Industrial equipment full life cycle management system based on AI

Abstract

The invention relates to the technical field of industrial equipment management, in particular to an AI-based industrial equipment full-life-cycle management system. The defect that the traditional S-N calculation neglects on-site environment degradation is overcome by combining the dust concentration, the environment temperature and the influence of a vibration transfer matrix between equipment; a Gaussian distribution model is introduced to capture a nonlinear damage effect of load fluctuation, a load sequence correction factor to reflect historical accumulation of a repeated load and an impact load, and an environment and vibration correction factor, and fatigue damage of the equipment is accurately calculated; the degraded stress is coupled with a Paris crack propagation model, so that the problems of life depletion and crack propagation disjunction of traditional fatigue prediction are solved; a plurality of feasible path sets are calculated in combination with the current position of the maintainer and the high-risk equipment, buffer driving time is determined according to weather influence, and balance of latest departure and saving waiting is realized by constructing time window constraints and calculating the latest departure time of the maintainer, so that safety management of the equipment life cycle is realized.

Description

Technical Field

[0001] The present invention relates to the technical field of industrial equipment management, and in particular to an AI-based industrial equipment full life cycle management system. Background Art

[0002] In the lifecycle management of mining equipment, most methods rely on fixed SN curves or simple empirical rules to assess equipment's remaining life. These methods ignore the profound impact of the complex working conditions typical of underground mines—high dust concentrations, low or high temperature environments, and mechanical vibration coupling between multiple pieces of equipment—on fatigue behavior. Consequently, current technologies still have the following shortcomings:

[0003] The fatigue life parameters calculated from the SN curve under standard conditions cannot reflect the increased friction, heat accumulation, and resulting material performance degradation caused by dust entering the equipment. They also cannot reflect the additional dynamic load increase imposed on surrounding equipment when the equipment resonates or amplifies vibrations.

[0004] Secondly, fatigue damage accumulation often uses the linear Miner method, which accumulates the proportion of cycles under each working condition. This method does not consider the nonlinear effects of repeated loads and the sudden increase and loss of impact loads, and cannot handle the differences in fatigue sensitivity caused by dust environments and temperature changes.

[0005] Regarding vibration coupling between devices, there is a lack of a high-precision solution that combines frequency response function data with FEM simulation results to correct fatigue parameters.

[0006] In addition, traditional equipment lifecycle management scheduling is usually based on static point-to-point distance or shortest path algorithms, and does not fully consider the characteristics of limited maintenance personnel resources. As a result, it is very easy for improper path selection or inaccurate departure timing to lead to delays in the maintenance of high-risk equipment, which in turn causes production stagnation and safety hazards, affecting the safe management of the equipment lifecycle. Summary of the Invention

[0007] In response to the above-mentioned shortcomings of the existing technology, the present invention provides an AI-based industrial equipment full life cycle management system, which can effectively solve the problem that the existing technology does not combine the resonance transmission effect between the mining environment and the equipment, resulting in the inability to accurately judge the remaining fatigue life of the equipment, affecting the safe maintenance and disposal of the equipment cycle.

[0008] To achieve the above objectives, the present invention is implemented through the following technical solutions:

[0009] The present invention provides an AI-based industrial equipment full life cycle management system, which at least includes:

[0010] Life analysis submodule, including:

[0011] Construct stress amplitude based on the interaction between the dust environment in which the equipment is located and the equipment;

[0012] Building fatigue strength by combining friction and load in dusty environments;

[0013] Determine the fatigue index of equipment materials under different working conditions based on the interaction between environmental factors and equipment;

[0014] Establish SN curve to estimate the fatigue life of equipment in initial working condition;

[0015] Damage analysis submodule, including:

[0016] The Gaussian damage model is introduced to capture the effect of changes in load amplitude on fatigue damage;

[0017] Introducing load sequence correction factors to consider the historical impact of loads;

[0018] Construct environmental correction factors based on the damage effects of dust, temperature, and changes in the equipment's working environment on the equipment;

[0019] Establish vibration transfer correction factors based on the transmission of vibration between devices;

[0020] Establish a comprehensive fatigue damage calculation model to calculate the fatigue damage of equipment;

[0021] The crack analysis submodule analyzes the impact of crack growth on equipment and calculates the crack growth rate to predict the remaining fatigue life of the equipment.

[0022] The equipment management module establishes an equipment set based on the remaining fatigue life of each equipment, and generates the latest departure time for equipment maintenance based on the location distribution of maintenance personnel and equipment, as well as weather influences.

[0023] This solution achieves closed-loop management from high-precision fatigue prediction to on-site maintenance scheduling through four interconnected modules:

[0024] Under the influence of dust concentration, temperature and vibration transmission between equipment, stress amplitude, fatigue strength and index are adjusted in real time to make the life calculation fit the extreme environment of the mine;

[0025] Secondly, the nonlinear Gaussian combined with load sequence cumulative damage analysis accurately captures load fluctuations, repetition and impact effects, and combines environmental and vibration correction factors to significantly improve the calculation accuracy of fatigue damage;

[0026] Third, post-degradation stress and coupled Paris crack growth closely correlate the on-site accumulated damage with the crack growth rate, and the true remaining life is calculated by integrating the remaining cycles, ensuring a scientific and reliable warning time window;

[0027] Finally, multi-path, weather buffering, and last-departure single-person scheduling enable maintenance personnel, limited by a single person, to achieve both a "latest departure" and on-time arrival at each red-zone device, leveraging confidence-buffered travel times and mathematical time window constraints. Dynamic rescheduling capabilities also allow the system to flexibly respond to sudden weather changes and road congestion. This synergistic combination of technologies not only significantly improves the accuracy of remaining life predictions for mining equipment but also enables safe, economical, and efficient single-person closed-loop maintenance during on-site repairs, ensuring safe management of the equipment lifecycle and providing a solid foundation for the stability and safety of mine production.

[0028] Furthermore, the final stress amplitude is determined as follows:

[0029] The final stress amplitude is established based on the stress amplitude of the equipment in a standard environment, the influence of the dust environment, the influence of temperature, and the vibration transmission between equipment.

[0030] Furthermore, the vibration transmission coefficient is determined as follows:

[0031] The vibration transmission between devices is described by the frequency response function, which reflects how the input vibration signal is transmitted through the device and affects the output vibration of the device;

[0032] Model device geometry and physical properties, including:

[0033] Define the device geometry, material properties, contact points, and support structures;

[0034] Apply external excitation forces, including:

[0035] Simulate external loads or vibration sources on the equipment and calculate the vibration response of the equipment;

[0036] By simulating the vibration coupling between devices, the vibration transfer coefficient is determined based on the frequency response function and the peak frequency of the device.

[0037] Furthermore, the fatigue index is determined as follows:

[0038] It is determined based on the equipment’s fatigue index, environmental impact coefficient and dust concentration under standard conditions.

[0039] Furthermore, the transmission of vibration between devices will lead to increased load or vibration coupling effect, aggravating fatigue damage. A vibration transmission correction factor is established based on the influence coefficient of vibration on damage and the vibration transmission coefficient between devices.

[0040] Furthermore, the crack analysis submodule predicts the remaining fatigue life by:

[0041] Calculate the dynamic stress intensity factor ΔK of the crack:

[0042]

[0043] Where Y represents the geometric correction factor, σ represents the actual stress applied to the material, and u represents the half-length of the crack;

[0044] Establish nonlinear correction factor f based on fatigue damage stress (D) Introducing a coupled model of fatigue damage and crack growth to dynamically calculate the crack growth rate N represents the loading cycle count during crack growth, and C and g represent material constants;

[0045] According to the crack growth rate, calculate the number of cycles N required for crack growth fail ;

[0046] Determine the remaining fatigue life T rema =N fail Δt, Δt represents the time interval of each cycle.

[0047] Compared with the prior art, the technical solution provided by the present invention has the following beneficial effects:

[0048] Combined with the influence of dust concentration, ambient temperature and vibration transfer matrix between equipment, the stress amplitude, fatigue strength and fatigue index of the equipment are corrected in real time, solving the problem of traditional SN calculation ignoring the deterioration of the on-site environment;

[0049] The Gaussian distribution model is introduced to capture the nonlinear damage effect of load fluctuations, the load sequence correction factor is used to reflect the historical accumulation of repeated loads and impact loads, and the environmental and vibration correction factors are used to accurately calculate the fatigue damage of the equipment, overcoming the limitations of the linear accumulation of the Miner method.

[0050] By coupling post-degradation stress with the Paris crack growth model, the crack growth rate is calculated and the remaining fatigue life is derived, solving the problem of the disconnect between traditional fatigue prediction life exhaustion and crack growth.

[0051] Combining the maintenance personnel's current location with high-risk equipment, multiple feasible path sets are calculated. The driving time distribution on sunny and inclement days is linearly interpolated according to the weather weight to determine the buffer driving time. By constructing a time window constraint, the latest departure time of the maintenance personnel is calculated to achieve a balance between the latest departure and saving waiting time, improve the efficiency of human resource utilization, and realize the safe management of the equipment life cycle. BRIEF DESCRIPTION OF THE DRAWINGS

[0052] To more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are only some embodiments of the present invention. Those skilled in the art can also derive other drawings based on these drawings without inventive effort.

[0053] Figure 1 It is the overall module block diagram of the present invention. DETAILED DESCRIPTION

[0054] To make the purpose, technical solutions, and advantages of the embodiments of the present invention more clear, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making any creative efforts shall fall within the scope of protection of the present invention.

[0055] Current mining equipment lifecycle management solutions have achieved some success in online equipment monitoring, basic lifecycle prediction, and automated maintenance work orders. However, they still suffer from the following core deficiencies:

[0056] Insufficient environmental coupling

[0057] Traditional fatigue life calculations are mostly based on SN curves under laboratory conditions, ignoring the combined effects of high dust concentrations in mines on increased surface friction, heat accumulation, and material corrosion of mechanical components. They also fail to fully consider the degradation of metal fatigue strength and hardness caused by extreme temperatures. As a result, in actual mine operations, prediction errors are often too large, and life estimates cannot reflect the performance degradation induced by the on-site environment.

[0058] Lack of vibration interaction between devices

[0059] When a group of equipment is operating, vibration will be transmitted between adjacent equipment through foundation and structural coupling, forming secondary or tertiary loads. This "vibration amplification" or "vibration resonance" effect is often ignored in traditional models, or is only roughly treated as an empirical coefficient, resulting in a lack of precision in the judgment of local stress and fatigue tendency.

[0060] Fatigue damage accumulation model is too linear

[0061] Although the existing Miner linear accumulation method is simple and practical, its linear assumption ignores the "softening" or "hardening" effects that occur in subsequent working conditions after large loads or impact loads. It cannot correctly quantify the nonlinear influence of the order of the load sequence on the formation and propagation of microcracks. Therefore, in a high-cyclic stress fluctuation environment, the cumulative damage estimation deviates seriously from reality.

[0062] The scheduling scheme is simple or not robust

[0063] For maintenance scheduling at mine sites, most systems are based only on the shortest path or simple priority sorting, and are unable to incorporate uncertain factors such as weather changes, sudden road conditions, and time window constraints for single-person multi-point services. As a result, problems such as "low punctuality and poor delay risk control" often occur in actual implementation. It is also impossible to ensure that maintenance personnel can reach critical equipment safely and efficiently in extremely harsh environments, thus affecting the safety management of the equipment life cycle.

[0064] In summary, the present invention is proposed to solve the above-mentioned problems.

[0065] The present invention will be further described below with reference to the embodiments.

[0066] Example 1 (see Figure 1 ): An AI-based industrial equipment full life cycle management system, including at least:

[0067] The life analysis submodule analyzes the environment in which the equipment is located in the mine, including:

[0068] 1) If the equipment is located in a dusty environment in an underground mine, this environment will not only cause external corrosion of the equipment, but also increase friction, wear, and heat accumulation during long-term operation. Dust may enter the equipment, increase friction of mechanical components, increase equipment load, and thus affect the fatigue behavior of the equipment.

[0069] 2) If there is mutual influence between devices, that is, the physical contact and joint operation of multiple devices may cause resonance, vibration amplification and other problems, thereby affecting fatigue strength and fatigue life. For example, when a device vibrates due to excessive load, it may affect other devices, causing their stress levels to increase.

[0070] Therefore, considering the impact of the dust environment in underground mines on the fatigue behavior of equipment and the mutual influence between equipment, stress amplitude, fatigue strength and fatigue index are defined to more accurately predict the fatigue life of equipment, including:

[0071] Determination of stress amplitude S:

[0072]

[0073] Among them, S0 represents the stress amplitude of the equipment under standard environment, k dust Indicates the influence coefficient of dust concentration on stress amplitude, C dust Indicates dust concentration, which directly affects the friction of equipment components, k tempIt represents the influence coefficient of temperature on stress amplitude, describing the influence of heat accumulation caused by temperature on the stress of the equipment, T env Indicates temperature, increases the brittleness of the material, and affects the fatigue strength of the equipment. trabs (i, j) represents the vibration transmission coefficient between devices (the influence of the vibration of device j on device i), reflecting the interaction between devices, n represents the number of devices involved in the interaction, α j The influence coefficient of the vibration of equipment j on equipment i is expressed, and the stress amplitude is adjusted according to the dust concentration, temperature and the interaction between the equipment;

[0074] Among them, for the vibration transmission coefficient M between equipment trabs (i,j):

[0075] The vibration transmission between devices is described in advance by the frequency response function. The frequency response function reflects how the input vibration signal is transmitted through the device and affects the output vibration of the device.

[0076] Frequency response function V i (ω) represents the vibration response of device i at frequency ω, F j (ω) represents the external excitation force on device j;

[0077] Modeling device geometry and physical properties: Define the device geometry, material properties, contact points, and support structures;

[0078] Apply external excitation force: simulate external loads or vibration sources on the device and calculate the vibration response of the device;

[0079] Calculate the vibration transmission coefficient by simulating the vibration coupling between devices ω peak Indicates the peak frequency of the device, that is, the resonant frequency or the main excitation frequency.

[0080] Determination of fatigue strength:

[0081] Increased friction and load in dusty environments will reduce the fatigue strength of the equipment. f :

[0082] S f =S f0 ·(1-β·C dust ·T env )

[0083] Among them, S f0 It represents the fatigue strength of the equipment under standard environment, β represents the environmental impact coefficient, which describes the influence of dust and temperature on fatigue strength, C dust Indicates dust concentration, affects equipment friction and reduces material fatigue strength;

[0084] Determination of fatigue index:

[0085] The interaction between environmental factors (such as dust concentration, temperature, etc.) and equipment determines the fatigue sensitivity of equipment materials under different working conditions, that is, the fatigue index m:

[0086]

[0087] Among them, m0 represents the fatigue index of the equipment under standard environment, γ represents the environmental impact coefficient, which describes the influence of dust concentration and vibration between equipment on the fatigue index, and C dust Indicates dust concentration, which can reflect the sensitivity of the equipment to dust environment;

[0088] Based on the SN curve analysis, the fatigue life of the equipment in the initial working state is estimated as follows:

[0089] Calculate the fatigue life of the equipment based on the stress amplitude of the equipment and the fatigue strength of the material:

[0090] N f It represents fatigue life. Then, by analyzing the mining environment in which the equipment is located, environmental factors such as dust concentration and temperature are combined with factors such as the equipment's workload and mutual influence to determine the actual application environment and status of the mining equipment, thereby improving the accuracy of fatigue life prediction.

[0091] The damage analysis submodule, after obtaining the fatigue life of the equipment, considers that in actual environments, the loads faced by the equipment are varied, nonlinear, and historically dependent. Load changes and alternating effects (such as impact loads and repeated loads) usually have a more serious impact on fatigue damage.

[0092] In addition, the equipment working environment (such as dust concentration and temperature) and the vibration transmission effect between equipment will further affect the fatigue strength of the material. Especially in complex industrial environments, these factors will significantly accelerate the accumulation of damage. Therefore, the determination of fatigue damage is based on a variety of conditions, as follows:

[0093] 1) During long-term operation, equipment will experience loads of varying magnitudes. Generally, the greater the load, the more severe the fatigue damage. However, the change in load magnitude often follows certain rules. Therefore, considering the volatility of load changes and the nonlinear damage relationship, the Gaussian damage model is introduced to describe the distribution characteristics of the load and capture the impact of changes in load magnitude on fatigue damage. The following are the results:

[0094]

[0095] Among them, n iIndicates the actual number of cycles under working condition i, μ i , σ i Represent the load average and standard deviation under the corresponding working conditions, D i represents the damage contribution coefficient under the i-th working condition, exp() represents the exponential function, which describes the attenuation process of the damage contribution;

[0096] 2) Furthermore, during the operation of the equipment, the load is not simply repeated or single, but has a certain historical effect. For example, repeated loads will accelerate the accumulation of damage, and impact loads may cause sudden and huge damage. The influence of load history effects (for example, the sequence of loads, changes in amplitude, etc.) on fatigue damage is very important. Therefore, considering the load sequence effect (the load change sequence under different working conditions will have different effects on equipment fatigue damage, especially in the case of repeated loads and impact loads) and nonlinear damage accumulation (load history will lead to nonlinear damage accumulation, and load changes (such as from low load to high load) will accelerate fatigue damage), a load sequence correction factor is introduced to consider the historical influence of the load, and then:

[0097]

[0098] Among them, f load (n prev ) represents the load sequence correction factor, λ represents the strength coefficient of the load sequence correction, N i represents the fatigue life under working condition i, α i It represents the nonlinear correction factor related to working condition i, υ represents the nonlinear influence coefficient of impact load, sign(n i ) represents the sign function, which describes the direction of the load. If it is 1, it indicates a positive load, and if it is -1, it indicates a reverse load. This term considers the impact of the positive and negative directions of the load on the damage. Δn i It represents the load change between working condition i and the previous working condition, and l represents the number of historical working conditions, that is, the total number of load working conditions considered.

[0099] 3) In actual operation, environmental factors (such as dust concentration and temperature) will significantly affect the fatigue damage of the equipment. In mining environments and other working conditions, the increased friction caused by dust on the equipment surface and the increased brittleness of the material due to high temperature will accelerate fatigue damage. Secondly, the working environment of the equipment will also change. Therefore, based on the impact of dust, temperature and changes in the equipment working environment on equipment damage, an environmental correction factor f is constructed. env (C dust ,T env ):

[0100] f env (C dust ,T env)=exp(β dust ·C dust +β temp ·T env )

[0101] Among them, β dust , β temp Represent the correction factors of dust concentration and temperature respectively.

[0102] 4) Furthermore, the transmission of vibration between devices may lead to increased load or vibration coupling effect, thereby exacerbating fatigue damage. Moreover, the vibration interaction between devices is usually very complex and may be affected by multiple factors such as working conditions and equipment design. Therefore, a vibration transmission correction factor f is constructed. vib (M trans (i,j)):

[0103] f vib (M trans (i,j))=exp(γ vib ·M trans (i,j)), γ vib Represents the influence coefficient of vibration on damage.

[0104] In summary, 1)-4) in the damage analysis submodule can be combined to establish a comprehensive fatigue damage calculation model to simultaneously consider factors such as load history, environmental impact, and interaction between devices to calculate the fatigue damage D of the equipment:

[0105]

[0106] p represents the number of working conditions that need to be calculated. If the fatigue damage D is greater than 1, it usually means that the equipment has reached the fatigue limit and needs to be repaired or replaced. Therefore, by introducing the Gaussian damage model to capture the nonlinear effects of load fluctuations, the load sequence correction factor considers the role of load history, and the environmental factors and vibration transmission effects solve the additional fatigue damage caused by the interaction between the environment and the equipment in actual working conditions. This can more comprehensively and accurately predict the fatigue damage that the equipment may encounter during long-term operation.

[0107] The crack analysis submodule considers the impact of the equipment crack growth process and predicts the crack growth rate to predict the remaining fatigue life of the equipment. It includes the following steps:

[0108] Calculate the dynamic stress intensity factor ΔK of the crack:

[0109]

[0110] Where Y represents the geometric correction factor, σ represents the actual stress applied to the material, and u represents the half-length of the crack. In this embodiment, σ can be calculated based on the fatigue damage D, the life consumption ratio (the number of cycles currently used N), and the fatigue damage D, the life consumption ratio (the number of cycles currently used N). use Ratio of fatigue life to load history characteristics Combined with, defining the actual stress, can be σ0 represents the baseline stress of the equipment in the initial state, α1, α2, and α3 represent the damage weight, life weight, and load weight, respectively, and ωi represents the relative importance of the influence of the i-th historical working condition (load condition) on the current stress;

[0111] Establish nonlinear correction factor f based on fatigue damage stress (D)=(1+αD·D),α D Represents the adjustment coefficient related to material characteristics and working conditions. Therefore, a coupling model of fatigue damage and crack growth is introduced to combine the crack propagation process with factors such as fatigue damage and load history calculated in the first two steps to dynamically adjust the crack propagation rate. N represents the loading cycle count during crack growth, and C and g represent material constants;

[0112] Calculate the number of cycles required for crack growth based on the crack growth rate u0 represents the initial length of the crack, uf represents the length when the crack expands to the critical point, usually the failure point. By integrating the crack expansion process, the number of cycles required for the crack to grow from the current length to the critical failure length is obtained, and then the remaining fatigue life T is obtained. rema =N fail Δt, Δt represents the time interval of each cycle.

[0113] The equipment management module is used to determine the remaining fatigue life of all equipment in the mine. In this way, the remaining fatigue life of each equipment can be obtained, and the fatigue margin is added to the remaining fatigue life to determine the new remaining fatigue life. This method can ensure that the equipment takes into account errors or environmental mutations or unexpected impacts in subsequent applications, thereby ensuring that maintenance begins before the critical point is actually reached to avoid direct damage to the equipment. Therefore, when multiple new remaining fatigue lives are obtained, multiple life classification intervals are set, usually low risk (large remaining fatigue life), medium risk and high risk (small remaining fatigue life). Considering the loss and conservation of resources, maintenance work orders are usually only generated for high-risk equipment in the mine. At the same time, considering that there are fewer operators available to repair the equipment, such as when there is only one maintenance operator left to repair the equipment, in order to facilitate the maintenance personnel to conduct safe inspections of the equipment one by one to avoid the risk of equipment downtime and failure, the following steps are also included:

[0114] Determine the equipment corresponding to the high-risk-life classification interval, and obtain the equipment set [i1,···,i k ], k represents the total number of devices;

[0115] Determine the path set:

[0116] For each pair of points (a, b) (a represents the current position of the maintenance personnel, b represents the coordinates of the next device to be repaired m ), including (P0,i1), (i m-1 ,i m ), calculate a set of optional paths based on the mining area road network model represents the first candidate path, Represents the second candidate path, each path contains the average travel time of the path under sunny weather and the standard deviation of the travel time for this route in clear weather The average travel time for this route during inclement weather and the standard deviation of travel time for this route in severe weather conditions (ω represents weight, ω is 0 for sunny weather, 1 for bad weather);

[0117] Determine buffered travel time:

[0118] Calculate the average travel time for each route:

[0119] Average driving time under current weather conditions after interpolation (linear interpolation)

[0120] Interpolated travel time standard deviation

[0121] Calculating buffered travel times α represents the confidence level of the travel time, and zα represents the standard normal quantile corresponding to the confidence level;

[0122] For each segment a→b, select the path p that arrives the earliest in the path set, that is, take the minimum

[0123] Therefore, calculate the time T1 when the maintenance personnel finish the maintenance of the first equipment i1 (with the minimum remaining fatigue life) and set off to the next point:

[0124] T0 represents the time when the maintenance personnel departs from the current position P0. Indicates the maintenance time of i1, Indicates the minimum buffer travel time from P0 to i1;

[0125] Subsequent equipment i j ≥2, T j Indicates that the maintenance personnel have completed the equipment i j The moment of overhaul and departure to the next point, Indicates that the device is from i j-1 To device i j Buffered travel time

[0126] Time window constraints:

[0127] In order to ensure that the maintenance of each equipment is completed before the end of the remaining fatigue life, determine the equipment i j Remaining fatigue life cut-off time Then it must meet the following requirements:

[0128] Calculate the latest departure time of maintenance personnel:

[0129] Determine multiple upper bounds for T0:

[0130]

[0131] Let i0=P0, take the minimum value of the upper bound as the latest departure time That is, maintenance personnel should Starting from the current position P0, if If it is less than the current time, it means that it is impossible to complete all equipment maintenance tasks on time without adding manpower, so as to facilitate subsequent planning and management.

[0132] Therefore, through multi-path, weather buffer and latest departure planning, maintenance personnel can not only complete high-risk equipment maintenance on time with the most optimized route and controlled time margin, but also maximize resource utilization efficiency and significantly reduce the risks of equipment downtime and failure due to delays. At the same time, the route and departure time are optimized according to weather conditions, thereby achieving safe and stable equipment maintenance management in harsh mining environments.

[0133] A computer-readable storage medium having a computer program stored thereon, characterized in that the computer program implements the above-mentioned system when executed by a processor. Specific reference may be made to the above-mentioned system and will not be repeated here.

[0134] The above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit the same. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements will not cause the essence of the corresponding technical solutions to deviate from the protection scope of the technical solutions of the various embodiments of the present invention.

Claims

1. An AI-based industrial equipment full life cycle management system, characterized by: include: Life analysis submodule, including: Construct stress amplitude based on the interaction between the dust environment in which the equipment is located and the equipment; Building fatigue strength by combining friction and load in dusty environments; Determine the fatigue index of equipment materials under different working conditions based on the interaction between environmental factors and equipment; Establish SN curve to estimate the fatigue life of equipment in initial working condition; Damage analysis submodule, including: The Gaussian damage model is introduced to capture the effect of changes in load amplitude on fatigue damage; Considering the historical impact of load, a load sequence correction factor is introduced; Construct environmental correction factors based on the damage effects of dust, temperature, and changes in the equipment's working environment on the equipment; Establish vibration transfer correction factors based on the transmission of vibration between devices; Establish a comprehensive fatigue damage calculation model to calculate the fatigue damage of equipment; The crack analysis submodule analyzes the impact of crack growth on equipment and calculates the crack growth rate to predict the remaining fatigue life of the equipment. The equipment management module establishes an equipment set based on the remaining fatigue life of each equipment, and generates the latest departure time for equipment maintenance based on the location distribution of maintenance personnel and equipment, as well as weather influences.

2. The AI-based industrial equipment full life cycle management system according to claim 1 is characterized in that: The formula for determining the stress amplitude S is: Among them, S0 represents the stress amplitude of the equipment under standard environment, k dust Indicates the influence coefficient of dust concentration on stress amplitude, C dust Indicates dust concentration, k temp Indicates the influence coefficient of temperature on stress amplitude, T env Indicates temperature, M trabs (i, j) represents the vibration transmission coefficient between devices, n represents the number of devices involved in the interaction, and αj represents the influence coefficient of the vibration of device j on device i.

3. The AI-based industrial equipment full life cycle management system according to claim 2 is characterized in that: The vibration transmission coefficient is determined as follows: The vibration transmission between devices is described by the frequency response function. The frequency response function reflects how the input vibration signal is transmitted through the device and affects the output vibration of the device. Frequency response function V i (ω) represents the vibration response of device i at frequency ω, F j (ω) represents the external excitation force on device j; Model device geometry and physical properties, including: Define the device geometry, material properties, contact points, and support structures; Apply external excitation forces, including: Simulate external loads or vibration sources on the equipment and calculate the vibration response of the equipment; By simulating the vibration coupling between devices, the vibration transfer coefficient is determined based on the frequency response function and the peak frequency of the device.

4. The AI-based industrial equipment full life cycle management system according to claim 2 is characterized in that: The formula for determining the fatigue index m is: Among them, m0 represents the fatigue index of the equipment under standard environment, γ represents the environmental impact coefficient, C dust Indicates dust concentration.

5. The AI-based industrial equipment full life cycle management system according to claim 1 is characterized in that: Considering the load sequence effect and nonlinear damage accumulation, the load sequence correction factor is introduced, and then: Among them, f load (n prev ) represents the load sequence correction factor, λ represents the strength coefficient of the load sequence correction, N i represents the fatigue life under working condition i, αi represents the nonlinear correction factor related to working condition i, υ represents the nonlinear influence coefficient of impact load, sign(n i ) represents the sign function, Δn i It represents the load change between working condition i and the previous working condition, and l represents the number of historical working conditions.

6. The AI-based industrial equipment full life cycle management system according to claim 1 is characterized in that: The transmission of vibration between the devices may lead to increased load or vibration coupling effect, thereby aggravating fatigue damage. A vibration transmission correction factor is established based on the influence coefficient of vibration on damage and the vibration transmission coefficient between the devices.

7. The AI-based industrial equipment full life cycle management system according to claim 1 is characterized in that: The method for predicting the remaining fatigue life of the crack analysis submodule is: Calculate the dynamic stress intensity factor ΔK of the crack: Where Y represents the geometric correction factor, σ represents the actual stress applied to the material, and u represents the half-length of the crack; Establish nonlinear correction factor f based on fatigue damage stress (D) Introducing a coupled model of fatigue damage and crack growth to dynamically calculate the crack growth rate N represents the loading cycle count during crack growth, and C and g represent material constants; Calculate the number of cycles required for crack growth based on the crack growth rate u0 represents the initial length of the crack, and uf represents the length of the crack when it expands to the critical point; Determine the remaining fatigue life T rema =N fail Δt, Δt represents the time interval of each cycle.

8. The AI-based industrial equipment full life cycle management system according to claim 1 is characterized in that: The method for generating the latest departure time for equipment maintenance is: According to the remaining fatigue life from low to high, the equipment set [i1,···,i k ], k represents the total number of devices; For each pair of points (a, b), a represents the current location of the maintenance personnel, and b represents the coordinates of the next device to be repaired. m , including (P0,i1), (i m-1 ,i m ), calculate a set of optional paths ρ based on the mining area road network model ab ; Calculate the average travel time for each route: Average driving time μ under current weather conditions p (ω); Travel time standard deviation σ p (ω); Calculating buffered travel times Calculate the time T1 when the maintenance personnel finishes maintenance on the first equipment i1 and leaves for the next point: T0 represents the time when the maintenance personnel departs from the current position P0, s i1 Indicates the maintenance time of i1, Indicates the minimum buffer travel time from P0 to i1; Subsequent equipment i j ≥2, T j Indicates that the maintenance personnel have completed the equipment i j The time for inspection and departure to the next point, Indicates that the device is from i j-1 To device i j Buffer travel time; Calculate the latest departure time of maintenance personnel That is, maintenance personnel The time starts from the current position P0.

9. The AI-based industrial equipment full life cycle management system according to claim 8, characterized in that: Each path contains: The average travel time for this route in clear weather The standard deviation of the travel time for this route in clear weather The average travel time for this route during inclement weather The standard deviation of the route's travel time during inclement weather 10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the system according to any one of claims 1 to 9 is implemented.

Images