Intelligent power plant infrastructure settlement early warning system based on BIM + GIS and multi-source data fusion

Abstract

An intelligent power plant infrastructure settlement early warning system based on BIM + GIS and multi-source data fusion comprises a high-precision Internet of Things sensing settlement monitoring module, a BIM + GIS three-dimensional model construction module, a multi-source data fusion analysis module, a settlement early warning model construction module and a virtual simulation module. The method has the following beneficial effects: 1, comprehensive coverage monitoring: high-precision Internet of Things sensing settlement monitoring sensors are densely arranged at key parts of a power plant building and an underground pipeline, and a three-dimensional model of the power plant is constructed in combination with a BIM + GIS technology, so that comprehensive monitoring of settlement of the infrastructure of the power plant is realized, and a monitoring blind area is effectively avoided; 2, real-time accurate monitoring: data acquired by the sensor is analyzed in real time by using a computer algorithm, and the sampling frequency of the sensor is dynamically adjusted according to the importance and historical settlement conditions of different parts, so that the real-time performance and accuracy of settlement monitoring are improved;

Description

Smart power plant infrastructure settlement early warning system based on BIM+GIS and multi-source data fusion Technical Field

[0001] This invention relates to a smart power plant infrastructure settlement early warning system based on BIM+GIS and multi-source data fusion, which is applicable to digital smart power plant management. Background Technology

[0002] With the rise of a new round of industrial revolution, including the Industrial Internet and artificial intelligence, smart power plants represent a new initiative for my country's power generation enterprises to transform and upgrade, and to cope with energy changes. However, the current system architecture and information system construction of power generation enterprises are struggling to meet the demands of intelligent construction, making the system architecture and intelligent system construction of smart power plants a pressing problem to be solved. As a systems engineering project based on digital models, smart power plants influence the success or failure of future digital and intelligent construction for power generation enterprises.

[0003] Currently, power companies face a complex and challenging operating environment and are at a critical juncture of transformation. As technology-intensive enterprises supporting the economic and social operation, thermal power companies must adapt to the times and accelerate their own information technology transformation and upgrading to better achieve the goals of safety, reliability, efficiency, environmental protection, scientific management, and standardized intelligence. The construction of smart power plants is a crucial breakthrough direction for power generation companies in their digital transformation and a concentrated manifestation of the industrial internet concept in the power generation field.

[0004] With the development of the power industry, the stability and safety of power plant infrastructure are receiving increasing attention. Settlement issues in power plant infrastructure, if not monitored and addressed promptly, can lead to equipment damage or even safety accidents, causing significant economic losses and environmental impacts. Therefore, how to achieve real-time monitoring and early warning of power plant infrastructure settlement has become an important research direction in the power industry.

[0005] Currently, settlement monitoring of power plant infrastructure mainly relies on traditional manual inspections and simple sensor monitoring. Manual inspections are inefficient, cannot achieve real-time monitoring, and are greatly affected by human factors. While simple sensor monitoring can achieve some real-time monitoring, its monitoring range is limited, and its data collection and analysis capabilities are insufficient, making it difficult to achieve comprehensive and accurate settlement monitoring and early warning.

[0006] While existing technologies can monitor settlement of power plant infrastructure to some extent, several problems and shortcomings remain. First, current monitoring methods cannot achieve comprehensive coverage of power plant infrastructure, resulting in numerous blind spots and incomplete settlement monitoring. Second, existing data acquisition and analysis capabilities are insufficient, making real-time and accurate settlement monitoring and early warning difficult. Furthermore, existing technical solutions have limitations in practical applications and cannot meet the settlement monitoring needs of different power plant infrastructures. Therefore, developing a novel power plant infrastructure settlement monitoring and early warning system has significant practical importance and application value. Summary of the Invention

[0007] The technical problems to be solved by this invention are: 1) the incomplete coverage of power plant infrastructure settlement monitoring and the existence of monitoring blind spots; 2) the insufficient data acquisition and analysis capabilities of existing monitoring methods, making it difficult to achieve real-time and accurate settlement monitoring and early warning; 3) the inability of existing technical solutions to meet the settlement monitoring needs of different power plant infrastructures and the large limitations of their application. Therefore, this invention provides a smart power plant infrastructure settlement early warning system based on BIM+GIS and multi-source data fusion.

[0008] To address the aforementioned problems, this invention is implemented through the following technical solution: a smart power plant infrastructure settlement early warning system based on BIM+GIS and multi-source data fusion, comprising: a high-precision IoT sensor settlement monitoring module, a BIM+GIS 3D model construction module, a multi-source data fusion analysis module, a settlement early warning model construction module, and a virtual simulation module. The implemented early warning method includes the following steps: Step 1: The high-precision IoT sensor settlement monitoring module includes settlement monitoring sensors, a data acquisition unit, and a communication module; the settlement monitoring sensors are located at key locations in the power plant buildings and underground pipelines, and the settlement monitoring sensors transmit the collected settlement data to the data processing center in real time via the communication module through the data acquisition unit; Step 2: A BIM+GIS 3D model is established in the BIM+GIS 3D model construction module. First, BIM modeling software is used to construct 3D models of the power plant buildings and underground pipelines, and the location information of the settlement monitoring sensors is accurately marked in the 3D model; then, GIS software is used to integrate geographic information... Step 3: Collect historical settlement data and environmental data of the power plant from the high-precision IoT sensor settlement monitoring module, and train the 3D model of the power plant infrastructure established in Step 2; Step 4: Set different settlement warning thresholds for each settlement monitoring sensor according to the importance of different parts of the infrastructure and historical settlement conditions. When the settlement data collected by the settlement monitoring sensor exceeds the warning threshold, the system will immediately issue an alarm; Step 5: Use InSAR technology to obtain large-scale surface deformation data and Beidou positioning communication to obtain power plant facility location data. Combine the settlement data, large-scale surface deformation data and power plant facility location data into a fusion analysis to obtain future settlement prediction values; Step 6: Input the settlement monitoring data and future settlement prediction values ​​into the power plant BIM+GIS 3D model constructed in Step 2 in real time; Use computer simulation technology to dynamically simulate the settlement of the power plant infrastructure, flood control and drainage effects, and the operating status of key equipment under different operating conditions.

[0009] Step 4 specifically includes: Step 4.1: Analyzing the settlement data collected by the settlement monitoring sensors to obtain the settlement rate V and settlement acceleration A information, wherein, Where V is the average settlement rate (mm / year), and d i The cumulative settlement (in millimeters) is t for the nth observation. n -t1 represents the total time span (years); V2 and V1 are the average settlement rates for two different time periods, and t2-t1 is the time interval between the two time periods; Step 4.2: Establish a settlement early warning model using machine learning algorithms, and set different settlement thresholds based on the design standards and historical data of different infrastructures of the power plant; Step 4.3: Monitor settlement data in real time. When the monitored data reaches or exceeds the threshold, the system automatically sends early warning information to the mobile terminals of relevant personnel through Beidou positioning and communication technology.

[0010] Step 5 specifically includes: Step 5.1: In the multi-source data fusion analysis module, preprocess the settlement data, large-scale surface deformation data, and power plant facility location data respectively; Step 5.2: Then, construct a deep neural network model through a deep learning fusion algorithm. This model contains multiple hidden layers and is trained with a large amount of historical data to learn the inherent relationships and characteristics between data from different data sources, outputting accurate fused settlement information; Specifically: Step 5.2.1: Data preprocessing and alignment: Unify the settlement data, large-scale surface deformation data, and power plant facility location data to the same spatiotemporal reference; Step 5.2.2: Primary fusion - Deformation field optimization: Use spatiotemporal Kriging or Kalman filtering to fuse multi-track large-scale surface deformation data and power plant facility location data to generate a more reliable and accurate three-dimensional deformation field; Step 5.2.3: Feature extraction: Extract settlement rate, settlement acceleration, and cumulative settlement features from the optimized three-dimensional deformation field; Step 5.2.4: Advanced fusion - Prediction and Diagnosis: Input all extracted features into a multi-branch deep learning model. The multi-branch deep learning model automatically learns the relationship between deformation and driving factors and outputs the predicted value of future settlement. Step 5.2.5: Decision Fusion - Early Warning Generation: Combine the predicted value of future settlement output by the model with the expert knowledge base and management rules.

[0011] Compared with the prior art, the present invention has the following beneficial effects: 1. Comprehensive monitoring coverage: By densely deploying high-precision IoT-based settlement monitoring sensors in key parts of power plant buildings and underground pipelines, and combining BIM + GIS technology to construct a three-dimensional model of the power plant, comprehensive monitoring of the settlement of power plant infrastructure is realized, effectively avoiding monitoring blind spots.

[0012] 2. Real-time and accurate monitoring: Computer algorithms are used to analyze the data collected by the sensors in real time, and the sampling frequency of the sensors is dynamically adjusted according to the importance of different parts and historical settlement conditions, which improves the real-time performance and accuracy of settlement monitoring.

[0013] 3. Intelligent early warning: By combining settlement data collected by InSAR, measurement robots and IoT sensing technologies, a settlement early warning model is established using machine learning algorithms. When the monitored data reaches or exceeds the threshold, the system automatically sends an early warning message, improving the timeliness and accuracy of the early warning.

[0014] 4. Refined Management: A virtual simulation model of the power plant is built based on BIM + GIS technology. Settlement monitoring data, flood control and drainage data, and key equipment operation data are input into the model in real time. Computer simulation technology is used to dynamically simulate the settlement of power plant infrastructure, flood control and drainage effects, and the operating status of key equipment under different operating conditions, thus realizing refined management of the power plant.

[0015] 5. High adaptability: Based on the differences in geological conditions in different regions, the density of sensor deployment can be appropriately increased in geologically unstable areas, thereby improving the adaptability of the monitoring scheme.

[0016] 6. Data Fusion: The data fusion analysis combines the settlement data collected by sensors with the large-scale surface deformation data obtained by InSAR technology and the location data obtained by BeiDou positioning and communication, which improves the accuracy and comprehensiveness of settlement monitoring.

[0017] 7. Deep Learning Optimization: The adoption of a deep learning-based multi-source data fusion algorithm improves the accuracy and comprehensiveness of settlement monitoring.

[0018] 8. Distributed computing: The distributed computing architecture is used to process the data collected by the sensors, which improves the efficiency and stability of data processing. Attached Figure Description

[0019] Figure 1 is a flowchart of the present invention; Figure 2 is a flowchart of BIM+GIS 3D model construction; Figure 3 is a flowchart of multi-source data fusion analysis; Figure 4 is a flowchart of settlement early warning model operation. Detailed Implementation

[0020] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0021] In the description of this invention, it should be understood that the terms "upper", "lower", "front", "rear", "left", "right", "top", "bottom", "inner", "outer", etc., indicate the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings. They are only for the convenience of describing this invention and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation. Therefore, they should not be construed as limitations on this invention.

[0022] As shown in Figure 1, the smart power plant infrastructure settlement early warning system based on BIM+GIS and multi-source data fusion includes: a high-precision IoT sensor settlement monitoring module, a BIM+GIS 3D model construction module, a multi-source data fusion analysis module, a settlement early warning model construction module, and a virtual simulation module. The implemented early warning method includes the following steps: Step 1: The high-precision IoT sensor settlement monitoring module includes settlement monitoring sensors, a data acquisition unit, and a communication module. The settlement monitoring sensors are located at key locations in the power plant buildings and underground pipelines, such as the four corners of buildings, bends in pipelines, and interfaces. The settlement monitoring sensors transmit the collected settlement data to the data processing center in real time via the communication module through the data acquisition unit. It should be noted that the density of the settlement monitoring sensors is adjusted according to the geological conditions of different areas of the power plant; the density of sensors is appropriately increased in areas with less stable geology.

[0023] Step 2: Create a BIM+GIS 3D model in the BIM+GIS 3D model building module, as shown in Figure 2. First, use BIM modeling software to build a 3D model of the power plant building and underground pipelines, and accurately mark the location information of the settlement monitoring sensors in the 3D model. Then, use GIS software to integrate geographic information to form a complete 3D model of the power plant infrastructure.

[0024] Step 3: Collect historical settlement data and environmental data of the power plant, such as rainfall and temperature, from the high-precision IoT-based settlement monitoring module. Use this data to train the 3D model of the power plant infrastructure established in Step 2. During real-time monitoring, the model can more accurately predict settlement trends based on current environmental and settlement data, and dynamically adjust the sampling frequency of the settlement monitoring sensors.

[0025] Step 4: Based on the importance of different parts of the infrastructure and historical settlement data, set different settlement warning thresholds for each settlement monitoring sensor. When the settlement data collected by the settlement monitoring sensor exceeds the warning threshold, the system immediately issues an alarm, and the alarm information is sent to relevant management personnel via SMS, email, etc. Simultaneously, the areas experiencing settlement anomalies are marked with a prominent color in the 3D model of the power plant infrastructure. The specific process is shown in Figure 4: Step 4.1: Analyze the settlement data collected by the settlement monitoring sensor to obtain information such as settlement rate V and settlement acceleration A. Where V is the average settlement rate (mm / year), and d i The cumulative settlement (in millimeters) is t for the nth observation. n -t1 represents the total time span (years); V2 and V1 are the average settlement rates over two different time periods, and t2-t1 is the time interval between the two time periods. For example: A≈0: uniform settlement; A>0: accelerated settlement, requiring high vigilance and possibly triggering a higher-level warning; A<0: decelerated settlement.

[0026] Step 4.2: Establish a settlement early warning model using machine learning algorithms, and set different settlement thresholds based on the design standards and historical data of different infrastructures in the power plant; Step 4.3: Monitor settlement data in real time, and when the monitored data reaches or exceeds the threshold, the system automatically sends early warning information to the mobile terminals of relevant personnel through Beidou positioning and communication technology.

[0027] Step 5: As shown in Figure 3, the large-scale surface deformation data obtained through InSAR technology and the power plant facility location data obtained through BeiDou positioning and communication are fused together. The data fusion algorithm integrates the advantages of different data sources to improve the accuracy and comprehensiveness of settlement monitoring. The specific process is as follows: Step 5.1: In the multi-source data fusion analysis module, the settlement data, large-scale surface deformation data, and power plant facility location data are preprocessed, including data cleaning and normalization. Step 5.2: Then, a deep neural network model is constructed using a deep learning fusion algorithm. This model contains multiple hidden layers and is trained with a large amount of historical data to learn the inherent relationships and characteristics between different data sources, outputting accurate fused settlement information. In practical applications, real-time data is input into the trained model, and the model outputs fused accurate settlement information, improving the accuracy and comprehensiveness of settlement monitoring. Specifically: Step 5.2.1: Data preprocessing and alignment: Settlement data, large-scale surface deformation data, and power plant facility location data are unified to the same spatiotemporal reference (same coordinate system, same timestamp, same grid); Step 5.2.2: Primary fusion - deformation field optimization: Using spatiotemporal kriging or Kalman filtering, multi-track (ascending and descending tracks) large-scale surface deformation data and power plant facility location data are fused to generate a more reliable and accurate three-dimensional deformation field.

[0028] Step 5.2.3: Feature Extraction: Extract features such as settlement rate, settlement acceleration, and cumulative settlement from the optimized 3D deformation field. Even better, extract features such as water level changes, soil and rock parameters, and building loads from auxiliary data.

[0029] Step 5.2.4: Advanced Fusion - Prediction and Diagnosis: Input all extracted features into a multi-branch deep learning model. The multi-branch deep learning model automatically learns the relationship between deformation and driving factors and outputs the predicted value of future settlement.

[0030] Step 5.2.5: Decision Fusion - Early Warning Generation: Combine the future settlement prediction values ​​output by the model with the expert knowledge base and management rules.

[0031] Step 6: In the power plant BIM+GIS 3D model constructed in Step 2, input settlement monitoring data, flood control and drainage data, and key equipment operation data into the model in real time. Using computer simulation technology, dynamically simulate the settlement of the power plant infrastructure, the effectiveness of flood control and drainage, and the operating status of key equipment under different operating conditions. Through virtual simulation, predict potential problems in advance and formulate corresponding countermeasures to achieve refined management of the power plant.

[0032] Compared with existing technologies, the beneficial effects of this technical solution are as follows: 1. Comprehensive monitoring coverage: By densely deploying high-precision IoT-based settlement monitoring sensors in key parts of power plant buildings and underground pipelines, and combining BIM + GIS technology to construct a three-dimensional model of the power plant, comprehensive monitoring of the settlement of power plant infrastructure is achieved, effectively avoiding monitoring blind spots.

[0033] 2. Real-time and accurate monitoring: Computer algorithms are used to analyze the data collected by the sensors in real time, and the sampling frequency of the sensors is dynamically adjusted according to the importance of different parts and historical settlement conditions, which improves the real-time performance and accuracy of settlement monitoring.

[0034] 3. Intelligent early warning: By combining settlement data collected by InSAR, measurement robots and IoT sensing technologies, a settlement early warning model is established using machine learning algorithms. When the monitored data reaches or exceeds the threshold, the system automatically sends an early warning message, improving the timeliness and accuracy of the early warning.

[0035] 4. Refined Management: A virtual simulation model of the power plant is built based on BIM + GIS technology. Settlement monitoring data, flood control and drainage data, and key equipment operation data are input into the model in real time. Computer simulation technology is used to dynamically simulate the settlement of power plant infrastructure, flood control and drainage effects, and the operating status of key equipment under different operating conditions, thus realizing refined management of the power plant.

[0036] 5. High adaptability: Based on the differences in geological conditions in different regions, the density of sensor deployment can be appropriately increased in geologically unstable areas, thereby improving the adaptability of the monitoring scheme.

[0037] 6. Data Fusion: The data fusion analysis combines the settlement data collected by sensors with the large-scale surface deformation data obtained by InSAR technology and the location data obtained by BeiDou positioning and communication, which improves the accuracy and comprehensiveness of settlement monitoring.

[0038] 7. Deep Learning Optimization: The adoption of a deep learning-based multi-source data fusion algorithm improves the accuracy and comprehensiveness of settlement monitoring.

[0039] 8. Distributed computing: The distributed computing architecture is used to process the data collected by the sensors, which improves the efficiency and stability of data processing.

[0040] The above description is only a preferred embodiment of the present invention. It should be noted that those skilled in the art can make several changes and improvements without departing from the overall concept of the present invention, and these should also be considered within the scope of protection of the present invention.

Claims

1. A smart power plant infrastructure settlement early warning system based on BIM+GIS and multi-source data fusion, characterized in that: include: The high-precision IoT-based settlement monitoring module, BIM+GIS 3D model building module, multi-source data fusion analysis module, settlement early warning model building module, and virtual simulation module implement an early warning method comprising the following steps: Step 1: The high-precision IoT-based settlement monitoring module includes settlement monitoring sensors, a data acquisition unit, and a communication module. The settlement monitoring sensors are located at key locations in the power plant building and underground pipelines. The settlement monitoring sensors transmit the collected settlement data to the data processing center in real time via the communication module through the data acquisition unit. Step 2: A BIM+GIS 3D model is established in the BIM+GIS 3D model building module. First, BIM modeling software is used to construct 3D models of the power plant building and underground pipelines, and the location information of the settlement monitoring sensors is accurately marked in the 3D model. Then, GIS software is used to integrate geographic information to form a complete... Step 3: Collect historical settlement data and environmental data of the power plant from the high-precision IoT sensor settlement monitoring module, and train the 3D model of the power plant infrastructure established in Step 2; Step 4: Set different settlement warning thresholds for each settlement monitoring sensor according to the importance of different parts of the infrastructure and historical settlement conditions. When the settlement data collected by the settlement monitoring sensor exceeds the warning threshold, the system immediately issues an alarm; Step 5: Use InSAR technology to obtain large-scale surface deformation data and Beidou positioning communication to obtain power plant facility location data. Integrate the settlement data, large-scale surface deformation data and power plant facility location data into the fusion analysis to obtain future settlement prediction values; Step 6: Input the settlement monitoring data and future settlement prediction values ​​into the power plant BIM+GIS 3D model constructed in Step 2 in real time; Computer simulation technology is used to dynamically simulate the settlement of power plant infrastructure, flood control and drainage effects, and the operating status of key equipment under different operating conditions.

2. The smart power plant infrastructure settlement early warning system based on BIM+GIS and multi-source data fusion as described in claim 1, characterized in that: Step 4 specifically includes: Step 4.1: Analyzing the settlement data collected by the settlement monitoring sensors to obtain the settlement rate V and settlement acceleration A information, wherein, Where V is the average settlement rate (mm / year), and d i The cumulative settlement (in millimeters) is t for the nth observation. n -t1 represents the total time span (years); V2 and V1 are the average settlement rates for two different time periods, and t2-t1 is the time interval between the two time periods; Step 4.2: Establish a settlement early warning model using machine learning algorithms, and set different settlement thresholds based on the design standards and historical data of different infrastructures of the power plant; Step 4.3: Monitor settlement data in real time. When the monitored data reaches or exceeds the threshold, the system automatically sends early warning information to the mobile terminals of relevant personnel through Beidou positioning and communication technology.

3. The smart power plant infrastructure settlement early warning system based on BIM+GIS and multi-source data fusion as described in claim 1, characterized in that: Step 5 specifically includes: Step 5.1: In the multi-source data fusion analysis module, preprocess the settlement data, large-scale surface deformation data, and power plant facility location data respectively; Step 5.2: Then, construct a deep neural network model through a deep learning fusion algorithm. This model contains multiple hidden layers and is trained with a large amount of historical data to learn the inherent relationships and characteristics between data from different data sources, outputting accurate fused settlement information; Specifically: Step 5.2.1: Data preprocessing and alignment: Unify the settlement data, large-scale surface deformation data, and power plant facility location data to the same spatiotemporal reference; Step 5.2.2: Primary fusion - Deformation field optimization: Use spatiotemporal Kriging or Kalman filtering to fuse multi-track large-scale surface deformation data and power plant facility location data to generate a more reliable and accurate three-dimensional deformation field; Step 5.2.3: Feature extraction: Extract settlement rate, settlement acceleration, and cumulative settlement features from the optimized three-dimensional deformation field; Step 5.2.4: Advanced fusion - Prediction and Diagnosis: Input all extracted features into a multi-branch deep learning model. The multi-branch deep learning model automatically learns the relationship between deformation and driving factors and outputs the predicted value of future settlement. Step 5.2.5: Decision Fusion - Early Warning Generation: Combine the predicted value of future settlement output by the model with the expert knowledge base and management rules.

Images