A method and system for auxiliary decision-making in operation and maintenance of ship equipment

Abstract

The present invention relates to the technical field of ship equipment operation and maintenance, and discloses an operation and maintenance auxiliary decision method and system for ship equipment, the method comprising: collecting historical status data of ship equipment; preprocessing the historical status data to obtain preprocessed historical status data; based on the preprocessed historical status data, using a pre-trained fault diagnosis model to perform fault diagnosis, output the fault type diagnosis result, and generate a diagnosis report; formulating and executing a maintenance plan according to the diagnosis report, and recording the maintenance data of the maintenance process; iteratively updating the fault diagnosis model according to the maintenance data to obtain the optimal fault diagnosis model, and re-diagnosing the fault using the optimal fault diagnosis model to improve the operation and maintenance auxiliary decision of ship equipment. The fault diagnosis model is iteratively updated to ensure that the model can adapt to changes in the equipment status. By updating the fault diagnosis model, the prediction accuracy can be improved, thereby better supporting operation and maintenance decisions and reducing equipment downtime.

Description

Technical Field

[0001] The present invention relates to the technical field of ship equipment operation and maintenance, and in particular to an auxiliary decision-making method and system for operation and maintenance of ship equipment. Background Art

[0002] With the development of the shipping industry and technological advancements, the complexity and degree of automation of ship equipment are constantly increasing, which places higher demands on the operation and maintenance of equipment. In order to improve the efficiency and accuracy of equipment operation and maintenance, more and more research has begun to focus on how to use machine learning technology for fault prediction and decision-making assistance. Existing methods for predicting equipment failures for operation and maintenance based on machine learning models are mainly based on machine learning models of historical data: this type of method uses historical status data to train machine learning models (such as support vector machines, random forests, neural networks, etc.), and predicts future failures by learning the characteristics of the equipment in normal and faulty states. Although this method has improved the accuracy of diagnosis to a certain extent, the following problems still exist:

[0003] Due to limitations in training data, the model may not generalize well to unseen fault scenarios, resulting in reduced prediction accuracy. This is particularly true in the field of ship equipment operation and maintenance, where factors such as equipment aging, wear, and environmental changes can cause unforeseen changes in equipment status that may not be present in the training data, thus reducing the model's generalization ability.

[0004] When the model's predictive accuracy decreases, it can lead to incorrect diagnostic results, which directly impacts maintenance planning. Incorrect diagnoses can lead to unnecessary maintenance activities, wasting time and resources. They can also miss actual faults, exacerbating equipment failures and even causing accidents, impacting operational decisions. As equipment status changes over time, the original training data may no longer be applicable. Summary of the Invention

[0005] In order to solve the above technical problems, the present invention provides an operation and maintenance auxiliary decision-making method and system for ship equipment to achieve iterative updating of fault diagnosis models and improve operation and maintenance auxiliary decision-making of ship equipment.

[0006] The present invention provides an auxiliary decision-making method for operation and maintenance of ship equipment, comprising:

[0007] Step S1, collecting historical status data of ship equipment;

[0008] Step S2, preprocessing the historical state data to obtain preprocessed historical state data;

[0009] Step S3, based on the pre-processed historical state data, using the pre-trained fault diagnosis model to perform fault diagnosis, output the fault type diagnosis result, and generate a diagnosis report;

[0010] Step S4, formulate and execute a maintenance plan based on the diagnostic report, and record maintenance data during the maintenance process;

[0011] In step S5, the fault diagnosis model is iteratively updated according to the maintenance data to obtain an optimal fault diagnosis model, and the fault is re-diagnosed using the optimal fault diagnosis model to improve the auxiliary decision-making of operation and maintenance of ship equipment.

[0012] Furthermore, the training set of the pre-trained fault diagnosis model is: historical status data with fault type labels.

[0013] Furthermore, the fault diagnosis model is iteratively updated according to the maintenance data to obtain the optimal fault diagnosis model, which specifically includes the following steps:

[0014] Step S51, obtaining the fault type output accuracy according to the fault type diagnosis result;

[0015] Step S52: Obtaining the correlation between the maintenance plan and the fault type label based on the fault type diagnosis result and the maintenance plan;

[0016] Step S53, obtaining the contribution of the maintenance data to the fault diagnosis model performance, and sorting the contribution of each maintenance data from high to low, and selecting the top L maintenance data as the effective maintenance data;

[0017] Step S54, defining weight coefficients for the output accuracy of the fault type, the correlation between the maintenance plan and the fault type label, and the maintenance validity data;

[0018] Step S55, calculating a comprehensive score of the fault diagnosis model performance based on the fault type output accuracy, the correlation between the maintenance plan and the fault type label, and the effective maintenance data;

[0019] Step S56 , adding the effective maintenance data to the training set to repeatedly train the fault diagnosis model, and adjusting the weight coefficient of the fault diagnosis model to obtain the optimal fault diagnosis model.

[0020] Furthermore, the comprehensive score is:

[0021]

[0022] Where, represents the comprehensive score of the fault diagnosis model performance at the t-th time point, represents the output accuracy of the fault type at the t-th time point, 、 、 Represent the weight coefficients of the fault diagnosis model performance, specifically, The weight coefficient representing the accuracy of the fault type output, represents the correlation between the maintenance plan and the fault type label at the tth time point, The weight coefficient representing the correlation between the maintenance plan and the fault type label, Represents the contribution of maintenance effective data at the t-th time point, The weight coefficient representing the contribution of maintenance effective data.

[0023] Furthermore, the weight coefficient is adjusted, including the following steps:

[0024] Step S561, obtaining a change in the fault type output accuracy based on the fault type output accuracy after each maintenance plan execution and the fault type output accuracy after the last maintenance plan execution;

[0025] Step S562: Obtain a change in the correlation between the maintenance plan and the fault type label based on the correlation between the maintenance plan and the fault type label after each execution of the maintenance plan and the correlation between the maintenance plan and the fault type label after the last execution of the maintenance plan.

[0026] Step S563, obtaining a contribution degree change of the maintenance effective data based on the maintenance effective data obtained after each execution of the maintenance plan and the maintenance effective data obtained after the last execution of the maintenance plan;

[0027] Step S564, calculating an updated value of the weight coefficient based on the output accuracy change, the correlation change, the contribution change, and the comprehensive score;

[0028] Step S565 , adjusting the weight coefficient of the fault diagnosis model according to the updated value of the weight coefficient.

[0029] Furthermore, the weight coefficient of the fault diagnosis model is adjusted according to the updated value of the weight coefficient, including:

[0030] Define the update step size of the fault diagnosis model;

[0031] According to the weight coefficient, the weight coefficient update value, and the update step size, the weight coefficient of the fault diagnosis model is adjusted to obtain the optimized weight coefficient;

[0032] The optimal fault diagnosis model is obtained according to the optimized weight coefficient.

[0033] The present invention further provides an operation and maintenance decision-making support system for ship equipment, which is used to execute the above-mentioned operation and maintenance decision-making support method for ship equipment, and includes the following modules:

[0034] Historical status data acquisition module: used to collect historical status data of ship equipment;

[0035] Preprocessing module: connected to the historical state data acquisition module, used to preprocess the historical state data and obtain preprocessed historical state data;

[0036] Diagnostic report acquisition module: connected to the preprocessing module, used to perform fault diagnosis based on preprocessed historical status data using a pretrained fault diagnosis model, output fault type diagnosis results, and generate a diagnostic report;

[0037] Maintenance plan execution module: connected to the diagnosis report acquisition module, used to formulate and execute maintenance plans based on the diagnosis report and record maintenance data during the maintenance process;

[0038] Optimal fault diagnosis model: connected to the maintenance plan execution module, used to iteratively update the fault diagnosis model according to maintenance data, obtain the optimal fault diagnosis model, and use the optimal fault diagnosis model to re-diagnose the fault.

[0039] The embodiments of the present invention have the following technical effects:

[0040] 1. This invention uses effective maintenance data to iteratively update the fault diagnosis model to ensure that the model can adapt to changes in equipment status. By updating the fault diagnosis model, we can improve prediction accuracy, thereby better supporting operation and maintenance decisions, reducing equipment downtime, and improving equipment availability and safety.

[0041] 2. The present invention calculates a comprehensive score of the fault diagnosis model performance based on the output accuracy of the fault type, the correlation between the maintenance plan and the fault type label, and the effective maintenance data, and iteratively updates the model weight coefficient based on the comprehensive score to obtain the optimal fault diagnosis model, so that the model can continuously learn new maintenance data and adapt to changes in equipment status, thereby improving the generalization ability of the model and enabling it to maintain a high level of diagnostic accuracy when facing unprecedented fault scenarios, taking preventive maintenance measures in a timely manner, reducing the probability of equipment failure, extending the service life of the equipment, and improving the availability and safety of the equipment.

[0042] 3. The fault diagnosis model provided by the present invention can provide operation and maintenance personnel with more accurate and reliable fault diagnosis results, and provide strong data support for the formulation of maintenance plans. By continuously optimizing the performance of the model, the present invention enhances the intelligence of decision support, enabling operation and maintenance personnel to make decisions more confidently. BRIEF DESCRIPTION OF THE DRAWINGS

[0043] In order to more clearly illustrate the specific embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the specific embodiments or the description of the prior art. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0044] Figure 1 This is a flow chart of an auxiliary decision-making method for operation and maintenance of ship equipment provided by an embodiment of the present invention;

[0045] Figure 2 This is a structural diagram of an auxiliary decision-making method for operation and maintenance of ship equipment provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0046] To make the objectives, technical solutions, and advantages of the present invention more clear, the technical solutions of the present invention are described clearly and completely below. Obviously, the embodiments described are only some of the embodiments of the present invention, not all of them. All other embodiments derived by persons of ordinary skill in the art based on the embodiments of the present invention without inventive effort are also within the scope of protection of the present invention.

[0047] Figure 1 This is a flow chart of a method for auxiliary decision-making for operation and maintenance of marine equipment provided by an embodiment of the present invention. Figure 1 , specifically including:

[0048] Step S1: collecting historical status data of ship equipment.

[0049] Historical status data includes but is not limited to: equipment operating parameters: temperature, pressure, vibration frequency, etc.; operation log: equipment start and stop time, operator operation records, etc.; alarm record: all warnings and error codes issued by the equipment; maintenance record: historical maintenance activity records, including information on replaced parts, maintenance time, etc.; environmental data: seawater temperature, wind speed, and other external conditions that affect equipment operation.

[0050] Step S2: pre-process the historical state data to obtain pre-processed historical state data.

[0051] Preprocessing includes: data cleaning: removing invalid or missing data points; outlier detection: identifying and processing abnormal data points; data standardization / normalization: converting data in different ranges to the same range to facilitate model processing.

[0052] Step S3, based on the pre-processed historical state data, using the pre-trained fault diagnosis model to perform fault diagnosis, output the fault type diagnosis result, and generate a diagnosis report;

[0053] Pretrained fault diagnosis models can include random forest models, support vector machine models, neural network models, and so on. The training set for a pretrained fault diagnosis model consists of historical status data labeled with fault types. The selected model is trained using the training set, and hyperparameters are adjusted through cross-validation to optimize model performance, resulting in a pretrained fault diagnosis model.

[0054] The system obtains status data from ship equipment in real time, inputs it into a pre-trained fault diagnosis model, and outputs a diagnosis report that includes: fault type, diagnosis results, maintenance recommendations, etc.

[0055] Step S4: formulate and execute a maintenance plan based on the diagnosis report, and record maintenance data of the maintenance process.

[0056] Develop a maintenance plan based on the maintenance recommendations in the diagnostic report and perform actual maintenance work according to the maintenance plan.

[0057] Maintenance data includes: maintenance records: maintenance date (used to analyze the frequency and pattern of failures), maintenance personnel, and maintenance time (reflecting the severity of the failure and the complexity of the repair); parts replacement records: part name / model (used to analyze the failure rate and reliability of specific parts), supplier information (used to evaluate the supplier's quality and service level), part status (used to understand the degree of wear and service life of parts); working hour records: technician's working hours (reflecting the technician's skill level and work efficiency), and the total working hours of the entire maintenance process (used to evaluate the efficiency and cost of the maintenance process).

[0058] In step S5, the fault diagnosis model is iteratively updated according to the maintenance data to obtain an optimal fault diagnosis model, and the fault is re-diagnosed using the optimal fault diagnosis model to improve the auxiliary decision-making of operation and maintenance of ship equipment.

[0059] The fault diagnosis model is iteratively updated according to the maintenance data to obtain the optimal fault diagnosis model, which specifically includes the following steps:

[0060] Step S51: Obtaining the fault type output accuracy according to the fault type diagnosis result.

[0061] The prediction results of the pre-trained fault diagnosis model for the fault types of ship equipment are collected, and the prediction results are compared with the fault type labels that have occurred in the past. The number of correctly predicted fault types is counted and divided by the total number of predictions to obtain the fault type output accuracy.

[0062] Step S52: Obtain the correlation between the maintenance plan and the fault type label according to the fault type diagnosis result and the maintenance plan.

[0063] Using a pre-trained fault diagnosis model, the maintenance effectiveness of repaired ship equipment is tracked. Specifically, real-time status data is used to re-predict the status of the ship equipment to determine the success of the maintenance plan execution. Correlations are calculated based on the maintenance plan execution results and the fault type labels. If the prediction accuracy of the re-prediction improves compared to the pre-repair prediction accuracy, the correlation between the maintenance plan execution results and the fault type labels is considered strong, indicating successful maintenance plan execution. If the prediction accuracy of the re-prediction results does not improve but instead decreases compared to the pre-repair prediction accuracy, the correlation between the maintenance plan execution results and the fault type labels is considered weak, indicating a failure.

[0064] Step S53, obtaining the contribution of the maintenance data to the fault diagnosis model performance, and sorting the contribution of each maintenance data from high to low, and selecting the top L maintenance data as the effective maintenance data;

[0065] The contribution of the maintenance data to the performance of the fault diagnosis model is determined. Specifically, a contribution index is defined to evaluate the impact of the maintenance data on the performance of the pre-trained fault diagnosis model. In this embodiment, the contribution index is preferably the change in fault type output accuracy before and after maintenance. For each maintenance data item, the contribution index to the performance of the pre-trained fault diagnosis model is calculated. The contribution of each maintenance data item is then ranked from highest to lowest, and the top L maintenance data items are selected as the effective maintenance data.

[0066] The contribution of each maintenance data is the change in the accuracy of the fault type output before and after the maintenance. The greater the change in the accuracy of the fault type output before and after the maintenance, the greater the effect of the maintenance data on improving the model performance after the maintenance. Therefore, it is determined as effective maintenance data.

[0067] Step S54 , defining weight coefficients for the output accuracy of the fault type, the correlation between the maintenance plan and the fault type label, and the maintenance validity data.

[0068] Determine the weight coefficient based on business needs, experimental results, expert opinions, etc.

[0069] Step S55, calculating a comprehensive score of the fault diagnosis model performance based on the fault type output accuracy, the correlation between the maintenance plan and the fault type label, and the effective maintenance data;

[0070] The composite score is:

[0071]

[0072] Where, represents the comprehensive score of the fault diagnosis model performance at the t-th time point, represents the output accuracy of the fault type at the t-th time point, The weight coefficient representing the performance of the fault diagnosis model, represents the correlation between the maintenance plan and the fault type label at the tth time point, The weight coefficient representing the correlation between the maintenance plan and the fault type label, Represents the contribution of maintenance effective data at the t-th time point, The weight coefficient representing the contribution of maintenance effective data.

[0073] Step S56: Add the effective maintenance data to the training set to repeatedly train the fault diagnosis model, and adjust the weight coefficient to obtain the optimal fault diagnosis model.

[0074] Adjusting the weight coefficient includes the following steps:

[0075] Step S561 : obtaining a change in the output accuracy of the fault type according to the output accuracy of the fault type after each execution of the maintenance plan and the output accuracy of the fault type after the last execution of the maintenance plan.

[0076] The change in output accuracy of fault type is:

[0077]

[0078] in, represents the output accuracy change of the fault type, n represents the maximum number of repairs, represents the output accuracy of the fault type at time point t of the fault diagnosis model after the i-th maintenance, Represents the fault type output accuracy of the fault diagnosis model at time point t after the i-1th (last) maintenance.

[0079] Step S562 , obtaining a change in the correlation between the maintenance plan and the fault type label based on the correlation between the maintenance plan and the fault type label after each execution of the maintenance plan and the correlation between the maintenance plan and the fault type label after the last execution of the maintenance plan.

[0080] Change in the correlation between maintenance plan and fault type label:

[0081]

[0082] in, Represents the change in the correlation between the maintenance plan and the fault type label, represents the correlation between the maintenance plan and the fault type label at the tth time point of the fault diagnosis model after the i-th maintenance, Represents the correlation between the maintenance plan and the fault type label at time point t in the fault diagnosis model after the i-1th (last) maintenance.

[0083] Step S563 , obtaining a contribution degree change of the maintenance effective data based on the maintenance effective data obtained after each execution of the maintenance plan and the maintenance effective data obtained after the last execution of the maintenance plan.

[0084] Change in contribution of maintenance effective data:

[0085]

[0086] in, Represents the contribution degree change of maintenance effective data, represents the change in the contribution of the fault diagnosis model to the effective maintenance data at time point t after the i-th maintenance, Represents the change in the contribution of the effective maintenance data to the fault diagnosis model at time point t after the i-1th (last) maintenance.

[0087] Step S564, calculating the updated value of the weight coefficient according to the change in output accuracy, the change in relevance, the change in contribution degree, and the comprehensive score.

[0088] First, the updated value of the weight coefficient is calculated using the output accuracy change, relevance change, contribution change, and comprehensive score:

[0089]

[0090]

[0091]

[0092] Where, The updated value of the weight coefficient representing the accuracy of the fault type output; The updated value of the weight coefficient representing the correlation between the maintenance plan and the fault type label, Represents the updated value of the weight coefficient of maintenance effective data.

[0093] Step S565 , adjusting the weight coefficient of the fault diagnosis model according to the updated value of the weight coefficient.

[0094] The weight coefficient of the fault diagnosis model is adjusted according to the updated value of the weight coefficient, including:

[0095] Define the update step size of the fault diagnosis model;

[0096] According to the weight coefficient, the weight coefficient update value, and the update step size, the weight coefficient of the fault diagnosis model is adjusted to obtain the optimized weight coefficient;

[0097]

[0098]

[0099]

[0100] Where, The optimized weight coefficient representing the accuracy of the fault type output, The optimization weight coefficient representing the correlation between the maintenance plan and the fault type label, The optimization weight coefficient representing the contribution of maintenance effective data, Represents the update step size of the fault diagnosis model.

[0101] The optimal fault diagnosis model is obtained according to the optimized weight coefficient.

[0102] Apply the optimized weight coefficient to the fault diagnosis model, that is, use 、 、 Replace the corresponding weight coefficients in the fault diagnosis model. Retrain the fault diagnosis model using the optimized weight coefficients to ensure that the fault diagnosis model can adapt to the new weights. After multiple iterations and optimizations, until the model performance no longer significantly improves, the model has reached optimal performance and is considered the optimal fault diagnosis model.

[0103] The optimal fault diagnosis model is used to re-diagnose ship equipment to improve the accuracy of diagnosis.

[0104] The optimal fault diagnosis model can more accurately identify the fault type of ship equipment and obtain more accurate diagnostic results, thereby providing more reliable information for maintenance plans, formulating more targeted maintenance plans, reducing unnecessary maintenance activities, and reducing the downtime of equipment due to failures by identifying potential faults early, thereby improving auxiliary decision-making.

[0105] Figure 2 This is a structural diagram of an auxiliary decision-making method for operation and maintenance of marine equipment provided by an embodiment of the present invention, referring to Figure 2 This embodiment further provides an operation and maintenance decision-making support system for ship equipment, which is used to execute the above-mentioned operation and maintenance decision-making support method for ship equipment, and includes the following modules:

[0106] Historical status data acquisition module: used to collect historical status data of ship equipment;

[0107] Preprocessing module: connected to the historical state data acquisition module, used to preprocess the historical state data and obtain preprocessed historical state data;

[0108] Diagnostic report acquisition module: connected to the preprocessing module, used to perform fault diagnosis based on preprocessed historical status data using a pretrained fault diagnosis model, output fault type diagnosis results, and generate a diagnostic report;

[0109] Maintenance plan execution module: connected to the diagnosis report acquisition module, used to formulate and execute maintenance plans based on the diagnosis report, and record maintenance data during the maintenance process;

[0110] Optimal fault diagnosis model: connected to the maintenance plan execution module, used to iteratively update the fault diagnosis model according to maintenance data, obtain the optimal fault diagnosis model, and use the optimal fault diagnosis model to re-diagnose the fault.

[0111] This embodiment further provides an electronic device, which includes one or more processors and a memory.

[0112] The processor may be a central processing unit (CPU) or other forms of processing units having data processing capabilities and / or instruction execution capabilities, and may control other components in the electronic device to perform desired functions.

[0113] The memory may include one or more computer program products, which may include various forms of computer-readable storage media, such as volatile memory and / or non-volatile memory. The volatile memory may include, for example, random access memory (RAM) and / or cache memory. The non-volatile memory may include, for example, read-only memory (ROM), a hard disk, flash memory, etc. One or more computer program instructions may be stored on the computer-readable storage medium, and the processor may execute the program instructions to implement the operation and maintenance decision-making support method for ship equipment according to any embodiment of the present application described above and / or other desired functions. Various contents such as initial external parameters and threshold values ​​may also be stored in the computer-readable storage medium.

[0114] In one example, the electronic device may further include an input device and an output device, these components interconnected via a bus system and / or other connection mechanisms (not shown). The input device may include, for example, a keyboard, a mouse, etc. The output device may output various information to the outside, including warning information, braking force, etc. The output device may include, for example, a display, a speaker, a printer, a communication network, and remote output devices connected thereto.

[0115] In addition, the electronic device may further include any other appropriate components depending on specific applications.

[0116] In addition to the above-mentioned methods and devices, an embodiment of the present application may also be a computer program product, which includes computer program instructions, which, when executed by a processor, enable the processor to execute the steps of a method for auxiliary decision-making for operation and maintenance of ship equipment provided in any embodiment of the present application.

[0117] The computer program product may be written in any combination of one or more programming languages ​​to implement the program code for performing the operations of the embodiments of the present application, including object-oriented programming languages ​​such as Java, C++, and conventional procedural programming languages ​​such as "C" or similar programming languages. The program code may be executed entirely on the user's computing device, partially on the user's computing device, as a standalone software package, partially on the user's computing device and partially on a remote computing device, or entirely on a remote computing device or server.

[0118] In addition, an embodiment of the present application may also be a computer-readable storage medium on which computer program instructions are stored. When the computer program instructions are executed by a processor, the processor executes the steps of a method for auxiliary decision-making for operation and maintenance of ship equipment provided in any embodiment of the present application.

[0119] The computer-readable storage medium may be any combination of one or more readable media. The readable medium may be a readable signal medium or a readable storage medium. The readable storage medium may include, for example, but is not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, device, or device, or any combination thereof. More specific examples (a non-exhaustive list) of readable storage media include: an electrical connection with one or more wires, a portable disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination thereof.

[0120] It should be noted that the terms used in the present invention are only for describing specific embodiments and are not intended to limit the scope of this application. As shown in the present specification, unless the context clearly indicates an exception, the words "one", "a", "a kind of" and / or "the" do not specifically refer to the singular and may also include the plural. The terms "comprise", "include" or any other variants thereof are intended to cover non-exclusive inclusion, so that the process, method or device comprising a series of elements includes not only those elements, but also includes other elements not explicitly listed, or also includes elements inherent to such process, method or device. In the absence of further restrictions, the elements defined by the sentence "comprise a..." do not exclude the presence of other identical elements in the process, method or device comprising the elements.

[0121] It should also be noted that the terms "center", "up", "down", "left", "right", "vertical", "horizontal", "inside", "outside", etc., indicating orientations or positional relationships, are based on the orientations or positional relationships shown in the accompanying drawings. They are only for the convenience of describing the present invention and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation. Therefore, they cannot be understood as limitations on the present invention. Unless otherwise clearly specified and limited, the terms "installed", "connected", "connected", etc. should be understood in a broad sense. For example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be a direct connection, or an indirect connection through an intermediate medium, or it can be a communication between the internal parts of two elements. For those of ordinary skill in the art, the specific meanings of the above terms in the present invention can be understood according to specific circumstances.

[0122] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the above embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the above embodiments, or replace some or all of the technical features therein with equivalents. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the technical solutions of the embodiments of the present invention.

Claims

1. A method for auxiliary decision making for operation and maintenance of ship equipment, characterized in that: include: Step S1, collecting historical status data of ship equipment; Step S2, preprocessing the historical state data to obtain preprocessed historical state data; Step S3, based on the pre-processed historical state data, using the pre-trained fault diagnosis model to perform fault diagnosis, output the fault type diagnosis result, and generate a diagnosis report; Step S4, formulate and execute a maintenance plan based on the diagnostic report, and record maintenance data of the maintenance process; Step S5, iteratively updating the fault diagnosis model according to the maintenance data to obtain the optimal fault diagnosis model, and re-diagnosing the fault using the optimal fault diagnosis model to improve the auxiliary decision-making of operation and maintenance of ship equipment; Step S51, obtaining the fault type output accuracy according to the fault type diagnosis result; Step S52, obtaining the correlation between the maintenance plan and the fault type label according to the fault type diagnosis result and the maintenance plan; Step S53, obtaining the contribution of the maintenance data to the performance of the fault diagnosis model, and arranging the contribution of each maintenance data from high to low, and selecting the first L maintenance data as effective maintenance data; Step S54, defining weight coefficients for the output accuracy of the fault type, the correlation between the maintenance plan and the fault type label, and the maintenance validity data; Step S55, calculating the comprehensive score of the fault diagnosis model performance according to the output accuracy of the fault type, the correlation between the maintenance plan and the fault type label, and the effective maintenance data; ； In the formula, represents the comprehensive score of the fault diagnosis model performance at the tth time point, represents the output accuracy of the fault type at the tth time point, , , Represent the weight coefficients of the fault diagnosis model performance, The weight coefficient representing the accuracy of the fault type output, represents the correlation between the maintenance plan and the fault type label at the tth time point, The weight coefficient representing the correlation between the maintenance plan and the fault type label, Represents the contribution of effective maintenance data at the tth time point, The weight coefficient representing the contribution of maintenance effective data; Step S56, adding the effective maintenance data to the training set to repeatedly train the fault diagnosis model, and adjusting the weight coefficient of the fault diagnosis model to obtain the optimal fault diagnosis model; Step S561, according to the output accuracy of the fault type after each execution of the maintenance plan and the output accuracy of the fault type after the last execution of the maintenance plan, obtain the change in the output accuracy of the fault type: ； represents the change in output accuracy of the fault type, n represents the maximum number of repairs, represents the output accuracy of the fault type of the fault diagnosis model at the tth time point after the i-th maintenance, represents the output accuracy of the fault type at the tth time point of the fault diagnosis model after the i-1th maintenance; Step S562, according to the correlation between the maintenance plan and the fault type label after each execution of the maintenance plan, and the correlation between the maintenance plan and the fault type label after the last execution of the maintenance plan, obtain the correlation change between the maintenance plan and the fault type label: ； Represents the change in the correlation between the maintenance plan and the fault type label, represents the correlation between the maintenance plan and the fault type label of the fault diagnosis model after the i-th maintenance at the t-th time point, represents the correlation between the maintenance plan and the fault type label of the fault diagnosis model after the i-1th maintenance at the tth time point; Step S563, according to the maintenance effective data obtained after each execution of the maintenance plan and the maintenance effective data obtained after the last execution of the maintenance plan, the contribution degree change of the maintenance effective data is obtained: ； Represents the contribution degree change of maintenance effective data, represents the change in contribution of the fault diagnosis model to the effective maintenance data at time point t after the i-th maintenance, Represents the change in contribution of the fault diagnosis model to the effective maintenance data at time point t after the i-1th maintenance; Step S564, calculate the updated value of the weight coefficient according to the output accuracy change, relevance change, contribution change, and comprehensive score: ； ； ； An updated value of the weight coefficient representing the accuracy of the fault type output; The updated value of the weight coefficient representing the correlation between the maintenance plan and the fault type label, Represents the updated value of the weight coefficient of the maintenance effective data; Step S565, adjusting the weight coefficient of the fault diagnosis model according to the updated value of the weight coefficient, including: Define the update step size of the fault diagnosis model; According to the weight coefficient, weight coefficient update value, and update step size, the weight coefficient of the fault diagnosis model is adjusted to obtain the optimized weight coefficient: ； ； ； The optimized weight coefficient representing the accuracy of the fault type output, The optimization weight coefficient representing the correlation between the maintenance plan and the fault type label, The optimization weight coefficient representing the contribution of effective maintenance data, Represents the update step size of the fault diagnosis model; The optimal fault diagnosis model is obtained according to the optimized weight coefficient.

2. The operation and maintenance auxiliary decision-making method of ship equipment according to claim 1 is characterized in that: The training set of the pre-trained fault diagnosis model is: historical status data with fault type labels.

3. An operation and maintenance auxiliary decision system for ship equipment, used to execute an operation and maintenance auxiliary decision method for ship equipment according to any one of claims 1 to 2, characterized in that: Includes the following modules: Historical status data collection module: used to collect historical status data of ship equipment; Preprocessing module: connected with the historical state data acquisition module, used for preprocessing the historical state data to obtain preprocessed historical state data; Diagnostic report acquisition module: connected to the preprocessing module, used to perform fault diagnosis based on preprocessed historical status data using a pretrained fault diagnosis model, output fault type diagnosis results, and generate a diagnostic report; Maintenance plan execution module: connected with the diagnosis report acquisition module, used to formulate and execute maintenance plans according to the diagnosis report, and record maintenance data of the maintenance process; Optimal fault diagnosis model: connected to the maintenance plan execution module, used to iteratively update the fault diagnosis model according to the maintenance data, obtain the optimal fault diagnosis model, and re-diagnose the fault using the optimal fault diagnosis model.

Images