A fire early warning monitoring method and system for intelligent workshops based on artificial intelligence

Abstract

The present invention provides an intelligent workshop fire early warning monitoring method and system based on artificial intelligence, which relates to the field of data processing technology. The method includes: acquiring sensor data in real time; determining the current fire risk state according to the sensor data; generating an early warning alarm and generating an immediate inspection task when the current fire risk state exceeds the risk threshold; predicting the posterior failure probability distribution of each workshop equipment based on a Bayesian network when the current fire risk state does not exceed the risk threshold; calculating the expected life cycle cost according to the posterior failure probability distribution of each workshop equipment; generating a dynamic inspection plan with the goal of minimizing the expected life cycle cost, so that the staff can perform fire inspections according to the dynamic inspection plan. The present invention can dynamically adjust the inspection plan according to the actual state and usage of the equipment, reduce the cost of fire early warning, avoid unnecessary inspections and maintenance, and improve the inspection efficiency.

Description

Technical Field

[0001] The present invention relates to the field of data processing technology, and in particular to an intelligent workshop fire early warning monitoring method and system based on artificial intelligence. Background Art

[0002] Workshop fire safety is not only a basic requirement for protecting people’s lives and property, but also an important guarantee for the sustainable and healthy development of the enterprise.

[0003] At present, the fire safety of workshops is mainly guaranteed by regular manual inspections. Special personnel are arranged to regularly inspect the fire-fighting equipment in the workshop, such as fire extinguishers, fire hydrants, sprinkler systems, fire alarms, etc., to ensure that they are in good condition and can be used normally in emergency situations. Focus on inspecting the areas where flammable and explosive items are stored in the workshop to check whether the storage of items complies with safety regulations and whether there are dangerous factors such as open flames and fire sources around. Regularly check the electrical lines and equipment in the workshop to ensure that there are no aging, damage, overload operation, etc., to prevent the occurrence of electrical fires.

[0004] However, regular inspections are time-consuming and labor-intensive, which may lead to unnecessary inspections and maintenance, wasting human and material resources. Regular inspections are performed at fixed time intervals and cannot be dynamically adjusted according to the actual status and usage of the equipment, which may lead to over-maintenance or under-maintenance. Summary of the invention

[0005] In order to solve the technical problems that the existing regular inspection method is time-consuming and labor-intensive, may lead to unnecessary inspections and maintenance, waste human and material resources, is carried out at fixed time intervals, cannot be dynamically adjusted according to the actual status and usage of the equipment, and may lead to excessive or insufficient maintenance, the present invention provides an intelligent workshop fire early warning monitoring method and system based on artificial intelligence.

[0006] The technical solution provided by the embodiment of the present invention is as follows:

[0007] First aspect:

[0008] An embodiment of the present invention provides an artificial intelligence-based intelligent workshop fire early warning monitoring method, comprising:

[0009] S1: Get sensor data in real time;

[0010] S2: Determine the current fire risk status according to the sensor data;

[0011] S3: When the current fire risk status exceeds the risk threshold, a warning alarm is generated and an immediate inspection task is generated, so that the staff can perform a fire inspection according to the immediate inspection task;

[0012] S4: When the current fire risk status does not exceed the risk threshold, the posterior failure probability distribution of each workshop equipment is predicted based on the Bayesian network;

[0013] S5: Calculate the expected life cycle cost based on the posterior failure probability distribution of each workshop equipment;

[0014] S6: With the goal of minimizing the expected life cycle cost, a dynamic inspection plan is generated so that staff can perform fire inspections according to the dynamic inspection plan.

[0015] In the artificial intelligence-based intelligent workshop fire early warning monitoring method, preferably, the fire risk state includes: fire risk state, dust risk state and machine risk state; S2 specifically includes:

[0016] S201: Determine the current fire risk status by mutual verification between the fire detection visual sensor data and the temperature sensor data;

[0017] S202: Determine the current dust risk status by mutual verification between the dust concentration sensor data and the UWB breathing sensor data;

[0018] S203: Determine the current risk status of the machine through mutual verification between the motor load sensor data and the vibration sensor data.

[0019] In the artificial intelligence-based intelligent workshop fire early warning monitoring method, preferably, S201 specifically includes:

[0020] S2011: Determine visual anomalies through fire detection visual sensor data;

[0021] S2012: determining a temperature abnormality value through temperature sensor data;

[0022] S2013: Determine a fire risk value according to the visual abnormality value and the temperature abnormality value:

[0023] r f =β v a v +β t a t

[0024] Among them, r f represents the fire risk value, a v represents a visual outlier, a t represents the temperature anomaly value, β v Represents the weight coefficient of visual outliers, β t Represents the weight coefficient of temperature anomalies.

[0025] In the artificial intelligence-based intelligent workshop fire early warning monitoring method, preferably, S202 specifically includes:

[0026] S2021: determining an abnormal dust concentration value through dust concentration sensor data;

[0027] S2022: Determine abnormal breathing values ​​of personnel through UWB breathing sensor data;

[0028] S2023: Determine a dust risk value according to the abnormal dust concentration value and the abnormal breathing value of the personnel:

[0029] r d =β c a c +β b a b

[0030] Among them, r d represents the dust risk value, a c Indicates the abnormal value of dust concentration, a b Indicates the abnormal value of personnel breathing, β c Represents the weight coefficient of dust concentration abnormality, β b Indicates the weight coefficient of abnormal breathing value of personnel.

[0031] In the artificial intelligence-based intelligent workshop fire early warning monitoring method, preferably, S203 specifically includes:

[0032] S2031: Determine the abnormal value of the motor load through the motor load sensor data;

[0033] S2032: Determine a vibration abnormality value through vibration sensor data;

[0034] S2033: Determine a motor risk value according to the motor load abnormality value and the vibration abnormality value:

[0035] r e =β r a r +β z a z

[0036] Among them, r e represents the motor risk value, a r Indicates the abnormal value of motor load, a z represents the vibration abnormality value, β r Represents the weight coefficient of the abnormal value of the motor load, β z Indicates the weight coefficient of the vibration abnormal value.

[0037] In the intelligent workshop fire early warning monitoring method based on artificial intelligence, preferably, the prediction of the posterior failure probability distribution of each workshop equipment based on the Bayesian network in S4 specifically includes:

[0038] S401: define node status, the nodes include: equipment status node, inspection result node and maintenance operation node;

[0039] S402: Constructing causal relationships between nodes: the current moment device status node depends on the previous moment inspection result node and the previous moment maintenance operation node, the current moment inspection result node depends on the current moment device status node, and the current moment maintenance operation node depends on the current moment inspection result node;

[0040] S403: Using the inspection results of the performed regular inspection tasks and immediate inspection tasks as evidence of the Bayesian network, defining the conditional probabilities between the equipment status, the inspection results and the maintenance operations, and constructing a conditional probability table;

[0041] S404: Using Bayesian reasoning, update the posterior failure probability distribution of each device state.

[0042] In the artificial intelligence-based intelligent workshop fire early warning monitoring method, preferably, S404 is specifically:

[0043] According to the following formula, Bayesian reasoning is used to update the posterior failure probability distribution of each device status:

[0044]

[0045] Among them, S t represents the device status at time t, I t-1 represents the inspection result at time t-1, M t-1 represents the maintenance operation at time t-1, P(S t |I t-1 ,M t-1 ) indicates the inspection result I at a given time t-1 t-1 and maintenance operation M at time t-1 t-1 The posterior failure probability distribution of the current device state, P(S t ,I t-1 ,M t-1 ) represents the equipment status at time t, and the inspection result at time t-1 t-1 and maintenance operation M at time t-1 t-1 The joint probability distribution of t-1 ,M t-1 ) represents the inspection result I at time t-1 t-1 and maintenance operation M at time t-1 t-1The joint probability distribution, P(S t ) represents the device state S at time t t The prior probability, P(I t-1 ∣S t ) represents the conditional probability of the inspection result at time t - 1 given the device state at time t, P(M t-1 ∣S t ,I t-1 ) represents the conditional probability of the maintenance operation at time t - 1 given the device state S t and the inspection result I at time t - 1 t-1 , P(I t-1 ) represents the marginal probability of the inspection result I at time t - 1 t-1 , P(M t-1 ∣I t-1 ) represents the conditional probability of the maintenance operation M at time t - 1 given the inspection result I at time t - 1 t-1 . t-1 t-1 In the above - mentioned intelligent workshop fire warning and monitoring method based on artificial intelligence, preferably, the S5 is specifically as follows:

[0046] Calculate the expected life - cycle cost according to the following formula:

[0047]

[0048]

[0049] where ELC represents the expected life - cycle cost, π represents the dynamic inspection plan, the dynamic inspection plan includes inspection time and maintenance strategy, c 0 represents the initial cost, t L represents the remaining life of the device, R I (t) represents the inspection cost at time t, R M (t) represents the maintenance cost at time t, R F (t) represents the failure cost at time t;

[0050] The failure cost is specifically:

[0051] R F (t) = P(t, π)c F

[0052] where P(t, π) represents the posterior failure probability at time t when adopting the inspection plan π, c F represents the average failure cost when failure occurs.

[0053] In the above - mentioned intelligent workshop fire warning and monitoring method based on artificial intelligence, preferably, the S6 is specifically as follows: ​​

[0054] With the goal of minimizing the expected life cycle cost, a dynamic inspection plan is generated by combining the whale optimization algorithm with the simulated annealing algorithm, so that staff can conduct fire inspections according to the dynamic inspection plan.

[0055] Second aspect:

[0056] An embodiment of the present invention provides an artificial intelligence-based intelligent workshop fire early warning monitoring system, comprising:

[0057] processor;

[0058] A memory having computer-readable instructions stored thereon, wherein when the computer-readable instructions are executed by the processor, the intelligent workshop fire early warning monitoring method based on artificial intelligence as described in the first aspect is implemented.

[0059] The third aspect:

[0060] An embodiment of the present invention provides a computer-readable storage medium, on which a computer program is stored. When the program is executed by a processor, the method for early warning and monitoring of fire prevention in an intelligent workshop based on artificial intelligence as described in the first aspect is implemented.

[0061] The beneficial effects brought about by the technical solution provided by the embodiment of the present invention include at least:

[0062] In the present invention, based on the Bayesian network, the posterior failure probability distribution of each workshop equipment is predicted, and then the expected life cycle cost is calculated. Then, with the goal of minimizing the expected life cycle cost, a dynamic inspection plan is generated so that the staff can perform fire inspections according to the dynamic inspection plan. The inspection plan can be dynamically adjusted according to the actual status and usage of the equipment, reducing the cost of fire warning, avoiding unnecessary inspections and maintenance, and improving inspection efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0063] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. 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.

[0064] Figure 1 A flow chart of an intelligent workshop fire early warning monitoring method based on artificial intelligence provided by an embodiment of the present invention;

[0065] Figure 2 A schematic structural diagram of an artificial intelligence-based intelligent workshop fire early warning monitoring system provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0066] The technical solution of the present invention is described below in conjunction with the accompanying drawings.

[0067] In the embodiments of the present invention, words such as "exemplarily" and "for example" are used to indicate examples, illustrations or explanations. Any embodiment or design described as "example" in the present invention should not be interpreted as being more preferred or more advantageous than other embodiments or designs. Specifically, the use of the word "example" is intended to present the concept in a specific way. In addition, in the embodiments of the present invention, the meaning expressed by "and / or" can be both, or it can be either of the two.

[0068] In the embodiments of the present invention, "image" and "picture" can sometimes be used interchangeably. It should be noted that when the difference between them is not emphasized, the meanings they intend to express are the same. "of", "corresponding, relevant" and "corresponding" can sometimes be used interchangeably. It should be noted that when the difference between them is not emphasized, the meanings they intend to express are the same.

[0069] In the embodiments of the present invention, sometimes the subscripts such as W 1 It may be mistakenly written as a non-subscript form such as W1. When the difference is not emphasized, the meanings they express are the same.

[0070] In order to make the technical problems, technical solutions and advantages to be solved by the present invention more clear, a detailed description will be given below with reference to the accompanying drawings and specific embodiments.

[0071] Reference Manual Attached Figure 1 , which shows a flow chart of an artificial intelligence-based intelligent workshop fire early warning monitoring method provided in an embodiment of the present invention.

[0072] An embodiment of the present invention provides an artificial intelligence-based intelligent workshop fire early warning monitoring method, which can be implemented by an artificial intelligence-based intelligent workshop fire early warning monitoring system, and the artificial intelligence-based intelligent workshop fire early warning monitoring system can be a terminal or a server.

[0073] The home page interface of the intelligent workshop fire warning monitoring system includes related inspection work, plan name, inspection cycle, time period, start and end dates, no inspection dates, inspectors, work name, inspection category, start date, end date, inspection requirements, inspection content, attachments, inspection points, serial number, system number, inspection point type, specific location, detailed address, responsible department, responsible person and other information.

[0074] The processing flow of the intelligent workshop fire early warning monitoring method based on artificial intelligence may include the following steps:

[0075] S1: Get sensor data in real time.

[0076] Among them, the sensor data includes fire detection visual sensor data, temperature sensor data, dust concentration sensor data, UWB breathing sensor data, motor load sensor data, vibration sensor data, humidity sensor data, etc.

[0077] Furthermore, users can add sensor devices as checkpoints in the system according to actual needs and obtain sensor data through wireless transmission technology.

[0078] Specifically, you can click the [Add] button in the system interface to display the "Add" pop-up window, and the user can enter the following in the pop-up window: serial number, equipment number, equipment name, workshop, equipment status, and monitoring status; then click the [Add] button to complete the addition, and click the [Cancel] button to close the current window. Of course, you can also execute the [Delete] command, click the [Delete] button in the system interface to display the delete edit pop-up window, and delete the table content information.

[0079] If there is wrong or duplicate data in the table, the simpler and quicker way is to select the data and then delete it.

[0080] Checkpoints can be set up with details for each address. Optionally, the checkpoint information includes:

[0081] System number, inspection point type, specific location, detailed address, responsible department, and responsible person.

[0082] S2: Determine the current fire risk status based on sensor data.

[0083] In a possible implementation, the fire risk status includes: fire risk status, dust risk status and machine risk status. S2 specifically includes sub-steps S201 to S203:

[0084] S201: Determine the current fire risk status through mutual verification between the fire detection visual sensor data and the temperature sensor data.

[0085] Optionally, S201 specifically includes:

[0086] S2011: Determine visual anomalies through fire detection visual sensor data.

[0087] Specifically, the workshop images taken by the fire detection visual sensor can be used to identify the various flame areas in the workshop images using target detection algorithms such as YOLO and UNet, and the visual anomaly values ​​can be calculated based on the area and centroid dispersion of each flame area:

[0088] av =λ 1 S+λ 2 L

[0089] Among them, a v represents the visual outlier value, S represents the flame area, L represents the centroid dispersion of the flame area, and λ 1 Represents the weight coefficient of the flame area, λ 2 The weight coefficient representing the dispersion of the center of mass of the flame area.

[0090] Among them, those skilled in the art can set the weight coefficient λ of the flame area according to actual conditions. 1 And the weight coefficient λ of the flame area centroid dispersion 2 The present invention does not limit the size.

[0091] Furthermore, the centroid dispersion is specifically:

[0092] L=σ x (t)σ y (t)

[0093] Among them, L represents the centroid dispersion, σ x (t) represents the standard deviation of the horizontal coordinates of all centroids at time t, σ y (t) represents the standard deviation of all centroid ordinates at time t.

[0094] It should be noted that by comprehensively considering the area of ​​the flame region and the centroid dispersion, the characteristic information of the fire can be more comprehensively reflected, the fire risk can be monitored more accurately, dynamically and comprehensively, the accuracy and robustness of the fire detection system can be improved, and intelligent and automated fire warning and response can be supported.

[0095] S2012: Determine temperature abnormality values ​​through temperature sensor data.

[0096] Specifically, the difference between the temperature value measured by the temperature sensor and the maximum temperature warning value may be used as the temperature abnormality value.

[0097] S2013: Determine the fire risk value based on the visual anomaly value and the temperature anomaly value:

[0098] r f =β v a v +β t a t

[0099] Among them, r f represents the fire risk value, a v represents a visual outlier, a t represents the temperature anomaly value, β vRepresents the weight coefficient of visual outliers, β t Represents the weight coefficient of temperature anomalies.

[0100] Among them, those skilled in the art can set the weight coefficient β of the visual outlier value according to the actual situation. v and the weight coefficient β of the temperature anomaly t The present invention does not limit the size.

[0101] In the present invention, visual sensors and temperature sensors have different dependence on the environment. By verifying the fire detection visual sensor data and temperature sensor data to determine the fire risk status, the accuracy, robustness and real-time performance of the fire detection system can be significantly improved, providing more comprehensive and reliable fire warning and management capabilities. This multi-sensor data fusion method not only reduces false alarms and missed alarms, but also maintains efficient fire risk assessment in complex environments, improving the overall fire safety level.

[0102] S202: Determine the current dust risk status by mutual verification between the dust concentration sensor data and the UWB breathing sensor data.

[0103] Optionally, S202 specifically includes:

[0104] S2021: Determine an abnormal dust concentration value through dust concentration sensor data.

[0105] Specifically, the difference between the dust concentration measured by the dust concentration sensor and the maximum dust concentration warning value may be used as the dust concentration abnormality value.

[0106] S2022: Determine abnormal breathing values ​​of the person through UWB breathing sensor data.

[0107] Among them, the UWB (ultra-wideband) breathing sensor uses UWB technology and short pulse signals, has the characteristics of high resolution and strong penetration, and can accurately detect tiny movements in complex environments, such as human breathing. The UWB breathing sensor can efficiently and accurately monitor the breathing rate of people, providing reliable data support for various health and safety applications.

[0108] Specifically, the difference between the person's breathing frequency measured by the UWB breathing sensor data and the maximum person's breathing frequency warning value can be used as the person's breathing abnormality value.

[0109] S2023: Determine the dust risk value based on the abnormal dust concentration value and the abnormal breathing value of personnel:

[0110] r d =β c a c +β b ab

[0111] Among them, r d represents the dust risk value, a c Indicates the abnormal value of dust concentration, a b Indicates the abnormal value of personnel breathing, β c Represents the weight coefficient of dust concentration abnormality, β b Indicates the weight coefficient of abnormal breathing value of personnel.

[0112] Among them, those skilled in the art can set the weight coefficient β of the dust concentration abnormal value according to the actual situation. c and the weight coefficient β of abnormal breathing value of personnel b The present invention does not limit the size.

[0113] In the present invention, the current dust risk status is determined by mutual verification between the dust concentration sensor data and the UWB breathing sensor data, which can significantly improve the accuracy, real-time and robustness of dust risk detection, provide a comprehensive and quantitative dust risk assessment, support intelligent and automated early warning and management, improve the health and safety level of employees, and optimize the overall safety management of the workshop.

[0114] S203: Determine the current risk status of the machine through mutual verification between the motor load sensor data and the vibration sensor data.

[0115] Optionally, S203 specifically includes:

[0116] S2031: Determine the abnormal value of the motor load through the motor load sensor data.

[0117] Specifically, the difference between the motor load measured by the motor load sensor and the maximum motor load warning value may be used as the motor load abnormality value.

[0118] S2032: Determine the vibration abnormality value through the vibration sensor data.

[0119] Specifically, the difference between the motor vibration measured by the vibration sensor data and the maximum motor vibration warning value may be used as the vibration abnormality value.

[0120] S2033: Determine the motor risk value based on the motor load abnormality value and the vibration abnormality value:

[0121] r e =β r a r +β z a z

[0122] Among them, r e represents the motor risk value, ar Indicates the abnormal value of motor load, a z represents the vibration abnormality value, β r Represents the weight coefficient of the abnormal value of the motor load, β z Indicates the weight coefficient of the vibration abnormal value.

[0123] Among them, those skilled in the art can set the weight coefficient β of the abnormal value of the motor load according to the actual situation. z and the weight coefficient β of the vibration abnormality r The present invention does not limit the size.

[0124] In the present invention, the current machine risk status is determined by mutual verification between the motor load sensor data and the vibration sensor data, which can significantly improve the accuracy, real-time and robustness of machine status monitoring, provide comprehensive and quantitative machine risk assessment, support intelligent and automated early warning and management, improve equipment maintenance efficiency, and optimize the overall safety management of the workshop.

[0125] S3: When the current fire risk status exceeds the risk threshold, a warning alarm is generated and an immediate inspection task is generated so that the staff can perform fire inspections according to the immediate inspection task.

[0126] It should be noted that the inspection plan consists of two parts, one is the immediate inspection task, and the other is the dynamic inspection task that will be introduced later. The inspection plan table parameters include: inspection items, inspection date, plan theme, implementation status, person in charge, etc. The inspection plan section displays the generated dynamic inspection tasks, and can also be manually modified to edit the table's add, delete, modify, and query operations; users only need to enter the correct value in the table interface to perform the operation.

[0127] Among them, those skilled in the art can set the size of the risk threshold according to actual conditions, and the present invention does not limit it.

[0128] In the present invention, when the current fire risk status exceeds the risk threshold, a warning alarm is generated and an immediate inspection task is generated, which can significantly improve the timeliness and effectiveness of fire risk response, enhance safety assurance, improve inspection efficiency, reduce losses, optimize resource allocation, and improve employees' sense of security and satisfaction, which can significantly improve the overall fire safety management level of the workshop.

[0129] S4: When the current fire risk status does not exceed the risk threshold, the posterior failure probability distribution of each workshop equipment is predicted based on the Bayesian network.

[0130] Among them, the Bayesian Network, also known as the belief network or directed acyclic graph (DAG), is a probabilistic model for representing and reasoning about uncertain knowledge. The Bayesian Network represents the conditional dependency between variables through a graphical structure and combines the Bayesian theorem for probabilistic reasoning.

[0131] In a possible implementation, the prediction of the posterior failure probability distribution of each workshop equipment based on the Bayesian network in S4 specifically includes sub-steps S401 to S404:

[0132] S401: Define node status.

[0133] The nodes include: equipment status node, inspection result node and maintenance operation node. The equipment status node indicates the running status or health status of the equipment at a certain moment. The inspection result node indicates the equipment status evaluation result obtained after a certain inspection. The maintenance operation node indicates the maintenance action taken after a certain inspection or failure.

[0134] It should be noted that the relationship between these nodes can be dynamically updated to reflect the health status and maintenance needs of the equipment in real time. The equipment status node reflects the current health status of the equipment, the inspection result node provides status feedback after the equipment inspection, and the maintenance operation node records the response actions to the equipment status. The combination of the three constitutes the core framework of equipment management and fault prediction.

[0135] S402: Construct causal relationships between nodes: the current device status node depends on the previous inspection result node and the previous maintenance operation node, the current inspection result node depends on the current device status node, and the current maintenance operation node depends on the current inspection result node.

[0136] S403: Using the inspection results of the performed regular inspection tasks and immediate inspection tasks as evidence of the Bayesian network, defining the conditional probabilities between the equipment status, the inspection results and the maintenance operations, and constructing a conditional probability table.

[0137] S404: Using Bayesian reasoning, update the posterior failure probability distribution of each device state.

[0138] Optionally, S404 specifically includes:

[0139] According to the following formula, Bayesian reasoning is used to update the posterior failure probability distribution of each device status:

[0140]

[0141] Among them, S t represents the device status at time t, I t-1represents the inspection result at time t-1, M t-1 represents the maintenance operation at time t-1, P(S t |I t-1 ,M t-1 ) indicates the inspection result I at a given time t-1 t-1 and maintenance operation M at time t-1 t-1 The posterior failure probability distribution of the current device state, P(S t ,I t-1 ,M t-1 ) represents the equipment status at time t, and the inspection result at time t-1 t-1 and maintenance operation M at time t-1 t-1 The joint probability distribution of t-1 ,M t-1 ) represents the inspection result I at time t-1 t-1 and maintenance operation M at time t-1 t-1 The joint probability distribution of t ) represents the device state S at time t t The prior probability, P(I t-1 ∣S t ) represents the conditional probability of the inspection result at time t-1 given the device status at time t, P(M t-1 ∣S t ,I t-1 ) represents the device state S at a given time t t The inspection result I at time t-1 t-1 Maintenance operation M at time t-1 later t-1 The conditional probability, P(I t-1 ) represents the inspection result I at time t-1 t-1 The marginal probability, P(M t-1 ∣I t-1 ) represents the inspection result I at a given time t-1 t-1 Maintenance operation M at time t-1 later t-1 The conditional probability of .

[0142] In the present invention, by predicting the posterior failure probability distribution of each workshop equipment based on the Bayesian network, the accuracy of equipment failure prediction can be significantly improved, quantitative risk assessment can be provided, maintenance and inspection efficiency can be improved, early warning can be provided, flexible adaptation to changing environments can be achieved, and the level of system intelligence can be improved. This method comprehensively considers multiple factors, dynamically updates probabilities, and systematically models models, which can effectively improve the management and maintenance effects of workshop equipment, optimize resource allocation, and reduce failure rates and maintenance costs.

[0143] S5: Calculate the expected life cycle cost based on the posterior failure probability distribution of each workshop equipment.

[0144] In a possible implementation, S5 is specifically:

[0145] Calculate the expected life cycle cost according to the following formula:

[0146]

[0147] Where ELC represents the expected life cycle cost, π represents the dynamic inspection plan, which includes the inspection time and maintenance strategy, and c 0 represents the initial cost, t L Remaining life of the device, R I (t) represents the inspection cost at time t, R M (t) represents the maintenance cost at time t, R F (t) represents the failure cost at time t.

[0148] The specific failure cost is:

[0149] R F (t)=P(t,π)c F

[0150] Where P(t,π) represents the posterior failure probability at time t when the inspection plan π is adopted, c F Represents the average failure cost when failure occurs.

[0151] In the present invention, by adopting the expected life cycle cost (ELC) method and using the Bayesian network to dynamically predict and optimize the inspection plan, the economic benefits, reliability and safety of equipment management can be significantly improved. This method can not only comprehensively consider various costs within the life cycle of the equipment, but also dynamically adjust the plan according to real-time data to ensure the optimal allocation of resources and the efficient operation of the equipment, and ultimately achieve a dual improvement in management efficiency and economic benefits.

[0152] S6: With the goal of minimizing the expected life cycle cost, a dynamic inspection plan is generated so that staff can conduct fire inspections according to the dynamic inspection plan.

[0153] In a possible implementation, S6 specifically includes: taking minimization of expected life cycle cost as the goal, generating a dynamic inspection plan by combining the whale optimization algorithm with the simulated annealing algorithm, so that staff can perform fire inspections according to the dynamic inspection plan.

[0154] Among them, the Whale Optimization Algorithm (WOA) is a bionic optimization algorithm based on the whale's predation behavior. It simulates the bubble net predation behavior of humpback whales during the predation process to search for the global optimal solution.

[0155] Among them, the simulated annealing algorithm (SA) is a random optimization algorithm based on the physical annealing process, which searches for the local optimal solution by simulating the slow cooling process during metal annealing.

[0156] Specifically, the inverse of the expected life cycle cost is used as the fitness function of the whale optimization algorithm and the simulated annealing algorithm.

[0157] Initialize the whale population, which contains multiple whale individuals. The position of each whale individual in the search space represents a feasible inspection plan.

[0158] Generate a random number r between 0 and 1 1 , calculate the first coefficient A:

[0159] A=2ar 1 -a

[0160] Where a represents the adaptive convergence factor.

[0161] Optionally, the adaptive convergence factor a is specifically:

[0162]

[0163] Among them, t represents the current iteration number, and T represents the maximum iteration number.

[0164] It should be noted that in the initial stage of the iteration, the value of factor a is relatively large, which is conducive to the whale optimization algorithm to conduct global search, explore more areas of the search space, and avoid falling into the local optimal solution too early. In the later stage of the iteration, the value of factor a gradually decreases, which enhances the local search ability of the whale optimization algorithm, fine-tunes the quality of the solution, and improves the accuracy of the solution. Through the design of the adaptive factor a, the algorithm can smoothly transition from global search to local optimization at different iteration stages, and improve the convergence of the algorithm.

[0165] When |A|<1, go to the next step. When |A|<1, enter the random search mechanism.

[0166] Generates a random variable p.

[0167] When p < 0.5, it enters the contraction and encirclement mechanism. When p ≥ 0.5, it enters the spiral mechanism.

[0168] Determine the current global optimal position, and other whales move towards the global optimal position, and update their own positions through the shrinking and surrounding mechanism:

[0169] X(t+1)=X best (t)-A·D

[0170] D=|C·X best(t)-X(t)|

[0171] C=2r 1

[0172] Among them, X(t+1) represents the individual position of the whale at the t+1th iteration, X best (t) represents the global optimal position at the tth iteration, A represents the first coefficient, D represents the randomized distance between the individual position of the whale and the current global optimal position, C represents the second coefficient, t represents the current iteration number, and X(t) represents the individual position of the whale at the tth iteration.

[0173] It should be noted that by determining the current global optimal position and using the shrinking and encircling mechanism to move other whale individuals to the global optimal position, the convergence speed can be effectively accelerated, the search efficiency can be improved, the global search and local optimization can be balanced, the robustness of the algorithm can be enhanced, and the algorithm can be simple and easy to implement. These characteristics enable the whale optimization algorithm to efficiently find the global optimal solution when dealing with complex optimization problems, and show good performance and adaptability in different application scenarios.

[0174] Generate a random number r between 0 and 1 2 , the whale updates its position through the spiral mechanism by moving in an upward spiral and continuously shrinking the encirclement:

[0175]

[0176] D * =|X best (t)-X(t)|

[0177] Among them, D * represents the distance between the individual position of the whale and the current global optimal position, e represents a natural constant, b represents a logarithmic spiral constant, and l represents a random number between -1 and 1.

[0178] It should be noted that according to the random number r 2 The value of , selects different spiral paths (cosine or sine) to increase path diversity. By moving in an upward spiral and continuously shrinking the encirclement, the whale optimization algorithm can effectively balance global search and local optimization, improve the diversity and flexibility of the search path, enhance the robustness and adaptability of the algorithm, quickly approach the global optimal solution, and ultimately improve the optimization effect and efficiency. This method combines the spiral mechanism and the shrinking encirclement mechanism, and further enhances the comprehensiveness and effectiveness of the search by randomly selecting the movement mode.

[0179] Based on the positions of individual whales relative to each other, they update their own positions through a random search mechanism:

[0180] X(t+1)=X rand-A.D rand

[0181] D rand =|C·X rand -X(t)|

[0182] Among them, X rand Represents the position of a randomly selected individual whale.

[0183] It should be noted that through the random search mechanism, updating the position of each whale according to the position of each other can significantly enhance the global search capability of the algorithm, improve robustness and adaptability, balance the exploration and development process, and ensure that the algorithm can effectively cover the search space and gradually converge to the global optimal solution.

[0184] Update the global optimal position.

[0185] Perform a Cauchy mutation perturbation on the global optimal position:

[0186]

[0187] in, represents the global optimal position after the Cauchy mutation perturbation, and cauchy(0,1) represents the Cauchy operator.

[0188] It should be noted that performing Cauchy mutation perturbation on the global optimal position can significantly enhance the global search capability of the optimization algorithm, avoid falling into the local optimum, accelerate convergence, and improve the robustness and adaptability of the algorithm.

[0189] Determine whether the current number of iterations has reached the maximum number of iterations. If so, output the inspection plan represented by the whale with the highest current fitness as the suboptimal solution. Otherwise, return to continue iterating.

[0190] Initialize the initial temperature T 0 , maximum number of iterations, termination temperature T m .

[0191] The whale individual with the highest fitness determined by the whale optimization algorithm is used as the current individual, and the neighborhood individuals are generated according to the suboptimal inspection plan.

[0192] Compare the fitness values ​​of the current individual with those of the neighboring individuals. When the fitness value of the neighboring individual is greater than the fitness value of the current individual, use the neighboring individual to replace the current individual. When the fitness value of the neighboring individual is less than the fitness value of the current individual, use the neighboring individual to replace the current individual with a replacement probability P.

[0193] The replacement probability P is specifically:

[0194]

[0195] Among them, P represents the replacement probability, exp represents the exponential function with e as the base, and X new represents the neighborhood individuals, F(X new ) represents the fitness value of the neighborhood individual, X t represents the current individual at the tth iteration, F(X t ) represents the fitness value of the current individual at the tth iteration, and T represents the current temperature.

[0196] It should be noted that by accepting poor solutions with a certain probability, the simulated annealing algorithm can jump out of the local optimum, thereby increasing the chance of finding the global optimal solution. At the same time, high temperature in the early stage promotes global search and the probability of accepting poor solutions is higher; low temperature in the later stage promotes local search and the probability of accepting poor solutions is reduced.

[0197] Determine whether the current number of iterations has reached the maximum number of iterations, or whether the current temperature has reached the termination temperature T m If yes, output the optimal model parameters. Otherwise, update the temperature and continue to generate neighborhood individuals for iteration.

[0198] In a possible implementation, the temperature update method is:

[0199] T t+1 =α·T t

[0200] Where α represents the temperature drop coefficient, T t+1 represents the temperature at the t+1th iteration, T t represents the temperature at the tth iteration, T 0 Indicates the initial temperature.

[0201] It should be noted that by gradually lowering the temperature, the search process gradually converges from the initial broad search to the later fine search. At the initial high temperature, the algorithm explores more search space; at the later low temperature, the algorithm focuses on local optimization to improve the accuracy of the solution.

[0202] The beneficial effects brought about by the technical solution provided by the embodiment of the present invention include at least:

[0203] In the present invention, based on the Bayesian network, the posterior failure probability distribution of each workshop equipment is predicted, and then the expected life cycle cost is calculated. Then, with the goal of minimizing the expected life cycle cost, a dynamic inspection plan is generated so that the staff can perform fire inspections according to the dynamic inspection plan. The inspection plan can be dynamically adjusted according to the actual status and usage of the equipment, reducing the cost of fire warning, avoiding unnecessary inspections and maintenance, and improving inspection efficiency.

[0204] Reference Manual Attached Figure 2, showing a structural schematic diagram of an artificial intelligence-based intelligent workshop fire early warning monitoring system provided by the present invention.

[0205] The present invention also provides an artificial intelligence-based intelligent workshop fire early warning monitoring system 20, which is applied to the above-mentioned artificial intelligence-based intelligent workshop fire early warning monitoring method, including:

[0206] Processor 201;

[0207] The memory 202 stores computer-readable instructions, and when the computer-readable instructions are executed by the processor 201, the intelligent workshop fire early warning monitoring method based on artificial intelligence as described in the method embodiment is implemented.

[0208] The artificial intelligence-based intelligent workshop fire early warning monitoring system 20 provided by the present invention can execute the above-mentioned artificial intelligence-based intelligent workshop fire early warning monitoring method and achieve the same or similar technical effects. To avoid repetition, the present invention will not go into details.

[0209] The beneficial effects brought about by the technical solution provided by the embodiment of the present invention include at least:

[0210] In the present invention, based on the Bayesian network, the posterior failure probability distribution of each workshop equipment is predicted, and then the expected life cycle cost is calculated. Then, with the goal of minimizing the expected life cycle cost, a dynamic inspection plan is generated so that the staff can perform fire inspections according to the dynamic inspection plan. The inspection plan can be dynamically adjusted according to the actual status and usage of the equipment, reducing the cost of fire warning, avoiding unnecessary inspections and maintenance, and improving inspection efficiency.

[0211] It should be understood that the processor in the embodiment of the present invention may be a central processing unit (CPU), and the processor may also be other general-purpose processors, digital signal processors (DSP), application specific integrated circuits (ASIC), field programmable gate arrays (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor may be a microprocessor or the processor may also be any conventional processor, etc.

[0212] It should also be understood that the memory in the embodiments of the present invention may be a volatile memory or a non-volatile memory, or may include both volatile and non-volatile memories. Among them, the non-volatile memory may be a read-only memory (ROM), a programmable read-only memory (PROM), an erasable programmable read-only memory (EPROM), an electrically erasable programmable read-only memory (EEPROM), or a flash memory. The volatile memory may be a random access memory (RAM), which is used as an external cache. By way of example and not limitation, many forms of random access memory (RAM) are available, such as static RAM (SRAM), dynamic random access memory (DRAM), synchronous DRAM (SDRAM), double data rate synchronous dynamic random access memory (DDR SDRAM), enhanced synchronous dynamic random access memory (ESDRAM), synchronous link DRAM (SLDRAM), and direct rambus RAM (DR RAM).

[0213] The above embodiments can be implemented in whole or in part by software, hardware (such as circuits), firmware or any other combination. When implemented by software, the above embodiments can be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions or computer programs. When the computer instructions or computer programs are loaded or executed on a computer, the process or function described in the embodiment of the present invention is generated in whole or in part. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable devices. The computer instructions can be stored in a computer-readable storage medium, or transmitted from one computer-readable storage medium to another computer-readable storage medium. For example, the computer instructions can be transmitted from one website, computer, server or data center to another website, computer, server or data center by wired (such as infrared, wireless, microwave, etc.). The computer-readable storage medium can be any available medium that can be accessed by a computer or a data storage device such as a server or data center that contains one or more available media sets. The available medium can be a magnetic medium (for example, a floppy disk, a hard disk, a tape), an optical medium (for example, a DVD), or a semiconductor medium. The semiconductor medium can be a solid-state hard disk.

[0214] It should be understood that the term "and / or" in this article is only a description of the association relationship of associated objects, indicating that there can be three relationships. For example, A and / or B can represent: A exists alone, A and B exist at the same time, and B exists alone. A and B can be singular or plural. In addition, the character " / " in this article generally indicates that the associated objects before and after are in an "or" relationship, but it may also indicate an "and / or" relationship. Please refer to the context for specific understanding.

[0215] In the present invention, "at least one" means one or more, and "more than one" means two or more. "At least one of the following" or similar expressions refers to any combination of these items, including any combination of single or plural items. For example, at least one of a, b, or c can be represented by: a, b, c, ab, ac, bc, or abc, where a, b, c can be single or multiple.

[0216] It should be understood that in various embodiments of the present invention, the size of the serial numbers of the above-mentioned processes does not mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present invention.

[0217] Those skilled in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of the present invention.

[0218] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working processes of the above-described equipment, devices and units can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0219] In the several embodiments provided by the present invention, it should be understood that the disclosed devices, apparatuses and methods can be implemented in other ways. For example, the device embodiments described above are only schematic. For example, the division of the units is only a logical function division. There may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another device, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or units, which can be electrical, mechanical or other forms.

[0220] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0221] In addition, each functional unit in each embodiment of the present invention may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit.

[0222] If the functions are implemented in the form of software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention can be essentially or partly embodied in the form of a software product that contributes to the prior art. The computer software product is stored in a storage medium and includes several instructions for enabling a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the methods described in various embodiments of the present invention. The aforementioned storage medium includes: various media that can store program codes, such as a USB flash drive, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk.

[0223] An embodiment of the present invention provides a computer-readable storage medium having a computer program stored thereon, characterized in that when the program is executed by a processor, the program implements the artificial intelligence-based intelligent workshop fire warning monitoring method as described in the method embodiment.

[0224] A computer-readable storage medium provided by the present invention can implement the steps and effects of the intelligent workshop fire warning monitoring method based on artificial intelligence in the above method embodiment. To avoid repetition, the present invention will not go into details.

[0225] The beneficial effects brought about by the technical solution provided by the embodiment of the present invention include at least:

[0226] In the present invention, based on the Bayesian network, the posterior failure probability distribution of each workshop equipment is predicted, and then the expected life cycle cost is calculated. Then, with the goal of minimizing the expected life cycle cost, a dynamic inspection plan is generated so that the staff can perform fire inspections according to the dynamic inspection plan. The inspection plan can be dynamically adjusted according to the actual status and usage of the equipment, reducing the cost of fire warning, avoiding unnecessary inspections and maintenance, and improving inspection efficiency.

[0227] The above is only a specific embodiment of the present invention, but the protection scope of the present invention is not limited thereto. Any person skilled in the art can easily think of changes or substitutions within the technical scope disclosed by the present invention, which should be included in the protection scope of the present invention. Therefore, the protection scope of the present invention should be based on the protection scope of the claims.

[0228] There are a few points to note:

[0229] (1) The drawings of the embodiments of the present invention only relate to the structures related to the embodiments of the present invention, and other structures may refer to the general design.

[0230] (2) For the sake of clarity, in the drawings used to describe the embodiments of the present invention, the thickness of the layers or regions is exaggerated or reduced, that is, these drawings are not drawn according to the actual scale. It is understood that when an element such as a layer, film, region or substrate is referred to as being "on" or "under" another element, the element may be "directly" "on" or "under" the other element or there may be intermediate elements.

[0231] (3) In the absence of conflict, the embodiments of the present invention and the features therein may be combined with each other to obtain new embodiments.

[0232] The above are only specific embodiments of the present invention, but the protection scope of the present invention is not limited thereto. The protection scope of the present invention shall be based on the protection scope of the claims.

Claims

1. An intelligent workshop fire early warning monitoring method based on artificial intelligence, characterized in that: include: S1: Get sensor data in real time; S2: Determine the current fire risk status according to the sensor data; S3: When the current fire risk status exceeds the risk threshold, a warning alarm is generated and an immediate inspection task is generated, so that the staff can perform a fire inspection according to the immediate inspection task; S4: When the current fire risk status does not exceed the risk threshold, the posterior failure probability distribution of each workshop equipment is predicted based on the Bayesian network; S5: Calculate the expected life cycle cost based on the posterior failure probability distribution of each workshop equipment; S6: With the goal of minimizing the expected life cycle cost, a dynamic inspection plan is generated, so that staff can perform fire inspections according to the dynamic inspection plan; Specifically, S6 is as follows: with the goal of minimizing the expected life cycle cost, a dynamic inspection plan is generated by combining the whale optimization algorithm with the simulated annealing algorithm, so that the staff can conduct fire inspections according to the dynamic inspection plan; Initialize the whale population, which contains multiple whale individuals. The position of each whale individual in the search space represents a feasible inspection plan. Generate a random number r1 between 0 and 1 and calculate the first coefficient A: A=2ar1-a Where a represents the adaptive convergence factor; The adaptive convergence factor a is specifically: Among them, t represents the current number of iterations, and T represents the maximum number of iterations; When |A|<1, proceed to the next step; when |A|≥1, enter the random search mechanism; Generate a random variable p; When p < 0.5, it enters the contraction and encirclement mechanism; when p ≥ 0.5, it enters the spiral mechanism; Determine the current global optimal position, and other whales move towards the global optimal position and update their own positions through the shrinking and surrounding mechanism; Generate a random number r2 between 0 and 1. Based on the random number r2, the whale moves in an upward spiral and continuously shrinks the encirclement, updating its position through the spiral mechanism. Based on the positions of individual whales relative to each other, they update their own positions through a random search mechanism; Update the global optimal position; Perform Cauchy mutation perturbation on the global optimal position; Determine whether the current number of iterations has reached the maximum number of iterations; if so, output the inspection plan represented by the whale individual with the highest current fitness as the suboptimal solution; otherwise, return to continue iterating; Initialize the initial temperature, maximum number of iterations, and termination temperature; The whale individual with the highest fitness determined by the whale optimization algorithm is used as the current individual, and the neighborhood individuals are generated according to the suboptimal inspection plan; Compare the fitness values ​​of the current individual with those of the neighboring individuals; when the fitness value of the neighboring individual is greater than the fitness value of the current individual, use the neighboring individual to replace the current individual; when the fitness value of the neighboring individual is less than the fitness value of the current individual, use the neighboring individual to replace the current individual with the replacement probability; Determine whether the current number of iterations has reached the maximum number of iterations, or whether the current temperature has reached the termination temperature; if so, output the optimal model parameters; otherwise, update the temperature and continue to generate neighborhood individuals for iteration.

2. The method for early warning and monitoring firefighting in intelligent workshops based on artificial intelligence according to claim 1 is characterized in that: The fire risk status includes: fire risk status, dust risk status and machine risk status; S2 specifically includes: S201: Determine the current fire risk status by mutual verification between the fire detection visual sensor data and the temperature sensor data; S202: Determine the current dust risk status by mutual verification between the dust concentration sensor data and the UWB breathing sensor data; S203: Determine the current risk status of the machine through mutual verification between the motor load sensor data and the vibration sensor data.

3. The method for early warning and monitoring firefighting in intelligent workshops based on artificial intelligence according to claim 2 is characterized in that: The S201 specifically includes: S2011: Determine visual anomalies through fire detection visual sensor data; S2012: determining a temperature abnormality value through temperature sensor data; S2013: Determine a fire risk value according to the visual abnormality value and the temperature abnormality value: r f =b v a v +b t a t Among them, r f represents the fire risk value, a v represents a visual outlier, a t represents the temperature anomaly value, β v Represents the weight coefficient of visual outliers, β t Represents the weight coefficient of temperature anomalies.

4. The method for early warning and monitoring firefighting in intelligent workshops based on artificial intelligence according to claim 2 is characterized in that: The S202 specifically includes: S2021: determining an abnormal dust concentration value through dust concentration sensor data; S2022: Determine abnormal breathing values ​​of personnel through UWB breathing sensor data; S2023: Determine a dust risk value according to the abnormal dust concentration value and the abnormal breathing value of the personnel: r d =b c a c +b b a b Among them, r d represents the dust risk value, a c Indicates the abnormal value of dust concentration, a b Indicates the abnormal value of personnel breathing, β c Represents the weight coefficient of dust concentration abnormality, β b Indicates the weight coefficient of abnormal breathing value of personnel.

5. The method for early warning and monitoring firefighting in intelligent workshops based on artificial intelligence according to claim 2 is characterized in that: The S203 specifically includes: S2031: Determine the abnormal value of the motor load through the motor load sensor data; S2032: Determine a vibration abnormality value through vibration sensor data; S2033: Determine a motor risk value according to the motor load abnormality value and the vibration abnormality value: r e =b r a r +b z a z Among them, r e represents the motor risk value, a r Indicates the abnormal value of motor load, a z represents the vibration abnormality value, β r Represents the weight coefficient of the abnormal value of the motor load, β z Indicates the weight coefficient of the vibration abnormal value.

6. The method for early warning and monitoring of firefighting in intelligent workshops based on artificial intelligence according to claim 1 is characterized in that: The prediction of the posterior failure probability distribution of each workshop equipment based on the Bayesian network in S4 specifically includes: S401: define node status, the nodes include: equipment status node, inspection result node and maintenance operation node; S402: Constructing causal relationships between nodes: the current moment device status node depends on the previous moment inspection result node and the previous moment maintenance operation node, the current moment inspection result node depends on the current moment device status node, and the current moment maintenance operation node depends on the current moment inspection result node; S403: Using the inspection results of the performed regular inspection tasks and immediate inspection tasks as evidence of the Bayesian network, defining the conditional probabilities between the equipment status, the inspection results and the maintenance operations, and constructing a conditional probability table; S404: Using Bayesian reasoning, update the posterior failure probability of each device state.

7. The method for early warning and monitoring firefighting in intelligent workshops based on artificial intelligence according to claim 6 is characterized in that: The S404 is specifically as follows: According to the following formula, Bayesian reasoning is used to update the posterior failure probability distribution of each device status: Among them, S t represents the device status at time t, I t-1 represents the inspection result at time t-1, M t-1 represents the maintenance operation at time t-1, P(S t |I t-1 ,M t-1 ) indicates the inspection result I at a given time t-1 t-1 and maintenance operation M at time t-1 t-1 The posterior failure probability distribution of the current device state, P(S t ,I t-1 ,M t-1 ) represents the equipment status at time t, and the inspection result at time t-1 t-1 and maintenance operation M at time t-1 t-1 The joint probability distribution of t-1 ,M t-1 ) represents the inspection result I at time t-1 t-1 and maintenance operation M at time t-1 t-1 The joint probability distribution of t ) represents the device state S at time t t The prior probability, P(I t-1 ∣S t ) represents the conditional probability of the inspection result at time t-1 given the device status at time t, P(M t-1 ∣S t ,I t-1 ) represents the device state S at a given time t t The inspection result I at time t-1 t-1 Maintenance operation M at time t-1 later t-1 The conditional probability, P(I t-1 ) represents the inspection result I at time t-1 t-1 The marginal probability, P(M t-1 ∣I t-1 ) represents the inspection result I at a given time t-1 t-1 Maintenance operation M at time t-1 later t-1 The conditional probability of .

8. The method for early warning and monitoring firefighting in intelligent workshops based on artificial intelligence according to claim 1 is characterized in that: The S5 is specifically: Calculate the expected life cycle cost according to the following formula: Where ELC represents the expected life cycle cost, π represents the dynamic inspection plan, which includes the inspection time and maintenance strategy, c0 represents the initial cost, t L Remaining life of the device, R I (t) represents the inspection cost at time t, R M (t) represents the maintenance cost at time t, R F (t) represents the failure cost at time t; The failure cost is specifically: R F (t)=P(t,π)c F Where P(t,π) represents the posterior failure probability at time t when the inspection plan π is adopted, c F Represents the average failure cost when failure occurs.

9. An intelligent workshop fire early warning monitoring system based on artificial intelligence, characterized in that: include: processor; A memory having computer-readable instructions stored thereon, wherein when the computer-readable instructions are executed by the processor, the intelligent workshop fire early warning monitoring method based on artificial intelligence as described in any one of claims 1 to 8 is implemented.

Images