Intelligent chassis energy management system for mountainous city

Abstract

The invention, which belongs to the technical field of the automobile industry, discloses an intelligent chassis energy management system for a mountain city, and the system comprises a sensing module which is used for collecting topographic data and vehicle state data; the decision module is used for predicting the kinetic energy recovery demand of the vehicle in real time by taking the topographic data as input according to a pre-constructed deep learning prediction model, and generating dynamic decision information based on the kinetic energy recovery demand and the vehicle state data; the execution module is used for controlling each subsystem of the intelligent chassis to execute corresponding adjustment operation according to the dynamic decision information output by the decision module in real time; and the optimization module is used for acquiring the actual execution effect of the execution module in real time, adjusting related decision parameters in the dynamic decision information based on a difference result between the actual execution effect and the expected effect of the dynamic decision information, and optimizing and training the deep learning prediction model through the actual execution effect and the corresponding dynamic decision information.

Description

Technical Field

[0001] The present invention belongs to the technical field of the automotive industry, and particularly relates to an intelligent chassis energy management system for mountain cities. Background Art

[0002] With the rapid development of the electric vehicle industry, the battery capacity and cruising range of electric vehicles have gradually become the focus of attention of many consumers; in order to extend the cruising range as much as possible for new energy vehicles, the prior art recovers some energy by setting up an energy recovery device.

[0003] However, in the mountain city environment, there are many long downhill slopes and steep uphill starts, which pose high requirements for the energy management of vehicles. The existing general energy management systems are difficult to achieve efficient energy recovery, coordinated power output and thermal management. Summary of the Invention

[0004] In view of this, the purpose of the present invention is to provide an intelligent chassis energy management system for mountain cities to solve the above technical problems.

[0005] To achieve the above purpose, the present invention provides the following technical solutions:

[0006] An intelligent chassis energy management system for mountain cities, comprising:

[0007] A sensing module, configured to collect terrain data and vehicle state data;

[0008] A decision-making module, configured to use the terrain data as input according to a pre-constructed deep learning prediction model, and in real time predict the kinetic energy recovery requirement of the vehicle, and generate dynamic decision-making information based on the kinetic energy recovery requirement and the vehicle state data;

[0009] An execution module, configured to control each subsystem of the intelligent chassis to perform corresponding adjustment operations respectively according to the dynamic decision-making information output by the decision-making module in real time;

[0010] An optimization module, configured to obtain the actual execution effect of the execution module in real time, adjust the relevant decision-making parameters in the dynamic decision-making information based on the difference result between the actual execution effect and the expected effect of the dynamic decision-making information, and optimize and train the deep learning prediction model through the actual execution effect and the corresponding dynamic decision-making information.

[0011] Further, the sensing module includes a terrain data collection sub-module and a vehicle data collection sub-module; wherein,

[0012] The terrain data collection sub-module, including a lidar, a camera and a GPS, is configured to collect terrain data; the terrain data includes slope, slope length, curve radius and surface friction coefficient;

[0013] The vehicle data acquisition sub-module, including an IMU and a battery sensor, is used to acquire vehicle state data; the vehicle state data includes vehicle speed, vehicle acceleration, direction data, and battery state.

[0014] Furthermore, a deep learning prediction model is constructed based on the mapping relationship between historical terrain data and the vehicle power system.

[0015] Furthermore, dynamic decision information is generated based on the kinetic energy recovery demand and vehicle state data, including:

[0016] Obtain vehicle state data, which is divided into driving data and battery state; among them, the driving data includes vehicle speed, vehicle acceleration, and direction data;

[0017] Based on the kinetic energy recovery demand and driving data, dynamically adjust the driving force distribution of the vehicle's front and rear wheels and the energy recovery intensity of the braking system to generate initial decision information;

[0018] Obtain the battery state and set the battery temperature threshold, and based on the safety mechanism, adjust the relevant decision parameters in the initial decision information to generate dynamic decision information; among them, the safety mechanism is to prevent overcharging or overheating of the battery.

[0019] Furthermore, an intelligent chassis energy management system for mountain cities further includes: monitoring the temperature of the braking system, and setting the braking temperature threshold, and based on the second safety mechanism, modulating the relevant decision parameters in the dynamic decision information to generate new dynamic decision information; among them, the second safety mechanism is to prevent overheating of the braking system.

[0020] Furthermore, each subsystem of the intelligent chassis includes an energy recovery subsystem, a power output subsystem, and a temperature protection subsystem; among them,

[0021] The energy recovery subsystem is used to dynamically adjust the current system working state according to the energy recovery demand included in the dynamic decision information;

[0022] The power output subsystem is used to adjust the output power of the engine or motor according to the vehicle driving demand included in the dynamic decision information;

[0023] The temperature protection subsystem is used to selectively activate the current limiting protection or cooling system according to the battery state information and the braking system temperature included in the dynamic decision information.

[0024] Furthermore, each subsystem of the intelligent chassis further includes a chassis adaptive adjustment subsystem, which is used to automatically adjust the suspension system according to the change information of the terrain data.

[0025] Furthermore, an intelligent chassis energy management system for mountain cities further includes: The sensing module, based on the vehicle's GPS positioning, the vehicle's traveling direction, and a preset sensing range, real-time calls the corresponding terrain data stored in the map APP database as estimated terrain data, and feeds it back to the decision-making module; The decision-making module inputs the estimated terrain data into a pre-constructed deep learning prediction model to obtain the vehicle's second kinetic energy recovery requirement for assisting in generating dynamic decision-making information; At the same time, the sensing module generates an actual terrain feature map based on the real-time collected terrain data, and a comparison terrain feature map based on the estimated terrain data; Align the features of the actual terrain feature map and the comparison terrain feature map according to the GPS positioning, and determine whether the terrain feature similarity between the two is less than a preset similarity threshold. If so, feed the terrain data back to the decision-making module to regenerate the dynamic decision-making information, and based on the coverage range of the current terrain data, re-call the estimated terrain data that does not cover the coverage range for generating subsequent dynamic decision-making information.

[0026] The beneficial effects of the present invention are as follows:

[0027] The present invention provides an intelligent chassis energy management system for mountain cities. Compared with the prior art's method of dealing with complex terrains through simple energy recovery modes or fixed power distribution strategies, this application integrates terrain perception, thermal management, and power distribution to generate a dynamic control strategy that combines the four dimensions of perception, decision-making, execution, and optimization, solving the problems of the traditional system lacking the integration of terrain perception, thermal management, and power distribution, the inability to smoothly switch between regenerative braking energy recovery and mechanical braking, and the poor driving comfort of vehicles in mixed downhill and uphill scenarios.

[0028] Other advantages, objectives, and features of the present invention will be elaborated in the subsequent specification, and to some extent, will be obvious to those skilled in the art, or those skilled in the art can obtain teachings from the practice of the present invention. The objectives and other advantages of the present invention can be achieved and obtained through the structures specifically pointed out in the written specification and the accompanying drawings.

[0029] The technical solutions of the present invention will be further described in detail below through the accompanying drawings and embodiments. Description of the Drawings

[0030] The drawings are used to provide a further understanding of the present invention, and constitute a part of the specification. Together with the embodiments of the present invention, they are used to explain the present invention, and do not constitute a limitation to the present invention. In the drawings:

[0031] Figure 1 It is a schematic diagram of the system modules of an intelligent chassis energy management system for mountain cities in an embodiment of the present invention. Detailed Embodiments

[0032] The preferred embodiments of the present invention will be described below in conjunction with the accompanying drawings. It should be understood that the preferred embodiments described herein are only for the purpose of illustrating and explaining the present invention, and are not intended to limit the present invention.

[0033] As Figure 1 shown, the present invention provides an intelligent chassis energy management system for mountain cities, including:

[0034] A sensing module for collecting terrain data and vehicle state data;

[0035] A decision-making module for taking the terrain data as input according to a pre-constructed deep learning prediction model, and in real-time predicting the kinetic energy recovery demand of the vehicle, and generating dynamic decision-making information based on the kinetic energy recovery demand and the vehicle state data;

[0036] An execution module for controlling each subsystem of the intelligent chassis to perform corresponding adjustment operations respectively according to the dynamic decision-making information output by the decision-making module in real-time;

[0037] An optimization module for obtaining the actual execution effect of the execution module in real-time, adjusting the relevant decision-making parameters in the dynamic decision-making information based on the difference result between the actual execution effect and the expected effect of the dynamic decision-making information, and optimizing and training the deep learning prediction model through the actual execution effect and the corresponding dynamic decision-making information;

[0038] The working principle of the above technical solution is as follows: In view of the limitations of the existing technology in energy management by adopting a simple energy recovery mode (energy recovery mode based on a fixed emphasis) or a fixed power distribution strategy (regenerative braking recovery mode) when facing complex terrains, the present application provides an intelligent chassis energy management system for mountain cities. Compared with the existing technology, the present application generates a dynamic control strategy integrating the four dimensions of sensing, decision-making, execution, and optimization through the integration of terrain sensing, thermal management, and power distribution, solving the problems of the traditional system lacking the integration of terrain sensing, thermal management, and power distribution, the inability to smoothly switch between braking energy recovery and mechanical braking, and the poor driving comfort of the vehicle in the mixed downhill and uphill scenarios;

[0039] Specifically, the present application includes a sensing module, a decision-making module, an execution module, and an optimization module. Among them, the sensing layer combines a variety of sensors such as lidar, cameras, GPS, IMU, and battery sensors. Through lidar and cameras, the system based on the sensing layer can accurately sense the surrounding environment, obtain terrain data, and be used to support the system to judge the slope change and friction condition of the road, and infer the possible vehicle power demand;

[0040] The GPS positioning module provides accurate vehicle position, speed, and motion trajectory data to assist the system in judging changes in the driving environment; the IMU sensor monitors the vehicle's attitude and driving state in real time by measuring dynamic parameters such as acceleration, vehicle speed, and direction changes;

[0041] The battery sensor can monitor the battery status (such as battery level, temperature, and battery health) in real time and transmit the dynamic data of the battery for the decision-making layer to adjust the energy management strategy; by integrating these sensor data, the system can obtain the state information of the vehicle in the complex road environment of mountainous cities in real time and provide real-time data support for subsequent decisions;

[0042] The decision-making layer constructs a deep learning model based on the mapping relationship between terrain data (slope length, slope) and the powertrain, predicts the kinetic energy recovery demand in real time, dynamically adjusts the driving force distribution between the front and rear wheels and the energy recovery intensity of the braking system, and allocates the temperature threshold of the mechanical braking mode / kinetic energy recovery model to avoid overcharging of the battery / overheating of the braking system;

[0043] Based on the output instructions of the decision-making layer, the execution layer executes energy recovery, power output, temperature protection, and chassis adaptive adjustment in each subsystem of the chassis respectively to adapt to different mountainous city roads;

[0044] The optimization layer monitors the execution effect of the system in real time, performs closed-loop optimization on the energy management strategy and power output strategy, calculates the feedback based on the difference model between the actual energy recovery and power distribution effect and the expected value, and adjusts the decision-making parameters in real time to improve the energy recovery efficiency, power output efficiency, and driving comfort; furthermore, the system dynamically adjusts the decision-making parameters according to the difference between the actual energy recovery and power output effect and the expected goal, and optimizes the training of the deep learning model with the feedback data to improve the accuracy of the strategy. The entire optimization process also includes personalized adjustment according to different road conditions, driving habits, and battery health status and other factors to ensure that the system can maintain the best operating state in the complex and changeable mountainous city road environment;

[0045] The beneficial effects of the above technical solution are as follows: Through the above technical solution, compared with the prior art, this method smoothly adjusts the braking force and recovery intensity during long downhill slopes through terrain perception and dynamic energy recovery optimization, improving the sudden braking feeling caused by excessive recovery force or energy waste caused by insufficient recovery in the traditional system; combined with real-time torque distribution and slope prediction, it solves the problems of insufficient driving force or response lag caused by traditional fixed-logic power distribution; based on the intelligent recovery method, it realizes the collaborative optimization of kinetic energy recovery and battery protection, endowing the chassis with the ability to adapt to the high-frequency long downhill scenarios in mountainous cities.

[0046] In one embodiment, the sensing module includes a terrain data acquisition sub-module and a vehicle data acquisition sub-module; among them,

[0047] The terrain data acquisition sub-module, including lidar, cameras, and GPS, is used to acquire terrain data; the terrain data includes slope, slope length, curve radius, and surface friction coefficient;

[0048] The vehicle data acquisition sub-module, including IMU and battery sensors, is used to acquire vehicle state data; the vehicle state data includes vehicle speed, vehicle acceleration, direction data, and battery state.

[0049] In one embodiment, a deep learning prediction model is constructed based on the mapping relationship between historical terrain data and the vehicle power system;

[0050] The working principle and beneficial effects of the above technical solution are as follows: Its principle is mainly to determine the friction force according to the slope length and slope, combined with the change in vehicle gravity, calculate the power demand required to meet the vehicle's driving at a certain speed, generate the mapping relationship between historical terrain data and the vehicle power system, and train the initial deep learning model based on this mapping relationship until the preset training conditions are met to generate a deep learning prediction model.

[0051] In one embodiment, dynamic decision information is generated based on the kinetic energy recovery demand and vehicle state data, including:

[0052] Obtain vehicle state data, which is divided into driving data and battery state; among them, the driving data includes vehicle speed, vehicle acceleration, and direction data;

[0053] Based on the kinetic energy recovery demand and driving data, dynamically adjust the driving force distribution of the vehicle's front and rear wheels and the energy recovery intensity of the braking system to generate initial decision information;

[0054] Obtain the battery state and set the battery temperature threshold, and based on the safety mechanism, adjust the relevant decision parameters in the initial decision information to generate dynamic decision information; among them, the safety mechanism is to prevent overcharging or overheating of the battery;

[0055] It also includes: monitoring the temperature of the braking system and setting the braking temperature threshold, and based on the second safety mechanism, modulating the relevant decision parameters in the dynamic decision information to generate new dynamic decision information; among them, the second safety mechanism is to prevent overheating of the braking system;

[0056] The working principle and beneficial effects of the above technical solution are as follows: Based on the constructed deep learning model, calculate the appropriate driving force distribution of the front and rear wheels according to the road surface conditions, and determine the energy recovery intensity of the braking system to maximize the energy recovery efficiency. Based on the current state and temperature of the battery, dynamically adjust the temperature protection strategy to avoid overcharging of the battery or overheating of the braking system. The decision-making layer is based on the real-time monitoring of the terrain, power demand, and vehicle state, and makes timely adjustments according to the actual situation.

[0057] In one embodiment, the subsystems of the intelligent chassis include an energy recovery subsystem, a power output subsystem, and a temperature protection subsystem; wherein,

[0058] The energy recovery subsystem is used to dynamically adjust the current system working state according to the energy recovery requirements included in the dynamic decision information;

[0059] The power output subsystem is used to adjust the output power of the engine or motor according to the vehicle driving requirements included in the dynamic decision information;

[0060] The temperature protection subsystem is used to selectively activate the current limiting protection or cooling system according to the battery state information and brake system temperature included in the dynamic decision information;

[0061] In addition, the subsystems of the intelligent chassis further include a chassis adaptive adjustment subsystem, which is used to automatically adjust the suspension system according to the change information of the terrain data;

[0062] The working principle and beneficial effects of the above technical solutions are as follows: The energy recovery subsystem adjusts the working state of the kinetic energy recovery system according to the real-time energy recovery requirements. The power output system adjusts the output power of the engine or motor according to the real-time requirements of the vehicle. At the same time, the system monitors the temperatures of the battery and the brake system, activates the cooling system or current limiting protection according to the temperature data. The chassis adaptive adjustment subsystem automatically adjusts the suspension system according to the terrain changes to adapt to the changes of mountain city roads.

[0063] In one embodiment, an intelligent chassis energy management system for mountain cities further includes: The sensing module, according to the vehicle's GPS positioning, vehicle traveling direction, and preset sensing range, real-time calls the corresponding terrain data stored in the map APP database as the estimated terrain data, and feeds it back to the decision module; The decision module inputs the estimated terrain data into a pre-constructed deep learning prediction model to obtain the second kinetic energy recovery requirement of the vehicle for assisting in generating dynamic decision information; At the same time, the sensing module generates an actual terrain feature map based on the real-time collected terrain data, and a comparison terrain feature map based on the estimated terrain data; Align the features of the actual terrain feature map and the comparison terrain feature map according to the GPS positioning, and determine whether the terrain feature similarity between the two is less than the preset similarity threshold. If so, feed the terrain data back to the decision module to regenerate the dynamic decision information, and based on the coverage range of the current terrain data, re-call the estimated terrain data that does not cover the coverage range for generating subsequent dynamic decision information;

[0064] The working principle and beneficial effects of the above technical solution are as follows: The perception module, based on the vehicle's GPS positioning, the vehicle's traveling direction, and a preset perception range, calls in real time the corresponding terrain data stored in the map APP database as estimated terrain data, and feeds it back to the decision-making module. It should be noted that the acquisition range of the estimated terrain data is equal to the preset perception range. To better understand this technical feature, an explanation is given here: The current vehicle position is determined according to the GPS positioning, and a preset straight-line distance and an estimated terrain range with a preset width are obtained based on the vehicle's traveling direction. The estimated terrain range is divided into multiple intervals based on the preset perception range, and for each interval, the corresponding terrain data stored in the map APP database is called as the estimated terrain data; for the estimated terrain data of each interval, the decision-making module inputs it into a pre-constructed deep learning prediction model to obtain the vehicle's second kinetic energy recovery requirement for assisting in generating dynamic decision-making information, that is, using the second kinetic energy recovery requirement to replace the original kinetic energy recovery requirement, achieving the purpose of predicting before collection and pre-starting the energy management system before entering the target area to offset the defect of low recovery efficiency during the time period of collecting terrain data when entering the target area. The specific generation process has been described in the above technical solution and will not be elaborated here; at the same time, after generating the dynamic decision-making information, to better improve the energy recovery efficiency, the perception module generates an actual terrain feature map based on the real-time collected terrain data and a comparison terrain feature map based on the estimated terrain data. It should be noted that the terrain features in the terrain feature map include but are not limited to slope features, slope length features, curve radius features, surface friction coefficient features, etc.; then, according to the GPS positioning, the actual terrain feature map and the comparison terrain feature map are feature-aligned, that is, each terrain feature in the two feature maps is corresponding according to its corresponding positioning information, facilitating the judgment of whether the similarity of the terrain features between the two is less than a preset similarity threshold. Since the estimated terrain data of the comparison terrain feature map has been fixed in the perception range when retrieved from the map APP database, the range of the comparison terrain feature map is fixed, while the actual terrain feature map is generated based on the real-time collected terrain data, and its perception range changes with the increase of the collected data, so the range of the actual terrain feature map is variable. Then, the preset similarity threshold proposed in this application will change in real time according to the range difference between the actual terrain feature map and the comparison terrain feature map, but when changing, it is necessary to ensure that the similarity of the terrain features in the actual terrain feature map and the corresponding terrain features in the comparison terrain feature map is above 90%, preventing the original energy recovery decision-making accuracy from decreasing due to too large terrain changes;After each actual topographic feature map is generated, a judgment is made. If the similarity of the topographic features between the two is less than the preset similarity threshold at any time, the topographic data is fed back to the decision-making module to regenerate the dynamic decision-making information. At the same time, within the current preset perception range, the estimated topographic data is no longer used to assist in generating the dynamic decision-making information. Based on the coverage range of the current topographic data, which is the preset perception range corresponding to the current topographic data, the estimated topographic data that does not cover the current coverage range is re-called to generate the subsequent dynamic decision-making information. Through the above technical solution, compared with the method of actual measurement and calculation (actual measurement of topographic data and real-time calculation of the kinetic energy recovery requirement according to the measured data), this technical solution can realize the pre-start of the energy management system before entering the target area, offset the defect of low recovery efficiency caused by the time period of collecting topographic data when entering the target area, and at the same time, a correction mechanism is set. When the difference in the estimated topographic data is too large, the dynamic decision-making information is automatically corrected according to the measured topographic data to achieve efficient and reliable chassis energy management.

[0065] Finally, it should be noted that the above preferred embodiments are only used to illustrate the technical solutions of the present invention and not to limit them. Although the present invention has been described in detail through the above preferred embodiments, those skilled in the art should understand that various changes can be made in form and details without departing from the scope defined by the claims of the present invention.

Claims

1. An intelligent chassis energy management system for mountainous cities, characterized in that: include: Perception module, used to collect terrain data and vehicle status data; A decision module is used to predict the vehicle's kinetic energy recovery demand in real time based on the pre-built deep learning prediction model and terrain data as input, and to generate dynamic decision information based on the kinetic energy recovery demand and vehicle status data; An execution module, used to control each subsystem of the intelligent chassis to perform corresponding adjustment operations according to the dynamic decision information output in real time by the decision module; The optimization module obtains the actual execution effect of the execution module in real time, adjusts the relevant decision parameters in the dynamic decision information based on the difference between the actual execution effect and the expected effect of the dynamic decision information, and optimizes the training of the deep learning prediction model through the actual execution effect and the corresponding dynamic decision information.

2. The intelligent chassis energy management system for mountainous cities according to claim 1 is characterized in that: The perception module includes a terrain data collection submodule and a vehicle data collection submodule; wherein, The terrain data acquisition submodule includes a laser radar, a camera, and a GPS, which are used to collect terrain data; the terrain data includes slope, slope length, curve radius, and surface friction coefficient; The vehicle data acquisition submodule includes an IMU and a battery sensor, which are used to collect vehicle status data; the vehicle status data includes vehicle speed, vehicle acceleration, direction data, and battery status.

3. The intelligent chassis energy management system for mountainous cities according to claim 1 is characterized in that: A deep learning prediction model is constructed based on the mapping relationship between historical terrain data and vehicle power systems.

4. The intelligent chassis energy management system for mountainous cities according to claim 1 is characterized in that: Generate dynamic decision information based on kinetic energy recovery requirements and vehicle status data, including: Acquire vehicle status data, which is divided into driving data and battery status; wherein the driving data includes vehicle speed, vehicle acceleration and direction data; Based on the kinetic energy recovery requirements and driving data, the vehicle's front and rear wheel drive force distribution and the energy recovery intensity of the braking system are dynamically adjusted to generate initial decision information; Obtain the battery status and set the battery temperature threshold, adjust the relevant decision parameters in the initial decision information based on the safety mechanism, and generate dynamic decision information; wherein the safety mechanism is to prevent the battery from overcharging or overheating.

5. The intelligent chassis energy management system for mountainous cities according to claim 4 is characterized in that: Also includes: Monitor the temperature of the brake system and set a brake temperature threshold, and based on the second safety mechanism, modulate the relevant decision parameters in the dynamic decision information to generate new dynamic decision information; wherein the second safety mechanism is to prevent the brake system from overheating.

6. The intelligent chassis energy management system for mountainous cities according to claim 1 is characterized in that: The subsystems of the intelligent chassis include energy recovery subsystem, power output subsystem and temperature protection subsystem; among them, An energy recovery subsystem, used to dynamically adjust the current system working state according to the energy recovery demand contained in the dynamic decision information; A power output subsystem for adjusting the output power of the engine or the electric motor according to the vehicle driving demand contained in the dynamic decision information; The temperature protection subsystem is used to selectively start the current limiting protection or cooling system according to the battery status information and the brake system temperature contained in the dynamic decision information.

7. The intelligent chassis energy management system for mountainous cities according to claim 6 is characterized in that: The subsystems of the intelligent chassis also include a chassis adaptive adjustment subsystem, which is used to automatically adjust the suspension system according to changes in terrain data.

8. The intelligent chassis energy management system for mountainous cities according to claim 1 is characterized in that: Also includes: The perception module calls the corresponding terrain data stored in the map APP database in real time according to the vehicle's GPS positioning, the vehicle's direction of travel and the preset perception range, as estimated terrain data, and feeds it back to the decision module; the decision module inputs the estimated terrain data into the pre-built deep learning prediction model to obtain the vehicle's second kinetic energy recovery demand to assist in generating dynamic decision information; at the same time, the perception module generates an actual terrain feature map based on the terrain data collected in real time, and generates a comparison terrain feature map based on the estimated terrain data; the actual terrain feature map and the comparison terrain feature map are feature aligned according to the GPS positioning, and it is determined whether the terrain feature similarity between the two is less than the preset similarity threshold. If so, the terrain data is fed back to the decision module to regenerate dynamic decision information, and based on the coverage of the current terrain data, the estimated terrain data that does not cover the coverage range is re-called to generate subsequent dynamic decision information.

Images