A DCS system maintenance system based on networked intelligent monitoring and a method thereof

Abstract

The application provides a DCS system maintenance system and method based on networked intelligent monitoring, and relates to the technical fields of intelligent manufacturing and high-voltage circuit breaker health management. First, an independent classification identifier and a unique incremental number are allocated to each type of high-voltage circuit breaker, multi-dimensional operation data of temperature, pressure, vibration and current are collected at a fixed time interval, and the operation data is stored in a high-voltage circuit breaker state data set according to the high-voltage circuit breaker number and time stamp. Then, a dynamic weight distribution formula is constructed, a health status score is output and recorded to a health status database. The operation data in the fault state is extracted to construct an abnormal data set, the health status score is combined with the abnormal data set, an LSTM model is used for deep training and prediction, and thus the future fault occurrence probability of the high-voltage circuit breaker is obtained. Once the predicted fault occurrence probability of the high-voltage circuit breaker is higher than the fault threshold, the high-voltage circuit breaker is immediately marked as a potential fault high-voltage circuit breaker, a detailed and targeted maintenance plan is generated, and the operation parameters of the potential fault high-voltage circuit breaker are quickly adjusted.

Description

Technical Field

[0001] This invention relates to the field of intelligent manufacturing and high-voltage circuit breaker health management technology, specifically to a DCS system maintenance system and method based on networked intelligent monitoring. Background Technology

[0002] In the crucial link of the modern industrial system of automatic control system for vaporized liquid argon, the high-voltage circuit breaker, as the core equipment for ensuring a stable power supply, shoulders the important mission of controlling and protecting the circuit. Its reliable operation is of paramount importance for maintaining the stable operation of the automatic control system for vaporized liquid argon, ensuring the continuity of the liquid argon vaporization process, and protecting the safety of various electrical equipment within the system. From the cryogenic pump drive motor at the liquid argon storage end, to the power supply for the heating device in the vaporization process, and to various sensors and actuators in the control system, the high-voltage circuit breaker runs through the entire automatic control system for vaporized liquid argon, and its performance directly affects the stability and safety of the system.

[0003] In automatic control systems for vaporized liquid argon, high-voltage circuit breakers often face environments characterized by extremely low temperatures, high pressures, and strong electromagnetic interference. For example, motors and cooling systems in distillation towers may require high-voltage circuit breaker protection circuits to prevent overload or short circuits. The vaporization process of liquid argon involves a large amount of heat exchange, leading to rapid changes in local ambient temperature, which may degrade the performance of the insulation materials of the high-voltage circuit breaker. The high-pressure environment inside the system places higher demands on the sealing performance and mechanical structure of the circuit breaker. Furthermore, the strong electromagnetic interference generated by electrical equipment in the system during operation may affect the transmission of control signals from the high-voltage circuit breaker, thereby affecting its normal operation. These complex and harsh operating conditions significantly increase the risk of high-voltage circuit breaker failure.

[0004] Traditional high-voltage circuit breaker maintenance strategies primarily rely on manual periodic inspections and reactive repairs. However, in vaporized liquid argon automatic control systems, manual periodic inspections face numerous challenges. Firstly, high-voltage circuit breakers in these systems are typically located in complex environments, with some areas potentially posing safety risks such as low-temperature freezing and asphyxiation, making it difficult for inspectors to conduct comprehensive and in-depth inspections within limited time. Secondly, the accuracy of manual assessments of equipment status is significantly affected by the inspectors' professional competence and experience. Different inspectors have varying perceptions and judgment criteria regarding high-voltage circuit breaker fault signs, resulting in a lack of unified standards and reliable data for assessing the health status of high-voltage circuit breakers.

[0005] In today's trend of digital and intelligent industrial development, the shortcomings of existing high-voltage circuit breaker maintenance technologies are becoming increasingly apparent. Firstly, there are limitations in data acquisition. Many existing monitoring systems can only monitor single-dimensional data, such as focusing solely on current or voltage, failing to comprehensively acquire operational information about the high-voltage circuit breaker and making it difficult to accurately assess its health status. Secondly, regarding fault prediction, most existing models rely on a single data source or simple algorithmic models, failing to fully consider the complex factors involved in the operation of high-voltage circuit breakers, resulting in poor prediction accuracy. Thirdly, maintenance planning is lagging behind. Although some intelligent monitoring systems can monitor the status of high-voltage circuit breakers in real time, the lack of dynamic adjustment mechanisms in the models prevents them from developing reasonable maintenance plans in a timely manner based on changes in equipment operating status, making it difficult to effectively cope with the dynamic changes in the operating status of high-voltage circuit breakers. Summary of the Invention

[0006] The purpose of this invention is to provide a DCS system maintenance system and method based on network intelligent monitoring to solve the problems mentioned in the background art.

[0007] To achieve the above objectives, the present invention further provides a DCS system maintenance system based on network-connected intelligent monitoring. The system is used to execute a DCS system maintenance method based on network-connected intelligent monitoring, comprising:

[0008] The status acquisition module is used to number the high-voltage circuit breakers in the DCS system in an incremental manner, collect multi-dimensional operating data at fixed time intervals, and store them as a high-voltage circuit breaker status dataset according to the number and timestamp. The operating data includes temperature, pressure, vibration and current.

[0009] The status scoring module is used to construct a dynamic weight allocation formula by combining the high-voltage circuit breaker status dataset with the real-time changes in the high-voltage circuit breaker's operating status, and output the health status score in real time and record it to the health status database.

[0010] The fault prediction module is used to analyze the status dataset of high-voltage circuit breakers, extract the operational data of fault status to construct an abnormal dataset, and construct an LSTM model to predict the probability of future faults of high-voltage circuit breakers by combining the real-time changes of health status scores with the abnormal dataset.

[0011] The fault determination module is used to pre-set fault thresholds, identify high-voltage circuit breakers with a predicted fault probability higher than the fault threshold as potentially faulty high-voltage circuit breakers, generate maintenance plans, and adjust the operating parameters of potentially faulty high-voltage circuit breakers.

[0012] This invention provides the following technical solution:

[0013] A DCS system maintenance method based on network-connected intelligent monitoring includes the following steps:

[0014] Step 1: Number the high-voltage circuit breakers in the DCS system in an incremental manner, collect multi-dimensional operating data at fixed time intervals, and store them as a high-voltage circuit breaker status dataset according to the number and timestamp. The operating data includes temperature, pressure, vibration and current.

[0015] Step 2: Using the high-voltage circuit breaker status dataset and combining it with the real-time changes in the high-voltage circuit breaker's operating status, construct a dynamic weight allocation formula, output the health status score in real time, and record it in the health status database.

[0016] Step 3: Analyze the high-voltage circuit breaker status dataset, extract the operational data of fault status to construct an abnormal dataset, and construct an LSTM model to predict the probability of future fault occurrence of the high-voltage circuit breaker by combining the real-time changes of the health status score with the abnormal dataset.

[0017] Step 4: Pre-set the fault threshold, mark the high-voltage circuit breakers with a predicted fault probability higher than the fault threshold as potential fault high-voltage circuit breakers, generate a maintenance plan and adjust the operating parameters of the potential fault high-voltage circuit breakers.

[0018] Furthermore, based on the type of high-voltage circuit breaker in the industry, each type of high-voltage circuit breaker is assigned a unique classification identifier, and then a unique incremental number is assigned to each high-voltage circuit breaker, thus assigning a unique high-voltage circuit breaker number W. ij Where i represents the high-voltage circuit breaker classification number, and j represents the incremental sequence number under that classification.

[0019] Furthermore, at a fixed time interval t gd The data acquisition cycle involves collecting multi-dimensional operational data and recording the temperature of each high-voltage circuit breaker as T. ij The pressure is P ij The vibration is V ij The current is I ij After that, the high-voltage circuit breaker was numbered W. ij With the corresponding timestamp t, the operating data of each high-voltage circuit breaker is stored in real time as a high-voltage circuit breaker status dataset, where the time interval t is... gd Given based on experience, the default is 5 minutes.

[0020] Furthermore, operational data from the high-voltage circuit breaker status dataset is extracted, and a dynamic weight allocation formula is constructed:

[0021] s ij (t)=ω1·f1(T ij (t))+ω2·f2(P ij (t))+ω3·f3(V ij (t))+ω4·f4(I ij (t))

[0022] In the formula, S ij (t) represents the high-voltage circuit breaker W. ij The health status score at time t; f1, f2, f3 and f4 are the respective calculation functions of temperature, pressure, vibration and current, ω1, ω2, ω3 and ω4 are the weighting coefficients of the high voltage circuit breaker health status score, and ω1+ω2+ω3+ω4=1;

[0023] The formula for the temperature calculation function is as follows:

[0024]

[0025] In the formula, T ij，ref The reference temperature of the high-voltage circuit breaker is represented by α1, which is a sensitivity constant ≥0, indicating the sensitivity of the high-voltage circuit breaker to the health status of the temperature.

[0026] The formula for the pressure calculation function is:

[0027]

[0028] In the formula, P ij，ref It is expressed as the reference pressure of the high-voltage circuit breaker, and α2 is a sensitivity constant ≥0, which represents the sensitivity of the pressure to the health status of the high-voltage circuit breaker.

[0029] The formula for the vibration calculation function is:

[0030]

[0031] In the formula, V ij，ref The reference vibration value of the high-voltage circuit breaker is represented by γ1 and γ2, which are sensitivity constants. γ1 < 0 and γ2 > 0, indicating the impact of vibration on the health status of the high-voltage circuit breaker.

[0032] The formula for the current calculation function is:

[0033]

[0034] In the formula, I ij，ref It is represented as the reference current of the high-voltage circuit breaker, and α3 is a sensitivity constant ≥0, which represents the influence of the current on the health status of the high-voltage circuit breaker.

[0035] Furthermore, during a period when the high-voltage circuit breaker is known to be in good health, n sets of operating data are collected, and the collected temperature data of the high-voltage circuit breaker is denoted as T. ijN The pressure data is P. ijN Vibration data is V ijN The current data is I ijN The health status score at this stage is S. maxThe value is 1, where N = 1, 2, ..., n, and n is a positive integer ≥ 2.

[0036] Calculate the average value of the temperature data:

[0037]

[0038] In the formula, This represents the average of n sets of temperature data collected from the high-voltage circuit breaker during a phase where the circuit is known to be in good health, where n represents the number of sets of operational data collected.

[0039] Constructing the intermediate temperature formula:

[0040]

[0041] The formula for calculating the temperature sensitivity constant can be obtained by rearranging the equations:

[0042]

[0043] Calculate the average value of the pressure data:

[0044]

[0045] In the formula, This represents the average value of n sets of pressure data collected from a high-voltage circuit breaker during a known healthy state phase.

[0046] Constructing the intermediate formula for pressure:

[0047]

[0048] The formula for calculating the pressure sensitivity constant can be obtained by rearranging the equations:

[0049]

[0050] Calculate the average value of the vibration data:

[0051]

[0052] In the formula, This represents the average value of n sets of vibration data collected from a high-voltage circuit breaker during a known healthy state phase.

[0053] when At that time, construct the intermediate formula for vibration:

[0054]

[0055] The formula for calculating the sensitivity constant of vibration can be obtained by rearranging the equations:

[0056]

[0057] when At that time, construct the intermediate formula for vibration:

[0058]

[0059] The formula for calculating the sensitivity constant of vibration can be obtained by rearranging the equations:

[0060]

[0061] Calculate the average value of the current data:

[0062]

[0063] In the formula, This represents the average value of n sets of current data collected from a high-voltage circuit breaker during a known healthy phase.

[0064] Constructing the intermediate formula for current:

[0065]

[0066] The formula for calculating the current sensitivity constant is obtained by reorganization:

[0067]

[0068] Furthermore, through the high-voltage circuit breaker status dataset, the operational data of each high-voltage circuit breaker fault state is extracted. When the temperature, pressure, vibration, and current exceed the safe range, they are marked as fault states, and a fault dataset is created. This dataset includes the operational data of the high-voltage circuit breaker marked as fault states from 12 hours before the fault occurs to 12 hours after the fault occurs, along with the corresponding high-voltage circuit breaker number and timestamp. Based on the operational data collected within this time period, the health status score of the high-voltage circuit breaker is calculated according to the dynamic weight allocation formula.

[0069] Then, the operational data of the fault dataset, the corresponding high-voltage circuit breaker number, timestamp, and the calculated health status score are used as input features.

[0070] The time step is:

[0071]

[0072] In the formula, timesteps represents the time step size of the LSTM model, T represents the duration of calibration for the fault state, and t gd This represents the time interval for collecting multi-dimensional runtime data;

[0073] The LSTM hidden layer is set to 64 units, the hyperbolic tangent activation function is selected, the LSTM has 2 layers, and each LSTM has 128 hidden units. Two fully connected layers are added after the LSTM layers. The last layer uses the sigmoid activation function to output the probability of failure. The loss function is the binary cross-entropy loss function. The Adam optimizer is used, the initial learning rate is set to 0.001, the batch size is set to 32-128, and the number of training epochs is set to 50-200. A Dropout layer with a ratio of 0.2 is added after each LSTM layer and fully connected layer, and the gradient threshold is set to 5. The future failure probability of the high-voltage circuit breaker is used as the output label with a value range of [0,1] to construct the LSTM model.

[0074] Furthermore, the trained LSTM model is used to predict the future fault probability of each high-voltage circuit breaker on the test dataset, with a pre-set fault threshold of P. thresh For high-voltage circuit breaker W ij When the probability P of a future fault of a high-voltage circuit breaker occurs ij >P thresh Then the high-voltage circuit breaker W ij High-voltage circuit breakers identified as potentially faulty.

[0075] Furthermore, for high-voltage circuit breakers identified as having potential faults, adjustment commands are issued through the DCS to modify the operating parameters of the high-voltage circuit breakers, namely, increasing the cooling efficiency of the high-voltage circuit breakers, reducing the operating pressure, reducing the speed of the high-voltage circuit breakers, and reducing the load current of the high-voltage circuit breakers. At the same time, an early warning is issued and a maintenance plan is generated.

[0076] Compared with the prior art, the beneficial effects of the present invention are:

[0077] This invention addresses the problem of insufficient data collection in traditional high-voltage circuit breaker management by periodically collecting and real-time monitoring data from multiple dimensions, including temperature, pressure, vibration, and current. Utilizing this multi-dimensional data, the operating status of the high-voltage circuit breaker can be more comprehensively assessed. Furthermore, by combining real-time status changes with dynamic adjustments to the weights of various parameters and calculating a health status score, the invention reflects the health status of the high-voltage circuit breaker in real time. Finally, by constructing an anomaly dataset based on fault status data and using an LSTM (Long Short-Term Memory) model to predict the probability of future faults, the invention enables early identification of potential faults, achieving early warning and adjusting operating parameters. This solves the problem of lack of real-time performance in traditional methods, reducing the occurrence and losses associated with high-voltage circuit breaker faults. Attached Figure Description

[0078] Figure 1This is a schematic diagram of the overall method flow of the present invention;

[0079] Figure 2 This is a schematic diagram of the overall system modules of the present invention. Detailed Implementation

[0080] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to specific embodiments.

[0081] It should be noted that, unless otherwise defined, the technical or scientific terms used in this invention should have the ordinary meaning understood by one of ordinary skill in the art to which this invention pertains. The terms "first," "second," and similar terms used in this invention do not indicate any order, quantity, or importance, but are merely used to distinguish different components. Terms such as "comprising" or "including" mean that the element or object preceding the word encompasses the elements or objects listed following the word and their equivalents, without excluding other elements or objects. Terms such as "connected" or "linked" are not limited to physical or mechanical connections, but can include electrical connections, whether direct or indirect. Terms such as "upper," "lower," "left," and "right" are used only to indicate relative positional relationships; when the absolute position of the described object changes, the relative positional relationship may also change accordingly.

[0082] Example:

[0083] Please see Figure 1 The present invention provides a technical solution:

[0084] A DCS system maintenance system based on networked intelligent monitoring includes a status acquisition module, a status scoring module, a fault prediction module, and a fault determination module. The status acquisition module numbers the high-voltage circuit breakers in the DCS system in an incremental manner, collects multi-dimensional operating data at fixed time intervals, and stores the data as a high-voltage circuit breaker status dataset according to the number and timestamp. The operating data includes temperature, pressure, vibration, and current.

[0085] In the automatic control system for vaporized liquid argon, various types of high-voltage circuit breakers are used. To achieve precise management and data monitoring, each type of high-voltage circuit breaker needs to be clearly identified and numbered. Specifically, each type of high-voltage circuit breaker is first assigned a unique classification identifier i. Then, a unique incremental number j is assigned to each high-voltage circuit breaker, resulting in a unique high-voltage circuit breaker number W for each circuit breaker. ij Where i represents the high-voltage circuit breaker classification number, such as oil circuit breaker, gas circuit breaker, and vacuum circuit breaker, and j represents the incrementing sequence number under that classification, incremented at fixed time intervals t. gdFor data acquisition cycles, in the power industry, temperature is usually an important factor affecting the health status of high-voltage circuit breakers, especially when high-voltage circuit breakers are in high-temperature environments, the risk of damage to high-voltage circuit breakers increases. Pressure also has a significant impact on the operating status of high-voltage circuit breakers, especially in power systems that include equipment such as pumps and valves, where abnormal pressure may lead to high-voltage circuit breaker failure. Vibration is often a precursor to damage to high-voltage circuit breakers, especially for high-voltage circuit breakers, where excessive vibration may cause damage to components. Current, as an operating parameter of electrical high-voltage circuit breakers, reflects changes in current load, power, etc., affecting the stable operation of high-voltage circuit breakers.

[0086] Therefore, multi-dimensional operational data is collected, and the temperature of each high-voltage circuit breaker is recorded as T. ij The pressure is P ij The vibration is V ij The current is I ij After that, the high-voltage circuit breaker was numbered W. ij With the corresponding timestamp t, the operating data of each high-voltage circuit breaker is stored in real time as a high-voltage circuit breaker status dataset.

[0087] For example, taking a liquid argon vaporization station as an example, this method is used to monitor the operating status of high-voltage circuit breakers. Classification numbers are assigned according to the type of high-voltage circuit breaker. In the high-voltage circuit breakers of the fractionation towers, unique, incrementally increasing high-voltage circuit breaker numbers are assigned sequentially. The high-voltage circuit breaker number for the first fractionation tower is W. 11 The high-voltage circuit breaker of the second distillation tower is numbered W. 12 The 5-minute time interval is a common data acquisition cycle adopted in many industry standards and practices. Therefore, the system collects multi-dimensional operating data of each high-voltage circuit breaker at 5-minute intervals, including temperature, pressure, vibration and current, and stores them as a high-voltage circuit breaker status dataset according to the corresponding high-voltage circuit breaker number and timestamp.

[0088] The status scoring module uses the high-voltage circuit breaker status dataset and combines it with the real-time changes in the high-voltage circuit breaker's operating status to construct a dynamic weight allocation formula, output a health status score in real time, and record it in the health status database.

[0089] During the operation of the extractor argon vaporization station, the operational data of the high-voltage circuit breaker status dataset is used to construct a dynamic weight allocation formula:

[0090] S ij (t)=ω1·f1(T ij (t))+ω2f2(P ij (t))+ω3·f3(V ij (t))+ω4·f4(I ij (t))

[0091] In the formula, S ij (t) represents the high-voltage circuit breaker W. ij The health status score is calculated at time t. f1, f2, f3, and f4 are the respective calculation functions for temperature, pressure, vibration, and current. ω1, ω2, ω3, and ω4 are the weighting coefficients for the high-voltage circuit breaker's health status score, and ω1 + ω2 + ω3 + ω4 = 1, with default values ​​of ω1 = 0.3, ω2 = 0.4, ω3 = 0.2, and ω4 = 0.1. Different types of high-voltage circuit breakers exhibit varying sensitivities to different operating parameters. In different application scenarios, some parameters may be more critical to the health status of the high-voltage circuit breaker than others. For example, in a distillation tower operating under high temperature (liquid argon boiling point approximately -186℃) and high pressure (working pressure up to 2-3 MPa), abnormal fluctuations in temperature and pressure may directly lead to equipment damage or liquid argon leakage. Temperature and pressure may be more critical than vibration. In mechanical high-voltage circuit breakers, vibration may be a more important fault indicator. The weights of each parameter can be appropriately adjusted according to the characteristics of each high-voltage circuit breaker. For example, if a high-voltage circuit breaker is highly sensitive to temperature changes, it may be assigned a higher temperature weight, but always ω2>ω1>ω3>ω4>0. Changes in each parameter directly affect the health status score. If temperature, pressure, vibration, or current is abnormal (exceeding safety thresholds), it will lead to a significant decrease in the health score, reflecting a deterioration in the health status of the high-voltage circuit breaker. By comprehensively considering the dynamic weighted scoring of multiple factors, the health status of the high-voltage circuit breaker can be assessed in real time.

[0092] The formula for the temperature calculation function is as follows:

[0093]

[0094] In the formula, T ij，ref The reference temperature of the high-voltage circuit breaker is represented by α1, which is a sensitivity constant ≥0, indicating the sensitivity of the high-voltage circuit breaker to the health status of the temperature.

[0095] The formula for the pressure calculation function is:

[0096]

[0097] In the formula, P ij，ref It is expressed as the reference pressure of the high-voltage circuit breaker, and α2 is a sensitivity constant ≥0, which represents the sensitivity of the pressure to the health status of the high-voltage circuit breaker.

[0098] The formula for the vibration calculation function is:

[0099]

[0100] In the formula, V ij，refThe reference vibration value of the high-voltage circuit breaker is represented by γ1 and γ2, which are sensitivity constants. γ1 < 0 and γ2 > 0, indicating the impact of vibration on the health status of the high-voltage circuit breaker.

[0101] The formula for the current calculation function is:

[0102]

[0103] In the formula, I ij，ref The reference current of the high-voltage circuit breaker is represented by α3, which is a sensitivity constant ≥0, representing the influence of the current on the health status of the high-voltage circuit breaker. The setting of the sensitivity constant reflects the sensitivity of the high-voltage circuit breaker to parameter changes under different operating conditions. It is adjusted according to the operating data and actual maintenance of the high-voltage circuit breaker to ensure that the health status assessment of the high-voltage circuit breaker under different operating conditions is more accurate.

[0104] In actual operation, various data of high-voltage circuit breakers may be subject to errors due to environmental interference and the accuracy of measuring instruments. By collecting multiple sets of data and taking the average value, the operating status of the high-voltage circuit breaker at different times can be comprehensively considered, so that the sensitivity constant can adapt to this dynamic change and more comprehensively and accurately reflect the relationship between the health status of the high-voltage circuit breaker and the data indicators.

[0105] Therefore, during the period when the high-voltage circuit breaker is known to be in good health, n sets of operating data are collected, and the collected temperature data of the high-voltage circuit breaker is denoted as T. ijN The pressure data is P. ijN Vibration data is V ijN The current data is I ijN The health status score at this stage is S. max The value is 1, where N = 1, 2, ..., n, and n is a positive integer ≥ 2.

[0106] Calculate the average value of the temperature data:

[0107]

[0108] In the formula, This represents the average of n sets of temperature data collected from the high-voltage circuit breaker during a phase where the circuit is known to be in good health, where n represents the number of sets of operational data collected.

[0109] Constructing the intermediate temperature formula:

[0110]

[0111] The formula for calculating the temperature sensitivity constant can be obtained by rearranging the equations:

[0112]

[0113] Calculate the average value of the pressure data:

[0114]

[0115] In the formula, This represents the average value of n sets of pressure data collected from a high-voltage circuit breaker during a known healthy state phase.

[0116] Constructing the intermediate formula for pressure:

[0117]

[0118] The formula for calculating the pressure sensitivity constant can be obtained by rearranging the equations:

[0119]

[0120] Calculate the average value of the vibration data:

[0121]

[0122] In the formula, This represents the average value of n sets of vibration data collected from a high-voltage circuit breaker during a known healthy state phase.

[0123] when At that time, construct the intermediate formula for vibration:

[0124]

[0125] The formula for calculating the sensitivity constant of vibration can be obtained by rearranging the equations:

[0126]

[0127] when At that time, construct the intermediate formula for vibration:

[0128]

[0129] The formula for calculating the sensitivity constant of vibration can be obtained by rearranging the equations:

[0130]

[0131] Calculate the average value of the current data:

[0132]

[0133] In the formula, This represents the average value of n sets of current data collected from a high-voltage circuit breaker during a known healthy phase.

[0134] Constructing the intermediate formula for current:

[0135]

[0136] The formula for calculating the current sensitivity constant is obtained by reorganization:

[0137]

[0138] The high-voltage circuit breaker W of the distillation tower 11 For example, to calculate a health status score at a certain moment, the collected data includes: timestamp: 2025-01-09 10:00:00, T 11 =450, P 11 =2.5, V 11 =2.0, I 11 =200, reference temperature T ij，ref =450, reference pressure P ij，ref =2, reference vibration value V ij，ref =1.5, reference current I ij，ref =180. Under normal circumstances, temperature has a significant impact on the high-voltage circuit breaker of the fractionation tower. Temperature fluctuations affect combustion efficiency, thermal expansion performance, and the structural safety of the fractionation tower. Therefore, the weight of temperature is set to ω1 = 0.35. During operation, the internal pressure of the fractionation tower is a key parameter for maintaining normal operation. Excessively high or low pressure can lead to tower malfunctions and, in severe cases, explosions. Therefore, the weight of pressure is usually set secondary, ω2 = 0.3. Abnormal vibration changes in the high-voltage circuit breaker of the fractionation tower during long-term operation... Vibration may cause wear, loosening of connections, or other mechanical damage to the high-voltage circuit breaker. The impact of vibration on the high-voltage circuit breaker in the fractionation tower is relatively smaller than that of temperature and pressure; therefore, the weight of vibration is set to ω3 = 0.15. Current typically reflects the load condition of the high-voltage circuit breaker in the fractionation tower. Excessive or insufficient current may indicate abnormal operation of the high-voltage circuit breaker, indirectly affecting the working state of the fractionation tower. Therefore, the weight of current is set to ω4 = 0.2. Five sets of operating data for the high-voltage circuit breaker were collected during a period when its health was known to be good, and the average value of the data was calculated.

[0139] Average temperature:

[0140]

[0141] Average pressure:

[0142]

[0143] Vibration average value:

[0144]

[0145] Average current:

[0146]

[0147] Given S max =1, which can be obtained using the formula for calculating the temperature sensitivity constant:

[0148]

[0149] The pressure sensitivity constant can be calculated using the following formula:

[0150]

[0151] The sensitivity constant of vibration can be calculated using the following formula:

[0152]

[0153] The following can be obtained from the formula for calculating the sensitivity constant of the current:

[0154]

[0155] The result of the temperature calculation function is as follows:

[0156]

[0157] The result of the pressure calculation function is:

[0158]

[0159] The result of the vibration calculation function is:

[0160] f3(V 11 (t))=1+γ2·(V 11 (t)-V 11，ref The result of the current calculation function is: )=1+32.2×(2.0-1.5)=17.1

[0161] f4(I 11 (t))=1+α3·(I 11 (t)-I 11，ref ) = 1 + 0.57 × (200 - 180) = 12.4

[0162] Therefore, the health status score of the high-voltage circuit breaker of the distillation tower at this moment is:

[0163] S 11 (t)=ω1·f1(T 11 (t))+ω2·f2(P 11 (t))+ω3·f3(V 11 (t))+ω4·f4(I 11 (t))

[0164] = 0.35×1 + 0.3×5.228 + 0.15×17.1 + 0.2×12.4

[0165] =0.35 + 1.568 + 2.565 + 2.48 = 6.963

[0166] Finally, a health status score is output in real time and recorded in the health status database.

[0167] The fault prediction module analyzes the high-voltage circuit breaker status dataset, extracts operational data on fault states to construct an anomaly dataset, and constructs an LSTM model to predict the probability of future faults in the high-voltage circuit breaker by combining real-time changes in health status scores with the anomaly dataset.

[0168] Using the high-voltage circuit breaker status dataset, operational data for each high-voltage circuit breaker in fault state is extracted. When temperature, pressure, vibration, or current exceeds the safe range, it is marked as a fault state, and a fault dataset is created. This dataset includes operational data of the high-voltage circuit breaker marked as fault state from 12 hours before the fault occurs to 12 hours after the fault occurs, along with the corresponding high-voltage circuit breaker number and timestamp. Based on the operational data collected within this time period, the health status score of the high-voltage circuit breaker is calculated according to the dynamic weight allocation formula.

[0169] Then, the operational data of the fault dataset, the corresponding high-voltage circuit breaker number, timestamp, and the calculated health status score are used as input features.

[0170] The time step is:

[0171]

[0172] In the formula, timesteps represents the time step size of the LSTM model, T represents the duration of calibration for the fault state, and t gd This represents the time interval for collecting multi-dimensional runtime data;

[0173] Extract a set of 24-hour operating data (temperature, pressure, vibration, and current) from a known healthy phase and perform Min-Max normalization.

[0174]

[0175] In the formula, T norm Represented as normalized temperature data, min(T) ij ) represents the minimum value of the temperature data within this set, maxT ij This represents the maximum temperature value within this set.

[0176]

[0177] In the formula, P norm Represented as normalized pressure data, min(P) ij ) represents the minimum pressure value within this set of data, maxP ij This represents the maximum value of the pressure data within this group;

[0178]

[0179] In the formula, V norm Represented as normalized vibration data, min(V ij ) represents the minimum value of the vibration data within this set, maxV ij This represents the maximum value of the vibration data within this set;

[0180]

[0181] middle, I norm Represented as normalized current data, min(I ij ) represents the minimum value of the current data within this set, maxI ij This represents the maximum value of the current data within this group;

[0182] The fault dataset is divided into multiple time-series samples according to time order. A sliding window with a window size of timesteps is set, sliding one time step at a time. The data from the previous time point is used as the input feature (training data), and the data from the next time point is used as the output label (prediction result), which is the probability of future fault occurrence of the high-voltage circuit breaker, with a value range of [0, 1]. Since the LSTM model is suitable for time-series data, a sliding window needs to be created, and the window size depends on the requirements of the prediction model. The data sampling interval is 5 minutes (which can be adjusted according to actual needs), so each window contains data from the past 60 minutes, i.e., 12 data points. After that, the label corresponding to the faulty high-voltage circuit breaker is 1, and the label corresponding to the non-faulty high-voltage circuit breaker is 0. Input layer: The input includes multiple top data (temperature, pressure, vibration, current) and health status scores of the high-voltage circuit breaker. LSTM layer: A 2-layer LSTM is selected, with the number of hidden units in each layer set to 128. The LSTM layer processes the input time-series data and captures the long-term dependencies in the data. Batch size is set to 32-128, training epochs to 50-200, Dropout layer: A Dropout layer is added after the LSTM layer with a ratio of 0.2 to prevent overfitting. Fully connected layers: Two fully connected layers, with the last layer using the sigmoid activation function to output the probability of future high-voltage circuit breaker failures. Loss function: A binary cross-entropy loss function is used, suitable for binary classification problems. Optimizer: The Adam optimizer is used with an initial learning rate of 0.001. Output layer: Using the sigmoid activation function, the output layer shows the probability of high-voltage circuit breaker failures, with a value range of [0,1].

[0183] The fault determination module pre-sets a fault threshold, identifies high-voltage circuit breakers with a predicted fault probability higher than the threshold as potentially faulty high-voltage circuit breakers, generates a maintenance plan, and adjusts the operating parameters of the potentially faulty high-voltage circuit breakers.

[0184] Using a trained LSTM model, predictions are made on the test dataset to obtain the future failure probability of each high-voltage circuit breaker. In machine learning and deep learning, setting a threshold of 0.5 is a common practice, especially in binary classification problems, where the model output is the probability of a high-voltage circuit breaker failure. The probability value output by the model is usually in the range [0,1], representing the probability of a high-voltage circuit breaker failure.

[0185] If the output probability is greater than 0.5, it means that the high-voltage circuit breaker is more likely to fail.

[0186] If the output probability is less than 0.5, it means that the high-voltage circuit breaker is more likely to be in normal operating condition.

[0187] During the operation of the liquid argon vaporization station, in order to respond promptly and effectively to potential faults in the high-voltage circuit breaker, a fault threshold of P is pre-set. thresh =0.5, but the threshold can be adjusted according to business needs based on actual circumstances. For high-voltage circuit breakers W ij When the probability P of a future fault of a high-voltage circuit breaker occurs ij >P thresh Then the high-voltage circuit breaker W ij High-voltage circuit breakers are identified as potentially faulty. Based on the potential fault prediction results of the high-voltage circuit breakers, the DCS issues a series of instructions to adjust the operating parameters of the high-voltage circuit breakers to reduce their workload and lower the risk of failure. Therefore, for high-voltage circuit breakers identified as potentially faulty, the DCS issues adjustment instructions to modify the operating parameters of the high-voltage circuit breakers, thereby increasing the cooling efficiency, reducing the operating pressure, decreasing the speed, and reducing the load current. At the same time, an early warning is issued and a maintenance plan is generated.

[0188] Please see Figure 2 The present invention also provides a DCS system maintenance method based on network intelligent monitoring, wherein the method is executed by the aforementioned DCS system maintenance system based on network intelligent monitoring, and includes:

[0189] Step 1: Number the high-voltage circuit breakers in the DCS system in an incremental manner, collect multi-dimensional operating data at fixed time intervals, and store them as a high-voltage circuit breaker status dataset according to the number and timestamp. The operating data includes temperature, pressure, vibration and current.

[0190] Step 2: Using the high-voltage circuit breaker status dataset and combining it with the real-time changes in the high-voltage circuit breaker's operating status, construct a dynamic weight allocation formula, output the health status score in real time, and record it in the health status database.

[0191] Step 3: Analyze the high-voltage circuit breaker status dataset, extract the operational data of fault status to construct an abnormal dataset, and construct an LSTM model to predict the probability of future fault occurrence of the high-voltage circuit breaker by combining the real-time changes of the health status score with the abnormal dataset.

[0192] Step 4: Pre-set the fault threshold, mark the high-voltage circuit breakers with a predicted fault probability higher than the fault threshold as potential fault high-voltage circuit breakers, generate a maintenance plan and adjust the operating parameters of the potential fault high-voltage circuit breakers.

[0193] The above embodiments can be implemented, in whole or in part, by software, hardware, firmware, or any other combination thereof. When implemented in software, the above embodiments can be implemented, in whole or in part, as a computer program product. Those skilled in the art will recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented by electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution.

[0194] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment, depending on actual needs.

[0195] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any changes or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application.

Claims

1. A DCS system maintenance system based on networked intelligent monitoring, characterized by, Comprise: The state acquisition module is used for numbering the high-voltage circuit breaker in the DCS system in an incremental manner, collecting multi-dimensional operation data at fixed time intervals, and storing as a high-voltage circuit breaker state data set according to the number and time stamp, wherein the operation data includes temperature, pressure, vibration and current; The state scoring module is used for constructing a dynamic weight distribution formula through the high-voltage circuit breaker state data set, combining the real-time changes of the high-voltage circuit breaker operating state, and outputting the health state score in real time and recording to the health state database; The fault prediction module is used for analyzing the high-voltage circuit breaker state data set, extracting the operating data of the fault state to construct an abnormal data set, and predicting the future fault occurrence probability of the high-voltage circuit breaker through the LSTM model combined with the abnormal data set according to the real-time changes of the health state score; The fault determination module is used for pre-setting a fault threshold, labeling the high-voltage circuit breaker with a predicted fault occurrence probability higher than the fault threshold as a potential fault high-voltage circuit breaker, generating a maintenance plan and adjusting the operating parameters of the potential fault high-voltage circuit breaker The high-voltage circuit breaker is assigned an independent classification mark, and then a unique incremental number is assigned in sequence, i.e. a unique high-voltage circuit breaker number W is assigned to each high-voltage circuit breaker ij wherein i represents the high-voltage circuit breaker classification number, and j represents the incremental serial number under the classification. At fixed time interval t gd , as data collection period, collect multi-dimensional operation data, record the temperature T ij , pressure P ij , vibration V ij , current I ij of each high-voltage circuit breaker, then store the operation data of each high-voltage circuit breaker as high-voltage circuit breaker state data set in real time with high-voltage circuit breaker number W ij and corresponding timestamp t, wherein the time interval t gd is given by experience, and the default is 5 minutes; Extract the operating data of the high-voltage circuit breaker state data set, and construct a dynamic weight distribution formula: S ij (t) = ω1 · f1(T ij (t)) + ω2 · f2(P ij (t)) + ω3 · f3(V ij (t)) + ω4 · f4(I ij (t)) where S ij (t) represents the high voltage circuit breaker W ij the health status score at time t; f1, f2, f3 and f4 are the respective calculation functions of temperature, pressure, vibration and current, ω1, ω2, ω3 and ω4 are the weight coefficients of the high voltage circuit breaker health status score, and ω1+ω2+ω3+ω4=1, ω2>ω1>ω3>ω4>0; The formula of the temperature calculation function is: In the formula, T ij，ref denotes the reference temperature of the high-voltage circuit breaker, and a1 is a sensitivity constant ≥ 0, which indicates the sensitivity of the temperature to the health status of the high-voltage circuit breaker; The formula of the pressure calculation function is: where P ij，ref denotes the reference pressure of the high voltage circuit breaker, and a2 is a sensitivity constant > 0, which indicates the sensitivity of the pressure to the health state of the high voltage circuit breaker; The formula of the vibration calculation function is: In the formula, V ij，ref The reference vibration value of the high-voltage circuit breaker is represented, and γ1, γ2 are sensitivity constants, wherein γ1<0 and γ2>0, and the influence of vibration on the health state of the high-voltage circuit breaker is represented. The formula of the current calculation function is: where I ij，ref Irefis the reference current for the high voltage circuit breaker, and a3is a sensitivity constant > 0, which indicates the influence of the current on the health state of the high voltage circuit breaker.

2. The DCS system maintenance system based on networked intelligent monitoring according to claim 1, characterized in that: The algorithm method of the sensitivity constant is: In a known good health phase, n sets of operating data of the high-voltage circuit breaker are collected, and the collected temperature data of the high-voltage circuit breaker is T ijN , the pressure data is P ijN , the vibration data is V ijN , and the current data is I ijN . The health state score S max of this phase is 1, where N = 1, 2, …, n, and n is a positive integer ≥ 2. Calculate the average value of the temperature data: wherein n sets of temperature data of high voltage circuit breakers collected at a stage representing a known good health status, n representing the number of sets of operational data collected; Construct the temperature intermediate formula: The sensitivity constant calculation formula of the temperature is arranged as: Calculate the average value of the pressure data: wherein n sets of pressure data collected from high voltage circuit breakers in a known good health phase are represented by the average values; Construct the pressure intermediate formula: The sensitivity constant calculation formula of the pressure is arranged as: Calculate the average value of the vibration data: wherein n sets of vibration data averages of high voltage circuit breakers collected at a known good health phase are represented; When the intermediate formula for the vibration is constructed: The sensitivity constant calculation formula of the vibration is arranged as: When the intermediate formula for vibration is constructed: The sensitivity constant calculation formula of the vibration is arranged as: Calculate the average value of the current data: wherein n sets of current data averages of high voltage circuit breakers collected at a known good health phase are represented; Construct the current intermediate formula: The sensitivity constant calculation formula of the current is arranged as:

3. The DCS system maintenance system based on networked intelligent monitoring according to claim 1, characterized in that: The method for constructing the LSTM model to predict the future fault occurrence probability of the high-voltage circuit breaker is: Through the high-voltage circuit breaker state data set, the operating data of each high-voltage circuit breaker fault state is extracted, and when the temperature, pressure, vibration and current exceed the safe range, it is labeled as a fault state to create a fault data set. This data set includes the operating data of the high-voltage circuit breaker within 12 hours before and after the fault occurs, and the corresponding high-voltage circuit breaker number, time stamp, based on the operating data collected within this time period, the health state score of the high-voltage circuit breaker is calculated according to the dynamic weight distribution formula; Then, the operating data of the fault data set and the corresponding high-voltage circuit breaker number, time stamp, and the calculated health state score are used as input features, The time step is: In the formula, timesteps represents the time step of the LSTM model, T represents the time length calibrated as the fault state, t gd represents the time interval for collecting multi-dimensional operation data; The fault data set is divided into multiple time sequence samples in chronological order, a sliding window with a window size of timesteps is set, and the data of the previous time point is used as the input feature (training data) and the data of the next time point is used as the output label (prediction result), i.e. the future fault occurrence probability of the high-voltage circuit breaker, and the value range is [0, 1]. The number of LSTM hidden layer units is set to 64, the activation function is selected as the hyperbolic tangent activation function, the number of hidden units of each LSTM layer is set to 128, two fully connected layers are added after the LSTM layer, the last layer uses the sigmoid activation function to output the fault occurrence probability, the binary cross-entropy loss function is used as the loss function, the Adam optimizer is used, the initial learning rate is set to 0.001, the batch size is set to 32-128, the training round is set to 50-200 rounds, a Dropout layer is added after each LSTM layer and fully connected layer, the proportion is 0.2, the gradient threshold is set to 5, the future fault occurrence probability of the high-voltage circuit breaker is used as the output label, the value range is [0, 1], and the LSTM model is constructed. After training, the trained model is used to predict the next time point to obtain the predicted fault occurrence probability.

4. The DCS system maintenance system based on networked intelligent monitoring according to claim 1, characterized in that: The high-voltage circuit breaker with a predicted fault occurrence probability higher than the fault threshold is labeled as a potential fault high-voltage circuit breaker, including the following steps: Using the trained LSTM model, the test data set is predicted to obtain the future failure probability of each high-voltage circuit breaker, and the failure threshold is preset as P thresh , for the high-voltage circuit breaker W ij , when the future failure probability P ij of the high-voltage circuit breaker is greater than the failure threshold P thresh , the high-voltage circuit breaker W ij is calibrated as a potential failure high-voltage circuit breaker.

5. The DCS system maintenance system based on networked intelligent monitoring according to claim 4, characterized in that: Generating a maintenance plan and adjusting the operating parameters of the potential fault high-voltage circuit breaker includes the following steps: For the high-voltage circuit breaker labeled as a potential fault, the DCS issues adjustment instructions to modify the operating parameters of the high-voltage circuit breaker, such as increasing the cooling efficiency of the high-voltage circuit breaker, reducing the working pressure, reducing the speed of the high-voltage circuit breaker, and reducing the load current of the high-voltage circuit breaker, while issuing a warning to generate a maintenance plan.

6. A DCS system maintenance method based on networked intelligent monitoring, characterized in that, The maintenance method is performed by a DCS system maintenance system based on networked intelligent monitoring according to any one of claims 1-5, and the specific steps include: Step 1: Number the high-voltage circuit breakers in the DCS system in an increasing manner, collect multi-dimensional operating data at fixed time intervals, and store them as high-voltage circuit breaker state data sets according to the number and time stamp. The operating data includes temperature, pressure, vibration, and current; Step 2: Construct a dynamic weight distribution formula based on the high-voltage circuit breaker state data set and the real-time changes of the high-voltage circuit breaker operating state, and output the health status score in real time and record it to the health status database; Step 3: Analyze the high-voltage circuit breaker state data set, extract the operating data of the fault state to construct an abnormal data set, and predict the future fault occurrence probability of the high-voltage circuit breaker based on the real-time changes of the health status score and the abnormal data set; Step 4: Pre-set a fault threshold, label the high-voltage circuit breaker with a predicted fault occurrence probability higher than the fault threshold as a potential fault high-voltage circuit breaker, generate a maintenance plan, and adjust the operating parameters of the potential fault high-voltage circuit breaker.

Images