Residue hydroprocessing unit run length prediction method and apparatus

Abstract

The application provides a residue oil hydrogenation device operation cycle prediction method and device, the method comprises the following steps: determining the feed oil property data of each device in the residue oil hydrogenation device; inputting the feed oil property data of each device into the operation state prediction model of each device respectively to obtain the operation state of each device output by the operation state prediction model of each device; inputting the operation state of each device into the operation cycle prediction model to obtain the operation cycle of the residue oil hydrogenation device output by the operation cycle prediction model. Based on the feed oil property data of each device, the operation state prediction model is adopted, the operation state of each device can be accurately predicted, the operation state model is adopted, the operation state of each device is analyzed, and the operation cycle of the residue oil hydrogenation device can be accurately and quickly predicted.

Description

Technical Field

[0001] This invention relates to the field of residue hydrotreating technology, and in particular to a method and apparatus for predicting the operating cycle of a residue hydrotreating unit. Background Technology

[0002] The trend of heavy oil deterioration is becoming increasingly serious. How to efficiently convert and utilize heavy oil is a key focus of the current oil refining industry. Among the current residue hydrotreating solutions, fixed-bed residue hydrotreating units are relatively mature and widely used, but they have problems such as reactor pressure drop and catalyst deactivation.

[0003] To address these issues, operating cycle equations for residue hydrotreating units are often constructed based on chemical reaction mechanisms. These equations are then used to predict the operating cycle of the unit, and relevant parameters are optimized accordingly. However, when the parameters in the constructed equations change, the obtained operating cycles become inaccurate. Constructing new equations based on the changed parameters is time-consuming and inefficient. Summary of the Invention

[0004] This invention provides a method and apparatus for predicting the operating cycle of a residue oil hydrotreating unit, which solves the problem of low accuracy in predicting the operating cycle of residue oil hydrotreating units in the prior art.

[0005] This invention provides a method for predicting the operating cycle of a residue hydrotreating unit, comprising:

[0006] Determine the feed oil properties data for each unit in the residue hydrotreating unit;

[0007] Input the feed oil property data of each device into the operation status prediction model of each device to obtain the operation status of each device output by the operation status prediction model.

[0008] The operating status of each device is input into the operating cycle prediction model to obtain the operating cycle of the residue hydrotreating unit output by the operating cycle prediction model.

[0009] The operation status prediction model for each device is trained based on the sample feed oil property data and the sample operation status of each device, while the operation cycle prediction model is trained based on the sample operation status of each device and the sample operation cycle of the residue hydrotreating unit.

[0010] According to the method for predicting the operating cycle of a residue hydrotreating unit provided by the present invention, after obtaining the operating status of each device output by the operating status prediction model of each device, the method further includes:

[0011] Obtain the current operating status of each device;

[0012] Based on the operating status of each device, the current operating status of each device is adjusted to ensure that the current operating status of each device is consistent with the operating status of each device.

[0013] According to the present invention, a method for predicting the operating cycle of a residue hydrotreating unit includes adjusting the current operating status of each piece of equipment based on their operating status, comprising:

[0014] Based on the operating status of each device and its current operating status, determine the adjustment amount for the current operating status of each device.

[0015] Adjust the current operating status of each device based on the adjustment amount of the current operating status of each device.

[0016] According to the method for predicting the operating cycle of a residue hydrotreating unit provided by the present invention, the training steps of the prediction model for the operating status of each piece of equipment include:

[0017] Determine the first initial model for predicting the operating status of each piece of equipment;

[0018] Data preprocessing is performed on the sample feed oil properties data and sample operating status of each device. Based on the preprocessed sample feed oil properties data and sample operating status of each device, the first initial model is trained to obtain the operating status prediction model of each device.

[0019] According to the present invention, a method for predicting the operating cycle of a residue hydrotreating unit includes the following training steps for the operating cycle prediction model:

[0020] Determine the second initial model of the operation cycle prediction model;

[0021] The sample operating status of each device and the sample operating cycle of the residue hydrotreating unit are preprocessed, and the second initial model is trained based on the preprocessed sample operating status of each device and the sample operating cycle of the residue hydrotreating unit to obtain the operating cycle prediction model.

[0022] According to the present invention, a method for predicting the operating cycle of a residue hydrotreating unit includes the following data preprocessing steps: sequentially performing data dimensionality reduction and feature normalization.

[0023] The present invention also provides a device for predicting the operating cycle of a residue hydrotreating unit, comprising:

[0024] The data determination unit is used to determine the feed oil properties data for each piece of equipment in the residue hydrotreating unit.

[0025] The status prediction unit is used to input the feed oil property data of each device into the operation status prediction model of each device, and obtain the operation status of each device output by the operation status prediction model of each device.

[0026] The cycle prediction unit is used to input the operating status of each device into the operation cycle prediction model to obtain the operation cycle of the residue hydrotreating unit output by the operation cycle prediction model.

[0027] The operation status prediction model for each device is trained based on the sample feed oil property data and the sample operation status of each device, while the operation cycle prediction model is trained based on the sample operation status of each device and the sample operation cycle of the residue hydrotreating unit.

[0028] The present invention also provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the program to implement the operation cycle prediction method of any of the above-described residue hydrotreating unit.

[0029] The present invention also provides a non-transitory computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the method for predicting the operating cycle of a residue hydrotreating unit as described above.

[0030] The present invention also provides a computer program product, including a computer program that, when executed by a processor, implements the method for predicting the operating cycle of a residue hydrotreating unit as described above.

[0031] The present invention provides a method and apparatus for predicting the operating cycle of a residue oil hydrotreating unit. Based on the feed oil property data of each device, an operating state prediction model is adopted to accurately predict the operating state of each device. Thus, the operating state model can be used to analyze the operating state of each device and accurately and quickly predict the operating cycle of the residue oil hydrotreating unit. Attached Figure Description

[0032] To more clearly illustrate the technical solutions in this invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of this invention. For those skilled in the art, other drawings can be obtained from these drawings without creative effort.

[0033] Figure 1 This is a flowchart illustrating the method for predicting the operating cycle of a residue hydrotreating unit provided by the present invention.

[0034] Figure 2This is a flowchart illustrating another method for predicting the operating cycle of a residue hydrotreating unit provided by the present invention.

[0035] Figure 3 This is a flowchart illustrating the model training method provided by the present invention;

[0036] Figure 4 This is a schematic diagram of the structure of the residual oil hydrotreating unit operation cycle prediction device provided by the present invention; Figure 5 This is a schematic diagram of the structure of the electronic device provided by the present invention. Detailed Implementation

[0037] To make the objectives, technical solutions, and advantages of this invention clearer, the technical solutions of this invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of this invention. All other embodiments obtained by those skilled in the art based on the embodiments of this invention without creative effort are within the scope of protection of this invention.

[0038] Currently, most methods rely on chemical reaction mechanisms to construct operating cycle equations for residue hydrotreating units. These equations are then used to predict the operating cycle of the unit, and relevant parameters are optimized based on these cycles. However, when the parameters in the constructed equations change, the obtained operating cycles become inaccurate. Constructing new equations based on the changed parameters is time-consuming and inefficient.

[0039] In response, this invention provides a method for predicting the operating cycle of a residue hydrotreating unit. Figure 1 This is a flowchart illustrating the method for predicting the operating cycle of a residue hydrotreating unit provided by the present invention, as shown below. Figure 1 As shown, the method includes the following steps:

[0040] Step 110: Determine the feed oil properties data for each piece of equipment in the residue hydrotreating unit.

[0041] Here, the equipment in the residue hydrotreating unit may include a heater, reactor, hot high-efficiency fractionator, hot low-efficiency fractionator, cold high-efficiency fractionator, cold low-efficiency fractionator, flash tank, distillation column, etc. Feed oil property data for each unit is used to characterize the feed oil properties of each unit. The content of sulfur (S), residual carbon (CCR), viscosity (N), nickel (Ni), vanadium (V), etc., in the feed oil can be used as feed oil property data.

[0042] Step 120: Input the feed oil property data of each device into the operation status prediction model of each device to obtain the operation status of each device output by the operation status prediction model of each device; the operation status prediction model of each device is trained based on the sample feed oil property data and sample operation status of each device.

[0043] Specifically, the operating status of each piece of equipment is used to characterize the operating status of each piece of equipment, which can be characterized by temperature (T), pressure (P), and flow rate (F). After determining the feed oil property data of each piece of equipment, the feed oil property data of each piece of equipment is input into the operating status prediction model of each piece of equipment. The operating status prediction model of each piece of equipment predicts the operating status of each piece of equipment based on the feed oil property data of each piece of equipment.

[0044] For example, in a residual oil hydrotreating unit, the feed oil properties data can be input into the furnace's operating status prediction model to obtain the furnace's operating status. Similarly, in a residual oil hydrotreating unit, the feed oil properties data can be input into the reactor's operating status prediction model to obtain the reactor's operating status.

[0045] The sample feed oil properties data for each device can be obtained from the Laboratory Information Management System (LIMS), and the sample operation status of each device can be obtained from the Distributed Control System (DCS).

[0046] Understandably, before inputting the feed oil property data of each device into the operation status prediction model of each device, the operation status prediction model of each device can be pre-trained. This can be achieved by performing the following steps: First, collect a large amount of sample feed oil property data for each device, and determine the corresponding sample operation status of each device through manual annotation. Then, train the first initial model based on the sample feed oil property data of each device and its corresponding sample operation status, thereby obtaining the operation status prediction model of each device.

[0047] Step 130: Input the operating status of each device into the operating cycle prediction model to obtain the operating cycle of the residue hydrotreating unit output by the operating cycle prediction model; the operating cycle prediction model is trained based on the sample operating status of each device and the sample operating cycle of the residue hydrotreating unit.

[0048] Specifically, the operating cycle of each piece of equipment is used to characterize the operating cycle of the residue hydrotreating unit. After determining the operating status of each piece of equipment, the operating status of all equipment is input into the operating cycle prediction model, which then predicts the operating cycle of the residue hydrotreating unit based on the operating status of each piece of equipment.

[0049] For example, after determining the operating status of the heater, reactor, hot high-performance fractionator, hot low-performance fractionator, cold high-performance fractionator, cold low-performance fractionator, flash tank, and distillation column in the residue hydrotreating unit, the operating status of the heater, reactor, hot high-performance fractionator, hot low-performance fractionator, cold high-performance fractionator, cold low-performance fractionator, flash tank, and distillation column can be input into the operating cycle prediction model to obtain the operating cycle of the residue hydrotreating unit.

[0050] Understandably, before inputting the operating status of each piece of equipment into the operating cycle prediction model, the operating cycle prediction model for each piece of equipment can be pre-trained. This can be achieved by performing the following steps: First, collect a large number of sample operating statuses for each piece of equipment, and determine the sample operating cycle of the corresponding residue hydrotreating unit through manual annotation. Then, based on the sample operating statuses of each piece of equipment and the sample operating cycles of the corresponding residue hydrotreating unit, train the second initial model to obtain the operating cycle prediction model for each piece of equipment.

[0051] Compared to traditional methods that rely on chemical reaction mechanisms to construct operating cycle equations for cycle prediction, this invention not only avoids the need to retrain the operating state prediction model and operating cycle model for each device when the feed oil properties data change, thus avoiding the time-consuming problem of constructing new equations based on the changed parameters in traditional methods, but also accurately predicts the operating cycle of the residue hydrotreating unit based on the changed feed oil properties data of each device.

[0052] The method for predicting the operating cycle of a residue oil hydrotreating unit provided in this invention uses an operating state prediction model based on the feed oil property data of each device to accurately predict the operating state of each device. Thus, the operating state model can be used to analyze the operating state of each device and accurately and quickly predict the operating cycle of the residue oil hydrotreating unit.

[0053] Based on the above embodiments, the operating status of each device is obtained from the operating status prediction model of each device, and then the process further includes:

[0054] Obtain the current operating status of each device;

[0055] Based on the operating status of each device, the current operating status of each device is adjusted to ensure that the current operating status of each device is consistent with the operating status of each device.

[0056] Specifically, the current operating status of each piece of equipment refers to its actual current operating status. The predicted operating status of each piece of equipment also refers to its predicted operating status. If the current operating status of each piece of equipment is inconsistent with its predicted operating status, it indicates that the current operating status does not match the feed oil property data, and adjustments to the current operating status are required. If the current operating status of each piece of equipment is consistent with its predicted operating status, it indicates that the current operating status matches the feed oil property data, and no adjustments are needed.

[0057] like Figure 2 As shown, the equipment in the residue hydrotreating unit includes a heater, reactor, hot high-efficiency fractionator, hot low-efficiency fractionator, cold high-efficiency fractionator, cold low-efficiency fractionator, flash tank, and distillation column. The corresponding operating state models for each piece of equipment are M1, M2, M3, M4, M5, M6, M7, and M8, respectively. The operating cycle prediction model for the residue hydrotreating unit is M. c Operating cycle prediction model M c The operating cycle C is obtained by predicting the operating state output from each operating state model. Additionally, the current operating temperature t of each device is acquired. i Pressure p i Flow f i If the operating status output by the operating status model of each device is inconsistent with the current operating status of each device, the adjustment trend of the current operating status of each device (such as increasing or decreasing) is determined based on the difference between the two, thereby providing decision support for manually adjusting the current operating status of each device.

[0058] Based on any of the above embodiments, the current operating state of each device is adjusted according to its operating state, including:

[0059] Based on the operating status of each device and its current operating status, determine the adjustment amount for the current operating status of each device.

[0060] Adjust the current operating status of each device based on the adjustment amount of the current operating status of each device.

[0061] Specifically, after determining the operating status of each piece of equipment and its current operating status, the adjustment amount for the current operating status of each piece of equipment can be determined based on the difference between the two. That is, the adjustment amount for the current operating status of each piece of equipment can be determined based on the following formula:

[0062] Δ=AB

[0063] In the formula, Δ represents the adjustment amount of the current operating status of each device, A represents the operating status of each device, and B represents the current operating status of each device.

[0064] After determining the adjustment amount for the current operating status of each device, the current operating status of each device can be adjusted based on the adjustment amount. Optionally, the operating status of each device may include predicted temperature t1, predicted pressure p1, and predicted flow rate f1, and the current operating status of each device may include current temperature t2, current pressure p2, and current flow rate f2. Accordingly, the adjustment amount for the current operating status may include temperature adjustment Δt, pressure adjustment Δp, and flow rate adjustment Δf, which are calculated based on the following formulas:

[0065] Δt=t1-t2

[0066] Δp=p1-p2

[0067] Δf=f1-f2

[0068] After determining the adjustment amount for the current operating status, the current operating status of each device is adjusted based on the adjustment amount to make the current operating status of each device the same as the operating status of each device.

[0069] Based on any of the above embodiments, the training steps for the operation status prediction model of each device include:

[0070] Determine the first initial model for predicting the operating status of each piece of equipment;

[0071] Data preprocessing is performed on the sample feed oil properties data and sample operating status of each device. Based on the preprocessed sample feed oil properties data and sample operating status of each device, the first initial model is trained to obtain the operating status prediction model of each device.

[0072] Specifically, the first initial model refers to the basic model of the operation status prediction model. After acquiring sample feed oil property data and sample operation status of each device, the sample feed oil property data and sample operation status of each device are preprocessed. Based on the preprocessed sample feed oil property data of each device, the first initial model is trained using the sample operation status of each device as a label, to obtain the operation status prediction model for each device pair. The first initial model can be trained using methods such as multilayer linear perceptrons, support vector machines, and decision trees; this embodiment of the invention does not specifically limit the methods used.

[0073] In addition, when preprocessing the sample feed oil properties data and sample operating status of each device, data dimensionality reduction and feature normalization can be performed on each data in sequence.

[0074] For example, the sample operation status of each device can be represented by matrix λ1=(T i P i F i ) n That is, it is denoted as:

[0075]

[0076] The sample feed oil property data for each device can be represented as matrix λ2=(S j CCR j N j Ni j V j ) n That is, it is denoted as:

[0077]

[0078] Based on any of the above embodiments, the training steps of the operation cycle prediction model include:

[0079] Determine the second initial model for the operating cycle prediction model;

[0080] The sample operating status of each device and the sample operating cycle of the residue hydrotreating unit were preprocessed. Based on the preprocessed sample operating status of each device and the sample operating cycle of the residue hydrotreating unit, the second initial model was trained to obtain the operating cycle prediction model.

[0081] Specifically, the second initial model refers to the base model of the operation cycle prediction model. After obtaining the sample operation status of each device and the sample operation cycle of the residue hydrotreating unit, the sample operation status of each device and the sample operation cycle of the residue hydrotreating unit are preprocessed. Based on the preprocessed sample operation status of each device, the model is used as input, and the sample operation cycle of the residue hydrotreating unit is used as a label to train the second initial model, thereby obtaining the operation cycle prediction model. The sample operation status of each device can be determined using the operation status prediction model for each device. When training the second initial model, methods such as multilayer linear perceptrons, support vector machines, and decision trees can be used; this embodiment of the invention does not specifically limit the methods used.

[0082] In addition, when preprocessing the data on the sample operating status of each device and the sample operating cycle of the residue hydrotreating unit, data dimensionality reduction and feature normalization can be performed on each data in sequence.

[0083] For example, the sample operation status of each device can be represented by matrix λ1=(T i ,P i ,F i ) nThat is, it is denoted as:

[0084]

[0085] The sample operating cycle of the residue hydrotreating unit can be represented by matrix λ3, i.e.:

[0086]

[0087] Based on any of the above embodiments, the data preprocessing steps include: performing data dimensionality reduction processing and feature normalization processing in sequence.

[0088] Specifically, data dimensionality reduction methods can include principal component analysis (PCA), discrete cosine transform (DCT), etc. After performing dimensionality reduction on each data point, the computational load of the model during training can be reduced, thus improving the model's convergence speed. To reduce data variability and facilitate subsequent feature extraction, feature normalization is performed on the dimensionality-reduced data, which facilitates feature extraction and data analysis.

[0089] Based on any of the above embodiments, such as Figure 3 As shown, the operating status prediction model and operating cycle prediction model of each device are trained based on the following process:

[0090] First, obtain the sample operating status of each device from the DCS, and obtain the sample feed oil property data of each device from the LIMS, and determine the sample operating cycle corresponding to the sample operating status of each device.

[0091] Next, the sample operating status of each device, the sample feed oil properties of each device, and the sample operating cycle were sequentially subjected to data dimensionality reduction and feature normalization.

[0092] Then, based on the processed sample operating status of each device and the sample feed oil properties data of each device, the initial model of the operating status prediction model of each device is trained to obtain the operating status prediction model of each device. Based on the processed sample operating status and sample operating cycle of each device, the initial model of the operating cycle prediction model is trained to obtain the operating cycle prediction model.

[0093] Finally, based on the operational status prediction model of each device after training, the operational status of each device is predicted, and combined with the current operational status of each device, it is determined whether to adjust the current operational status of each device in order to complete the device dynamic modulation and auxiliary decision-making.

[0094] The following describes the residual oil hydrotreating unit operation cycle prediction device provided by the present invention. The residual oil hydrotreating unit operation cycle prediction device described below can be referred to in correspondence with the residual oil hydrotreating unit operation cycle prediction method described above.

[0095] Based on any of the above embodiments, the present invention also provides a device for predicting the operating cycle of a residue hydrotreating unit, such as... Figure 4 As shown, the device includes:

[0096] Data determination unit 410 is used to determine the feed oil properties data of each device in the residue hydrotreating unit;

[0097] The status prediction unit 420 is used to input the feed oil property data of each device into the operation status prediction model of each device, and obtain the operation status of each device output by the operation status prediction model of each device.

[0098] The cycle prediction unit 430 is used to input the operating status of each device into the operating cycle prediction model to obtain the operating cycle of the residue hydrotreating unit output by the operating cycle prediction model.

[0099] The operation status prediction model for each device is trained based on the sample feed oil property data and the sample operation status of each device, while the operation cycle prediction model is trained based on the sample operation status of each device and the sample operation cycle of the residue hydrotreating unit.

[0100] Based on any of the above embodiments, the device further includes:

[0101] The acquisition unit is used to acquire the current operating status of each device after obtaining the operating status of each device from the operating status prediction model of each device.

[0102] The adjustment unit is used to adjust the current operating status of each device based on the operating status of each device, so that the current operating status of each device is consistent with the operating status of each device.

[0103] Based on any of the above embodiments, the adjustment unit includes:

[0104] The adjustment amount determination unit is used to determine the adjustment amount of the current operating state of each device based on the operating state of each device and the current operating state of each device.

[0105] The adjustment subunit is used to adjust the current operating status of each device based on the adjustment amount of the current operating status of each device.

[0106] Based on any of the above embodiments, the device further includes:

[0107] The first determining unit is used to determine the first initial model of the prediction model of the operating status of each device.

[0108] The first training unit is used to preprocess the sample feed oil property data and sample operating status of each device, and to train the first initial model based on the preprocessed sample feed oil property data and sample operating status of each device to obtain the operating status prediction model of each device.

[0109] Based on any of the above embodiments, the device further includes:

[0110] The second determining unit is used to determine the second initial model of the operation cycle prediction model;

[0111] The second training unit is used to preprocess the sample operating status of each device and the sample operating cycle of the residue hydrotreating unit, and to train the second initial model based on the preprocessed sample operating status of each device and the sample operating cycle of the residue hydrotreating unit to obtain the operating cycle prediction model.

[0112] Based on any of the above embodiments, the data preprocessing steps include: performing data dimensionality reduction processing and feature normalization processing sequentially.

[0113] Figure 5 This is a schematic diagram of the structure of the electronic device provided by the present invention, such as... Figure 5 As shown, the electronic device may include: a processor 510, a memory 520, a communication interface 530, and a communication bus 540, wherein the processor 510, memory 520, and communication interface 530 communicate with each other through the communication bus 540. The processor 510 can call logical instructions in the memory 520 to execute a method for predicting the operating cycle of the residue hydrotreating unit. This method includes: determining the feed oil properties data of each device in the residue hydrotreating unit; inputting the feed oil properties data of each device into the operating state prediction model of each device to obtain the operating state of each device output by the operating state prediction model; inputting the operating state of each device into the operating cycle prediction model to obtain the operating cycle of the residue hydrotreating unit output by the operating cycle prediction model; wherein the operating state prediction model of each device is trained based on sample feed oil properties data and sample operating states of each device, and the operating cycle prediction model is trained based on sample operating states of each device and sample operating cycles of the residue hydrotreating unit.

[0114] Furthermore, the logical instructions in the aforementioned memory 520 can be implemented as software functional units and, when sold or used as independent products, can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, essentially, or the part that contributes to the prior art, or a part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of the present invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0115] On the other hand, the present invention also provides a computer program product, the computer program product comprising a computer program stored on a non-transitory computer-readable storage medium, the computer program comprising program instructions, wherein when the program instructions are executed by a computer, the computer is able to execute the method for predicting the operating cycle of a residue hydrotreating unit provided by the above methods, the method comprising: determining the feed oil property data of each device in the residue hydrotreating unit; inputting the feed oil property data of each device into the operating state prediction model of each device respectively, to obtain the operating state of each device output by the operating state prediction model of each device; inputting the operating state of each device into the operating cycle prediction model, to obtain the operating cycle of the residue hydrotreating unit output by the operating cycle prediction model; wherein the operating state prediction model of each device is trained based on the sample feed oil property data and the sample operating state of each device, and the operating cycle prediction model is trained based on the sample operating state of each device and the sample operating cycle of the residue hydrotreating unit.

[0116] In another aspect, the present invention also provides a non-transitory computer-readable storage medium storing a computer program thereon. When executed by a processor, the computer program implements the above-described methods for predicting the operating cycle of a residue hydrotreating unit. The method includes: determining feed oil property data for each device in the residue hydrotreating unit; inputting the feed oil property data for each device into an operating state prediction model for each device to obtain the operating state of each device output by the operating state prediction model; inputting the operating state of each device into an operating cycle prediction model to obtain the operating cycle of the residue hydrotreating unit output by the operating cycle prediction model; wherein the operating state prediction model for each device is trained based on sample feed oil property data and sample operating states of each device, and the operating cycle prediction model is trained based on sample operating states of each device and sample operating cycles of the residue hydrotreating unit.

[0117] The device embodiments described above are merely illustrative. 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; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs. Those skilled in the art can understand and implement this without any creative effort.

[0118] Through the above description of the embodiments, those skilled in the art can clearly understand that each embodiment can be implemented by means of software plus necessary general-purpose hardware platforms, and of course, it can also be implemented by hardware. Based on this understanding, the above technical solutions, in essence or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute the methods described in the various embodiments or some parts of the embodiments.

[0119] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for predicting the operating cycle of a residue hydrotreating unit, characterized in that, include: Determine the feed oil properties of each unit in the residue hydrotreating plant; the feed oil properties include the content of sulfur, residual carbon, viscosity, nickel and vanadium in the feed oil; The feed oil properties data of each device are input into the operation status prediction model of each device to obtain the operation status of each device output by the operation status prediction model; the operation status of each device is characterized by the temperature, pressure and flow rate of each device. The operating status of each device is input into the operating cycle prediction model to obtain the operating cycle of the residue hydrotreating unit output by the operating cycle prediction model. The operation status prediction model of each device is trained based on the sample feed oil property data and the sample operation status of each device, and the operation cycle prediction model is trained based on the sample operation status of each device and the sample operation cycle of the residue hydrotreating unit. When training the predictive model for the operating status of each device, the sample operating status of each device is represented as a matrix. The sample feed oil property data for each device are represented as a matrix. The sample operating cycle of the residue hydrotreating unit is represented by a matrix. ; The process of obtaining the operating status of each device from the predictive model of each device's operating status further includes: Obtain the current operating status of each device; If the operating status of each device is inconsistent with the current operating status of each device, the adjustment trend of the current operating status of each device is determined based on the difference between the two, and the current operating status of each device is adjusted accordingly; if the operating status of each device is consistent with the current operating status of each device, the current operating status of each device is not adjusted.

2. The method for predicting the operating cycle of a residue hydrotreating unit according to claim 1, characterized in that, The adjustment of the current operating status of each device based on its operating status includes: Based on the operating status of each device and its current operating status, determine the adjustment amount for the current operating status of each device. Adjust the current operating status of each device based on the adjustment amount of the current operating status of each device.

3. The method for predicting the operating cycle of a residue hydrotreating unit according to claim 1, characterized in that, The training steps for the operational status prediction model of each piece of equipment include: Determine the first initial model for predicting the operating status of each piece of equipment; Data preprocessing is performed on the sample feed oil properties data and sample operating status of each device. Based on the preprocessed sample feed oil properties data and sample operating status of each device, the first initial model is trained to obtain the operating status prediction model of each device.

4. The method for predicting the operating cycle of a residue hydrotreating unit according to claim 1, characterized in that, The training steps for the operation cycle prediction model include: Determine the second initial model of the operation cycle prediction model; The sample operating status of each device and the sample operating cycle of the residue hydrotreating unit are preprocessed, and the second initial model is trained based on the preprocessed sample operating status of each device and the sample operating cycle of the residue hydrotreating unit to obtain the operating cycle prediction model.

5. The method for predicting the operating cycle of a residue hydrotreating unit according to claim 3 or 4, characterized in that, The data preprocessing steps include: performing data dimensionality reduction and feature normalization in sequence.

6. A device for predicting the operating cycle of a residue oil hydrotreating unit, characterized in that, include: The data determination unit is used to determine the feed oil property data of each device in the residue hydrotreating unit; the feed oil property data includes the content of sulfur, residual carbon, viscosity, nickel and vanadium in the feed oil; The status prediction unit is used to input the feed oil property data of each device into the operation status prediction model of each device to obtain the operation status of each device output by the operation status prediction model; the operation status of each device is characterized by the temperature, pressure and flow rate of each device. The cycle prediction unit is used to input the operating status of each device into the operation cycle prediction model to obtain the operation cycle of the residue hydrotreating unit output by the operation cycle prediction model. The operation status prediction model of each device is trained based on the sample feed oil property data and the sample operation status of each device, and the operation cycle prediction model is trained based on the sample operation status of each device and the sample operation cycle of the residue hydrotreating unit. When training the predictive model for the operating status of each device, the sample operating status of each device is represented as a matrix. The sample feed oil property data for each device are represented as a matrix. The sample operating cycle of the residue hydrotreating unit is represented by a matrix. ; The process of obtaining the operating status of each device from the predictive model of each device's operating status further includes: Obtain the current operating status of each device; If the operating status of each device is inconsistent with the current operating status of each device, the adjustment trend of the current operating status of each device is determined based on the difference between the two, and the current operating status of each device is adjusted accordingly; if the operating status of each device is consistent with the current operating status of each device, the current operating status of each device is not adjusted.

7. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the program, it implements the method for predicting the operating cycle of the residue hydrotreating unit as described in any one of claims 1 to 5.

8. A non-transitory computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, it implements the method for predicting the operating cycle of the residue hydrotreating unit as described in any one of claims 1 to 5.

9. A computer program product, comprising a computer program, characterized in that, When the computer program is executed by the processor, it implements the method for predicting the operating cycle of the residue hydrotreating unit as described in any one of claims 1 to 5.

Images