Real-time environmental protection risk early warning method and system for power transmission and transformation project construction period

Abstract

The application provides a power transmission and transformation project construction period environmental protection risk real-time early warning method and system, which comprises the following steps: determining the real-time monitoring focus of a monitoring area according to the project construction progress, configuring corresponding monitoring sensors for the real-time monitoring focus, performing feature analysis on the real-time collection data of the monitoring sensors by using a machine learning algorithm, determining the potential environmental risks of the monitoring area, constructing an early warning model of the monitoring area according to the real-time collection data, supervising the burst probability corresponding to each potential environmental risk, finding the emergency measures corresponding to the dangerous environmental risks with a burst probability higher than a specified probability, constructing an emergency scheme in combination with the real-time collection data of the monitoring area and displaying the emergency scheme, performing real-time monitoring of multi-source environmental protection risks by fusing multi-sensor technology and real-time monitoring technology, adopting high-precision sensors and data analysis technology and real-time risk early warning, realizing rapid analysis and early warning of monitoring data, and timely issuing an early warning signal.

Description

Technical Field

[0001] This invention relates to the fields of environmental protection and power transmission and transformation engineering monitoring technology, and in particular to a method and system for real-time early warning of environmental risks during the construction period of power transmission and transformation projects. Background Technology

[0002] With the development of the national power grid, the capacity of the power grid is increasing, and the voltage level of transmission lines is constantly rising. The construction of power transmission and transformation projects poses a risk of environmental damage to the local environment. To protect the environment, environmental protection regulations for power transmission and transformation projects have been established to ensure that the impact on the local environment is minimized. However, the construction of power transmission and transformation projects may still cause certain impacts on the surrounding environment, such as air pollution, soil erosion, vegetation destruction, and tower foundation disturbance. Existing environmental risk monitoring methods often rely on manual inspections, which suffer from problems such as untimely monitoring and inaccurate data, making it difficult to meet the needs of rapid early warning and effective prevention and control. These methods have the following shortcomings:

[0003] 1. Limited monitoring methods: Current environmental risk monitoring during the construction period of power transmission and transformation projects mainly relies on manual inspections, which is a relatively limited method and cannot achieve comprehensive monitoring of multiple environmental risk factors.

[0004] 2. Poor monitoring timeliness: Manual inspections have problems such as long cycles and slow response, making it difficult to achieve real-time monitoring and resulting in an inability to respond to environmental risk events in a timely manner.

[0005] 3. Low data accuracy: Manual monitoring is greatly affected by subjective factors, resulting in low data accuracy, making it difficult to serve as an effective basis for early warning.

[0006] 4. Lack of early warning system: There is a lack of an integrated early warning system in the current technology, which makes it impossible to quickly analyze and issue early warnings for the collected data.

[0007] 5. Weak emergency response capability: Due to untimely monitoring and early warning, existing technologies are unable to quickly take effective emergency measures when environmental risks occur.

[0008] Therefore, this invention provides a method and system for real-time early warning of environmental risks during the construction period of power transmission and transformation projects. Summary of the Invention

[0009] This invention relates to a method and system for real-time early warning of environmental risks during the construction period of power transmission and transformation projects. By integrating multi-sensor technology and real-time monitoring technology, it enables real-time monitoring of multi-source environmental risks. It employs high-precision sensors and data analysis technology for real-time risk early warning, thereby achieving rapid analysis and early warning of monitoring data and issuing timely warning signals.

[0010] This invention provides a method for real-time early warning of environmental risks during the construction period of power transmission and transformation projects, including:

[0011] Step 1: Determine the key monitoring areas in real time based on the project construction progress, and control the preset mobile monitoring platform to configure the corresponding monitoring sensors for the key monitoring areas in real time;

[0012] Step 2: Acquire the real-time data corresponding to each of the monitoring sensors, and use machine learning algorithms to perform feature analysis on the real-time data to determine the potential environmental risks of the monitoring area;

[0013] Step 3: Construct an early warning model for the monitoring area based on the real-time collected data, and monitor the probability of occurrence corresponding to each potential environmental risk in the early warning model;

[0014] Step 4: Identify emergency measures corresponding to hazardous environmental risks with a probability of occurrence higher than the prescribed probability, construct an emergency plan based on the real-time data collected in the monitoring area, and display it.

[0015] In one feasible approach

[0016] Step 1 includes:

[0017] Step 11: Based on the engineering blueprint of this project, construct the estimated construction progress of this project, determine the construction key areas corresponding to different construction progress, obtain the map of the monitoring area, and determine the environmental layout corresponding to each construction key area.

[0018] Step 12: Determine the current critical construction area based on the current construction progress of this project, and determine several monitoring locations and the monitoring type corresponding to each monitoring location based on the environmental layout corresponding to the current critical construction area.

[0019] Step 13: Construct a sensor configuration scheme according to the monitoring type corresponding to each monitoring location, control the preset mobile monitoring platform to arrive at each monitoring location in sequence according to the sensor configuration scheme, and configure the corresponding monitoring sensor for each monitoring location.

[0020] In one feasible approach

[0021] Also includes:

[0022] When the current construction progress reaches the next stage, a sensor retrieval scheme is constructed based on the sensor configuration scheme. The mobile monitoring platform is controlled to sequentially reach each monitoring location and retrieve the monitoring sensor corresponding to each monitoring location.

[0023] In one feasible approach

[0024] Step 2 includes:

[0025] Step 21: Control the preset drone to conduct regular inspections of the monitoring area, construct an engineering construction image of the monitoring area, acquire real-time data corresponding to each monitoring sensor, and establish dimensional data attributes for each real-time data based on the monitoring dimension corresponding to each monitoring sensor.

[0026] Step 22: Perform cluster analysis on the real-time collected data according to the dimensional data attributes to obtain several monitoring data classes, analyze the data logic information between different monitoring data in each monitoring data class, construct a clustering tree corresponding to each monitoring data class, and determine the current dimensional space corresponding to each clustering tree respectively;

[0027] Step 23: Determine the low-dimensional space corresponding to each clustering tree, compress each clustering tree from the corresponding current dimension space to the corresponding low-dimensional space using a preset autoencoder, and extract the low-dimensional tree features corresponding to each clustering tree in each low-dimensional space.

[0028] Step 24: Construct real-time presentation features corresponding to the real-time collected data based on the low-dimensional tree features corresponding to each clustering tree, construct dynamic presentation information corresponding to each monitoring dimension, perform anomaly analysis on each dynamic presentation information, determine several abnormal dynamics contained in the monitoring area, determine the potential environmental risks of the monitoring area, and mark the risk range corresponding to each potential environmental risk in the engineering construction image and display it.

[0029] In one feasible approach

[0030] Step 24 includes:

[0031] Step 241: Determine several nonlinear features of the corresponding monitoring data class based on the low-dimensional tree features of each clustering tree, construct a nonlinear model of the corresponding monitoring data class using the nonlinear features, run the nonlinear data model to determine the nonlinear structure of each real-time acquired data, and determine the real-time presentation features of each real-time acquired data.

[0032] Step 242: Sort several real-time presentation features corresponding to the same real-time collected data in time sequence to obtain dynamic presentation information corresponding to each monitoring dimension, obtain the dynamic period corresponding to each dynamic presentation information respectively, and establish anomaly screening rules corresponding to each dynamic presentation information by using a preset isolated forest combined with the dynamic period.

[0033] Step 243: Use the anomaly screening rules to screen the anomaly information contained in the dynamic presentation information respectively, construct several overall anomaly features of the monitoring area, analyze the anomaly dimension corresponding to the monitoring area based on the overall anomaly features, analyze the anomaly level corresponding to the monitoring area based on the anomaly feature value corresponding to the overall anomaly features, and determine the potential environmental risks of the monitoring area.

[0034] In one feasible approach

[0035] Step 3 includes:

[0036] Step 31: Obtain the risk monitoring list of the monitoring area, establish several risk monitoring conditions based on the risk monitoring list, and construct an early warning model for the monitoring area based on the risk monitoring conditions and the real-time collected data;

[0037] Step 32: In the early warning model, determine the risk occurrence range corresponding to each potential environmental risk, and perform model sampling on each risk occurrence range to obtain several sampling information of each risk occurrence range;

[0038] Step 33: Construct several risk factors corresponding to the risk occurrence range using the sampling information, run the early warning model, and determine the trigger time corresponding to each risk factor;

[0039] Step 34: Estimate the probability of sudden occurrence of potential environmental risks within the specified construction period based on the time difference between the trigger time and the current time.

[0040] In one feasible approach

[0041] Step 4 includes:

[0042] Step 41: Obtain the hazardous environmental risks in the monitoring area with a probability of sudden occurrence higher than the specified probability, find several emergency measures corresponding to each hazardous environmental risk, input each emergency measure into the early warning model for measure simulation, and obtain the first degree of integration between each emergency measure and the monitoring area;

[0043] Step 42: Based on the first degree of fusion, select several available emergency measures corresponding to each of the hazardous environmental risks, extract one available emergency measure corresponding to each of the hazardous environmental risks, and combine the measures to obtain several risk-resistance combinations;

[0044] Step 43: Simulate each of the risk mitigation combinations using the early warning model to obtain a second degree of integration between each risk mitigation combination and the monitoring area, and select the target risk mitigation combination for the monitoring area based on the second degree of integration;

[0045] Step 44: Determine the current risk value of the monitoring area based on the real-time collected data, adjust the execution starting point of the target risk mitigation combination using the current risk value, obtain the emergency plan for the monitoring area, and display it.

[0046] In one feasible approach

[0047] Also includes:

[0048] When there is a hazardous environmental risk within the monitoring area, the range of the hazardous environmental risk is obtained, real-time early warning information is generated and displayed.

[0049] This invention provides a real-time early warning system for environmental risks during the construction period of power transmission and transformation projects, including:

[0050] The real-time acquisition module is used to determine the real-time monitoring focus of the monitoring area according to the progress of the project construction, and to control the preset mobile monitoring platform to configure the corresponding monitoring sensors for the real-time monitoring focus.

[0051] The risk assessment module is used to acquire real-time data corresponding to each of the monitoring sensors, and to perform feature analysis on the real-time data using machine learning algorithms to determine the potential environmental risks of the monitoring area.

[0052] The monitoring and analysis module is used to construct an early warning model for the monitoring area based on the real-time collected data, and to monitor the probability of occurrence of each potential environmental risk in the early warning model.

[0053] The early warning execution module is used to identify emergency measures corresponding to hazardous environmental risks with a probability of occurrence higher than the specified probability, and to construct and display emergency plans based on real-time data collected from the monitoring area.

[0054] In one feasible approach

[0055] The risk assessment module includes:

[0056] The data acquisition unit is used to control a preset drone to conduct regular inspections of the monitoring area, construct engineering construction images of the monitoring area, acquire real-time data corresponding to each monitoring sensor, and establish dimensional data attributes of each real-time data according to the monitoring dimension corresponding to each monitoring sensor.

[0057] The data processing unit is used to perform cluster analysis on the real-time collected data according to the dimensional data attributes to obtain several monitoring data classes, analyze the data logic information between different monitoring data in each monitoring data class, construct a clustering tree corresponding to each monitoring data class, and determine the current dimensional space corresponding to each clustering tree respectively.

[0058] The data dimensionality reduction unit is used to determine the low-dimensional dimension space corresponding to each clustering tree, compress each clustering tree from the corresponding current dimension space to the corresponding low-dimensional dimension space using a preset autoencoder, and extract the low-dimensional tree features corresponding to each clustering tree in each low-dimensional dimension space.

[0059] The risk determination unit is used to construct real-time presentation features corresponding to real-time collected data based on the low-dimensional tree features corresponding to each clustering tree, construct dynamic presentation information corresponding to each monitoring dimension, perform anomaly analysis on each dynamic presentation information, determine several abnormal dynamics contained in the monitoring area, determine the potential environmental risks of the monitoring area, and mark and display the risk range corresponding to each potential environmental risk in the engineering construction image.

[0060] The beneficial effects of the above technical solution are as follows: In order to achieve rapid analysis and early warning of monitoring data, and to reduce the impact of environmental risks on the environment through intervention after early warning, firstly, corresponding monitoring sensors are configured at the corresponding construction locations in the monitoring area for key monitoring as the project construction progresses. During the monitoring process, real-time data from each monitoring sensor is collected, and then machine algorithms are used to perform feature analysis to determine potential environmental risks in the monitoring area. Furthermore, an early warning model for the monitoring area is constructed using the real-time data. The early warning model is used to analyze the probability of occurrence of each potential environmental risk in the monitoring area. In order to avoid the interference of sudden risks on the environment and reduce environmental pollution, when there are dangerous environmental risks in the monitoring area, corresponding emergency measures are identified, and an emergency plan is constructed. In this way, relevant personnel can adjust the construction in the monitoring area according to the emergency plan, and guide project managers to quickly take emergency measures to reduce the damage of construction to the environment.

[0061] Other features and advantages of the invention will be set forth in the following description, and will be apparent in part from the description, or may be learned by practicing the invention. The objects and other advantages of the invention may be realized and obtained by means of the structures particularly pointed out in the written description and the accompanying drawings.

[0062] The technical solution of the present invention will be further described in detail below with reference to the accompanying drawings and embodiments. Attached Figure Description

[0063] The accompanying drawings are provided to further illustrate the invention and form part of the specification. They are used together with the embodiments of the invention to explain the invention and do not constitute a limitation thereof. In the drawings:

[0064] Figure 1 This is a schematic diagram illustrating the workflow of the real-time early warning method for environmental risks during the construction period of power transmission and transformation projects in this embodiment of the invention.

[0065] Figure 2 This is a schematic diagram of the composition of the real-time early warning system for environmental risks during the construction period of power transmission and transformation projects in an embodiment of the present invention. Detailed Implementation

[0066] The preferred embodiments of the present invention will be described below with reference to the accompanying drawings. It should be understood that the preferred embodiments described herein are for illustration and explanation only and are not intended to limit the present invention.

[0067] Example 1:

[0068] This embodiment provides a method and system for real-time early warning of environmental risks during the construction period of power transmission and transformation projects, such as... Figure 1 As shown, it includes:

[0069] Step 1: Determine the key monitoring areas in real time based on the project construction progress, and control the preset mobile monitoring platform to configure the corresponding monitoring sensors for the key monitoring areas in real time;

[0070] Step 2: Acquire the real-time data corresponding to each of the monitoring sensors, and use machine learning algorithms to perform feature analysis on the real-time data to determine the potential environmental risks of the monitoring area;

[0071] Step 3: Construct an early warning model for the monitoring area based on the real-time collected data, and monitor the probability of occurrence corresponding to each potential environmental risk in the early warning model;

[0072] Step 4: Identify emergency measures corresponding to hazardous environmental risks with a probability of occurrence higher than the prescribed probability, construct an emergency plan based on the real-time data collected in the monitoring area, and display it.

[0073] In this example, as the construction progresses, different tasks are performed at different times, so the focus of the monitoring area is constantly changing. Therefore, as the construction progresses, the real-time monitoring focus corresponding to different progress stages must be determined.

[0074] In this example, the monitoring sensors include: PM2.5 sensor, nitrogen oxide sensor, water quality detection sensor, soil pollution sensor, vegetation status sensor (such as chlorophyll sensor), and vibration sensor;

[0075] In this example, fixed monitoring sensors are installed on key areas for real-time monitoring (such as tower bases, substations, construction roads, etc.).

[0076] In this example, portable sensors are deployed on a mobile monitoring platform for temporary or mobile monitoring.

[0077] In this example, potential environmental risk refers to the potential dangers that may occur to the environment of the monitoring area due to the impact of engineering construction on the monitoring area;

[0078] In this example, the probability is specified as 35%.

[0079] In this example, a hazardous environmental risk may correspond to multiple or one emergency measures.

[0080] The working principle and beneficial effects of the above technical solution are as follows: To achieve rapid analysis and early warning of monitoring data, and to intervene after early warning to reduce the impact of environmental risks on the environment, firstly, corresponding monitoring sensors are configured at the relevant construction locations in the monitoring area for key monitoring as the project construction progresses. During the monitoring process, real-time data emitted by each monitoring sensor is collected, and then machine algorithms are used to perform feature analysis to determine potential environmental risks within the monitoring area. Furthermore, an early warning model for the monitoring area is constructed using the real-time collected data. The early warning model is used to analyze the probability of occurrence of each potential environmental risk in the monitoring area. To avoid interference with the environment from sudden risks and reduce environmental pollution, when dangerous environmental risks exist in the monitoring area, corresponding emergency measures are identified, thereby constructing an emergency plan. In this way, relevant personnel can adjust the construction in the monitoring area according to the emergency plan, and guide project managers to quickly take emergency measures to reduce the damage of construction to the environment.

[0081] Example 2:

[0082] Based on Example 1, the real-time early warning method for environmental risks during the construction period of power transmission and transformation projects, step 1 includes:

[0083] Step 11: Based on the engineering blueprint of this project, construct the estimated construction progress of this project, determine the construction key areas corresponding to different construction progress, obtain the map of the monitoring area, and determine the environmental layout corresponding to each construction key area.

[0084] Step 12: Determine the current critical construction area based on the current construction progress of this project, and determine several monitoring locations and the monitoring type corresponding to each monitoring location based on the environmental layout corresponding to the current critical construction area.

[0085] Step 13: Construct a sensor configuration scheme according to the monitoring type corresponding to each monitoring location, control the preset mobile monitoring platform to arrive at each monitoring location in sequence according to the sensor configuration scheme, and configure the corresponding monitoring sensor for each monitoring location.

[0086] In this example, the monitoring type matches the type of monitoring sensor required.

[0087] The working principle and beneficial effects of the above technical solution are as follows: In order to save costs and reduce waste, the location and monitoring type of the current construction key area are determined according to the construction progress, so that the mobile monitoring platform can configure the sensors, thereby reducing the investment in sensor costs without affecting the monitoring effect.

[0088] Example 3:

[0089] Based on Example 2, the real-time early warning method for environmental risks during the construction period of power transmission and transformation projects further includes:

[0090] When the current construction progress reaches the next stage, a sensor retrieval scheme is constructed based on the sensor configuration scheme. The mobile monitoring platform is controlled to sequentially reach each monitoring location and retrieve the monitoring sensor corresponding to each monitoring location.

[0091] The working principle and beneficial effects of the above technical solution are as follows: In order to establish a complete monitoring process, when the project construction reaches the next stage, the sensors configured in the previous stage are retrieved and await reconfiguration.

[0092] Example 4:

[0093] Based on Example 1, the real-time early warning method for environmental risks during the construction period of power transmission and transformation projects, step 2 includes:

[0094] Step 21: Control the preset drone to conduct regular inspections of the monitoring area, construct an engineering construction image of the monitoring area, acquire real-time data corresponding to each monitoring sensor, and establish dimensional data attributes for each real-time data based on the monitoring dimension corresponding to each monitoring sensor.

[0095] Step 22: Perform cluster analysis on the real-time collected data according to the dimensional data attributes to obtain several monitoring data classes, analyze the data logic information between different monitoring data in each monitoring data class, construct a clustering tree corresponding to each monitoring data class, and determine the current dimensional space corresponding to each clustering tree respectively;

[0096] Step 23: Determine the low-dimensional space corresponding to each clustering tree, compress each clustering tree from the corresponding current dimension space to the corresponding low-dimensional space using a preset autoencoder, and extract the low-dimensional tree features corresponding to each clustering tree in each low-dimensional space.

[0097] Step 24: Construct real-time presentation features corresponding to the real-time collected data based on the low-dimensional tree features corresponding to each clustering tree, construct dynamic presentation information corresponding to each monitoring dimension, perform anomaly analysis on each dynamic presentation information, determine several abnormal dynamics contained in the monitoring area, determine the potential environmental risks of the monitoring area, and mark the risk range corresponding to each potential environmental risk in the engineering construction image and display it.

[0098] In this example, the frequency of regular inspections is once every 12 hours;

[0099] In this example, the monitoring dimensions include: surface dimension, vegetation dimension, and air dimension;

[0100] In this example, the dimensional data attributes are related to the monitoring dimensions, and different monitoring dimensions correspond to different dimensional data attributes.

[0101] In this example, the data logic information represents the linear and non-linear relationships between different monitoring data within the same monitoring data class;

[0102] In this example, the cluster analysis is hierarchical clustering;

[0103] In this example, the clustering tree represents a binary tree built by clustering the monitoring data in the monitoring data class obtained from the clustering analysis according to its corresponding data logic information;

[0104] In this example, the low-dimensional space corresponding to a clustering tree is one dimension lower than its current dimension space;

[0105] In this example, the predefined autoencoder represents the neural network used to compress the clustering tree into a low-dimensional space before reconstruction.

[0106] In this example, the real-time presentation feature refers to the data features presented in the real-time acquired data;

[0107] In this example, the low-dimensional tree features represent the characteristics of the clustering tree in a low-dimensional space, and the clustering tree can be denoised after dimensionality reduction.

[0108] The working principle and beneficial effects of the above technical solution are as follows: In order to monitor and warn before danger occurs, drones are used for regular inspections to build engineering construction images of the monitoring area. Then, cluster analysis is performed on the real-time data collected by each monitoring sensor to construct a cluster tree using the data logic information between different monitoring data. The low-dimensional tree features of each cluster tree are analyzed by dimensionality reduction to determine the real-time presentation features corresponding to each real-time data collection. By constructing dynamic presentation information for each monitoring dimension, abnormal dynamics in the monitoring area are analyzed. Finally, the potential environmental risks in the monitoring area are determined, and the risk range of each potential environmental risk is marked in the engineering construction image. This guides managers to keep abreast of the situation in the monitoring area and realize remote monitoring and early warning.

[0109] Example 5:

[0110] Based on Example 4, the real-time early warning method for environmental risks during the construction period of power transmission and transformation projects, step 24 includes:

[0111] Step 241: Determine several nonlinear features of the corresponding monitoring data class based on the low-dimensional tree features of each clustering tree, construct a nonlinear model of the corresponding monitoring data class using the nonlinear features, run the nonlinear data model to determine the nonlinear structure of each real-time acquired data, and determine the real-time presentation features of each real-time acquired data.

[0112] Step 242: Sort several real-time presentation features corresponding to the same real-time collected data in time sequence to obtain dynamic presentation information corresponding to each monitoring dimension, obtain the dynamic period corresponding to each dynamic presentation information respectively, and establish anomaly screening rules corresponding to each dynamic presentation information by using a preset isolated forest combined with the dynamic period.

[0113] Step 243: Use the anomaly screening rules to screen the anomaly information contained in the dynamic presentation information respectively, construct several overall anomaly features of the monitoring area, analyze the anomaly dimension corresponding to the monitoring area based on the overall anomaly features, analyze the anomaly level corresponding to the monitoring area based on the anomaly feature value corresponding to the overall anomaly features, and determine the potential environmental risks of the monitoring area.

[0114] In this example, a nonlinear structure represents a structure between different monitoring data in a monitoring data class that cannot be simply described by a linear model;

[0115] In this example, the real-time presentation feature refers to the characteristics that a real-time acquired data exhibits under the interference or influence of other real-time acquired data.

[0116] In this example, temporal sorting refers to the result of sorting the real-time presented features according to the order of time;

[0117] In this example, dynamic periodicity refers to the period during which data in dynamically presented information repeats;

[0118] In this example, the default isolated forest represents an algorithm used for anomaly analysis.

[0119] The working principle and beneficial effects of the above technical solution are as follows: To further ensure the efficiency and accuracy of analyzing potential environmental risks, the nonlinear characteristics of the monitoring data class are first determined based on the low-dimensional tree characteristics of the clustering tree, thereby constructing a nonlinear model. This model is used to analyze the nonlinear structure of the real-time acquired data, thereby determining the real-time presentation characteristics of each real-time acquired data. By sorting the data according to time order, dynamic presentation information corresponding to each monitoring dimension is constructed. Then, based on the dynamic cycle of the dynamic presentation information, anomaly screening rules are constructed in conjunction with isolated forests. The anomaly screening rules are then used to complete the screening work, obtaining the overall anomaly characteristics of the monitoring area. Finally, the level and dimension of potential environmental risks are determined based on the overall anomaly characteristics. This method can quickly filter and process anomaly characteristics without requiring long-term monitoring of the dynamic presentation information, reducing the workload of the backend and improving the efficiency of anomaly screening.

[0120] Example 6:

[0121] Based on Example 1, the real-time early warning method for environmental risks during the construction period of power transmission and transformation projects, step 3 includes:

[0122] Step 31: Obtain the risk monitoring list of the monitoring area, establish several risk monitoring conditions based on the risk monitoring list, and construct an early warning model for the monitoring area based on the risk monitoring conditions and the real-time collected data;

[0123] Step 32: In the early warning model, determine the risk occurrence range corresponding to each potential environmental risk, and perform model sampling on each risk occurrence range to obtain several sampling information of each risk occurrence range;

[0124] Step 33: Construct several risk factors corresponding to the risk occurrence range using the sampling information, run the early warning model, and determine the trigger time corresponding to each risk factor;

[0125] Step 34: Estimate the probability of sudden occurrence of potential environmental risks within the specified construction period based on the time difference between the trigger time and the current time.

[0126] In this example, the risk monitoring list refers to a list set up in advance before construction to monitor the risks of various tasks in the area. It includes several risk names and the risk level classification corresponding to each risk.

[0127] In this example, the specified time period represents 12 hours × 30 days from the current moment;

[0128] In this example, the probability of a potential environmental risk occurring within the specified time period is greater than 35%, and the closer it is to the current time, the higher the probability of occurrence. Conversely, the probability of an external potential environmental risk occurring within the specified time period is less than 35%, and the closer it is to the specified time, the higher the probability of occurrence.

[0129] In this example, the risk factor represents the factors within the scope of the risk occurrence that are likely to cause potential environmental risks.

[0130] The working principle and beneficial effects of the above technical solution are as follows: Risk monitoring conditions are constructed by using a risk monitoring list of the monitoring area, and an early warning model is constructed by combining real-time collected data. Then, the risk occurrence range of each potential marriage risk is analyzed. Each risk occurrence range is sampled in the early warning model, and the risk factors of each risk occurrence range are analyzed by using the sampled information. The trigger time of each risk factor is determined by running the early warning model, and the probability of the sudden occurrence of the potential risk is determined by combining the current time, which facilitates subsequent early warning work.

[0131] Example 7:

[0132] Based on Example 1, the real-time early warning method for environmental risks during the construction period of power transmission and transformation projects, step 4 includes:

[0133] Step 41: Obtain the hazardous environmental risks in the monitoring area with a probability of sudden occurrence higher than the specified probability, find several emergency measures corresponding to each hazardous environmental risk, input each emergency measure into the early warning model for measure simulation, and obtain the first degree of integration between each emergency measure and the monitoring area;

[0134] Step 42: Based on the first degree of fusion, select several available emergency measures corresponding to each of the hazardous environmental risks, extract one available emergency measure corresponding to each of the hazardous environmental risks, and combine the measures to obtain several risk-resistance combinations;

[0135] Step 43: Simulate each of the risk mitigation combinations using the early warning model to obtain a second degree of integration between each risk mitigation combination and the monitoring area, and select the target risk mitigation combination for the monitoring area based on the second degree of integration;

[0136] Step 44: Determine the current risk value of the monitoring area based on the real-time collected data, adjust the execution starting point of the target risk mitigation combination using the current risk value, obtain the emergency plan for the monitoring area, and display it.

[0137] The working principle and beneficial effects of the above technical solution are as follows: When there are hazardous environmental risks in the monitoring area, the emergency measures corresponding to each hazardous environmental risk are quickly identified. Then, available emergency measures that are compatible with the monitoring area are initially screened. The available emergency measures are then combined and screened again to obtain a target risk-resistant combination that can be integrated with the monitoring area and will not produce side effects between them. Finally, the target risk-resistant combination is fine-tuned using real-time data to generate an emergency plan, which can help guide personnel to resolve hazards in a timely manner.

[0138] Example 8:

[0139] Based on Example 7, the real-time early warning method for environmental risks during the construction period of power transmission and transformation projects further includes:

[0140] When there is a hazardous environmental risk within the monitoring area, the range of the hazardous environmental risk is obtained, real-time early warning information is generated and displayed.

[0141] Example 9:

[0142] This embodiment provides a real-time early warning system for environmental risks during the construction period of power transmission and transformation projects, such as... Figure 2 As shown, it includes:

[0143] The real-time acquisition module is used to determine the real-time monitoring focus of the monitoring area according to the progress of the project construction, and to control the preset mobile monitoring platform to configure the corresponding monitoring sensors for the real-time monitoring focus.

[0144] The risk assessment module is used to acquire real-time data corresponding to each of the monitoring sensors, and to perform feature analysis on the real-time data using machine learning algorithms to determine the potential environmental risks of the monitoring area.

[0145] The monitoring and analysis module is used to construct an early warning model for the monitoring area based on the real-time collected data, and to monitor the probability of occurrence of each potential environmental risk in the early warning model.

[0146] The early warning execution module is used to identify emergency measures corresponding to hazardous environmental risks with a probability of occurrence higher than the specified probability, and to construct and display emergency plans based on real-time data collected from the monitoring area.

[0147] In this example, as the construction progresses, different tasks are performed at different times, so the focus of the monitoring area is constantly changing. Therefore, as the construction progresses, the real-time monitoring focus corresponding to different progress stages must be determined.

[0148] In this example, the monitoring sensors include: PM2.5 sensor, nitrogen oxide sensor, water quality detection sensor, soil pollution sensor, vegetation status sensor (such as chlorophyll sensor), and vibration sensor;

[0149] In this example, fixed monitoring sensors are installed on key areas for real-time monitoring (such as tower bases, substations, construction roads, etc.).

[0150] In this example, portable sensors are deployed on a mobile monitoring platform for temporary or mobile monitoring.

[0151] In this example, potential environmental risk refers to the potential dangers that may occur to the environment of the monitoring area due to the impact of engineering construction on the monitoring area;

[0152] In this example, the probability is specified as 35%.

[0153] In this example, a hazardous environmental risk may correspond to multiple or one emergency measures.

[0154] The working principle and beneficial effects of the above technical solution are as follows: To achieve rapid analysis and early warning of monitoring data, and to intervene after early warning to reduce the impact of environmental risks on the environment, firstly, corresponding monitoring sensors are configured at the relevant construction locations in the monitoring area for key monitoring as the project construction progresses. During the monitoring process, real-time data emitted by each monitoring sensor is collected, and then machine algorithms are used to perform feature analysis to determine potential environmental risks within the monitoring area. Furthermore, an early warning model for the monitoring area is constructed using the real-time collected data. The early warning model is used to analyze the probability of occurrence of each potential environmental risk in the monitoring area. To avoid interference with the environment from sudden risks and reduce environmental pollution, when dangerous environmental risks exist in the monitoring area, corresponding emergency measures are identified, thereby constructing an emergency plan. In this way, relevant personnel can adjust the construction in the monitoring area according to the emergency plan, and guide project managers to quickly take emergency measures to reduce the damage of construction to the environment.

[0155] Example 10:

[0156] Based on Example 9, the real-time environmental risk early warning system during the construction period of power transmission and transformation projects includes a risk assessment module comprising:

[0157] The data acquisition unit is used to control a preset drone to conduct regular inspections of the monitoring area, construct engineering construction images of the monitoring area, acquire real-time data corresponding to each monitoring sensor, and establish dimensional data attributes of each real-time data according to the monitoring dimension corresponding to each monitoring sensor.

[0158] The data processing unit is used to perform cluster analysis on the real-time collected data according to the dimensional data attributes to obtain several monitoring data classes, analyze the data logic information between different monitoring data in each monitoring data class, construct a clustering tree corresponding to each monitoring data class, and determine the current dimensional space corresponding to each clustering tree respectively.

[0159] The data dimensionality reduction unit is used to determine the low-dimensional dimension space corresponding to each clustering tree, compress each clustering tree from the corresponding current dimension space to the corresponding low-dimensional dimension space using a preset autoencoder, and extract the low-dimensional tree features corresponding to each clustering tree in each low-dimensional dimension space.

[0160] The risk determination unit is used to construct real-time presentation features corresponding to real-time collected data based on the low-dimensional tree features corresponding to each clustering tree, construct dynamic presentation information corresponding to each monitoring dimension, perform anomaly analysis on each dynamic presentation information, determine several abnormal dynamics contained in the monitoring area, determine the potential environmental risks of the monitoring area, and mark and display the risk range corresponding to each potential environmental risk in the engineering construction image.

[0161] In this example, the frequency of regular inspections is once every 12 hours;

[0162] In this example, the monitoring dimensions include: surface dimension, vegetation dimension, and air dimension;

[0163] In this example, the dimensional data attributes are related to the monitoring dimensions, and different monitoring dimensions correspond to different dimensional data attributes.

[0164] In this example, the data logic information represents the linear and non-linear relationships between different monitoring data within the same monitoring data class;

[0165] In this example, the cluster analysis is hierarchical clustering;

[0166] In this example, the clustering tree represents a binary tree built by clustering the monitoring data in the monitoring data class obtained from the clustering analysis according to its corresponding data logic information;

[0167] In this example, the low-dimensional space corresponding to a clustering tree is one dimension lower than its current dimension space;

[0168] In this example, the predefined autoencoder represents the neural network used to compress the clustering tree into a low-dimensional space before reconstruction.

[0169] In this example, the real-time presentation feature refers to the data features presented in the real-time acquired data;

[0170] In this example, the low-dimensional tree features represent the characteristics of the clustering tree in a low-dimensional space, and the clustering tree can be denoised after dimensionality reduction.

[0171] The working principle and beneficial effects of the above technical solution are as follows: In order to monitor and warn before danger occurs, drones are used for regular inspections to build engineering construction images of the monitoring area. Then, cluster analysis is performed on the real-time data collected by each monitoring sensor to construct a cluster tree using the data logic information between different monitoring data. The low-dimensional tree features of each cluster tree are analyzed by dimensionality reduction to determine the real-time presentation features corresponding to each real-time data collection. By constructing dynamic presentation information for each monitoring dimension, abnormal dynamics in the monitoring area are analyzed. Finally, the potential environmental risks in the monitoring area are determined, and the risk range of each potential environmental risk is marked in the engineering construction image. This guides managers to keep abreast of the situation in the monitoring area and realize remote monitoring and early warning.

[0172] Obviously, those skilled in the art can make various modifications and variations to this invention without departing from its spirit and scope. Therefore, if these modifications and variations fall within the scope of the claims of this invention and their equivalents, this invention also intends to include these modifications and variations.

Claims

1. A method for real-time early warning of environmental risks during the construction period of power transmission and transformation projects, characterized in that, include: Step 1: Determine the key monitoring areas in real time based on the project construction progress, and control the preset mobile monitoring platform to configure the corresponding monitoring sensors for the key monitoring areas in real time; Step 2: Acquire the real-time data corresponding to each of the monitoring sensors, and use machine learning algorithms to perform feature analysis on the real-time data to determine the potential environmental risks of the monitoring area; Step 3: Construct an early warning model for the monitoring area based on the real-time collected data, and monitor the probability of occurrence corresponding to each potential environmental risk in the early warning model; Step 4: Identify emergency measures corresponding to hazardous environmental risks with a probability of occurrence higher than the prescribed probability, construct an emergency plan based on the real-time data collected in the monitoring area, and display it. Step 2 includes: Step 21: Control the preset drone to conduct regular inspections of the monitoring area, construct an engineering construction image of the monitoring area, acquire real-time data collected by each monitoring sensor, and establish dimensional data attributes of each real-time data collected according to the monitoring dimension corresponding to each monitoring sensor. Step 22: Perform cluster analysis on the real-time collected data according to the dimensional data attributes to obtain several monitoring data classes, analyze the data logic information between different monitoring data in each monitoring data class, construct a clustering tree corresponding to each monitoring data class, and determine the current dimensional space corresponding to each clustering tree respectively; Step 23: Determine the low-dimensional space corresponding to each clustering tree, compress each clustering tree from the corresponding current dimension space to the corresponding low-dimensional space using a preset autoencoder, and extract the low-dimensional tree features corresponding to each clustering tree in each low-dimensional space. Step 24: Construct real-time presentation features corresponding to the real-time collected data based on the low-dimensional tree features corresponding to each clustering tree, construct dynamic presentation information corresponding to each monitoring dimension, perform anomaly analysis on each dynamic presentation information, determine several abnormal dynamics contained in the monitoring area, determine the potential environmental risks of the monitoring area, and mark the risk range corresponding to each potential environmental risk in the engineering construction image and display it. Step 24 includes: Step 241: Determine several nonlinear features of the corresponding monitoring data class based on the low-dimensional tree features of each clustering tree, construct a nonlinear model of the corresponding monitoring data class using the nonlinear features, run the nonlinear model to determine the nonlinear structure of each real-time acquired data, and determine the real-time presentation features of each real-time acquired data. Step 242: Sort several real-time presentation features corresponding to the same real-time collected data in time sequence to obtain dynamic presentation information corresponding to each monitoring dimension, obtain the dynamic period corresponding to each dynamic presentation information respectively, and establish anomaly screening rules corresponding to each dynamic presentation information by using a preset isolated forest combined with the dynamic period. Step 243: Use the anomaly screening rules to screen the anomaly information contained in the dynamic presentation information respectively, construct several overall anomaly features of the monitoring area, analyze the anomaly dimension corresponding to the monitoring area based on the overall anomaly features, analyze the anomaly level corresponding to the monitoring area based on the anomaly feature value corresponding to the overall anomaly features, and determine the potential environmental risks of the monitoring area.

2. The real-time early warning method for environmental risks during the construction period of power transmission and transformation projects as described in claim 1, characterized in that, Step 1 includes: Step 11: Based on the engineering blueprint of this project, construct the estimated construction progress of this project, determine the construction key areas corresponding to different construction progress, obtain the map of the monitoring area, and determine the environmental layout corresponding to each construction key area. Step 12: Determine the current critical construction area based on the current construction progress of this project, and determine several monitoring locations and the monitoring type corresponding to each monitoring location based on the environmental layout corresponding to the current critical construction area. Step 13: Construct a sensor configuration scheme according to the monitoring type corresponding to each monitoring location, control the preset mobile monitoring platform to arrive at each monitoring location in sequence according to the sensor configuration scheme, and configure the corresponding monitoring sensor for each monitoring location.

3. The real-time early warning method for environmental risks during the construction period of power transmission and transformation projects as described in claim 2, characterized in that, Also includes: When the current construction progress reaches the next stage, a sensor retrieval scheme is constructed based on the sensor configuration scheme. The mobile monitoring platform is controlled to sequentially reach each monitoring location and retrieve the monitoring sensor corresponding to each monitoring location.

4. The real-time early warning method for environmental risks during the construction period of power transmission and transformation projects as described in claim 1, characterized in that, Step 3 includes: Step 31: Obtain the risk monitoring list of the monitoring area, establish several risk monitoring conditions based on the risk monitoring list, and construct an early warning model for the monitoring area based on the risk monitoring conditions and the real-time collected data; Step 32: In the early warning model, determine the risk occurrence range corresponding to each potential environmental risk, and perform model sampling on each risk occurrence range to obtain several sampling information of each risk occurrence range; Step 33: Construct several risk factors corresponding to the risk occurrence range using the sampling information, run the early warning model, and determine the trigger time corresponding to each risk factor; Step 34: Estimate the probability of sudden occurrence of potential environmental risks within the specified construction period based on the time difference between the trigger time and the current time.

5. The method for real-time early warning of environmental risks during the construction period of power transmission and transformation projects as described in claim 1, characterized in that, Step 4 includes: Step 41: Obtain the hazardous environmental risks in the monitoring area with a probability of sudden occurrence higher than the specified probability, find several emergency measures corresponding to each hazardous environmental risk, input each emergency measure into the early warning model for measure simulation, and obtain the first degree of integration between each emergency measure and the monitoring area; Step 42: Based on the first degree of fusion, select several available emergency measures corresponding to each of the hazardous environmental risks, extract one available emergency measure corresponding to each of the hazardous environmental risks, and combine the measures to obtain several risk-resistance combinations; Step 43: Simulate each of the risk mitigation combinations using the early warning model to obtain a second degree of integration between each risk mitigation combination and the monitoring area, and select the target risk mitigation combination for the monitoring area based on the second degree of integration; Step 44: Determine the current risk value of the monitoring area based on the real-time collected data, adjust the execution starting point of the target risk mitigation combination using the current risk value, obtain the emergency plan for the monitoring area, and display it.

6. The real-time early warning method for environmental risks during the construction period of power transmission and transformation projects as described in claim 5, characterized in that, Also includes: When there is a hazardous environmental risk within the monitoring area, the range of the hazardous environmental risk is obtained, real-time early warning information is generated and displayed.

7. A real-time early warning system for environmental risks during the construction period of power transmission and transformation projects, characterized in that, include: The real-time acquisition module is used to determine the real-time monitoring focus of the monitoring area according to the progress of the project construction, and to control the preset mobile monitoring platform to configure the corresponding monitoring sensors for the real-time monitoring focus. The risk assessment module is used to acquire real-time data corresponding to each of the monitoring sensors, and to perform feature analysis on the real-time data using machine learning algorithms to determine the potential environmental risks of the monitoring area. The monitoring and analysis module is used to construct an early warning model for the monitoring area based on the real-time collected data, and to monitor the probability of occurrence of each potential environmental risk in the early warning model. The early warning execution module is used to identify emergency measures corresponding to hazardous environmental risks with a probability of occurrence higher than the specified probability, and to construct and display emergency plans based on real-time data collected from the monitoring area. The risk assessment module includes: The data acquisition unit is used to control a preset drone to conduct regular inspections of the monitoring area, construct engineering construction images of the monitoring area, acquire real-time data corresponding to each monitoring sensor, and establish dimensional data attributes of each real-time data according to the monitoring dimension corresponding to each monitoring sensor. The data processing unit is used to perform cluster analysis on the real-time collected data according to the dimensional data attributes to obtain several monitoring data classes, analyze the data logic information between different monitoring data in each monitoring data class, construct a clustering tree corresponding to each monitoring data class, and determine the current dimensional space corresponding to each clustering tree respectively. The data dimensionality reduction unit is used to determine the low-dimensional dimension space corresponding to each clustering tree, compress each clustering tree from the corresponding current dimension space to the corresponding low-dimensional dimension space using a preset autoencoder, and extract the low-dimensional tree features corresponding to each clustering tree in each low-dimensional dimension space. The risk determination unit is used to construct real-time presentation features of the corresponding real-time collected data based on the low-dimensional tree features corresponding to each clustering tree, construct dynamic presentation information corresponding to each monitoring dimension, perform anomaly analysis on each dynamic presentation information, determine several abnormal dynamics contained in the monitoring area, determine the potential environmental risks of the monitoring area, and mark and display the risk range corresponding to each potential environmental risk in the engineering construction image. The working process of the risk determination unit includes: Based on the low-dimensional tree features corresponding to each clustering tree, several nonlinear features are determined for the corresponding monitoring data class. The nonlinear features are used to construct a nonlinear model for the corresponding monitoring data class. The nonlinear model is run to determine the nonlinear structure corresponding to each real-time acquired data and to determine the real-time presentation features corresponding to each real-time acquired data. The real-time presentation features corresponding to the same real-time collected data are sorted in time sequence to obtain the dynamic presentation information corresponding to each monitoring dimension. The dynamic period corresponding to each dynamic presentation information is obtained respectively. An anomaly screening rule corresponding to each dynamic presentation information is established by using a preset isolated forest combined with the dynamic period. The abnormal information contained in the dynamic presentation information is screened using the aforementioned abnormal screening rules, and several overall abnormal features of the monitoring area are constructed. The abnormal dimension of the monitoring area is analyzed based on the overall abnormal features, and the abnormal level of the monitoring area is analyzed based on the abnormal feature value of the overall abnormal features, thereby determining the potential environmental risks of the monitoring area.

Images