A dynamic management method and system for optimizing the full-life cycle cost of special equipment

Abstract

The present invention discloses a dynamic management method and system for optimizing the whole-life cycle cost of special equipment, which relates to the technical field of equipment management and includes: collecting the operation data and environmental data of special equipment, inputting them into a fault prediction sub-model for preprocessing to predict the probability of fault occurrence and the faulty components of special equipment; constructing a fuzzy comprehensive evaluation model to divide the risk levels of special equipment; combining the risk levels of special equipment and the predicted benefits of a benefit prediction sub-model, formulating a dynamic maintenance strategy for special equipment through maintenance costs; performing maintenance on special equipment according to the dynamic maintenance strategy, and optimizing the data model and the dynamic maintenance strategy according to the actual maintenance costs. By predicting the probability of equipment fault occurrence, the present invention dynamically determines the optimal maintenance time of equipment, reduces the maintenance cost and the cost of downtime loss, improves the maintenance efficiency, extends the service life of the equipment, and controls the whole-life cycle cost of the equipment from the source.

Description

Technical Field

[0001] The present invention relates to the technical field of equipment management, and more specifically to a dynamic management method and system for optimizing the full life cycle cost of special equipment. Background Art

[0002] The full life cycle cost refers to the sum of all costs generated during the life cycle of equipment from initial investment, operation and maintenance to final retirement and scrapping. Special equipment usually has large investment, complex technology and harsh operating environment. The life cycle cost of equipment involves multiple links. For special equipment enterprises, the service life of equipment is long. For the equipment that has been purchased, the upfront requisition cost has been determined. However, for the equipment with an extended operation period, the maintenance cost of the equipment will increase significantly. By optimizing the equipment maintenance cost, the optimization of the equipment life cycle cost can be achieved, and the lean management of the equipment full life cycle cost can be realized.

[0003] At present, most enterprises mainly have two ways of maintaining equipment. One is to conduct regular maintenance on the equipment, and the other is to rely on emergency maintenance after a failure occurs. These two maintenance systems are not perfect, resulting in additional downtime losses, thereby increasing the life cycle cost of the equipment. There is an urgent need for an optimized management method for the full life cycle cost of equipment to reduce the sudden downtime caused by equipment failures during the equipment production process and increase the full life cycle cost of the equipment.

[0004] The existing Chinese patent application with the publication number CN117455710A discloses a method, device and computer equipment for determining the full life cycle cost of equipment, which includes obtaining the resource cost of the full life cycle of equipment under various maintenance strategies according to the equipment management conditions; obtaining the carbon cost of the full life cycle of equipment under various maintenance strategies; and obtaining the full life cycle cost of equipment under various maintenance strategies according to the resource cost and carbon cost of the full life cycle of the equipment. Using this method can accurately calculate the full life cycle cost of equipment.

[0005] The existing Chinese patent with the publication number CN111882162A discloses a method and device for managing the full life cycle cost of equipment, which includes the following steps: synchronizing the equipment hierarchical structure in the production management system to the asset management system; obtaining maintenance work orders from the production management system and importing the maintenance work orders into the asset management system; obtaining the equipment associated with the maintenance work order according to the equipment hierarchical result and attributing the maintenance cost to the equipment associated with the maintenance work order; and realizing the cost management of the full life cycle of each equipment by combining the maintenance cost of each equipment and the purchase cost of each equipment in the asset management system. The invention can manage the cost of the full life cycle of equipment, and there is no need to enter information repeatedly, and the management efficiency is high.

[0006] The above prior art manages the full - life - cycle cost of equipment by calculating the maintenance cost of the equipment. However, it does not arrange a maintenance plan in advance for the possible failures during the operation of the equipment, resulting in additional downtime loss costs and increasing the additional full - life - cycle cost of the equipment caused by unreasonable equipment failure maintenance. Summary of the Invention

[0007] Aiming at the deficiencies of the prior art, the purpose of the present invention is to provide a dynamic management method and system for optimizing the full - life - cycle cost of special equipment. Through the operation data and environmental data of the equipment, the probability of equipment failure is predicted, the risk levels of special equipment are classified, and a dynamic maintenance strategy is formulated according to the probability of failure and risk levels to determine the optimal maintenance time or maintenance method of special equipment, reduce the maintenance cost of special equipment and the production losses caused by equipment downtime, extend the service life of special equipment, and control the full - life - cycle cost of special equipment from the source.

[0008] To achieve the above purpose, the present invention provides the following technical solutions:

[0009] On the one hand, the present invention provides a dynamic management method for optimizing the full - life - cycle cost of special equipment, specifically including the following steps:

[0010] Construct a data model for the full life cycle of special equipment, where the data model includes a failure prediction sub - model and a benefit prediction sub - model;

[0011] Collect the operation data and environmental data of special equipment, and after pre - processing, input them into the failure prediction sub - model to predict the probability of equipment failure and the faulty components;

[0012] Based on the probability of failure and the faulty components, combined with the characteristics of special equipment, construct a fuzzy comprehensive evaluation model to classify the risk levels of special equipment;

[0013] Combined with the risk levels of special equipment and the predicted benefits of the benefit prediction sub - model, formulate a dynamic maintenance strategy for special equipment through maintenance costs;

[0014] Perform maintenance on special equipment according to the dynamic maintenance strategy, record the actual maintenance cost, and optimize the data model and dynamic maintenance strategy according to the actual maintenance cost.

[0015] As a further improvement of the present invention, the specific steps of constructing the fuzzy comprehensive evaluation model to classify the risk levels of special equipment include:

[0016] Obtain the probability of failure and the faulty components, as well as the operation data, select evaluation factors, and determine the evaluation factor set;

[0017] The risk levels of special equipment are divided into three levels, namely normal operation, low risk and high risk, and an evaluation level set is constructed.

[0018] The relative importance of each evaluation factor in the evaluation factor set is compared pairwise to construct a judgment matrix, and the eigenvector and the maximum eigenvalue in the judgment matrix are calculated to obtain the weight of each evaluation factor.

[0019] Determine the membership degree of each evaluation factor to different risk levels, and establish a fuzzy relation matrix.

[0020] Through fuzzy composition operation, a comprehensive evaluation result vector is obtained, and according to the principle of maximum membership degree, the risk level of special equipment is determined.

[0021] As a further improvement of the present invention, the dynamic maintenance strategy includes a cost-priority maintenance strategy and a time-priority maintenance strategy. The specific steps of formulating the dynamic maintenance strategy of special equipment through maintenance costs are as follows:

[0022] Configure a failure probability threshold. When the failure probability of special equipment is less than or equal to the failure probability threshold or the risk level is normal operation, no dynamic maintenance strategy is formulated.

[0023] When the failure probability of special equipment is greater than the failure probability threshold and the risk level is normal operation or low risk, a cost-priority maintenance strategy is formulated.

[0024] When the risk level of special equipment is high risk, a time-priority maintenance strategy is formulated.

[0025] As a further improvement of the present invention, the specific steps of formulating the cost-priority maintenance strategy include:

[0026] S41: When the failure probability is greater than the failure probability threshold, trigger a warning signal to provide equipment maintenance reminder.

[0027] S42: Take the moment of receiving the equipment maintenance reminder as the start time node of the equipment maintenance process, take the latest maintenance time point as the end time node, and divide it at intervals of a preset first fixed duration to form a first maintenance time node sequence.

[0028] S43: Calculate the total maintenance cost expected to be generated when the maintenance work starts at each maintenance time node in the first maintenance time node sequence, and associate the total maintenance cost with the first maintenance time node to form a first total maintenance cost set.

[0029] S44: Configure a total cost threshold, and add the maintenance time nodes corresponding to the total maintenance cost in the first total maintenance cost set that is less than the total cost threshold to the time node set.

[0030] S45: Judge and analyze the time node set, form a second maintenance time node sequence, and calculate and obtain a second total maintenance cost set;

[0031] S46: Sort the total maintenance costs in the first total maintenance cost set and the second total maintenance cost set to form a maintenance cost sequence from low to high;

[0032] S47: Select the maintenance time node corresponding to the lowest total maintenance cost from the maintenance cost sequence as the optimal equipment maintenance time.

[0033] As a further improvement of the present invention, the judging and analyzing the time node set, forming a second maintenance time node sequence, and calculating and obtaining a second total maintenance cost set includes:

[0034] Judge whether there are adjacent maintenance time nodes in the time node set. If there are adjacent maintenance time nodes, merge the adjacent maintenance time nodes;

[0035] Take the earliest maintenance time node among the adjacent maintenance time nodes as the start time node for secondary judgment, take the latest maintenance time node among the adjacent maintenance time nodes as the end time node for secondary judgment, and take the duration between the start time node for secondary judgment and the end time node for secondary judgment as the maintenance time period for secondary judgment;

[0036] Divide the maintenance time period for secondary judgment at intervals of a second fixed duration to form a second maintenance time node sequence, calculate the total maintenance cost expected to be generated when each maintenance time node in the second maintenance time node sequence starts to perform maintenance work, and associate the total maintenance cost with the maintenance time node to form a second total maintenance cost set;

[0037] If there are no adjacent maintenance time nodes, define the second total maintenance cost set as an empty set.

[0038] As a further improvement of the present invention, the calculation of the total maintenance cost expected to be generated when starting to perform maintenance work includes:

[0039] Obtain the shutdown loss cost caused by shutdown during the maintenance process through the cumulative production benefit of special equipment;

[0040] Calculate the equipment maintenance cost generated by special equipment during the expected maintenance process through labor costs, equipment part replacement costs, and non-linear equipment health status impact factors; among them, the labor cost is the product of the unit labor hour cost and the labor hours spent;

[0041] Obtain the probability of failure occurrence at each maintenance time node, and obtain the potential future cost by combining with the average cost of historical maintenance costs. The potential future cost is used to represent the loss cost caused by the expected failures from the initial time node to the maintenance time node due to non-maintenance;

[0042] Perform a weighted sum of the downtime loss cost, equipment maintenance cost, and potential future cost to obtain the total expected maintenance cost.

[0043] As a further improvement of the present invention, the obtaining of the downtime loss cost caused by downtime during maintenance specifically includes:

[0044] Obtain the cumulative production benefit value corresponding to the initial time node; among them, the cumulative production benefit is the cumulative production volume multiplied by the economic benefit that can be created by a unit production volume;

[0045] Predict the cumulative production benefit value corresponding to each maintenance time node through the benefit prediction sub-model;

[0046] Determine the downtime loss cost based on the cumulative benefit difference between the cumulative production benefit value corresponding to the initial time node and the cumulative production benefit value corresponding to the maintenance time node, the time length between the initial time node and the maintenance time node, and the expected downtime duration.

[0047] As a further improvement of the present invention, the specific steps for formulating the time-priority maintenance strategy include:

[0048] Take the maintenance method as the decision node, the cost factor as the branch condition, and the final total maintenance cost as the leaf node; among them, the maintenance methods include comprehensive maintenance and key component maintenance, and the cost factor data includes direct maintenance cost, downtime loss cost, and safety risk cost;

[0049] Take minimizing the total maintenance cost as the goal, perform branch division according to the value range of different cost factors, and train the decision tree model;

[0050] When the risk level of the special equipment is high risk, collect the cost factor data of the special equipment, perform preprocessing, input the preprocessed cost factor data into the trained decision tree model, start from the root node, traverse the decision tree downward in sequence according to the values of the cost factors until reaching the leaf node, and the maintenance method corresponding to the leaf node is the optimal maintenance method output.

[0051] As a further improvement of the present invention, the failure prediction sub-model includes: an input layer, an LSTM layer, an attention layer, a fully connected layer, and an output layer;

[0052] The input layer is used to input the preprocessed data;

[0053] The LSTM layer contains a hidden layer with multiple LSTM units, which is used to capture long-term dependencies in dynamic data;

[0054] The attention layer is used to calculate attention weights for the output sequence of the LSTM layer, and the attention output is obtained through weighted summation;

[0055] The fully connected layer integrates features by connecting the attention output of the attention layer;

[0056] The output layer uses the Softmax function to predict the probability of fault occurrence and the faults of special equipment components.

[0057] In a second aspect, the present invention provides a dynamic management system for optimizing the full life cycle cost of special equipment, including a model construction module, a data acquisition module, a fault prediction module, a risk division module, and a dynamic maintenance module, wherein:

[0058] The model construction module is used to construct a data model for the full life cycle of special equipment, and the data model includes a fault prediction sub-model and a benefit prediction sub-model;

[0059] The data acquisition module is used to collect the operation data, environmental data, and cumulative production benefits of special equipment in real time and perform preprocessing;

[0060] The fault prediction module is used to predict the probability of fault occurrence and the faulty components of special equipment;

[0061] The risk division module is used to construct a fuzzy comprehensive evaluation model based on the probability of fault occurrence and the faulty components, combined with the characteristics of special equipment, to divide the risk level of special equipment;

[0062] The dynamic maintenance module is used to formulate a dynamic maintenance strategy for special equipment by combining the risk level of special equipment and the predicted benefits of the benefit prediction sub-model through maintenance costs;

[0063] The feedback optimization module is used to record the actual maintenance cost and optimize the data model and the dynamic maintenance strategy according to the actual maintenance cost.

[0064] Advantages of the present invention:

[0065] By continuously collecting and analyzing the real-time operation status and environmental data of the equipment, a fault prediction sub-model of the equipment is constructed using machine learning technology to predict the probability of equipment failure occurrence and the faulty components, enabling equipment managers to anticipate potential problems before the failure occurs and take preventive maintenance measures, significantly reducing the downtime caused by sudden failures and ensuring the continuity and efficiency of equipment operation;

[0066] Combined with the probability of equipment failure and the risk level of special equipment, different dynamic management strategies are formulated. The optimal maintenance time of the equipment is dynamically determined through the total maintenance cost. This strategy can avoid unnecessary premature maintenance, reduce maintenance costs, and at the same time prevent the high repair costs caused by excessive equipment wear and reduce the production loss costs caused by equipment downtime. Through accurate prediction and reasonable arrangement of maintenance time, the optimal allocation of maintenance resources is achieved, the maintenance efficiency is improved, the service life of the equipment is extended, and the total life cycle cost of the equipment is controlled from the source. BRIEF DESCRIPTION OF THE DRAWINGS

[0067] Figure 1 It is a flowchart of a dynamic management method for optimizing the total life cycle cost of special equipment provided by the present invention;

[0068] Figure 2 It is a flowchart of a cost-priority maintenance strategy provided by the present invention;

[0069] Figure 3 It is a module interaction diagram of a dynamic management system for optimizing the total life cycle cost of special equipment provided by the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0070] In order to make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are some, but not all, of the embodiments of the present invention. All other embodiments obtained by those of ordinary skill in the art based on the described embodiments of the present invention without creative efforts shall fall within the scope of protection of the present invention.

[0071] Embodiment 1

[0072] Refer to Figure 1 and Figure 2 As shown, it is a specific implementation manner of a dynamic management method for optimizing the total life cycle cost of special equipment of the present invention.

[0073] Specifically, special equipment refers to equipment that involves life safety and has relatively great danger during use. Such equipment usually needs to follow strict regulations and standards, and requires regular inspections, maintenance, and monitoring to ensure the safe operation of special equipment. Common special equipment includes boilers, pressure vessels, elevators, cranes, etc.

[0074] A dynamic management method for optimizing the total life cycle cost of special equipment specifically includes the following steps:

[0075] S1. Construct a data model for the total life cycle of special equipment, and the data model includes a failure prediction sub-model and a benefit prediction sub-model;

[0076] By constructing a data model for special equipment, reliable analysis results can be provided for the subsequent analysis of special equipment during operation. Based on the analysis results, data support can be provided for the maintenance and repair plan, improving the efficiency and accuracy of special equipment detection, thereby formulating a reasonable maintenance and repair plan and optimizing the management cost of special equipment during operation.

[0077] S2. Collect the operation data and environmental data of special equipment, preprocess them, and input them into the fault prediction sub-model to predict the probability of fault occurrence and the faulty components of special equipment at a preset time point;

[0078] By comprehensively collecting the operation data of special equipment, such as operation time, load rate, and fatigue strength, etc., a comprehensive data source is provided for the evaluation of the operation status of special equipment. Through the influence of environmental data on special equipment during operation, such as high environmental humidity will corrode special equipment, and high environmental temperature will affect the material properties of special equipment, increasing the risk or probability of failure. By combining operation data with environmental data to comprehensively predict potential faults or problems of special equipment, the accuracy and reliability of special equipment fault prediction can be increased.

[0079] Preferably, the operation data of the special equipment includes static data and dynamic data. The dynamic data is used to record the time series data monitored during the operation of the special equipment, including operation time, load rate, vibration, corrosion rate, stress, load, and fatigue strength, which are respectively collected through corresponding sensors;

[0080] The static data is used to record the historical data of special equipment, including the expected service life of special equipment, the used service life of special equipment, historical maintenance data, the maintenance type of special equipment, records of replaced or repaired parts, the cost of each maintenance, and industry standard data; among them, the maintenance type of special equipment includes preventive maintenance, corrective maintenance, and emergency maintenance, and the industry standard data includes national standards and regulations or industry association guidelines. The historical data of special equipment is obtained through the technical documents of special equipment or the enterprise's internal management system.

[0081] Preferably, the fault prediction sub-model introduces the long short-term memory network in time series analysis combined with the attention mechanism to deeply analyze the dynamic data and static data of special equipment and calculate the future fault probability more accurately.

[0082] Furthermore, the specific steps of S2 include:

[0083] S21: Preprocess the operation data and environmental data, specifically including: using the sliding window algorithm to identify and remove outliers. For missing values, adopt the multiple imputation method based on deep learning, and use the generative adversarial network (GAN) combined with historical data to generate reasonable imputed values to ensure the integrity and accuracy of the data; normalize different types of data;

[0084] S22: Construct a fault prediction sub-model, including: an input layer, an LSTM layer, an attention layer, a fully connected layer, and an output layer. Among them, the input layer is used to input the preprocessed data; the LSTM layer contains a hidden layer with multiple LSTM units, which is used to capture the long-term dependencies in the time series data (dynamic data) and perform in-depth feature extraction on the input time series data; the attention layer is used to calculate the attention weights for the output sequence of the LSTM layer, and then obtain the attention output through weighted summation, enabling the model to focus on the time step features that are most important for fault prediction; the fully connected layer integrates the features by connecting the attention output of the attention layer. The fully connected layer includes at least two layers, and the number of neurons in each layer gradually decreases; the output layer uses the Softmax function for fault occurrence probability prediction and special equipment component fault prediction. For fault occurrence probability prediction, output the probabilities of fault occurrence and non-occurrence. For special equipment component fault prediction, output the probability distribution of possible faults for each component.

[0085] S23: Divide the preprocessed data into a training set, a validation set, and a test set according to a preset ratio. Use the validation set to train the fault prediction sub-model, perform multiple iterations on the training set to update the model parameters, use the validation set to verify the fault prediction sub-model, and adopt the early stopping method to prevent overfitting. When the loss function on the validation set converges, stop training to obtain the trained fault prediction sub-model. Among them, the loss function uses the cross-entropy loss function, and the Adam W optimizer is adopted to prevent the model from overfitting.

[0086] S24: Input the dynamically collected and preprocessed data into the trained fault prediction sub-model to output the probability value of fault occurrence. At the same time, compare the predicted fault probability distribution of each equipment component in the output layer with the preset component probability threshold for each equipment component. When the predicted fault probability of the equipment component is greater than the corresponding component probability threshold, it is determined that a possible component fault has occurred.

[0087] S3: Based on the fault occurrence probability and the faulty components, combined with the characteristics of special equipment, construct a fuzzy comprehensive evaluation model to divide the risk levels of special equipment;

[0088] Based on the probability of failure occurrence output by the failure prediction sub-model and the equipment components that may fail, combined with other data of special equipment, including operation data, historical failure rates, and the complexity of the equipment operation environment, etc., comprehensively evaluate the risk level of special equipment, providing a basis for formulating corresponding dynamic maintenance strategies in the follow-up, so as to formulate different maintenance methods or times according to different risk levels, optimize the costs generated by failures during the operation of special equipment, and thus optimize the life-cycle cost of special equipment.

[0089] Furthermore, constructing a fuzzy comprehensive evaluation model to divide the risk levels of special equipment specifically includes the following steps:

[0090] S31: Obtain the probability of failure occurrence and the failed components, as well as the operation data, select the evaluation factors, and determine the evaluation factor set;

[0091] Preferably, in addition to the probability of failure occurrence and the failed components of special equipment, key data are also selected as evaluation factors according to the operation data in combination with the characteristics of special equipment. For example, for boilers, select the degree of pressure abnormality (the deviation ratio of the pressure exceeding the normal range), temperature deviation (the interpolation of the actual temperature and the equipment temperature), water quality condition (indicators such as impurities and acidity in water), historical probability of failure occurrence, etc. as key data to determine the evaluation factor set. For cranes, select the degree of deviation of the lifting weight (the difference ratio between the actual lifting weight and the rated lifting weight), the degree of abnormality of the lifting speed (the deviation of the actual lifting speed from the standard lifting speed), the wear rate of the walking wheels (the degree of wear of the walking wheels, determined by measuring the ratio of the reduction in the diameter of the walking wheels to the initial diameter), etc. as key data to determine the evaluation factor set. The specific evaluation factor set is determined by those skilled in the art according to the characteristics of special equipment by selecting appropriate key factors. By selecting evaluation factors, the safety risk status of special equipment is reflected from multiple dimensions, and considering these factors comprehensively avoids the one-sidedness of a single factor.

[0092] S32: Divide the risk levels of special equipment into three levels, namely normal operation, low risk, and high risk, and construct an evaluation level set;

[0093] S33: Compare the relative importance of each evaluation factor in the evaluation factor set pairwise, construct a judgment matrix, calculate the eigenvector and the maximum eigenvalue in the judgment matrix, and obtain the weight of each evaluation factor;

[0094] S33 specifically includes the following steps:

[0095] Adopt the expert investigation method to compare the relative importance of each evaluation factor in the evaluation factor set pairwise. For example, compare the influence of the probability of failure occurrence and the equipment operation data on the safety risk, and the experts give judgments on the relative importance according to their experience;

[0096] S331: Construct a judgment matrix based on the judgment results of experts , where n is the total number of evaluation factors represents the evaluation factor and the evaluation factor 's relative importance, with values ranging from 1 - 9 and their reciprocals. 1 indicates equal importance of both, and 9 indicates is extremely more important than ;

[0097] S332: Calculate the product of each row element of the judgment matrix, then take the nth root of the obtained product to get the initial weights. Normalize the initial weights to obtain the weights of each evaluation factor, and form the eigenvector of the judgment matrix

[0098] S333: Calculate the maximum eigenvalue based on the judgment matrix and the eigenvector; The calculation formula for the maximum eigenvalue is: , where is the maximum eigenvalue, W represents the eigenvector of the judgment matrix represents the weight of the i-th evaluation factor

[0099] S334: Calculate the consistency ratio through the maximum eigenvalue; The calculation formula for the consistency ratio is: , where represents the consistency ratio represents the average random consistency index, which is obtained by referring to a table

[0100] S335: Compare the consistency ratio with a preset threshold. If the consistency ratio is less than the preset threshold, it is determined that the consistency of the judgment matrix is acceptable, indicating that the calculated eigenvector as the weights of each evaluation factor is reasonable and reliable. If the consistency ratio is greater than or equal to the preset threshold, it means that the consistency of the judgment matrix is poor, and it is necessary to readjust the judgment of experts or correct the judgment matrix. Repeat S332 - S334 to recalculate the eigenvector and the maximum eigenvalue until the consistency test passes

[0101] The weights reflect the relative importance of each evaluation factor in the safety risk assessment. By constructing a judgment matrix to determine the weights of the evaluation factors and through the consistency test to judge the rationality of the weight allocation, the evaluation results can be made more in line with the actual situation, avoiding estimation deviations caused by unreasonable weights of evaluation factors. For example, if the probability of a fault occurrence is high, it indicates that its impact on the risk level is greater, and it should be focused on during the assessment

[0102] S34: Determine the membership degrees of each evaluation factor to different risk levels, and establish a fuzzy relation matrix

[0103] By statistically analyzing historical data, the membership degree of each evaluation factor to different risk levels is determined to form a membership degree vector, and the membership degree vectors of all evaluation factors are combined to form a fuzzy relation matrix. The fuzzy relation matrix reflects the degree of association between each evaluation factor and different risk levels and is the key data for fuzzy comprehensive evaluation. It converts the specific state of each factor into a membership relationship with different risk levels, providing a data basis for subsequent comprehensive evaluation.

[0104] S35: Through fuzzy composition operation, a comprehensive evaluation result vector is obtained. According to the principle of maximum membership degree, the risk level of special equipment is determined.

[0105] Through fuzzy composition operation, all evaluation factors and their weights are comprehensively considered to obtain a comprehensive evaluation result vector that comprehensively reflects the safety risk status of special equipment. The principle of maximum membership degree makes the determination of the risk level more clear and objective, and can quickly and accurately judge the risk level of the equipment.

[0106] S4. Combining the risk level of special equipment and the predicted benefits of the benefit prediction sub-model, a dynamic maintenance strategy for special equipment is formulated through maintenance costs;

[0107] Based on the risk level of special equipment and the probability of equipment failure, different maintenance strategies are formulated. For different situations, different maintenance methods are adopted to reduce unnecessary resource allocation and waste. For special equipment operating normally, no maintenance is required. For special equipment with a certain risk, different maintenance times and methods are determined according to maintenance costs, so as to control the maintenance costs of special equipment.

[0108] The dynamic maintenance strategy includes a cost-priority maintenance strategy and a time-priority maintenance strategy;

[0109] Configure a failure probability threshold. When the failure probability of special equipment is less than or equal to the failure probability threshold or the risk level is normal operation, no dynamic maintenance strategy is formulated; this indicates that the operating state of special equipment is good and it can maintain stable operation for a long time without maintenance, saving management costs;

[0110] When the failure probability of special equipment is greater than the failure probability threshold and the risk level is normal operation or low risk, a cost-priority maintenance strategy is formulated; this indicates that there may be some potential problems or failures in the operating state of special equipment and a maintenance plan needs to be arranged. Due to the production plan arrangements of special equipment at different times, the accumulation of equipment operating states, etc., the costs generated by maintenance at different times may be different. Therefore, choosing a suitable maintenance time node can optimize and reduce costs;

[0111] When the risk level of special equipment is high risk, a time-priority maintenance strategy is formulated; this indicates that the special equipment is very likely to have a malfunction. If it continues to operate, it may have a greater impact, and it is necessary to immediately stop the machine for maintenance. However, the maintenance costs generated by different maintenance methods are different. By formulating a time-priority maintenance strategy and selecting an appropriate maintenance method for maintenance, the cost can be further optimized.

[0112] Furthermore, the specific steps of the cost-priority maintenance strategy include: S41: When the probability of failure occurrence is greater than the failure occurrence probability threshold, trigger a warning signal to provide equipment maintenance reminders;

[0113] S42: Take the moment of receiving the equipment maintenance reminder as the start time node of the equipment maintenance process, take the latest maintenance time point as the end time node, and divide it at intervals of a preset first fixed duration to form a first maintenance time node sequence; among them, the first fixed duration and the latest maintenance time point are set by management or production personnel according to actual needs;

[0114] S43: Calculate the total maintenance cost expected to be generated when each maintenance time node in the first maintenance time node sequence starts to perform maintenance work, and associate the total maintenance cost with the first maintenance time node to form a first total maintenance cost set;

[0115] S44: Configure a total cost threshold, and add the maintenance time nodes in the first total maintenance cost set whose total maintenance cost is less than the total cost threshold to the time node set; among them, the total cost threshold is set by those skilled in the art according to actual needs;

[0116] S45: Conduct judgment and analysis on the time node set to form a second maintenance time node sequence, and calculate and obtain a second total maintenance cost set;

[0117] The specific steps of S45 include:

[0118] Judge whether there are adjacent maintenance time nodes in the time node set. If there are adjacent maintenance time nodes, merge the adjacent maintenance time nodes;

[0119] Take the earliest maintenance time node among the adjacent maintenance time nodes as the start time node for secondary judgment, take the latest maintenance time node among the adjacent maintenance time nodes as the end time node for secondary judgment, and take the duration between the start time node for secondary judgment and the end time node for secondary judgment as the secondary judgment maintenance time period;

[0120] The secondary judgment maintenance time period is divided at intervals of a second fixed duration to form a second sequence of maintenance time nodes. Calculate the total expected maintenance cost when starting maintenance work at each maintenance time node in the second sequence of maintenance time nodes, and associate the total maintenance cost with the maintenance time node to form a second set of total maintenance costs;

[0121] If there are no adjacent maintenance time nodes, define the second set of total maintenance costs as an empty set.

[0122] Among them, the second fixed duration is at least less than half of the first fixed duration, so as to add new maintenance time nodes among adjacent maintenance time nodes to determine more detailed maintenance time nodes, avoid missing the optimal maintenance time nodes due to too large a step size of the first fixed duration, and also reduce the calculation burden increased due to too small a step size of the first fixed duration. The second fixed duration is specifically set by management or production personnel according to actual needs;

[0123] S46: Sort the total maintenance costs in the first set of total maintenance costs and the second set of total maintenance costs to form a sequence of maintenance costs from low to high;

[0124] S47: Select the maintenance time node corresponding to the lowest total maintenance cost from the sequence of maintenance costs as the optimal equipment maintenance time.

[0125] Exemplarily, if a special equipment, such as a boiler, detects that the probability of a fault occurring is greater than the fault occurrence probability threshold (such as 0.7) at 9:00, a warning signal is triggered to provide an equipment maintenance reminder. Taking 9:00 as the start time node of the equipment maintenance process, management or production personnel set to complete the maintenance within 12 hours, that is, 21:00 as the end time node. Divide it at intervals of the first fixed duration (such as 3 hours), and the obtained maintenance time nodes are 9:00, 12:00, 15:00, 18:00, 21:00, forming a first sequence of maintenance time nodes. Calculate the total expected maintenance cost when starting maintenance work at each maintenance time node, such as corresponding to 16,000 yuan, 14,000 yuan, 17,000 yuan, 12,000 yuan, 10,000 yuan respectively, to form a first set of total maintenance costs. Configure the total cost threshold to be 15,000 yuan. Then, the maintenance time nodes in the first set of total maintenance costs where the total maintenance cost is less than the corresponding total cost threshold are 12:00, 18:00, and 21:00. Add them to the set of time nodes. It is judged that 18:00 and 21:00 are adjacent maintenance time nodes and are merged. Take 18:00 as the start time node for secondary judgment and 21:00 as the end time node for secondary judgment. Take the three hours between 18:00 and 21:00 as the secondary judgment maintenance time period;

[0126] Divide the secondary judgment maintenance time period at intervals of a second fixed duration (such as 1 hour), and the obtained maintenance time nodes are 18:00, 19:00, 20:00, 21:00, forming a second maintenance time node sequence. Calculate the total maintenance cost expected to be generated when the maintenance work starts at each maintenance time node in the second maintenance time node sequence, such as corresponding to 12,000 yuan, 8,000 yuan, 9,000 yuan, and 10,000 yuan respectively, to form a second total maintenance cost set. Sort all the total maintenance costs to form a maintenance cost sequence from low to high, and select the maintenance time node (19:00) corresponding to the lowest total maintenance cost (8,000 yuan) as the optimal equipment maintenance time.

[0127] By dividing different maintenance time nodes and determining the optimal equipment maintenance time according to the total maintenance cost of each maintenance time node, it is possible to optimize the maintenance cost of special equipment after sending a device maintenance reminder and avoid cost waste caused by inappropriate selection of maintenance time.

[0128] The calculation of the total maintenance cost expected to be generated when the maintenance work starts at each maintenance time node includes:

[0129] Obtain the shutdown loss cost caused by shutdown during the maintenance process through the cumulative production benefit of special equipment; specifically, it includes the following steps:

[0130] Obtain the cumulative production benefit value corresponding to the initial time node; among them, the cumulative production benefit is the cumulative production volume multiplied by the economic benefit that can be created by a unit of production volume;

[0131] Predict the cumulative production benefit value corresponding to each maintenance time node through the benefit prediction sub-model; the benefit prediction sub-model is trained by obtaining historical production data and market data, and the benefit prediction sub-model is an ARIMA model or an LSTM model;

[0132] Determine the shutdown loss cost through the cumulative benefit difference between the cumulative production benefit value corresponding to the initial time node and the cumulative production benefit value corresponding to the maintenance time node, the time length between the initial time node and the maintenance time node, and the expected shutdown duration; among them, the expected shutdown duration is represented by the average shutdown duration in actual maintenance for faults that are the same or similar to the current fault situation in the maintenance database; the specific calculation method of the shutdown loss cost is:

[0133]

[0134] Among them, represents the shutdown loss cost, represents the cumulative production benefit value corresponding to the initial time node, represents the cumulative production benefit value corresponding to the kth time node, represents the time length from the initial time node to the k-th maintenance time node, and t represents the estimated shutdown maintenance duration.

[0135] The equipment maintenance cost generated during the estimated maintenance of special equipment is calculated through labor costs, equipment part replacement costs, and non-linear equipment health status impact factors; among them, the labor cost is the product of the unit labor hour cost and the labor hours spent.

[0136]

[0137] Among them, represents the equipment maintenance cost, H represents the number of labor hours required for equipment maintenance, L represents the unit labor hour cost, R represents the equipment part replacement cost, represents the non-linear equipment health status impact factor, is the equipment health status decay rate constant, which is obtained by calculating the average rate of equipment health status degradation over time using the regression model based on historical maintenance data in the maintenance database;

[0138] It should be noted that the non-linear equipment health status impact factor is obtained by fitting the equipment maintenance cost, maintenance time nodes, and related health status indicators. The model parameters are estimated by methods such as the least squares method, so as to obtain the quantitative relationship between the non-linear equipment health status impact factor and the maintenance time nodes. Due to potential problems or faults that may occur inside special equipment, the later the maintenance time node, the greater the impact on the equipment health status may be, and at the same time, the greater the maintenance difficulty and the longer the labor hours spent may be. By introducing the non-linear equipment health status impact factor, the non-linear impact of the maintenance time node on the equipment maintenance cost is effectively considered, increasing the accuracy of the equipment maintenance cost.

[0139] Obtain the failure occurrence probability at each maintenance time node, and combine the average cost of historical maintenance costs to obtain the potential future cost, which is used to represent the loss cost caused by the failure that is expected to occur from the initial time node to the maintenance time node due to non-maintenance; the calculation formula for the potential future cost is:

[0140]

[0141] Among them, represents the potential future cost, represents the failure occurrence probability of the equipment at the k-th maintenance time node predicted by the failure prediction sub-model, represents the average cost of historical maintenance costs, which is represented by the average value of the costs spent in historical maintenance for faults that are the same or similar to the current fault situation in the maintenance database.

[0142] The total maintenance cost to be incurred is obtained by weighted summation of the downtime loss cost, equipment maintenance cost, and potential future cost.

[0143] Furthermore, the specific steps of the time - priority maintenance strategy include:

[0144] Taking the maintenance method as the decision node, cost factors as the branch conditions, and the final total maintenance cost as the leaf node. A decision tree model is constructed through historical data and experience. During the model training process, with the goal of minimizing the total cost, branch division is carried out according to the value ranges and inter - relationships of different cost factors. When facing a new high - risk equipment maintenance decision, the cost factor data is input into the decision tree model, and the model outputs the optimal maintenance method according to the rules obtained from training.

[0145] Comprehensively collect the maintenance case data of special equipment with a high - risk level in the past, including data of different types and specifications. The maintenance case data includes the maintenance methods adopted each time and cost factor data. Maintenance methods such as emergency comprehensive maintenance and key component maintenance, and cost factor data includes direct maintenance cost, downtime loss cost, and safety risk cost. Among them, the direct maintenance cost is the maintenance labor cost and the cost of replacing parts, the downtime loss cost is the production cost loss caused by equipment downtime during maintenance, and the safety risk cost is the risk loss estimated to still exist after maintenance, such as the accident loss caused by the recurrence of faults in the short term after maintenance. It can be calculated through historical data statistics and risk assessment models. For example, it is measured by the product of the probability of fault recurrence after maintenance of similar historical equipment and the average loss caused by the recurrence of faults.

[0146] Organize the collected data to ensure that the data format is unified and standardized. Label the maintenance method as the category label and cost factors as the feature data. At the same time, check the integrity and accuracy of the data, and process missing values and outliers.

[0147] Normalize cost factors such as direct maintenance cost, downtime loss cost, and safety risk cost, and select cost factors that have a greater impact on the decision of maintenance methods as features.

[0148] Initialize the decision tree. Taking the maintenance method as the decision node, for each cost factor, calculate its information gain or Gini index. Select the cost factor with the largest information gain or the smallest Gini index as the optimal splitting feature of the current node. According to the value range of the selected splitting feature, divide the sample set into different sub - nodes, and each sub - node contains samples in the corresponding interval. For example, if it is calculated that the information gain of the direct maintenance cost is the largest, then take the direct maintenance cost as the first splitting feature. If the direct maintenance cost is divided into three intervals: high, medium, and low, then divide the sample set into three sub - nodes.

[0149] Repeat the above process of calculating the information gain or Gini index, selecting the optimal splitting feature, and splitting the node for each child node, and recursively construct the subtree of the decision tree until the stopping condition is met. The stopping condition can be that the number of samples in the node is less than the preset threshold, or the depth of the decision tree reaches the preset value;

[0150] Use the sorted maintenance case data to train the constructed decision tree. By continuously adjusting the structure and parameters of the decision tree, the decision tree can accurately predict the corresponding maintenance method according to the cost factors. During the training process, the goal is to minimize the total maintenance cost, that is, when the decision tree predicts the maintenance method, make the total cost corresponding to the prediction result as close as possible to the actual minimum total cost.

[0151] To prevent the decision tree from overfitting, pruning operations are performed on the trained decision tree. Pruning is divided into pre-pruning and post-pruning. Pre-pruning is to stop splitting the node when a certain condition is met (such as the information gain is less than a certain threshold) during the construction of the decision tree; post-pruning is to start from the leaf node and gradually evaluate the nodes upward after the decision tree is constructed. If the performance of the model (such as accuracy, F1 value, etc.) on the validation set does not decrease after cutting off a certain node, then cut off the node;

[0152] Deploy the trained decision tree to the decision application. When the risk level of the special equipment is high risk, collect the data of various cost factors of the special equipment, and perform preprocessing. Input the preprocessed data into the trained decision tree model. Starting from the root node, traverse the decision tree downward in turn according to the values of the cost factors until reaching the leaf node. The maintenance method corresponding to the leaf node is the optimal maintenance method output.

[0153] S5. Perform maintenance on the special equipment according to the dynamic maintenance strategy, record the actual maintenance cost, and optimize the data model and dynamic maintenance strategy according to the actual maintenance cost.

[0154] Establish a real-time data feedback mechanism to timely feedback information such as the operation data of the equipment after maintenance, the change of risk level, and the actual maintenance cost. Through the analysis of these real-time data, continuously optimize the dynamic maintenance strategy for balancing the risk and cost of special equipment. For example, if it is found that the actual operating condition of the special equipment is better than expected, the risk level drops significantly, and the actual maintenance cost is lower than the expected cost after a certain maintenance, then when formulating the maintenance strategy for similar equipment in the future, this experience can be appropriately referred to and the maintenance time and content can be adjusted.

[0155] Regularly conduct a comprehensive assessment of the dynamic maintenance strategy for balancing the risks and costs of special equipment, and update and improve the strategy in combination with the latest safety regulations, technical standards, and industry best practices. For example, conduct a comprehensive assessment of the maintenance strategy every quarter or every six months, analyze the problems and deficiencies in the implementation process of the strategy, and promptly adjust the parameters and rules in the strategy to ensure that the maintenance strategy always meets the safety operation and cost management requirements of special equipment.

[0156] Embodiment 2

[0157] Reference Figure 3 As shown, this embodiment is the second embodiment of the present invention. Based on the same inventive concept as Embodiment 1, this embodiment introduces a specific implementation method of a dynamic management system for optimizing the full life cycle cost of special equipment, including a model construction module, a data acquisition module, a fault prediction module, a risk division module, a dynamic maintenance module, a feedback optimization module, and a maintenance record module, where:

[0158] The model construction module is used to construct a data model for the full life cycle of special equipment, and the data model includes a fault prediction sub-model and a benefit prediction sub-model;

[0159] The data acquisition module is used to collect the operation data, environmental data, and cumulative production benefits of special equipment in real time and perform preprocessing;

[0160] The fault prediction module is used to input the preprocessed data into the fault prediction sub-model to predict the probability of equipment failure and the faulty components;

[0161] The risk division module is used to construct a fuzzy comprehensive evaluation model based on the probability of equipment failure and the faulty components, combined with the characteristics of special equipment, to divide the risk level of special equipment;

[0162] The dynamic maintenance module is used to formulate a dynamic maintenance strategy for special equipment through maintenance costs in combination with the risk level of special equipment and the predicted benefits of the benefit prediction sub-model;

[0163] The feedback optimization module is used to record the actual maintenance costs and optimize the data model and the dynamic maintenance strategy according to the actual maintenance costs.

[0164] The maintenance record module is used to record the actual maintenance data, including the maintenance time, replaced parts, maintenance effect, and the actual maintenance cost of this maintenance; so that the staff can view the real-time operation status of the equipment, the probability of equipment failure, the maintenance plan, and the actual maintenance data.

[0165] Working principle and its effect:

[0166] By continuously collecting and analyzing the real-time operating status and environmental data of the equipment, a fault prediction sub-model of the equipment is constructed using machine learning technology to predict the occurrence probability of equipment faults and faulty components, enabling equipment management personnel to foresee potential problems before the occurrence of faults and take preventive maintenance measures, significantly reducing the downtime caused by sudden faults and ensuring the continuity and efficiency of equipment operation;

[0167] Combined with the probability of equipment failure and the risk level of special equipment, different dynamic management strategies are formulated. By the total maintenance cost, the optimal maintenance time of the equipment is dynamically determined. This strategy can avoid unnecessary premature maintenance, reduce maintenance costs, and at the same time prevent the high repair costs caused by excessive wear of the equipment and reduce the production loss costs caused by equipment downtime. Through accurate prediction and reasonable arrangement of maintenance time, the optimal allocation of maintenance resources is achieved, the maintenance efficiency is improved, the service life of the equipment is extended, and the total life cycle cost of the equipment is controlled from the source.

[0168] In addition, parts of the above technical solutions provided in the embodiments of the present application that are consistent with the implementation principles of the corresponding technical solutions in the prior art are not described in detail to avoid excessive elaboration.

[0169] The above preset parameters or preset thresholds are all set by those skilled in the art according to the actual situation or obtained through a large amount of data simulation.

[0170] The embodiments of the present invention have been described above in conjunction with the accompanying drawings. However, the present invention is not limited to the above specific embodiments. The above specific embodiments are merely illustrative and not restrictive. Under the inspiration of the present invention, those of ordinary skill in the art can make changes, modifications, substitutions, and variations to the above embodiments without departing from the spirit of the present invention and the scope protected by the claims. All of these are within the protection scope of the present invention.

Claims

1. A dynamic management method for optimizing the full - life - cycle cost of special equipment, characterized in that: It includes the following steps: Construct a data model for the whole life cycle of special equipment, where the data model includes a fault prediction sub-model and a benefit prediction sub-model; Collect the operation data and environmental data of special equipment, and input them into the fault prediction sub-model after preprocessing to predict the probability of fault occurrence and the faulty components of special equipment; Based on the probability of fault occurrence and the faulty components, combined with the characteristics of special equipment, construct a fuzzy comprehensive evaluation model to divide the risk levels of special equipment; Combined with the risk level of special equipment and the predicted benefits of the benefit prediction sub-model, formulate a dynamic maintenance strategy for special equipment through maintenance costs; Carry out maintenance on special equipment according to the dynamic maintenance strategy, record the actual maintenance costs, and optimize the data model and the dynamic maintenance strategy according to the actual maintenance costs; The dynamic maintenance strategy includes a cost-priority maintenance strategy. The steps for formulating the cost-priority maintenance strategy specifically include: S41: When the probability of fault occurrence is greater than the fault occurrence probability threshold, trigger a warning signal to provide equipment maintenance reminders; S42: Take the moment of receiving the equipment maintenance reminder as the start time node of the equipment maintenance process, take the latest maintenance time point as the end time node, and divide it at intervals of a preset first fixed duration to form a first maintenance time node sequence; S43: Calculate the total maintenance cost expected to be generated when starting the maintenance work at each maintenance time node in the first maintenance time node sequence, associate the total maintenance cost with the maintenance time node, and form a first total maintenance cost set; S44: Configure the total cost threshold, and add the maintenance time nodes corresponding to the total maintenance cost less than the total cost threshold in the first total maintenance cost set to the time node set; S45: Conduct judgment and analysis on the time node set to form a second maintenance time node sequence, and calculate and obtain a second total maintenance cost set; S46: Sort the total maintenance costs in the first total maintenance cost set and the second total maintenance cost set to form a maintenance cost sequence from low to high; S47: Select the maintenance time node corresponding to the lowest total maintenance cost in the maintenance cost sequence as the optimal equipment maintenance time.

2. The dynamic management method for optimizing the full - life - cycle cost of special equipment according to claim 1, wherein: The specific steps of constructing the fuzzy comprehensive evaluation model to divide the risk levels of special equipment specifically include: Obtain the probability of fault occurrence, the faulty components, and the operation data, select evaluation factors, and determine the evaluation factor set; Divide the risk levels of special equipment into three levels, namely normal operation, low risk, and high risk, and construct an evaluation level set; Compare the relative importance of each evaluation factor in the evaluation factor set pairwise, construct a judgment matrix, calculate the eigenvector and the largest eigenvalue in the judgment matrix, and obtain the weights of each evaluation factor; Determine the membership degree of each evaluation factor to different risk levels, and establish a fuzzy relation matrix; Through fuzzy composition operation, obtain a comprehensive evaluation result vector, and determine the risk level of special equipment according to the maximum membership degree principle.

3. The dynamic management method for optimizing the life cycle cost of special equipment according to claim 2, characterized in that: The dynamic maintenance strategy includes a cost-priority maintenance strategy and a time-priority maintenance strategy. The specific steps of formulating the dynamic maintenance strategy for special equipment through maintenance costs specifically include: Configure the failure occurrence probability threshold. When the failure occurrence probability of special equipment is less than or equal to the failure occurrence probability threshold or the risk level is normal operation, no dynamic maintenance strategy is formulated; When the failure occurrence probability of special equipment is greater than the failure occurrence probability threshold and the risk level is normal operation or low risk, a cost-priority maintenance strategy is formulated; When the risk level of special equipment is high risk, a time-priority maintenance strategy is formulated.

4. A dynamic management method for optimizing the full - life - cycle cost of special equipment according to claim 1, characterized in that: The judgment and analysis of the time node set to form the second maintenance time node sequence and calculate the second total maintenance cost set include: Judge whether there are adjacent maintenance time nodes in the time node set. If there are adjacent maintenance time nodes, merge the adjacent maintenance time nodes; Take the earliest maintenance time node among the adjacent maintenance time nodes as the start time node for secondary judgment, take the latest maintenance time node among the adjacent maintenance time nodes as the end time node for secondary judgment, and take the duration between the start time node for secondary judgment and the end time node for secondary judgment as the maintenance time period for secondary judgment; Divide the maintenance time period for secondary judgment at intervals of the second fixed duration to form the second maintenance time node sequence, calculate the total maintenance cost expected to be generated when each maintenance time node in the second maintenance time node sequence starts to perform maintenance work, and associate the total maintenance cost with the maintenance time node to form the second total maintenance cost set; If there are no adjacent maintenance time nodes, define the second total maintenance cost set as an empty set.

5. The dynamic management method for optimizing the life-cycle cost of special equipment according to claim 4, characterized in that: The calculation of the total maintenance cost expected to be generated when starting to perform maintenance work includes: Obtain the shutdown loss cost caused by shutdown during the maintenance process through the cumulative production benefit of special equipment; Calculate the equipment maintenance cost generated by special equipment during the expected maintenance process through labor costs, replacement costs of equipment parts, and the non-linear equipment health status impact factor; among them, the labor cost is the product of the unit labor hour cost and the labor hours spent; Obtain the failure occurrence probability of each maintenance time node, and obtain the potential future cost in combination with the average cost of historical maintenance costs. The potential future cost is used to represent the loss cost caused by the expected failure from the initial time node to the maintenance time node; Perform a weighted sum of the shutdown loss cost, equipment maintenance cost, and potential future cost to obtain the total maintenance cost expected to be generated.

6. The dynamic management method for optimizing the life-cycle cost of special equipment according to claim 5, characterized in that: The specific method for obtaining the shutdown loss cost caused by shutdown during the maintenance process includes: Obtain the cumulative production benefit value corresponding to the initial time node; Predict the cumulative production benefit value corresponding to each maintenance time node through the benefit prediction sub-model; Determine the shutdown loss cost through the cumulative benefit difference between the cumulative production benefit value corresponding to the initial time node and the cumulative production benefit value corresponding to the maintenance time node, the time length between the initial time node and the maintenance time node, and the expected shutdown duration.

7. A dynamic management method for optimizing the life-cycle cost of special equipment according to claim 3, characterized in that: The specific steps for formulating the time-priority maintenance strategy include: Take the maintenance method as the decision node, the cost factor as the branch condition, and the final total maintenance cost as the leaf node; among them, the maintenance methods include comprehensive maintenance and key component maintenance, and the cost factor data includes direct maintenance cost, downtime loss cost, and safety risk cost; Aiming at minimizing the total maintenance cost, divide the branches according to the value ranges of different cost factors, and train the decision tree model; When the risk level of special equipment is high risk, collect the cost factor data of special equipment, preprocess it, and input the preprocessed cost factor data into the trained decision tree model. Starting from the root node, traverse the decision tree downward in sequence according to the values of the cost factors until reaching the leaf node. The maintenance method corresponding to the leaf node is the optimal maintenance method output.

8. A dynamic management method for optimizing the life-cycle cost of special equipment according to claim 1, characterized in that: The fault prediction sub-model includes: an input layer, an LSTM layer, an attention layer, a fully connected layer, and an output layer; The input layer is used to input the preprocessed data; The LSTM layer contains a hidden layer with multiple LSTM units, which is used to capture the long-term dependencies in dynamic data; The attention layer is used to calculate the attention weights for the output sequence of the LSTM layer, and obtain the attention output through weighted summation; The fully connected layer integrates the features by connecting the attention output of the attention layer; The output layer uses the Softmax function, which is used for predicting the probability of fault occurrence and predicting the faults of special equipment components.

9. A dynamic management system for optimizing the life-cycle cost of special equipment, which is used to implement a dynamic management method for optimizing the life-cycle cost of special equipment according to any one of claims 1-8, characterized in that: It includes a model construction module, a data acquisition module, a fault prediction module, a risk division module, and a dynamic maintenance module, where: The model construction module is used to construct a data model for the whole life cycle of special equipment, and the data model includes a fault prediction sub-model and a benefit prediction sub-model; The data acquisition module is used to collect the operation data, environmental data, and cumulative production benefits of special equipment in real time, and preprocess them; The fault prediction module is used to predict the probability of fault occurrence and the faulty components of special equipment; The risk division module is used to construct a fuzzy comprehensive evaluation model based on the probability of fault occurrence and the faulty components, and combine the characteristics of special equipment to divide the risk level of special equipment; The dynamic maintenance module is used to formulate a dynamic maintenance strategy for special equipment through maintenance costs in combination with the risk level of special equipment and the predicted benefits of the benefit prediction sub-model; The feedback optimization module is used to record the actual maintenance cost, and optimize the data model and the dynamic maintenance strategy according to the actual maintenance cost.

Images