CN-121973322-A - Concrete condensation state prediction and maintenance method based on multi-sensor fusion

CN121973322ACN 121973322 ACN121973322 ACN 121973322ACN-121973322-A

Abstract

The invention relates to the technical field of concrete performance detection, and discloses a concrete condensation state prediction and maintenance method based on multi-sensor fusion. The invention aims to solve the problems that the existing concrete coagulation state judging method is destructive, the real state inside the structure cannot be reflected in real time, and the maintenance measures cannot be dynamically adapted to the change of the actual working conditions. The invention collects data in real time by arranging multiple sensors in the concrete, inputs the data into a long-short-period memory network model after pretreatment, predicts the coagulation state and early strength, and automatically controls maintenance equipment according to the prediction result. The method also utilizes the final measured intensity to perform feedback optimization on the model. The method realizes nondestructive, real-time monitoring and accurate prediction of the internal state of the concrete, can carry out automatic and refined closed-loop regulation and control on maintenance operation, and improves the scientificity and efficiency of construction management.

Inventors

XU XIAOHUI
Lv Qianghua
JI FEI
BAO YANG
CHEN GANG
MEI JIANGTAO
XIAO HAN
YUAN CHENG
MA HUAIZHANG
QUAN YOUWEI
WANG JINGXIONG
CHEN YONGHUI

Assignees

中建八局第三建设有限公司

Dates

Publication Date: 20260505
Application Date: 20251219

Claims (10)

1. The concrete setting state prediction and maintenance method based on multi-sensor fusion is characterized by comprising the following steps of: s1, before concrete pouring, a sensing unit combination comprising a temperature sensor, a resistivity sensor and an ultrasonic transceiver probe is deployed inside a concrete structure, and temperature data, resistivity data and ultrasonic propagation data reflecting the hydration process of the concrete are collected in real time; S2, preprocessing temperature data, resistivity data and ultrasonic propagation data acquired in real time, wherein the preprocessing comprises noise filtering and data normalization; s3, inputting the preprocessed data into a pre-trained long-period and short-period memory network model, and outputting a predicted result of the coagulation state and early strength of the concrete in real time by the long-period and short-period memory network model; S4, controlling curing equipment associated with the concrete structure to perform intelligent curing operation according to the prediction result; and S5, drilling a core sample of the concrete structure after the engineering is finished, performing compressive strength test to determine an actual strength value, comparing the actual strength test value with the prediction result, and retraining the long-short-period memory network model by using the comparison result to optimize the prediction performance.
2. The method for predicting and maintaining a concrete setting state based on multi-sensor fusion according to claim 1, wherein the combination of sensor units is deployed inside a concrete structure, specifically, in a center position with a thickest cross section of the concrete structure.
3. The method for predicting and maintaining the coagulation status of concrete based on multi-sensor fusion according to claim 2, wherein the deployment further comprises binding and fixing the sensing unit combination on the steel bars before pouring the concrete, and performing waterproof sealing treatment on the junction and the cable.
4. The method for predicting and maintaining a concrete setting state based on multi-sensor fusion according to claim 1, wherein the noise filtering adopts a smoothing filtering algorithm, and the data normalization adopts a minimum-maximum normalization process.
5. The method for predicting and maintaining the coagulation status of concrete based on multi-sensor fusion according to claim 1, wherein the pre-training comprises the steps of synchronously collecting sensor data of concrete test blocks in a laboratory, and performing supervised learning training on the long-term and short-term memory network model by using coagulation time data measured by a penetration resistance meter as a label.
6. The method for predicting and curing a concrete setting state based on multi-sensor fusion according to claim 1, wherein the prediction results comprise qualitative judgment of the current setting stage, quantitative prediction of the remaining time to reach the initial setting and final setting state, and prediction curves for future strength development.
7. The method for predicting and curing the concrete condensation state based on the multi-sensor fusion according to claim 1, wherein the sensing unit combination further comprises a surface humidity sensor, the curing equipment comprises an atomization spraying device, and the intelligent curing operation comprises automatically starting the atomization spraying device when the concrete is predicted to be in a water loss risk stage according to a preset rule and the detection value of the surface humidity sensor is lower than a preset humidity threshold value.
8. The method for predicting and curing the concrete condensation state based on the multi-sensor fusion according to claim 1, wherein the curing equipment comprises an electric heating and heat preserving device, and the intelligent curing operation comprises automatically starting the electric heating and heat preserving device to heat and cure the concrete structure according to a preset rule when the temperature data is lower than a preset minimum curing temperature.
9. The concrete setting state prediction and maintenance method based on multi-sensor fusion according to claim 1, wherein the retraining of the long-short-period memory network model is completed by supplementing the actual strength value and the corresponding sensor data thereof as new samples into an original training set in an incremental learning mode.
10. The method for predicting and curing the coagulation status of concrete based on multi-sensor fusion according to claim 1, wherein the combination of sensing units further comprises a strain sensor for collecting strain data of the concrete in real time to monitor shrinkage and cracking risk thereof.

Description

Concrete condensation state prediction and maintenance method based on multi-sensor fusion Technical Field The invention relates to the technical field of concrete performance detection, in particular to a concrete coagulation state prediction and maintenance method based on multi-sensor fusion. Background In modern civil engineering, concrete is the most basic and core building material. The process of transition from a freshly mixed plastic state to a rigid solid with load-bearing capacity, i.e. the setting and hardening process, has a decisive influence on the construction progress, the structural safety and the long-term durability of the entire project. At present, an engineering site generally depends on a traditional method to judge the coagulation state of concrete, for example, a side test block is manufactured according to the standard requirement, and tests are carried out at fixed time through a penetration resistance meter to determine initial coagulation and final coagulation time, and a method for subjectively judging the coagulation state of concrete depends on personal experience of constructors. Meanwhile, for curing of concrete, fixed age criteria are generally adopted, such as continuous watering and curing for 7 to 14 days after reaching final setting. These established methods, while guaranteeing engineering quality to some extent, have inherent limitations. The actual setting and hardening process of concrete is a complex physicochemical reaction, and the speed is dynamically influenced by various factors such as cement variety, admixture variety, tiny fluctuation of mixing proportion, changeable environment temperature, humidity and the like in a construction site. The traditional immobilized test method and maintenance system can not respond to the changes accurately in real time, so that the person skilled in the art always faces the essential technical problem to be solved in urgent need, namely, how to realize nondestructive, real-time tracking and accurate prediction of the evolution process of the real performance of the concrete in the main structure, and on the basis of the nondestructive, real-time tracking and accurate prediction, the maintenance measures are subjected to dynamic and intelligent closed-loop regulation and control, and the construction efficiency and the resource utilization rate are improved to the greatest extent on the premise of ensuring the engineering quality. Disclosure of Invention The present invention has been made in view of the above-described problems occurring in the prior art. Therefore, in order to solve the problems that the existing concrete condensation state judging method is destructive, the real state inside the structure cannot be reflected in real time, and the maintenance measures cannot be dynamically adapted to the change of the actual working conditions, the invention provides a concrete condensation state prediction and maintenance method based on multi-sensor fusion. In order to solve the technical problems, the invention provides the following technical scheme: in a first aspect, the invention provides a method for predicting and maintaining a concrete setting state based on multi-sensor fusion, which comprises the following steps, S1, before concrete pouring, a sensing unit combination comprising a temperature sensor, a resistivity sensor and an ultrasonic transceiver probe is deployed inside a concrete structure, and temperature data, resistivity data and ultrasonic propagation data reflecting the hydration process of the concrete are collected in real time; S2, preprocessing temperature data, resistivity data and ultrasonic propagation data acquired in real time, wherein the preprocessing comprises noise filtering and data normalization; s3, inputting the preprocessed data into a pre-trained long-period and short-period memory network model, and outputting a predicted result of the coagulation state and early strength of the concrete in real time by the long-period and short-period memory network model; S4, controlling curing equipment associated with the concrete structure to perform intelligent curing operation according to the prediction result; and S5, drilling a core sample of the concrete structure after the engineering is finished, performing compressive strength test to determine an actual strength value, comparing the actual strength test value with the prediction result, and retraining the long-short-period memory network model by using the comparison result to optimize the prediction performance. The concrete condensation state prediction and maintenance method based on multi-sensor fusion is characterized in that a sensing unit combination comprising a temperature sensor, a resistivity sensor and an ultrasonic transceiver probe is deployed inside a concrete structure before concrete pouring, and temperature data, resistivity data and ultrasonic propagation data reflecting a concrete hydration process are collecte