Heavy truck vehicle abnormity identification and analysis method based on distributed cluster and machine learning

Abstract

The invention relates to the technical field of intelligent network connection application of heavy-duty commercial vehicles, in particular to a heavy truck vehicle abnormity identification and analysis method based on distributed cluster and machine learning. According to the system, by integrating an advanced distributed computing technology and a machine learning algorithm, efficient processing and accurate analysis of mass data are realized; the method comprises the following steps: firstly, storing collected data in a high-performance distributed file system by utilizing a distributed cluster technology, and ensuring high availability of the system through load balancing; then, carrying out preprocessing and feature extraction on the data by utilizing a distributed computing framework, and providing a high-quality data basis for subsequent anomaly recognition; then, constructing an anomaly recognition model by adopting an isolated forest algorithm, and realizing accurate recognition of anomalies such as oil consumption, high temperature and torque limitation through data training and optimization; and finally, through a visual interface and an API (Application Program Interface), displaying an exception identification result. And the big data intelligence level and the operation efficiency of the heavy truck are greatly improved.

Description

Technical Field

[0001] This invention relates to the field of intelligent connected vehicle application technology for heavy-duty commercial vehicles, specifically to a method for anomaly identification and analysis of heavy-duty trucks based on distributed clusters and machine learning. Background Technology

[0002] With the rapid development of the logistics industry, heavy-duty trucks, as the main force in logistics transportation, are of paramount importance in terms of their operational safety, economy, and environmental friendliness. However, heavy-duty trucks may experience abnormal phenomena due to various factors during use.

[0003] In the logistics and transportation sector, the monitoring and management of heavy-duty trucks has always been an important research topic. With the development of IoT technology, more and more systems are beginning to use sensors to collect vehicle operational data and combine it with data analysis techniques to identify anomalies. In existing technologies, a common approach is to centrally store and process the collected operational data, and then use statistical methods to identify anomalies.

[0004] While these systems can perform basic monitoring functions, they have the following drawbacks:

[0005] 1. Low data processing efficiency: Because centralized data processing requires all data to be transmitted to a central server for processing, the efficiency of data transmission and processing is low, which cannot meet the needs of large-scale data processing.

[0006] 2. Lack of proactive anomaly identification and intelligent analysis: Some systems can only provide basic parameter display functions, lacking proactive anomaly identification, in-depth analysis of the causes of anomalies, and handling suggestions.

[0007] Therefore, developing a method that can proactively identify and analyze anomalies such as fuel consumption, high temperature, and torque limitation in heavy-duty trucks is of great significance for improving the efficiency and safety of logistics transportation. Summary of the Invention

[0008] To address the problems existing in the prior art, the purpose of this invention is to provide a method for anomaly identification and analysis of heavy-duty trucks based on distributed clusters and machine learning.

[0009] The technical solution adopted by this invention to solve its technical problem is: a method for identifying and analyzing anomalies in heavy-duty trucks based on distributed clusters and machine learning, including a data acquisition and parsing module, a distributed storage module, a distributed computing module, an anomaly identification module, an analysis module, and an application module;

[0010] The distributed storage module utilizes the HDFS distributed file system to store the massive amounts of collected data.

[0011] The distributed computing module utilizes the Apache Spark distributed computing framework to efficiently process and analyze data stored in the distributed file system;

[0012] The anomaly detection module constructs an anomaly detection model for anomaly information using the isolated forest algorithm from machine learning.

[0013] The analysis module performs in-depth analysis of the identified abnormal information and combines it with relevant vehicle signals for joint analysis.

[0014] Preferably, the distributed computing module preprocesses the data, extracts statistical and temporal features for identifying anomalies from the original data using a feature extraction algorithm, and uses the extracted statistical and temporal features as input to train the isolated forest machine learning model.

[0015] Preferably, the application module has a visual interface and an API interface, which can view the abnormal information identification results and receive early warning information.

[0016] Preferably, the data acquisition and analysis module uses an in-vehicle data terminal to transmit and analyze vehicle operation data via a 4G network.

[0017] Preferably, the abnormal information includes information on fuel consumption, high temperature, and torque limitation.

[0018] Preferably, the vehicle-related signals include engine speed, fan speed, torque, water temperature, load, and throttle opening.

[0019] The present invention has the following beneficial effects:

[0020] 1) Real-time performance: The system can collect and process operational data in a timely manner, promptly detect anomalies such as fuel consumption, high temperature, and torque limitation, and reduce the impact of anomalies on operations;

[0021] 2) Accuracy: The isolated forest algorithm, a machine learning method, is used for anomaly identification, which has high accuracy and stability.

[0022] 3) Scalability: Utilizing a distributed cluster architecture, it can easily handle large-scale data and high-concurrency requests;

[0023] 4) Intelligent: Automatically learns and optimizes models through machine learning algorithms to improve the accuracy and efficiency of anomaly identification;

[0024] 5) Flexibility: Provides a visual interface and API interface for easy viewing and analysis of abnormal data, and supports integration with other systems. Attached Figure Description

[0025] Figure 1 This is a schematic diagram of the overall framework structure of the present invention.

[0026] In the diagram: 1-1 Data acquisition and parsing module; 1-2 Distributed storage module; 1-3 Distributed computing module; 1-4 Anomaly detection module; 1-5 Analysis module; 1-6 Application module. Detailed Implementation

[0027] The technical solutions of the embodiments of the present invention will be described in further detail below with reference to the accompanying drawings. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without creative effort are within the scope of protection of the present invention.

[0028] The method for identifying and analyzing anomalies in heavy-duty trucks based on distributed clusters and machine learning includes a system architecture consisting of six core modules: data acquisition and parsing module 1-1, distributed storage module 1-2, distributed computing module 1-3, anomaly identification module 1-4, analysis module 1-5, and application module 1-6. Each module ensures the efficient operation and accuracy of the system.

[0029] Data acquisition and analysis module 1-1: Uses an on-board data terminal to transmit and analyze vehicle operation data via a 4G network.

[0030] Distributed Storage Module 1-2: This module utilizes distributed file systems such as the Hadoop Distributed File System (HDFS) to reliably store the massive amounts of collected data. Through technologies such as data sharding, redundant backup, and load balancing, this module ensures high availability, scalability, and fault tolerance, maintaining stable performance even when facing large-scale data growth.

[0031] Distributed Computing Modules 1-3: The distributed computing module is the core of the entire system. It utilizes distributed computing frameworks such as Apache Spark to efficiently process and analyze data stored in a distributed file system. This module first performs data preprocessing, including data cleaning, deduplication, and standardization, to improve data quality. Then, through feature extraction algorithms, it extracts valuable statistical features and time-series features for anomaly identification from the raw data. Finally, the extracted features are used as input to train machine learning models such as Isolation Forest.

[0032] Anomaly Detection Modules 1-4: The anomaly detection module is a crucial part of the system, employing the Isolation Forest algorithm (a machine learning algorithm) to construct anomaly detection models for fuel consumption, high temperature, and torque limitation. The Isolation Forest algorithm is a tree-based anomaly detection algorithm that identifies outliers by constructing multiple isolated trees. During the training phase, this module utilizes statistical and temporal features extracted by the distributed computing module to train the model, resulting in a model capable of accurately identifying anomalies. During the testing phase, this module inputs real-time collected data such as fuel consumption, engine speed, fan speed, engine torque, accelerator pedal opening, and vehicle load into the model to obtain anomaly detection results. By continuously adjusting model parameters and optimizing the algorithm, this module can achieve efficient identification of anomalies such as fuel consumption, high temperature, and torque limitation while maintaining accuracy.

[0033] Analysis modules 1-5: Conduct in-depth analysis of identified abnormal information such as fuel consumption, high temperature, and torque limitation. Combine this with signals such as vehicle engine speed, fan speed, torque, water temperature, load, and throttle opening for joint analysis. Through analysis, provide decision support for vehicle maintenance.

[0034] Application Modules 1-6: These modules serve as a bridge between the system and users / enterprises. They provide a visual interface and API, allowing users to intuitively view anomaly identification results such as fuel consumption, high temperature, and torque limitation, and receive early warning information. The visual interface provides detailed information on anomalies like fuel consumption, high temperature, and torque limitation, offering strong support for intelligent decision-making within enterprises.

[0035] This invention combines distributed cluster technology and the isolated forest algorithm, and innovatively applies these two technologies to the identification of anomalies such as fuel consumption, high temperature, and torque limitation in heavy-duty trucks.

[0036] Efficient processing of massive amounts of data: Preprocessing and feature extraction of massive amounts of data through a distributed computing module improves data quality and feature effectiveness. This facilitates the subsequent training of the Isolation Forest algorithm and anomaly detection, enhancing the accuracy and reliability of anomaly detection.

[0037] Construction and Application of the Anomaly Detection Model: An anomaly detection model is constructed using the Isolation Forest algorithm, and the accuracy of anomaly detection is improved through model training and optimization. Simultaneously, a visual interface and API are provided through the application module, enabling the anomaly detection results to be intuitively displayed to users and integrated with other systems for application.

[0038] This invention is not limited to the above-described embodiments. Anyone should know that any structural changes made under the guidance of this invention, and any technical solutions that are the same as or similar to this invention, fall within the protection scope of this invention.

[0039] The technologies, shapes, and structures not described in detail in this invention are all known technologies.

Claims

1. A method for anomaly identification and analysis of heavy-duty trucks based on distributed clusters and machine learning, characterized in that, It includes a data acquisition and parsing module, a distributed storage module, a distributed computing module, an anomaly detection module, an analysis module, and an application module; The distributed storage module utilizes the HDFS distributed file system to store the massive amounts of collected data. The distributed computing module utilizes the Apache Spark distributed computing framework to efficiently process and analyze data stored in the distributed file system; The anomaly detection module constructs an anomaly detection model for anomaly information using the isolated forest algorithm from machine learning. The analysis module performs in-depth analysis of the identified abnormal information and combines it with relevant vehicle signals for joint analysis.

2. The method for anomaly identification and analysis of heavy-duty trucks based on distributed clusters and machine learning according to claim 1, characterized in that, The distributed computing module preprocesses the data and extracts statistical and temporal features for identifying anomalies from the original data using a feature extraction algorithm. The extracted statistical and temporal features are then used as input to train the isolated forest machine learning model.

3. The method for anomaly identification and analysis of heavy-duty trucks based on distributed clusters and machine learning according to claim 1, characterized in that, The application module has a visual interface and API interface, which can view the abnormal information identification results and receive early warning information.

4. The method for anomaly identification and analysis of heavy-duty trucks based on distributed clusters and machine learning according to claim 1, characterized in that, The data acquisition and analysis module uses an in-vehicle data terminal to transmit and analyze vehicle operation data via a 4G network.

5. The method for anomaly identification and analysis of heavy-duty trucks based on distributed clusters and machine learning according to claim 1, characterized in that, The abnormal information includes information on fuel consumption, high temperature, and torque limitation.

6. The method for anomaly identification and analysis of heavy-duty trucks based on distributed clusters and machine learning according to claim 1, characterized in that, The vehicle-related signals include engine speed, fan speed, torque, water temperature, load, and throttle opening.

Images