A safety risk identification method for mine ecological restoration projects

Abstract

The present invention discloses a method for identifying safety risks in mine ecological restoration projects, which relates to the technical field of engineering safety risk identification. It can monitor the environmental conditions at the construction site and the behaviors of construction workers in real time, discover problems in a timely manner and take corresponding measures, thereby improving the response speed and handling efficiency; automation and unmanned operation: realizing automated monitoring, reducing the dependence on manual supervision, reducing the possibility of human errors, and at the same time reducing the work burden of supervisors; improving safety, through comprehensive monitoring of the construction site, irregular behaviors and potential dangers can be discovered and corrected in a timely manner, thereby reducing the risk of accidents; reducing supervision costs, the automated and unmanned supervision mode reduces the demand for human resources, thereby reducing the overall cost of safety supervision.

Description

Technical Field

[0001] The present invention belongs to the technical field of engineering safety risk identification, and specifically relates to a method for identifying safety risks in mine ecological restoration projects. Background Art

[0002] Mine ecological restoration projects are carried out in the wild. The qualities of workers are uneven, the personnel mobility is large, and the construction is random and non-standard, which are very important reasons for causing safety risks. At the same time, under the action of the external environment, the working environment is relatively complex and harsh. It is very difficult for workers to concentrate highly during operation. In addition, if the weather suddenly changes, with strong winds or heavy rains, causing rolling stones on the slope, it will also cause the scaffolding or tall mechanical equipment to shake, bringing danger to workers.

[0003] At present, the safety control of mine ecological restoration construction mostly relies on people for management, lacking auxiliary complete sets of equipment for full-process management of construction safety. The subjective factors of people have a great impact on construction safety. Some key indicators still rely on manual supervision, and the supervision ability of safety supervision personnel is limited. It is difficult to carry out on-site supervision of multiple construction operations simultaneously, forming safety monitoring loopholes and easily leading to construction accidents. At the same time, the existing monitoring methods in other fields are not suitable for the ecological restoration construction field. Their installation process is cumbersome, the sensor deployment method lacks flexibility, the deployment pace is slow, the data information cannot be transmitted in a timely and effective manner, and the monitoring object can often only be a single target such as not wearing a helmet or a reflective vest, unable to provide comprehensive mine ecological restoration safety risk control data. Based on this, a solution is provided. Summary of the Invention

[0004] The present invention aims to solve at least one of the technical problems existing in the prior art;

[0005] For this purpose, the present invention proposes a method for identifying safety risks in mine ecological restoration projects, which specifically includes the following steps:

[0006] Perform data fusion and risk assessment before operation. Before workers enter the work area for operation, first perform face recognition, dangerous area recognition, meteorological disasters, and landslide rolling stone data collection processes, and then fuse the collected data, and perform preliminary risk assessment based on the fused data;

[0007] Fuse multi-source information and model safety risks in construction operations. By sorting and collecting data and picture materials of mine ecological restoration projects, analyze safety risks and influencing factors, fuse multi-source information on safety risks in construction operations, and respectively perform unified modeling on the data flow, business flow, and control flow affecting construction safety according to the types of data in the information system;

[0008] Quick acquisition and automatic identification of safety risks, identifying potential safety hazards in the real-time construction process, discerning on-site risks, evaluating using the analytic hierarchy process, and obtaining the evaluation index score values of construction risk factors according to the classification standard of construction risk evaluation indicators;

[0009] Simultaneously detecting and identifying unsafe behaviors of construction workers, thereby conducting safety risk data fusion and risk hazard assessment.

[0010] Compared with the prior art, the beneficial effects of the present invention are:

[0011] This application can monitor the environmental conditions at the construction site and the behaviors of construction workers in real time, discover problems in a timely manner and take corresponding measures, thereby improving the response speed and processing efficiency; automation and unmanned operation: realizing automated monitoring, reducing the dependence on manual supervision, reducing the possibility of human errors, and at the same time reducing the work burden of supervisors;

[0012] Improving safety, through comprehensive monitoring of the construction site, it is possible to discover and correct non-standard behaviors and potential dangers in a timely manner, thereby reducing the risk of accidents; reducing supervision costs, the automated and unmanned supervision mode reduces the demand for human resources, thereby reducing the overall cost of safety supervision. BRIEF DESCRIPTION OF THE DRAWINGS

[0013] Figure 1 It is a flowchart of the present invention. DETAILED DESCRIPTION OF THE INVENTION

[0014] The technical solutions of the present invention will be clearly and completely described below in conjunction with the embodiments. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all of the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.

[0015] Please refer to Figure 1 , this application provides a method for identifying safety risks in mine ecological restoration projects, and the method specifically includes the following steps:

[0016] I. Data fusion and risk assessment before operation; providing the necessary data basis for subsequent analysis;

[0017] Data fusion and risk assessment before operation refer to the data collection work of face recognition, dangerous area identification, meteorological disasters, landslides, rockfalls, etc. before workers enter the work area for operation;

[0018] Step 1, conduct data collection;

[0019] Face recognition data collection: Use high-definition face recognition devices to collect the facial features of workers before they enter the work area, and record data such as the identity information of workers;

[0020] Hazardous area identification data collection: Through sensors and image recognition systems installed in the work area, monitor the terrain, landform, equipment layout, etc. of the work area in real time, and identify data such as the location and scope of hazardous areas;

[0021] Meteorological disaster data collection: Use meteorological monitoring instruments to collect meteorological data such as wind speed, rainfall, temperature, and humidity in the work area;

[0022] Landslide and rockfall data collection: With the help of geological monitoring sensors, monitor the geological conditions in the work area and collect data related to landslides and rockfalls;

[0023] Step 2: Data integration and fusion;

[0024] Use the threshold method to screen and clean various types of collected data, and remove invalid and incorrect data;

[0025] Standardize the data after screening and cleaning to make it have a unified data format and measurement standard;

[0026] According to the following data fusion formula, fuse various types of processed data to obtain the fused data; The data fusion formula is as follows:

[0027] Assign a weight to various types of collected video image data. Various types of video image data include: face recognition data index Dr, hazardous area identification data index Dd, meteorological disaster data index Dw, landslide and rockfall data index Ds; The corresponding assigned weights are w r 、w d 、w w and w s, And satisfy: w r +w d +w w +w s = 1; Use the following method to fuse various types of video image data. The specific fusion method is:

[0028] Df = w r Dr + w d Dd + w w Dw + w s Ds;

[0029] The fused data index is Df ， Comprehensively reflects the dataset of various risk factors and is used to evaluate the overall safety risk level;

[0030] Calculate the risk assessment index R' according to the corresponding weights of the historical risk operation data index Dh, the operation guidance data index Dz, and the fusion data index Df, which are α, β, and γ respectively; the historical risk operation data includes historical data such as accident records, the number of dangerous situations, and the severity of accidents in history; the operation guidance data includes operation procedures, operation guides, safety standards, etc., which are used to guide operations and evaluate whether the operations meet safety standards

[0031] Then the risk assessment index R' is:

[0032] R' = αDh + βDz + γDf;

[0033] Step 3: Risk assessment

[0034] Determine the risk level according to the calculated risk assessment index R'.

[0035] Step 4: Risk warning and pre-control

[0036] According to the risk assessment result, when the risk assessment index exceeds the preset threshold, the system issues a risk warning signal.

[0037] Based on the risk warning result, formulate corresponding risk pre-control measures, such as adjusting the operation plan, increasing safety protection facilities, strengthening personnel training, etc., to reduce the operation risk.

[0038] II. Multi-source information fusion and modeling of construction operation safety risks provide deeper data understanding and context information for risk assessment by constructing a knowledge graph;

[0039] By sorting out and collecting data, pictures and other materials of projects such as mine ecological restoration, analyzing safety risks and influencing factors, and integrating multi-source information of construction operation safety risks, mainly aiming at the types of data in the information system, the data flow, business flow, and control flow affecting construction safety are respectively modeled uniformly.

[0040] Step 1: Data flow modeling. The generation of data is mainly related to cameras and sensors. Therefore, the control of the data flow mainly takes on-site videos and sensors as the core, and uses the device ID as the association field between data of the same device object to integrate the data of the unified device.

[0041] Step 2: Business flow modeling, which is mainly related to specific construction operations. Modeling work will be carried out around the construction stage and construction specifications, and the business stage number is used as the association field for different business handovers.

[0042] Step 3: Control flow modeling, which is mainly related to personnel decisions. Modeling will be carried out around personnel and specific decisions, and the user account that issues the decision is used as the association for decision data.

[0043] Step 4: After locking the device ID, business number information, and user account information, correlate video data, sensor data, business information, and personnel control information with each other to form a complete business loop.

[0044] Step 5: After establishing a complete unified naming and coding rule, standardize the naming of data information in the data flow, business flow, and control flow respectively, and improve the missing information and data structure.

[0045] Step 6: Adopt the PFCA model to conduct construction safety intelligent control modeling; aiming at realizing the specific functions of the project organization, with the business process as the core, using advanced information technology and modern management measures, maximize the integration of the organization in technology and management, establish a process-based organizational structure and an information-based communication channel, create a favorable organizational atmosphere, and make the project organization respond quickly and produce agilely.

[0046] III. Rapid acquisition and automatic identification of safety risks, using the results of the first two steps, conduct risk assessment and decision support to achieve precise management and control of construction risks;

[0047] 1) First, carry out a special risk assessment

[0048] Sort out the construction process of mine ecological restoration and common safety influencing factors, and fully understand the safety hazards in the construction process, such as personnel, equipment, materials, systems, and environment, etc.; the main content includes two parts: risk source identification and risk assessment.

[0049] Step 1: Risk source identification. Risk assessment personnel obtain relevant information and data and combine with on-site engineering investigations. Based on the expert investigation method and the system analysis method, the expert investigation method refers to that the administrator provides insights and predictions on potential risks by means of experience and understanding of the industry; the system analysis method is to adopt a systematic method to comprehensively analyze all components of the project, including physical equipment, operation processes, personnel behaviors, environmental factors, etc.;

[0050] Identify, analyze, and file possible risk sources or risk events.

[0051] Step 2: Summarize the construction safety risk assessment index system, select factors such as personnel, mechanical and electrical equipment, materials, management systems, and environment as the core factors of evaluation indicators, and obtain the evaluation index values of each risk factor according to the risk factor evaluation criteria, expert suggestions, and previous research experiences. Quantify the occurrence probability and impact degree of risk events to form a risk matrix, so as to rank and control risks.

[0052] Step 3: Risk assessment. Use the analytic hierarchy process for evaluation. According to the classification criteria of construction risk evaluation indicators and the opinions of consulting experts, obtain the evaluation index score values of construction risk factors. First, determine the evaluation indicators, which are the basis for evaluating the risk level. Then, through expert scoring or data analysis, compare each indicator pairwise to construct a judgment matrix

[0053] Next, calculate the eigenvalues and corresponding eigenvectors of the judgment matrix to obtain the weights of each indicator. Consult the opinions of experts: Combine the knowledge and experience of experts to evaluate the risk factors to obtain the preliminary score values of each risk factor. Substitute the quantified values after quantifying the directional concept into the forward cloud generator and repeatedly calculate to obtain the membership degree values under each risk level. According to a large number of sample data, determine the membership function through statistical methods. For example, if the value of an indicator is within a certain interval, its membership degree can be determined by statistically calculating the proportion of samples in this interval. The membership degree matrix is A, and then use the entropy weight method to determine the weight W of each evaluation indicator. Finally, substitute A and W into the designed comprehensive risk evaluation expression to calculate the risk value. The steps to determine the weight W of each evaluation indicator by the entropy weight method are as follows:

[0054] 1) Data standardization: Standardize the original data to eliminate the influence of dimensions between different indicators.

[0055] 2) Calculate information entropy: According to the standardized data, calculate the information entropy of each indicator to reflect the dispersion degree of each indicator.

[0056] 3) Calculate the weight: Calculate the weight of each indicator according to the information entropy. The greater the weight, the higher the importance of the indicator in the evaluation system. Standardize the original data to eliminate the influence of dimensions between different indicators.

[0057] The comprehensive risk evaluation expression is usually a weighted summation formula, in the following form:

[0058]

[0059] where, is the comprehensive risk value; n is the total number of indicators in the integrated indicators; is the weight of the i-th indicator, which can be determined by methods such as cross-validation; is the prediction result of the i-th indicator for the input feature set ; is the number of features considered in feature engineering; is the importance weight of the j-th feature, which can be determined by feature selection algorithms; is the transformation result of the j-th feature on the input dataset y, which can be a non-linear transformation; is the base of the natural logarithm. In the feature engineering stage, not only the original data but also data transformation and feature selection should be considered to enhance the prediction ability of the model:

[0060]

[0061] Among them, is the feature set after feature engineering processing; are the original features; are the new features obtained through feature selection and transformation, which can be polynomial features, interaction features, etc. In the index integration stage, multiple machine learning algorithms are used and weight allocation is performed for each index:

[0062]

[0063] Among them, is the accuracy rate of the performance index of the i-th index on the validation set; is the average value of all performance indicators.

[0064] Through the above formula, we can calculate the comprehensive risk value, so as to evaluate the risk level of the project;

[0065] 2) Detect and identify the unsafe behaviors of construction workers:

[0066] Step 1: Sort out the unsafe behaviors of construction workers. Classify typical construction unsafe behaviors. One is the construction unsafe behaviors caused by the on-site environment or social environment; the other is the unsafe factors in construction organization and management, which are mainly reflected in the process management of construction, the safety status management of construction materials, safety training, and safety investment; the third is the unsafe factors at the construction site, which are mainly reflected in the operation safety, clothing safety, and safety awareness of construction workers at the construction site.

[0067] Step 2: Use vision-based detection equipment for the illegal behaviors of construction workers. Segment the image into meaningful target region blocks, and apply specific neural network models, such as the YOLO series; identify the illegal behaviors of construction workers. By identifying some relatively representative and meaningful objects in these sub-block scenes, determine which category of behavior this type of scene belongs to, and mark the most likely illegal operation behavior in this area.

[0068] Step 3: Establish unsafe behaviors and their detection framework. By analyzing the main functional requirements of the video monitoring system during ecological restoration construction, count unsafe behaviors, and study image preprocessing algorithms suitable for ecological restoration construction. Mainly introduce the Retinex algorithm to comprehensively process construction images, study the description of functional points, analyze and design the system database, and finally develop the system through the OpenCV image recognition library to achieve the purpose of identifying unsafe behaviors.

[0069] 3) Conduct safety risk data fusion and risk hazard assessment

[0070] Step 1: The decision-making basis for business control, including various business management systems and information platforms, and the construction safety intelligent control model, is based on various current national and enterprise safety-related regulatory documents, etc. These are the data sources and interaction objects for intelligent control.

[0071] Step 2: First, use the data flow diagram (DFD) to represent the input, processing, and output of data, clarify the data input, output, and processing names for each processing step, and avoid over-refinement. At the same time, use the id-mapping technology to map data from different sources to a unified device identifier to associate the data flows of the same device object. Through the ETL process, we integrate the data from cameras and sensors into a central database for subsequent analysis and processing.

[0072] Then, apply natural language processing (NLP) technology to extract structured data, namely entities, relationships, and attributes, from unstructured text and store them in a graph database such as Neo4j to manage and query complex relationship data. Create indexes in the graph database to improve query efficiency.

[0073] On this basis, extract entities and relationships from the cleaned data, use deep learning models for named entity recognition (NER) and relationship extraction, and build a knowledge graph framework. Integrate knowledge from different sources through knowledge fusion technology to solve knowledge conflicts and redundancy problems. At the same time, link the extracted entities with the entities in the knowledge base to enrich and expand the knowledge graph.

[0074] Furthermore, perform semantic annotation on data instances, combined with manual semi-automatic keyword semantic annotation to enhance the semantic understanding of data. By defining the spatio-temporal data association state matrix, calculate the content object semantic association state, time association state, and space association state between instances to achieve the associated organization of multi-modal spatio-temporal data.

[0075] Finally, use deep learning techniques for relationship extraction and knowledge reasoning to identify potential risk patterns and associations. Build a risk assessment model, combine historical data and real-time data, evaluate the risk level, and formulate corresponding preventive measures. Automatically generate risk prevention plans, including adjusting the operation plan, increasing safety protection facilities, strengthening personnel training, etc., to reduce operation risks.

[0076] Step 3: Achieve intelligent extraction of multi-source information such as data flow, business flow, and control flow, so as to achieve accurate assessment of potential risks and provide decision-making assistance for risk prevention during construction.

[0077] Data collection: Collect data streams related to construction projects, including but not limited to environmental data, equipment status, personnel behavior, historical accident records, etc.; environmental data includes environmental parameters such as temperature, humidity, and wind speed, which are obtained in real time through sensors and normalized to eliminate the influence of dimensions; equipment status includes operation parameters, fault records, and maintenance logs of equipment; personnel behavior is obtained by collecting and tracking the behavior patterns of construction personnel through videos and sensors; historical accident records are extracted from the database, specifically including accident types, times, locations, and results.

[0078] Data integration: Integrate data from different sources to form a unified data warehouse.

[0079] Feature engineering: Extract features from the original data. The features include equipment failure frequency, number of personnel violations, etc.; these features may be related to risk assessment.

[0080] Risk model construction: Use machine learning or statistical methods to build a risk assessment model.

[0081] Risk assessment: Apply the risk model to evaluate potential risks during the construction process. The risk assessment model may use various algorithms, such as logistic regression, decision tree, random forest, etc. Its basic form is:

[0082] Risk score = log(1 / (1 - p)) = β0 + β1x1 + β2x2 +... + βnxn, where p is the probability of the occurrence of risk, β is the model parameter, and x is the feature.

[0083] Decision support: Provide decision support based on the risk assessment results, such as adjusting the construction plan, strengthening safety measures, etc.

[0084] The above embodiments are only used to illustrate the technical method of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those of ordinary skill in the art should understand that the technical method of the present invention can be modified or equivalently replaced without departing from the spirit and scope of the technical method of the present invention.

Claims

1. A safety risk identification method for mine ecological restoration projects, characterized in that, Including: Before workers enter the work area for operation, perform data fusion and risk assessment before the operation. First, carry out the processes of face recognition, dangerous area recognition, meteorological disaster, and landslide and rolling stone data collection. Then, fuse the collected data and conduct a preliminary risk assessment based on the fused data; Multi-source information fusion and modeling of construction operation safety risks. By sorting and collecting data and picture materials of mine ecological restoration projects, analyze safety risks and influencing factors, fuse multi-source information of construction operation safety risks, and conduct unified modeling of the data streams, business flows, and control flows affecting construction safety respectively according to the types of data in the information system; Rapid acquisition and automatic identification of safety risks. For the real-time construction process, identify potential safety hazards, distinguish existing risks, and use the analytic hierarchy process for evaluation. According to the classification standard of construction risk evaluation indicators, obtain the evaluation index score values of construction risk factors. The specific method is as follows: Carry out special risk assessment. Substitute the evaluation index score values of the preliminary construction risk factors entered by the administrator into the forward cloud generator, and repeatedly calculate to obtain the membership degree values under each risk level. The membership degree matrix is A, and then use the entropy weight method to determine the weights W of each evaluation index; Finally, substitute A and W into the designed comprehensive risk evaluation expression to calculate the risk value. The comprehensive risk evaluation expression is: ； Among them, is the comprehensive risk value; n is the total number of indicators in the integrated indicators; is the weight of the i-th indicator, determined by the cross-validation method; is the prediction result of the i-th indicator for the input feature set ; is the number of features considered in feature engineering; is the importance weight of the j-th feature, determined by the feature selection algorithm; is the conversion result of the j-th feature for the input data set y, which is the base of the natural logarithm; In the feature engineering stage, the predictive ability of the model also needs to be enhanced through data transformation and feature selection: ； Among them, is the feature set after feature engineering; is the original feature; is the new feature obtained through feature selection and transformation. The new features include polynomial features and interaction features; In the index integration stage, use a variety of machine learning algorithms and assign weights to each index: ； Among them, is the accuracy rate of the performance index of the i-th index on the validation set, is the average value of all index performance indicators; Simultaneously detect and identify the unsafe behaviors of construction workers, so as to conduct safety risk data fusion and risk hazard assessment.

2. The safety risk identification method for a mine ecological restoration project according to claim 1, wherein The specific method of data fusion and risk assessment before operation is as follows: Collect data to obtain face recognition data, dangerous area recognition data, meteorological disaster data, and landslide and rolling stone data; Screen and clean the collected various data to remove invalid and incorrect data; Standardize the data after screening and cleaning to make it have a unified data format and measurement standard; According to the following data fusion formula, fuse the processed various data to obtain the fused data. The fusion formula is: ； In the formula, the face recognition data index is Dr, the dangerous area recognition data index is Dd, the meteorological disaster data index is Dw, and the landslide and rockfall data index is Ds, and the weights are respectively , and the fused data index is Df; Conduct risk assessment based on the fused data.

3. The safety risk identification method for a mine ecological restoration project according to claim 2, characterized in that, The specific method of risk assessment is as follows: Input the fused data, historical risk operation data, and operation guidance data into the safety risk control system for intelligent behavior recognition; conduct assessment according to the risk assessment formula. The formula is specifically: ； In the formula, the risk assessment index is R’, and the weights corresponding to the historical risk operation data index Dh and the operation guidance data index Dz are α and β respectively, and γ is the weight of Df.

4. A method for identifying safety risks in a mine ecological restoration project according to claim 3, characterized in that, According to the risk assessment result, when the risk assessment index exceeds the preset threshold, the system issues a risk warning signal.

5. A method for identifying safety risks in a mine ecological restoration project according to claim 1, characterized in that, The specific method of multi-source information fusion and modeling of construction operation safety risks is as follows: Specifically conduct data stream modeling, business flow modeling, and control flow modeling. After locking the device ID, business number information, and user account information, correlate video data, sensor data, business information, and personnel control information with each other and form a complete business closed-loop; After establishing a complete unified naming and coding rule, standardize the naming of data information in the data flow, business flow, and control flow respectively, and improve the missing information and data structure; Adopt the PFCA model to build a construction safety intelligent control model.

6. The safety risk identification method for a mine ecological restoration project according to claim 1, characterized in that, The specific methods for rapid acquisition and automatic identification of safety risks are as follows: With the help of the YOLO series neural network model, detect and identify the unsafe behaviors of construction workers, and establish a framework for unsafe behaviors and their detection; by analyzing the main functional requirements of the video monitoring system for the ecological restoration construction process, count the unsafe behaviors, introduce the Retinex algorithm to comprehensively process the construction images, study the description of functional points, analyze and design the system database, and finally develop the system through the OpenCV image recognition library to achieve the purpose of identifying unsafe behaviors.

7. A method for identifying safety risks in a mine ecological restoration project according to claim 6, characterized in that The steps for determining the weight W of each evaluation index by the entropy weight method are as follows: Standardize the original data to eliminate the influence of dimensions between different indicators; According to the standardized data, calculate the information entropy of each index to reflect the dispersion degree of each index; Calculate the weight of each index according to the information entropy. The greater the weight, the higher the importance of the index in the evaluation system. Standardize the original data to eliminate the influence of dimensions between different indicators.

8. A method for identifying safety risks in a mine ecological restoration project according to claim 6, characterized in that, After the detection and identification of unsafe behaviors, it is also necessary to conduct safety risk data fusion and risk hazard assessment.

9. The safety risk identification method for a mine ecological restoration project according to claim 7, characterized in that, The specific methods for safety risk data fusion and risk hazard assessment are as follows: Fuse the data flow, business flow, and control flow, interact based on text information, establish a database based on text information, thus build a knowledge graph framework related to safety risks, and finally give the data analysis results and risk prevention plans through text semantic association and spatio-temporal coupling internal relations.

Images