Method and system for failure prediction and prevention in intelligent device retrofitting technology

Abstract

This invention relates to the field of industrial intelligence, and proposes a method and system for fault prediction and prevention in intelligent equipment transformation technology. The method includes: acquiring real-time data from the intelligent equipment; filtering and selecting abnormal entropy values ​​and extracting abnormal parameter sets; constructing a fault prediction framework based on this data to predict fault probability, determine remaining service life, and calculate equipment maintenance cycles; evaluating the operating status to obtain status evaluation parameters; querying fault types and analyzing occurrence times; constructing preventive maintenance strategies based on the current status; generating a maintenance task list and querying maintenance tasks; analyzing prevention objectives in maintenance tasks; identifying prevention obstacles; monitoring operational feedback data based on this; and finally generating an operational report. This invention can improve the operational stability of intelligent equipment.

Description

Technical Field

[0001] This invention relates to the field of industrial intelligence, and in particular to fault prediction and prevention methods and systems in intelligent equipment transformation technology. Background Technology

[0002] In today's industrial sector, the application of intelligent equipment is becoming increasingly widespread, involving many industries such as manufacturing, energy, and transportation. In the transformation technology of intelligent equipment, effective fault prediction and prevention methods are crucial, which are of great significance for ensuring the reliable operation of equipment and extending its service life.

[0003] Currently, traditional fault detection for intelligent equipment mainly relies on regular maintenance and post-incident repair. This approach not only consumes a lot of time and manpower but also often fails to detect potential faults in a timely manner, leading to increased equipment downtime and impacting production progress. Furthermore, conventional fault detection methods are insufficient to meet the increasingly complex structural and functional requirements of intelligent equipment, making it impossible to accurately predict and prevent faults in new intelligent equipment. Therefore, a fault prediction and prevention method is needed in intelligent equipment transformation technology to improve the operational stability of intelligent equipment. Summary of the Invention

[0004] This invention provides a method and system for fault prediction and prevention in intelligent equipment transformation technology, the main purpose of which is to improve the operational stability of intelligent equipment.

[0005] Real-time data during the operation of intelligent equipment is acquired, the real-time data is filtered to obtain filtered real-time data, the abnormal entropy parameters corresponding to the filtered real-time data are selected, and the abnormal entropy parameters are feature extracted to obtain an abnormal parameter set.

[0006] Based on the abnormal parameter set, a fault prediction framework corresponding to the intelligent device is constructed. The fault prediction framework is used to predict the fault probability of the intelligent device. Based on the fault probability, the remaining operating life of the intelligent device is determined. Based on the remaining operating life, the equipment maintenance cycle of the intelligent device is calculated. The calculation of the equipment maintenance cycle based on the remaining operating life includes:

[0007] The equipment maintenance cycle corresponding to the intelligent device is calculated using the following formula:

[0008]

[0009] Wherein, WT represents the equipment operation and maintenance cycle corresponding to the intelligent device, k represents the adjustment coefficient corresponding to the intelligent device, L represents the remaining operating life, t represents the time variable, f(t) represents the weight coefficient of the time variable t, R(t) represents the reliability coefficient at time t, and dt represents the small increment of the time variable t.

[0010] Based on the equipment operation and maintenance cycle, the operating status of the intelligent equipment is evaluated to obtain status evaluation parameters. Based on the status evaluation parameters, the fault types of the intelligent equipment are queried, and the occurrence time of the fault types is analyzed.

[0011] Based on the fault type and the occurrence time, and combined with the current state of the intelligent device, a preventive maintenance strategy corresponding to the intelligent device is constructed. Based on the preventive maintenance strategy, a maintenance task list corresponding to the intelligent device is generated, and maintenance tasks in the maintenance task list are queried.

[0012] The preventive objectives in the maintenance task are analyzed, the preventive obstacles in the preventive objectives are identified, the operational feedback data of the intelligent equipment is monitored based on the preventive obstacles, and the corresponding operational report of the intelligent equipment is generated based on the operational feedback data.

[0013] Optionally, the abnormal entropy parameter corresponding to the filtered real-time data includes:

[0014] Generate the entropy sequence corresponding to the filtered real-time data;

[0015] Based on the entropy sequence, calculate the average entropy value corresponding to the filtered real-time data;

[0016] Based on the average entropy value, the entropy fluctuation range of the filtered real-time data is determined;

[0017] Based on the entropy fluctuation range, the abnormal entropy parameters corresponding to the filtered real-time data are selected.

[0018] Optionally, calculating the average entropy value corresponding to the filtered real-time data based on the entropy value sequence includes:

[0019] The average entropy value corresponding to the filtered real-time data is calculated using the following formula:

[0020]

[0021] in, H represents the average entropy value corresponding to the filtered real-time data, N represents the total number of entropy values ​​in the entropy value sequence, i represents the index of the number of entropy values ​​in the entropy value sequence, and H represents the average entropy value. i This represents the i-th entropy value in the entropy value sequence.

[0022] Optionally, constructing the fault prediction framework corresponding to the intelligent device based on the abnormal parameter set includes:

[0023] Analyze the parameter distribution patterns corresponding to the abnormal parameter set;

[0024] Based on the parameter distribution pattern, the importance of the parameters in the abnormal parameter set is determined;

[0025] Based on the degree of importance, key abnormal parameters corresponding to the abnormal parameter set are selected;

[0026] Based on the key anomaly parameters, a fault feature vector corresponding to the intelligent device is generated;

[0027] Based on the fault feature vector, the fault classification standard corresponding to the intelligent device is determined;

[0028] Based on the fault classification criteria, a fault prediction framework corresponding to the intelligent device is constructed.

[0029] Optionally, determining the remaining operational lifespan of the intelligent device based on the failure probability includes:

[0030] Based on the failure probability, obtain the historical operating data corresponding to the intelligent device;

[0031] Analyze the performance degradation trend in the historical operating data;

[0032] Based on the aforementioned performance degradation trend, predict the time span from performance degradation to failure of the intelligent device;

[0033] Analyze the uncertainties corresponding to the time span;

[0034] Based on the aforementioned uncertainties, the remaining operational lifespan of the intelligent device is determined.

[0035] Optionally, the step of evaluating the operating status of the intelligent device based on the device operation and maintenance cycle to obtain status evaluation parameters includes:

[0036] Based on the equipment maintenance cycle, query the original operating data corresponding to the intelligent equipment;

[0037] Remove invalid data from the original running data to obtain valid running data;

[0038] Analyze the degree of deviation between the effective operating data and the historical normal operating data corresponding to the intelligent device;

[0039] Based on the degree of deviation, mark the abnormal data segments in the valid operating data;

[0040] Based on the abnormal data segment, calculate the performance degradation value corresponding to the intelligent device;

[0041] Based on the magnitude of performance degradation, the operating status of the intelligent device is assessed to obtain status assessment parameters.

[0042] Optionally, calculating the performance degradation magnitude of the intelligent device based on the abnormal data segment includes:

[0043] The performance degradation of the intelligent device is calculated using the following formula:

[0044]

[0045] Where XF represents the performance degradation of the intelligent device, m represents the number of data points in the abnormal data segment, j represents the index of the data point in the abnormal data segment, and P nj P represents the actual value of the performance metric corresponding to the j-th data point in the abnormal data segment. hj This represents the expected value of the performance index corresponding to the j-th data point under normal operating conditions.

[0046] Optionally, querying the fault type of the intelligent device based on the state assessment parameters includes:

[0047] Based on the aforementioned status assessment parameters, analyze the operational fluctuations of the intelligent equipment.

[0048] Extract the operational fluctuation characteristics corresponding to the aforementioned operational fluctuation conditions;

[0049] Query the fluctuation indicators that show faults in the operational fluctuation characteristics;

[0050] The fluctuation index is matched with a preset fault index library to obtain matching data;

[0051] Filter the subset of fault indicators with high similarity from the matched data;

[0052] Based on the subset of fault indicators, query the types of faults existing in the intelligent device.

[0053] Optionally, the step of constructing a preventive maintenance strategy for the intelligent device based on the fault type and the time of occurrence, combined with the current state of the intelligent device, includes:

[0054] Based on the fault type and occurrence time, analyze the fault development trend corresponding to the intelligent device;

[0055] Extract the key change nodes in the fault development trend;

[0056] Query the node operation parameters corresponding to the key change nodes;

[0057] The node's operating parameters are compared with normal standard parameters to obtain the difference parameters;

[0058] Analyze the difference maintenance items corresponding to the difference parameters;

[0059] Based on the aforementioned differential maintenance items and the current state of the intelligent device, a preventive maintenance strategy corresponding to the intelligent device is constructed.

[0060] Optionally, to address the above problems, the present invention provides a fault prediction and prevention system in intelligent equipment transformation technology, the system comprising:

[0061] The feature extraction module is used to acquire real-time data during the operation of intelligent equipment, filter the real-time data to obtain filtered real-time data, filter the abnormal entropy parameters corresponding to the filtered real-time data, and extract features from the abnormal entropy parameters to obtain an abnormal parameter set.

[0062] The operation and maintenance cycle calculation module is used to construct a fault prediction framework corresponding to the intelligent device based on the abnormal parameter set, predict the fault probability corresponding to the intelligent device using the fault prediction framework, determine the remaining operating life of the intelligent device based on the fault probability, and calculate the equipment operation and maintenance cycle corresponding to the intelligent device based on the remaining operating life. The step of calculating the equipment operation and maintenance cycle corresponding to the intelligent device based on the remaining operating life includes:

[0063] The equipment maintenance cycle corresponding to the intelligent device is calculated using the following formula:

[0064]

[0065] Wherein, WT represents the equipment operation and maintenance cycle corresponding to the intelligent device, k represents the adjustment coefficient corresponding to the intelligent device, L represents the remaining operating life, t represents the time variable, f(t) represents the weight coefficient of the time variable t, R(t) represents the reliability coefficient at time t, and dt represents the small increment of the time variable t.

[0066] The time analysis module is used to evaluate the operating status of the intelligent equipment based on the equipment operation and maintenance cycle, obtain status evaluation parameters, query the fault types of the intelligent equipment based on the status evaluation parameters, and analyze the occurrence time of the fault types.

[0067] The task query module is used to construct a preventive maintenance strategy for the intelligent device based on the fault type and the occurrence time, combined with the current status of the intelligent device, generate a maintenance task list for the intelligent device based on the preventive maintenance strategy, and query the maintenance tasks in the maintenance task list.

[0068] The report generation module is used to analyze the prevention objectives in the maintenance task, identify the prevention obstacles in the prevention objectives, monitor the operation feedback data of the intelligent equipment based on the prevention obstacles, and generate the corresponding operation report of the intelligent equipment based on the operation feedback data.

[0069] First, this invention, by acquiring real-time data during the operation of intelligent equipment, reflects the current operating status of the equipment, including changes in various parameters. This helps to promptly detect anomalies and allow for proactive measures to prevent malfunctions. Second, based on the set of abnormal parameters, this invention constructs a fault prediction framework for the intelligent equipment. This framework fully utilizes the information provided by the abnormal parameters to conduct a comprehensive and in-depth analysis of the intelligent equipment's operating status, thereby more accurately predicting potential faults. Third, based on the equipment's maintenance cycle, this invention assesses the operating status of the intelligent equipment to obtain status assessment parameters. This helps to promptly identify potential problems and intervene with targeted measures before the problems escalate, effectively preventing equipment failures and ensuring stable equipment operation. This invention, based on the fault type and occurrence time, combined with the current state of the intelligent equipment, constructs a corresponding preventive maintenance strategy for the intelligent equipment. This allows for a comprehensive and in-depth understanding of the equipment's operational evolution, clearly grasping the entire process from normal operation to fault occurrence. Based on the fault type, occurrence time, and current state, maintenance personnel can be rationally assigned, and necessary parts and tools prepared, thereby significantly improving maintenance efficiency and reducing equipment downtime. Furthermore, by analyzing the preventive objectives in the maintenance tasks, this invention can accurately identify the key points and critical aspects of maintenance work, rationally allocate resources, avoid resource waste and misallocation, and help predict future fault modes and performance degradation trends, enabling advance preparation and timely identification, adjustment, and improvement of maintenance strategies. Therefore, the fault prediction and prevention method and system in the intelligent equipment transformation technology proposed in this invention can improve the operational stability of intelligent equipment. Attached Figure Description

[0070] Figure 1 A flowchart illustrating a fault prediction and prevention method in an intelligent equipment retrofitting technology provided by an embodiment of the present invention;

[0071] Figure 2This is a schematic diagram of a fault prediction and prevention system in an intelligent equipment transformation technology provided in an embodiment of the present invention.

[0072] The realization of the objective, functional features and advantages of the present invention will be further explained in conjunction with the embodiments and with reference to the accompanying drawings. Detailed Implementation

[0073] It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention.

[0074] This application provides a fault prediction and prevention method in intelligent equipment transformation technology. The execution subject of the fault prediction and prevention method in intelligent equipment transformation technology includes, but is not limited to, at least one of the following electronic devices that can be configured to execute the method provided in this application embodiment: a server, a terminal, etc. In other words, the fault prediction and prevention method in intelligent equipment transformation technology can be executed by software or hardware installed on a terminal device or a server device, and the software can be a blockchain platform. The server includes, but is not limited to, a single server, a server cluster, a cloud server, or a cloud server cluster. The server can be an independent server or a cloud server that provides basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communication, middleware services, domain name services, security services, content delivery networks (CDN), and big data and artificial intelligence platforms.

[0075] Reference Figure 1 The diagram shown is a flowchart illustrating a fault prediction and prevention method in an intelligent equipment retrofitting technology according to an embodiment of the present invention. In this embodiment, the fault prediction and prevention method in the intelligent equipment retrofitting technology includes:

[0076] S1. Acquire real-time data during the operation of the intelligent device, filter the real-time data to obtain filtered real-time data, filter the abnormal entropy parameters corresponding to the filtered real-time data, extract features from the abnormal entropy values ​​to obtain an abnormal parameter set.

[0077] This invention acquires real-time data during the operation of intelligent equipment, reflecting the current operating status of the equipment, including changes in various parameters. This helps to detect abnormal situations in the equipment in a timely manner, thereby taking measures in advance to avoid failures.

[0078] The intelligent equipment refers to equipment that applies advanced information technology and intelligent control technology and can achieve automated and intelligent operation, such as intelligent robots, automated production lines, and intelligent sensors. The real-time data refers to various instantaneous data generated by the intelligent equipment during operation, including but not limited to the equipment's operating parameters (such as temperature, pressure, speed, current, etc.), working status (such as on, off, operating mode, etc.), fault information (such as fault codes, fault time, etc.), and environmental data (such as humidity, air quality, etc.). Optionally, the acquisition of real-time data during the operation of the intelligent equipment can be achieved through sensor technology, such as temperature sensors, pressure sensors, speed sensors, etc., to collect the equipment's operating parameters in real time.

[0079] Furthermore, by filtering the real-time data, the present invention obtains filtered real-time data, which can effectively remove noise and interference from the real-time data, making the data more accurate and reliable. This helps to accurately judge the operating trend and potential problems of the equipment, reduces the impact of abnormal data on fault prediction and prevention, and improves the accuracy and reliability of fault diagnosis.

[0080] The filtered real-time data refers to the data obtained after filtering the real-time data acquired during the operation of the intelligent device. This is achieved by removing or reducing noise, interference, and abnormal fluctuations in the original real-time data, thereby making the data smoother, more stable, and more reliable. Optionally, the filtering of the real-time data can be achieved by digital filters, such as low-pass filters, high-pass filters, and band-pass filters.

[0081] Furthermore, by filtering the abnormal entropy parameters corresponding to the filtered real-time data, the present invention can help capture abnormal patterns and potential fault signs in the data, promptly detect abnormal situations in equipment operation, provide early warnings of faults, and thus take corresponding measures to avoid production stoppages and economic losses caused by equipment failures.

[0082] The preset threshold refers to an entropy limit set based on experience or specific standards, used to determine whether the entropy value is abnormal; the abnormal entropy refers to the entropy value exceeding the preset threshold, which indicates that there are abnormal situations or potential fault information in the filtered real-time data.

[0083] As an embodiment of the present invention, the step of filtering the abnormal entropy parameters corresponding to the filtered real-time data includes: generating an entropy value sequence corresponding to the filtered real-time data; calculating the average entropy value corresponding to the filtered real-time data based on the entropy value sequence; determining the entropy value fluctuation range of the filtered real-time data based on the average entropy value; and filtering the abnormal entropy parameters corresponding to the filtered real-time data based on the entropy value fluctuation range.

[0084] The entropy sequence refers to the arrangement of a series of entropy values ​​obtained by processing filtered real-time data according to certain rules; the average entropy value refers to the average value of all entropy values ​​in the entropy sequence, which reflects the overall entropy level of the filtered real-time data; the entropy fluctuation range refers to the range of changes in entropy values ​​in the entropy sequence, which can usually be determined by calculating the difference between the maximum and minimum entropy values.

[0085] Furthermore, the generation of the entropy sequence corresponding to the filtered real-time data can be achieved using an information entropy algorithm, such as calculating the entropy value based on the probability of each data point appearing in the filtered real-time data, thereby obtaining the entropy sequence; the calculation of the average entropy value corresponding to the filtered real-time data can be achieved using the following algorithm; the determination of the entropy fluctuation range of the filtered real-time data can be achieved using the NumPy library, such as using the max and min functions in the NumPy library to find the maximum and minimum values ​​of the entropy sequence, thereby determining the entropy fluctuation range; the filtering of abnormal entropy parameters corresponding to the filtered real-time data can be achieved using anomaly detection algorithms, such as statistical anomaly detection algorithms, machine learning-based anomaly detection algorithms, etc.

[0086] As an embodiment of the present invention, the step of calculating the average entropy value corresponding to the filtered real-time data based on the entropy value sequence includes:

[0087] The average entropy value corresponding to the filtered real-time data is calculated using the following formula:

[0088]

[0089] in, H represents the average entropy value corresponding to the filtered real-time data, N represents the total number of entropy values ​​in the entropy value sequence, i represents the index of the number of entropy values ​​in the entropy value sequence, and H represents the average entropy value. i This represents the i-th entropy value in the entropy value sequence.

[0090] In detail, the average entropy value refers to the sum of all entropy values ​​in the entropy value sequence, divided by the total number of entropy values ​​N. It reflects the overall entropy level of the filtered real-time data and is a measure of data uncertainty. The larger the average entropy value, the higher the data uncertainty; the smaller the average entropy value, the lower the data uncertainty. The entropy value refers to the degree of disorder or chaos in the filtered real-time data at the i-th position. The higher the entropy value, the more disordered and chaotic the data at that position, and the less information it contains; the lower the entropy value, the more ordered the data at that position, and the more information it contains.

[0091] S2. Based on the abnormal parameter set, construct a fault prediction framework corresponding to the intelligent device, use the fault prediction framework to predict the fault probability corresponding to the intelligent device, determine the remaining operating life of the intelligent device based on the fault probability, and calculate the equipment operation and maintenance cycle corresponding to the intelligent device based on the remaining operating life.

[0092] Based on the aforementioned abnormal parameter set, this invention constructs a fault prediction framework for the intelligent device, which can fully utilize the information provided by the abnormal parameter set to conduct a comprehensive and in-depth analysis of the operating status of the intelligent device, thereby more accurately predicting potential faults.

[0093] The fault prediction framework refers to a model or system built based on fault feature vectors and fault classification standards for predicting faults in intelligent equipment. It can perform fault prediction and early warning based on real-time data.

[0094] As an embodiment of the present invention, the step of constructing a fault prediction framework for the intelligent device based on the abnormal parameter set includes: analyzing the parameter distribution pattern corresponding to the abnormal parameter set; determining the importance of parameters in the abnormal parameter set based on the parameter distribution pattern; filtering key abnormal parameters corresponding to the abnormal parameter set based on the importance; generating a fault feature vector corresponding to the intelligent device based on the key abnormal parameters; determining a fault classification standard for the intelligent device based on the fault feature vector; and constructing a fault prediction framework for the intelligent device based on the fault classification standard.

[0095] The parameter distribution pattern refers to the distribution of the abnormal parameter set across various dimensions, including the parameter value range, central tendency, and dispersion. The importance refers to the influence of each parameter in the abnormal parameter set on the fault prediction of intelligent equipment, which is usually determined through data analysis and expert experience. The key abnormal parameters are those parameters in the abnormal parameter set that have a significant impact on equipment fault prediction; changes in these parameters directly lead to equipment faults. The fault feature vector is a vector composed of key abnormal parameters, used to describe the fault characteristics of intelligent equipment. The fault classification standard is the basis for classifying faults of intelligent equipment, usually determined based on factors such as the nature and severity of the fault.

[0096] Furthermore, the analysis of the parameter distribution patterns corresponding to the abnormal parameter set can be achieved through data visualization tools, such as Matplotlib and Seaborn; the determination of the importance of parameters in the abnormal parameter set can be achieved through Python tools, such as using the Scikit-learn library in Python to perform principal component analysis to determine the importance of each parameter in the abnormal parameter set; the screening of key abnormal parameters corresponding to the abnormal parameter set can be achieved through machine learning algorithms, such as decision trees and random forests; the generation of fault feature vectors corresponding to the intelligent device can be achieved through feature engineering methods, such as combining and transforming key abnormal parameters to generate fault feature vectors; the determination of fault classification criteria corresponding to the intelligent device can be achieved through clustering analysis algorithms, such as performing cluster analysis on fault data and determining fault classification criteria based on the clustering results; and the construction of the fault prediction framework corresponding to the intelligent device can be achieved through deep learning frameworks, such as TensorFlow and PyTorch models.

[0097] This invention uses the fault prediction framework to predict the fault probability of the intelligent equipment, which can provide early warning of equipment failures, giving relevant personnel enough time to take preventive measures to avoid failures or reduce losses caused by failures. It can also help to reasonably arrange equipment maintenance plans, and timely understanding of equipment failure probabilities helps to optimize production processes, avoid production interruptions due to equipment failures, and reduce maintenance costs.

[0098] The failure probability refers to the likelihood of an intelligent device failing within a specific time period in the future. It is a value calculated by analyzing the device's historical data, real-time operating data, and a failure prediction framework. The failure probability reflects the device's health status and potential risk level; a higher probability indicates a greater likelihood of failure. Optionally, the failure probability of the intelligent device can be predicted using the failure prediction framework through Monte Carlo simulation, such as by randomly selecting various operating parameters of the device, simulating the device's operation multiple times, and counting the number of failures to estimate the failure probability.

[0099] Furthermore, by determining the remaining operating life of the intelligent equipment based on the failure probability, the present invention can more accurately plan the maintenance and replacement schedule of the equipment, avoid resource waste caused by premature replacement, or sudden equipment failure that affects production due to late replacement, and ensure that the equipment can fully perform its function while it is still running stably, thus guaranteeing the continuity of production.

[0100] The remaining operating life refers to the remaining time that the intelligent device is expected to be able to operate normally in its current state.

[0101] As an embodiment of the present invention, determining the remaining operating life of the intelligent device based on the failure probability includes: obtaining historical operating data of the intelligent device based on the failure probability; analyzing the performance degradation trend in the historical operating data; predicting the time span from performance degradation to failure of the intelligent device based on the performance degradation trend; analyzing the uncertainties corresponding to the time span; and determining the remaining operating life of the intelligent device based on the uncertainties.

[0102] The historical operating data refers to relevant information generated and recorded by the intelligent device during its past operation, including the device's runtime, workload, maintenance records, number and type of faults, and operating environment conditions. The performance degradation trend refers to the gradual decline in the performance of the intelligent device during use, which is usually derived by analyzing and summarizing changes in the device's key performance indicators. The time span refers to the estimated duration from the current moment until the intelligent device's performance deteriorates to the point of failure. The uncertainties refer to various variables and unknown factors that can affect the accuracy of the prediction of the remaining operating life of the intelligent device, such as sudden changes in the external environment and potential internal problems of the device that have not been taken into account.

[0103] Furthermore, obtaining the historical operating data corresponding to the intelligent device can be achieved by querying a relational database, such as extracting the historical operating data of the intelligent device from a relational database (e.g., MySQL, Oracle); analyzing the performance degradation trend in the historical operating data can be achieved using the Scikit-learn library, such as using PolynomialFeatures combined with linear regression in the Scikit-learn library to capture non-linear performance degradation trends; predicting the time span from performance degradation to failure of the intelligent device can be achieved using time series prediction models, such as ARIMA, SARIMA, etc.; analyzing the uncertainties corresponding to the time span can be achieved using Python tools, such as using the Sobol sensitivity analysis library in Python to analyze the sensitivity of uncertainties; determining the remaining operating life of the intelligent device can be achieved using R language, such as using the reliability package in R language to determine the remaining operating life of the device.

[0104] Based on the remaining operating life, this invention calculates the equipment maintenance cycle corresponding to the intelligent equipment, which can help enterprises plan the investment of equipment maintenance resources more accurately, avoid waste or shortage of resources, effectively improve the availability and stability of equipment, perform maintenance and repair at the appropriate time, reduce the probability of equipment failure, and ensure the continuity and stability of production.

[0105] The equipment maintenance cycle refers to the time interval required for the maintenance and operation management of intelligent equipment, which reflects the ideal time length between two maintenance operations under normal operating conditions.

[0106] As an embodiment of the present invention, calculating the equipment maintenance cycle corresponding to the intelligent device based on the remaining operating life includes:

[0107] The equipment maintenance cycle corresponding to the intelligent device is calculated using the following formula:

[0108]

[0109] Wherein, WT represents the equipment operation and maintenance cycle corresponding to the intelligent device, k represents the adjustment coefficient corresponding to the intelligent device, L represents the remaining operating life, t represents the time variable, f(t) represents the weight coefficient of the time variable t, R(t) represents the reliability coefficient of the time variable t, and dt represents the small increment of the time variable t.

[0110] In detail, the adjustment coefficient refers to the parameter used to correct and adapt the formula for calculating the operation and maintenance cycle of the equipment; the weighting coefficient refers to the different weight values ​​assigned to the time variable t, which reflects the differences in the importance of various factors to the equipment operation and maintenance cycle at different points in time; the reliability coefficient refers to the numerical value describing the reliability of the intelligent equipment to maintain normal operation at a specific time t. The higher the reliability coefficient R(t), the greater the reliability of the equipment to operate normally at that point in time, and vice versa.

[0111] S3. Based on the equipment operation and maintenance cycle, assess the operating status of the intelligent equipment to obtain status assessment parameters. Based on the status assessment parameters, query the fault types of the intelligent equipment and analyze the occurrence time of the fault types.

[0112] Based on the equipment operation and maintenance cycle, this invention assesses the operating status of the intelligent equipment and obtains status assessment parameters. This helps to promptly identify potential problems in the equipment and take targeted intervention measures before the problems worsen, thereby effectively preventing equipment failures and ensuring stable equipment operation.

[0113] The status assessment parameters refer to a set of quantitative indicators and qualitative descriptions used to comprehensively describe and measure the operating status of intelligent devices.

[0114] As an embodiment of the present invention, the step of evaluating the operating status of the intelligent device based on the device operation and maintenance cycle to obtain status evaluation parameters includes: querying the original operating data of the intelligent device based on the device operation and maintenance cycle; removing invalid data from the original operating data to obtain valid operating data; analyzing the degree of deviation between the valid operating data and the historical normal operating data of the intelligent device; marking abnormal data segments in the valid operating data according to the degree of deviation; calculating the performance degradation magnitude of the intelligent device based on the abnormal data segments; and evaluating the operating status of the intelligent device based on the performance degradation magnitude to obtain status evaluation parameters.

[0115] Wherein, the raw operating data refers to all operation-related data directly collected from the intelligent device without processing or filtering; the effective operating data refers to useful data that can truly reflect the operating status of the device after filtering and processing to remove invalid, erroneous, or interfering information; the historical normal operating data refers to a representative set of data recorded by the intelligent device in its previous normal operating states; the deviation degree refers to a measure of the difference between the effective operating data and the historical normal operating data in various key indicators; the abnormal data segment refers to a continuous data segment in the effective operating data that is marked as having a large deviation from the normal situation; and the performance degradation magnitude refers to a quantitative value of the magnitude of the performance degradation of the intelligent device from the normal state to the current state.

[0116] Furthermore, the querying of the original operating data corresponding to the intelligent device can be achieved through a data acquisition interface, such as obtaining the original operating data using the OPC UA interface of industrial equipment; removing invalid data from the original operating data can be achieved through data cleaning tools, such as OpenRefine; analyzing the degree of deviation between the valid operating data and the historical normal operating data corresponding to the intelligent device can be achieved through clustering algorithms, such as clustering the valid operating data and the historical normal operating data separately and comparing the differences in the clustering results; marking abnormal data segments in the valid operating data can be achieved through the isolated forest algorithm, such as identifying abnormal data segments by constructing isolated trees; calculating the performance degradation magnitude corresponding to the intelligent device can be achieved through the following calculation formula; evaluating the operating status corresponding to the intelligent device can be achieved through the fuzzy comprehensive evaluation method, such as fuzzifying the qualitative evaluation indicators and comprehensively obtaining the evaluation data of the operating status.

[0117] As an embodiment of the present invention, the step of calculating the performance degradation value corresponding to the intelligent device based on the abnormal data segment includes:

[0118] The performance degradation of the intelligent device is calculated using the following formula:

[0119]

[0120] Where XF represents the performance degradation of the intelligent device, m represents the number of data points in the abnormal data segment, j represents the index of the data point in the abnormal data segment, and P nj P represents the actual value of the performance metric corresponding to the j-th data point in the abnormal data segment. hj This represents the expected value of the performance index corresponding to the j-th data point under normal operating conditions.

[0121] In detail, the performance degradation magnitude refers to a value used to quantify the degree of performance degradation of intelligent devices, which reflects the extent of performance reduction of the device when abnormal data segments occur, relative to the normal operating state; the actual value of the performance index refers to the value of the specific performance index measured in the abnormal data segment during the operation of the intelligent device; the expected value of the performance index refers to the theoretical or empirical value of the performance index expected to be achieved for the corresponding data point under normal operating conditions.

[0122] Based on the aforementioned status assessment parameters, this invention queries the types of faults present in the intelligent equipment, enabling rapid and accurate location of equipment faults. This significantly shortens troubleshooting time, improves maintenance efficiency, and helps in preparing necessary maintenance tools and parts in advance, allowing for preventative measures to avoid further deterioration of equipment faults.

[0123] The fault type refers to the specific types and forms of faults that occur in intelligent devices, such as hardware faults, software faults, and mechanical faults.

[0124] As an embodiment of the present invention, the step of querying the fault type of the intelligent device based on the state assessment parameters includes: analyzing the operational fluctuation of the intelligent device based on the state assessment parameters; extracting the operational fluctuation features corresponding to the operational fluctuation; querying the fluctuation indicators with faults in the operational fluctuation features; matching the fluctuation indicators with a preset fault indicator library to obtain matching data; filtering the fault indicator subset with high similarity in the matching data; and querying the fault type of the intelligent device based on the fault indicator subset.

[0125] The operational fluctuation refers to the changes and fluctuations in various performance parameters and operating status of intelligent equipment during operation; the operational fluctuation characteristics refer to the key attributes extracted from the operational fluctuations that can reflect the operating rules and characteristics of the equipment; the fluctuation index refers to the specific quantitative values ​​or feature identifiers used to measure the operational fluctuation characteristics that indicate faults; the matching data refers to the relevant result data obtained by comparing the fluctuation index with a preset fault index library, including information such as similarity and matching items; and the fault index subset refers to the set of fault indicators with high similarity selected from the matching data.

[0126] Furthermore, the analysis of the operational fluctuations of the intelligent equipment can be achieved through time series analysis methods, such as moving average and exponential smoothing; the extraction of operational fluctuation features can be achieved using Python tools, such as wavelet transform using the PyWavelets library in Python; the query for fluctuation indicators with faults in the operational fluctuation features can be achieved through pattern recognition algorithms, such as decision trees and support vector machines; the matching of the fluctuation indicators with a preset fault indicator library can be achieved through similarity algorithms, such as cosine similarity and Euclidean distance; the filtering of fault indicator subsets with high similarity in the matched data can be achieved through sorting algorithms, such as bubble sort and quicksort; and the query for the fault types existing in the intelligent equipment can be achieved through deep learning models, such as convolutional neural networks and recurrent neural networks.

[0127] By analyzing the occurrence time corresponding to the fault type, this invention helps to accurately trace the source of the fault and understand at which stage of equipment operation the problem occurred, thus providing key clues for in-depth analysis of the cause of the fault.

[0128] The occurrence time refers to the specific moment or time period when the fault type first appears or recurs. Optionally, the analysis of the occurrence time corresponding to the fault type can be achieved by data mining algorithms, such as the Apriori algorithm.

[0129] S4. Based on the fault type and the occurrence time, and combined with the current state of the intelligent device, construct a preventive maintenance strategy corresponding to the intelligent device, generate a maintenance task list corresponding to the intelligent device based on the preventive maintenance strategy, and query the maintenance tasks in the maintenance task list.

[0130] Based on the fault type and the time of occurrence, combined with the current state of the intelligent device, this invention constructs a corresponding preventive maintenance strategy for the intelligent device. It can comprehensively and deeply understand the evolution of the device's operating status, clearly grasp the entire process from normal operation to failure, and rationally arrange maintenance personnel and prepare the necessary parts and tools according to the fault type, time of occurrence, and current state, thereby greatly improving maintenance efficiency and reducing equipment downtime.

[0131] The aforementioned preventive maintenance strategy refers to a series of targeted maintenance plans, measures, and methods formulated based on the current status of the equipment, the trend of fault development, and the different maintenance items, in order to avoid similar faults from recurring in intelligent equipment or to reduce the occurrence of faults.

[0132] As an embodiment of the present invention, the step of constructing a preventive maintenance strategy for the intelligent device based on the fault type and the occurrence time, combined with the current state of the intelligent device, includes: analyzing the fault development trend of the intelligent device based on the fault type and the occurrence time; extracting key change nodes in the fault development trend; querying the node operating parameters corresponding to the key change nodes; comparing the node operating parameters with normal standard parameters to obtain difference parameters; analyzing the difference maintenance items corresponding to the difference parameters; and constructing a preventive maintenance strategy for the intelligent device based on the difference maintenance items and the current state of the intelligent device.

[0133] The fault development trend refers to the direction and extent of change of the fault type over time, including the evolution of the fault's severity, frequency, and scope of impact. The key change node refers to a specific point in time or state transition that is significant in the fault development trend and can significantly influence the direction or extent of the fault's development. The node operating parameters refer to the various operating indicators and data of the intelligent equipment at the key change node, such as temperature, pressure, and speed. The difference parameters refer to the different values ​​or characteristics between the node operating parameters and the normal standard parameters. The difference maintenance items refer to the specific items and contents that require special maintenance, adjustment, or improvement, as reflected by the difference parameters.

[0134] Furthermore, the analysis of the fault development trend corresponding to the intelligent equipment can be achieved through time series prediction models, such as ARIMA and SARIMA models; the extraction of key change nodes in the fault development trend can be achieved through clustering analysis algorithms, such as K-Means clustering; the query of the node operating parameters corresponding to the key change nodes can be achieved through principal component analysis algorithms, such as using principal component analysis (PCA) to extract the main feature parameters; the comparison of the node operating parameters with normal standard parameters can be achieved through hypothesis testing methods, such as t-tests and analysis of variance; the analysis of the differential maintenance items corresponding to the differential parameters can be achieved through fishbone diagram analysis, such as using fishbone diagrams to identify factors such as personnel operation, equipment aging, and environmental changes when analyzing the reasons for equipment performance degradation; the construction of the preventive maintenance strategy corresponding to the intelligent equipment can be achieved through the analytic hierarchy process (AHP), such as using AHP to determine the optimal maintenance strategy when considering factors such as equipment maintenance costs, equipment reliability, and maintenance personnel skills.

[0135] Based on the aforementioned preventive maintenance strategy, this invention generates a maintenance task list corresponding to the intelligent equipment, which helps to prepare the necessary resources for maintenance in advance, including manpower, material resources, and financial resources, to ensure that maintenance work can be carried out smoothly, reduce equipment downtime, and ensure the normal operation of the equipment, thereby minimizing production losses caused by equipment failure.

[0136] The maintenance task list refers to a list that details a series of specific maintenance operations, required resources, execution time, and responsible persons for the preventive maintenance strategy of intelligent equipment. Optionally, the generation of the maintenance task list corresponding to the intelligent equipment can be achieved by project management tools, such as Microsoft Project or Trello.

[0137] Furthermore, by querying the maintenance tasks in the maintenance task list, the present invention can clarify the work priorities and responsibilities, which helps to plan equipment resource arrangements in advance, ensure that maintenance work can be carried out efficiently and orderly, facilitate the tracking and supervision of the execution of maintenance tasks, promptly identify potential problems and risks, and take corresponding adjustment measures.

[0138] The maintenance task refers to the specific operations and work content that need to be performed in order to keep the intelligent equipment running normally, prevent malfunctions, or repair existing problems. Optionally, the querying of the maintenance task list can be achieved through a database query language, such as using SQL statements to filter maintenance tasks based on specific conditions from the database storing the maintenance task list.

[0139] S5. Analyze the prevention objectives in the maintenance task, identify the prevention obstacles in the prevention objectives, monitor the operation feedback data of the intelligent equipment based on the prevention obstacles, and generate the operation report corresponding to the intelligent equipment based on the operation feedback data.

[0140] By analyzing the prevention objectives in the maintenance tasks, this invention can accurately determine the focus and key aspects of maintenance work, rationally allocate resources, avoid resource waste and misallocation, help predict future equipment failure modes and performance degradation trends, make preparations in advance, and facilitate timely detection of deficiencies in maintenance strategies for adjustment and improvement.

[0141] The prevention objective refers to the specific results and expected effects that are expected to be achieved in the maintenance task to avoid equipment failure, performance degradation or other adverse conditions. Optionally, the analysis of the prevention objective in the maintenance task can be achieved by the objective decomposition method, such as decomposing the overall prevention objective into multiple specific and measurable sub-objectives for clearer analysis.

[0142] Furthermore, by identifying the obstacles to prevention in the prevention objectives, the present invention can anticipate potential problems that may affect the achievement of prevention objectives in maintenance tasks, thereby developing targeted countermeasures, optimizing maintenance processes and improving maintenance effectiveness, and ensuring the stable operation and long-term reliability of intelligent equipment.

[0143] The aforementioned preventive obstacles refer to the specific factors that are ultimately identified and can clearly hinder the achievement of the prevention goals, including various internal and external obstacles.

[0144] As an embodiment of the present invention, identifying the prevention obstacles in the prevention target includes: analyzing the target environment corresponding to the prevention target; extracting potential adverse factors in the target environment; assessing the degree of influence of the potential adverse factors on the prevention target; screening the obstacle factors among the potential adverse factors according to the degree of influence; and identifying the prevention obstacles in the prevention target based on the obstacle factors.

[0145] The target environment refers to the sum of all external and internal conditions related to the prevention objective, including the physical environment of equipment operation, workflow, and personnel operation. The potential adverse factors refer to various factors in the target environment that negatively affect the achievement of the prevention objective, such as equipment aging, non-standard operation, and environmental changes. The degree of influence refers to the magnitude of the negative impact of the potential adverse factors on the prevention objective, which can be assessed quantitatively or qualitatively. The hindering factors refer to the factors among the potential adverse factors that have a significant impact on the achievement of the prevention objective and become the main obstacles to achieving the prevention objective.

[0146] Furthermore, the analysis of the target environment corresponding to the prevention objective can be achieved through system analysis methods, such as Petri and ISM; the extraction of potential adverse factors in the target environment can be achieved through impact analysis methods, such as FMEA analysis using Excel spreadsheets; the assessment of the impact of the potential adverse factors on the prevention objective can be achieved through hierarchical analysis tools, such as AHP analysis tools like Expert Choice; the screening of hindering factors among the potential adverse factors can be achieved through simulation analysis tools, such as Crystal Ball; and the identification of prevention hindering factors in the prevention objective can be achieved through data mining tools, such as Weka.

[0147] Based on the aforementioned preventive obstacles, this invention monitors the operational feedback data of the intelligent device, enabling timely detection of any situations that hinder the prevention objective during the operation of the intelligent device, so as to quickly take countermeasures and prevent the problem from worsening.

[0148] The operational feedback data refers to various data information generated by the intelligent device during operation that reflects its current status, performance, and factors related to preventing obstacles. Optionally, the monitoring of the operational feedback data of the intelligent device can be achieved through network monitoring tools, such as SolarWinds, to monitor the device's network traffic and connection status in real time.

[0149] Based on the operational feedback data, this invention generates operational reports corresponding to the intelligent equipment, which helps to rationally allocate resources, promotes intelligent and refined equipment management based on predicted equipment needs and potential problems, continuously improves equipment management level, and ensures the stable and efficient operation of intelligent equipment.

[0150] The operation report refers to a comprehensive report based on the operation feedback data of intelligent devices, derived through analysis and prediction methods, regarding the device's operating status, performance trends, potential fault risks, and corresponding maintenance recommendations for a future period. Specifically, the operation report covers multiple aspects. Regarding operating status, it clearly indicates the device's current working mode, load, and operating efficiency, allowing users to clearly understand the device's current operating condition. For performance trends, the report predicts the device's performance changes over a future period through analysis of historical data and trend extrapolation, such as whether performance degradation or efficiency reduction will occur. In the section on potential fault risks, it assesses the likelihood of various faults based on abnormal signals in the feedback data, changes in key indicators, and comparisons with known fault modes, specifically indicating the type and location of the fault. Optionally, the generation of the operation report for the intelligent device can be achieved using data visualization tools, such as Tableau or Power BI, to display the device's key indicators and trend predictions.

[0151] First, this invention, by acquiring real-time data during the operation of intelligent equipment, reflects the current operating status of the equipment, including changes in various parameters. This helps to promptly detect anomalies and allow for proactive measures to prevent malfunctions. Second, based on the set of abnormal parameters, this invention constructs a fault prediction framework for the intelligent equipment. This framework fully utilizes the information provided by the abnormal parameters to conduct a comprehensive and in-depth analysis of the intelligent equipment's operating status, thereby more accurately predicting potential faults. Third, based on the equipment's maintenance cycle, this invention assesses the operating status of the intelligent equipment to obtain status assessment parameters. This helps to promptly identify potential problems and intervene with targeted measures before the problems escalate, effectively preventing equipment failures and ensuring stable equipment operation. This invention, based on the fault type and occurrence time, combined with the current state of the intelligent equipment, constructs a corresponding preventive maintenance strategy for the intelligent equipment. This allows for a comprehensive and in-depth understanding of the equipment's operational evolution, clearly grasping the entire process from normal operation to fault occurrence. Based on the fault type, occurrence time, and current state, maintenance personnel can be rationally assigned, and necessary parts and tools prepared, thereby significantly improving maintenance efficiency and reducing equipment downtime. Furthermore, by analyzing the preventive objectives in the maintenance tasks, this invention can accurately identify the key points and critical aspects of maintenance work, rationally allocate resources, avoid resource waste and misallocation, and help predict future fault modes and performance degradation trends, enabling advance preparation and timely identification, adjustment, and improvement of maintenance strategies. Therefore, the fault prediction and prevention method and system in the intelligent equipment transformation technology proposed in this invention can improve the operational stability of intelligent equipment.

[0152] like Figure 2 The diagram shown is a functional block diagram of a fault prediction and prevention system in an intelligent equipment transformation technology provided by an embodiment of the present invention.

[0153] The fault prediction and prevention system 200 in the intelligent equipment transformation technology of this invention can be installed in electronic devices. Depending on the functions implemented, the fault prediction and prevention system 200 in the intelligent equipment transformation technology may include a feature extraction module 201, an operation and maintenance cycle calculation module 202, a time analysis module 203, a task query module 204, and a report generation module 205. The module described in this invention can also be called a unit, which refers to a series of computer program segments that can be executed by the processor of an electronic device and can perform a fixed function, and which are stored in the memory of the electronic device.

[0154] In this embodiment, the functions of each module / unit are as follows:

[0155] The feature extraction module 201 is used to acquire real-time data during the operation of intelligent equipment, filter the real-time data to obtain filtered real-time data, filter the abnormal entropy parameters corresponding to the filtered real-time data, and extract features from the abnormal entropy parameters to obtain an abnormal parameter set.

[0156] The maintenance cycle calculation module 202 is used to construct a fault prediction framework corresponding to the intelligent device based on the abnormal parameter set, predict the fault probability corresponding to the intelligent device using the fault prediction framework, determine the remaining operating life of the intelligent device based on the fault probability, and calculate the equipment maintenance cycle corresponding to the intelligent device based on the remaining operating life. The step of calculating the equipment maintenance cycle corresponding to the intelligent device based on the remaining operating life includes:

[0157] The equipment maintenance cycle corresponding to the intelligent device is calculated using the following formula:

[0158]

[0159] Wherein, WT represents the equipment operation and maintenance cycle corresponding to the intelligent device, k represents the adjustment coefficient corresponding to the intelligent device, L represents the remaining operating life, t represents the time variable, f(t) represents the weight coefficient of the time variable t, R(t) represents the reliability coefficient at time t, and dt represents the small increment of the time variable t.

[0160] Time analysis module 203 is used to evaluate the operating status of the intelligent equipment based on the equipment operation and maintenance cycle, obtain status evaluation parameters, query the fault types of the intelligent equipment based on the status evaluation parameters, and analyze the occurrence time of the fault types.

[0161] The task query module 204 is used to construct a preventive maintenance strategy for the intelligent device based on the fault type and the occurrence time, combined with the current state of the intelligent device, generate a maintenance task list for the intelligent device based on the preventive maintenance strategy, and query the maintenance tasks in the maintenance task list.

[0162] The report generation module 205 is used to analyze the prevention objectives in the maintenance task, identify the prevention obstacles in the prevention objectives, monitor the operation feedback data of the intelligent equipment based on the prevention obstacles, and generate an operation report corresponding to the intelligent equipment based on the operation feedback data.

[0163] In detail, each module in the fault prediction and prevention system 200 of the intelligent equipment transformation technology described in the embodiments of the present invention adopts the same technical means as the fault prediction and prevention method in the intelligent equipment transformation technology described in the accompanying drawings, and can produce the same technical effect, which will not be repeated here.

[0164] The above description is merely a specific embodiment of the present invention, enabling those skilled in the art to understand or implement the invention. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the invention. Therefore, the invention is not to be limited to the embodiments shown herein, but is to be accorded the widest scope consistent with the principles and novel features of the invention herein.

Claims

1. A method for failure prediction and prevention in intelligent equipment retrofitting technology, characterized by, The method includes: Real-time data during the operation of intelligent equipment is acquired, the real-time data is filtered to obtain filtered real-time data, the abnormal entropy parameters corresponding to the filtered real-time data are selected, and the abnormal entropy parameters are feature extracted to obtain an abnormal parameter set. Based on the abnormal parameter set, a fault prediction framework corresponding to the intelligent device is constructed. The fault prediction framework is used to predict the fault probability of the intelligent device. Based on the fault probability, the remaining operating life of the intelligent device is determined. Based on the remaining operating life, the equipment maintenance cycle of the intelligent device is calculated. The calculation of the equipment maintenance cycle based on the remaining operating life includes: The equipment maintenance cycle corresponding to the intelligent device is calculated using the following formula: Wherein, WT represents the equipment operation and maintenance cycle corresponding to the intelligent device, k represents the adjustment coefficient corresponding to the intelligent device, L represents the remaining operating life, t represents the time variable, f(t) represents the weight coefficient of the time variable t, R(t) represents the reliability coefficient at time t, and dt represents the small increment of the time variable t. Based on the equipment operation and maintenance cycle, the operating status of the intelligent equipment is evaluated to obtain status evaluation parameters. Based on the status evaluation parameters, the fault types of the intelligent equipment are queried, and the occurrence time of the fault types is analyzed. Based on the fault type and the occurrence time, and combined with the current state of the intelligent device, a preventive maintenance strategy corresponding to the intelligent device is constructed. Based on the preventive maintenance strategy, a maintenance task list corresponding to the intelligent device is generated, and maintenance tasks in the maintenance task list are queried. The preventive objectives in the maintenance task are analyzed, the preventive obstacles in the preventive objectives are identified, the operational feedback data of the intelligent equipment is monitored based on the preventive obstacles, and the corresponding operational report of the intelligent equipment is generated based on the operational feedback data.

2. The method for failure prediction and prevention in intelligent equipment reengineering technology according to claim 1, wherein, The abnormal entropy parameter corresponding to the filtered real-time data includes: Generate the entropy sequence corresponding to the filtered real-time data; Based on the entropy sequence, calculate the average entropy value corresponding to the filtered real-time data; Based on the average entropy value, determine the entropy fluctuation range of the filtered real-time data; Based on the entropy fluctuation range, the abnormal entropy parameters corresponding to the filtered real-time data are selected.

3. The method of claim 2, wherein the method further comprises: The step of calculating the average entropy value corresponding to the filtered real-time data based on the entropy value sequence includes: The average entropy value corresponding to the filtered real-time data is calculated using the following formula: in, H represents the average entropy value corresponding to the filtered real-time data, N represents the total number of entropy values ​​in the entropy value sequence, i represents the index of the number of entropy values ​​in the entropy value sequence, and H represents the average entropy value. i This represents the i-th entropy value in the entropy value sequence.

4. The method of claim 1, wherein the method is characterized by, The step of constructing a fault prediction framework for the intelligent device based on the abnormal parameter set includes: Analyze the parameter distribution patterns corresponding to the abnormal parameter set; Based on the parameter distribution pattern, the importance of the parameters in the abnormal parameter set is determined; Based on the degree of importance, key abnormal parameters corresponding to the abnormal parameter set are selected; Based on the key anomaly parameters, a fault feature vector corresponding to the intelligent device is generated; Based on the fault feature vector, the fault classification standard corresponding to the intelligent device is determined; Based on the fault classification criteria, a fault prediction framework corresponding to the intelligent device is constructed.

5. The method of claim 1, wherein the method is characterized by, Determining the remaining operational life of the intelligent device based on the failure probability includes: Based on the failure probability, obtain the historical operating data corresponding to the intelligent device; Analyze the performance degradation trend in the historical operating data; Based on the aforementioned performance degradation trend, predict the time span from performance degradation to failure of the intelligent device; Analyze the uncertainties corresponding to the time span; Based on the aforementioned uncertainties, the remaining operational lifespan of the intelligent device is determined.

6. The method for failure prediction and prevention in intelligent reengineering of equipment according to claim 1, wherein, The process of assessing the operational status of the intelligent device based on the device's maintenance cycle to obtain status assessment parameters includes: Based on the equipment maintenance cycle, query the original operating data corresponding to the intelligent equipment; Remove invalid data from the original running data to obtain valid running data; Analyze the degree of deviation between the effective operating data and the historical normal operating data corresponding to the intelligent device; Based on the degree of deviation, mark the abnormal data segments in the valid operating data; Based on the abnormal data segment, calculate the performance degradation value corresponding to the intelligent device; Based on the magnitude of performance degradation, the operating status of the intelligent device is assessed to obtain status assessment parameters.

7. The method of predictive maintenance of claim 6, wherein, The step of calculating the performance degradation of the intelligent device based on the abnormal data segment includes: The performance degradation of the intelligent device is calculated using the following formula: Where XF represents the performance degradation of the intelligent device, m represents the number of data points in the abnormal data segment, j represents the index of the data point in the abnormal data segment, and P nj P represents the actual value of the performance metric corresponding to the j-th data point in the abnormal data segment. hj This represents the expected value of the performance index corresponding to the j-th data point under normal operating conditions.

8. The method for failure prediction and prevention in intelligent reengineering of equipment according to claim 1, wherein, The step of querying the fault types of the intelligent device based on the status assessment parameters includes: Based on the aforementioned status assessment parameters, analyze the operational fluctuations of the intelligent equipment. Extract the operational fluctuation characteristics corresponding to the aforementioned operational fluctuation conditions; Query the fluctuation indicators that show faults in the operational fluctuation characteristics; The fluctuation index is matched with a preset fault index library to obtain matching data; Filter the subset of fault indicators with high similarity from the matched data; Based on the subset of fault indicators, query the types of faults existing in the intelligent device.

9. The method for failure prediction and prevention in intelligent reengineering of equipment according to claim 1, wherein, The method for constructing a preventive maintenance strategy for the intelligent device based on the fault type and the time of occurrence, combined with the current state of the intelligent device, includes: Based on the fault type and occurrence time, analyze the fault development trend corresponding to the intelligent device; Extract the key change nodes in the fault development trend; Query the node operation parameters corresponding to the key change nodes; The node's operating parameters are compared with normal standard parameters to obtain the difference parameters; Analyze the difference maintenance items corresponding to the difference parameters; Based on the aforementioned differential maintenance items and the current state of the intelligent device, a preventive maintenance strategy corresponding to the intelligent device is constructed.

10. A failure prediction and prevention system in intelligent device retrofit technology, characterized by, The system is used to perform the fault prediction and prevention method in the intelligent equipment transformation technology as described in any one of claims 1-9, the system comprising: The feature extraction module is used to acquire real-time data during the operation of intelligent equipment, filter the real-time data to obtain filtered real-time data, filter the abnormal entropy parameters corresponding to the filtered real-time data, and extract features from the abnormal entropy parameters to obtain an abnormal parameter set. The operation and maintenance cycle calculation module is used to construct a fault prediction framework corresponding to the intelligent device based on the abnormal parameter set, predict the fault probability corresponding to the intelligent device using the fault prediction framework, determine the remaining operating life of the intelligent device based on the fault probability, and calculate the equipment operation and maintenance cycle corresponding to the intelligent device based on the remaining operating life. The step of calculating the equipment operation and maintenance cycle corresponding to the intelligent device based on the remaining operating life includes: The equipment maintenance cycle corresponding to the intelligent device is calculated using the following formula: Wherein, WT represents the equipment operation and maintenance cycle corresponding to the intelligent device, k represents the adjustment coefficient corresponding to the intelligent device, L represents the remaining operating life, t represents the time variable, f(t) represents the weight coefficient of the time variable t, R(t) represents the reliability coefficient at time t, and dt represents the small increment of the time variable t. The time analysis module is used to evaluate the operating status of the intelligent equipment based on the equipment operation and maintenance cycle, obtain status evaluation parameters, query the fault types of the intelligent equipment based on the status evaluation parameters, and analyze the occurrence time of the fault types. The task query module is used to construct a preventive maintenance strategy for the intelligent device based on the fault type and the occurrence time, combined with the current status of the intelligent device, generate a maintenance task list for the intelligent device based on the preventive maintenance strategy, and query the maintenance tasks in the maintenance task list. The report generation module is used to analyze the prevention objectives in the maintenance task, identify the prevention obstacles in the prevention objectives, monitor the operation feedback data of the intelligent equipment based on the prevention obstacles, and generate the corresponding operation report of the intelligent equipment based on the operation feedback data.

Images