Mine scheduling risk prediction method and system based on personnel and vehicle trajectory prediction

Abstract

The present invention relates to a method and system for predicting risks in mine scheduling based on prediction of personnel and vehicle trajectories, comprising: an information collection unit, a communication unit, a storage unit, a risk prediction unit and an analysis and presentation unit; the information collection unit effectively collects and processes the hazardous source information in the process of personnel and vehicle scheduling and stores it in the storage unit, and transmits the data signal to the risk prediction unit through the communication unit; the risk prediction unit performs data mining analysis, algorithm operation analysis and comprehensive prediction analysis on the collected information using a risk prediction model constructed by a risk prediction index system, and draws a risk prediction conclusion, which is finally sent to the analysis and presentation unit through the communication unit for comprehensive visual early warning presentation of the prediction result. The present invention can improve production efficiency, reduce costs, and enhance safety and environmental benefits by predicting the trajectories of personnel and vehicles in production scheduling, combining task allocation and real-time data monitoring.

Description

Technical Field

[0001] The present invention belongs to the technical field of mine dispatching, and in particular relates to a mine dispatching risk prediction method and system based on personnel and vehicle trajectory prediction. Background Art

[0002] The main responsibility of a mine dispatcher is to coordinate the production activities of a mine, including mining, transportation, mineral processing, etc. It is necessary to ensure the smooth implementation of the production plan, monitor the safety of the production site, coordinate equipment maintenance and material supply, etc.

[0003] The job responsibilities of a dispatcher in the mining industry mainly include the formulation and implementation of production plans to ensure that production tasks are completed on time. The dispatcher is responsible for monitoring and adjusting the production process, coordinating communication and collaboration between production departments, collecting, organizing and analyzing production data, and handling and responding to emergencies.

[0004] With the advancement of intelligent Internet of Things, more and more mines are now using intelligent scheduling of the Internet of Things. The intelligent mine production scheduling system is a software system that uses advanced technology and algorithms to optimize the planning and scheduling of mine production processes and improve production efficiency and resource utilization. The system integrates sensor technology, data mining, artificial intelligence and other technologies, and can monitor and analyze mine production data in real time, and perform intelligent production scheduling according to specified goals and constraints.

[0005] The dispatching system in the existing technology adopts intelligent dispatching, but it can only perform basic dispatching of vehicles, equipment, personnel and production tasks. During the dispatching process, the running trajectories of vehicles and personnel cannot be predicted, nor can the risks in the dispatching process be predicted, which makes the dispatching system unable to predict risks. Summary of the invention

[0006] Purpose of the invention: To provide a mine scheduling risk prediction method and system based on personnel and vehicle trajectory prediction to solve the above-mentioned problems.

[0007] In the first aspect, a mine dispatching risk prediction system based on personnel and vehicle trajectory prediction is provided, wherein the dispatching risk prediction system comprises: an information collection unit, a communication unit, a storage unit, a risk prediction unit and an analysis and presentation unit; the information collection unit effectively collects and processes the hazardous source information in the personnel and vehicle dispatching process and stores it in the storage unit, and transmits the data signal to the risk prediction unit through the communication unit; the risk prediction unit uses the risk prediction model constructed by the risk prediction index system to perform data mining analysis, algorithm operation analysis and comprehensive prediction analysis on the collected information, and draws a risk prediction conclusion, which is finally sent to the analysis and presentation unit through the communication unit, and a comprehensive visual early warning presentation of the prediction result is performed.

[0008] In a further embodiment, the information collection unit collects, aggregates and manages different types of data in the mine; the information collection unit automatically performs real-time statistical processing on the collected data information, and constructs a spatial database and an attribute database to provide data support for prediction and early warning in the risk prediction unit.

[0009] In a further embodiment, the risk prediction unit determines the risk level of the collected data by establishing a risk prediction model, and divides the scheduling risk level into several ranges, namely ;in represents the set of scheduling risk levels, Indicates the dispatch risk level from high to low; the input information is analyzed, judged and scored through the risk prediction model to determine the dispatch risk level. The lower the score, the higher the dispatch risk level; the higher the score, the lower the dispatch risk level.

[0010] In a further embodiment, the analysis and presentation unit establishes a three-dimensional mining scene consistent with reality through a three-dimensional model, presents basic mining information, analysis and evaluation information, prediction and warning information, and emergency resource information, and performs a visual analysis thereof; at the same time, the risk prediction results output by the risk prediction unit are displayed in the three-dimensional model.

[0011] In a further embodiment, the analysis and presentation unit comprises:

[0012] Comprehensive information visualization display module, used for comprehensive visualization of mine basic information, mine production status information, personnel and equipment information, and safety monitoring information;

[0013] The risk assessment and presentation module cleans, integrates and filters the data collected by the information collection unit, and obtains comprehensive information on mine risk accidents by combining the risk assessment grading table constructed by the risk prediction model, and uses deep learning to learn and update previous data;

[0014] A three-dimensional visualization display module for prediction results is used to display the prediction results output by the risk prediction unit in three dimensions, and to display the level, category, status details, and risk disposal information of the risk information in the form of a variety of thematic charts, and to display the mine risk distribution and risk information statistics in detail in a list format;

[0015] The emergency rescue resource display module is used to visualize mine risk rescue resources, rescue plans, rescue routes and rescue strategies.

[0016] In a second aspect, a mine scheduling risk prediction method based on personnel and vehicle trajectory prediction is implemented using the above-mentioned mine scheduling risk prediction system based on personnel and vehicle trajectory prediction, and the mine scheduling risk prediction method specifically includes the following steps:

[0017] Step S1, collecting data in the target mine area according to the scheduling task, including: collecting personnel and vehicle trajectory information, road condition collection, and production task collection; collecting the collected data and sending it to the storage unit and the risk prediction unit;

[0018] Step S2: perform risk classification and risk monitoring based on the collected data, screen and collect risk data based on the detection results of the dispatching operation risk, combine with standardization processing, build a risk warning model and output the prediction results;

[0019] Step S3: Display the prediction results through the analysis and presentation unit, and conduct risk analysis, deep learning and risk avoidance plans.

[0020] In a further embodiment, the specific steps of step S3 are as follows:

[0021] Step S21: After the collected data is input, the personnel and vehicle trajectories in the risk data are extracted and clustered to timely control the operation risk of the dispatch. The behavioral characteristics and attributes of the personnel and vehicle trajectories are abstracted into a labeled model. By establishing a label system, the freight locomotives are labeled to describe the real status of the dispatch operation in the form of labels.

[0022] Step S22, quantify the data after clustering processing, remove redundant features, extract the features of operation risks and normalize them;

[0023] Step S23: construct a scheduling operation risk prediction model, substitute the processed data into risk assessment and output the results.

[0024] In a further embodiment, in step S21, the clustering process specifically includes: firstly establishing a relationship between a target object label and a target object image, the target object includes a person and a vehicle, and obtaining a scheduling operation risk data label according to the operation data in the input data, that is:

[0025]

[0026] in, is the risk data label, Represents custom attributes and functional attributes, For the Risk indicators of the dispatching system, The number of risks for the scheduling operation.

[0027] In a further embodiment, the step S21 uses big data technology to obtain the scheduling operation risk data and performs clustering processing on it, including the following steps:

[0028] Step S211: for A dataset of scheduling operation risk samples , in the dataset Select Risk samples are used as the center of clustering , then the clustering categories of scheduling operation risks are: ;

[0029] Step S212: In the data set Calculate the risk sample data and Cluster processing centers The distance between them is calculated, and according to the label category corresponding to the target object image, the risk sample is divided into the category corresponding to the cluster center with the minimum distance;

[0030] Step S213: traverse all risk samples, and calculate the cluster center of each type of risk samples again after traversal;

[0031] Step S214, return to step S212, and re-execute the operations of step S212 and step S213 until the cluster center of each category of risk data no longer changes.

[0032] In a further embodiment, in step S21, the characteristics of the scheduling operation risk are extracted from the data that has completed the clustering process, so as to improve the efficiency of the scheduling operation risk detection;

[0033] First, perform data quantification: ; In the formula The minimum characteristics of the dispatch operation risk; is the maximum signal-to-noise ratio of risk data;

[0034] Secondly, the redundant features of risk data are removed, and the distribution characteristics of scheduling operation risks are predicted based on the probability of risk occurrence, namely: ,in is the weighted value of risk distribution; is the probability of the risk occurring;

[0035] According to the distribution characteristics of the dispatching operation risk, the risk characteristics are mined and the characteristics of the risk data are analyzed; and the data after the feature mining and analysis are subjected to risk detection. The specific steps are as follows:

[0036] Normalize the characteristics of scheduling operation risks:

[0037] ;

[0038] in Expressed as risk characteristics; is the minimum eigenvalue; is the maximum eigenvalue;

[0039] Determine the risk level of dispatching operations;

[0040] Introduce activation functions, calculate the output matrix of risk data in big data analysis, and use the output matrix to establish a scheduling operation risk prediction model;

[0041] Detect and schedule operational risks.

[0042] The present invention has the following advantages:

[0043] 1. Real-time data monitoring and early warning

[0044] The present invention combines real-time production data and equipment status to predict the risks of personnel and vehicles in scheduling. At the same time, the system can also achieve coordination and information sharing between tasks, thereby improving production efficiency and flexibility.

[0045] 2. Improve production efficiency and resource utilization

[0046] The system of the present invention can monitor and analyze the production data of the mine in real time, and issue early warnings and reminders in time. The system uses data collection equipment to monitor the personnel and vehicles in the mine dispatch in real time, and issues early warning information when danger occurs or is predicted, helping mine managers make timely decisions and adjustments to ensure production safety and efficiency;

[0047] 3. Reduce production costs and energy consumption

[0048] The system of the present invention can reduce the occurrence of risks, thereby preventing waste of resources, improving the utilization rate and production capacity of equipment, and maximizing the collection rate and output.

[0049] In general, the present invention can improve production efficiency, reduce costs, and enhance safety and environmental benefits by predicting the trajectories of personnel and vehicles in production scheduling, combining task allocation with real-time data monitoring. BRIEF DESCRIPTION OF THE DRAWINGS

[0050] Figure 1 It is a flow chart of the mine scheduling risk prediction system of the present invention.

[0051] Figure 2 It is a schematic diagram of the mine scheduling risk prediction system of the present invention.

[0052] Figure 3 It is a flow chart of the mine scheduling risk prediction method of the present invention.

[0053] Figure 4 It is a risk prediction flow chart in the mine scheduling risk prediction method of the present invention.

[0054] Figure 5 It is a flow chart of risk data clustering processing in step S21 of the mine scheduling risk prediction method of the present invention.

[0055] Figure 6 It is a risk data risk monitoring flow chart of step S21 in the mine scheduling risk prediction method of the present invention. DETAILED DESCRIPTION

[0056] The technical solution of the present invention will be described clearly and completely below 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 creative work are within the scope of protection of the present invention. Example 1

[0057] A mine scheduling risk prediction system based on personnel and vehicle trajectory prediction includes: an information collection unit, a communication unit, a storage unit, a risk prediction unit and an analysis and presentation unit; the information collection unit effectively collects and processes the hazard source information in the personnel and vehicle scheduling process and stores it in the storage unit, and transmits the data signal to the risk prediction unit through the communication unit; the risk prediction unit uses the risk prediction model constructed by the risk prediction index system to perform data mining analysis, algorithm operation analysis and comprehensive prediction analysis on the collected information, and draws a risk prediction conclusion, which is finally sent to the analysis and presentation unit through the communication unit for comprehensive visual early warning presentation of the prediction result.

[0058] In one embodiment, Figure 1 As shown, the information collection unit collects, aggregates and manages different types of data in the mine; the information collection unit automatically performs real-time statistical processing on the collected data information, and constructs a spatial database and an attribute database to provide data support for the prediction and early warning in the risk prediction unit.

[0059] In one embodiment, Figure 1 As shown, the risk prediction unit determines the risk level of the collected data by establishing a risk prediction model, and divides the scheduling risk level into several ranges, namely ;in represents the set of scheduling risk levels, Indicates the dispatch risk level from high to low; the input information is analyzed, judged and scored through the risk prediction model to determine the dispatch risk level. The lower the score, the higher the dispatch risk level; the higher the score, the lower the dispatch risk level.

[0060] In one embodiment, Figure 1 As shown, the analysis and presentation unit establishes a three-dimensional mine scene consistent with reality through a three-dimensional model, presents basic mine information, analysis and evaluation information, prediction and warning information, and emergency resource information, and performs visual analysis on them; at the same time, the risk prediction results output by the risk prediction unit are displayed in the three-dimensional model.

[0061] In one embodiment, Figure 1 As shown, the analysis and presentation unit includes:

[0062] Comprehensive information visualization display module, used for comprehensive visualization of mine basic information, mine production status information, personnel and equipment information, and safety monitoring information;

[0063] The risk assessment and presentation module cleans, integrates and filters the data collected by the information collection unit, and obtains comprehensive information on mine risk accidents by combining the risk assessment grading table constructed by the risk prediction model, and uses deep learning to learn and update previous data;

[0064] A three-dimensional visualization display module for prediction results is used to display the prediction results output by the risk prediction unit in three dimensions, and to display the level, category, status details, and risk disposal information of the risk information in the form of a variety of thematic charts, and to display the mine risk distribution and risk information statistics in detail in a list format;

[0065] The emergency rescue resource display module is used to visualize mine risk rescue resources, rescue plans, rescue routes and rescue strategies. Example 2

[0066] A mine scheduling risk prediction method based on personnel and vehicle trajectory prediction, the mine scheduling risk prediction method specifically comprises the following steps:

[0067] Step S1, collecting data in the target mine area according to the scheduling task, including: collecting personnel and vehicle trajectory information, road condition collection, and production task collection; collecting the collected data and sending it to the storage unit and the risk prediction unit;

[0068] Step S2: perform risk classification and risk monitoring based on the collected data, screen and collect risk data based on the detection results of the dispatching operation risk, combine with standardization processing, build a risk warning model and output the prediction results;

[0069] Step S3: Display the prediction results through the analysis and presentation unit, and conduct risk analysis, deep learning and risk avoidance plans.

[0070] In one embodiment, Figure 1 As shown, the specific steps of step S2 are as follows:

[0071] Step S21: After the collected data is input, the personnel and vehicle trajectories in the risk data are extracted and clustered to timely control the operation risk of the dispatch. The behavioral characteristics and attributes of the personnel and vehicle trajectories are abstracted into a labeled model. By establishing a label system, the freight locomotives are labeled to describe the real status of the dispatch operation in the form of labels.

[0072] Step S22, quantify the data after clustering processing, remove redundant features, extract the features of operation risks and normalize them;

[0073] Step S23: construct a scheduling operation risk prediction model, substitute the processed data into risk assessment and output the results.

[0074] In one embodiment, Figure 1 As shown, in step S21, the clustering process specifically includes: establishing a target object label and a target object image relationship, the target object includes personnel and vehicles, and obtaining a scheduling operation risk data label according to the operation data in the input data, that is:

[0075] ;

[0076] in, is the risk data label, Represents custom attributes and functional attributes, For the Risk indicators of the dispatching system, The number of risks for the scheduling operation.

[0077] In one embodiment, Figure 1 As shown, in step S21, the scheduling operation risk data is obtained by using big data technology and clustered, including the following steps:

[0078] Step S211: for A dataset of scheduling operation risk samples , in the dataset Select Risk samples are used as the center of clustering , then the clustering categories of scheduling operation risks are: ;

[0079] Step S212: In the data set Calculate the risk sample data and Cluster processing centers The distance between them is calculated, and according to the label category corresponding to the target object image, the risk sample is divided into the category corresponding to the cluster center with the minimum distance;

[0080] Step S213: traverse all risk samples, and calculate the cluster center of each type of risk samples again after traversal;

[0081] Step S214, return to step S212, and re-execute the operations of step S212 and step S213 until the cluster center of each category of risk data no longer changes.

[0082] In one embodiment, Figure 1 As shown, in step S21, the characteristics of scheduling operation risk are extracted from the data that has completed the clustering process, so as to improve the efficiency of scheduling operation risk detection;

[0083] First, perform data quantification: ; In the formula The minimum characteristics of the dispatch operation risk; is the maximum signal-to-noise ratio of risk data;

[0084] Secondly, the redundant features of risk data are removed, and the distribution characteristics of scheduling operation risks are predicted based on the probability of risk occurrence, namely: ,in is the weighted value of risk distribution; is the probability of the risk occurring;

[0085] According to the distribution characteristics of the dispatching operation risk, the risk characteristics are mined and the characteristics of the risk data are analyzed; and the data after the feature mining and analysis are subjected to risk detection. The specific steps are as follows:

[0086] Step 1: Normalize the characteristics of scheduling operation risks:

[0087] ;

[0088] in Expressed as risk characteristics; is the minimum eigenvalue; is the maximum eigenvalue;

[0089] Step 1: Normalize the characteristics of scheduling operation risks;

[0090] Step 3: Introduce the activation function, calculate the output matrix of risk data in big data analysis, and use the output matrix to establish a scheduling operation risk prediction model;

[0091] Step 4: Detect scheduling operation risks.

[0092] Obviously, the above embodiments are merely examples for clear explanation, and are not intended to limit the implementation methods. For those skilled in the art, other different forms of changes or modifications can be made based on the above description. It is not necessary and impossible to list all the implementation methods here. The obvious changes or modifications derived from them are still within the protection scope of the invention.

Claims

1. A mine scheduling risk prediction system based on personnel and vehicle trajectory prediction, characterized in that: The dispatch risk prediction system comprises: an information collection unit, a communication unit, a storage unit, a risk prediction unit and an analysis and presentation unit; The information collection unit collects and processes the information of the sources of danger in the process of personnel and vehicle dispatching and stores it in the storage unit, and transmits the data signal to the risk prediction unit through the communication unit; The risk prediction unit uses the risk prediction model constructed by the risk prediction index system to perform data mining analysis, algorithm operation analysis and comprehensive prediction analysis on the collected information, and draws a risk prediction conclusion, which is sent to the analysis and presentation unit through the communication unit to perform a comprehensive visual early warning presentation of the prediction result; The scheduling risk prediction system executes a mine scheduling risk prediction method; the mine scheduling risk prediction method specifically includes the following steps: Step S1, collecting data in the target mine area according to the scheduling task, including: collecting personnel and vehicle trajectory information, road condition collection, and production task collection; collecting the collected data and sending it to the storage unit and the risk prediction unit; Step S2: perform risk classification and risk monitoring based on the collected data, screen and collect risk data based on the detection results of the dispatching operation risk, combine with standardization processing, build a risk warning model and output the prediction results; the specific steps of step S2 are as follows: Step S21: After the collected data is input, the personnel and vehicle trajectories in the risk data are extracted and clustered to control the operational risk of the dispatch. The behavioral characteristics and attributes of the personnel and vehicle trajectories are abstracted into a labeled model. By establishing a label system, the freight locomotives are labeled to describe the real status of the dispatch operation in the form of labels. The clustering process specifically includes: establishing the relationship between the target object label and the target object image, the target objects include people and vehicles, and obtaining the scheduling operation risk data label based on the operation data in the input data, that is: ; in, is the risk data label, Represents custom attributes and functional attributes, For the The risk index of the dispatching system, N is the number of dispatching operation risks; Extract the characteristics of scheduling operation risks from the data that have completed clustering processing to improve the efficiency of scheduling operation risk detection; First, perform data quantification: ;in, Data for quantitative processing; The minimum characteristics of the dispatch operation risk; is the maximum signal-to-noise ratio of risk data; Secondly, the redundant features of risk data are removed, and the distribution characteristics of scheduling operation risks are predicted based on the probability of risk occurrence, namely: ,in is the weighted value of risk distribution; is the probability of the risk occurring; According to the distribution characteristics of the dispatching operation risk, the risk characteristics are mined and the characteristics of the risk data are analyzed; the data after feature mining and analysis are subjected to risk detection. The specific steps are as follows: Normalize the characteristics of scheduling operation risks: ; in Expressed as risk characteristics; is the minimum eigenvalue; is the maximum eigenvalue; Determine the risk level of dispatching operations; Introduce activation functions, calculate the output matrix of risk data in big data analysis, and use the output matrix to establish a scheduling operation risk prediction model; Detection and scheduling operation risks; Obtaining the scheduling operation risk data label includes the following steps: Step S211: for A dataset of scheduling operation risk samples , in the dataset Select Risk samples are used as the center of clustering , determine the clustering categories of dispatching operation risks; the clustering categories of dispatching operation risks are: ; Step S212: In the data set Calculate the risk sample data and Cluster processing centers The distance between them is calculated, and according to the label category corresponding to the target object image, the risk sample is divided into the category corresponding to the cluster center with the minimum distance; Step S213: traverse all risk samples, and calculate the cluster center of each type of risk samples again after traversal; Step S214, return to step S212, and re-execute the operations of step S212 and step S213 until the cluster center of each category of risk data no longer changes; Step S22, quantify the data after clustering processing, remove redundant features, extract the features of operation risks and normalize them; Step S23: construct a scheduling operation risk prediction model, substitute the processed data into risk assessment and output the results; Step S3: Display the prediction results through the analysis and presentation unit, and conduct risk analysis, deep learning and risk avoidance plans.

2. The mine scheduling risk prediction system based on personnel and vehicle trajectory prediction according to claim 1 is characterized in that: The information collection unit collects, aggregates and manages different types of data in the mine; The information collection unit automatically performs real-time statistical processing on the collected data information, and constructs a spatial database and an attribute database to provide data support for the prediction and early warning in the risk prediction unit.

3. The mine scheduling risk prediction system based on personnel and vehicle trajectory prediction according to claim 1 is characterized in that: The risk prediction unit determines the risk level of the collected data by establishing a risk prediction model, and divides the scheduling risk level into several ranges, namely ; in represents the set of scheduling risk levels, Indicates the scheduling risk level from high to low; The risk prediction model is used to analyze and judge the input information and score it to determine the dispatch risk level. The lower the score, the higher the dispatch risk level. The higher the score, the lower the scheduling risk level.

4. The mine scheduling risk prediction system based on personnel and vehicle trajectory prediction according to claim 1 is characterized in that: The analysis and presentation unit establishes a three-dimensional mine scene consistent with reality through a three-dimensional model, presents basic mine information, analysis and evaluation information, prediction and warning information, and emergency resource information, and performs visual analysis on them; and displays the risk prediction results output by the risk prediction unit in the three-dimensional model.

5. The mine scheduling risk prediction system based on personnel and vehicle trajectory prediction according to claim 4 is characterized in that: The analysis and presentation unit comprises: Comprehensive information visualization display module, used for comprehensive visualization of mine basic information, mine production status information, personnel and equipment information, and safety monitoring information; The risk assessment and presentation module cleans, integrates and filters the data collected by the information collection unit, and obtains comprehensive information on mine risk accidents by combining the risk assessment grading table constructed by the risk prediction model, and uses deep learning to learn and update previous data; A three-dimensional visualization display module for prediction results is used to display the prediction results output by the risk prediction unit in three dimensions, and to display the level, category, status details, and risk disposal information of the risk information in the form of a variety of thematic charts, and to display the mine risk distribution and risk information statistics in a list format; The emergency rescue resource display module is used to visualize mine risk rescue resources, rescue plans, rescue routes and rescue strategies.

Images