Operation and maintenance management method and system based on digital twinning

Abstract

The invention relates to the technical field of operation and maintenance management, in particular to an operation and maintenance management method based on digital twinning, and the method comprises the following steps: S1, building a corresponding digital twinning model through collecting the structure data and operation data of an actual operation and maintenance system; s2, collecting operation data of the operation and maintenance system in real time, transmitting the operation data to the cloud platform, and preprocessing the operation data transmitted in real time in the cloud platform; s3, performing real-time analysis on the preprocessed data through a digital twinborn model, and detecting a system operation state and an abnormal condition; s4, based on the historical data and the real-time data, using a machine learning algorithm to construct a fault prediction model, and predicting possible faults in the future; and S5, generating an optimization management scheme according to the fault prediction result and the system operation state. Through accurate data acquisition and real-time monitoring, potential faults and hidden dangers of equipment can be found in time, accidental shutdown of the equipment is avoided, and production loss caused by faults is greatly reduced.

Description

Technical Field

[0001] The present invention relates to the field of operation and maintenance management technology, and in particular to an operation and maintenance management method and system based on digital twins. Background Art

[0002] With the rapid development of digital and intelligent technologies, operation and maintenance management technologies are also constantly innovating. The operation and maintenance management methods and systems based on digital twins build virtual models to achieve real-time monitoring, fault prediction and optimization management of actual operation and maintenance systems, greatly improving operation and maintenance efficiency and management level. However, existing operation and maintenance management technologies still have some shortcomings in comprehensive monitoring, intelligent analysis and prediction, resulting in problems such as low operation and maintenance efficiency and slow response speed in actual applications.

[0003] After searching, an operation and maintenance management system with publication number CN114637640B was disclosed, and the publication date is April 19, 2024. This patent provides an operation and maintenance management system, including a cloud platform, an operation and maintenance terminal and a management server. The management server has a monitoring module storage unit, a module acquisition unit, a monitoring module output unit, a device status judgment unit and a management side communication unit. The system realizes real-time monitoring and maintenance of cloud servers through a modular monitoring and alarm mechanism. However, in this technical solution, the acquisition and operation of the monitoring module rely on the manual selection of the operation and maintenance personnel, and lacks automated and intelligent management methods, which may lead to low operation and maintenance efficiency. In addition, the adaptability and compatibility of the system with multiple types of cloud servers needs to be improved, and it is difficult to meet the complex and changeable operation and maintenance needs.

[0004] After searching, an IT operation and maintenance management method and an IT operation and maintenance management device with publication number CN118550573B were disclosed, and the publication date is October 29, 2024. This patent reflects the operating status and potential risks of the data center through data processing analysis and feature value extraction, builds a risk prediction model, and realizes risk warning and emergency response. However, in this technical solution, the data processing and analysis process is relatively complicated, especially in the big data environment, there are problems with the real-time and accuracy of data processing. In addition, the risk prediction model of the system relies on historical data, and its ability to respond to emerging operation and maintenance problems and abnormal situations is limited, which may lead to insufficient flexibility and intelligence in operation and maintenance management.

[0005] The above problems show that the existing operation and maintenance management technology still has certain shortcomings in automation, intelligence, real-time monitoring and risk prediction. Summary of the invention

[0006] Based on the above purpose, the present invention provides an operation and maintenance management method based on digital twins, comprising the following steps: S1: Build the corresponding digital twin model by collecting the structural data and operation data of the actual operation and maintenance system; S2: collects the operation data of the operation and maintenance system in real time and transmits it to the cloud platform, and pre-processes the real-time transmitted operation data in the cloud platform; S3: Use the digital twin model to analyze the pre-processed data in real time to detect system operation status and abnormal conditions; S4: Based on historical data and real-time data, a fault prediction model is constructed using a machine learning algorithm to predict possible future faults. The input of the fault prediction model is historical data and real-time data, and the output is the fault prediction result. S5: Generate an optimization management plan based on the fault prediction results and system operation status, including system maintenance, resource scheduling and fault prevention measures.

[0007] Preferably, S1 specifically includes the following steps: S1.1: Collect structural data and operation data of the operation and maintenance system through sensors and data acquisition equipment, including equipment status, environmental parameters and system performance indicators; S1.2: Preprocess the collected data, including data cleaning, normalization and format conversion. Data cleaning is used to remove invalid data and outliers in the data, normalization is used to normalize the data to the range of [0, 1], and format conversion is used to convert the preprocessed data into a unified format; S1.3: Build a physical model based on the structural data of the operation and maintenance system, and build a behavioral model based on the operating data of the operation and maintenance system. Combine the physical model and the behavioral model to form a digital twin model.

[0008] Preferably, in S1.3, the physical model includes geometric parameters, material properties and connection relationships of the device; The geometric parameters of the device are described using the following formula: ;in, For the The geometric parameters of each device, including the length, width, height, weight, material properties, and connection relationships of the device; The behavioral model describes the operating status and performance indicators of the device. The operating status of the device is described using the following formula: ;in, For the The operating status of each device, including the operating temperature, operating current, operating voltage, vibration frequency and energy consumption of the device; The physical model and the behavioral model are combined to form a digital twin model, which can be expressed using the following formula: ;in, It is a digital twin model, which is composed of physical models and behavioral models Comprehensive.

[0009] Preferably, S3 specifically includes the following steps: S3.1: Extract key features from the preprocessed data, including equipment status, environmental parameters, and system performance indicators; S3.2: Map the extracted features into the digital twin model; S3.3: Perform status detection based on the mapped data; S3.4: Classify detected anomalies.

[0010] Preferably, in S3.1, the feature extraction process is represented by the following formula: ;in, For the Features In S3.2, the extracted features are mapped into the digital twin model, and the data mapping process is represented by the following formula: ;in, For the Mapping data.

[0011] Preferably, in S3.3, based on the mapped data, the following formula is used to perform state detection:

[0012] The threshold is a preset boundary value between a normal state and an abnormal state; In S3.4, The detected anomalies are classified using the following formula: ; Among them, threshold 1, threshold 2, threshold 3, threshold 4, etc. are preset abnormality type judgment values.

[0013] Preferably, the specific steps of S4 include: 4.1: Collect a large amount of pre-processed historical data and real-time data of the operation and maintenance system, including temperature, current, voltage, vibration frequency and energy consumption; 4.2: Construct a long short-term memory network based on the attention mechanism, including the input layer, LSTM layer, attention mechanism layer and output layer. The input layer receives preprocessed data in the format of [time step, number of features], where the time step represents the length of the time series and the number of features represents the number of features in each time step. LSTM layer: contains multiple LSTM units, each of which consists of an input gate, an output gate, and a forget gate. LSTM units are used to capture long-term dependencies in time series data. The parameters of the LSTM layer are set as follows: Input dimension: number of features; Hidden layer dimension: 128; Time steps: 100; Attention mechanism layer: Introducing the attention mechanism to highlight the data of key time steps in the time series in a weighted manner; Output layer: Use a fully connected layer to map the weighted hidden state to the fault prediction result. The activation function of the output layer is the sigmoid function, and the output range is [0, 1], which represents the probability of fault occurrence. The output data is the fault prediction result, including fault type, fault location, and fault time. 4.3: Establish training sets and test sets to train and test the long short-term memory network based on the attention mechanism.

[0014] Preferably, in S4.3, the training set includes historical operation and maintenance data, the amount of which is , the data format is , each piece of data Includes historical operating parameters and fault tags of the equipment ; The test set includes real-time operation and maintenance data, the data volume is , the data format is , each piece of data Includes real-time operating parameters and fault tags of the equipment .

[0015] Preferably, in S5, the following steps are included: S5.1: The cloud platform generates an optimization management plan based on the fault prediction results and system operation status; S5.2: The optimization management plan includes system maintenance plan, resource scheduling plan and fault prevention measures; S5.3: The specific system maintenance plan includes regular inspection and maintenance of equipment, replacement of aging or damaged parts, and ensuring the normal operation of the equipment; S5.4: The resource scheduling scheme dynamically adjusts resource allocation and optimizes system performance based on the system operation status and prediction results; S5.5: Fault prevention measures include strengthening monitoring and alarms, and promptly handling possible fault hazards to avoid the occurrence of faults.

[0016] Accordingly, the embodiment of the present invention also provides an operation and maintenance management system based on digital twins, which is used to run the operation and maintenance management method based on digital twins described in the embodiment of the present invention, including a data acquisition module, a data preprocessing module, a digital twin module, a fault prediction module, an optimization management module, a cloud platform and an operation and maintenance terminal, wherein the data acquisition module, the data preprocessing module, the digital twin module, the fault prediction module, the optimization management module and the operation and maintenance terminal are all electrically connected to the cloud platform; The data acquisition module includes sensors and data collectors. The sensors are used to collect the structural data and operation data of the operation and maintenance system in real time, including equipment status, environmental parameters and system performance indicators; the data collectors are used to perform preliminary processing on the data collected by the sensors and transmit them to the cloud platform via wired or wireless means. The cloud platform includes a data storage unit, a data processing unit, a communication unit, and a management unit. The data storage unit is used to store data collected by the data acquisition module and preprocessed data. The data processing unit is used to execute the functions of the data preprocessing module, the digital twin module, and the fault prediction module. The communication unit is used to transmit data with the data acquisition module and the operation and maintenance terminal. The management unit is used to manage various functional units of the cloud platform. The operation and maintenance terminal includes a display unit, an operation unit and a communication unit. The display unit is used to display the working status and optimization management plan of the operation and maintenance management system, the operation unit is used for operation and management by operation and maintenance personnel, and the communication unit is used to transmit data with the cloud platform.

[0017] Beneficial effects of the present invention: 1. The operation and maintenance management method of digital twins can not only improve the predictability of equipment, but also realize intelligent fault prevention and maintenance management. Through accurate data collection and real-time monitoring, potential equipment failures can be discovered in time, avoiding unexpected equipment downtime and greatly reducing production losses caused by failures.

[0018] 2. Through precise data collection, rigorous data preprocessing and multi-dimensional modeling, the efficient construction of the digital twin model is ensured, thereby providing accurate and reliable information support for equipment operation and maintenance management, which can significantly improve the reliability of equipment, extend its service life, and reduce operation and maintenance costs.

[0019] 3. By integrating physical models and behavioral models, the digital twin model can accurately simulate the physical structure and dynamic operating status of the equipment. This simulation can not only help operation and maintenance personnel understand the working principle of the equipment, but also predict the performance of the equipment under different working conditions. For example, physical models based on geometric parameters and material properties can help evaluate the durability of equipment in extreme environments such as high temperature and high pressure, while behavioral models can predict the temperature, vibration and other characteristics of the equipment under specific operating conditions and provide real-time operation feedback.

[0020] 4. Through reasonable data extraction, mapping, detection and classification, a complete operation and maintenance management process is formed, which not only improves the accuracy and timeliness of equipment status monitoring, but also enhances abnormal diagnosis and response capabilities, thereby providing efficient and reliable support for equipment operation and maintenance.

[0021] 5. By establishing an accurate connection between the actual operating status of the equipment and the virtual model, it provides strong support for subsequent status detection, fault diagnosis and predictive maintenance. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] In order to more clearly illustrate the technical solutions in the present invention or the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings in the following description are only for the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0023] Figure 1 is a flow chart of the steps of the method of the present invention; Figure 2 is a flow chart of the steps of method S1 of the present invention; Figure 3 A block diagram of the system of the present invention. DETAILED DESCRIPTION

[0024] The present invention is described in detail below in conjunction with the accompanying drawings and specific embodiments. At the same time, it is explained here that in order to make the embodiments more detailed, the following embodiments are the best and preferred embodiments, and those skilled in the art may also adopt other alternatives to implement some known technologies; and the accompanying drawings are only for more specific description of the embodiments, and are not intended to specifically limit the present invention.

[0025] See also Figure 1-Figure 3 , an embodiment of the present invention provides an operation and maintenance management method based on digital twins, comprising the following steps: S1: Build the corresponding digital twin model by collecting the structural data and operation data of the actual operation and maintenance system; S2: collects the operation data of the operation and maintenance system in real time and transmits it to the cloud platform, and pre-processes the real-time transmitted operation data in the cloud platform; S3: Use the digital twin model to analyze the pre-processed data in real time to detect system operation status and abnormal conditions; S4: Based on historical data and real-time data, a fault prediction model is constructed using a machine learning algorithm to predict possible future faults. The input of the fault prediction model is historical data and real-time data, and the output is the fault prediction result. S5: Generate an optimization management plan based on the fault prediction results and system operation status, including system maintenance, resource scheduling and fault prevention measures.

[0026] In one possible implementation, before starting operation and maintenance management, it is first necessary to build a digital twin model by collecting the structural data and operating data of the operation and maintenance system. The structural data includes the geometric dimensions, material properties, and composition structure of the equipment, while the operating data involves the operating status, environmental conditions, and load conditions of the equipment. This digital twin model is the core of the subsequent operation and maintenance process and can provide accurate physical and behavioral data support for subsequent analysis and decision-making. By virtualizing the equipment, the digital twin model provides a basis for the entire operation and maintenance management to truly reflect the equipment status.

[0027] Furthermore, during the normal operation of the equipment, real-time collection of operating data is crucial. Through sensors and monitoring equipment, the system can obtain real-time data. After being transmitted to the cloud platform, these data undergo preprocessing processes such as denoising, normalization, and formatting to ensure the data quality of subsequent analysis. This step provides a high-quality and accurate data foundation for subsequent real-time analysis and fault prediction.

[0028] Furthermore, after data preprocessing is completed, the digital twin model analyzes the data in real time and monitors the operating status of the system. In this process, the digital twin model provides instant feedback on the actual operating status of the equipment through virtual mapping, which can effectively detect abnormal conditions during operation. For example, when the temperature or vibration value of the equipment exceeds the predetermined threshold, the system can immediately identify potential failure risks. Through abnormality detection, timely warnings can be issued to avoid further development of equipment failures.

[0029] Furthermore, based on historical data and real-time collected operation data, a fault prediction model is constructed using machine learning algorithms (such as LSTM networks). The input of this model is historical data and real-time data. The model continuously optimizes the prediction accuracy by training a large amount of historical data, thereby realizing the prediction of future faults. The fault prediction model can not only predict the time, location and type of faults, but also identify potential risks in advance. The implementation of this step greatly improves the reliability of the equipment, avoids sudden failures and reduces downtime.

[0030] Furthermore, based on the results of fault prediction and the real-time operating status of the equipment, the system generates optimized management plans. These plans may include regular maintenance plans for equipment, dynamic resource scheduling strategies, and preventive measures for predicted faults. These optimized management plans can help operation and maintenance personnel prepare in advance and deal with problems in a timely manner, thereby improving the operating efficiency of equipment, extending its service life, and effectively reducing maintenance costs.

[0031] In the embodiment of the present invention, S1 specifically includes the following steps: S1.1: Collect structural data and operation data of the operation and maintenance system through sensors and data acquisition equipment, including equipment status, environmental parameters and system performance indicators; S1.2: Preprocess the collected data, including data cleaning, normalization and format conversion. Data cleaning is used to remove invalid data and outliers in the data, normalization is used to normalize the data to the range of [0, 1], and format conversion is used to convert the preprocessed data into a unified format; S1.3: Build a physical model based on the structural data of the operation and maintenance system, and build a behavioral model based on the operating data of the operation and maintenance system. Combine the physical model and the behavioral model to form a digital twin model.

[0032] In one possible implementation, in step S1.1, the operation and maintenance system is fully monitored through sensors and data acquisition equipment to collect the structural data, environmental parameters and system performance indicators of the equipment. The structural data includes the geometric characteristics, components, configuration, etc. of the equipment; the operating data involves the real-time status of the equipment, operating parameters (such as temperature, pressure, current, vibration, etc.), environmental factors (such as humidity, temperature, air pressure, etc.) and system performance indicators (such as power, efficiency, etc.). These data are the basis for building a digital twin model to ensure that the information obtained truly reflects the working condition and operating environment of the equipment.

[0033] Furthermore, step S1.2 cleans, normalizes and converts the data collected in S1.1. Data cleaning is to remove invalid data (such as erroneous data generated by sensor failure) and outliers in the original data to ensure that the data used in subsequent analysis is accurate and valid. The normalization process maps the data to a standard range (usually [0, 1]) to facilitate comparison and model processing between different features. Format conversion converts the cleaned and normalized data into a unified format to accommodate subsequent data analysis, modeling and system integration. The preprocessed data is of high quality and consistency, and can provide reliable input for building models.

[0034] Furthermore, in step S1.3, a physical model and a behavioral model are constructed based on the preprocessed structural data and operating data, respectively. The physical model reflects the actual geometric structure, component properties and their interrelationships of the equipment, ensuring that the digital twin model can accurately reproduce the physical characteristics of the equipment. The behavioral model constructs the dynamic behavior of the equipment by analyzing operating data (such as real-time data such as temperature and pressure), and expresses the response characteristics of the equipment under different environmental conditions. By combining the physical model and the behavioral model, a comprehensive digital twin model is formed, which enables the virtual model to simulate the behavior of the equipment in real time during dynamic operation and reflect its current state.

[0035] In the embodiment of the present invention, in S1.3, the physical model includes geometric parameters, material properties and connection relationships of the device; The geometric parameters of the device are described using the following formula: ;in, For the The geometric parameters of each device, including the length, width, height, weight, material properties, and connection relationships of the device; The behavioral model describes the operating status and performance indicators of the device. The operating status of the device is described using the following formula: ;in, For the The operating status of each device, including the operating temperature, operating current, operating voltage, vibration frequency and energy consumption of the device; The physical model and the behavioral model are combined to form a digital twin model, which can be expressed using the following formula: ;in, It is a digital twin model, which is composed of physical models and behavioral models Comprehensive.

[0036] In a possible implementation, the physical model includes geometric parameters, material properties, and connection relationships of the device, which together constitute the basic physical characteristics of the device. The geometric parameters of the device are described by the following formula: ; For the The geometric parameters of a device include the length, width, height, weight, material properties and connection relationships of the device. The geometric parameters provide basic structural information of the device, the material properties reflect the physical properties of each component of the device (such as strength, temperature resistance, etc.), and the connection relationship describes how different components work together through physical contact or connection. This information helps build a three-dimensional model of the device and simulate the physical performance of the device under various environmental conditions.

[0037] The behavioral model describes the performance of the device through its operating status, reflecting the dynamic performance of the device in actual operation. The specific parameters of the behavioral model include the device's operating temperature, operating current, operating voltage, vibration frequency, and energy consumption. The following formula is used to describe the device's operating status: ;in, For the The operating status of each device, including the real-time working conditions and performance data of the device. The behavioral model reflects the operating characteristics of the device under different loads, environments and working conditions through the real-time changes of these parameters, helping to predict possible failures or performance degradation trends of the device.

[0038] Combine the physical model with the behavioral model to form a comprehensive digital twin model to represent the real state of the equipment in actual operation: ;in, It is a digital twin model, which is composed of physical models and behavioral models Through this integration, the digital twin model can simultaneously display the physical structure of the equipment and its dynamic operating status, providing all-round support for further simulation, prediction, analysis and decision-making.

[0039] In the embodiment of the present invention, S3 specifically includes the following steps: S3.1: Extract key features from the preprocessed data, including equipment status, environmental parameters, and system performance indicators; S3.2: Map the extracted features into the digital twin model; S3.3: Perform status detection based on the mapped data; S3.4: Classify detected anomalies.

[0040] In one possible implementation, in S3.1, key features related to equipment health and operating status need to be extracted from the original equipment data. These features include equipment status (such as temperature, pressure, speed, etc.), environmental parameters (such as humidity, temperature, air quality, etc.) and system performance indicators (such as power consumption, vibration frequency, fault codes, etc.). The purpose of extracting these features is to find the core information that can accurately reflect the equipment's operating health and potential risks from massive amounts of data, usually using data processing and feature selection techniques such as statistical analysis, Fourier transform, principal component analysis, etc. This step ensures the validity of the data and the efficiency of processing, and removes redundant or irrelevant noise information.

[0041] Furthermore, in S3.2, the key features extracted from the equipment operation data are mapped to the constructed digital twin model. The digital twin model contains the physical model and behavior model of the equipment, and can show the current operating status of the equipment by combining with the real-time data of the actual equipment. In this step, the extracted features are converted into a data format that can be recognized by the digital twin model to ensure the accuracy of the mapping. This mapping process is the key to realizing the interaction between the digital twin model and the real equipment, and can accurately reflect the actual situation of the equipment into the virtual model, thereby providing a real and reliable data basis for subsequent analysis.

[0042] Further, in S3.3, based on the data mapping in step S3.2, the system begins to detect and analyze the equipment status. This process uses the mapped data to monitor the health of the equipment and detect whether there are any abnormalities in real time. By comparing with the normal operating range of the equipment, the system can detect abnormal conditions in the operation of the equipment, such as excessive temperature, excessive vibration, etc. Status detection uses algorithm models for data analysis, which may include machine learning models, statistical analysis methods, or rule-driven methods. The purpose of this step is to evaluate the operating status of the equipment in real time and identify potential faults or performance issues in a timely manner.

[0043] Furthermore, in S3.4, the system classifies the detected anomalies based on the results of the status detection. Abnormal conditions may be caused by a variety of factors, including equipment failure, environmental changes, or external interference. Through classification, the system can further determine the severity of the anomaly, possible causes, and corresponding treatment measures. For example, if the device temperature is detected to be too high, the system can classify it as a possible cooling system failure or excessive ambient temperature, and take corresponding operation and maintenance measures based on the classification results. This classification step is crucial for operation and maintenance decisions. It helps operation and maintenance personnel quickly identify the cause of the anomaly and make targeted responses.

[0044] In the embodiment of the present invention, in S3.1, the feature extraction process is represented by the following formula: ;in, For the Features In S3.2, the extracted features are mapped into the digital twin model, and the data mapping process is represented by the following formula: ;in, For the Mapping data.

[0045] In a possible implementation, in S3.1, key features are first extracted from the original device data. A set of feature sets representing the device status is generated through a series of mathematical and statistical methods. The feature set is expressed as: ;in, For the Features include the device's temperature, pressure, vibration frequency, power consumption, etc. The purpose of feature extraction is to filter out the key information that best represents the health status of the device from a large amount of raw data. The selection of these features is based on previous experience, engineering knowledge, and data analysis, with the goal of making the extracted features accurately reflect the health status, performance changes, or failure risks of the device.

[0046] Further, in S3.2, the features extracted in S3.1 are mapped to the digital twin model. The digital twin model is a mathematical model that virtually represents the state of the device, which can simulate the behavior of the device under various operating conditions. The mapping process is expressed by the formula: ;in, For the The core of this step is to extract the features It is converted into a format that can be processed by the digital twin model and mapped to the digital twin model through mathematical functions or algorithms. This process enables the real-time operation data of the equipment to be reflected in the virtual model, thereby simulating and analyzing the actual operation status of the equipment.

[0047] The feature extraction in S3.1 is closely linked to the data mapping process in S3.2. Each feature extracted by S3.1 , will be mapped into the digital twin model as input to form mapping data This connection ensures the correspondence between the actual equipment operating status and the digital twin model, enabling the digital twin model to reflect the current health status, performance changes, and potential failure risks of the equipment. The accuracy of feature and data mapping directly affects the effectiveness of subsequent status detection and fault diagnosis.

[0048] Through mathematical methods and algorithms, features Converted into mapping data This process can be regarded as a transformation from high-dimensional device state features to low-dimensional model representation. The feature extraction process obtains simplified high-dimensional features by analyzing a large amount of raw data, while the data mapping process converts these features into parameters or state variables of the digital twin model, thereby achieving the unification of virtual and real device states.

[0049] In the embodiment of the present invention, in S3.3, based on the mapped data, the following formula is used to perform status detection: ; The threshold is a preset boundary value between a normal state and an abnormal state; In S3.4, The detected anomalies are classified using the following formula: ; Among them, threshold 1, threshold 2, threshold 3, threshold 4, etc. are preset abnormality type judgment values.

[0050] In one possible implementation, in S3.3, by using the mapped data , the operation and maintenance system will determine whether the current status of the device is "normal" or "abnormal". The specific operation is to calculate the sum of the mapping data: ; In this step, the operation and maintenance system will sum all the mapped data. If the sum is lower than the preset threshold, the device status is judged to be normal; if the sum exceeds the threshold, it is judged to be abnormal. The threshold is set based on the historical operation data of the device or expert experience, and the purpose is to determine under what conditions the device exhibits an abnormal state. This step is the key to the entire operation and maintenance management process because it can quickly identify the health status of the device and discover potential faults in a timely manner.

[0051] Furthermore, once the device status is determined to be abnormal, the next task is to classify the abnormality in detail. Step S3.4 uses a specific formula to classify the abnormality type:

[0052] This process is based on the specific eigenvalues The anomalies are classified based on the information provided by the operator and their relationship with the preset thresholds. Different types of anomalies may indicate different failure modes or equipment states. For example, changes in certain feature values ​​may indicate excessive temperature, while others may indicate excessive equipment vibration. In this way, maintenance personnel can take different countermeasures for different types of anomalies.

[0053] The status detection in S3.3 determines whether the device status is normal based on the sum of the device mapping data, while the abnormality classification in S3.4 further refines the type of abnormality. The connection between the two is that S3.3 first provides a rough judgment (normal or abnormal) for the device status, and S3.4 further refines and analyzes the specific type of abnormality after detecting the abnormality, thereby providing more precise guidance for subsequent treatment measures. The classification of abnormal status can help operation and maintenance personnel accurately identify the problem and take targeted treatment plans.

[0054] The overall status of the device (normal or abnormal) is obtained through S3.3. If the device is judged to be abnormal, S3.4 will enter a deeper level of analysis, that is, classify the abnormality by judging the characteristic value. The key to this process is the setting of the threshold. The reasonable selection of the threshold can effectively distinguish different types of abnormalities and reduce the possibility of misjudgment or missed judgment.

[0055] In the embodiment of the present invention, the specific steps of S4 include: 4.1: Collect a large amount of pre-processed historical data and real-time data of the operation and maintenance system, including temperature, current, voltage, vibration frequency and energy consumption; 4.2: Construct a long short-term memory network based on the attention mechanism, including the input layer, LSTM layer, attention mechanism layer and output layer. The input layer receives preprocessed data in the format of [time step, number of features], where the time step represents the length of the time series and the number of features represents the number of features in each time step. LSTM layer: contains multiple LSTM units, each of which consists of an input gate, an output gate, and a forget gate. LSTM units are used to capture long-term dependencies in time series data. The parameters of the LSTM layer are set as follows: Input dimension: number of features; Hidden layer dimension: 128; Time steps: 100; Attention mechanism layer: Introducing the attention mechanism to highlight the data of key time steps in the time series in a weighted manner; Output layer: Use a fully connected layer to map the weighted hidden state to the fault prediction result. The activation function of the output layer is the sigmoid function, and the output range is [0, 1], which represents the probability of fault occurrence. The output data is the fault prediction result, including fault type, fault location, and fault time. 4.3: Establish training sets and test sets to train and test the long short-term memory network based on the attention mechanism.

[0056] In a possible implementation, step 4.1 mainly collects historical data and real-time data in the operation and maintenance system, which include the temperature, current, voltage, vibration frequency, and energy consumption of the equipment. The goal of data collection is to obtain the operating status of the equipment at different time points, and use these data to provide input for fault prediction. All collected data needs to be preprocessed, and the processed data format is a time series, where the time step represents the length of the time series, and the number of features represents the number of features in each time step. By collecting and preprocessing data, it is ensured that the data input to the model is of high quality and suitable for time series prediction.

[0057] The data collection and preprocessing in step 4.1 provide the necessary input for the LSTM network in step 4.2. The preprocessed data can be input into the LSTM model in the form of a time series, so that the model can effectively learn the operating status and potential fault information of the equipment at different times. The quality of the data directly affects the training effect of the LSTM network. Therefore, high-quality data collection and processing is a prerequisite for the model to successfully predict faults.

[0058] In step 4.2, a long short-term memory network (LSTM) based on the attention mechanism is constructed. The core of the LSTM network is to capture the long-term dependencies in time series data. In particular, for the changing trends of the equipment's operating status and fault data, the LSTM network can effectively predict future fault trends. The LSTM layer processes input data and memory status through its input gate, output gate, and forget gate. The LSTM settings here include input dimension (number of features), hidden layer dimension (128), and time step (100). These parameters help LSTM better capture the potential time patterns during equipment operation.

[0059] By introducing the attention mechanism, the model can automatically learn and highlight the time steps that are most influential for fault prediction, thereby avoiding information loss or redundancy when processing long time series. The role of the attention mechanism is to weight the different time steps of the input sequence, adjust the weights according to their contribution to the prediction results, and help the model pay more attention to the data at those critical moments.

[0060] After being processed by LSTM and the attention mechanism, the output layer uses a fully connected layer to map the weighted hidden state to the fault prediction result. The activation function of the output layer is the sigmoid function, and its output range is [0, 1], which represents the probability of a fault. The output results include specific information such as fault type, fault location, and fault time.

[0061] In 4.2, the LSTM layer and the attention mechanism layer work together. The LSTM layer is mainly responsible for capturing the time dependency in the data, while the attention mechanism can give higher weights to important time steps according to the contribution of historical data. This synergy enhances the model's ability to understand time series and can effectively cope with the challenges of long time series in equipment failure prediction.

[0062] In step 4.3, the collected historical data needs to be divided into a training set and a test set. The training set is used to train the LSTM model, and the test set is used to verify the prediction ability of the model. Through the training process, the LSTM model is able to adjust its internal parameters to better adapt to the operation mode and failure mode of different devices. During the training process, the network optimizes its prediction accuracy by minimizing the loss function. The cross entropy loss function is usually used to evaluate the accuracy of fault prediction.

[0063] In 4.3, the establishment of training sets and test sets enables the LSTM model to conduct targeted learning and evaluation. The training set helps the model capture the operating mode of the equipment, while the test set verifies the actual prediction ability of the model. Through continuous training and adjustment, the LSTM model can improve the accuracy of its fault prediction, thereby providing more reliable prediction results for operation and maintenance personnel.

[0064] In the embodiment of the present invention, in S4.3, the training set includes historical operation and maintenance data, and the data volume is , the data format is , each piece of data Includes historical operating parameters and fault tags of the equipment ; The test set includes real-time operation and maintenance data, the data volume is , the data format is: , each piece of data Includes real-time operating parameters and fault tags of the equipment .

[0065] In a possible implementation, the training set includes historical operation and maintenance data, the amount of which is , the data format is , each piece of data Includes historical operating parameters and fault tags of the equipment The historical operating parameters of the device may include temperature, vibration, current, power consumption and other characteristics, which reflect the status of the device at different time points. It is labeled based on the actual fault conditions of the equipment, usually a binary value (for example, 0 indicates no fault, 1 indicates a fault) or a multi-class value (indicating different types of faults). The historical data in the training set provides enough samples for the LSTM model to help the model learn the different characteristic patterns of the equipment in normal operation and fault state, so as to effectively predict future equipment failures.

[0066] The test set includes real-time operation and maintenance data, the data volume is , the data format is: , each piece of data Includes real-time operating parameters and fault tags of the equipment . This data is collected when the equipment is currently running, and is usually provided by the equipment's real-time monitoring system. The purpose of the test set is to verify the performance of the trained LSTM model in actual applications, especially its predictive ability on unknown data (real-time data). Since real-time data usually fluctuates, the model needs to have strong generalization capabilities to make accurate fault predictions even in unfamiliar equipment states.

[0067] The training process is done by inputting the data in the training set Learning is performed by minimizing the loss function (such as the cross entropy loss function) and adjusting the parameters of the LSTM network (such as weights and biases) so that the model can fit the historical operation and maintenance data and learn the complex relationship between normal and fault states. After training, the model is verified using the test set to evaluate its prediction effect on real-time data. By comparing the prediction results with the actual fault labels The difference between them can be used to evaluate the performance of the model and make further optimization accordingly.

[0068] In the embodiment of the present invention, in S5, the following steps are included: S5.1: The cloud platform generates an optimization management plan based on the fault prediction results and system operation status; S5.2: The optimization management plan includes system maintenance plan, resource scheduling plan and fault prevention measures; S5.3: The specific system maintenance plan includes regular inspection and maintenance of equipment, replacement of aging or damaged parts, and ensuring the normal operation of the equipment; S5.4: The resource scheduling scheme dynamically adjusts resource allocation and optimizes system performance based on the system operation status and prediction results; S5.5: Fault prevention measures include strengthening monitoring and alarms, and promptly handling possible fault hazards to avoid the occurrence of faults.

[0069] In one possible implementation, in S5.1, the cloud platform receives the results from the fault prediction model and the current system operation status of the equipment, combines real-time data with historical data for analysis, and generates a globally optimized management plan. This plan aims to prevent and minimize equipment failures, and covers subsequent maintenance, resource scheduling, and fault prevention. The role of the cloud platform is to centrally process data from multiple devices and systems, and to analyze and derive the optimal management plan through powerful computing power.

[0070] In S5.2, the generated optimization management plan includes three main parts: system maintenance plan, resource scheduling plan and fault prevention measures. Each part is formulated based on the fault prediction results and the operating status of the equipment to ensure efficient and timely intervention and adjustment during the operation of the system.

[0071] The system maintenance plan is in S5.3, which specifically includes regular inspection and maintenance of equipment, replacement of aging or damaged parts, and ensuring the continuous operation of the equipment. Through real-time monitoring of equipment status and analysis of historical failure data, the maintenance plan can conduct early inspection and replacement of key parts of the equipment to reduce the risk of equipment aging or failure. The formulation of a maintenance plan can not only extend the service life of the equipment, but also reduce the impact of sudden failures on production.

[0072] In S5.4, the resource scheduling scheme is dynamically adjusted according to the system's operating status and prediction results. The cloud platform can monitor the equipment load and resource usage in real time, and dynamically adjust the allocation of manpower, material resources, equipment and funds according to changes in the system's operating status (such as equipment load, usage frequency, fault prediction, etc.). By optimizing resource allocation, the overall system performance can be effectively improved, ensuring that key equipment can operate smoothly under high load conditions and avoiding equipment overload or shutdown due to insufficient resources.

[0073] In S5.5, fault prevention measures are to avoid faults and enhance the system's early warning capabilities. The cloud platform promptly strengthens equipment monitoring and alarms based on real-time monitoring of operating data and fault prediction results, ensuring that operation and maintenance personnel can respond promptly at the early stages of equipment failure hazards to prevent further deterioration of the failure. For example, when the equipment shows signs of abnormal temperature or current fluctuations, the system will automatically issue an early warning and suggest that the operation and maintenance personnel conduct further inspections or adjustments.

[0074] Accordingly, the embodiment of the present invention also provides an operation and maintenance management system based on digital twins, which is used to run the operation and maintenance management method based on digital twins in the embodiment of the present invention, including a data acquisition module, a data preprocessing module, a digital twin module, a fault prediction module, an optimization management module, a cloud platform and an operation and maintenance terminal, wherein the data acquisition module, the data preprocessing module, the digital twin module, the fault prediction module, the optimization management module and the operation and maintenance terminal are all electrically connected to the cloud platform; The data acquisition module includes sensors and data collectors. The sensors are used to collect the structural data and operation data of the operation and maintenance system in real time, including equipment status, environmental parameters and system performance indicators; the data collectors are used to perform preliminary processing on the data collected by the sensors and transmit them to the cloud platform via wired or wireless means. The cloud platform includes a data storage unit, a data processing unit, a communication unit, and a management unit. The data storage unit is used to store data collected by the data acquisition module and preprocessed data. The data processing unit is used to execute the functions of the data preprocessing module, the digital twin module, and the fault prediction module. The communication unit is used to transmit data with the data acquisition module and the operation and maintenance terminal. The management unit is used to manage various functional units of the cloud platform. The operation and maintenance terminal includes a display unit, an operation unit and a communication unit. The display unit is used to display the working status and optimization management plan of the operation and maintenance management system, the operation unit is used for operation and management by operation and maintenance personnel, and the communication unit is used to transmit data with the cloud platform.

[0075] The present invention covers any substitution, modification, equivalent method and scheme made on the essence and scope of the present invention. In order to make the public have a thorough understanding of the present invention, specific details are described in detail in the following preferred embodiments of the present invention, but those skilled in the art can fully understand the present invention without the description of these details. In addition, in order to avoid unnecessary confusion about the essence of the present invention, well-known methods, processes, procedures, components and circuits are not described in detail.

[0076] The above is only a preferred embodiment of the present invention. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the principle of the present invention. These improvements and modifications should also be regarded as the scope of protection of the present invention.

Claims

1. The operation and maintenance management method based on digital twins is characterized by: The following steps are involved: S1: Build the corresponding digital twin model by collecting the structural data and operation data of the actual operation and maintenance system; S2: collects the operation data of the operation and maintenance system in real time and transmits it to the cloud platform, and pre-processes the real-time transmitted operation data in the cloud platform; S3: Use the digital twin model to analyze the pre-processed data in real time to detect system operation status and abnormal conditions; S4: Based on historical data and real-time data, a fault prediction model is constructed using a machine learning algorithm to predict possible future faults. The input of the fault prediction model is historical data and real-time data, and the output is the fault prediction result. S5: Generate an optimization management plan based on the fault prediction results and system operation status, including system maintenance, resource scheduling and fault prevention measures.

2. The operation and maintenance management method based on digital twin according to claim 1 is characterized in that: S1 specifically includes the following steps: S1.1: Collect structural data and operation data of the operation and maintenance system through sensors and data acquisition equipment, including equipment status, environmental parameters and system performance indicators; S1.2: Preprocess the collected data, including data cleaning, normalization and format conversion. Data cleaning is used to remove invalid data and outliers in the data, normalization is used to normalize the data to the range of [0, 1], and format conversion is used to convert the preprocessed data into a unified format; S1.3: Build a physical model based on the structural data of the operation and maintenance system, and build a behavioral model based on the operating data of the operation and maintenance system. Combine the physical model and the behavioral model to form a digital twin model.

3. The operation and maintenance management method based on digital twin according to claim 2 is characterized in that: In S1.3, the physical model includes the geometric parameters, material properties, and connection relationships of the equipment; The geometric parameters of the device are described using the following formula: ;in, For the The geometric parameters of each device, including the length, width, height, weight, material properties, and connection relationships of the device; The behavior model describes the operating status and performance indicators of the device. The following formula is used to describe the operating status of the device: ;in, For the The operating status of each device, including the operating temperature, operating current, operating voltage, vibration frequency and energy consumption of the device; The physical model and the behavioral model are combined to form a digital twin model, which is expressed using the following formula: ;in, It is a digital twin model, which is composed of physical models and behavioral models Comprehensive.

4. The operation and maintenance management method based on digital twin according to claim 1, characterized in that: S3 specifically includes the following steps: S3.1: Extract key features from the preprocessed data, including equipment status, environmental parameters, and system performance indicators; S3.2: Map the extracted features into the digital twin model; S3.3: Perform status detection based on the mapped data; S3.4: Classify detected anomalies.

5. The operation and maintenance management method based on digital twins according to claim 4 is characterized in that: In S3.1, the feature extraction process is expressed using the following formula: ;in, For the Features In S3.2, the extracted features are mapped into the digital twin model, and the data mapping process is represented by the following formula: ;in, For the Mapping data.

6. The operation and maintenance management method based on digital twins according to claim 5 is characterized in that: In S3.3, based on the mapped data, the following formula is used for status detection: ； The threshold is a preset boundary value between a normal state and an abnormal state; In S3.4, The detected anomalies are classified using the following formula: ； Among them, threshold 1, threshold 2, threshold 3, threshold 4, etc. are preset abnormality type judgment values.

7. The operation and maintenance management method based on digital twin according to claim 1 is characterized in that: The specific steps of S4 include: 4.1: Collect a large amount of pre-processed historical data and real-time data of the operation and maintenance system, including temperature, current, voltage, vibration frequency and energy consumption; 4.2: Construct a long short-term memory network based on the attention mechanism, including the input layer, LSTM layer, attention mechanism layer and output layer. The input layer receives preprocessed data in the format of [time step, number of features], where the time step represents the length of the time series and the number of features represents the number of features in each time step. LSTM layer: contains multiple LSTM units, each of which consists of an input gate, an output gate, and a forget gate. LSTM units are used to capture long-term dependencies in time series data. The parameters of the LSTM layer are set as follows: Input dimension: number of features; Hidden layer dimension: 128; Time steps: 100; Attention mechanism layer: Introducing the attention mechanism to highlight the data of key time steps in the time series in a weighted manner; Output layer: Use a fully connected layer to map the weighted hidden state to the fault prediction result. The activation function of the output layer is the sigmoid function, and the output range is [0, 1], which represents the probability of fault occurrence. The output data is the fault prediction result, including fault type, fault location, and fault time. 4.3: Establish training sets and test sets to train and test the long short-term memory network based on the attention mechanism.

8. The operation and maintenance management method based on digital twin according to claim 7 is characterized in that: In S4.3, the training set includes historical operation and maintenance data, and the data volume is , the data format is , each piece of data Includes historical operating parameters and fault tags of the equipment ; The test set includes real-time operation and maintenance data, the data volume is , the data format is , each piece of data Includes real-time operating parameters and fault tags of the equipment .

9. The operation and maintenance management method based on digital twin according to claim 1, characterized in that: In S5, the following steps are included: S5.1: The cloud platform generates an optimization management plan based on the fault prediction results and system operation status; S5.2: The optimization management plan includes system maintenance plan, resource scheduling plan and fault prevention measures; S5.3: The specific system maintenance plan includes regular inspection and maintenance of equipment, replacement of aging or damaged parts, and ensuring the normal operation of the equipment; S5.4: The resource scheduling scheme dynamically adjusts resource allocation and optimizes system performance based on the system operation status and prediction results; S5.5: Fault prevention measures include strengthening monitoring and alarms, and promptly handling possible fault hazards to avoid the occurrence of faults.

10. An operation and maintenance management system based on digital twins, used to run the operation and maintenance management method based on digital twins according to claims 1-9, characterized in that: It includes a data acquisition module, a data preprocessing module, a digital twin module, a fault prediction module, an optimization management module, a cloud platform and an operation and maintenance terminal, wherein the data acquisition module, the data preprocessing module, the digital twin module, the fault prediction module, the optimization management module and the operation and maintenance terminal are all electrically connected to the cloud platform; The data acquisition module includes sensors and data collectors. The sensors are used to collect the structural data and operation data of the operation and maintenance system in real time, including equipment status, environmental parameters and system performance indicators; the data collectors are used to perform preliminary processing on the data collected by the sensors and transmit them to the cloud platform via wired or wireless means. The cloud platform includes a data storage unit, a data processing unit, a communication unit, and a management unit. The data storage unit is used to store data collected by the data acquisition module and preprocessed data. The data processing unit is used to execute the functions of the data preprocessing module, the digital twin module, and the fault prediction module. The communication unit is used to transmit data with the data acquisition module and the operation and maintenance terminal. The management unit is used to manage various functional units of the cloud platform. The operation and maintenance terminal includes a display unit, an operation unit and a communication unit. The display unit is used to display the working status and optimization management plan of the operation and maintenance management system, the operation unit is used for operation and management by operation and maintenance personnel, and the communication unit is used to transmit data with the cloud platform.

Images