Failure risk management and control method of unmanned mine card system and related device

Abstract

The invention belongs to the technical field of mining industry safety, and discloses a failure risk management and control method of an unmanned mine card system and a related device, and the method comprises the steps: constructing a failure risk factor list of the unmanned mine card system; combining the unmanned mine card system failure cause list to construct a systematic failure cause classification system for unmanned truck control system failure cause identification; taking the systematic failure cause classification system as a node of a fuzzy Bayesian network, and constructing a visual Bayesian network based on a generative interaction environment; performing failure risk evaluation through a visual Bayesian network to obtain a risk evaluation result; and performing system failure countermeasure analysis according to a risk evaluation result, and obtaining an unmanned mine card system failure risk management and control measure based on a bow-tie theory. The safety system engineering theory and the analysis method are innovatively applied to the operation scene of the unmanned mine truck system, the blank in the field of reliability analysis of the unmanned mine truck system is filled, and a foundation is laid for safety analysis of the intelligent mine unmanned truck operation system.

Description

Technical Field

[0001] The present invention belongs to the field of mining safety technology, and specifically relates to a failure risk control method and related devices for an unmanned mining truck system. Background Art

[0002] With the rapid development of automation and intelligent technology, unmanned mining truck systems are increasingly used in mining operations. The system integrates advanced sensors, communication technologies, and control systems to achieve remote control and autonomous navigation of mining trucks, greatly improving the efficiency and safety of mining operations. However, due to the complex and diverse equipment involved in the unmanned mining truck system, the harsh environment of mine roads, and the lack of corresponding management experience on the part of the miners, many factors lead to the risk of failure of the unmanned mining truck system. Therefore, how to comprehensively identify and control the failure modes and key causes of the unmanned mining truck system is a technical problem that needs to be urgently solved in the deployment of unmanned mining truck applications. Summary of the invention

[0003] The purpose of the present invention is to overcome the shortcomings of the prior art and propose a failure risk management method and related devices for an unmanned mining truck system. The present invention takes a 5G unmanned mining truck system as the analysis object, comprehensively identifies and controls the failure modes and key causes of the unmanned mining truck system, and improves the safety of the unmanned mining truck system.

[0004] To achieve the above object, the present invention adopts the following technical solutions:

[0005] A failure risk control method for an unmanned mining truck system includes the following processes:

[0006] Construct a list of risk factors for failure of unmanned mining truck systems;

[0007] Combined with the unmanned mining truck system failure cause list, a systematic failure cause classification system (SFCCS) for identifying failure causes of unmanned truck control systems is constructed;

[0008] The systematic failure cause classification system is used as a node of a fuzzy Bayesian network, and a visual Bayesian network is constructed based on a generative interactive environment;

[0009] Failure risk assessment is performed through visual Bayesian network to obtain risk assessment results;

[0010] According to the risk assessment results, a system failure countermeasure analysis is carried out, and the failure risk control measures of the unmanned mining truck system are obtained based on the bowtie theory.

[0011] Preferably, the process of constructing a list of risk factors for failure of an unmanned mining truck system includes:

[0012] Determine the organizational structure and workflow of the unmanned mining truck system, and build the control feedback structure of the unmanned mining truck system based on the organizational structure and workflow;

[0013] Determining system-level accidents and system-level hazards according to the unmanned mining truck system control feedback structure;

[0014] Identify unsafe control behaviors from the control process and feedback process that cause system-level accidents and system-level hazards, and obtain the causal factors that lead to unsafe control behaviors based on the unsafe control behaviors;

[0015] The causal factors are classified to obtain a list of risk factors for failure of the unmanned mining truck system.

[0016] Preferably, the causal factors are classified into four categories: management, people, things and environment, where:

[0017] For management, the causal factors include monitoring and supervision, safety hazards, safety atmosphere, enterprise management, site management, emergency management, system construction, personnel recruitment, safety training and line design;

[0018] For people, the causal factors include theoretical level, physical and mental state, practical experience, technical level, communication level, operational errors and operational violations;

[0019] For things, the causal factors include hardware equipment quality, software system algorithms, network delay failures, and data transmission errors;

[0020] For the environment, the causative factors include severe weather, extreme temperatures, rugged mountains, and harsh conditions, among which harsh environments include dusty environments or corrosive environments.

[0021] Preferably, in combination with the unmanned mining truck system failure cause list, the process of constructing a systematic failure cause classification system for unmanned truck control system failure cause identification includes:

[0022] Combined with the unmanned mine truck system failure cause list, the unmanned mine truck system failure accidents are subdivided into several levels of factors, each of which contains causal factors, and the relationship between the causal factors is hierarchically divided, ultimately forming a systematic failure cause classification system.

[0023] Preferably, the hierarchical factors include seven factors: organizational influence, unsafe supervision, preconditions of unsafe behavior, unsafe behavior of personnel, technical system factors, environmental factors and external factors; organizational influence includes three subcategories: improper resource management, bad organizational process and bad safety atmosphere; unsafe supervision includes two subcategories: wrong supervision and illegal supervision; preconditions of unsafe behavior include three subcategories: unqualified personal conditions, unqualified personnel practices and environmental factors; unsafe behavior includes two subcategories: wrong behavior and illegal behavior; technical system factors include two subcategories: system function failure and system hardware failure; environmental factors include two subcategories: natural environment and operating environment; external factors include two subcategories: external interference and external attack;

[0024] The relationship between the causal factors is divided into levels, and the results are as follows:

[0025] Improper resource management includes four factors: insufficient pertinence of safety training, insufficient investment in safety funds, insufficient emergency resources and insufficient investment in safety equipment; organizational processes include three factors: imperfect safety management system, unqualified safety protection and unreasonable employee performance appraisal; organizational atmosphere includes two factors: inadequate implementation of safety management and poor safety atmosphere; incorrect supervision includes three factors: ineffective government supervision, improper supervision plan and failure to correct unsafe behaviors in a timely manner; illegal supervision includes two factors: daily violations and individual violations; unqualified personal conditions include three factors: poor psychological or physiological state, unqualified professional skills and weak safety awareness; unqualified personnel practices include three factors: insufficient training, insufficient emergency response capabilities and insufficient personal preparation; environmental factors include four factors: physical environment, technical environment, organizational environment and human environment; incorrect behaviors include three factors: decision-making errors, skill errors and perception errors; illegal behaviors include three factors: habitual violations, accidental violations and risky operations; natural environment includes five factors: extreme weather, extreme temperature, dust particles, sunlight and rain; operating environment includes two factors: rugged roads and regional obstacles; external interference includes two factors: electromagnetic interference and wireless signal interference; external attacks are mainly caused by data tampering or interception.

[0026] Preferably, the systematic failure cause classification system is used as a node of a fuzzy Bayesian network, and the process of constructing a visual Bayesian network based on a generative interactive environment includes:

[0027] Each causal factor of the systematic failure cause classification system is taken as a node of the fuzzy Bayesian network, and a topological structure is constructed. The states of the nodes in the topological structure are represented by occurrence and non-occurrence.

[0028] Based on expert knowledge, the probability of event occurrence at each node is obtained, and the probability of event occurrence is described by seven levels: very low, low, relatively low, medium, relatively high, high, and very high;

[0029] The triangular fuzzy number is selected as the membership function for analyzing the accident events. According to the corresponding relationship between the level and the probability of the event, the natural language of the experts is converted into quantitative probability values, and the fuzzy Bayesian network model is constructed.

[0030] The nodes of the fuzzy Bayesian network model are created in the generative interactive environment software; then directed arcs are drawn to connect the nodes; then the prior probability of the root node and the conditional probability of the child nodes are set, and then the probability of each node is updated to obtain the visual Bayesian network.

[0031] Preferably, a system failure countermeasure analysis is performed according to the risk assessment result, and a process of obtaining unmanned mining truck system failure risk control measures based on the bowtie theory includes:

[0032] Combined with the risk assessment results, a bow-tie theoretical model of unmanned mining truck system failure is established: the failure of the unmanned mining truck system is taken as the top event and placed in the center of the bow tie; then the causes of the accident are listed on the left side of the bow tie, and the consequences of the accident are listed on the right side; the cause of the accident on the left side of the bow tie is taken as the event hazard source, the left side of the bow tie is connected with the top event, obstacles are placed on the connection path between the hazard source and the event, and preventive measures are set for the hazard source to be triggered and induce the accident; on the right side of the bow tie, starting from the accident, lines are drawn between the consequences, and control measures are set on the path where the impact of the consequences is further expanded;

[0033] Through the failure tie theoretical model of unmanned mining truck system, for each failure cause, pre-emptive prevention measures and post-event mitigation measures are proposed from four dimensions: staff, operation management, system equipment and environmental conditions.

[0034] The present invention also provides a failure risk management system of an unmanned mining truck system, which is used to implement the failure risk management method of the unmanned mining truck system of the present invention. The system includes:

[0035] List building unit: used to build a list of risk factors for failure of unmanned mining truck systems;

[0036] The first model building unit is used to build a systematic failure cause classification system for identifying failure causes of unmanned truck control systems in combination with the failure cause list of the unmanned mining truck system;

[0037] A second model building unit is used for taking the systematic failure cause classification system as a node of a fuzzy Bayesian network and building a visual Bayesian network based on a generative interactive environment;

[0038] Evaluation unit: used to evaluate failure risk through a visual Bayesian network and obtain risk evaluation results;

[0039] Risk control unit: used to analyze system failure countermeasures according to the risk assessment results, and obtain unmanned mining truck system failure risk control measures based on the bowtie theory.

[0040] The present invention also provides an electronic device, comprising:

[0041] one or more processors;

[0042] a storage device having one or more programs stored thereon;

[0043] When the one or more programs are executed by the one or more processors, the one or more processors implement the failure risk management method of the unmanned mining truck system of the present invention as described above.

[0044] The present invention also provides a storage medium having a computer program stored thereon, wherein when the computer program is executed by a processor, the failure risk management method of the unmanned mining truck system of the present invention as described above is implemented.

[0045] The technical solution provided by the present invention has the following beneficial effects:

[0046] The present invention innovatively applies the theory and analysis methods of safety system engineering to the operation scenario of the unmanned mine truck system, fills the gap in the field of reliability analysis of the unmanned mine truck system, and lays a theoretical, path and technical foundation for the safety analysis of the operation system of unmanned trucks in smart mines. In addition, the present invention constructs a list of failure risk factors of the unmanned mine truck system based on the structure and characteristics of the unmanned mine truck system, and establishes a systematic failure cause classification system for the identification of failure causes of unmanned truck control systems; on the other hand, it combines expert safety evaluation methods, fuzzy mathematics theory and Bayesian networks, which not only ensures the authority of the evaluation and provides implementation support, but also eliminates the subjectivity of judgment to a certain extent, ensuring the accuracy and effectiveness of the Bayesian risk evaluation model. At the same time, the bow-tie risk analysis method is innovatively applied to the failure prevention and control of the unmanned mine truck system, providing an effective failure risk management solution for the unmanned mine truck system. BRIEF DESCRIPTION OF THE DRAWINGS

[0047] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the drawings required for use in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0048] Figure 1 It is a flow chart of the failure risk control method of the unmanned mining truck system of the present invention;

[0049] Figure 2Schematic diagram of control feedback structure of unmanned mining truck system in an embodiment of the present invention;

[0050] Figure 3 Schematic diagram of a failure accident of an unmanned mining truck system in an embodiment of the present invention;

[0051] Figure 4 Schematic diagram of the BN model of the unmanned mining truck system failure in an embodiment of the present invention;

[0052] Figure 5 is a schematic diagram of the failure probability reasoning of the unmanned mining truck system in an embodiment of the present invention;

[0053] Figure 6 is a schematic diagram of the collision sensitivity analysis of an unmanned truck in an embodiment of the present invention;

[0054] Figure 7 It is a schematic diagram of the BOW-tie model of the unmanned mining truck system in an embodiment of the present invention. DETAILED DESCRIPTION

[0055] The specific implementation modes of the present invention are described in detail below in conjunction with the accompanying drawings.

[0056] See also Figure 1 The failure risk control method of the unmanned mining truck system of the present invention comprises the following steps:

[0057] S1. Build a list of risk factors for failure of unmanned mining truck systems from four aspects: management, people, objects and environment; specifically including:

[0058] Determine the organizational structure and workflow of the unmanned mining truck system, and build the control feedback structure of the unmanned mining truck system;

[0059] According to the control feedback structure of the unmanned mining truck system, four system-level accidents (i.e., unmanned truck collision, unmanned truck rollover, unmanned truck damage, and unmanned truck loss) and system-level hazards (i.e., loss of vehicle control, ground communication interruption, vehicle deviation from the route, and vehicle external impact) are identified, see Tables 1 and 2;

[0060] Unsafe control behaviors were identified from 16 control processes and 12 feedback processes, and the causal factors leading to unsafe control behaviors were analyzed. The results are shown in Table 3;

[0061] According to the four categories of management, people, objects and environment, the causal factors of the unmanned mine truck system are initially integrated and divided into four categories with a total of 25 risk factors, and a list of risk factors for failure of the unmanned mine truck system is constructed. The results are shown in Table 4;

[0062] S2. Combined with the list of failure causes of unmanned mining truck system, a systematic failure cause classification system (SFCCS) for failure cause identification of unmanned truck control system is constructed; specifically, it includes:

[0063] Combined with the list of causes of failure of unmanned mining truck system, the failure accidents of unmanned mining truck system are subdivided into 7 levels (i.e. organizational influence, unsafe supervision, preconditions of unsafe behavior, unsafe behavior of personnel, technical system factors, environmental factors and external factors), with a total of 42 factors. The relationship between the causal factors is hierarchically divided, and finally a systematic failure cause classification system for failure cause identification of SFCCS unmanned truck control system is formed. Figure 3 ;

[0064] S3. The systematic failure cause classification system is used as the node of the fuzzy Bayesian network, and a visual Bayesian network is constructed based on GeNIe (generative interactive environment), and failure risk assessment is performed; specifically, the following are included:

[0065] Each causal factor of the systematic failure cause classification system for failure cause identification of unmanned truck control system is used as a node of the fuzzy Bayesian network;

[0066] Fuzzy language variables are introduced to describe the expert's judgment results, and the probability of this fuzzy event is described in seven levels: "very low, low, relatively low, medium, relatively high, high and very high";

[0067] Furthermore, triangular fuzzy numbers are selected as the membership function for analyzing accident events, and the formula is as follows:

[0068]

[0069] in, is a triangular fuzzy number, x is a possible value in the middle, a and b are the lower and upper limits of the triangular fuzzy number respectively, and n is the value of the membership function when the membership degree is 1.

[0070] Convert qualitative expert knowledge into quantitative probability values ​​and construct a fuzzy Bayesian network model;

[0071] Based on the GeNIe software, a visual Bayesian network was constructed, and the Bayesian reasoning calculation and analysis were performed through formula 1 to obtain the risk probability of each accident cause.

[0072]

[0073] Based on the main inducing path analysis of four system-level accidents, the main inducing path of each accident is determined in turn according to the probability, and the sensitivity coefficients of the associated nodes are compared to determine the key factors leading to the occurrence of the target node;

[0074] The main influencing factors of unmanned mining truck system failure accidents were summarized, and finally 16 key control factors were obtained. Figure 3 ;

[0075] S4. According to the risk assessment results, the system failure countermeasures are analyzed, and the unmanned mining truck system failure risk control measures are proposed based on the Bow-tie theory, including:

[0076] Combining 42 causal factors and 16 key control factors of unmanned mining truck system failure, a Bow-tie analysis model of unmanned mining truck system failure is established;

[0077] By using the Bow-tie analysis model of unmanned mining truck system failure, preventive measures are proposed for each failure cause from four dimensions: staff, operation management, system equipment and environmental conditions. At the same time, control measures are proposed for the four accident consequences.

[0078] Example

[0079] like Figure 1 As shown, the failure risk control method of the unmanned mining truck system of this embodiment includes the following steps:

[0080] S1, build a list of risk factors for failure of unmanned mining truck system from four aspects: management, people, objects and environment:

[0081] Furthermore, the control feedback structure of the unmanned mining truck system is constructed, including:

[0082] The unmanned mining truck system is regarded as a control model. According to the relevant specifications of the unmanned mining truck and the actual situation of the unmanned mining truck system at the operation site, the unmanned mining truck system is analyzed, including the elements of enterprise managers, unmanned mining truck suppliers, professional and technical personnel, safety managers, maintenance personnel, unmanned mining trucks and auxiliary mechanical equipment.

[0083] The unmanned mining truck operation system includes three parts: vehicle-mounted unmanned driving subsystem, communication management subsystem and ground management subsystem. Combined with the field survey, the perception system, monitoring system, electrical system, power operation system and some important components are added to form the following Figure 2 The unmanned mining truck system control feedback structure is shown.

[0084] According to the established control feedback structure of the unmanned mining truck system, four system-level hazards are identified. The results are shown in Table 1:

[0085] Table 1

[0086] Serial number Danger <![CDATA[H 1 ]]> Loss of vehicle control <![CDATA[H 2 ]]> Ground communications disrupted <![CDATA[H 3 ]]> Vehicle deviates from the route <![CDATA[H 4 ]]> The vehicle is hit by external impact

[0087] System-level accidents are determined by causing overall system failure instead of causing losses in the system. System-level accidents are recorded as L and are divided into four types. The correlation between system-level hazards and system-level accidents is shown in Table 2:

[0088] Table 2

[0089] System-level incidents System-level dangers <![CDATA[Driverless truck collision L 1 > <![CDATA[H 1 ,H 2 ,H 3 ,H 4 ]]> <![CDATA[Unmanned truck rollover L 2 > <![CDATA[H 1 ,H 3 ,H 4 ]]> <![CDATA[Autonomous truck damaged L 3 > <![CDATA[H 1 ,H 2 ,H 3 ,H 4 ]]> <![CDATA[Driverless truck loses L 4 > <![CDATA[H 2 ,H 3 ]]>

[0090] Unsafe control behaviors are divided into four categories: no control behavior, incorrect control behavior, wrong control behavior time or wrong order, and incorrect control behavior duration. Unsafe control behaviors in the interaction of unmanned mining truck system components are identified. The unsafe control behaviors in the control process are shown in Table 3, and the unsafe control behaviors in the feedback process are shown in Table 4.

[0091] Table 3

[0092]

[0093]

[0094]

[0095] Table 4

[0096]

[0097]

[0098] By analyzing the causal factors of system failure from the unsafe control behaviors obtained and merging them with the causal factors of external influences on the system, we can obtain a list of 25 factors of unmanned mining truck system failure causes as shown in Table 5.

[0099] Table 5

[0100]

[0101]

[0102] S2, combined with the list of failure causes of unmanned mining truck system, builds a systematic failure cause classification system (SFCCS) for failure cause identification of unmanned truck control system:

[0103] based on Figure 3 It can be obtained that there are seven major factors that contribute to the failure risk of unmanned mining truck systems, including organizational influence, unsafe supervision, preconditions for unsafe behavior, unsafe behavior of personnel, technical system factors, environmental factors, and external factors. Organizational influence is divided into three subcategories: improper resource management, poor organizational process, and poor safety atmosphere; unsafe supervision is divided into two subcategories: wrong supervision and illegal supervision; preconditions for unsafe behavior are divided into three subcategories: unqualified personal conditions, unqualified personnel practices, and environmental factors; unsafe behavior is divided into two subcategories: wrong behavior and illegal behavior; technical system factors are divided into two subcategories: system function failure and system hardware failure; environmental factors are divided into two subcategories: natural environment and operating environment; external factors are divided into two subcategories: external interference and external attack.

[0104] Combined with the list of causes of failure of unmanned mining truck system, each subcategory is further subdivided: improper resource management is divided into four factors: insufficient pertinence of safety training, insufficient investment in safety funds, insufficient emergency resources and insufficient investment in safety equipment; organizational process is divided into three factors: imperfect safety management system, unqualified safety protection and unreasonable employee performance appraisal; organizational atmosphere is divided into two factors: inadequate implementation of safety management and poor safety atmosphere; wrong supervision is divided into three factors: ineffective government supervision, improper supervision plan and failure to correct unsafe behavior in time; illegal supervision is divided into two factors: daily violations and individual violations; unqualified personal conditions are divided into poor psychological or physiological state, unqualified professional skills and safety The three factors of weak full awareness are as follows: insufficient training, insufficient emergency response ability and insufficient personal preparation; environmental factors are divided into four factors: physical environment, technical environment, organizational environment and human environment; wrong behavior is divided into three factors: decision-making errors, skill errors and perception errors; violations are divided into three factors: habitual violations, accidental violations and risky operations; the natural environment is divided into five factors: extreme weather, extreme temperature, dust particles, sun exposure and rain; the operating environment is divided into two factors: rugged roads and regional obstacles; external interference is divided into two factors: electromagnetic interference and wireless signal interference; external attacks are mainly one factor: data tampering or interception. Based on the hierarchical division of 42 causal factors, a systematic failure cause classification system (SFCCS) for failure cause identification of unmanned truck control systems is constructed.

[0105] S3, taking the systematic failure cause classification system as the node of the fuzzy Bayesian network, constructing a visual Bayesian network based on the GeNIe software, and performing failure risk assessment.

[0106] Furthermore, a fuzzy Bayesian network model is constructed, including:

[0107] Each causal factor of the systematic failure cause classification system is taken as a node of the fuzzy Bayesian network, and a topological structure is constructed. Figure 4 Among them, the status of the node is represented by occurrence (Y) and non-occurrence (N).

[0108] The probability values ​​in the Bayesian network model are obtained based on expert knowledge. The parameters of the model include the initial probability of the root node and the conditional probability of the intermediate nodes. The probability of occurrence of each node is calculated by conducting a questionnaire survey on experts. The experts use seven language variables, namely very low (VL), low (L), low (FL), medium (M), high (FH), high (H) and very high (VH), to represent the probability of occurrence of the node, and use triangular fuzzy numbers as the membership function for analyzing the occurrence of accidents. The natural language of the experts is converted into triangular fuzzy numbers according to the corresponding relationship.

[0109] Furthermore, the fuzzy opinions of five experts were collected according to the expert survey method, and the SAM method was used to integrate the expert opinions. The similarity of the expert opinions was evaluated, the average consistency, relative consistency and consistency coefficient of the expert opinions were calculated, the fuzzy opinions of the experts were summarized, and the mean area method was used to defuzzify the expert opinions, so as to obtain the prior probability of the node. Table 6 shows an example of expert opinion integration for the organizational atmosphere node, and Table 7 shows an example of conditional probability for system function failure.

[0110] Table 6

[0111]

[0112]

[0113] Table 7

[0114]

[0115] Using GeNIe software to construct a visual Bayesian network includes the following steps:

[0116] Step 1, create nodes: draw a total of 5 network nodes (or called nodes), including system hardware failure, system function failure, technical system factors, vehicle deviation from the route, and unmanned truck loss.

[0117] Step 2, draw directed arcs: connect the five network nodes according to the four sets of causal relationships;

[0118] Step 3, set the prior probability of the root node: input the probability of the node factor not appearing (NO) and appearing (YES);

[0119] Step 4, set the conditional probability of the child node: input the probability of the child node occurring or not when one or more parent nodes appear or not, and ensure that the sum of the probabilities in each column is 1;

[0120] Step 5, network update: Update the probability of each node in the Bayesian network, and perform Bayesian reasoning analysis based on this. The failure probability reasoning of the unmanned mining truck system is obtained as follows: Figure 5 shown.

[0121] Based on the established visual Bayesian network, reverse reasoning is performed to obtain the probability of other network node factors appearing when each system-level accident occurs, so as to determine the main inducing path of the system-level accident in turn according to the probability. The main process is as follows:

[0122] (1) The node of a certain system-level accident is set to the state of inevitable occurrence, that is, YES=100%, and the other nodes of the layer are set to YES=0%, indicating that the other nodes are in a state of non-occurrence;

[0123] (2) Update the network model to obtain the probability of occurrence of each node factor related to the system-level accident under the condition that the system-level accident is bound to occur;

[0124] (3) Starting from each factor node, determine the parent node with the largest posterior probability upward, thus forming the main inducing path of the accident.

[0125] Furthermore, the system-level accidents are taken as target nodes in turn, and the sensitivity analysis tool is called to obtain Bayesian nodes of different shades. The redder the node, the higher the sensitivity, and the gray represents no correlation. By performing sensitivity analysis on the four target nodes, the sensitivity level is obtained. Figure 6 An example of sensitivity analysis for an unmanned truck crash is shown.

[0126] According to the color depth of each node variable in the Bayesian network diagram, a sensitivity coefficient table is established, and the sensitivity intensity is roughly divided into four levels. Nodes with a sensitivity coefficient ≥ 0.1 are displayed in dark red and can be defined as highly sensitive nodes; nodes with a sensitivity coefficient ≥ 0.05 and < 0.1 are displayed in light red and are defined as medium sensitive nodes; nodes with a sensitivity coefficient ≥ 0.02 and < 0.05 are displayed in light red and are low sensitive nodes; nodes with a sensitivity coefficient < 0.02 are almost white and are considered insensitive nodes. When the unmanned truck collision is the target node, loss of vehicle control, ground communication interruption, vehicle deviation from the route, technical system factors and external factors are highly sensitive nodes, and unsafe behavior of personnel is a medium sensitive node.

[0127] By analyzing the main accident-inducing paths and factor sensitivity, the main influencing factors of unmanned mining truck system failure accidents were analyzed and summarized, and 16 key control factors including the control of unsafe personnel behaviors were obtained.

[0128] S4, conduct system failure countermeasures analysis based on risk assessment results, and propose unmanned mining truck system failure risk control measures based on Bow-tie theory.

[0129] Furthermore, construct Figure 7 The Bow-tie analysis model of unmanned mining truck system failure is shown in the figure. The main process is as follows:

[0130] (1) The failure of the unmanned mining truck system is taken as the top event and placed in the center of the bow tie.

[0131] (2) List the causes of the accident on the left side of the bow tie and the consequences of the accident on the right side.

[0132] (3) The cause of the accident on the left side of the bow tie is taken as the event hazard source, and the accident hazard source is connected to the top event. Obstacles are placed on the connection path between the hazard source and the event, and "preventive measures" are set up for the hazard source to be triggered and induce the accident, so as to control the occurrence of the hazard source and reduce the possibility of the occurrence of the accident hazard source.

[0133] (4) On the right side of the bow tie, take the accident as the starting point and draw lines between each of the consequences. Set up obstacles on the path where the impact of the consequences will further expand. That is, set up "control measures" to reduce the possibility of the event results and the severity of the consequences.

[0134] In view of the 42 causal factors and 16 key control factors of the failure of unmanned mining truck systems, we propose pre-emptive prevention measures and post-event mitigation measures according to the four categories of people, machines, management and environment. On the one hand, we reduce the possibility of accidents, and on the other hand, we reduce the losses caused by accidents and prevent the further expansion of accident hazards.

[0135] The present invention also provides a system for implementing the failure risk control method of the unmanned mining truck system provided in any of the above embodiments, the system comprising:

[0136] List building unit: used to build a list of risk factors for failure of unmanned mining truck systems;

[0137] The first model building unit is used to build a systematic failure cause classification system (SFCCS) for unmanned truck control system failure cause identification in combination with the unmanned mining truck system failure cause list;

[0138] A second model building unit is used for taking the systematic failure cause classification system as a node of a fuzzy Bayesian network and building a visual Bayesian network based on a generative interactive environment;

[0139] Evaluation unit: used to evaluate failure risk through a visual Bayesian network and obtain risk evaluation results;

[0140] Risk control unit: used to analyze system failure countermeasures according to the risk assessment results, and obtain unmanned mining truck system failure risk control measures based on the bowtie theory.

[0141] The embodiments of the present invention also provide corresponding electronic devices and computer-readable storage media for implementing the solutions provided by the embodiments of the present invention.

[0142] Among them, the device includes a memory and a processor, the memory is used to store instructions or codes, and the processor is used to execute the instructions or codes, so that the device executes the failure risk management method of the unmanned mining truck system described in any embodiment of the present application.

[0143] A computer program is stored on the storage medium, wherein the computer program, when executed by the processor, implements the failure risk management method of the unmanned mining truck system described in any embodiment of the present application.

[0144] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the above embodiments, ordinary technicians in the relevant field should understand that the specific implementation methods of the present invention can still be modified or replaced by equivalents. Any modification or equivalent replacement that does not depart from the spirit and scope of the present invention should be covered within the scope of protection of the claims of the present invention.

Claims

1. A failure risk management method for an unmanned mining truck system, characterized in that: The process includes the following: Construct a list of risk factors for failure of unmanned mining truck systems; Combined with the unmanned mining truck system failure cause list, a systematic failure cause classification system for unmanned truck control system failure cause identification is constructed; The systematic failure cause classification system is used as a node of a fuzzy Bayesian network, and a visual Bayesian network is constructed based on a generative interactive environment; Failure risk assessment is performed through visual Bayesian network to obtain risk assessment results; According to the risk assessment results, a system failure countermeasure analysis is carried out, and the unmanned mining truck system failure risk control measures are obtained based on the bowtie theory.

2. The failure risk management method of an unmanned mining truck system according to claim 1 is characterized in that: The process of building a list of risk factors for failure of unmanned mining truck systems includes: Determine the organizational structure and workflow of the unmanned mining truck system, and build the control feedback structure of the unmanned mining truck system based on the organizational structure and workflow; Determining system-level accidents and system-level hazards according to the unmanned mining truck system control feedback structure; Identify unsafe control behaviors from the control process and feedback process that cause system-level accidents and system-level hazards, and obtain the causal factors that lead to unsafe control behaviors based on the unsafe control behaviors; The causal factors are classified to obtain a list of risk factors for failure of the unmanned mining truck system.

3. The failure risk management method of an unmanned mining truck system according to claim 2 is characterized in that: The causal factors are classified into four categories: management, people, things and environment, among which: For management, the causal factors include monitoring and supervision, safety hazards, safety atmosphere, enterprise management, site management, emergency management, system construction, personnel recruitment, safety training and line design; For people, the causal factors include theoretical level, physical and mental state, practical experience, technical level, communication level, operational errors and operational violations; For things, the causal factors include hardware equipment quality, software system algorithms, network delay failures, and data transmission errors; For the environment, the causative factors include severe weather, extreme temperatures, rugged mountains, and harsh conditions, among which harsh environments include dusty environments or corrosive environments.

4. The failure risk management method of an unmanned mining truck system according to claim 1 is characterized in that: In combination with the unmanned mining truck system failure cause list, the process of constructing a systematic failure cause classification system for unmanned truck control system failure cause identification includes: Combined with the unmanned mine truck system failure cause list, the unmanned mine truck system failure accidents are subdivided into several levels of factors, each of which contains causal factors, and the relationship between the causal factors is hierarchically divided, ultimately forming a systematic failure cause classification system.

5. The failure risk management method of an unmanned mining truck system according to claim 4 is characterized in that: The hierarchical factors include seven factors: organizational influence, unsafe supervision, preconditions for unsafe behavior, unsafe behavior of personnel, technical system factors, environmental factors and external factors; organizational influence includes three subcategories: improper resource management, poor organizational process and poor safety atmosphere; unsafe supervision includes two subcategories: wrong supervision and illegal supervision; preconditions for unsafe behavior include three subcategories: unqualified personal conditions, unqualified personnel practices and environmental factors; unsafe behavior includes two subcategories: wrong behavior and illegal behavior; Technical system factors include two subcategories: system function failure and system hardware failure; environmental factors include two subcategories: natural environment and operating environment; external factors include two subcategories: external interference and external attack; The relationship between the causal factors is divided into levels, and the results are as follows: Improper resource management includes four factors: insufficient safety training, insufficient safety funding, insufficient emergency resources, and insufficient safety equipment investment; The organizational process includes three factors: imperfect safety management system, unqualified safety protection and unreasonable employee performance appraisal; Organizational atmosphere includes two factors: inadequate implementation of safety management and poor safety atmosphere; faulty supervision includes three factors: inadequate government supervision, inappropriate supervision plan and failure to correct unsafe behaviors in a timely manner; illegal supervision includes two factors: daily violations and individual violations; unqualified personal conditions include three factors: poor psychological or physiological state, unqualified professional skills and weak safety awareness; unqualified personnel practices include three factors: insufficient training, insufficient emergency response capabilities and insufficient personal preparation; Environmental factors include four factors: physical environment, technical environment, organizational environment and human environment; erroneous behaviors include three factors: decision-making errors, skill errors and perception errors; illegal behaviors include three factors: habitual violations, accidental violations and risky operations; the natural environment includes five factors: extreme weather, extreme temperature, dust particles, sunlight and rain; the operating environment includes two factors: rugged roads and regional obstacles; external interference includes two factors: electromagnetic interference and wireless signal interference; external attacks mainly include one factor: data tampering or interception.

6. The failure risk management method of an unmanned mining truck system according to claim 1 is characterized in that: The systematic failure cause classification system is used as a node of a fuzzy Bayesian network, and the process of constructing a visual Bayesian network based on a generative interactive environment includes: Each causal factor of the systematic failure cause classification system is taken as a node of the fuzzy Bayesian network, and a topological structure is constructed. The states of the nodes in the topological structure are represented by occurrence and non-occurrence. Based on expert knowledge, the probability of event occurrence at each node is obtained, and the probability of event occurrence is described by seven levels: very low, low, relatively low, medium, relatively high, high, and very high; The triangular fuzzy number is selected as the membership function for analyzing the accident events. According to the corresponding relationship between the level and the probability of the event, the natural language of the experts is converted into quantitative probability values, and the fuzzy Bayesian network model is constructed. The nodes of the fuzzy Bayesian network model are created in the generative interactive environment software; then directed arcs are drawn to connect the nodes; then the prior probability of the root node and the conditional probability of the child nodes are set, and then the probability of each node is updated to obtain the visual Bayesian network.

7. The failure risk management method of an unmanned mining truck system according to claim 1 is characterized in that: According to the risk assessment results, the system failure countermeasures are analyzed. The process of obtaining the unmanned mining truck system failure risk control measures based on the bowtie theory includes: Combined with the risk assessment results, a bow-tie theoretical model of unmanned mining truck system failure is established: the failure of the unmanned mining truck system is taken as the top event and placed in the center of the bow tie; then the causes of the accident are listed on the left side of the bow tie, and the consequences of the accident are listed on the right side; the cause of the accident on the left side of the bow tie is taken as the event hazard source, the left side of the bow tie is connected with the top event, obstacles are placed on the connection path between the hazard source and the event, and preventive measures are set for the hazard source to be triggered and induce the accident; on the right side of the bow tie, starting from the accident, lines are drawn between the consequences, and control measures are set on the path where the impact of the consequences is further expanded; Through the failure tie theoretical model of unmanned mining truck system, for each failure cause, pre-emptive prevention measures and post-event mitigation measures are proposed from four dimensions: staff, operation management, system equipment and environmental conditions.

8. A failure risk management system for an unmanned mining truck system, characterized in that: include: List building unit: used to build a list of risk factors for failure of unmanned mining truck systems; The first model building unit is used to build a systematic failure cause classification system for identifying failure causes of unmanned truck control systems in combination with the failure cause list of the unmanned mining truck system; A second model building unit is used for taking the systematic failure cause classification system as a node of a fuzzy Bayesian network and building a visual Bayesian network based on a generative interactive environment; Evaluation unit: used to evaluate failure risk through a visual Bayesian network and obtain risk evaluation results; Risk control unit: used to analyze system failure countermeasures according to the risk assessment results, and obtain unmanned mining truck system failure risk control measures based on the bowtie theory.

9. An electronic device, characterized in that: include: one or more processors; a storage device having one or more programs stored thereon; When the one or more programs are executed by the one or more processors, the one or more processors implement the failure risk management method of the unmanned mining truck system as described in any one of claims 1 to 7.

10. A storage medium, characterized in that: A computer program is stored thereon, wherein when the computer program is executed by a processor, the failure risk management method of the unmanned mining truck system as described in any one of claims 1 to 7 is implemented.

Images