CN-118761629-B - Intelligent engineering monitoring method and system based on multi-sensor information correction

Abstract

The invention discloses an intelligent engineering monitoring method and system based on multi-sensor information correction, wherein the method comprises the steps of acquiring a plurality of pieces of sensor information of a target engineering area through a plurality of sensors; the method comprises the steps of obtaining corresponding correction sensing information by correcting each sensing information according to correction rules among sensing information values, determining a corresponding prediction neural network according to the engineering type corresponding to a target engineering area, and predicting engineering danger information corresponding to the target engineering area according to the prediction neural network and all the correction sensing information. Therefore, the invention can more efficiently and accurately predict and analyze the engineering danger information, improve the engineering danger monitoring effect and provide more accurate data guidance for engineering construction.

Inventors

• LEI HUA
• YUAN YUKANG

Assignees

• 广州城市职业学院

Dates

Publication Date

20260508

Application Date

20240708

Claims (6)

1. 1. An intelligent engineering monitoring method based on multi-sensor information correction is characterized by comprising the following steps: Acquiring a plurality of sensing information of a target engineering area through a plurality of sensors; correcting each piece of sensing information according to a correction rule among the sensing information values to obtain corresponding corrected sensing information; According to the engineering type corresponding to the target engineering area, determining a corresponding prediction neural network comprises the following steps: Inputting each piece of correction sensing information into a trained engineering type classifier network to obtain a corresponding engineering type prediction result, wherein the engineering type classifier network is obtained by training a training data set comprising a plurality of pieces of training sensing information and corresponding engineering type labels; Calculating mode items of all the project type prediction results to obtain the project type corresponding to the target project area, wherein the project type is electric power project, waterway project, decoration project, wall project or building main project; For each candidate neural network, calculating the similarity between the engineering type label and the engineering type in the training data set of the candidate neural network; The candidate neural network with the highest similarity is determined to be a prediction neural network corresponding to the target engineering area, and the prediction neural network is obtained through training a training data set comprising a plurality of training sensing data and corresponding engineering risk type labels; And predicting engineering danger information corresponding to the target engineering area according to the prediction neural network and all the correction sensing information.
2. 2. The intelligent engineering monitoring method based on multi-sensor information correction according to claim 1, wherein the sensor information is temperature sensor information, humidity sensor information, light intensity sensor information, precipitation sensor information, infrared sensor information or image sensor information.
3. 3. The intelligent engineering monitoring method based on multi-sensor information correction according to claim 1, wherein the correcting each sensor information according to the correction rule between the sensor information values to obtain the corresponding corrected sensor information comprises: For any two pieces of sensing information, determining a correlation parameter between information types of the two pieces of sensing information according to a preset correlation relation between the information types; establishing an information value mathematical relationship model of the two pieces of sensing information according to historical data corresponding to the information types of the two pieces of sensing information; and determining the corrected sensing information corresponding to the two sensing information according to the information value mathematical relationship model and the association degree parameter.
4. 4. The intelligent engineering monitoring method based on multi-sensor information correction according to claim 1, wherein predicting engineering risk information corresponding to the target engineering area according to the prediction neural network and all the corrected sensor information comprises: inputting each piece of correction sensing information into the prediction neural network to obtain a risk prediction value and a risk type corresponding to each piece of correction sensing information; judging whether the risk prediction value corresponding to the correction sensing information is larger than a preset threshold value or not for each correction sensing information, and if so, determining the correction sensing information as risk sensing information; And determining the position of the area corresponding to each risk sensing information and the corresponding risk type as engineering risk information corresponding to the target engineering area.
5. 5. An intelligent engineering monitoring system based on multi-sensor information correction, the system comprising: The acquisition module is used for acquiring a plurality of sensing information of the target engineering area through a plurality of sensors; The correction module is used for correcting each piece of sensing information according to the correction rule among the sensing information values to obtain corresponding corrected sensing information; the determining module is configured to determine, according to the engineering type corresponding to the target engineering area, a corresponding prediction neural network, and includes: Inputting each piece of correction sensing information into a trained engineering type classifier network to obtain a corresponding engineering type prediction result, wherein the engineering type classifier network is obtained by training a training data set comprising a plurality of pieces of training sensing information and corresponding engineering type labels; Calculating mode items of all the project type prediction results to obtain the project type corresponding to the target project area, wherein the project type is electric power project, waterway project, decoration project, wall project or building main project; For each candidate neural network, calculating the similarity between the engineering type label and the engineering type in the training data set of the candidate neural network; The candidate neural network with the highest similarity is determined to be a prediction neural network corresponding to the target engineering area, and the prediction neural network is obtained through training a training data set comprising a plurality of training sensing data and corresponding engineering risk type labels; And the prediction module is used for predicting engineering danger information corresponding to the target engineering area according to the prediction neural network and all the correction sensing information.
6. 6. An intelligent engineering monitoring system based on multi-sensor information correction, the system comprising: A memory storing executable program code; A processor coupled to the memory; the processor invokes the executable program code stored in the memory to perform the smart engineering monitoring method based on multi-sensor information correction as claimed in any one of claims 1-4.

Description

Intelligent engineering monitoring method and system based on multi-sensor information correction Technical Field The invention relates to the technical field of data processing, in particular to an intelligent engineering monitoring method and system based on multi-sensor information correction. Background Along with the increasing building demand, building engineering projects are also increasing, and in particular, the building engineering projects involve a plurality of factors such as personnel, machines, materials, environment and the like, the monitoring difficulty is high, the existing manufacturers start to attempt to introduce intelligent engineering concepts to realize intelligent monitoring, but the prior art generally only adopts single sensing information to monitor, engineering monitoring is not realized based on mutual correction and combined action among various sensing information, and the monitoring effect is poor. It can be seen that the prior art has defects and needs to be solved. Disclosure of Invention The technical problem to be solved by the invention is to provide the intelligent engineering monitoring method and system based on multi-sensor information correction, which can more efficiently and accurately predict and analyze engineering danger information, improve engineering danger monitoring effect and provide more accurate data guidance for engineering construction. In order to solve the technical problems, a first aspect of the present invention discloses an intelligent engineering monitoring method based on multi-sensor information correction, the method comprising: Acquiring a plurality of sensing information of a target engineering area through a plurality of sensors; correcting each piece of sensing information according to a correction rule among the sensing information values to obtain corresponding corrected sensing information; Determining a corresponding prediction neural network according to the engineering type corresponding to the target engineering region; And predicting engineering danger information corresponding to the target engineering area according to the prediction neural network and all the correction sensing information. As an optional implementation manner, in the first aspect of the present invention, the sensing information is temperature sensing information, humidity sensing information, light intensity sensing information, precipitation sensing information, infrared sensing information or image sensing information. As an alternative embodiment, in the first aspect of the present invention, the engineering type is electric power engineering, waterway engineering, decoration engineering, wall engineering or construction main engineering. In a first aspect of the present invention, the correcting each piece of sensing information according to the correction rule between the sensing information values to obtain corresponding corrected sensing information includes: For any two pieces of sensing information, determining a correlation parameter between information types of the two pieces of sensing information according to a preset correlation relation between the information types; establishing an information value mathematical relationship model of the two pieces of sensing information according to historical data corresponding to the information types of the two pieces of sensing information; and determining the corrected sensing information corresponding to the two sensing information according to the information value mathematical relationship model and the association degree parameter. In a first aspect of the present invention, the correction sensing information corresponding to the two sensing information is determined according to the information value mathematical relationship model and the association degree parameter. As an optional implementation manner, in the first aspect of the present invention, the determining, according to the engineering type corresponding to the target engineering area, a corresponding predicted neural network includes: determining the engineering type corresponding to the target engineering area; For each candidate neural network, calculating the similarity between the engineering type label and the engineering type in the training data set of the candidate neural network; And determining the candidate neural network with the highest similarity as a prediction neural network corresponding to the target engineering region, wherein the prediction neural network is obtained through training a training data set comprising a plurality of training sensing data and corresponding engineering risk type labels. As an optional implementation manner, in the first aspect of the present invention, the determining the engineering type corresponding to the target engineering area includes: Inputting each piece of correction sensing information into a trained engineering type classifier network to obtain a corresponding engineering t