Method and device for predicting activity of catalytic cracking catalyst, electronic equipment and storage medium

Abstract

This invention discloses a method and apparatus for predicting the activity of catalytic cracking catalysts. The method includes the following steps: A. Collecting product output data from the catalytic cracking unit and performing data preprocessing; B. Dividing the products into multiple categories based on their composition, required catalyst type, and activity range, and calculating the cracking cut-off ratio for each category; C. Using the data obtained from the cracking cut-off ratio calculation as input variables for a catalyst activity prediction model, and training the model; D. Predicting the catalyst activity in the catalytic cracking unit using the trained model. The method and apparatus of this invention can classify and calculate the output product data from the catalytic cracking unit, and use the processed data as input parameters for the algorithm model to predict catalyst activity, resulting in higher prediction accuracy.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of chemical process dynamic monitoring, and particularly relates to a catalytic cracking catalyst activity prediction method and device, an electronic equipment and a storage medium. BACKGROUND

[0002] The catalyst is used to change the reaction process, reduce the reaction activation energy, and thus improve the reaction speed, so as to promote the reaction speed of the target product. The catalyst can play an important role in the yield and quality of the product.

[0003] The catalytic cracking catalyst not only improves the speed of decomposition, aromatization and other reactions, but also improves the speed of isomerization, hydrogen transfer and other reactions, so that the production capacity of the catalytic cracking device is not only larger than that of the thermal cracking device, but also the octane number of the obtained gasoline is high and the stability is good. At the same time, the addition of the catalyst can make the catalytic cracking reaction produce the target product, which depends on the adjustment of the catalyst activity. According to different production requirements, selectively controlling the catalyst activity value index becomes an important part of the catalytic cracking process. Therefore, if the catalyst activity value can be accurately predicted, it is helpful to improve the yield of the target product, help enterprises avoid waste, and has important significance for improving economic benefits.

[0004] The catalyst of the catalytic cracking device is different from that of other oil refining devices. The catalyst is regenerated in the regenerator after participating in the reaction in the reactor, and then is recycled to the reactor. Fresh catalyst or balanced catalyst needs to be continuously added and removed, and the reaction activity is monitored by laboratory tests to ensure the performance of the catalyst. The catalyst of other oil refining devices is generally filled into the reaction unit once, and is replaced once after reaching the service life after a certain period of operation. Therefore, the change rule of the catalyst activity of the catalytic cracking device and other devices is quite different, and has certain particularity.

[0005] In the prior art, there is a research on predicting the catalyst activity of the catalytic cracking device by using a neural network model. For example, the paper "Application of BP Network in Catalyst Activity Prediction" (the author is Liu Xinping et al.) published in Computer Engineering and Design, Vol. 29, No. 11, takes the front-end data of the device as the input parameters of the BP neural network model, such as: coke burning load, regeneration temperature, reaction time, catalyst replacement rate, etc. The neural network model is trained by using these parameters, and the catalyst activity can be predicted. However, for the catalytic cracking device, the mixing, reaction, regeneration and circulation of the catalyst in the catalytic reactor and the regenerator are a series of complex processes. In addition, the type and amount of fresh catalyst are adjusted frequently, which causes uncertain interference to the catalyst activity prediction value. The catalyst activity prediction value output by the model trained by the front-end data has little guiding significance for production.

[0006] Therefore, there is an urgent need for a catalyst activity prediction method for a catalytic cracking device, which can have higher prediction accuracy and can guide real-time prediction of catalyst activity in actual production process and dynamic optimization of process.

[0007] The information disclosed in this BACKGROUND section is only for the purpose of increasing the understanding of the general background of the application and should not be taken as an acknowledgement or any form of suggestion that this information forms prior art that is already known to those of ordinary skill in the art. SUMMARY

[0008] The purpose of the present application is to provide a catalytic cracking catalyst activity prediction method and device, which classifies and calculates the product data of a catalytic cracking device, takes the data processed as input parameters of an algorithm model, and predicts the catalyst activity, thereby having higher prediction accuracy.

[0009] To achieve the above-mentioned purpose, according to a first aspect of the present application, a catalytic cracking catalyst activity prediction method is provided, which comprises the following steps: A, collecting product output data of a catalytic cracking device and performing data preprocessing; B, dividing products into multiple categories according to the composition of the products, the required catalyst types, and the activity value range, and calculating the cracking cutting proportion of the categories; C, taking the data after the cracking cutting proportion calculation as input variables of a catalyst activity prediction model, and training the model; D, predicting the catalyst activity in the catalytic cracking device through the trained model.

[0010] Further, in the above technical solution, the product output data can specifically be the yield of each product, and the products include but are not limited to propane, propylene, refined liquefied gas, purified dry gas, light diesel oil, carbon four, alkylated material, and stable gasoline, etc.

[0011] Further, in the above technical solution, the cracking cutting proportion calculation in step B can specifically be: calculating the proportion of the yield of each category of product relative to the total yield of the catalytic cracking device.

[0012] Further, in the above technical solution, the division standard of the multiple categories in step B includes the carbon chain length of the products.

[0013] Further, in the above technical solution, according to the carbon chain length, the above eight products can be divided into four categories, and the data after the cracking cutting proportion calculation in step C can include: a first characteristic parameter, which is the ratio of the yield of the purified dry gas to the total yield; a second characteristic parameter, which is the ratio of the sum of the yields of the refined liquefied gas, propane, propylene, alkylated material, and carbon four to the total yield; a third characteristic parameter, which is the ratio of the yield of the stable gasoline to the total yield; and a fourth characteristic parameter, which is the ratio of the yield of the light diesel oil to the total yield.

[0014] Further, in the technical solution, the step A further comprises: collecting the measured value of catalyst activity in the same period of product output data and performing same-dimension processing on the data.

[0015] Further, in the technical solution, the data preprocessing in the step A can comprise data extraction in normal working condition period and abnormal data elimination.

[0016] Further, in the technical solution, the data extraction in normal working condition period can specifically be: retaining the data in normal working condition period and deleting the data in shutdown state; and the abnormal data elimination can specifically be: retaining the collected product output data within a preset data threshold range and eliminating the data outside the threshold range.

[0017] Further, in the technical solution, the step D further comprises: controlling the addition amount and reuse amount of the catalyst according to the predicted value of the catalyst activity.

[0018] Further, in the technical solution, the catalyst activity prediction model can adopt a BP neural network model.

[0019] According to the second aspect of the present application, the present application provides a catalyst activity prediction device for catalytic cracking, comprising: a collection and processing module for collecting product output data of a catalytic cracking device and performing data preprocessing; a parameter calculation module for dividing products into multiple categories according to the composition of the products, the required catalyst types and the activity value range, and calculating the cracking cutting proportion of the categories; a model training module for training the model by taking the data after the cracking cutting proportion calculation as the input variable of the catalyst activity prediction model; and an activity prediction module for predicting the catalyst activity in the catalytic cracking device by using the trained model.

[0020] Further, in the technical solution, the collection and processing module can further comprise: a normal working condition data extraction submodule for extracting data in normal working condition period and deleting data in shutdown state; and an abnormal data elimination submodule for retaining the collected product output data within a preset data threshold range and eliminating the data outside the threshold range.

[0021] Further, in the technical solution, the device can further comprise an activity control module for controlling the addition amount and reuse amount of the catalyst according to the predicted value of the catalyst activity.

[0022] According to a third aspect of the present application, the present application provides a catalytic cracking catalyst activity prediction electronic device, comprising: at least one processor; and a memory connected in communication with the at least one processor; wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor to cause the at least one processor to perform the catalytic cracking catalyst activity prediction method as described above.

[0023] According to a fourth aspect of the present application, the present application provides a non-transitory computer-readable storage medium storing computer-executable instructions for causing a computer to perform the catalytic cracking catalyst activity prediction method as described above.

[0024] Compared with the prior art, the present application has one or more beneficial effects as follows:

[0025] 1) The present application has found, through a large amount of data analysis and business analysis, that the catalytic cracking catalyst activity prediction value output by the front-end data training model in the prior art has no obvious regularity and is not strong in guiding production. It is found that using product data output by the catalytic cracking device as an input parameter of an algorithm model can more effectively predict the catalyst activity, and further processing the product data cracking cutting ratio can make the prediction accuracy higher.

[0026] 2) The present application can avoid the long work cycle and the lagging defect caused by manual inspection in the addition and activity value determination of the catalytic cracking device catalyst, and can also avoid the interference caused by the use characteristics and complexity of the catalytic cracking device catalyst. The present application classifies each output product according to the carbon chain length by using the output product yield data, can analyze and calculate according to the cracking cutting ratio of the product in the production process, and obtains the predicted activity value of the catalyst through the BP neural network model. The present application not only greatly shortens the acquisition period of the catalyst activity value, but also effectively improves the accuracy of the catalyst activity prediction, and provides accurate catalyst micro-reaction activity prediction results for the actual production site staff.

[0027] The above description is only a summary of the technical solutions of the present application. In order to more clearly understand the technical means of the present application and can be implemented according to the content of the specification, and at the same time, in order to make the above and other purposes, technical features and advantages of the present application more easily understood, one or more preferred embodiments are listed below, and are described in detail as follows with reference to the accompanying drawings. BRIEF DESCRIPTION OF DRAWINGS

[0028] Figure 1 is a flowchart of the catalytic cracking catalyst activity prediction method of the embodiment 1 of the present application.

[0029] Figure 2is a flowchart of a catalytic cracking catalyst activity prediction method of embodiment 2 of the present application.

[0030] Figure 3 is a structural diagram of a catalytic cracking catalyst activity prediction device of embodiment 3 of the present application.

[0031] Figure 4 is a structural diagram of a catalytic cracking catalyst activity prediction electronic device of embodiment 4 of the present application. DETAILED DESCRIPTION

[0032] The specific embodiments of the present application will be described in detail below with reference to the drawings, but the scope of protection of the present application is not limited by the specific embodiments.

[0033] Unless otherwise explicitly indicated, throughout the specification and claims, the term "comprise" or its variants such as "comprises" or "comprising" will be understood to encompass the stated element or components, but not to exclude other elements or components.

[0034] In this document, spatially relative terms, such as "beneath", "below", "lower", "above", "upper", "on", "directly on", and the like, can be used to describe one element's or feature's relationship to another element or feature as illustrated in the drawings. The spatially relative terms are intended to encompass different orientations of the object in use or operation in addition to the orientations depicted in the drawings. For example, if an object is turned over in the drawings, a dependent object or feature can be below or on the object or feature in the drawings. Thus, the exemplary term "below" can encompass both an orientation of below and above. The object or feature can also be oriented in other ways (rotated 90 degrees or at other orientations) and be understood to encompass the spatially relative terms used herein. The terms "first", "second", and the like, do not necessarily denote any ordinal, chronological or other relative importance, but are used to distinguish one element from another.

[0035] In this document, the terms "first", "second", and the like are used to distinguish two different elements or parts, and are not used to limit a specific position or relative relationship. In other words, in some embodiments, the terms "first", "second", and the like can be interchanged with each other.

[0036] The method, system, electronic device and storage medium of the present application will be described in more detail below in the form of specific embodiments, and it should be understood that the embodiments are exemplary only, and the present application is not limited thereto.

[0037] Through data analysis and business analysis research, it is shown that, due to the mixing, reaction, regeneration and circulation of the catalyst in the catalytic reactor and the regenerator being a series of complex processes, and the continuous adjustment of the type and addition amount of fresh catalyst having a high frequency, all of which will cause uncertain interference to the catalyst activity value, therefore, the catalyst activity prediction value output by the front-end data training model in the prior art has little guiding significance to the production. It is found through research that, the product data output by the catalytic cracking device can be used as input parameters of an algorithm model to more effectively predict the catalyst activity, and further, through processing of the cracking cutting proportion of the product data, the prediction accuracy can be higher.

[0038] Example 1

[0039] As shown in FIG. 1, the catalytic cracking activity prediction method of the present embodiment 1 comprises the following steps: Figure 1

[0040] Step S101, collecting product output data of the catalytic cracking device and performing data preprocessing.

[0041] Specifically, the product output data can be specifically the yield of each product of the catalytic cracking device, and the products include but are not limited to propane, propylene, refined liquefied gas, purified dry gas, light diesel oil, carbon four, alkylated material and stabilized gasoline, etc. The product yield data includes historical product output data and real-time output data, so as to be used for subsequent model training and testing. The present application is to predict the activity of the catalyst by using the output data of the products through an algorithm model, therefore, in the data collection process, the historical catalyst activity data in the catalytic cracking device is also collected and recorded for subsequent training of the algorithm model.

[0042] Step S102, dividing the products into multiple categories according to the composition of the products, the required catalyst type and the activity value range, and performing cracking cutting proportion calculation on the categories. It needs to be explained here that: the catalyst of the catalytic cracking device is different from that of other oil refining devices, the catalyst thereof is regenerated after being burned to remove the attached carbon in the regenerator after participating in the reaction in the reactor, and then is recycled to the reactor, and fresh catalyst needs to be continuously added or balanced catalyst needs to be unloaded, and the reaction activity can be monitored through laboratory test to ensure the performance of the catalyst. The catalyst of other oil refining devices is generally filled into the reaction unit once, and is replaced once after running for 3-5 years, so the change rule of the catalyst activity of the catalytic cracking device and other devices is very different, and has particularity. Due to the particularity of the catalyst of the catalytic cracking device and the complexity in the use process, the prediction difficulty is relatively high.

[0043] ​The present application finds, through data analysis and business analysis, that by classifying catalytic cracking products, the yield proportion of each classified product has a stronger correlation with the catalyst activity value, which can better reflect the activity state of the catalyst. Therefore, the present embodiment divides different products of catalytic cracking into multiple categories, and calculates the cracking cutting proportion of these categories. Preferably but not limitatively, the cracking cutting proportion calculation is specifically calculating the proportion of the yield of each category of product relative to the total yield of the catalytic cracking device.

[0044] Step S103, the data after the cracking cutting proportion calculation is taken as an input variable of the catalyst activity prediction model, and the model is trained. In this step, the ratio of the yield proportion of each category of product relative to the total yield of the catalytic cracking device calculated by the product output historical data collected is taken as an input parameter of the model (i.e. calculated by the product output historical data collected). The training process of the model can be analyzed and adjusted by combining the actual measured catalyst activity value historical data with the corresponding predicted activity value obtained by the catalyst activity prediction model, has the function of autonomous learning, and through the forward and backward propagation in the model, the reliability of the predicted activity value is enhanced. The autonomous learning function of the prediction model is a continuous optimization process, which compares the actual measured catalyst activity value historical data with the calculated predicted activity value, and continuously optimizes the calculation combined with the input parameters, so that the predicted activity value converges and continuously approaches the actual measured catalyst activity value, thereby realizing the prediction training of the model.

[0045] Preferably but not limitatively, the catalyst activity prediction can adopt a neuron network algorithm, and the algorithm model can adopt a BP neural network model.

[0046] The construction process of the prediction model is as follows:

[0047] Firstly, set the initial weight value k and b.

[0048] Secondly, weighted sum the input layer nodes of the model to calculate the hidden layer value, specifically, the following formula (1) is adopted:

[0049]

[0050] Wherein, Ff,j is the hidden layer value, j is the value serial number of the hidden layer, n is the cracking cutting proportion of the output product under various working conditions, i.e. the corresponding value of each layer node, k nj is the weight value corresponding to the node value n.

[0051] Thirdly, the sigmoid activation formula (2) is executed as follows:

[0052]

[0053] Fourthly, the predicted value of the catalyst activity is calculated by using formula (3) based on formula (1) and formula (2):

[0054]

[0055] wherein f(x) is the predicted value of the catalyst activity of the catalytic cracking device, and b is the second layer weight value.

[0056] Fifthly, the error t = (f(x) - f b ) 2 Converge when the error is close to zero.

[0057] wherein t is the error, f b is the set catalyst activity sample value, and converge when t < 0.0001.

[0058] Sixthly, if the error value does not meet the standard, the two layer weight values k and b are updated by reverse calculation, and the calculation formula (4) is as follows:

[0059]

[0060] wherein X is the set learning rate.

[0061] Seventhly, the calculation in the first step to the sixth step is repeated by using the updated weight values until convergence, and the best predicted value of the catalyst activity is obtained.

[0062] Step S104: The catalyst activity in the catalytic cracking device is predicted by using the trained model.

[0063] The method for predicting the catalytic cracking catalyst activity in the embodiment 1 can avoid the long working period and the lagging defects caused by manual inspection, and can avoid the interference caused by the use characteristics and complexity of the catalytic cracking catalyst. The catalyst activity is predicted without using the data of the front end of the catalytic cracking reaction. In the embodiment, the yield data of the output products are used to analyze and calculate the cracking cutting proportion of the products in the production process, and the predicted activity value of the catalyst is obtained by using the algorithm model. The method can greatly shorten the acquisition period of the catalyst activity value, can effectively improve the accuracy of the catalyst activity prediction, and can provide accurate catalyst micro-reaction activity prediction results for the on-site staff in the actual production.

[0064] Example 2

[0065] The embodiment is a more optimized embodiment based on the embodiment 1. The following steps are used:

[0066] Step S201, collect product output data of the catalytic cracking unit and perform data preprocessing. The data collection process is the same as that in step S101 in embodiment 1, and will not be repeated here. The data preprocessing after data collection of this embodiment specifically includes normal working condition period data extraction and abnormal data elimination. Preferably but not limitedly, the normal working condition period data extraction can be specifically as follows: retaining the data of the normal working condition period and deleting the data under the shutdown state; the abnormal data elimination can be specifically as follows: according to the pre-set data threshold range, retaining the collected product output data within the threshold range and eliminating the data outside the threshold range. In addition, during the collection of product output data, the catalyst activity measured value of the same period of the product output data is further collected and the data is processed in the same dimension, which can facilitate the calculation and comparison in the model training process.

[0067] Step S202, according to the composition of the product, the required catalyst type and the activity value range, the product is divided into multiple categories and the cracking cutting proportion of these categories is calculated. Compared with step S102 in embodiment 1, this step specifically divides the eight products in embodiment 1 into four categories according to the carbon chain length of the catalytic cracking output products. It is known to those skilled in the art that the catalytic cracking reaction is carried out according to the carbonium ion mechanism, and the catalyst promotes the cracking, isomerization and aromatization reactions, and the isomerization and aromatization convert straight-chain hydrocarbons with low octane number into isomeric hydrocarbons and aromatic hydrocarbons with high octane number. According to the deep analysis of the catalytic cracking reaction by the inventors of the present application, the eight products of propane, propylene, refined liquefied gas, purified dry gas, light diesel oil, carbon four, alkylated material and stable gasoline are cut into components, wherein the main components of the purified dry gas are C1-C2 hydrocarbons, which are cut into one category; the main components of the liquefied gas, propane, propylene and alkylated material are C3-C4 hydrocarbons, which can be cut into one category with carbon four; the main components of the stable gasoline are C5-C12 hydrocarbons, and the main components of the stable diesel are C10-C22 hydrocarbons, which can be cut separately. The specific cutting and division are into four categories, which are the first category (purified dry gas), the second category (including refined liquefied gas, propane, propylene, alkylated material and carbon four), the third category (stable gasoline) and the fourth category (light diesel oil), and the yield and total yield proportion of the four categories are calculated to obtain the cracking cutting proportion data of the four categories.

[0068] The cracking cutting proportion data of the four categories is specifically as follows: the first characteristic parameter is the ratio of the yield of the purified dry gas to the total yield; the second characteristic parameter is the ratio of the sum of the yields of the refined liquefied gas, propane, propylene, alkylated material and carbon four to the total yield; the third characteristic parameter is the ratio of the yield of the stable gasoline to the total yield; and the fourth characteristic parameter is the ratio of the yield of the light diesel oil to the total yield.

[0069] Step S203, the four characteristic parameters calculated by the cracking cutting ratio in step S202 are taken as input variables of the catalyst activity prediction model, and the model is trained. The construction process and the training process of the prediction model are the same as those in step S103, and will not be described here.

[0070] Step S204, the catalyst activity in the catalytic cracking device is predicted by the trained model.

[0071] Step S205, the addition amount and the reuse amount of the catalyst are controlled according to the prediction result of the catalyst activity in step S204.

[0072] In this embodiment, through deep data analysis and business analysis, different eight products are divided into four categories according to the carbon chain length of different products of catalytic cracking, the input parameters of four algorithm models are obtained by calculating the cracking cutting ratio of the four categories, and the activity value of the catalyst is predicted by further training the algorithm model. Not only the acquisition period of the catalyst activity value is greatly shortened, but also the accuracy of the catalyst activity prediction is effectively improved, and accurate catalyst micro-reaction activity prediction results are provided for the actual production site staff.

[0073] The technical effects of the present application are illustrated by a specific example and a comparative example as follows:

[0074] Firstly, 300,000 pieces of original data of a catalytic cracking entity device from March 4, 2018 to February 24, 2021 are collected for data analysis and correlation analysis, and eight yield parameters of product output are reserved: propane, propylene, refined liquefied gas, purified dry gas, light diesel oil, carbon four, alkylated material and stable gasoline.

[0075] Secondly, the data during shutdown is deleted; the data is deleted according to the respective value range of product yield, and the values within the value range are reserved.

[0076] Thirdly, from the processed data, according to the actual detection value acquisition time of the catalyst activity, 207 pieces of data containing the catalyst activity value are selected, and the average data of the product yield six hours before and after the detection time point is calculated. The catalyst activity value and the product yield data are processed in the same dimension, and 150 pieces of data from March 4, 2018 to February 18, 2020 are selected as model training data, which are input into the model for training. In order to compare and verify the prediction effect of the present application, the data processed by the cracking cutting ratio, the product yield ratio data without cracking cutting and the yield calculation data of each product are taken as the input parameters of the algorithm model, and the model is trained in three ways. 57 pieces of data from February 19, 2020 to February 24, 2021 are taken as test data, and the catalyst activity measured data are taken as result comparison data to test the accuracy of the model.

[0077] Finally, the three methods are compared, and the prediction results show that the prediction accuracy of the BP neural network model using four input parameters (i.e., the first to fourth characteristic parameters after the cracking cutting ratio calculation) is 95.80%; using only the yield ratio data of eight products, the prediction accuracy is only 87.65%; using the yield calculation data of eight products, the prediction accuracy is 91.32%. That is, using the first to fourth characteristic parameters after the cracking cutting ratio calculation of the present application as the input parameters of the model for prediction, the prediction accuracy is obviously higher than that of using only the yield ratio data of eight products and the yield calculation data of eight products (see Table 1).

[0078] Table 1

[0079]

[0080] Example 3

[0081] As Figure 3 shown, the catalytic cracking catalyst activity prediction device of the present embodiment includes a collection and processing module 301, a parameter calculation module 302, a model training module 303, and an activity prediction module 304. Among them, the collection and processing module 301 is used to collect the product output data of the catalytic cracking device and perform data preprocessing; the parameter calculation module 302 is used to divide the products into multiple categories according to the composition of the products, the required catalyst types and the activity value range, and to calculate the cracking cutting ratio of these categories; the model training module 303 is used to use the data after the cracking cutting ratio calculation as the input variable of the catalyst activity prediction model, and train the model; the activity prediction module 304 is used to predict the catalyst activity in the catalytic cracking device through the trained model.

[0082] Further, the collection and processing module can further include: a normal working condition data extraction submodule for extracting data during normal working condition period and deleting data under shutdown and blowout state; an abnormal data elimination submodule for retaining the collected product output data within a pre-set data threshold range and eliminating data outside the threshold range.

[0083] Further, the prediction device of the present embodiment can further include an activity control module 305, which can be used to control the addition amount and reuse amount of the catalyst according to the predicted value of the catalyst activity.

[0084] The present embodiment is a virtual device embodiment corresponding to the methods of embodiments 1 and 2, which can achieve the same technical effects as the method embodiments.

[0085] Example 4

[0086] Figure 4 Fig. 1 is a schematic diagram of a hardware structure of a catalytic cracking catalyst activity prediction electronic device according to an embodiment of the present application. The device (such as a terminal, a server, etc.) comprises one or more processors 610 and a memory 620. The device can further comprise an input device 630 and an output device 640, taking the processor 610 as an example.

[0087] The processor 610, the memory 620, the input device 630 and the output device 640 can be connected by a bus or other means.

[0088] The memory 620, as a non-transitory computer readable storage medium, can be used to store non-transitory software programs, non-transitory computer executable programs and modules. The processor 610 executes various functional applications and data processing of the electronic device by running the non-transitory software programs, instructions and modules stored in the memory 620, that is, implements the processing method of the above method embodiments.

[0089] The memory 620 can comprise a program storage area and a data storage area, wherein the program storage area can store an operating system and at least one application required by a function; the data storage area can store data, etc. In addition, the memory 620 can comprise a high-speed random access memory, and can further comprise a non-transitory memory, such as at least one magnetic disk storage device, a flash memory device, or other non-transitory solid-state storage device. In some embodiments, the memory 620 can optionally comprise a memory disposed remotely with respect to the processor 610, which can be connected to the processing device through a network. Examples of the above network include but are not limited to the Internet, an intranet, a local area network, a mobile communication network, and a combination thereof.

[0090] The input device 630 can receive input digital or character information and generate signal input. The output device 640 can comprise a display device such as a display screen.

[0091] The one or more modules are stored in the memory 620, and when executed by the one or more processors 610, the following steps are performed: A, collecting product output data of a catalytic cracking device and performing data preprocessing; B, dividing the products into multiple categories according to the composition of the products, the required catalyst types and the activity value range, and calculating the cracking cutting proportion of the categories; C, taking the data after the cracking cutting proportion calculation as an input variable of a catalyst activity prediction model, and training the model; D, predicting the catalyst activity in the catalytic cracking device through the trained model.

[0092] The above electronic device can execute the method provided by the embodiments of the present application, and has the corresponding functional modules and beneficial effects of executing the method. Technical details not described in detail in the embodiments can be referred to the method provided by other embodiments of the present application.

[0093] The apparatus embodiments described above are merely illustrative, wherein the units described as separate components can or can not be physically separate, and the components displayed as units can or can not be physical units, i.e., can be located in one place, or can be distributed on multiple network units. Part or all of the modules can be selected according to actual needs to achieve the purpose of the embodiment.

[0094] From the above description of the embodiments, those skilled in the art can clearly understand that the embodiments can be implemented by means of software plus a general hardware platform, and of course, can also be implemented by hardware. Based on such understanding, the above technical solutions essentially or in other words, the part that contributes to the related art, can be embodied in the form of a software product, which can be stored in a computer readable storage medium, such as a ROM / RAM, a magnetic disk, an optical disk, etc., and includes a number of instructions to make a computer device (which can be a personal computer, a server, or a network device, etc.) execute the methods described in the various embodiments or some parts of the embodiments.

[0095] Example 5

[0096] The embodiment provides a non-transitory computer readable storage medium, which stores computer executable instructions, and the instructions are used to make a computer execute a catalytic cracking catalyst activity prediction method, including the following steps: A, collecting product output data of a catalytic cracking device and performing data preprocessing; B, according to the composition of the product, the required catalyst type and the activity value range, the product is divided into multiple categories and the cutting ratio of the categories is calculated; C, the data after the cutting ratio calculation is taken as an input variable of a catalyst activity prediction model, and the model is trained; D, the catalyst activity in the catalytic cracking device is predicted through the trained model.

[0097] The foregoing description of specific exemplary embodiments of the application is intended to be illustrative only and is not intended to limit the application to the precise forms described. Many modifications and variations are possible in light of the above teachings without departing from the spirit or essential characteristics of the application. The exemplary embodiments were chosen and described in order to explain the principles of the application and its practical application and to allow others skilled in the art to understand the application for various exemplary embodiments and with various modifications as are suited to the particular use contemplated. Any simple modification, equivalent replacement, and modification of the above-described exemplary embodiments should fall within the protection scope of the application.

Claims

1. A method for predicting the activity of a catalytic cracking catalyst, characterized in that, Includes the following steps: A. Collect product output data from the catalytic cracking unit and perform data preprocessing; the product output data specifically refers to the output of each product, including propane, propylene, refined liquefied petroleum gas, purified dry gas, light diesel oil, C4, alkylate feedstock, and stabilized gasoline. B. Based on the composition of the product, the required catalyst type, and the activity range, the product is divided into multiple categories, and the cracking ratio of each category is calculated. The criteria for classifying multiple categories include the carbon chain length of the product. The product is divided into four categories based on the carbon chain length. The cracking ratio calculation specifically involves calculating the proportion of the product output of each category relative to the total output of the catalytic cracking unit. C. Use the data calculated from the cracking and cutting ratio as the input variable for the catalyst activity prediction model, and train the model accordingly. The data calculated based on the cracking and cutting ratio includes: a first characteristic parameter, which is the ratio of the output of purified dry gas to the total output; a second characteristic parameter, which is the ratio of the sum of the outputs of refined liquefied petroleum gas, propane, propylene, alkylate, and C4 to the total output; a third characteristic parameter, which is the ratio of the output of stabilized gasoline to the total output; and a fourth characteristic parameter, which is the ratio of the output of light diesel oil to the total output. D. Predict the catalyst activity in the catalytic cracking unit using the trained model.

2. The method for predicting the activity of a catalytic cracking catalyst according to claim 1, characterized in that, Step A further includes: collecting the measured values ​​of catalyst activity for the same period of time based on product output data and performing data processing in the same dimension.

3. The method for predicting the activity of a catalytic cracking catalyst according to claim 1, characterized in that, The data preprocessing in step A includes data extraction during normal operating periods and removal of abnormal data.

4. The method for predicting the activity of a catalytic cracking catalyst according to claim 3, characterized in that, The extraction of data during normal operating periods specifically involves: retaining data during normal operating periods and deleting data during shutdown and furnace stoppage states; the removal of abnormal data specifically involves: retaining the product output data collected within a pre-set data threshold range and removing data outside the threshold range.

5. The method for predicting the activity of a catalytic cracking catalyst according to claim 1, characterized in that, Step D is followed by controlling the amount of catalyst added and the amount of catalyst recycled based on the predicted value of the catalyst activity.

6. The method for predicting the activity of a catalytic cracking catalyst according to claim 1, characterized in that, The catalyst activity prediction model uses a BP neural network model.

7. A device for predicting the activity of a catalytic cracking catalyst, characterized in that, The method described in any one of claims 1 to 6 includes: The data acquisition and processing module is used to collect product output data from the catalytic cracking unit and perform data preprocessing. The parameter calculation module is used to classify the product into multiple categories and calculate the cracking and cutting ratio of the categories according to the product's composition, the required catalyst type, and the activity value range. The model training module is used to train the model by using the data calculated from the cracking and cutting ratio as the input variable of the catalyst activity prediction model. An activity prediction module is used to predict the catalyst activity in the catalytic cracking unit using the trained model.

8. The catalytic cracking catalyst activity prediction device according to claim 7, characterized in that, The acquisition and processing module further includes: The normal operating condition data extraction submodule is used to extract data during normal operating periods and delete data under shutdown or furnace shutdown conditions. The abnormal data removal submodule retains the product output data collected within a pre-set data threshold range and removes data outside the threshold range.

9. The catalytic cracking catalyst activity prediction device according to claim 7, characterized in that, Also includes: An activity control module is used to control the amount of catalyst added and the amount of catalyst recycled based on the predicted value of the catalyst activity.

10. An electronic device for predicting the activity of a catalytic cracking catalyst, characterized in that, include: At least one processor; as well as A memory communicatively connected to the at least one processor; wherein, The memory stores instructions that can be executed by the at least one processor to cause the at least one processor to perform the catalytic cracking catalyst activity prediction method as described in any one of claims 1 to 6.

11. A non-transitory computer-readable storage medium, characterized in that, The non-transitory computer-readable storage medium stores computer-executable instructions for causing the computer to perform the catalytic cracking catalyst activity prediction method as described in any one of claims 1 to 6.

Images