CN-121994307-A - Multimode intelligent data acquisition system and method for automobile interior odor monitoring

CN121994307ACN 121994307 ACN121994307 ACN 121994307ACN-121994307-A

Abstract

The invention discloses a multimode intelligent data acquisition system and method for odor monitoring of automotive interiors, comprising a terminal shell, a multimode sensor array module, a mixed communication module and a power management module, wherein the multimode sensor array module is integrated in the terminal shell or externally connected through a standard interface and is used for synchronously acquiring various types of parameters, the array comprises an environment parameter sensor, a gas concentration sensor, an optional vision auxiliary sensor and a technological process sensor interface, the core processing and edge intelligent module comprises a main controller and an edge intelligent unit which are arranged in the terminal shell, the mixed communication module comprises a wired communication unit and a wireless communication unit which are arranged in the terminal shell, and the power management module is used for controlling the environment parameter sensor, the gas concentration sensor, the optional vision auxiliary sensor and the technological process sensor interface. The intelligent odor consistency monitoring system completes data acquisition, achieves real-time and efficient initial diagnosis of abnormality through a built-in special algorithm, forms a hierarchical and cooperative intelligent monitoring network with a cloud system, and remarkably improves performance and efficiency of the whole odor consistency monitoring system.

Inventors

ZHANG JINGMING
QU YIBO
HOU YIHANG
ZHUANG MENGMENG
ZHU ZHENYU
DENG CHENGPENG
TIAN BOYANG
LIU YUTONG
SUN FANGZHOU
LIANG XINNING

Assignees

中汽数据(天津)有限公司
中汽数据有限公司

Dates

Publication Date: 20260508
Application Date: 20260313

Claims (10)

1. Multimode intelligent data acquisition system towards automotive interior smell control, its characterized in that includes: A terminal housing; The multi-mode sensor array module is integrated in the terminal shell or externally connected through a standard interface and is used for synchronously collecting various types of parameters, and the array comprises an environment parameter sensor, a gas concentration sensor, an optional vision auxiliary sensor and a technological process sensor interface; the core processing and edge intelligent module comprises a main controller and an edge intelligent unit which are arranged in the terminal shell; The hybrid communication module comprises a wired communication unit and a wireless communication unit which are arranged in the terminal shell; The power management module comprises a wide-voltage power input circuit, a high-capacity lithium battery and a charge and discharge management circuit, wherein the wide-voltage power input circuit and the high-capacity lithium battery are arranged in the terminal shell.
2. The multimode intelligent data acquisition system for odor monitoring of automotive interiors of claim 1, wherein the environmental parameter sensor comprises a high-precision digital temperature and humidity sensor and an atmospheric pressure sensor; the gas concentration sensor aims at an electrochemical or photoionization sensor module of common odor substances of total volatile organic compounds, formaldehyde and benzene series internal decoration; The optional vision auxiliary sensor is integrated with a low-power-consumption miniature camera or an infrared thermal imaging module and is used for acquiring equipment state images or local temperature field distribution and assisting in abnormality judgment; The process sensor interface provides an analog input interface and a digital input interface and is used for connecting an existing flowmeter, a pressure transmitter and valve position feedback equipment on a production line.
3. The system for monitoring odor of automotive interiors according to claim 1, wherein the main controller is responsible for sensor driving, data reading, protocol processing and task scheduling; And a lightweight real-time sequence abnormality detection algorithm is arranged in the edge intelligent unit, the continuously acquired sensor data is subjected to online analysis, whether the current data point is abnormal or not is judged in real time, and an abnormality confidence coefficient is calculated.
4. The system for multi-modal intelligent data collection for automotive interior odor monitoring as set forth in claim 3 wherein said real-time timing anomaly detection algorithm comprises: The edge intelligence unit processes each preprocessed sensor data stream The following calculations are performed: 1) EWMA control map calculation: calculating an exponentially weighted moving average of current data And upper/lower control limits: Wherein, the The sensor data value after pretreatment acquired at the current moment is represented, and the unit is consistent with the measured physical quantity; an exponentially weighted moving average representing the current time reflects a smooth trend of recent changes in the data; representing a weight coefficient, controlling the weight of the current data point in a value range of 0< lambda less than or equal to 1; the closer to 1, the more interesting the latest data; the closer to 0, the more dependent the history data; A sample mean value representing baseline learning stage data; Sample standard deviation representing reference learning stage data; Representing an upper control limit; representing a lower control limit; If it is Beyond [ Triggering a drift abnormality flag; the control limit coefficient is used for adjusting a constant of the control limit width, and is related to the required confidence level, and the value range is 2-3; the larger the value is, the wider the control limit is, the lower the false alarm rate is, but the false alarm rate possibility is increased; 2) Fast local outlier factor Fast-LOF estimation: Maintaining a fixed length nearest neighbor data window For new points Calculate it before it reaches the window Average distance of adjacent points Simultaneously calculating average neighbor distances of all points in the window The fast LOF score is estimated as: Setting a threshold value If (if) ( ) The "mutation" flag is triggered.
5. The multimode intelligent data acquisition system for monitoring automobile interior trim smell according to claim 4, wherein the multimode intelligent data acquisition system is characterized in that according to the drift abnormality and mutation abnormality marks, a preset rule is combined, wherein the method comprises the steps of triggering one of drift abnormality and mutation abnormality marks, alarming or alarming only by continuous N points drift, and generating a final abnormal state identifier 0 Normal, 1 abnormal, abnormal confidence score In the range of 0-1.
6. The multimode intelligent data acquisition system for monitoring the odor of the automotive trim according to claim 1 is characterized in that the wired communication unit is integrated with an RS-485 and CAN bus interface for accessing a traditional industrial field bus network; the wireless communication unit integrates a 4G/5G cellular communication module for wide area wireless backhaul, integrates a Wi-Fi and Bluetooth module for flexible networking or near field configuration debugging in a workshop, and integrates a LoRa module for forming a star network with other low-power consumption sensor nodes in the workshop.
7. The multimode intelligent data acquisition system for odor monitoring of automotive interiors according to claim 1, wherein the wide-voltage power input circuit supports DC 9-36V industrial standard power input, and the high-capacity lithium battery and the charge and discharge management circuit support continuous operation of equipment terminals without external power supply and support the solar panel to be connected to charge.
8. An acquisition method based on the multimode intelligent data acquisition system for monitoring the odor of the automotive interior according to any one of claims 1 to 7, characterized by comprising the following steps: S1, configuring terminal equipment through Bluetooth or USB connection, setting an acquisition period and a communication strategy, automatically executing self-checking after the terminal is powered on, and entering a reference learning stage for initializing subsequent abnormal detection algorithm parameters; S2, synchronously triggering a sensor array in the terminal equipment to acquire data according to a preset period by a main controller in the terminal equipment, and performing basic cleaning on the original data; S3, for each preprocessed sensor data stream The edge intelligent unit in the terminal equipment performs calculation in parallel, and the method comprises the following steps: 1) EWMA control diagram calculation; 2) Fast local outlier factor Fast-LOF estimation; judging and triggering drift abnormality and mutation abnormality marks according to the calculation result, and generating a final abnormal state mark by combining with a preset rule Anomaly confidence score ; S4, the core processing and edge intelligent module processes the data, And Packaging according to a protocol, and dynamically adjusting a communication strategy according to an abnormal state: Normal mode Reporting aggregation characteristic values such as mean value, maximum value and state heartbeat according to low frequency, or reporting by adopting change; Abnormality early warning mode Immediately starting a high-speed reporting channel, uploading an alarm data packet containing original high-density data points, abnormal confidence coefficient and algorithm identification to a monitoring center in a wired/high-speed wireless manner, and meanwhile, locally storing detailed data of an abnormal time period; And S5, the equipment terminal periodically reports the state information of the equipment terminal to a monitoring center, wherein the state information comprises battery power, signal strength, health state of each sensor and internal temperature, receives a remote instruction from the monitoring center, realizes dynamic configuration update, firmware air upgrade and diagnosis test trigger, and receives algorithm parameter update or reference model fine adjustment instruction to adapt to production change.
9. A computer readable storage medium storing a computer program, which, when executed by a processor, causes the processor to perform the steps of the method as claimed in claim 8.
10. A computer device comprising a memory and a processor, the memory storing a computer program, characterized in that the computer program, when executed by the processor, causes the processor to perform the steps of the method as claimed in claim 8.

Description

Multimode intelligent data acquisition system and method for automobile interior odor monitoring Technical Field The invention relates to the technical field of data acquisition and edge intelligence of the Internet of things, in particular to a multimode intelligent data acquisition system and method for odor monitoring of automotive interiors. Background In the existing 'vehicle interior odor mass production consistency monitoring system based on the Internet of things', the real-time property, the accuracy and the comprehensiveness of data are the basis for the effective operation of a rear-end intelligent analysis model. However, when implementing the data acquisition layer, the following challenges are faced: (1) The existing acquisition scheme mostly adopts single-function sensor scattered deployment, data collection is carried out through an independent transmitter or PLC, wiring is complex, cost is high, and time synchronization is difficult to ensure; (2) The environment adaptability is poor, the environment of an interior production field is complex (such as a drying room is high in temperature and high in humidity, chemical gas exists in a spraying workshop, electromagnetic interference exists in a general assembly line), and the common commercial data acquisition equipment is insufficient in long-term stability, protection level and anti-interference capability; (3) The data preprocessing and intelligent judgment are missing, namely the original sensor data is often accompanied with noise and outliers, and the traditional acquisition equipment is only responsible for data transmission and does not have the capability of carrying out real-time quality judgment and preliminary abnormality identification at a data generation source. This results in a large amount of invalid or normal data occupying network bandwidth, and the actual abnormal signal may be submerged or delayed to report, affecting the response speed of the backend system; (4) The deployment is inflexible, and for a mobile station or a temporary monitoring point, an acquisition terminal which can be conveniently deployed, is powered by a battery and can adaptively return data through a wireless network is lacking; (5) The protocol conversion is complex, namely equipment on the production line is different in new and old, communication protocols are various (such as 4-20mA, 0-10V, modbus, CAN buses and the like), and an additional gateway is needed for carrying out the protocol conversion, so that the complexity of the system and fault points are increased; Therefore, a multi-mode intelligent data acquisition terminal and a method which are specially designed for the automobile interior trim odor consistency monitoring scene and integrate a lightweight real-time anomaly detection algorithm are needed. Disclosure of Invention The multimode intelligent data acquisition system and method for the odor monitoring of the automotive trim, provided by the invention, can at least solve one of the technical problems in the background technology. In order to achieve the above purpose, the present invention adopts the following technical scheme: Multimode intelligent data acquisition system towards automotive interior smell control includes: A terminal housing; The multi-mode sensor array module is integrated in the terminal shell or externally connected through a standard interface and is used for synchronously collecting various types of parameters, and the array comprises an environment parameter sensor, a gas concentration sensor, an optional vision auxiliary sensor and a technological process sensor interface; the core processing and edge intelligent module comprises a main controller and an edge intelligent unit which are arranged in the terminal shell; The hybrid communication module comprises a wired communication unit and a wireless communication unit which are arranged in the terminal shell; The power management module comprises a wide-voltage power input circuit, a high-capacity lithium battery and a charge and discharge management circuit, wherein the wide-voltage power input circuit and the high-capacity lithium battery are arranged in the terminal shell. As a preferable scheme of the multimode intelligent data acquisition system for the odor monitoring of the automotive interiors, the environment parameter sensor comprises a high-precision digital temperature and humidity sensor and an atmospheric pressure sensor; the gas concentration sensor aims at an electrochemical or photoionization sensor module of common odor substances of total volatile organic compounds, formaldehyde and benzene series internal decoration; The optional vision auxiliary sensor is integrated with a low-power-consumption miniature camera or an infrared thermal imaging module and is used for acquiring equipment state images or local temperature field distribution and assisting in abnormality judgment; The process sensor interface provides an analog input interfa