CN-121984508-A - Data acquisition and analysis module and method for multi-source data of building construction

CN121984508ACN 121984508 ACN121984508 ACN 121984508ACN-121984508-A

Abstract

The invention discloses a data acquisition analysis module and a method for multi-source data in building construction, wherein the module comprises an adaptive sampling interface, a logic control chip, an AI chip and a communication module, wherein the adaptive sampling interface comprises a multi-channel input interface, a front detection unit and a configuration module which are sequentially connected, the logic control chip is connected with the adaptive sampling interface and is used for detecting the front detection unit, judging that data signals accessed by the input interfaces of all channels are analog or digital signals, and adjusting sampling frequencies of the analog or digital signals through the configuration module so as to automatically adjust the sampling frequencies of the adaptive sampling interface, the AI chip is integrated with a lightweight neural network model and used for carrying out real-time analysis, noise reduction and lightweight processing on the data signals subjected to the automatic adjustment of the sampling frequencies of the adaptive sampling interface through the logic control chip, and the communication module is used for transmitting the data signals subjected to the lightweight processing through the AI chip. The invention has the advantages of low power consumption and small data volume.

Inventors

FANG TINGCHEN
CHEN YUANHONG
GONG JIAN
FENG YU
HE HONGYU

Assignees

上海建工集团股份有限公司

Dates

Publication Date: 20260505
Application Date: 20251231

Claims (10)

1. The utility model provides a construction multisource data's data acquisition analysis module which characterized in that includes: The self-adaptive sampling interface comprises a multi-channel input interface, a front detection unit and a configuration module which are sequentially connected, wherein the multi-channel input interface comprises a plurality of video channel input interfaces and a plurality of sensor channel input interfaces which are respectively used for connecting a video camera and various sensors in building construction; The logic control chip is connected with the self-adaptive sampling interface and is used for detecting the front-end detection unit, judging whether the data signals accessed by the video channel input interfaces and the sensor channel input interfaces are analog signals or digital signals, and adjusting the sampling frequency of the data signals passing through the front-end detection unit through the configuration module so as to automatically adjust the sampling frequency of the self-adaptive sampling interface; the AI chip adopts a low-power chip of RISC-V architecture, is integrated with a lightweight neural network model and is used for carrying out real-time analysis, noise reduction and lightweight treatment on data signals after the sampling frequency of the self-adaptive sampling interface is automatically regulated by the logic control chip; and the communication module is used for transmitting the data signals subjected to the light weight processing by the AI chip.
2. The data acquisition and analysis module for construction multisource data according to claim 1, further comprising: and the battery is provided with a charging interface and is used for supplying power.
3. The data acquisition and analysis module of multi-source data for building construction according to claim 1, wherein the video channel input interface comprises a digital video channel input interface, and the logic control chip and the AI chip are integrated with h.265/h.266 encoders to support dynamic resolution switching.
4. The data acquisition and analysis module of construction multisource data according to claim 1, wherein the video channel input interface comprises an analog video channel input interface, and the analog video channel input interface is connected with a programmable gain amplifier and an analog-to-digital converter, wherein the programmable gain amplifier is used for dynamically adjusting signal gain, and the analog-to-digital converter is used for converting an analog signal into a digital signal.
5. The data acquisition and analysis module of the construction multi-source data according to claim 1, wherein the front detection unit is an impedance matching component, and the logic control chip judges the data signal corresponding to each channel input interface to be an analog signal or a digital signal according to the value or handshake protocol for detecting the impedance matching corresponding to each channel input interface.
6. The data acquisition and analysis module of the multi-source data for the building construction according to claim 1, wherein the configuration module is a switching tube connected to each channel input interface of the multi-channel input interface, a control end of the switching tube is connected with the logic control unit, two non-control ends of the switching tube are respectively connected to the front detection unit and the AI chip, and the logic control unit controls the on-off of the switching tube to control the sampling frequency of the data signal passing through the front detection unit so as to automatically adjust the sampling frequency of the self-adaptive sampling interface.
7. The data acquisition and analysis module of construction multisource data according to claim 1, wherein the communication module is ‌ NB-IoT narrowband internet of things wireless transmission module or 5G communication module.
8. A method for collecting data of construction multisource data, which is characterized in that the data collecting and analyzing module of construction multisource data according to any one of claims 1-7 is adopted, and the method comprises the following steps: Connecting a data acquisition and analysis module of the construction multi-source data with a power supply, and connecting a video camera and various sensors of a construction site on the data acquisition and analysis module of the construction multi-source data through a self-adaptive sampling interface; Detecting data signals of all channels of a multi-channel input interface of the self-adaptive sampling interface through a front detection unit by a logic control chip, judging that the data signals corresponding to all video channel input interfaces and all sensor channel input interfaces are analog signals or digital signals, adjusting the sampling frequency of the data signals passing through the front detection unit through a configuration module to automatically adjust the sampling frequency of the self-adaptive sampling interface, detecting whether a moving object exists or not by the logic control chip through an optical flow method, adjusting the self-adaptive sampling interface to be a high-frequency sampling signal when the moving object exists, otherwise, adjusting the self-adaptive sampling interface to be a low-frequency sampling signal; The method comprises the steps of carrying out real-time analysis, noise reduction and light weight processing on a data signal with an adjusted sampling frequency through an AI chip, carrying out frame extraction mode storage key frames on a static scene by using the AI chip based on background difference motion detection on the video data signal to reduce the image resolution of the video data signal, carrying out denoising and compression on the data signal of a sensor class by wavelet transformation, carrying out low-frequency processing on a steady-state signal, and carrying out high-frequency processing on data in a certain time period near an abrupt feature point.
9. The method for collecting the multi-source data for the building construction according to claim 8, wherein the communication module is used for transmitting the light-weight data output by the AI chip to the data center, and the data center is used for carrying out non-real-time analysis on the received light-weight data to judge whether the technical parameters of the corresponding monitoring of each data signal collected in the building construction process meet the construction requirements or not, so as to judge whether the progress, quality, cost and safety of the building construction process in the time period where the data are located are within a controllable range or not according to the analysis result.
10. The method for data collection of construction multisource data according to claim 9, wherein the video channel input interface is turned off at night and during non-construction periods during the day.

Description

Data acquisition and analysis module and method for multi-source data of building construction Technical Field The invention relates to the technical field of building construction data acquisition, in particular to a data acquisition analysis module and method for multi-source data of building construction. Background Along with the development of informatization technology, parameters such as temperature, stress, deformation and vibration in the building construction process are acquired by using a sensor, the acquired data are analyzed by a controller, whether the technical parameters controlled in the construction process meet the requirements is judged, and then corresponding technical measures are taken to ensure the progress, quality, cost, safety and controllability in the construction process. In addition, when the existing controller analyzes the collected data, the data can only be simply analyzed, the construction is difficult to be directly guided, in addition, when the controller collects the data, the data is generally obtained at a fixed sampling frequency, the power consumption is high, the storage resource is occupied greatly, and the cost is high, so that the storage battery is difficult to meet the long-term service requirement. Disclosure of Invention The invention aims to provide a data acquisition and analysis module and method for multi-source data in building construction, which are used for solving the problems that the traditional controller consumes large power and occupies large storage resources for acquiring data. In order to solve the technical problems, the technical scheme provided by the invention is that a data acquisition and analysis module of multi-source data for building construction comprises: The self-adaptive sampling interface comprises a multi-channel input interface, a front detection unit and a configuration module which are sequentially connected, wherein the multi-channel input interface comprises a plurality of video channel input interfaces and a plurality of sensor channel input interfaces which are respectively used for connecting a video camera and various sensors in building construction; The logic control chip is connected with the self-adaptive sampling interface and is used for detecting the front-end detection unit, judging whether the data signals accessed by the video channel input interfaces and the sensor channel input interfaces are analog signals or digital signals, and adjusting the sampling frequency of the data signals passing through the front-end detection unit through the configuration module so as to automatically adjust the sampling frequency of the self-adaptive sampling interface; the AI chip adopts a low-power chip of RISC-V architecture, is integrated with a lightweight neural network model and is used for carrying out real-time analysis, noise reduction and lightweight treatment on data signals after the sampling frequency of the self-adaptive sampling interface is automatically regulated by the logic control chip; The communication module is used for transmitting the data signals subjected to the light weight processing by the AI chip; further, the data acquisition and analysis module of the construction multisource data provided by the invention further comprises: and the battery is provided with a charging interface and is used for supplying power. Furthermore, the data acquisition and analysis module for the construction multi-source data provided by the invention is characterized in that the video channel input interface comprises a digital video channel input interface, and the logic control chip and the AI chip are integrated with H.265/H.266 encoders to support dynamic resolution switching. Further, the data acquisition and analysis module for the construction multi-source data provided by the invention is characterized in that the video channel input interface comprises an analog video channel input interface, the analog video channel input interface is connected with a programmable gain amplifier and an analog-to-digital converter, the programmable gain amplifier is used for dynamically adjusting signal gain, and the analog-to-digital converter is used for converting an analog signal into a digital signal. Further, the data acquisition and analysis module for the construction multi-source data provided by the invention is characterized in that the front detection unit is an impedance matching component, and the logic control chip judges that the data signals corresponding to the input interfaces of all channels are analog signals or digital signals according to the impedance matching values or handshake protocols corresponding to the input interfaces of all channels. Further, the configuration module is a switching tube connected to each channel input interface of the multi-channel input interface, wherein a control end of the switching tube is connected with the logic control unit, two non-control ends of the switching