An industrial internet of things-based remote operation and maintenance system and method for petrochemical equipment

Abstract

This application provides a remote operation and maintenance system and method for petrochemical equipment based on the Industrial Internet of Things (IIoT). It collects operational monitoring data from petrochemical equipment and determines the explicit behavioral feature vectors corresponding to the equipment. Through feature learning using these explicit behavioral feature vectors, it determines the equipment health indicators. It collects coolant information corresponding to each reactor node in the petrochemical equipment and performs delayed correlation analysis on this information to obtain implicit behavioral feature vectors. Based on these implicit behavioral feature vectors, it performs implicit risk analysis. The remote operation and maintenance center then executes corresponding remote operation and maintenance strategies based on the implicit risk information. This system enables risk identification through the implicit behavioral analysis of petrochemical equipment, achieving real-time control of implicit risks and improving the operational safety of petrochemical equipment.

Description

Technical Field

[0001] This application relates to the field of remote operation and maintenance technology, and more specifically, to a remote operation and maintenance system and method for petrochemical equipment based on the Industrial Internet of Things. Background Technology

[0002] Petrochemical plants operate under high temperature, high pressure, and complex conditions for extended periods. The quality of equipment manufacturing, structural rationality, and the level of intelligent operation management directly impact the safety and stability of the plant. In actual production, some equipment accidents are not caused by sudden extreme conditions, but rather by the long-term accumulation of factors such as quality defects inherited from the manufacturing stage, unreasonable structural design, and inadequate operation management. These accidents are often characterized by their high degree of concealment and long evolution cycle; once a failure occurs, the impact is widespread and the losses are significant.

[0003] In existing technologies, maintenance is often carried out through regular inspections and operations by maintenance personnel. For example, during equipment operation, a system of fixed personnel, fixed time, fixed location, fixed route, and fixed standards is used for routine inspections by operators, regular inspections by professionals, and precise inspections by maintenance technicians. This system involves checking the operating data of each piece of equipment, analyzing and maintaining and repairing any abnormal operating data. This manual approach has the problems of high maintenance costs and late detection of equipment failures. Summary of the Invention

[0004] This application provides a remote operation and maintenance system and method for petrochemical equipment based on the Industrial Internet of Things (IIoT). It can identify risks through the analysis of the implicit behavior of petrochemical equipment, realize real-time control of implicit risks, and improve the operational safety of petrochemical equipment.

[0005] In the first aspect, this application provides a remote operation and maintenance method for petrochemical equipment based on the Industrial Internet of Things. This method can be executed by a network device, or it can be executed by a chip configured in the network device. This application does not limit the method in this regard.

[0006] Specifically, the method includes:

[0007] Data on the operation and monitoring of petrochemical equipment is collected through IoT sensors;

[0008] The operation monitoring data is standardized to obtain the operation status feature dataset of the petrochemical equipment. Multidimensional feature extraction is performed based on the operation status feature dataset to determine the explicit behavior feature vector corresponding to the petrochemical equipment.

[0009] The equipment health index of the petrochemical equipment is determined by performing feature learning through the explicit behavioral feature vectors.

[0010] When the health index of the equipment is lower than the preset threshold, the coolant information of each reactor node in the petrochemical equipment is collected, and the coolant information of each reactor node is subjected to delay correlation analysis according to the preset reactor delay time to obtain the implicit behavior feature vector of the petrochemical equipment.

[0011] Latent risk analysis is performed based on the latent behavioral feature vector, and latent risk information is sent to the remote operation and maintenance center of the petrochemical equipment; the remote operation and maintenance center executes the corresponding remote operation and maintenance strategy based on the latent risk information.

[0012] In conjunction with the first aspect, in certain implementations of the first aspect, the standardization processing of the operation monitoring data to obtain the operation status feature dataset of the petrochemical equipment specifically includes:

[0013] Acquire initial sensor data for each dimension of the operational monitoring data;

[0014] Outlier filtering and correction are performed on the initial sensor data for each dimension. The initial sensor data for each dimension are then timestamped and mapped to the same time series window to form a synchronous monitoring data matrix.

[0015] The synchronous monitoring data matrix is ​​normalized to obtain the operating status feature dataset of the petrochemical equipment.

[0016] In conjunction with the first aspect, in certain implementations of the first aspect, the extraction of multi-dimensional features based on the operational state feature dataset to determine the explicit behavioral feature vector corresponding to the petrochemical equipment specifically includes:

[0017] Obtain the aforementioned running status feature dataset;

[0018] Extract the statistical features corresponding to each dimension of the monitoring data in the operational status dataset;

[0019] The statistical features corresponding to each dimension of the monitoring data are combined to form an explicit behavioral feature vector.

[0020] In conjunction with the first aspect, in certain implementations of the first aspect, implicit risk analysis based on the implicit behavioral feature vector specifically includes:

[0021] Obtain the latent behavior feature vector and historical latent behavior feature vector, and predict the latent behavior of each reactor node based on the latent behavior feature vector and historical latent behavior feature vector.

[0022] Based on the prediction results of the latent behavior of each reactor node, latent risk information is sent to the remote operation and maintenance center of the petrochemical equipment. The latent risk information includes the latent risk level and the latent risk area.

[0023] In conjunction with the first aspect, in some implementations of the first aspect, a single hidden layer neural network is used to classify the explicit behavior feature vector; based on the classification results, a health degree mapping is performed to determine the equipment health degree index of the petrochemical equipment.

[0024] In conjunction with the first aspect, in some implementations of the first aspect, a PLC acquisition module is set as a data acquisition unit to perform signal conditioning and data acquisition processing on the raw signals output by the Internet of Things sensor.

[0025] In conjunction with the first aspect, in some implementations of the first aspect, the operational monitoring data includes temperature data, pressure data, vibration data, and current data of the petrochemical equipment.

[0026] Secondly, this application provides a remote operation and maintenance system for petrochemical equipment based on the Industrial Internet of Things, which includes an operation and maintenance control unit, the operation and maintenance control unit comprising:

[0027] The operation monitoring module is used to collect operation monitoring data of petrochemical equipment through IoT sensors;

[0028] The data analysis module is used to standardize the operation monitoring data to obtain the operation status feature dataset of the petrochemical equipment, and to perform multi-dimensional feature extraction based on the operation status feature dataset to determine the explicit behavior feature vector corresponding to the petrochemical equipment.

[0029] The data analysis module is also used to perform feature learning through the explicit behavioral feature vector to determine the equipment health index of the petrochemical equipment.

[0030] The data analysis module is also used to collect coolant information corresponding to each reactor node in the petrochemical equipment when the equipment health index is lower than a preset threshold, and to perform delay correlation analysis on the coolant information corresponding to each reactor node according to the preset reactor delay time to obtain the implicit behavior feature vector corresponding to the petrochemical equipment.

[0031] The operation and maintenance management module is used to perform implicit risk analysis based on the implicit behavioral feature vector and send implicit risk information to the remote operation and maintenance center of the petrochemical equipment; the remote operation and maintenance center executes the corresponding remote operation and maintenance strategy based on the implicit risk information.

[0032] Thirdly, this application provides a computer terminal device, which includes a memory and a processor. The memory stores code, and the processor is configured to acquire the code and execute the above-described method for remote operation and maintenance of petrochemical equipment based on the Industrial Internet of Things.

[0033] Fourthly, this application provides a computer-readable storage medium storing at least one computer program, which is loaded and executed by a processor to perform the operations described above in a remote operation and maintenance method for petrochemical equipment based on the Industrial Internet of Things.

[0034] The technical solutions provided by the embodiments disclosed in this application have the following beneficial effects:

[0035] This application provides a remote operation and maintenance system and method for petrochemical equipment based on the Industrial Internet of Things. First, operational monitoring data of the petrochemical equipment is collected. This data is then standardized to obtain an operational status feature dataset. Multi-dimensional feature extraction is performed on this dataset to determine the explicit behavioral feature vectors corresponding to the petrochemical equipment. Feature learning is then performed using these explicit behavioral feature vectors to determine the equipment health index. When the equipment health index is below a preset threshold, coolant information for each reactor node in the petrochemical equipment is collected. Delay correlation analysis is performed on the coolant information for each reactor node based on a preset reactor delay time to obtain the implicit behavioral feature vectors corresponding to the petrochemical equipment. Implicit risk analysis is then performed based on these implicit behavioral feature vectors, and implicit risk information is sent to the remote operation and maintenance center of the petrochemical equipment. The remote operation and maintenance center then executes the corresponding remote operation and maintenance strategy based on the implicit risk information.

[0036] Therefore, this application utilizes IoT sensors to collect real-time operational monitoring data of petrochemical equipment, forming a quantifiable dataset of operational status features and explicit behavioral feature vectors. This enables continuous, real-time monitoring of equipment operation, overcoming the limitations of traditional manual inspections which can only monitor at fixed times and locations. By using explicit behavioral feature vectors for feature learning, equipment health indicators are obtained, allowing for the quantification of equipment operation status and comparison with preset thresholds. This enables early identification of potential equipment anomalies. By collecting coolant information from each reactor node and combining it with reactor delay time for delay correlation analysis, implicit behavioral feature vectors are obtained. These vectors can identify potential anomalies and long-term accumulated risks that traditional explicit monitoring cannot detect. Implicit risk analysis is performed based on implicit behavioral feature vectors, generating implicit risk information (including risk level and risk area). The remote operation and maintenance center executes corresponding strategies based on the risk information, forming a complete remote intelligent operation and maintenance closed loop. This achieves real-time control and proactive intervention of implicit risks, improving the operational safety of petrochemical equipment. Attached Figure Description

[0037] Figure 1 This is an exemplary flowchart illustrating a remote operation and maintenance method for petrochemical equipment based on the Industrial Internet of Things, according to some embodiments of this application.

[0038] Figure 2 This is a schematic diagram of the operation and maintenance control unit according to some embodiments of this application;

[0039] Figure 3 This is a schematic diagram of the structure of a computer terminal device that implements a remote operation and maintenance method for petrochemical equipment based on the Industrial Internet of Things, according to some embodiments of this application. Detailed Implementation

[0040] This application collects operational monitoring data from petrochemical equipment; standardizes the operational monitoring data to obtain an operational status feature dataset for the petrochemical equipment; extracts multidimensional features based on the operational status feature dataset to determine the explicit behavioral feature vectors corresponding to the petrochemical equipment; performs feature learning through the explicit behavioral feature vectors to determine the equipment health index of the petrochemical equipment; when the equipment health index is lower than a preset threshold, collects coolant information corresponding to each reactor node in the petrochemical equipment; performs delay correlation analysis on the coolant information corresponding to each reactor node according to a preset reactor delay time to obtain the implicit behavioral feature vectors corresponding to the petrochemical equipment; performs implicit risk analysis based on the implicit behavioral feature vectors and sends implicit risk information to the remote operation and maintenance center of the petrochemical equipment; the remote operation and maintenance center executes corresponding remote operation and maintenance strategies based on the implicit risk information. This approach enables risk identification through the implicit behavioral analysis of petrochemical equipment, achieving real-time control of implicit risks and improving the operational safety of petrochemical equipment.

[0041] To better understand the above technical solutions, a detailed description of the solutions will be provided below in conjunction with the accompanying drawings and specific implementation methods. (Reference) Figure 1 The figure is an exemplary flowchart illustrating a remote operation and maintenance method for petrochemical equipment based on the Industrial Internet of Things (IIoT) according to some embodiments of this application. This remote operation and maintenance method 100 for petrochemical equipment based on the IIoT mainly includes the following steps:

[0042] In step S101, operational monitoring data of petrochemical equipment is collected through IoT sensors.

[0043] Preferably, in some embodiments, the operational monitoring data includes at least the temperature data, pressure data, vibration data, and current data of the petrochemical equipment.

[0044] In specific implementation, the data acquisition process for the operation monitoring can be achieved by deploying various types of industrial IoT sensors at predetermined operating locations of the petrochemical equipment. Temperature sensors can be deployed in locations such as the equipment housing, bearing areas, and the outer wall of the reaction chamber to detect temperature changes during equipment operation in real time. Pressure sensors can be deployed in locations such as fluid pipelines, reaction chambers, and pressure vessels to monitor pressure changes inside the equipment or pipelines. Vibration sensors are installed on the equipment base, bearing housing, or motor housing to collect vibration signals generated during equipment operation. Current sensors are installed in the equipment power supply line or motor control circuit to detect current changes during equipment operation. Optionally, in some embodiments, some IoT sensors in this application are connected to the local data acquisition unit via analog / digital interfaces.

[0045] Preferably, in some embodiments, this application sets a PLC acquisition module as a data acquisition unit to perform signal conditioning and data acquisition processing on the raw signals output by the IoT sensors. This includes filtering, amplifying, and converting analog signals to digital signals, and synchronously sampling various monitoring signals according to a preset sampling period to form corresponding equipment operation monitoring data. The acquired monitoring data is preliminarily processed by an embedded edge computing module, including timestamp marking, data standardization, and outlier filtering, thereby generating a structured equipment operation data packet. After data encapsulation, the equipment operation monitoring data is uploaded to a remote operation and maintenance platform through an industrial IoT communication network.

[0046] In specific implementation, the communication method adopted by the industrial IoT communication network in this application may include: wired industrial Ethernet, 5G communication network, NB-IoT network or LoRa wireless communication method, and this application does not limit it.

[0047] In step S102, the operation monitoring data is standardized to obtain the operation status feature dataset of the petrochemical equipment. Based on the operation status feature dataset, multi-dimensional feature extraction is performed to determine the explicit behavior feature vector corresponding to the petrochemical equipment.

[0048] Preferably, in some embodiments, the standardization processing of the operation monitoring data to obtain the operating status feature dataset of the petrochemical equipment specifically includes:

[0049] Acquire initial sensor data for each dimension of the operational monitoring data;

[0050] Outlier filtering and correction are performed on the initial sensor data for each dimension. The initial sensor data for each dimension are then timestamped and mapped to the same time series window to form a synchronous monitoring data matrix.

[0051] The synchronous monitoring data matrix is ​​normalized to obtain the operating status feature dataset of the petrochemical equipment.

[0052] In specific implementation, the operation monitoring data includes at least four dimensions of data: temperature data, pressure data, vibration data, and current data of the petrochemical equipment. In specific applications, the detection dimensions of the operation monitoring data can also be adjusted according to the type of petrochemical equipment. This application does not limit this. After obtaining the operation monitoring data of the petrochemical equipment, the remote operation and maintenance platform needs to perform standardization processing on the operation monitoring data. Standardization processing is used to eliminate the differences in the dimensions, sampling frequency, and sensor detection disturbances of data collected by different sensors.

[0053] The standardization process begins by identifying and removing outliers from the collected temperature, pressure, vibration, and current data. Specifically, this can be achieved using methods commonly found in existing technologies, such as statistical threshold detection and box plots, to identify data points exceeding the normal operating range. The outliers are then corrected using the mean of adjacent time windows or linear interpolation to obtain a continuous monitoring data sequence. Next, time series alignment is performed on the data from different types of sensors. Since the sampling periods of various sensors may differ, all monitoring data are resampled according to a preset sampling time interval. Temperature, pressure, vibration, and current data are then uniformly mapped to the same time series window using timestamp alignment, forming a synchronous monitoring data matrix. Finally, the aligned data is standardized to eliminate dimensional differences between different monitoring parameters. Specifically, the Z-score standardization method, commonly used in data normalization, can be employed, which will not be elaborated upon further in this application.

[0054] Preferably, in some embodiments, the process of extracting multidimensional features based on the operational status feature dataset to determine the explicit behavioral feature vector corresponding to the petrochemical equipment specifically includes:

[0055] Obtain the aforementioned running status feature dataset;

[0056] Extract the statistical features corresponding to each dimension of the monitoring data in the operational status dataset;

[0057] The statistical features corresponding to each dimension of the monitoring data are combined to form an explicit behavioral feature vector.

[0058] This process involves multidimensional feature extraction of multidimensional monitoring data within each time window to extract explicit feature parameters that characterize the changing patterns of equipment operation behavior. Specifically, for slowly changing parameters such as temperature and pressure data, statistical features can be extracted, including statistical indicators such as mean, maximum, minimum, standard deviation, and rate of change. For dynamic signals such as vibration data, time-domain features can be extracted, such as vibration characteristic parameters like root mean square value, peak factor, kurtosis, and skewness. For current data, load change characteristics can be extracted, such as characteristic indicators like mean current, current fluctuation amplitude, and current change slope. After extracting various feature parameters, the temperature, pressure, vibration, and current feature parameters are fused and combined according to a preset feature order to construct a multidimensional equipment behavior feature vector, thereby forming an explicit behavior feature vector that directly reflects the changing patterns of equipment operation status.

[0059] In step S103, feature learning is performed using the explicit behavioral feature vector to determine the equipment health index of the petrochemical equipment.

[0060] Optionally, in some embodiments, a single hidden layer neural network can be used to classify the explicit behavioral feature vector; based on the classification results, a health status mapping can be performed to determine the equipment health index of the petrochemical equipment.

[0061] The following is a specific embodiment of this application using a single hidden layer neural network to perform feature classification on the explicit behavior feature vector:

[0062] The explicit behavioral feature vector is input into a single hidden layer neural network in the form of a data vector for training and cluster analysis. The hidden layer of the single hidden layer neural network contains multiple activation function nodes for feature extraction and classification training, and the output layer is used to output the classification results of the equipment health status. The dimensions of the explicit feature vector may include temperature statistical features, pressure statistical features, vibration time-domain features, current load features, and other parameters that can reflect the operating status of the equipment.

[0063] In the specific implementation process, multiple equipment operating status samples and corresponding manual health scores are first prepared as a training set. Each explicit feature vector is input into the single hidden layer neural network for training. The hidden layer performs nonlinear mapping on the input features through an activation function to identify and classify different operating status patterns. The output layer generates classification results based on the mapping results of the hidden layer and compares the classification results with the manual scores to quantitatively evaluate the equipment health status. During training, when the correlation between the health classification results output by the single hidden layer neural network and the manual scores is lower than a preset threshold, iterative optimization can be performed by adjusting the parameters in the hidden layer activation function or modifying the weight coefficients until the correlation between the clustering results and the manual scores reaches a preset standard. Simultaneously, after training, the mapping relationship between explicit behavioral feature vectors and equipment health indicators can be obtained for subsequent real-time operating status monitoring and health assessment. When the equipment operating status is abnormal or the explicit behavioral feature vectors deviate from the normal pattern, the mapping relationship can output equipment health indicators in real time to assist maintenance personnel in making early warnings and maintenance decisions.

[0064] In some other embodiments, during the process of determining the equipment health index of the petrochemical equipment through feature learning using the explicit behavioral feature vector, multiple equipment operating status samples and corresponding manual health score results can be used as training sets. Other model training methods can be adopted, such as inputting the explicit behavioral feature vector into a multi-layer feedforward neural network or a convolutional neural network, performing nonlinear mapping and abstract learning on the features through multiple hidden layers to extract higher-order operating rule features, and generating equipment health indexes at the output layer. Optionally, support vector machines or mean clustering methods can also be used to perform regression classification on the training set for model training, and then the equipment health index can be determined through the trained model. This application does not limit this.

[0065] In step S104, when the equipment health index is lower than a preset threshold, the coolant information corresponding to each reactor node in the petrochemical equipment is collected, and the coolant information corresponding to each reactor node is subjected to delay correlation analysis according to the preset reactor delay time to obtain the implicit behavior feature vector corresponding to the petrochemical equipment.

[0066] It should be noted that the reactor node mentioned in this application is used to represent the location of an independently monitored functional unit in a reactor system. For petrochemical equipment with multiple reactors, such as petrochemical equipment with series reactor devices or multi-stage reaction systems, the reactor node refers to the data acquisition unit of the reactor equipment used for the reaction process in the industrial Internet of Things monitoring network. Each reactor node corresponds to coolant monitoring information, which is used to characterize the heat exchange state between the corresponding reactor and the coolant during operation. This application can also be adapted to petrochemical equipment with a single reactor. In this case, the reactor node is set at different monitorable locations in the single reactor, such as the reaction zone, jacket cooling zone, feed pipeline, discharge pipeline, and circulating heat exchange loop, and each monitoring point is regarded as a reactor node. Upstream reactor nodes, intermediate reactor nodes, and downstream reactor nodes can also be constructed according to the process flow or equipment structure relationship. This application will not elaborate on this further.

[0067] Preferably, in some embodiments, during the process of collecting coolant information corresponding to each reactor node in the petrochemical equipment, temperature sensors, flow sensors, and pressure sensors installed on the cooling pipes of each reactor node are used to collect the coolant information of the corresponding reactor node. The coolant information includes at least the coolant inlet temperature, coolant outlet temperature, coolant flow rate, and coolant pressure. The coolant information is continuously recorded according to timestamps to form coolant time series data corresponding to each reactor node.

[0068] Preferably, in some embodiments, the latent behavioral feature vector corresponding to the petrochemical equipment is obtained by performing a delay correlation analysis on the coolant information corresponding to each reactor node based on a preset reactor delay time, specifically including:

[0069] Obtain the coolant information corresponding to each reactor node. For any reactor node, determine the node cooling entropy corresponding to multiple time periods based on the coolant information corresponding to the reactor node, and form a node cooling entropy sequence according to the time sequence.

[0070] Based on the reactor delay time corresponding to the reactor node and the other reactor nodes respectively, the node cooling entropy sequence corresponding to the reactor node and the other reactor nodes is correlated and analyzed to determine the cooling stability corresponding to the reactor node.

[0071] The latent behavioral feature vector is determined based on the cooling stability of each reactor node.

[0072] Preferably, in some embodiments, the coolant time series is periodically divided by a preset time window, and within each time period, the amount of cooling heat exchange within that time period is determined as the node cooling entropy based on the coolant inlet temperature, coolant outlet temperature, and coolant flow rate.

[0073] It should be noted that the node cooling entropy mentioned in this application is used as a heat characterization index reflecting the heat exchange through coolant within a corresponding time period. Specifically, the node cooling entropy = (coolant flow rate × coolant density × coolant specific heat capacity × coolant inlet and outlet temperature difference within the time period) × pressure correction coefficient. The pressure correction coefficient is preset based on the coolant pressure value corresponding to the reactor node, and is usually 1~0.8. The corresponding node cooling entropy values ​​are calculated for multiple time periods and arranged in chronological order to form the node cooling entropy sequence corresponding to the reactor node. The node cooling entropy sequence should be able to reflect the cooling load change trend of the reactor node in different operating stages and serve as the basis data for subsequent delay correlation analysis between reactor nodes.

[0074] It should be noted that the reactor delay time mentioned in this application is used to represent the time interval from a change in the operating state of one reactor node (such as temperature, coolant inlet temperature, heat release, etc.) to the stabilization of the cooling system or heat load response of another reactor node. The flow, transfer, or diffusion of materials and heat between different reactors will produce a certain time lag, which is the reactor delay time. For example, in a series reactor, an increase in material in the upstream reactor will lead to an increase in the heat absorption of the coolant in the downstream reactor, but this effect will not be immediately apparent. Instead, it requires the time delay of heat transfer through pipelines and material flow. The reactor delay time is calibrated based on the physical pipeline length and flow rate, reactor thermal inertia, cooling system response time, and the production process characteristics of the petrochemical equipment. It can also be calibrated by the flow time between different reactor nodes during the production of multiple sets of materials. This application will not elaborate on this further.

[0075] When the equipment is operating normally and the cooling system is stable, the temperature of the coolant decreases, and the flow and pressure change smoothly. The cooling entropy sequence shows regularity and small fluctuations. When the equipment has hidden faults (such as local blockage, decreased heat exchange efficiency, aging of pumps or valves, etc.), the temperature, flow, or pressure of the coolant will fluctuate abnormally, causing the cooling entropy sequence to fluctuate more or exhibit abnormal patterns. By collecting the cooling entropy of different reactor nodes and performing delayed correlation analysis, potential coupling anomalies or long-term accumulated risks between nodes can be discovered. The regularity and stability changes of cooling entropy can indirectly reflect equipment manufacturing defects, unreasonable structures, local blockages, and operational status problems. These are all hidden behavioral trends that are difficult to directly capture with explicit indicators in existing technologies for manual inspection. The cooling stability described in this application is used to reflect the stability of the heat dissipation behavior of a single reactor node relative to the whole system. The closer the value is to 1, the higher the stability of the node's cooling behavior is to the overall system. Preferably, in some embodiments, correlation analysis is performed on the node cooling entropy sequences corresponding to the reactor node and the other reactor nodes respectively to determine the cooling stability of the reactor node. Specifically, this includes:

[0076] Based on the corresponding reactor delay time, after aligning the cooling entropy sequence of the reactor node with the other nodes, the average correlation coefficient between the cooling entropy sequence of the reactor node and the cooling entropy sequences of the other reactor nodes is extracted and used as the cooling stability of the reactor node. Specifically, this can be achieved by extracting the Pearson correlation coefficient of the node cooling entropy sequence after delay alignment. The average value of the Pearson correlation coefficient between the cooling entropy sequence of the reactor node and the cooling entropy sequences of the other reactor nodes is used as the average correlation coefficient. In some other embodiments, other parameters that can achieve correlation learning can also be used for characterization. This application does not limit this. The cooling stability of each reactor node is determined, and a latent behavioral feature vector is constructed based on the cooling stability of each reactor node.

[0077] Optionally, in some embodiments, when the device health index is higher than a preset threshold, the remote operation and maintenance center performs scheduled device inspections according to the scheduled inspection cycle.

[0078] In step S105, implicit risk analysis is performed based on the implicit behavior feature vector, and implicit risk information is sent to the remote operation and maintenance center of the petrochemical equipment; the remote operation and maintenance center executes the corresponding remote operation and maintenance strategy based on the implicit risk information.

[0079] Preferably, in some embodiments, the implicit risk analysis based on the implicit behavioral feature vector specifically includes:

[0080] Obtain the latent behavior feature vector and historical latent behavior feature vector, and predict the latent behavior of each reactor node based on the latent behavior feature vector and historical latent behavior feature vector.

[0081] Based on the prediction results of the latent behavior of each reactor node, latent risk information is sent to the remote operation and maintenance center of the petrochemical equipment. The latent risk information includes the latent risk level and the latent risk area.

[0082] In specific implementation, the implicit behavior feature vector of the petrochemical equipment at the current moment is obtained, and the historical implicit behavior feature vector of the corresponding equipment is also obtained. The historical feature vector can be constructed by the historical node cooling stability sequence and stored in the database or local cache in chronological order to form a training and reference dataset for trend prediction. Based on the current implicit behavior feature vector and the historical feature vector, the implicit behavior of each reactor node is predicted. In some embodiments, the moving average autoregressive model commonly used in data trend prediction in the prior art can be used to predict the implicit behavior of each reactor node.

[0083] The following is a specific embodiment of this application using a moving average autoregressive model to predict the implicit behavior of each reactor node: A preset prediction time window is set for the next 24 hours (in other embodiments, other time lengths can be selected based on the equipment's operating cycle or maintenance cycle), and the implicit behavior feature vectors of each reactor node over the past 720 hours (or other selectable time periods) are obtained to form a historical implicit behavior feature sequence. This historical sequence may include the cooling stability of each reactor node, node coupling correlation, and other implicit feature indicators. The historical implicit behavior feature sequence is plotted as a time series graph, where the horizontal axis represents time and the vertical axis represents the implicit feature values ​​of each node. To eliminate the trend of the sequence variance changing over time, the sequence can be exponentially smoothed or normalized to obtain a stationary historical sequence.

[0084] After obtaining the historical implicit behavioral feature sequence, an autocorrelation coefficient plot and a partial autocorrelation coefficient plot are drawn based on the historical implicit behavioral feature sequence. The autocorrelation coefficient plot is used to observe the autoregressive characteristics of the node features, and the partial autocorrelation coefficient plot is used to identify the relationship between the node features and the moving average process. By observing the significant truncation order in the plots, the order p of the autoregressive model and the order q of the moving average model can be preliminarily determined. For example, if the last significant autocorrelation coefficient in the autocorrelation coefficient plot is at order 3, then the order of the autoregressive model is 3; if the last significant partial autocorrelation coefficient in the partial autocorrelation coefficient plot is at order 2, then the order of the moving average model is 2. Finally, based on the characteristics of the autocorrelation coefficient plot and the partial autocorrelation coefficient plot, the order (p, q) of the autoregressive moving average model (ARMA model) is determined.

[0085] During model building, the least squares method can be used to estimate the autoregressive and moving average parameters, and a significance test can be performed (e.g., a significance level of 0.05). Simultaneously, the optimal model parameters can be selected by referring to the Schwarz-Bayes criterion (BIC). Using the obtained autoregressive moving average model, the historical latent behavior feature sequence and the current latent behavior feature vector can be input into the model to predict the latent behavior of each reactor node within the future prediction time window.

[0086] Finally, the predicted latent behavior values ​​of each reactor node are used as the latent behavior prediction results, which can be used to generate latent risk information at the node level and the overall equipment level. This includes the latent risk level and latent risk area determined after judgment based on fixed threshold mapping, and sent to the remote operation and maintenance center to realize prediction-based risk warning and remote operation and maintenance intervention.

[0087] In a preferred embodiment of this application, the remote operation and maintenance center executes corresponding remote operation and maintenance strategies based on the implicit risk information to achieve timely intervention and predictive maintenance for potential anomalies in petrochemical equipment. The implicit risk information includes the implicit risk level and implicit risk area of ​​each reactor node. The implicit risk level indicates the severity of potential anomalies at each node, and the implicit risk area identifies the specific space or functional area in the equipment system where the risk exists. After receiving the implicit risk information through an industrial IoT communication interface, the remote operation and maintenance center analyzes the implicit risk level and implicit risk area, and matches the analysis results with a pre-set remote operation and maintenance strategy library. The strategy library defines corresponding operational measures based on different risk levels and risk areas. For example, a low-risk level only generates an alarm notification and records risk data for trend analysis; a medium-risk level simultaneously triggers an alarm and remotely adjusts the cooling system or temperature control parameters to mitigate potential risks; a high-risk level activates an emergency alarm, simultaneously adjusts key control parameters, arranges on-site inspections or emergency maintenance, and triggers backup operation or shutdown protection measures when necessary. During strategy execution, the remote operation and maintenance center continuously monitors the status of each node and changes in key parameters, and uses the execution feedback to evaluate the effectiveness of the strategy and optimize subsequent operation and maintenance decisions. This achieves closed-loop management from implicit risk information to remote operation and maintenance strategies, thereby effectively improving the operational safety and reliability of petrochemical equipment and supporting predictive maintenance and intelligent operation and maintenance management.

[0088] Furthermore, in another aspect of this application, in some embodiments, this application provides a remote operation and maintenance system for petrochemical equipment based on the Industrial Internet of Things (IIoT). This system includes an operation and maintenance control unit, as referenced... Figure 2The figure is a schematic diagram of the exemplary hardware and / or software structure of an operation and maintenance control unit according to some embodiments of this application. The operation and maintenance control unit 200 includes: an operation monitoring module 201, a data analysis module 202, and an operation and maintenance management module 203, which are described below:

[0089] The operation monitoring module 201 is used to collect operation monitoring data of petrochemical equipment through IoT sensors;

[0090] Data analysis module 202 is used to standardize the operation monitoring data to obtain the operation status feature dataset of the petrochemical equipment, and to perform multi-dimensional feature extraction based on the operation status feature dataset to determine the explicit behavior feature vector corresponding to the petrochemical equipment.

[0091] The data analysis module 202 is also used to perform feature learning through the explicit behavior feature vector to determine the equipment health index of the petrochemical equipment.

[0092] The data analysis module 202 is also used to collect the coolant information corresponding to each reactor node in the petrochemical equipment when the equipment health index is lower than a preset threshold, and perform a delay correlation analysis on the coolant information corresponding to each reactor node according to the preset reactor delay time to obtain the implicit behavior feature vector corresponding to the petrochemical equipment.

[0093] The operation and maintenance management module 203 is used to perform implicit risk analysis based on the implicit behavior feature vector and send implicit risk information to the remote operation and maintenance center of the petrochemical equipment; the remote operation and maintenance center executes the corresponding remote operation and maintenance strategy based on the implicit risk information.

[0094] The foregoing detailed an example of a remote operation and maintenance system and method for petrochemical equipment based on the Industrial Internet of Things provided in the embodiments of this application. It is understood that, in order to achieve the above functions, the corresponding device includes hardware structures and / or software modules for performing each function.

[0095] Those skilled in the art should readily recognize that, based on the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein, this application can be implemented in hardware or a combination of hardware and computer software. Whether a certain function in the application is executed in a manner that drives hardware or computer software depends on the specific application and design constraints of the technical solution. Therefore, those skilled in the art can use different methods to implement the described function for each specific application, but such implementation should not be considered to be beyond the scope of this application.

[0096] In addition, this application also provides a computer terminal device, which includes a memory and a processor. The memory stores code, and the processor is configured to acquire the code and execute the above-described method for remote operation and maintenance of petrochemical equipment based on the Industrial Internet of Things.

[0097] In some embodiments, reference Figure 3 The figure is a schematic diagram of the structure of a computer terminal device implementing a remote operation and maintenance method for petrochemical equipment based on the Industrial Internet of Things (IIoT) according to some embodiments of this application. The remote operation and maintenance method for petrochemical equipment based on the Industrial Internet of Things (IIoT) in the above embodiments can... Figure 3 The computer terminal device 300 shown is used to implement this, and the computer terminal device 300 includes at least one communication bus 301, communication interface 302, processor 303 and memory 304.

[0098] The processor 303 can be a general-purpose central processing unit (CPU), an application-specific integrated circuit (ASIC), or one or more devices used to control the execution of a remote operation and maintenance method for petrochemical equipment based on the Industrial Internet of Things as described in this application.

[0099] The communication bus 301 may include a path for transmitting information between the aforementioned components.

[0100] Memory 304 may be a read-only memory (ROM) or other type of static storage device capable of storing static information and instructions, random access memory (RAM) or other type of dynamic storage device capable of storing information and instructions, or electrically erasable programmable read-only memory (EEPROM), compact disc read-only memory (CD-ROM) or other optical disc storage, optical disc storage (including compressed optical discs, laser discs, optical discs, digital universal optical discs, Blu-ray discs, etc.), magnetic disks or other magnetic storage devices, or any other medium capable of carrying or storing desired program code in the form of instructions or data structures and accessible by a computer, but not limited thereto. Memory 304 may exist independently and be connected to processor 303 via communication bus 301. Memory 304 may also be integrated with processor 303.

[0101] The memory 304 stores program code for executing the solution of this application, and its execution is controlled by the processor 303. The processor 303 executes the program code stored in the memory 304. The program code may include one or more software modules. In the above embodiments, the determination of the device health index can be achieved by the processor 303 and one or more software modules in the program code in the memory 304.

[0102] Communication interface 302 uses any transceiver-like device for communicating with other devices or communication networks, such as Ethernet, radio access network (RAN), wireless local area networks (WLAN), etc.

[0103] Optionally, the computer terminal device 300 may also include a power supply 305 for providing power to various devices or circuits in the real-time computer terminal device.

[0104] In a specific implementation, as one example, a computer terminal device may include multiple processors, each of which may be a single-core (single-CPU) processor or a multi-core (multi-CPU) processor. Here, a processor may refer to one or more devices, circuits, and / or processing cores used to process data (e.g., computer program instructions).

[0105] The aforementioned computer terminal device can be a general-purpose computer terminal device or a dedicated computer terminal device. In specific implementations, the computer terminal device can be a desktop computer, a portable computer, a network server, a handheld computer (PDA), a mobile phone, a tablet computer, a wireless terminal device, a communication device, or an embedded device. This application does not limit the type of computer terminal device.

[0106] In addition, other aspects of this application also provide a computer-readable storage medium storing at least one computer program, which is loaded and executed by a processor to perform the operations described above in a remote operation and maintenance method for petrochemical equipment based on the Industrial Internet of Things.

[0107] In summary, the remote operation and maintenance system and method for petrochemical equipment based on the Industrial Internet of Things disclosed in this application firstly collects operational monitoring data of the petrochemical equipment; standardizes the operational monitoring data to obtain an operational status feature dataset of the petrochemical equipment; performs multi-dimensional feature extraction based on the operational status feature dataset to determine the explicit behavioral feature vector corresponding to the petrochemical equipment; performs feature learning through the explicit behavioral feature vector to determine the equipment health index of the petrochemical equipment; when the equipment health index is lower than a preset threshold, collects the coolant information corresponding to each reactor node in the petrochemical equipment; performs delay correlation analysis on the coolant information corresponding to each reactor node according to the preset reactor delay time to obtain the implicit behavioral feature vector corresponding to the petrochemical equipment; performs implicit risk analysis based on the implicit behavioral feature vector and sends implicit risk information to the remote operation and maintenance center of the petrochemical equipment; the remote operation and maintenance center executes the corresponding remote operation and maintenance strategy based on the implicit risk information. This allows for risk identification through the implicit behavioral analysis of the petrochemical equipment, achieving real-time control of implicit risks and improving the operational safety of the petrochemical equipment.

[0108] The above descriptions are merely embodiments of this application, and common knowledge such as specific technical solutions or characteristics in the solutions are not described in detail here. It should be noted that those skilled in the art can make several modifications and improvements without departing from the technical solutions of this application, and these should also be considered within the scope of protection of this application, without affecting the effectiveness of the implementation of this application or the practicality of the patent.

[0109] The scope of protection claimed in this application shall be determined by the content of its claims. The specific embodiments described in the specification can be used to interpret the content of the claims. Obviously, those skilled in the art can make various modifications and variations to this application without departing from the spirit and scope of the invention. Therefore, if these modifications and variations of this application fall within the scope of the claims of this application and their equivalents, this application also intends to include these modifications and variations.

Claims

1. A method for remote operation and maintenance of petrochemical equipment based on the Industrial Internet of Things, characterized in that, include: Data on the operation and monitoring of petrochemical equipment is collected through IoT sensors; The operation monitoring data is standardized to obtain the operation status feature dataset of the petrochemical equipment. Multidimensional feature extraction is performed based on the operation status feature dataset to determine the explicit behavior feature vector corresponding to the petrochemical equipment. The equipment health index of the petrochemical equipment is determined by performing feature learning through the explicit behavioral feature vectors. When the health index of the equipment is lower than the preset threshold, the coolant information of each reactor node in the petrochemical equipment is collected, and the coolant information of each reactor node is subjected to delay correlation analysis according to the preset reactor delay time to obtain the implicit behavior feature vector of the petrochemical equipment. Based on the implicit behavioral feature vector, implicit risk analysis is performed, and implicit risk information is sent to the remote operation and maintenance center of the petrochemical equipment. The remote operation and maintenance center executes corresponding remote operation and maintenance strategies based on implicit risk information.

2. The method as described in claim 1, characterized in that, The standardized processing of the operation monitoring data to obtain the operation status feature dataset of the petrochemical equipment specifically includes: Acquire initial sensor data for each dimension of the operational monitoring data; Outlier filtering and correction are performed on the initial sensor data for each dimension. The initial sensor data for each dimension are then timestamped and mapped to the same time series window to form a synchronous monitoring data matrix. The synchronous monitoring data matrix is ​​normalized to obtain the operating status feature dataset of the petrochemical equipment.

3. The method as described in claim 1, characterized in that, Based on the aforementioned operational status feature dataset, multidimensional feature extraction is performed to determine the explicit behavioral feature vector corresponding to the petrochemical equipment. Specifically, this includes: Obtain the aforementioned running status feature dataset; Extract the statistical features corresponding to each dimension of the monitoring data in the operational status dataset; The statistical features corresponding to each dimension of the monitoring data are combined to form an explicit behavioral feature vector.

4. The method as described in claim 1, characterized in that, The implicit risk analysis based on the aforementioned implicit behavioral feature vector specifically includes: Obtain the latent behavior feature vector and historical latent behavior feature vector, and predict the latent behavior of each reactor node based on the latent behavior feature vector and historical latent behavior feature vector. Based on the prediction results of the latent behavior of each reactor node, latent risk information is sent to the remote operation and maintenance center of the petrochemical equipment. The latent risk information includes the latent risk level and the latent risk area.

5. The method as described in claim 1, characterized in that, A single hidden layer neural network is used to classify the explicit behavior feature vector; based on the classification results, a health status mapping is performed to determine the equipment health index of the petrochemical equipment.

6. The method as described in claim 1, characterized in that, The PLC acquisition module is set as the data acquisition unit to perform signal conditioning and data acquisition processing on the raw signals output by the IoT sensor.

7. The method as described in claim 1, characterized in that, The operational monitoring data includes temperature data, pressure data, vibration data, and current data of the petrochemical equipment.

8. A remote operation and maintenance system for petrochemical equipment based on the Industrial Internet of Things (IIoT), comprising an operation and maintenance control unit, wherein the operation and maintenance control unit is used to execute the remote operation and maintenance method for petrochemical equipment based on the Industrial Internet of Things as described in any one of claims 1 to 7, characterized in that, The operation and maintenance control unit includes: The operation monitoring module is used to collect operation monitoring data of petrochemical equipment through IoT sensors; The data analysis module is used to standardize the operation monitoring data to obtain the operation status feature dataset of the petrochemical equipment, and to perform multi-dimensional feature extraction based on the operation status feature dataset to determine the explicit behavior feature vector corresponding to the petrochemical equipment. The data analysis module is also used to perform feature learning through the explicit behavioral feature vector to determine the equipment health index of the petrochemical equipment. The data analysis module is also used to collect coolant information corresponding to each reactor node in the petrochemical equipment when the equipment health index is lower than a preset threshold, and to perform delay correlation analysis on the coolant information corresponding to each reactor node according to the preset reactor delay time to obtain the implicit behavior feature vector corresponding to the petrochemical equipment. The operation and maintenance management module is used to perform implicit risk analysis based on the implicit behavioral feature vector and send implicit risk information to the remote operation and maintenance center of the petrochemical equipment; the remote operation and maintenance center executes the corresponding remote operation and maintenance strategy based on the implicit risk information.

9. A computer terminal device, characterized in that, The computer terminal device includes a memory and a processor. The memory stores code, and the processor is configured to acquire the code and execute a remote operation and maintenance method for petrochemical equipment based on the Industrial Internet of Things as described in any one of claims 1 to 7.

10. A computer-readable storage medium storing at least one computer program, characterized in that, The computer program is loaded and executed by a processor to perform the operations described in any one of claims 1 to 7 of the remote operation and maintenance method for petrochemical equipment based on the Industrial Internet of Things.

Images