, the equivalent factor λ is initialized to the original equivalent factor λ. init ; When a vehicle returns to a historical fixed route and the duration is ≥t0, the dynamic adjustment of the equivalent factor based on the change of SOC is restored, and the equivalent factor λ corresponding to the historical route is used as the initial value. To ensure the validity of the initial values ​​when the vehicle returns to the historical route, the GPS coordinate sets of multiple historical fixed routes and their corresponding equivalent factors λ are recorded.

[0015] Example 2 The vehicle's GPS data is collected in real time by the vehicle-mounted GPS module. The GPS data includes at least GPS coordinates and heading angle. When the angle between two consecutively collected heading angles is greater than or equal to a preset heading angle change threshold, and the vehicle is in a start-stop condition within the time period corresponding to the two heading angle collection points, the location of the latter collection point in the consecutive collection points is marked as the driving endpoint. Preferably, the heading angle change threshold is 170°. That is, when the included angle between two consecutively collected heading angles is ≥170°, and the vehicle is in a start-stop condition within the time period corresponding to the two heading angle collection points, the location of the latter collection point in the consecutive collection points is marked as the driving endpoint.

[0016] Multiple GPS tracks are used to construct a sparse matrix representing the spatial distribution of each running segment's trajectory according to a preset grid size. The sparse matrix corresponding to each running segment is then processed. Preferably, the grid size is 0.1 degrees, resulting in a 0-1 sparse matrix representing the spatial distribution of the trajectory of each running segment, with the sparse matrix having a size of 132×257; The mean ΔSOC is calculated by taking the average change in SOC for a single round trip on a historical fixed route over multiple consecutive round trips. avg ; ΔSOC is less than or equal to ΔSOC b At this time, the equivalent factor λ adjustment is not triggered; ΔSOC is greater than ΔSOC b When, the equivalent factor λ is adjusted to λ n1 : λ n1 =λ×(1-k1×(ΔSOC-ΔSOC b) / 100); In the formula, k1 is the empirical coefficient for adjusting SOC increase; When ΔSOC continues to decrease, the equivalent factor λ is adjusted to λ. n2 : λ n2 =λ×(1+k2×(ΔSOC-ΔSOC b ) / 100); In the formula, k2 is the empirical coefficient for adjusting the SOC decrease.

[0017] Optionally, let ΔSOC b 5%, When ΔSOC is less than or equal to 5%, it indicates that the current equivalent factor value has been adapted to the working condition, and SOC is in a dynamic equilibrium state, without triggering the adjustment of the equivalent factor λ. ΔSOC b When the efficiency is >5%, it indicates insufficient energy utilization, requiring an increase in the proportion of electricity used and a reduction in the equivalence factor λ to λ0. n1 : λ n1 =λ×(1-k1×(ΔSOC-5%) / 100); This formula establishes a linear relationship between the equivalent factor λ and the increase in SOC. The greater the increase, the greater the decrease in the equivalent factor λ, and the higher the electricity consumption ratio.

[0018] The calibration principle for k1 considers that excessively low k1 sensitivity mainly affects economic efficiency, and a relatively fast response capability is required. Therefore, simulation analysis software is used to simulate the adjustment process, aiming to reach the SOC change threshold ΔSOC in 2-3 adjustments. b As long as it reaches the ideal range When ΔSOC continues to decline, it indicates that energy consumption is too rapid, which may affect power performance. Therefore, it is necessary to reduce the proportion of electricity used and increase the equivalent factor λ to λ0. n2 : λn2=λ×(1+k2×(ΔSOC-5%) / 100); This formula ensures that the equivalent factor λ is increased without sacrificing core dynamic performance.

[0019] The calibration principle for k2 is to consider that if the sensitivity of k2 is too low, it will mainly affect the power performance. For commercial vehicles, the required number of adjustments can be appropriately reduced. Calibration can be carried out after 3 to 5 adjustments until the SOC change reaches the ideal threshold.

Claims

1. A hybrid power train energy management method, characterized in that, include: S1, driving endpoint positioning, by collecting GPS data in real time during vehicle driving through the vehicle GPS module, the GPS data includes at least GPS coordinates and driving heading angle; when the included angle between two consecutively collected driving heading angles is greater than or equal to a preset heading angle change threshold, and the vehicle is in a start-stop condition within the time period corresponding to the two driving heading angle collection points, the location of the latter collection point in the consecutive collection points is marked as the driving endpoint. GPS coordinates include longitude and latitude coordinates; S2, fixed route modeling: obtain the GPS trajectory of multiple running segments based on the GPS coordinates in N continuously collected GPS data segments, construct a sparse matrix representing the spatial distribution of the trajectory of each running segment according to the preset grid size, and use an adaptive resonance theory neural network model to cluster the sparse matrix corresponding to each running segment according to the preset similarity threshold to identify similar paths as historical fixed routes. S3, SOC data acquisition, collects the SOC value of the vehicle in real time through the battery management system (BMS) during the driving process, and records the SOC status when reaching each driving endpoint and destination each time a fixed historical route is traveled. S4, Calculation of SOC change for round trip: Based on the SOC at the end of a single round trip and the SOC at the starting point, the change in SOC for a single round trip, ΔSOC, is calculated. S5, Multi-period SOC mean calculation: The mean value ΔSOC is calculated by averaging the change in SOC per round trip ΔSOC for multiple consecutive round trips on a historical fixed route. avg The mean ΔSOC avg As a quantitative basis for adjusting the equivalent factor λ; S6. Based on the SOC change, dynamically adjust the equivalent factor λ, obtain the current driving route mileage and the vehicle's power battery rated capacity, and dynamically set the SOC change threshold ΔSOC according to the mileage and rated capacity. b ; When ΔSOC is greater than ΔSOC b When this occurs, the equivalent factor λ is adjusted.

2. The hybrid train energy management method according to claim 1, characterized in that, Specifically, S6 is: ΔSOC is less than or equal to the stated ΔSOC b At this time, the equivalent factor λ adjustment is not triggered; ΔSOC is greater than the stated ΔSOC b When, the equivalent factor λ is adjusted to λ n1 : l n1 =λ×(1-k1×(ΔSOC-ΔSOC b ) / 100); In the formula, k1 is the empirical coefficient for adjusting SOC increase; When ΔSOC continues to decrease, the equivalent factor λ is adjusted to λ. n2 : l n2 =λ×(1+k2×(ΔSOC-ΔSOC b ) / 100); In the formula, k2 is the empirical coefficient for adjusting the SOC decrease.

3. The hybrid power train energy management method according to claim 2, characterized in that, The constraint condition for adjusting the equivalent factor λ is: l min ≤λ n1 ≤λ max ; l min ≤λ n2 ≤λ max ; In the formula, λ min The minimum equivalent factor set at the factory; λ max The maximum equivalent factor set at the factory.

4. The hybrid train energy management method according to claim 1, characterized in that, It also includes route deviation identification and adaptive processing, wherein the route deviation identification is: The system compares the current GPS coordinates with the GPS coordinate set of the historical fixed route in real time, calculates the shortest distance d between the current location and the historical fixed route, uses the shortest distance d as the coordinate offset, sets a distance threshold d0, and sets a deviation duration threshold t0. When the shortest distance d is continuously greater than d0 for a duration exceeding t0, it is determined to be a deviation from the historical fixed route. The adaptive processing is as follows: when it is determined that the route deviates from the historical fixed route, the equivalent factor λ is initialized to the original equivalent factor λ. init ; When a vehicle returns to a historical fixed route and the duration is ≥t0, the dynamic adjustment of the equivalent factor based on the change of SOC is restored, and the equivalent factor λ corresponding to the historical route is used as the initial value. To ensure the validity of the initial values ​​when the vehicle returns to the historical route, the GPS coordinate sets of multiple historical fixed routes and their corresponding equivalent factors λ are recorded.

Images