CN-122016088-A - Asphalt panel temperature calibration model training and calibrating method, system and device

CN122016088ACN 122016088 ACN122016088 ACN 122016088ACN-122016088-A

Abstract

The invention provides a training method of an asphalt panel temperature calibration model, which comprises the steps of firstly adopting a training sample set to respectively train and obtain a first calibration model and a second calibration model, then cascading the first calibration model and the second calibration model obtained by training to obtain an initial temperature calibration model, and finally training the initial temperature calibration model to obtain the temperature calibration model. The temperature calibration model obtained through training in a grading training mode is higher in accuracy and is more suitable for temperature calibration of the seepage-proofing panel. Meanwhile, the invention also provides a temperature calibration method by using the trained temperature calibration model, through the temperature calibration method, the temperature of different depths can be calibrated aiming at panels with different collecting equipment, different collecting environments, different composition components and structures, the temperature distribution of the longitudinal depths of the panels is obtained, the temperature measurement efficiency is improved, and the construction quality and safety of the asphalt concrete panels are ensured.

Inventors

ZHU GUOJIN
ZHAO JINGYA
LI ZHICHAO
LING LEI
WANG ZHONG
DING QI
LI XINDA
LI JINMIN
MA LONG
SONG CHAO
ZHAO YUEDONG
HU LIANG
WANG XINYAO
TIAN ZHENXING
WANG XING

Assignees

中国电建集团北京勘测设计研究院有限公司

Dates

Publication Date: 20260512
Application Date: 20260130

Claims (8)

1. An asphalt panel temperature calibration model training method is characterized by comprising the following steps: The method for constructing the initial first calibration model and the initial second calibration model specifically comprises the following steps: Acquiring a first training sample set, wherein each first training sample in the first training sample set comprises a first parameter matrix, a first measured temperature and a first real temperature; acquiring a second training sample set, wherein each second training sample in the second sample set comprises a second parameter matrix and a second real temperature, and the second parameter matrix indicates information related to temperature calibration; Training an initial first calibration model by adopting the first parameter matrix, the first measured temperature and the first real temperature to obtain a first calibration model; Training an initial second calibration model by adopting the second parameter matrix, the second real temperature and the first real temperature to obtain a second calibration model; Cascading the first calibration model and the second calibration model; and training a first calibration model and a second calibration model which are cascaded by adopting the first parameter matrix, the second parameter matrix, the first measured temperature and the second real temperature to obtain a temperature calibration model.
2. The method for training the asphalt panel temperature calibration model according to claim 1, wherein the first parameter matrix comprises a first acquisition equipment parameter and a first environment parameter, the first acquisition equipment parameter at least comprises equipment height, the first environment parameter at least comprises environment temperature, and the second parameter matrix at least comprises aggregate type, aggregate proportion, asphalt content, mineral powder content and target calibration depth of the panel.
3. The method for training the temperature calibration model of the asphalt panel according to claim 1, wherein the training of an initial first calibration model by using the first parameter matrix, the first measured temperature and the first real temperature is performed to obtain a first calibration model, and specifically comprises the steps of inputting the first parameter matrix and the first measured temperature into the initial first calibration model to obtain a first model output; The method comprises the steps of training an initial second calibration model by adopting the second parameter matrix, the second real temperature and the first real temperature to obtain a second calibration model, and specifically comprises the steps of inputting the second parameter matrix and the first real temperature to the initial second calibration model to obtain a second model output, judging whether to stop model training according to the distance between the second model output and the second real temperature, and stopping model training when the distance between the second model output and the second real temperature is smaller than a preset value, so that the second calibration model is obtained.
4. The method for training the asphalt panel temperature calibration model according to claim 1, wherein the first calibration model and the second calibration model are cascaded, and the first calibration model and the second calibration model are trained by adopting the first parameter matrix, the second parameter matrix, the first measured temperature and the second real temperature to obtain the temperature calibration model, specifically comprising the following steps: Cascading the first calibration model and the second calibration model to obtain an initial temperature calibration model; And judging whether to stop model training according to the distance between the third model output and the second real temperature, and stopping model training when the distance between the third model output and the second real temperature is smaller than a preset value, thereby obtaining the temperature calibration model.
5. A method of calibrating asphalt panel temperature using the temperature calibration model of any of claims 1-4, the method comprising: The method comprises the steps of collecting a third parameter matrix, a fourth parameter matrix and a second measured temperature, wherein the third parameter matrix indicates information related to temperature collection, and the fourth parameter matrix indicates information related to temperature calibration; And inputting the third parameter matrix, the fourth parameter matrix and the second measured temperature into a temperature calibration model to obtain temperature distribution at different depths of the panel.
6. The method for calibrating a temperature calibration model of an asphalt panel according to claim 5, further comprising: acquiring a digital model of the panel; and according to the temperature distribution, carrying out visual display of the temperature field in the digital model.
7. An asphalt panel temperature calibration model training system, the system comprising: The construction module is used for constructing an initial first calibration model and an initial second calibration model; the system comprises a first acquisition module, a second acquisition module and a first control module, wherein the first acquisition module is used for acquiring a first training sample set, and each training sample in the first training sample set comprises a first parameter matrix, a first measured temperature and a first real temperature; The system comprises a first acquisition module, a second acquisition module and a temperature calibration module, wherein the first acquisition module is used for acquiring a first training sample set, and each training sample in the first training sample set comprises a first parameter matrix and a first real temperature; The first training module is used for training an initial first calibration model by adopting the first parameter matrix, the first measured temperature and the first real temperature to obtain a first calibration model; The second training module is used for training an initial second calibration model by adopting the second parameter matrix, the second real temperature and the first real temperature to obtain a second calibration model; the cascade module is used for cascading the first calibration model and the second calibration model; and the third training module is used for training the cascaded first calibration model and second calibration model by adopting the first parameter matrix, the second parameter matrix, the first measured temperature and the second real temperature to obtain a temperature calibration model.
8. An apparatus for calibrating asphalt panel construction temperature using the temperature calibration model of claims 1-4, the apparatus comprising: The system comprises an acquisition module, a temperature calibration module and a temperature calibration module, wherein the acquisition module is used for acquiring a third parameter matrix, a fourth parameter matrix and a second measured temperature, wherein the third parameter matrix indicates information related to temperature acquisition, and the fourth parameter matrix indicates information related to temperature calibration; The calibration module is used for inputting the third parameter matrix, the fourth parameter matrix and the second measured temperature into a temperature calibration model to obtain temperature distribution at different depths of the panel; and the visualization module is used for acquiring a digital model of the panel and carrying out visual display of a temperature field in the digital model according to the temperature distribution.

Description

Asphalt panel temperature calibration model training and calibrating method, system and device Technical Field The invention belongs to the technical field of hydraulic and hydroelectric engineering construction, and particularly relates to a method, a system and a device for training and calibrating a temperature calibration model of an asphalt panel. Background The temperature control is a core control link of the construction of the asphalt concrete anti-seepage panel of the hydraulic and hydroelectric engineering, and is directly related to the construction quality of the panel and the stability of the anti-seepage function, namely, the panel is easy to shrink and crack in the later period when the temperature is too high, the aggregate bonding compactness is insufficient when the temperature is too low, the anti-seepage effect of the panel is obviously reduced, and the service life of the panel is shortened. Therefore, accurate temperature measurement is implemented in key construction processes such as paving, rolling and the like of the asphalt concrete seepage-proofing panel, and the method is a necessary premise for guaranteeing engineering quality. The temperature measurement modes of the current main stream mainly fall into two categories: (1) Manual insertion measurement, in which the measurement is completed by manually inserting a thermometer into the interior of the panel to a specified depth. The method has the core advantages that temperature data of the appointed depth inside the panel can be directly obtained, the measurement accuracy is high, however, a remarkable short plate exists, real-time and continuous measurement cannot be realized, the measurement efficiency is low, the whole area of the panel is difficult to cover, a local temperature abnormal area is easy to miss, meanwhile, the environment temperature of a construction site is usually 120-180 ℃, and the manual operation has extremely high scalding safety risk. (2) IoT infrared sensing measurement, along with the development of internet of things (IoT) technology, part of engineering attempts to install an infrared temperature sensor at a certain height from a panel on a paving machine, and infrared radiation signals on the surface of the panel are collected in real time through the sensor, so that non-contact real-time automatic measurement is realized. The method can cover all areas of the panel, greatly improves the temperature measurement efficiency, has outstanding technical limitations, can only acquire the surface temperature of the panel, cannot acquire the temperature distribution of the panel in the vertical direction, is inaccurate, is not enough to truly reflect the temperature state of key control points of the construction quality of the panel, and finally restricts the large-scale popularization and application of the technology in the construction of asphalt concrete anti-seepage panels. The existing asphalt concrete anti-seepage panel construction temperature measurement technology has core technical pain points that the high-efficiency construction requirement is difficult to adapt due to the fact that the real-time performance and the continuity are insufficient and the safety risk is high, the temperature distribution of the panel in the vertical direction cannot be obtained due to the fact that the measurement accuracy is insufficient, the accurate quality control is difficult to support, and the practical requirement of engineering on the omnibearing and high-precision monitoring of the construction temperature cannot be met. In view of this, the present invention has been made. Disclosure of Invention In order to solve the technical problems in the prior art, the invention provides a training and calibrating method, a system and a device for an asphalt panel temperature calibration model, which solve the technical problem that the temperature of an asphalt concrete anti-seepage panel in different vertical depths cannot be efficiently and accurately calibrated in the prior art. In order to achieve the above purpose, the technical scheme of the invention is as follows: A training method for an asphalt panel temperature calibration model comprises the following steps: The method for constructing the initial first calibration model and the initial second calibration model specifically comprises the following steps: Acquiring a first training sample set, wherein each first training sample in the first training sample set comprises a first parameter matrix, a first measured temperature and a first real temperature; Acquiring a second training sample set, wherein each second training sample in the second training sample set comprises a second parameter matrix and a second real temperature, and the second parameter matrix indicates information related to temperature calibration; Training an initial first calibration model by adopting the first parameter matrix, the first measured temperature and the first real te