Construction risk situation monitoring method and system based on deep neural network

Abstract

The present invention discloses a construction risk situation monitoring method and system based on a deep neural network, which relates to the technical field of construction risk monitoring. A construction safety assessment model is used to evaluate construction safety within a construction area. If safety falls short of expectations, an appropriate alarm method is recommended to the construction worker from an alarm method library. Based on the construction worker's construction status data and environmental condition data, corresponding operation reminders are matched from an operation reminder library to automatically adjust various regulating equipment within the construction area. The construction area is patrolled based on inspection time nodes that meet constraint conditions, and an emergency plan library matches corresponding emergency plans based on emergency characteristics. After determining the key factors affecting construction safety within the construction area, a construction plan optimization knowledge map outputs a corresponding construction optimization strategy. The patrol frequency is adapted to the current state of the construction area, improving patrol efficiency and reducing potential construction risks within the construction area.

Description

Technical Field

[0001] The present invention relates to the technical field of construction risk monitoring, and specifically to a construction risk situation monitoring method and system based on deep neural network. Background Art

[0002] Construction risk monitoring refers to the systematic, real-time monitoring and tracking of various potential or existing risks (such as structural safety, construction operations, environmental impacts, etc.) in construction projects. By setting up monitoring points at key locations on the construction site and utilizing advanced sensors, surveillance cameras, data analysis software and other instruments and equipment, key parameters during the construction process (such as structural stress, deformation, environmental quality, etc.) are collected, recorded, and analyzed in real time. The purpose of this process is to promptly discover and accurately assess potential quality defects, safety hazards, or environmental risks, and to provide construction managers with timely decision-making support so that they can quickly take targeted control and response measures to effectively curb the further development of risks, ensure the quality, progress, and safety of the construction project, and also protect the safety of the surrounding environment and personnel.

[0003] In the Chinese invention patent with authorization announcement number CN118313670B, a smart construction site risk monitoring and assessment method is disclosed, which relates to the field of construction site risk monitoring technology, specifically as follows: receiving construction site information, and transmitting the received construction site information to an information analysis module; analyzing the construction site information to obtain environmental data and construction site data, transmitting the environmental data and construction site data to a data calculation module for calculation to obtain construction site risk reference data and environmental risk reference data, receiving the construction site risk reference data to perform risk level classification to obtain a construction site risk level threshold, the information analysis module receives the environmental risk reference data to perform risk level classification to obtain an environmental risk level threshold; transmitting the differentiated risk level threshold to a risk judgment module, and performing risk judgment based on the risk level threshold; the present invention obtains and analyzes construction site information during construction, and judges the safety of construction site construction by acquiring construction site information in real time.

[0004] Combined with the above application and the contents of the prior art:

[0005] When construction needs to be carried out indoors, especially when working with scaffolding or carrying out interior decoration, if the construction area is large, there are many construction workers, and there are many construction tasks, construction safety is particularly important. Continuous safety management of the construction area is required, such as real-time video surveillance and environmental monitoring within the construction area.

[0006] The existing deep neural network construction risk situation monitoring method mainly relies on the real-time collection of construction data in the construction area, takes the construction data of construction workers as input, uses deep neural networks for safety evaluation, and issues alarms and responses when construction safety risks are identified, thereby ensuring construction safety. However, when there are many drilling and cutting operations in the construction area, the dust concentration in the air is high and the noise is also high, and the hearing and vision of the construction workers will be disturbed to a certain extent. In this construction scenario, when there is a safety risk in the construction area, the construction workers may not hear the alarm sound, which may cause the alarm to fail. At the same time, the existing risk situation perception method mainly focuses on safety alarms. When an alarm is actually generated, there is a lack of corresponding emergency management measures, resulting in incomplete risk situation perception effects.

[0007] To this end, the present invention provides a construction risk situation monitoring method and system based on deep neural network. Summary of the Invention

[0008] (1) Technical problems solved

[0009] In response to the shortcomings of the existing technology, the present invention provides a construction risk situation monitoring method and system based on a deep neural network. By using a construction safety assessment model to evaluate construction safety within the construction area, if the safety is lower than expected, an appropriate alarm method is recommended to the construction worker from the alarm method library. Based on the construction worker's construction status data and environmental condition data, corresponding operation reminders are matched from the operation reminder library to automatically adjust various regulating equipment in the construction area; the construction area is patrolled according to the inspection time nodes that meet the constraint conditions, and the emergency plan library matches the corresponding emergency plan based on the emergency characteristics; the construction plan optimization knowledge graph outputs the corresponding construction optimization strategy. The patrol frequency is adapted to the current status of the construction area, improving patrol efficiency and reducing possible construction risks in the construction area; thereby solving the technical problems described in the background technology.

[0010] (2) Technical solution

[0011] To achieve the above objectives, the present invention is implemented through the following technical solutions:

[0012] A construction risk situation monitoring method based on a deep neural network includes collecting environmental condition data within the construction area, conducting an environmental risk assessment, and generating an environmental risk level Hjx from a number of continuously acquired risk values. If the environmental risk level Hjx exceeds a risk threshold, a data collection instruction is issued to the outside world.

[0013] Use the construction safety assessment model to evaluate construction safety within the construction area. If safety is lower than expected, recommend appropriate alarm methods to construction workers from the alarm method library. After switching the alarm method based on the alarm feedback data, issue a first-level alarm command to the outside world.

[0014] After receiving the first-level alarm command, the system matches the corresponding operation reminder from the operation reminder library based on the construction worker's construction status data and environmental condition data, and uses the conditional automatic control model to automatically adjust various regulating equipment in the construction area;

[0015] The construction area is inspected based on inspection time nodes that meet the constraints. The warning value Jzp is generated by the status data of the continuous issuance of the first-level alarm command. If the warning value Jzp exceeds the expected value, the emergency plan library matches the corresponding emergency plan based on the emergency characteristics.

[0016] A construction safety report is automatically generated after the inspection cycle ends. After determining the key factors affecting construction safety in the construction area, the corresponding construction optimization strategy is output from the construction plan optimization knowledge graph based on the correspondence between the key factors and the construction optimization plan.

[0017] Furthermore, the sensor network in the construction area continuously collects environmental condition data, and the collected multi-source environmental condition data is fused to generate multi-modal environmental condition data;

[0018] Taking multimodal environmental condition data as input, the trained risk assessment deep model is used to perform environmental risk assessment and obtain risk values ​​at several consecutive time nodes.

[0019] Furthermore, the environmental risk level is generated from a number of consecutive risk values ​​in the following way:

[0020]

[0021] Where Xz(t) is the risk value at time t, λ is the time attenuation coefficient, β is the trend influence factor, and [a, b] is the time interval of the risk value distribution.

[0022] Furthermore, image data is collected by an image acquisition device arranged in the construction area, including the movement and posture data of the construction workers during construction, which are aggregated as construction status data;

[0023] Using construction status data as input, the trained construction safety assessment model is used to perform a safety assessment. If the safety score obtained is lower than expected, an external reminder is issued.

[0024] Furthermore, several alarm methods are collected in advance and summarized to generate an alarm method library; after receiving the reminder instruction, the worker's location information is determined, and the distribution status of dust and noise in the area where the construction worker is located and his / her reminder preference data are used to recommend an alarm method for the construction worker based on the filtered recommendation algorithm.

[0025] Furthermore, a personalized alarm method is implemented to alarm the construction workers, and the response time of each construction worker is collected. The switching value xH is constructed from several response times in the following manner:

[0026]

[0027] Where: xT i is the response time of the i-th construction worker, xT a is the mean of response time, IQR is the interquartile range of several response times; F1, F2 and F3 are weight coefficients.

[0028] Furthermore, the environmental risk level Hjx and alarm data in the construction area are collected during several consecutive inspection cycles, and the inspection frequency Cv in the next inspection cycle is constrained based on the environmental risk level Hjx and alarm data. The construction area is inspected according to the inspection time nodes that meet the constraints, and inspection feedback data is obtained.

[0029] Furthermore, the constraints for constraining the patrol frequency Cv in the next patrol cycle are as follows:

[0030]

[0031] Weight coefficient: 0≤ρ≤1, 0≤ζ≤1, and ρ+ζ=1; n is the number of first-level alarm instructions, U(i,j) is the time interval from the i-th first-level alarm instruction to the j-th first-level alarm instruction, Ut a is the average value of the time interval.

[0032] Furthermore, if the number of first-level alarm instructions received during the current patrol cycle exceeds the corresponding number threshold, an alert value Jzp is generated based on the status data of the continuously issued first-level alarm instructions. If the alert value Jzp exceeds the preset alert threshold, a second-level alarm instruction is issued to the outside.

[0033] Furthermore, the method of generating the warning value Jzp based on the status data of the continuously issued first-level alarm instructions is as follows:

[0034]

[0035] Where: Pf(t) is the average safety score of the construction workers at time t, Δt(t) is the time interval between two first-level alarm instructions, is the rate of change of the time interval, and F1, F2 and F3 are weight coefficients.

[0036] Furthermore, a construction safety report is automatically generated after the inspection cycle ends and sent to the inspectors. Combined with the construction safety report, the key factors affecting construction safety in the construction area are determined through principal component analysis, and each key factor is ranked according to the degree of influence. With construction plan optimization as the target word, a construction plan optimization knowledge map is pre-constructed.

[0037] Furthermore, the warning threshold is adjusted according to the optimized patrol frequency. The adjustment method is as follows:

[0038]

[0039] Among them, m is the number of inspection nodes in the inspection cycle, Xj(i,j) is the time interval from the i-th inspection node to the j-th inspection node, and Xj a is the average value of the time interval, and Pa is the adjustment ratio of the warning threshold.

[0040] The construction risk situation monitoring system based on deep neural networks includes an environmental risk analysis unit that collects environmental condition data within the construction area and then conducts an environmental risk assessment. The environmental risk level Hjx is generated from a number of continuously acquired risk values. If the environmental risk level Hjx exceeds the risk threshold, a data collection instruction is issued to the outside world.

[0041] The alarm mode matching unit uses the construction safety assessment model to evaluate the construction safety in the construction area. If the safety is lower than expected, it recommends an appropriate alarm mode for the construction workers from the alarm mode library. After switching the alarm mode based on the alarm feedback data, it issues a first-level alarm command to the outside world.

[0042] After receiving the first-level alarm command, the automatic control unit matches the corresponding operation reminder from the operation reminder library based on the construction worker's construction status data and environmental condition data, and automatically adjusts the various regulating equipment in the construction area using the conditional automatic control model;

[0043] The multi-level alarm unit patrols the construction area according to the inspection time nodes that meet the constraints. The status data of the continuous issuance of the first-level alarm instructions generates a warning value Jzp. If the warning value Jzp exceeds the expected value, the emergency plan library will match the corresponding emergency plan based on the emergency characteristics.

[0044] The scheme optimization unit automatically generates a construction safety report after the inspection cycle ends. After determining the key factors affecting construction safety in the construction area, the construction scheme optimization knowledge map outputs the corresponding construction optimization strategy based on the correspondence between the key factors and the construction optimization scheme.

[0045] (3) Beneficial effects

[0046] The present invention provides a construction risk situation monitoring method and system based on deep neural network, which has the following beneficial effects:

[0047] 1. Conduct a comprehensive risk assessment of the construction area to verify whether the construction environment conditions will affect the construction process; the environmental risk degree Hjx can be used to comprehensively evaluate the possible construction environment risks in the construction area. When the construction environment conditions are not conducive to the current construction, targeted measures can be taken in a timely manner.

[0048] 2. Evaluate the safety of the current construction operation, determine whether the current construction operation is compliant, and ensure that the current construction process can be carried out safely; consider the user's environment and personal preferences, and match the user with an appropriate alarm method to make the alarm method more adaptable to the current scenario and reduce the risk of user neglect.

[0049] 3. Feedback on the alarm effect at the current stage can be formed based on the switching value xH. When the current alarm mode fails to achieve full effect, the alarm mode can be switched to provide a comprehensive alarm in the construction area, thereby improving the effectiveness of the alarm. By issuing reminders or guidance to users, protection can be provided against possible construction risks. When there are abnormalities in the current construction environment, adaptive modifications can be made to the environmental conditions in the construction area, which can improve the environmental conditions in the construction area.

[0050] 4. Adaptively adjust the patrol frequency to match it with the current status of the construction area, avoid excessive or insufficient patrols within the construction area, improve patrol efficiency, and reduce potential construction risks within the construction area; construct a warning value Jzp based on the status of receiving a first-level alarm instruction. When the current alarm fails to achieve the expected effect, enable a higher-level alarm to facilitate targeted measures.

[0051] 5. As feedback and response to the secondary alarm command, when there is a high level of risk in the construction area, a targeted emergency response plan will be matched based on the construction data in the construction area to form a targeted response to the construction risks in the construction area and ensure the safety of construction personnel.

[0052] 6. By generating a construction safety report, a summary of the current stage of construction can be formed; by adjusting the inspection frequency in the construction area, the warning threshold can be adaptively adjusted, the warning threshold can be lowered and the inspection frequency can be increased, the sensitivity to construction risks can be increased, and the ability to perceive risks in the construction area can be increased.

[0053] 7. By identifying the corresponding influencing factors and outputting the corresponding construction optimization strategies from the construction plan optimization knowledge graph, the current construction and management plans can be optimized, the current construction status can be improved at a deeper level, and the subsequent construction risks can be reduced. BRIEF DESCRIPTION OF THE DRAWINGS

[0054] Figure 1 This is a flow chart of the construction risk situation monitoring method of the present invention;

[0055] Figure 2 It is a structural diagram of the construction risk situation monitoring system of the present invention. DETAILED DESCRIPTION

[0056] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0057] See also Figure 1 The present invention provides a construction risk situation monitoring method based on a deep neural network, comprising:

[0058] Step 1: After collecting environmental condition data in the construction area, an environmental risk assessment is performed. The environmental risk level Hjx is generated from a number of continuously acquired risk values. If the environmental risk level Hjx exceeds the risk threshold, a data collection instruction is issued to the outside world.

[0059] The step 1 includes the following:

[0060] Step 101: After determining the current construction area, a sensor network is deployed within the construction area, including dust sensors, acoustic sensors, and gas monitoring sensors. The sensor network within the construction area continuously collects environmental condition data, such as dust, noise, and air quality data. The collected multi-source environmental condition data is fused to generate multi-modal environmental condition data, and then a multi-source environmental condition data set is generated after aggregation.

[0061] Step 102: Combine neural networks such as CNN, RNN, Transformer, and GNN to generate a deep neural network. After training the deep neural network with the labeled sample data, a trained risk assessment deep model is obtained.

[0062] Taking multimodal environmental condition data as input, the trained deep risk assessment model is used to perform environmental risk assessment, obtain risk values ​​at several consecutive time nodes, and mark the risk values ​​with the corresponding time nodes;

[0063] During use, in order to ensure the construction safety of the construction area, various data in the construction area are collected, and the trained multimodal evaluation model is used to conduct a comprehensive risk assessment of the construction area to verify whether the current construction environment conditions will affect the construction process;

[0064] Step 103: Generate an environmental risk degree from a number of consecutive risk values, and evaluate the construction environment in the construction area using the obtained environmental risk degree. The environmental risk degree is generated in the following manner:

[0065]

[0066] Where Xz(t) is the risk value at time t, λ is the time attenuation coefficient, ranging from 0.1 to 1, β is the trend influence factor, ranging from 0 to 1, e can be 2.713, and [a, b] is the time interval of the risk value distribution;

[0067] Based on historical data and safety management of environmental conditions in the construction area, a risk threshold is pre-set. If the obtained environmental risk Hjx exceeds the risk threshold, it indicates that the current environmental conditions in the construction area may have a certain impact on the health of construction workers and construction safety. At this time, a data collection instruction is issued to the outside world.

[0068] When using, combine the contents in steps 101 to 103:

[0069] As a further content, after continuously obtaining several risk values ​​along the time axis, the environmental risk degree Hjx is constructed by the risk values. The environmental risk degree Hjx can be used to comprehensively evaluate the construction environmental risks that may exist in the construction area. When the construction environment conditions are not conducive to the current construction, targeted treatment can be carried out in a timely manner.

[0070] The existing deep neural network construction risk situation monitoring method mainly relies on the real-time collection of construction data in the construction area, takes the construction data of construction workers as input, uses deep neural networks for safety evaluation, and issues alarms and responses when construction safety risks are identified, thereby ensuring construction safety. However, when there are many drilling and cutting operations in the construction area, the dust concentration in the air is high and the noise is also high, and the hearing and vision of the construction workers will be disturbed to a certain extent. In this construction scenario, when there is a safety risk in the construction area, the construction workers may not hear the alarm sound, which may cause the alarm to fail. At the same time, the existing risk situation perception method mainly focuses on safety alarms. When an alarm is actually generated, there is a lack of corresponding emergency management measures, resulting in incomplete risk situation perception effects.

[0071] Step 2: Use the construction safety assessment model to evaluate construction safety within the construction area. If safety is lower than expected, recommend an appropriate alarm method to the construction workers from the alarm method library. After switching the alarm method based on the alarm feedback data, issue a first-level alarm command to the outside world.

[0072] The second step includes the following:

[0073] Step 201: An image acquisition device placed in the construction area acquires image data, including the movement and posture data of construction workers during construction, and aggregates the data as construction status data.

[0074] The machine learning algorithm is trained with labeled sample data to obtain a trained construction safety assessment model. The trained construction safety assessment model is used to perform a safety assessment using construction status data as input. If the safety score obtained is lower than expected, an external warning is issued.

[0075] When in use, as a further content, the construction status of the construction personnel is used as the starting point to evaluate the safety of the current construction operation, to determine whether the current construction operation is compliant and whether it meets the safety requirements and specifications. For example, timely reminders are issued when in an open operation state to ensure that the current construction process can be carried out safely.

[0076] Step 202: Pre-collect several alarm methods, such as using workers' personal devices for visual signal alarms and vibration prompts, and aggregate the acquired alarm methods to generate an alarm method library;

[0077] After receiving the reminder command, the worker's location information is determined. The distribution of dust and noise in the construction worker's area and their reminder preference data are summarized as feedback data. A filtering-based recommendation algorithm is used to recommend appropriate alarm methods for the construction worker from the alarm method library based on the feedback data:

[0078] During use, as a further content, taking into account the need to ensure that construction personnel may not be able to effectively receive alarm instructions in the current construction scenario, therefore, when it is necessary to issue an alarm instruction to the outside, the user's environment and personal preferences are taken into consideration to match the user with an appropriate alarm method, so that the alarm method is more adaptable to the current scenario, reducing the risk of user neglect, and enabling the user to deal with the alarm in a timely manner when it is received.

[0079] Step 203: Execute a personalized alarm mode to alarm the construction workers, collect the response time of each construction worker, and construct a switching value xH based on the multiple response times, in the following manner:

[0080]

[0081] Where: xTi is the response time of the i-th construction worker, xT a is the mean of the response time, IQR is the interquartile range of several response times; F1, F2 and F3 are weight coefficients, all of which fall between 0 and 1, and the sum of the three is 1;

[0082] Based on historical data and management expectations for alarm modes, a switching threshold is pre-set. If the obtained switching value xH exceeds the pre-set switching threshold, it indicates that the current alarm mode may not have the expected notification effect. The alarm mode is switched and a first-level alarm instruction is issued to the outside world.

[0083] When using, combine the contents in steps 201 to 203:

[0084] As a further processing method, the response and feedback data of the construction personnel when receiving the alarm are collected, and the switching value is constructed accordingly. Based on the switching value xH, feedback can be formed on the alarm effect at the current stage. When the current alarm method fails to achieve full effect, the alarm method can be switched. For example, an audible and visual alarm covering the entire construction area can be enabled, and the personalized alarm can be switched to a comprehensive alarm within the construction area, thereby improving the effectiveness of the alarm.

[0085] Step 3: After receiving the first-level alarm command, the corresponding operation reminder is matched from the operation reminder library based on the construction worker's construction status data and environmental condition data, and the conditional automatic control model is used to automatically adjust various regulating equipment in the construction area;

[0086] The step three includes the following:

[0087] Step 301: pre-set an operation reminder library, such as wearing protective equipment, starting noise reduction, lighting or ventilation equipment, etc.;

[0088] Based on the construction workers' construction status data and environmental conditions, the operation reminder library matches corresponding operation reminders and can provide customized risk warnings and suggestions. For example, it reminds workers to pay attention to environmental changes and start wearing basic protective equipment such as masks and earplugs. For some difficult operations, the intelligent voice assistant can also provide voice guidance to help workers evacuate safely or adjust their work.

[0089] Step 302: Train a machine learning algorithm using the labeled sample data to obtain a trained conditional automatic control model;

[0090] After setting corresponding qualified ranges for various environmental condition parameters in the construction area, the trained conditional automatic control model is used to control various regulatory equipment in the construction area, such as turning on or automatically adjusting ventilation equipment or noise reduction equipment;

[0091] When using, combine the contents in steps 301 and 302:

[0092] As feedback to the issuance of a first-level alarm command, when the current construction method or operation of the construction personnel is unqualified, reminders or guidance will be issued to the user to protect against possible construction risks. As a further content, when there are abnormalities in the current construction environment, the environmental conditions in the construction area can be adaptively modified to improve the environmental conditions in the construction area.

[0093] Step 4: Conduct inspections within the construction area based on inspection time nodes that meet the constraints. Generate a warning value Jzp based on the status data of continuously issuing first-level alarm instructions. If the warning value Jzp exceeds the expected value, the emergency plan library will match the corresponding emergency plan based on the emergency characteristics.

[0094] The step 4 includes the following contents:

[0095] Step 401: Collect environmental risk Hjx and alarm data in the construction area during several consecutive patrol cycles. Constrain the patrol frequency Cv in the next patrol cycle based on the environmental risk Hjx and alarm data. The constraint conditions are as follows:

[0096]

[0097] Weight coefficient: 0≤ρ≤1, 0≤ζ≤1, and ρ+ζ=1; the weight coefficient can be obtained by referring to the hierarchical analysis method;

[0098] Where n is the number of first-level alarm instructions, U(i,j) is the time interval from the i-th first-level alarm instruction to the j-th first-level alarm instruction, and Ut a is the average value of the time interval;

[0099] Conduct inspections within the construction area based on inspection time nodes that meet the constraints and obtain inspection feedback data;

[0100] When in use, as feedback on the possible risks in the current construction environment and construction status, when the construction safety officer conducts safety inspections in the construction area, the inspection frequency is adaptively adjusted to make it compatible with the current status of the construction area, avoiding too many or too few inspections in the construction area, improving inspection efficiency and reducing possible construction risks in the construction area.

[0101] Step 402: If the number of first-level alarm instructions received during the current patrol cycle exceeds the corresponding threshold, it indicates that the environmental conditions in the pre-construction area are poor. In this case, under dimensionless conditions, a warning value Jzp is generated based on the status data of the continuously issued first-level alarm instructions in the following manner:

[0102]

[0103] Where: Pf(t) is the average safety score of the construction workers at time t, Δt(t) is the time interval between two first-level alarm instructions, is the rate of change of the time interval, F1, F2 and F3 are weight coefficients, and their values ​​are consistent with the previous values;

[0104] If the warning value Jzp exceeds the preset warning threshold, a secondary alarm instruction is issued to the outside;

[0105] When in use, as a response to the first-level alarm command, the warning value Jzp is constructed according to the status of the received first-level alarm command, and this is used as the starting point to start the multi-level alarm mechanism. When the current alarm fails to achieve the expected effect, a higher-level alarm is enabled to facilitate the adoption of corresponding targeted measures.

[0106] Step 403: Develop a detailed emergency plan, including emergency handling plan, evacuation route, assembly location, and emergency contact person, etc., and aggregate the developed emergency plans to generate an emergency plan library;

[0107] After receiving the second-level alarm command, the system extracts features from the construction data in the construction area, such as construction environment data and construction status data, to obtain the corresponding emergency features. The emergency plan library matches the corresponding emergency plan based on the emergency features and pushes the emergency plan to the construction workers.

[0108] When using, combine the contents in steps 401 to 403:

[0109] As feedback and response to the secondary alarm command, when there is a high level of risk in the construction area, targeted emergency response plans are matched based on the construction data in the construction area, such as arranging the rapid evacuation of construction workers. This can form a targeted response to the construction risks in the construction area and ensure the safety of construction workers.

[0110] Step 5: After the inspection cycle is complete, a construction safety report is automatically generated. After determining the key factors affecting construction safety in the construction area, the corresponding construction optimization strategy is output from the construction plan optimization knowledge graph based on the correspondence between the key factors and the construction optimization plan.

[0111] The step five includes the following:

[0112] Step 501: After the inspection cycle ends, a construction safety report is automatically generated, including the alarm time and location, alarm triggering reason and executed emergency plan, environmental condition data and video records, etc.; the construction safety report is sent to the inspection personnel;

[0113] The warning threshold can be adjusted according to the optimized patrol frequency. The current alarm mechanism can be updated and adjusted as follows:

[0114]

[0115] Among them, m is the number of inspection nodes in the inspection cycle, Xj(i,j) is the time interval from the i-th inspection node to the j-th inspection node, and Xj a is the average value of the time interval, Pa is the adjustment ratio of the warning threshold;

[0116] When in use, as feedback to the secondary alarm instruction, after the current construction stage is completed, a construction safety report can be generated to summarize the current stage of construction. As another step of feedback, the inspection frequency in the construction area can be adjusted to adaptively adjust the warning threshold, lower the warning threshold and increase the inspection frequency, increase sensitivity to construction risks, and increase the ability to perceive risks in the construction area.

[0117] Step 502: Based on the construction safety report, determine the key factors affecting construction safety in the construction area through principal component analysis, and rank each key factor according to its impact.

[0118] Taking construction plan optimization as the target word, after deep retrieval and entity relationship building, a construction plan optimization knowledge graph is pre-built. Based on the correspondence between key factors and construction optimization plans, the construction plan optimization knowledge graph outputs the corresponding construction optimization strategy.

[0119] When using, combine the contents in steps 501 and 502:

[0120] As further feedback, when there are major construction risks in the construction area, such as construction environment risks and construction operation risks, by identifying the corresponding influencing factors and outputting the corresponding construction optimization strategies from the construction plan optimization knowledge graph, the current construction and management plans can be optimized, thereby improving the current construction status at a deeper level and reducing subsequent construction risks.

[0121] The Analytic Hierarchy Process (AHP) is a structured, systematic decision-support technique that decomposes complex, multi-criteria decision-making problems into a series of interconnected, hierarchical sub-problems or factors. It then uses expert judgment to compare and quantitatively evaluate the importance of each factor at each level. Ultimately, through a composite calculation, it determines the relative merits of each decision-making solution, providing decision-makers with a clear, quantitative basis for their decisions. This method combines qualitative and quantitative analysis, effectively improving the scientific nature and accuracy of decision-making.

[0122] The process of constructing a knowledge graph with "construction plan optimization" as the target word is as follows:

[0123] Clarify the construction goals - First, it is necessary to clarify the construction goals of the knowledge graph, that is, to collect and organize relevant knowledge, concepts, relationships, etc. around the theme of "construction plan optimization".

[0124] Select a construction method - Based on the scale, complexity, and available resources of the knowledge graph, you can choose one or more of the following construction methods: Manual construction method: Suitable for small-scale, domain-specific knowledge graphs, through manual collection, organization, and annotation of data. Automatic extraction method: Use natural language processing technology to automatically extract information such as entities, relationships, and attributes from large amounts of text. Semi-automatic construction method: Combine manual construction and automatic extraction methods, and manually review and correct the results of automatic extraction. Ontology-based construction method: Use ontology to model domain knowledge to improve the consistency and scalability of the knowledge graph.

[0125] Data Collection and Preprocessing—Data Collection: Collect data related to "Construction Plan Optimization" from relevant books, academic papers, engineering practice reports, and other channels. Data Cleaning: Remove duplicate, invalid, or erroneous data to ensure data accuracy and consistency. Data Annotation: Annotate the collected data, including entity and relationship annotations, to provide a foundation for subsequent automatic extraction and knowledge graph construction.

[0126] Build a knowledge graph - define entities and relationships: Based on the characteristics of the field of construction plan optimization, define relevant entities (such as construction methods, construction sequence, construction organization, etc.) and relationships (such as optimization methods, influencing factors, etc.). Build a model layer: Based on the defined entities and relationships, build the model layer of the knowledge graph, including entity types, relationship types, etc. Entity filling: Use automatic extraction or manual annotation methods to fill the collected data into the knowledge graph to form entities and relationships. Relationship linking: Based on the relationship information in the data, establish links between entities to form a complete knowledge graph.

[0127] Verification and Optimization — Verify the knowledge graph: Verify the accuracy and completeness of the knowledge graph through manual inspection, logical reasoning, and other methods. Optimize the knowledge graph: Based on the verification results, revise and optimize the knowledge graph to improve its quality and usability.

[0128] Application Scenarios and Extensions - Application Scenarios: Apply the constructed knowledge graph to scenarios such as decision support for construction plan optimization and knowledge query. Continuous Expansion: As new knowledge continues to emerge, the knowledge graph is continuously updated and expanded to maintain its timeliness and integrity.

[0129] Specific steps and content - Define the key elements of construction plan optimization: including construction methods, construction sequence, construction organization, construction labor organization, construction machinery organization, etc. Collect and organize relevant cases: Collect successful cases of construction plan optimization from actual engineering projects, and extract key information and optimization strategies. Establish an entity relationship model: Based on the collected cases and key elements, establish an entity relationship model to clarify the relationship between each element. Fill in the knowledge graph: Fill the collected cases and optimization strategies into the knowledge graph to form a complete construction plan optimization knowledge graph. Verify and optimize the knowledge graph: Verify the effectiveness of the knowledge graph through actual project applications, and optimize it based on feedback.

[0130] See also Figure 2 The present invention provides a construction risk situation monitoring system based on a deep neural network, comprising:

[0131] The environmental risk analysis unit collects environmental condition data within the construction area and then conducts an environmental risk assessment. It generates an environmental risk level Hjx based on a number of continuously acquired risk values. If the environmental risk level Hjx exceeds the risk threshold, it issues a data collection instruction to the outside world.

[0132] The alarm mode matching unit uses the construction safety assessment model to evaluate the construction safety in the construction area. If the safety is lower than expected, it recommends an appropriate alarm mode for the construction workers from the alarm mode library. After switching the alarm mode based on the alarm feedback data, it issues a first-level alarm command to the outside world.

[0133] After receiving the first-level alarm command, the automatic control unit matches the corresponding operation reminder from the operation reminder library based on the construction worker's construction status data and environmental condition data, and automatically adjusts the various regulating equipment in the construction area using the conditional automatic control model;

[0134] The multi-level alarm unit patrols the construction area according to the inspection time nodes that meet the constraints. The status data of the continuous issuance of the first-level alarm instructions generates a warning value Jzp. If the warning value Jzp exceeds the expected value, the emergency plan library will match the corresponding emergency plan based on the emergency characteristics.

[0135] The scheme optimization unit automatically generates a construction safety report after the inspection cycle ends. After determining the key factors affecting construction safety in the construction area, the construction scheme optimization knowledge map outputs the corresponding construction optimization strategy based on the correspondence between the key factors and the construction optimization scheme.

[0136] Those skilled in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0137] Those skilled in the art will clearly understand that, for the convenience and brevity of description, the specific working processes of the systems, devices and units described above can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0138] In the several embodiments provided in this application, it should be understood that the disclosed systems, devices and methods can be implemented in other ways. For example, the device embodiments described above are merely schematic. For example, the division of the units is only for some logical functions. In actual implementation, there may be other division methods, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or units, which can be electrical, mechanical or other forms.

[0139] The units described as separate components may or may not be physically separate, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed across multiple network units. Some or all of these units may be selected to achieve the purpose of this embodiment according to actual needs.

[0140] The above description is merely a specific embodiment of the present application, but the scope of protection of the present application is not limited thereto. Any changes or substitutions that can be easily conceived by a person skilled in the art within the technical scope disclosed in this application should be included in the scope of protection of this application. Therefore, the scope of protection of this application should be based on the scope of protection of the claims.

Claims

1. A construction risk situation monitoring method based on a deep neural network, characterized by: include, After collecting environmental condition data in the construction area, an environmental risk assessment is conducted and an environmental risk index is generated from a number of continuously acquired risk values. If the environmental risk If the risk threshold is exceeded, a data collection instruction will be issued to the outside world; Use the construction safety assessment model to evaluate construction safety within the construction area. If safety is lower than expected, recommend appropriate alarm methods to construction workers from the alarm method library. After switching the alarm method based on the alarm feedback data, issue a first-level alarm command to the outside world. After receiving the first-level alarm command, the system matches the corresponding operation reminder from the operation reminder library based on the construction worker's construction status data and environmental condition data, and uses the conditional automatic control model to automatically adjust various regulating equipment in the construction area; Carry out inspections in the construction area according to the inspection time nodes that meet the constraints, and generate warning values ​​based on the status data of the continuous issuance of the first-level alarm instructions If the warning value If the number exceeds expectations, the emergency plan library will match the corresponding emergency plan based on the emergency characteristics; After the inspection cycle is completed, a construction safety report is automatically generated. After determining the key factors affecting construction safety in the construction area, the corresponding construction optimization strategy is output from the construction plan optimization knowledge map based on the correspondence between the key factors and the construction optimization plan; The sensor network in the construction area continuously collects environmental condition data, and the collected multi-source environmental condition data is integrated to generate multi-modal environmental condition data; Taking multimodal environmental condition data as input, the trained deep risk assessment model is used to perform environmental risk assessment and obtain risk values ​​at several consecutive time points. Generate environmental risk from several consecutive risk values , as follows: ； Where, is the risk value at time t, is the time attenuation coefficient, is the trend influencing factor, is the time interval of risk value distribution.

2. The construction risk situation monitoring method based on deep neural network according to claim 1, characterized in that: Image acquisition devices placed in the construction area collect image data, including the movement and posture data of construction workers during construction, which are aggregated as construction status data; Using construction status data as input, the trained construction safety assessment model is used to perform a safety assessment. If the safety score obtained is lower than expected, an external reminder is issued.

3. The construction risk situation monitoring method based on deep neural network according to claim 2, characterized in that: Collect several alarm methods in advance and generate an alarm method library after summarizing them; After receiving the reminder instruction, the worker's location information is determined, and the distribution status of dust and noise in the construction worker's area and his / her reminder preference data are used to recommend an alarm method for the construction worker based on the filtering recommendation algorithm.

4. The construction risk situation monitoring method based on deep neural network according to claim 3 is characterized in that: Implement personalized alarm mode to alarm construction workers, collect the response time of each construction worker, and construct the switching value based on several response times , as follows: ； Where: is the response time of the i-th construction worker, is the mean response time, is the interquartile range of several response times; and is the weight coefficient.

5. The construction risk situation monitoring method based on deep neural network according to claim 4 is characterized in that: Collect environmental risk data within the construction area during several consecutive inspection cycles and alarm data, based on environmental risk And the alarm data affects the patrol frequency in the next patrol cycle Constraints are imposed, and inspections are conducted within the construction area based on inspection nodes that meet the constraints, and inspection feedback data is obtained.

6. The construction risk situation monitoring method based on deep neural network according to claim 5, characterized in that: The patrol frequency in the next patrol cycle The constraints are as follows: ； Weight coefficient: , ,and ; The number of first-level alarm instructions. It is A first-level alarm command to the The time interval between the first level alarm instructions, is the average value of the time interval.

7. The construction risk situation monitoring method based on deep neural network according to claim 6, characterized in that: If in the current patrol cycle The number of times the first-level alarm command is received exceeds the corresponding threshold, and the warning value is generated based on the status data of the continuously issued first-level alarm commands. If the warning value When the preset warning threshold is exceeded, a secondary alarm command will be issued to the outside.

8. The construction risk situation monitoring method based on deep neural network according to claim 7, characterized in that: Generate warning values ​​based on the status data of the continuously issued first-level alarm instructions The way is as follows: ； Where: is the mean safety score of the construction workers at time t, is the time interval between two first-level alarm instructions, is the rate of change of the time interval, and is the weight coefficient.

9. The construction risk situation monitoring method based on deep neural network according to claim 8, characterized in that: Automatically generate a construction safety report after the inspection cycle is completed and send it to the inspectors; Combined with the construction safety report, principal component analysis is used to determine the key factors affecting construction safety in the construction area, and each key factor is ranked according to the degree of influence. With construction plan optimization as the target word, a construction plan optimization knowledge map is pre-constructed.

10. The construction risk situation monitoring method based on deep neural network according to claim 9, characterized in that: Adjust the warning threshold according to the optimized patrol frequency as follows: ； in, is the number of inspection nodes in the inspection cycle, It is Inspection nodes to the The time interval for patrolling nodes, is the average value of the time interval, Adjust the scale for the alert threshold.

11. A construction risk situation monitoring system based on a deep neural network, applying the monitoring method according to any one of claims 1 to 10, characterized in that: include, The environmental risk analysis unit collects environmental condition data in the construction area and then conducts environmental risk assessment. The environmental risk degree is generated from a number of continuously acquired risk values. If the environmental risk If the risk threshold is exceeded, a data collection instruction will be issued to the outside world; The alarm mode matching unit uses the construction safety assessment model to evaluate the construction safety in the construction area. If the safety is lower than expected, it recommends an appropriate alarm mode for the construction workers from the alarm mode library. After switching the alarm mode based on the alarm feedback data, it issues a first-level alarm command to the outside world. After receiving the first-level alarm command, the automatic control unit matches the corresponding operation reminder from the operation reminder library based on the construction worker's construction status data and environmental condition data, and automatically adjusts the various regulating equipment in the construction area using the conditional automatic control model; The multi-level alarm unit patrols the construction area according to the inspection time nodes that meet the constraints, and generates warning values ​​based on the status data of the continuous issuance of the first-level alarm command If the warning value If the number exceeds expectations, the emergency plan library will match the corresponding emergency plan based on the emergency characteristics; The scheme optimization unit automatically generates a construction safety report after the inspection cycle ends. After determining the key factors affecting construction safety in the construction area, the construction scheme optimization knowledge map outputs the corresponding construction optimization strategy based on the correspondence between the key factors and the construction optimization scheme.

Images