Search

CN-121998529-A - Vehicle intelligent management method and system based on multidimensional data perception and dynamic analysis

CN121998529ACN 121998529 ACN121998529 ACN 121998529ACN-121998529-A

Abstract

The invention discloses a vehicle intelligent management method and system based on multidimensional data perception and dynamic analysis, and aims to solve the problems of hysteresis distortion, stiffness of an assessment mechanism, passive safety management and information island of vehicle management of the existing aluminum processing enterprises. The system comprises a vehicle-mounted intelligent terminal, a cloud management platform and a user interaction terminal, wherein the vehicle management system is constructed through an integration scheme of hardware perception, a software platform and an intelligent algorithm, and the management method comprises dynamic profit assessment, intelligent wage improvement, predictive maintenance, dynamic fuel consumption standard management, active safety management and intelligent cargo loading verification. The invention realizes the real-time data acquisition and automatic analysis, upgrades the management from the experience drive to the data and algorithm drive, converts the post-processing into the pre-early warning and the in-process intervention, opens up the multi-system data flow, realizes the accurate management and control of the transportation cost, the pre-prevention of the transportation safety and the optimal allocation of transportation resources, and remarkably improves the management efficiency and the decision scientificity.

Inventors

  • WANG WANHONG
  • Shao Sanyong
  • YAN SHUAIJIE

Assignees

  • 河南义瑞新材料科技有限公司

Dates

Publication Date
20260508
Application Date
20260126

Claims (10)

  1. 1. Vehicle intelligent management system based on multidimensional data perception and dynamic analysis, which is characterized by comprising: The vehicle-mounted intelligent terminal is arranged on each transport vehicle, integrates a GPS module, an Internet of things sensor, an ADAS auxiliary driving camera and a DMS driver monitoring system and is used for collecting vehicle position, oil consumption, oil quantity of an oil tank, engine working conditions, driving behavior data, driver state data and driving environment data in real time; The cloud management platform is in communication connection with the vehicle-mounted intelligent terminal, is used for receiving, storing and processing various data uploaded by the vehicle-mounted intelligent terminal, and is associated with a cost database, wherein the cost database stores road fee standards, subsidization standards and pro-proportional rules; The user interaction terminal comprises a PC end and a mobile end App, and is in communication connection with the cloud management platform, so that management personnel, drivers and financial staff can view data, receive early warning information and process business.
  2. 2. The intelligent vehicle management system based on multidimensional data perception and dynamic analysis according to claim 1 is characterized in that the internet of things sensor comprises a fuel consumption sensor, a fuel quantity sensor, an engine working condition sensor and an acceleration sensor which are respectively used for monitoring real-time fuel consumption, residual fuel quantity of a fuel tank, operation load and operation duration of the engine and sudden acceleration/sudden braking data of the vehicle, an ADAS auxiliary driving camera is used for monitoring lane departure, collision risk of a front vehicle and driving road conditions, and a DMS driver monitoring system is used for monitoring fatigue state and distraction driving behavior of a driver.
  3. 3. The intelligent vehicle management method based on multidimensional data perception and dynamic analysis is characterized by comprising the following steps: The method comprises a full-flow profit automatic accounting method, an intelligent wage proposal method, a maintenance cost management method based on predictive maintenance, a dynamic fuel consumption benchmark generation and rewarding method based on big data learning, an ADAS/DMS integrated active safety risk early warning and scoring method and an aluminum loading compliance automatic checking method based on image recognition.
  4. 4. The vehicle intelligent management system based on multidimensional data sensing and dynamic analysis according to claim 3, wherein the method for automatically accounting full-flow profit of a transportation task comprises the following steps: s11, acquiring actual mileage data of a current transportation task through a GPS module of a vehicle-mounted intelligent terminal, acquiring actual oil consumption data through an Internet of things sensor, and acquiring road toll data through an ETC system interface; S12, the cloud management platform associates a cost database, and extracts subsidy standard and cost accounting parameters corresponding to the current transportation task; S13, calculating the dynamic profit of the transportation task through a profit accounting model, wherein the dynamic profit=transportation income-oil consumption cost-road toll-auxiliary cost-other fixed cost; And S14, synchronizing the dynamic profit data to a driver side App of the user interaction terminal in real time to realize real-time visualization of the performance, and S45, automatically summarizing and generating periodic profit reports of each vehicle and each driver by financial staff through the user interaction terminal to provide data support for management decisions.
  5. 5. A vehicle intelligent management system based on multidimensional data sensing and dynamic analysis as claimed in claim 3, wherein the intelligent payroll method comprises the steps of: s21, the cloud management platform automatically matches corresponding basic scales according to the verified transportation task route; s22, acquiring the time rate, the good rate and the oil consumption efficiency data of a transportation task through a system, and calculating a comprehensive efficiency coefficient, wherein the comprehensive efficiency coefficient=alpha×the time rate+beta×the good rate+gamma×the oil consumption efficiency, wherein alpha, beta and gamma are weight coefficients, and alpha+beta+gamma=1; s23, calculating an actual withdrawal amount according to the basic withdrawal proportion and the comprehensive efficiency coefficient, wherein the actual withdrawal amount=the basic withdrawal multiplied by the comprehensive efficiency coefficient; and S24, automatically summarizing the sum of all the transportation tasks in the period of each driver, generating a payroll by combining the basic payroll and the rewarding sum, and pushing the payroll to the driver and checking by financial staff through a user interaction terminal.
  6. 6. The vehicle intelligent management system based on multidimensional data sensing and dynamic analysis according to claim 3, wherein the maintenance cost management method based on predictive maintenance comprises the steps of: S31, acquiring vehicle engine operation data in real time through a vehicle-mounted intelligent terminal, wherein the engine operation data comprises operation time, operation load, fault codes and historical maintenance records; S32, establishing an electronic digital maintenance file for each vehicle, and recording maintenance data, fault data and operation data of the whole life cycle of the vehicle; S33, inputting engine operation data into a maintenance prediction model, wherein the maintenance prediction model is a model trained based on a machine learning algorithm and is used for predicting potential fault types, fault occurrence probability and optimal maintenance period of a vehicle; S34, automatically generating maintenance suggestions according to the prediction result, wherein the maintenance suggestions comprise maintenance items, maintenance time and maintenance priority, and pushing the maintenance suggestions to management staff through a user interaction terminal; And S35, calculating the actual maintenance condition of the vehicle, carrying out deviation analysis on the actual maintenance condition and the predicted result, and formulating a rewarding and punishing mechanism based on the deviation value, wherein rewarding is given if the actual maintenance cost is lower than the predicted range, and punishing is carried out if the actual maintenance cost exceeds the predicted range and no reasonable reason exists.
  7. 7. The intelligent vehicle management system based on multidimensional data sensing and dynamic analysis of claim 6, wherein in S33, the machine learning algorithm is a random forest algorithm or a support vector machine algorithm, the model training data comprises historical fault data, maintenance data and engine operation data of similar vehicles, and the model prediction accuracy is not lower than 85%.
  8. 8. The vehicle intelligent management system based on multidimensional data sensing and dynamic analysis according to claim 3, wherein the method for generating and rewarding the dynamic fuel consumption reference based on big data learning comprises the following steps: s41, acquiring massive historical transportation data through a vehicle-mounted intelligent terminal, wherein the historical transportation data comprise vehicle information, route information, vehicle weight information, weather information, road condition information and actual fuel consumption data; S42, preprocessing the historical transportation data, and removing abnormal data to obtain effective sample data; S43, constructing a dynamic fuel consumption reference model through a regression analysis model based on the effective sample data, and dynamically generating personalized fuel consumption reference values for each vehicle, each route, different seasons and different weather conditions; s44, acquiring actual fuel consumption data of a current transportation task in real time, and comparing the actual fuel consumption data with corresponding personalized fuel consumption reference values; And S45, automatically calculating the rewarding amount according to the comparison result, giving rewards if the actual fuel consumption is lower than the reference value, carrying out punishment if the actual fuel consumption is higher than the reference value, and synchronizing to a driver and a manager through the user interaction terminal.
  9. 9. The vehicle intelligent management system based on multidimensional data sensing and dynamic analysis according to claim 3, wherein the active safety risk early warning and scoring method of the integrated ADAS/DMS comprises the following steps: S51, acquiring driving behavior data and driver state data in the driving process in real time through an ADAS auxiliary driving camera and a DMS driver monitoring system, wherein dangerous driving behaviors comprise too close distance, sharp turning, sharp acceleration and sharp braking; S52, setting a safety risk judgment threshold value, wherein the safety risk judgment threshold value comprises a fatigue driving judgment threshold value, a distraction driving judgment threshold value and a dangerous driving behavior judgment threshold value; S53, comparing the data acquired in real time with a corresponding judgment threshold value, if the threshold value is triggered, sending out voice early warning to a driver through a vehicle-mounted intelligent terminal, and uploading early warning information to a cloud management platform; s54, based on a security risk portrait model, generating dynamic security driving scores for each driver according to the historical early warning records, the dangerous driving behavior times, the fatigue driving duration and the security training conditions of the driver, wherein the score range is 0-100 points; And S55, taking the safe driving score as a core basis for issuing and evaluating the safe rewards, and pushing a targeted training plan for drivers with scores lower than preset gridlines.
  10. 10. The intelligent vehicle management system based on multidimensional data sensing and dynamic analysis according to claim 3, wherein the aluminum material loading compliance automatic checking method based on image recognition comprises the following steps: s61, installing a high-definition monitoring camera at a loading point, wherein the shooting range of the monitoring camera covers the whole loading area, and the shooting resolution is not lower than 1080P; s62, acquiring image data after cargo loading is completed by a monitoring camera, wherein the image data comprise an aluminum ingot bundling number image, a tarpaulin covering image and a cargo bundling image; S63, inputting image data into a trained image recognition model, wherein the image recognition model is a convolutional neural network model based on a deep learning algorithm and is used for automatically recognizing whether the number of aluminum ingot bundles meets the order requirement, whether tarpaulin covers well and whether goods bundling is standard; And S64, if the identification result is non-compliance, sending an early warning to a manager through the cloud management platform to inform on-site personnel of correction, and if the identification result is compliance, generating a loading compliance record and storing the loading compliance record into a transportation task file.

Description

Vehicle intelligent management method and system based on multidimensional data perception and dynamic analysis Technical Field The invention relates to the technical field of transportation vehicle management, in particular to a vehicle intelligent management method and system based on multidimensional data perception and dynamic analysis. Background The transportation vehicles of the aluminum processing enterprises are key links for connecting production, storage and sales, and the management efficiency directly influences the transportation cost, the cargo safety and the market response speed of the enterprises. At present, the existing vehicle management method for aluminum processing enterprises depends on manual recording, static assessment and post analysis, and has a plurality of remarkable defects: The data is lagged and distorted, that is, the core data such as the fuel consumption, the driving mileage, the driving behavior and the like of the vehicle are mainly reported manually by a driver, so that the problem of data delay exists, false report and report hiding are easy to occur, the data accuracy is difficult to ensure, and the management decision lacks reliable data support; The assessment mechanism is statically rigidified, namely assessment indexes such as oil consumption, maintenance cost standard and the like are mostly rated, dynamic factors such as real-time working conditions (such as engine load), driving road conditions (such as gradient and congestion condition), weather conditions and the like of a vehicle are not considered, and assessment results lack scientificity and rationality, so that a driver is difficult to be effectively stimulated to optimize driving behaviors; The existing management mode mainly relies on post investigation to monitor unsafe behaviors such as fatigue driving, dangerous driving and the like, cannot realize real-time early warning and intervention, has weak accident prevention capability, is easy to cause safety accidents such as cargo damage, casualties and the like, and brings economic loss to enterprises; The information island problem is that the vehicle management system, the dispatching system, the financial system, the warehouse system and other business systems are mutually independent, and data cannot be communicated and shared, so that links such as planning, execution tracking, cost accounting and the like of a transportation task are not smoothly connected, the management efficiency is low, and the whole-flow closed-loop management is difficult to realize. Therefore, there is an urgent need for a full-process, data-driven, dynamically optimized intelligent vehicle management method and system that addresses the above-described shortcomings of the prior art. Disclosure of Invention The invention aims to provide a vehicle intelligent management method and system based on multidimensional data perception and dynamic analysis, which realize accurate control of transportation cost, pre-prevention of transportation safety and optimal allocation of transportation capacity resources, and improve the management intelligent level and operation efficiency of transportation vehicles of aluminum processing enterprises. In order to achieve the above purpose, the present invention provides the following technical solutions: Vehicle intelligent management system based on multidimensional data perception and dynamic analysis includes: The vehicle-mounted intelligent terminal is arranged on each transport vehicle, integrates a GPS module, an Internet of things sensor, an ADAS auxiliary driving camera and a DMS driver monitoring system and is used for collecting vehicle position, oil consumption, oil quantity of an oil tank, engine working conditions, driving behavior data, driver state data and driving environment data in real time; The cloud management platform is in communication connection with the vehicle-mounted intelligent terminal, is used for receiving, storing and processing various data uploaded by the vehicle-mounted intelligent terminal, and is associated with a cost database, wherein the cost database stores road fee standards, subsidization standards and pro-proportional rules; The user interaction terminal comprises a PC end and a mobile end App, and is in communication connection with the cloud management platform, so that management personnel, drivers and financial staff can view data, receive early warning information and process business. The system comprises an Internet of things sensor, an ADAS auxiliary driving camera, a DMS driver monitoring system and a DMS driver monitoring system, wherein the Internet of things sensor comprises an oil consumption sensor, an oil quantity sensor, an engine working condition sensor and an acceleration sensor which are respectively used for monitoring driving behavior data such as real-time oil consumption of a vehicle, residual oil quantity of an oil tank, running load and running time of the engine, sudd