Process equipment status monitoring and alarm system based on digital twin model

Abstract

The present invention discloses a process equipment status monitoring and alarm system based on a digital twin model. The present invention classifies the operating conditions and conducts in-depth analysis on the historical fault records of the reactor, explores the parameter change patterns therein, and constructs a set of operation warning indexes based on the operating condition types. Based on the set of operation warning indexes corresponding to different operating condition types of the reactor equipment, a data-driven model corresponding to the reactor equipment is established, and the data-driven model is integrated with the geometric model of the reactor equipment to obtain the digital twin model corresponding to the reactor equipment. The digital twin model is used to perform real-time monitoring and alarm on the reactor equipment, which overcomes the problem of poor accuracy caused by traditional single threshold warnings, achieves more accurate and reliable fault warnings, and greatly improves the early diagnosis capability.

Description

Technical Field

[0001] The present invention relates to the technical field of equipment monitoring and alarm technology, and in particular to a process equipment status monitoring and alarm system based on a digital twin model. Background Art

[0002] In many industrial fields such as chemical industry, pharmaceutical industry, and material synthesis, reactor equipment is one of the core production devices, and its operating status is directly related to product quality, production efficiency, and personnel and environmental safety. With the continuous development of industrial automation and intelligence, higher requirements are placed on the status monitoring and fault warning of reactor equipment.

[0003] At present, the monitoring and alarming of the reactor equipment in the prior art still has the following deficiencies:

[0004] When issuing early warnings for reactor equipment, it is impossible to effectively analyze and consider a large amount of historical fault data of the reactor equipment, and to explore the parameter change patterns therein as the basis for establishing an early warning mechanism. As a result, the accuracy of the early warning cannot be guaranteed and it is difficult to achieve early diagnosis of reactor equipment failures.

[0005] To this end, a process equipment status monitoring and alarm system based on the digital twin model was introduced. Summary of the invention

[0006] In view of this, the present invention provides a process equipment status monitoring and alarm system based on a digital twin model to solve the problems raised by the above background technology.

[0007] The purpose of the present invention can be achieved through the following technical solutions: A process equipment status monitoring and alarm system based on a digital twin model, comprising:

[0008] Geometric modeling module: Use the CAD drawings of the reactor equipment as a basis to build a highly accurate geometric model of the reactor equipment;

[0009] Data collection module: collects historical fault records of the reactor and stores them in a pre-built database. First, each historical fault is classified according to the pre-set working condition type; the working condition type includes temperature condition, pressure condition and stirring condition; after the classification is completed, each historical fault record of different working condition types is analyzed respectively, so as to obtain the operation warning index set MFi of the reactor equipment corresponding to different working condition types, where i represents the number of different working conditions, i=1, 2 or 3, corresponding to temperature condition, pressure condition and stirring condition respectively; based on the operation warning index set MFi of the reactor equipment corresponding to different working condition types, a data-driven model corresponding to the reactor equipment is established;

[0010] Model integration module: integrates the geometric model of the reactor equipment and the data-driven model to obtain the digital twin model corresponding to the reactor equipment;

[0011] Alarm module: During the operation of the reactor equipment, the fused digital twin model is used to monitor the reactor equipment. If the monitoring process triggers an alarm signal, the pre-deployed alarm will be triggered to sound an alarm and execute the corresponding steps; the alarm signals include temperature alarm, pressure alarm and stirring alarm;

[0012] Model optimization module: During the operation of the reactor equipment, the fused digital twin model is used to monitor the reactor equipment. If the reactor equipment fails but the monitoring process does not trigger an alarm signal of the corresponding fault type, the corresponding steps are executed to optimize the data-driven model.

[0013] In some embodiments, each historical fault record of the temperature condition type is analyzed, specifically:

[0014] Temperature working condition: Analyze each historical fault belonging to the temperature working condition, take the time point of each historical fault as the starting point, and extract the temperature of each monitoring point of the reactor equipment within the set time period before the starting point of each historical fault;

[0015] For each historical fault, the temperature values ​​of each monitoring point of the reactor equipment at each time point in the set time period are obtained, and the temperature values ​​of each monitoring point of the reactor equipment at each time point are respectively integrated into a temperature sequence set; the highest temperature value, the lowest temperature value and the temperature mean are respectively extracted from each integrated temperature sequence set;

[0016] The difference between the highest temperature value and the lowest temperature value in each set of temperature series is calculated to obtain the temperature trend value corresponding to each set of temperature series;

[0017] Preset the trend allowable value and mean allowable value corresponding to the temperature trend value and temperature mean value of each monitoring point of the reactor equipment within the set time period;

[0018] The temperature trend value and the temperature mean in each set of temperature series are respectively calculated with the corresponding preset trend allowable value and mean allowable value, so as to obtain the trend ratio and mean ratio corresponding to each set of temperature series, that is, calculated by temperature trend value / trend allowable value and temperature mean / mean allowable value;

[0019] Preset the weight coefficients corresponding to the trend ratio and the mean ratio, respectively, multiply the trend ratio and the mean ratio by the corresponding preset weight coefficients, and then sum them to obtain the sub-assessment index of each monitoring point within the set time period;

[0020] Preset the weight coefficient of the sub-evaluation index corresponding to each monitoring point respectively, multiply the sub-evaluation index of each monitoring point within the set time period by the corresponding preset weight coefficient respectively, and then sum them up to obtain the temperature evaluation index corresponding to each historical fault;

[0021] Set the upper and lower fluctuation ranges corresponding to the temperature evaluation index, and integrate the set upper and lower fluctuation ranges with the temperature evaluation index corresponding to each historical fault, so as to obtain the temperature warning index range corresponding to each historical fault;

[0022] An index set of the temperature warning index range corresponding to each historical fault is constructed, and the constructed index set is used as the operation warning index set MF1 under the temperature operating condition type corresponding to the reactor equipment.

[0023] In some embodiments, each historical fault record of the pressure condition type is analyzed, specifically:

[0024] Pressure working condition: Analyze each historical fault belonging to the pressure working condition, take the time point of each historical fault as the starting point, and extract the pressure value in the reactor of the reactor equipment within the set time period before the starting point of each historical fault;

[0025] For each historical fault, the pressure value in the reactor of the reactor equipment at each time point in the set time period is obtained; the pressure value in the reactor at each time point in the set time period is calculated using the standard deviation formula, thereby obtaining the pressure fluctuation value of the reactor equipment in the set time period;

[0026] Extracting the highest pressure value and the lowest pressure value from the pressure values ​​in the reactor of the reactor equipment at each time point within the set time period, and calculating the difference between the highest pressure value and the lowest pressure value, thereby obtaining the pressure trend value of the reactor equipment within the set time period;

[0027] Preset the pressure fluctuation allowable value and pressure trend allowable value corresponding to the pressure fluctuation value and pressure trend value of the reactor equipment in the set time period; calculate the ratio of the pressure fluctuation value and pressure trend value of the reactor equipment in the set time period to the corresponding preset pressure fluctuation allowable value and pressure trend allowable value, and obtain the pressure fluctuation ratio and pressure trend ratio; that is, calculate through pressure fluctuation value / pressure fluctuation allowable value and pressure trend value / pressure trend allowable value;

[0028] Preset the weight coefficients corresponding to the pressure fluctuation ratio and the pressure trend ratio, respectively, multiply the pressure fluctuation ratio and the pressure trend ratio by the corresponding preset weight coefficients, and then sum them to obtain the pressure assessment index corresponding to each historical fault of the reactor equipment;

[0029] Set the upper and lower fluctuation ranges corresponding to the pressure assessment index, and integrate the set upper and lower fluctuation ranges with the pressure assessment index corresponding to each historical fault, so as to obtain the pressure warning index range corresponding to each historical fault;

[0030] An index set of the pressure warning index range corresponding to each historical fault is constructed, and the constructed index set is used as the operation warning index set MF2 under the pressure condition type corresponding to the reactor equipment.

[0031] In some embodiments, each historical fault record of the stirring condition type is analyzed, specifically:

[0032] Stirring condition: Analyze each historical fault belonging to the stirring condition, take the time point of each historical fault as the starting point, and extract the stirring speed of the reactor equipment within the set time period before the starting point of each historical fault;

[0033] For each historical fault, the stirring speed of the reactor equipment at each time point in the set time period is obtained, and a line graph of the stirring speed change of the reactor equipment in the set time period is constructed;

[0034] Draw the numerical points corresponding to the stirring speed of the reactor equipment at each time point in the line graph, preset the normal reference range of the stirring speed of the reactor equipment, and draw the range interval corresponding to the normal reference range in the line graph, traverse each group of drawn numerical points, mark the numerical points outside the range interval, construct the vertical line segments between each group of marked points and the range interval as abnormal line segments, and take the range line closest to the marked point in the upper and lower range lines of the range interval as the object of constructing the vertical line segment;

[0035] The lengths of each group of abnormal line segments are accumulated to obtain the abnormal stirring value of the reactor equipment within the set time period; the number of marked points is counted as the number of abnormalities, and the value range of each group of quantities corresponding to the number of marked points is preset, and each group of quantity value range is set to correspond to an additional coefficient;

[0036] Match the number of abnormalities with the value range of each group of numbers to obtain the corresponding additional coefficient, and multiply the additional coefficient with the stirring abnormality value to obtain the stirring evaluation index of the reactor equipment within the set time period;

[0037] The upper and lower fluctuation ranges corresponding to the stirring evaluation index are set, and the set upper and lower fluctuation ranges are integrated with the stirring evaluation index corresponding to each historical fault, so as to obtain the stirring warning index range corresponding to each historical fault;

[0038] An index set of the stirring warning index range corresponding to each historical fault is constructed, and the constructed index set is used as the operation warning index set MF3 of the reactor equipment under the stirring working condition type.

[0039] In some embodiments, the fused digital twin model is used to monitor the reactor equipment, specifically:

[0040] During the monitoring process, the temperature evaluation index, pressure evaluation index and stirring evaluation index of the reactor equipment within the set time period are analyzed in real time and input into the operation warning index set MFi for matching. If a group of evaluation indexes successfully matches the operation warning index set MFi, the corresponding alarm signal is triggered.

[0041] In some embodiments, a pre-deployed alarm is triggered to sound an alarm, and corresponding steps are performed, specifically:

[0042] First, the operation of the reactor equipment is stopped, and the historical processing information of each preparation personnel is obtained, and the historical processing times, the corresponding alarm signals of the processing, and the duration of each processing times are extracted therefrom;

[0043] Classify each processing number according to the corresponding alarm signal to obtain the temperature alarm processing number, the pressure alarm processing number and the stirring alarm processing number; take the average of the time taken by each group corresponding to the temperature alarm processing number to obtain the temperature processing average time p1, take the average of the time taken by each group corresponding to the pressure alarm processing number to obtain the pressure processing average time p2, take the average of the time taken by each group corresponding to the stirring alarm processing number to obtain the stirring processing average time p3;

[0044] Preset the value ranges of each group of times corresponding to the number of temperature alarm processing times, the number of pressure alarm processing times and the number of stirring alarm processing times, respectively, and set each group of times corresponding to the number of temperature alarm processing times, the number of pressure alarm processing times and the number of stirring alarm processing times to correspond to an additional value of times; match the number of temperature alarm processing times, the number of pressure alarm processing times and the number of stirring alarm processing times of each preparatory personnel with the corresponding value ranges of each group of times, so as to obtain the additional values ​​of the number of temperature alarm processing times, the number of pressure alarm processing times and the number of stirring alarm processing times corresponding to each preparatory personnel, which are represented by t1, t2 and t3 respectively;

[0045] The ideal average temperature treatment time w1, the ideal average pressure treatment time w2 and the average stirring treatment time w3 corresponding to the average temperature treatment time p1, the average pressure treatment time p2 and the average stirring treatment time p3 of each preparation personnel are preset; using the formula The work performance value Fq of each reserve personnel is calculated.

[0046] In some embodiments, triggering a pre-deployed alarm to sound an alarm and executing corresponding steps further includes:

[0047] The work performance values ​​Fq are sorted from large to small, and the top three reserve personnel in terms of work performance values ​​Fq are selected. Then, the location of the current reactor equipment is used as the reference point, and the distance between the three selected reserve personnel and the reference point is calculated. The one with the shortest distance is selected from the three reserve personnel as the processing personnel, and the triggered alarm signal and the location of the reactor equipment are sent to the processing personnel.

[0048] In some embodiments, corresponding steps are performed to optimize the data-driven model, specifically:

[0049] The operator with the largest work performance value Fq at the time when the fault was discovered is selected to go to the location of the reactor equipment and determine the specific type of the fault. After the determination is completed, the fault is recorded as a new historical fault record and the operation warning index set MFi is updated. After the update is completed, the data-driven model is synchronously updated to complete the optimization.

[0050] Compared with the prior art, the present invention has the following beneficial effects:

[0051] The present invention classifies and deeply analyzes the working conditions of the historical fault records of the reactor, explores the law of parameter changes, constructs an operation warning index set based on the working condition type, and establishes a data-driven model corresponding to the reactor equipment based on the operation warning index set corresponding to different working conditions of the reactor equipment. The data-driven model is integrated with the geometric model of the reactor equipment to obtain a digital twin model corresponding to the reactor equipment. The digital twin model is used to monitor and alarm the reactor equipment in real time, overcoming the problem of poor accuracy caused by the traditional single threshold warning, achieving more accurate and reliable fault warning, and greatly improving the early diagnosis capability.

[0052] The present invention calculates the work performance value of each reserve personnel through the historical processing information of each reserve personnel, including the historical processing times, the corresponding alarm signals and the time taken, and selects the most suitable processing personnel in combination with the distance to the faulty equipment, thus changing the previous unscientific practice of randomly arranging personnel and improving the fault handling efficiency;

[0053] The present invention monitors the reactor equipment by utilizing the fused digital twin model. If the reactor equipment fails but the monitoring process does not trigger an alarm signal of the corresponding fault type, corresponding steps are executed to optimize the data-driven model, thereby achieving continuous updating and optimization of the digital twin model. BRIEF DESCRIPTION OF THE DRAWINGS

[0054] Further details, features and advantages of the present application are disclosed in the following description of exemplary embodiments in conjunction with the accompanying drawings, in which:

[0055] Figure 1It is a principle block diagram of the present invention;

[0056] Figure 2 It is a schematic diagram of the line graph of stirring speed variation in the present invention. DETAILED DESCRIPTION

[0057] Several embodiments of the present application will be described in more detail below with reference to the accompanying drawings so that those skilled in the art can implement the present application. The present application can be embodied in many different forms and purposes and should not be limited to the embodiments described herein. These embodiments are provided to make the present application comprehensive and complete, and to fully convey the scope of the present application to those skilled in the art. The embodiments do not limit the present application.

[0058] Unless otherwise defined, all terms (including technical and scientific terms) used herein have the same meaning as commonly understood by those of ordinary skill in the art to which this application belongs. It will be further understood that terms such as those defined in commonly used dictionaries should be interpreted as having a meaning consistent with their meaning in the relevant art and / or the context of this specification, and will not be interpreted in an idealized or overly formal sense unless explicitly defined as such herein.

[0059] See also Figure 1-Figure 2 As shown, the process equipment status monitoring and alarm system based on the digital twin model includes a geometric modeling module, a data collection module, a model integration module, an alarm module and a model optimization module;

[0060] The geometric modeling module is used to build a highly accurate geometric model of the reactor equipment using the CAD drawings of the reactor equipment as a basis;

[0061] It should be noted that the model presents the complex internal structure of the reactor in detail, such as the shape, size and installation position of the stirring paddle, the internal contour of the reactor body, the connection method of the inlet and outlet pipes, etc., and also accurately depicts the external shape and size information of the reactor. This not only provides a geometric framework for subsequent physical modeling and visualization, but also helps to perform spatial analysis and virtual operation simulation of the equipment in the digital twin environment;

[0062] The data collection module is used to collect the historical fault records of the reactor and store them in a pre-built database. First, each historical fault is classified according to the pre-set working condition type; the working condition type includes temperature working condition, pressure working condition and stirring working condition; after the classification is completed, each historical fault record of different working condition types is analyzed respectively, so as to obtain the operation warning index set MFi of the reactor equipment corresponding to different working condition types, where i represents the number of different working conditions, i=1,2 or 3, corresponding to temperature working condition, pressure working condition and stirring working condition respectively; based on the operation warning index set MFi of the reactor equipment corresponding to different working condition types, a data-driven model corresponding to the reactor equipment is established;

[0063] It should be noted that after collecting the operation warning index set MFi corresponding to different working conditions of the reactor equipment, an artificial neural network is selected to establish a data-driven model corresponding to the reactor equipment;

[0064] Traverse the historical fault records and classify the reactor failures caused by abnormal temperature into temperature conditions. Similarly, classify each historical fault record according to the cause of the failure;

[0065] Specifically:

[0066] Temperature conditions: Analyze each historical fault belonging to the temperature condition, take the time point of each historical fault as the starting point, and extract the temperature of each monitoring point of the reactor equipment within the set time period before the starting point of each historical fault; each monitoring point is pre-arranged by the technicians, including the center of the reactor body, different height levels and other key positions, to obtain the comprehensive temperature change of the reactor equipment before each historical fault occurs;

[0067] For each historical fault, the temperature values ​​of each monitoring point of the reactor equipment at each time point in the set time period are obtained, and the temperature values ​​of each monitoring point of the reactor equipment at each time point are respectively integrated into a temperature sequence set;

[0068] The highest temperature value, the lowest temperature value and the average temperature are extracted from each integrated set of temperature series respectively; the average temperature is obtained by calculating the average value of each set of temperature values ​​in the temperature series set;

[0069] The difference between the highest temperature value and the lowest temperature value in each set of temperature series is calculated to obtain the temperature trend value corresponding to each set of temperature series;

[0070] Preset the trend allowable value and mean allowable value corresponding to the temperature trend value and temperature mean value of each monitoring point of the reactor equipment within the set time period;

[0071] The temperature trend value and the temperature mean in each set of temperature series are respectively calculated with the corresponding preset trend allowable value and mean allowable value, so as to obtain the trend ratio and mean ratio corresponding to each set of temperature series, that is, calculated by temperature trend value / trend allowable value and temperature mean / mean allowable value;

[0072] Preset the weight coefficients corresponding to the trend ratio and the mean ratio, respectively, multiply the trend ratio and the mean ratio by the corresponding preset weight coefficients, and then sum them to obtain the sub-assessment index of each monitoring point within the set time period;

[0073] Preset the weight coefficient of the sub-evaluation index corresponding to each monitoring point respectively, multiply the sub-evaluation index of each monitoring point within the set time period by the corresponding preset weight coefficient respectively, and then sum them up to obtain the temperature evaluation index corresponding to each historical fault;

[0074] Set the upper and lower fluctuation ranges corresponding to the temperature evaluation index, and integrate the set upper and lower fluctuation ranges with the temperature evaluation index corresponding to each historical fault, so as to obtain the temperature warning index range corresponding to each historical fault;

[0075] Construct an index set of the temperature warning index range corresponding to each historical fault, and use the constructed index set as the operation warning index set MF1 under the temperature working condition type corresponding to the reactor equipment;

[0076] It should be noted that constructing a set of these indexes as the operation warning index set MF1 realizes the risk assessment of the entire reactor equipment before each historical failure occurs under temperature conditions, so that potential temperature-related failure risks can be discovered in time and corresponding measures can be taken.

[0077] Pressure working condition: Analyze each historical fault belonging to the pressure working condition, take the time point of each historical fault as the starting point, and extract the pressure value in the reactor of the reactor equipment within the set time period before the starting point of each historical fault;

[0078] For each historical fault, the pressure value in the reactor of the reactor equipment at each time point in the set time period is obtained; the pressure value in the reactor at each time point in the set time period is calculated using the standard deviation formula, thereby obtaining the pressure fluctuation value of the reactor equipment in the set time period;

[0079] Extracting the highest pressure value and the lowest pressure value from the pressure values ​​in the reactor of the reactor equipment at each time point within the set time period, and calculating the difference between the highest pressure value and the lowest pressure value, thereby obtaining the pressure trend value of the reactor equipment within the set time period;

[0080] Preset the pressure fluctuation allowable value and pressure trend allowable value corresponding to the pressure fluctuation value and pressure trend value of the reactor equipment in the set time period; calculate the ratio of the pressure fluctuation value and pressure trend value of the reactor equipment in the set time period to the corresponding preset pressure fluctuation allowable value and pressure trend allowable value, and obtain the pressure fluctuation ratio and pressure trend ratio; that is, calculate through pressure fluctuation value / pressure fluctuation allowable value and pressure trend value / pressure trend allowable value;

[0081] Preset the weight coefficients corresponding to the pressure fluctuation ratio and the pressure trend ratio, respectively, multiply the pressure fluctuation ratio and the pressure trend ratio by the corresponding preset weight coefficients, and then sum them to obtain the pressure assessment index corresponding to each historical fault of the reactor equipment;

[0082] Set the upper and lower fluctuation ranges corresponding to the pressure assessment index, and integrate the set upper and lower fluctuation ranges with the pressure assessment index corresponding to each historical fault, so as to obtain the pressure warning index range corresponding to each historical fault;

[0083] Construct an index set of the pressure warning index range corresponding to each historical fault, and use the constructed index set as the operation warning index set MF2 under the corresponding pressure condition type of the reactor equipment;

[0084] It should be noted that constructing a set of these indexes as the operation warning index set MF2 realizes the risk assessment of the entire reactor equipment before each historical failure occurs under pressure conditions, so that potential pressure-related failure risks can be discovered in time and corresponding measures can be taken.

[0085] Stirring condition: Analyze each historical fault belonging to the stirring condition, take the time point of each historical fault as the starting point, and extract the stirring speed of the reactor equipment within the set time period before the starting point of each historical fault;

[0086] For each historical fault, the stirring speed of the reactor equipment at each time point in the set time period is obtained, and a line graph of the stirring speed change of the reactor equipment in the set time period is constructed;

[0087] Draw the numerical points corresponding to the stirring speed of the reactor equipment at each time point in the line graph, preset the normal reference range of the stirring speed of the reactor equipment, and draw the range interval corresponding to the normal reference range in the line graph, traverse each group of drawn numerical points, mark the numerical points outside the range interval, construct the vertical line segments between each group of marked points and the range interval as abnormal line segments, and take the range line closest to the marked point in the upper and lower range lines of the range interval as the object of constructing the vertical line segment;

[0088] The lengths of each group of abnormal line segments are accumulated to obtain the abnormal stirring value of the reactor equipment within the set time period; the number of marked points is counted as the number of abnormalities, and the value range of each group of numbers corresponding to the number of marked points is preset, and each group of number value range is set to correspond to an additional coefficient; the additional coefficient range is set to 1.157-1.237, and the higher the number of abnormalities, the higher the corresponding additional coefficient;

[0089] It should be noted that, see Figure 2 ,

[0090] With time as the horizontal axis, arrange each time point within the set time period on the horizontal axis in sequence.

[0091] Take the stirring speed as the vertical axis, and mark the stirring speed value corresponding to each time point obtained at the corresponding position on the vertical axis;

[0092] Then, in chronological order, the numerical points corresponding to the stirring speed of the reactor equipment at each time point in the line graph are plotted. On the constructed stirring speed change line graph, the interval corresponding to the preset normal reference range is marked on the vertical axis. The numerical points corresponding to the lower and upper limits of the normal range are found on the vertical axis, and then this range interval is represented by a line segment parallel to the horizontal axis.

[0093] For those numerical points outside the normal reference range, they are marked. For each marked abnormal point, a vertical line segment is constructed as the abnormal line segment. The construction method is: take the range line closest to the marked point in the upper and lower range lines of the normal reference range as the object for constructing the vertical line segment.

[0094] Match the number of abnormalities with the value range of each group of numbers to obtain the corresponding additional coefficient, and multiply the additional coefficient with the stirring abnormality value to obtain the stirring evaluation index of the reactor equipment within the set time period;

[0095] The upper and lower fluctuation ranges corresponding to the stirring evaluation index are set, and the set upper and lower fluctuation ranges are integrated with the stirring evaluation index corresponding to each historical fault, so as to obtain the stirring warning index range corresponding to each historical fault;

[0096] Construct an index set of the stirring warning index range corresponding to each historical fault, and use the constructed index set as the operation warning index set MF3 of the reactor equipment under the stirring working condition type;

[0097] It should be noted that constructing a set of these indexes as the operation warning index set MF3 realizes the risk assessment of the entire reactor equipment before each historical failure occurs under stirring conditions, so that potential stirring-related failure risks can be discovered in time and corresponding measures can be taken.

[0098] The model integration module is used to integrate the geometric model of the reactor equipment and the data-driven model to obtain the digital twin model corresponding to the reactor equipment;

[0099] It should be noted that the geometric model and the data-driven model are integrated based on the model interface, and a unified interface is defined for the geometric model and the data-driven model;

[0100] For example, in the geometric model software, an interface is defined for the temperature property of the reactor tank, so that it can receive temperature data from the data-driven model. At the same time, at the output end of the data-driven model, the temperature data is output in the format required by the geometric model interface.

[0101] During operation, the data-driven model calculates the temperature of the reactor in real time and transmits the temperature data to the geometric model through the interface. The geometric model dynamically renders the color of the reactor tank body according to the received temperature data. For example, if the temperature calculated by the data-driven model is within the normal range, the reactor tank body is displayed as green in the geometric model; when the temperature exceeds the normal range, the tank body color gradually changes to red, and the depth of the color can also be proportional to the degree to which the temperature exceeds the normal range, thereby intuitively displaying the temperature status of the reactor;

[0102] The physical model provides physical constraints and prior knowledge for the data-driven model, while the data-driven model makes up for the errors and uncertainties that may exist in the physical model under complex working conditions.

[0103] The alarm module is used to monitor the reactor equipment during its operation using the fused digital twin model. If the monitoring process triggers an alarm signal, the pre-deployed alarm will sound an alarm and execute the corresponding steps; the alarm signals include temperature alarm, pressure alarm and stirring alarm;

[0104] Specifically:

[0105] During the monitoring process, the temperature evaluation index, pressure evaluation index and stirring evaluation index of the reactor equipment within the set time period are analyzed in real time, and input into the operation warning index set MFi for matching. If a set of evaluation indexes successfully matches the operation warning index set MFi, the corresponding alarm signal is triggered;

[0106] Follow the appropriate steps after an alert is issued:

[0107] First, the operation of the reactor equipment is stopped, and the historical processing information of each preparation personnel is obtained, and the historical processing times, the corresponding alarm signals of the processing, and the duration of each processing times are extracted therefrom;

[0108] Classify each processing number according to the corresponding alarm signal to obtain the temperature alarm processing number, the pressure alarm processing number and the stirring alarm processing number; take the average of the time taken by each group corresponding to the temperature alarm processing number to obtain the temperature processing average time p1, take the average of the time taken by each group corresponding to the pressure alarm processing number to obtain the pressure processing average time p2, take the average of the time taken by each group corresponding to the stirring alarm processing number to obtain the stirring processing average time p3;

[0109] Preset the value ranges of each group of times corresponding to the number of temperature alarm processing times, the number of pressure alarm processing times and the number of stirring alarm processing times, respectively, and set each group of times corresponding to the number of temperature alarm processing times, the number of pressure alarm processing times and the number of stirring alarm processing times to correspond to an additional value of times; the additional value range of times is set at 1.172-1.348, and the more times, the higher the corresponding additional value of times; match the number of temperature alarm processing times, the number of pressure alarm processing times and the number of stirring alarm processing times of each preparatory personnel with the corresponding value ranges of each group of times, so as to obtain the additional values ​​of the number of temperature alarm processing times, the number of pressure alarm processing times and the number of stirring alarm processing times of each preparatory personnel, which are represented by t1, t2 and t3 respectively;

[0110] The ideal average temperature treatment time w1, the ideal average pressure treatment time w2 and the average stirring treatment time w3 corresponding to the average temperature treatment time p1, the average pressure treatment time p2 and the average stirring treatment time p3 of each preparation personnel are preset; using the formula Calculate the work performance value Fq of each reserve personnel;

[0111] Sort the work performance values ​​Fq from large to small, select the top three reserve personnel with the highest work performance values ​​Fq, and then use the current location of the reactor as the reference point to calculate the distance between the three selected reserve personnel and the reference point. Select the one with the shortest distance from the three reserve personnel as the processing personnel, and send the triggered alarm signal and the location of the reactor to the processing personnel.

[0112] It should be noted that the entire process is based on the analysis and statistics of a large amount of historical processing information. As new processing data continues to accumulate, the evaluation criteria for reserve personnel and the troubleshooting process can be continuously updated and improved.

[0113] The model optimization module is used to monitor the reactor equipment using the fused digital twin model during the operation of the reactor equipment. If the reactor equipment fails but the monitoring process does not trigger an alarm signal of the corresponding fault type, the corresponding steps are executed to optimize the data-driven model.

[0114] Specifically:

[0115] The operator with the largest work performance value Fq at the time of the fault is selected to go to the location of the reactor equipment and determine the specific type of the fault. After the determination is completed, the fault is recorded as a new historical fault record and the operation warning index set MFi is updated. After the update is completed, the data-driven model is synchronously updated to complete the optimization;

[0116] It should be noted that as new fault data accumulates, the early warning index set MFi can be continuously updated and improved;

[0117] It also includes an auxiliary maintenance module, which is used for maintenance personnel to wear AR devices to view the image information of the digital twin model superimposed on the reactor equipment when performing maintenance on the reactor, and obtain information about hidden components inside the reactor;

[0118] The preferred embodiments of the present invention disclosed above are only used to help explain the present invention. The preferred embodiments do not describe all the details in detail, nor do they limit the invention to only specific implementation methods. Obviously, many modifications and changes can be made according to the content of this specification. This specification selects and specifically describes these embodiments in order to better explain the principles and practical applications of the present invention, so that those skilled in the art can understand and use the present invention well. The present invention is limited only by the claims and their full scope and equivalents.

Claims

1. The process equipment status monitoring and alarm system based on the digital twin model is characterized by: include: Geometric modeling module: Use the CAD drawings of the reactor equipment as a basis to build a highly accurate geometric model of the reactor equipment; Data collection module: collects historical fault records of the reactor and stores them in a pre-built database. First, each historical fault is classified according to the pre-set working condition type; The working condition types include temperature conditions, pressure conditions and mixing conditions; After the classification is completed, the historical fault records of different working conditions are analyzed respectively to obtain the operation warning index set MFi of the reactor equipment corresponding to different working conditions, where i represents the number of different working conditions, i=1, 2 or 3, corresponding to temperature conditions, pressure conditions and stirring conditions respectively; based on the operation warning index set MFi of the reactor equipment corresponding to different working conditions, a data-driven model corresponding to the reactor equipment is established; Model integration module: integrates the geometric model of the reactor equipment and the data-driven model to obtain the digital twin model corresponding to the reactor equipment; Alarm module: During the operation of the reactor equipment, the fused digital twin model is used to monitor the reactor equipment. If the monitoring process triggers an alarm signal, the pre-deployed alarm will be triggered to sound an alarm and execute the corresponding steps; the alarm signals include temperature alarm, pressure alarm and stirring alarm; Trigger the pre-set alarm to sound an alarm and execute the corresponding steps, specifically: First, the operation of the reactor equipment is stopped, and the historical processing information of each preparation personnel is obtained, and the historical processing times, the corresponding alarm signals of the processing, and the duration of each processing times are extracted therefrom; Classify each processing number according to the corresponding alarm signal to obtain the temperature alarm processing number, the pressure alarm processing number and the stirring alarm processing number; take the average of the time taken by each group corresponding to the temperature alarm processing number to obtain the temperature processing average time p1, take the average of the time taken by each group corresponding to the pressure alarm processing number to obtain the pressure processing average time p2, take the average of the time taken by each group corresponding to the stirring alarm processing number to obtain the stirring processing average time p3; Preset the value ranges of each group of times corresponding to the number of temperature alarm processing times, the number of pressure alarm processing times and the number of stirring alarm processing times, respectively, and set each group of times corresponding to the number of temperature alarm processing times, the number of pressure alarm processing times and the number of stirring alarm processing times to correspond to an additional value of times; match the number of temperature alarm processing times, the number of pressure alarm processing times and the number of stirring alarm processing times of each preparatory personnel with the corresponding value ranges of each group of times, so as to obtain the additional values ​​of the number of temperature alarm processing times, the number of pressure alarm processing times and the number of stirring alarm processing times corresponding to each preparatory personnel, which are represented by t1, t2 and t3 respectively; The ideal average temperature treatment time w1, the ideal average pressure treatment time w2 and the average stirring treatment time w3 corresponding to the average temperature treatment time p1, the average pressure treatment time p2 and the average stirring treatment time p3 of each preparation personnel are preset; using the formula Calculate the work performance value Fq of each reserve personnel; Model optimization module: During the operation of the reactor equipment, the fused digital twin model is used to monitor the reactor equipment. If the reactor equipment fails but the monitoring process does not trigger an alarm signal of the corresponding fault type, the corresponding steps are executed to optimize the data-driven model.

2. The process equipment status monitoring and alarm system based on the digital twin model according to claim 1 is characterized in that: Analyze the historical fault records of temperature condition types, specifically: Temperature working condition: Analyze each historical fault belonging to the temperature working condition, take the time point of each historical fault as the starting point, and extract the temperature of each monitoring point of the reactor equipment within the set time period before the starting point of each historical fault; For each historical fault, the temperature values ​​of each monitoring point of the reactor equipment at each time point in the set time period are obtained, and the temperature values ​​of each monitoring point of the reactor equipment at each time point are respectively integrated into a temperature sequence set; Extract the maximum temperature value, the minimum temperature value and the average temperature from each integrated set of temperature series; The difference between the highest temperature value and the lowest temperature value in each set of temperature series is calculated to obtain the temperature trend value corresponding to each set of temperature series; Preset the trend allowable value and mean allowable value corresponding to the temperature trend value and temperature mean value of each monitoring point of the reactor equipment within the set time period; The temperature trend value and the temperature mean in each set of temperature series are respectively calculated with the corresponding preset trend allowable value and mean allowable value, so as to obtain the trend ratio and mean ratio corresponding to each set of temperature series, that is, calculated by temperature trend value / trend allowable value and temperature mean / mean allowable value; Preset the weight coefficients corresponding to the trend ratio and the mean ratio, respectively, multiply the trend ratio and the mean ratio by the corresponding preset weight coefficients, and then sum them to obtain the sub-assessment index of each monitoring point within the set time period; Preset the weight coefficient of the sub-evaluation index corresponding to each monitoring point respectively, multiply the sub-evaluation index of each monitoring point within the set time period by the corresponding preset weight coefficient respectively, and then sum them up to obtain the temperature evaluation index corresponding to each historical fault; Set the upper and lower fluctuation ranges corresponding to the temperature evaluation index, and integrate the set upper and lower fluctuation ranges with the temperature evaluation index corresponding to each historical fault, so as to obtain the temperature warning index range corresponding to each historical fault; An index set of the temperature warning index range corresponding to each historical fault is constructed, and the constructed index set is used as the operation warning index set MF1 under the temperature operating condition type corresponding to the reactor equipment.

3. The process equipment status monitoring and alarm system based on the digital twin model according to claim 2 is characterized in that: Analyze the historical fault records of each pressure condition type, specifically: Pressure working condition: Analyze each historical fault belonging to the pressure working condition, take the time point of each historical fault as the starting point, and extract the pressure value in the reactor of the reactor equipment within the set time period before the starting point of each historical fault; For each historical fault, the pressure value in the reactor of the reactor equipment at each time point in the set time period is obtained; the pressure value in the reactor at each time point in the set time period is calculated using the standard deviation formula, thereby obtaining the pressure fluctuation value of the reactor equipment in the set time period; Extracting the highest pressure value and the lowest pressure value from the pressure values ​​in the reactor of the reactor equipment at each time point within the set time period, and calculating the difference between the highest pressure value and the lowest pressure value, thereby obtaining the pressure trend value of the reactor equipment within the set time period; Preset the pressure fluctuation allowable value and pressure trend allowable value corresponding to the pressure fluctuation value and pressure trend value of the reactor equipment within the set time period; Calculate the ratios of the pressure fluctuation value and the pressure trend value of the reactor equipment within the set time period to the corresponding preset pressure fluctuation allowable value and pressure trend allowable value to obtain the pressure fluctuation ratio and the pressure trend ratio; That is, it is calculated through the pressure fluctuation value / pressure fluctuation allowable value and the pressure trend value / pressure trend allowable value; Preset the weight coefficients corresponding to the pressure fluctuation ratio and the pressure trend ratio, respectively, multiply the pressure fluctuation ratio and the pressure trend ratio by the corresponding preset weight coefficients, and then sum them to obtain the pressure assessment index corresponding to each historical fault of the reactor equipment; Set the upper and lower fluctuation ranges corresponding to the pressure assessment index, and integrate the set upper and lower fluctuation ranges with the pressure assessment index corresponding to each historical fault, so as to obtain the pressure warning index range corresponding to each historical fault; An index set of the pressure warning index range corresponding to each historical fault is constructed, and the constructed index set is used as the operation warning index set MF2 under the pressure condition type corresponding to the reactor equipment.

4. The process equipment status monitoring and alarm system based on the digital twin model according to claim 3 is characterized in that: The historical fault records of the mixing condition type are analyzed, specifically: Stirring condition: Analyze each historical fault belonging to the stirring condition, take the time point of each historical fault as the starting point, and extract the stirring speed of the reactor equipment within the set time period before the starting point of each historical fault; For each historical fault, the stirring speed of the reactor equipment at each time point in the set time period is obtained, and a line graph of the stirring speed change of the reactor equipment in the set time period is constructed; Draw the numerical points corresponding to the stirring speed of the reactor equipment at each time point in the line graph, preset the normal reference range of the stirring speed of the reactor equipment, and draw the range interval corresponding to the normal reference range in the line graph, traverse each group of drawn numerical points, mark the numerical points outside the range interval, construct the vertical line segments between each group of marked points and the range interval as abnormal line segments, and take the range line closest to the marked point in the upper and lower range lines of the range interval as the object of constructing the vertical line segment; The lengths of each group of abnormal line segments are accumulated to obtain the abnormal stirring value of the reactor equipment within the set time period; the number of marked points is counted as the number of abnormalities, and the value range of each group of quantities corresponding to the number of marked points is preset, and each group of quantity value range is set to correspond to an additional coefficient; Match the number of abnormalities with the value range of each group of numbers to obtain the corresponding additional coefficient, and multiply the additional coefficient with the stirring abnormality value to obtain the stirring evaluation index of the reactor equipment within the set time period; The upper and lower fluctuation ranges corresponding to the stirring evaluation index are set, and the set upper and lower fluctuation ranges are integrated with the stirring evaluation index corresponding to each historical fault, so as to obtain the stirring warning index range corresponding to each historical fault; An index set of the stirring warning index range corresponding to each historical fault is constructed, and the constructed index set is used as the operation warning index set MF3 of the reactor equipment under the stirring working condition type.

5. The process equipment status monitoring and alarm system based on the digital twin model according to claim 4 is characterized in that: The fused digital twin model is used to monitor the reactor equipment, specifically: During the monitoring process, the temperature evaluation index, pressure evaluation index and stirring evaluation index of the reactor equipment within the set time period are analyzed in real time and input into the operation warning index set MFi for matching. If a group of evaluation indexes successfully matches the operation warning index set MFi, the corresponding alarm signal is triggered.

6. The process equipment status monitoring and alarm system based on the digital twin model according to claim 5 is characterized in that: Triggering a pre-set alarm to sound an alarm and executing corresponding steps, including: The work performance values ​​Fq are sorted from large to small, and the top three reserve personnel in terms of work performance values ​​Fq are selected. Then, the location of the current reactor equipment is used as the reference point, and the distance between the three selected reserve personnel and the reference point is calculated. The one with the shortest distance is selected from the three reserve personnel as the processing personnel, and the triggered alarm signal and the location of the reactor equipment are sent to the processing personnel.

7. The process equipment status monitoring and alarm system based on the digital twin model according to claim 6 is characterized in that: Perform the corresponding steps to optimize the data-driven model, specifically: The operator with the largest work performance value Fq at the time when the fault was discovered is selected to go to the location of the reactor equipment and determine the specific type of the fault. After the determination is completed, the fault is recorded as a new historical fault record and the operation warning index set MFi is updated. After the update is completed, the data-driven model is synchronously updated to complete the optimization.

Images