Residual oil hydrogenation product yield determination method and device, electronic equipment and storage medium

Abstract

The invention provides a residual oil hydrogenation product yield determination method and device, electronic equipment and a storage medium, and the method comprises the steps: obtaining feeding information which at least comprises the feeding sulfur content, the feeding residual carbon content, the feeding nickel content, the feeding vanadium content, the feeding temperature and the feeding pressure; the feeding information is input into a pre-trained auto-encoder model, a plurality of initial yields of a plurality of residual oil hydrogenation products output by the auto-encoder model are obtained, and the residual oil hydrogenation products correspond to the feeding information; and obtaining a target yield of each residual oil hydrogenation product based on the initial yield of each residual oil hydrogenation product. The target yield of each residual oil hydrogenation product can be accurately and objectively determined.

Description

Technical Field

[0001] The invention relates to the technical field of filtered oil hydroprocessing, and in particular to a method, a device, an electronic device and a storage medium for determining the yield of a residual oil hydrogenation product. Background Art

[0002] The residue oil hydrogenation unit is an important part of the refinery. It is the leading unit for residue oil treatment. Its operation status and economic benefits are directly related to the overall benefits of the entire refinery. In the actual residue oil hydrogenation production process, in order to ensure the economic benefits of the entire plant and the stable operation of the unit, the blending ratio of the feed residue oil and wax oil of the residue oil hydrogenation unit will change on average every 3 to 5 days, and the corresponding yield of the residue oil hydrogenation product will also change.

[0003] It is known from the related art that the yield of the product side of the device (also known as the residue oil hydrogenation product) is often estimated based on human subjective experience, which will lead to inaccurate yield of the residue oil hydrogenation product, thereby affecting the subsequent process of residue oil hydrogenation treatment.

[0004] Currently, finding a method that can accurately and objectively determine the yield of residue oil hydrogenation products has become a research hotspot. Summary of the invention

[0005] The present invention provides a method, a device, an electronic device and a storage medium for determining the yield of a residue oil hydrogenation product, which can accurately and objectively determine the target yield of each residue oil hydrogenation product.

[0006] The present invention provides a method for determining the yield of a residue oil hydrogenation product, the method comprising: obtaining feed information, wherein the feed information at least comprises feed sulfur content, feed residual carbon content, feed nickel content, feed vanadium content, feed temperature and feed pressure; inputting the feed information into a pre-trained autoencoder model to obtain multiple initial yields of multiple residue oil hydrogenation products output by the autoencoder model, wherein the residue oil hydrogenation product corresponds to the feed information; and obtaining a target yield of each residue oil hydrogenation product based on the initial yield of each residue oil hydrogenation product.

[0007] According to a method for determining the yield of residue hydrogenation products provided by the present invention, the target yield of each residue hydrogenation product is obtained based on the initial yield of each residue hydrogenation product, specifically comprising: obtaining the sum of the initial yields based on the initial yields of each residue hydrogenation product; and performing quotient processing on the initial yield of each residue hydrogenation product and the sum of the initial yields to obtain the target yield of each residue hydrogenation product.

[0008] According to a method for determining the yield of a residue oil hydrogenation product provided by the present invention, the pre-trained autoencoder model is obtained in the following manner: a first training data set is obtained, wherein the first training data set includes multiple groups of first input data and multiple groups of first output data, wherein the first input data includes feed sulfur content training data, feed residual carbon content training data, feed nickel content training data, feed vanadium content training data, feed temperature training data and feed pressure training data; the first output data includes initial yield training data of residue oil hydrogenation product training data corresponding to the first input data; the multiple groups of first input data and the multiple groups of first output data in the first training data set are normalized to obtain a normalized first training data set; and the autoencoder model is unsupervisedly trained based on the normalized first training data set to obtain a pre-trained autoencoder model.

[0009] According to a method for determining the yield of a residue hydrogenation product provided by the present invention, before inputting the feed information into a pretrained autoencoder model, the method further includes: obtaining a plurality of pretrained autoencoder models to be coupled; setting the plurality of pretrained autoencoder models to be coupled in parallel with the pretrained autoencoder model to obtain a stacked autoencoder model; the inputting the feed information into the pretrained autoencoder model specifically includes: inputting the feed information into the pretrained autoencoder model in the stacked autoencoder model; before obtaining the target yield of each residue hydrogenation product based on the initial yield of each residue hydrogenation product, the method further includes: inputting the feed information into the plurality of pretrained autoencoder models to be coupled in the stacked autoencoder model to obtain the element content of the constituent elements corresponding to each residue hydrogenation product output by each pretrained autoencoder model, wherein the element content of the constituent elements corresponding to the residue hydrogenation product determines the physical properties of the residue hydrogenation product.

[0010] According to a method for determining the yield of a residue oil hydrogenation product provided by the present invention, the pre-trained autoencoder model to be coupled is obtained in the following manner: a second training data set is obtained, wherein the second training data set includes multiple groups of second input data and multiple groups of second output data, wherein the second input data includes feed sulfur content training data, feed residual carbon content training data, feed nickel content training data, feed vanadium content training data, feed temperature training data and feed pressure training data; the second output data includes element content training data of constituent elements corresponding to the residue oil hydrogenation product training data, wherein the residue oil hydrogenation product training data corresponds to the second input data; the multiple groups of second input data and the multiple groups of second output data in the second training data set are normalized to obtain the normalized second training data set; unsupervised training is performed on the autoencoder model to be coupled based on the normalized second training data set to obtain the pre-trained autoencoder model to be coupled.

[0011] According to a method for determining the yield of a residue oil hydrogenation product provided by the present invention, the residue oil hydrogenation product includes any one or more of acid gas, dry gas, naphtha, diesel, and tail oil; the constituent elements corresponding to the residue oil hydrogenation product include any one or more of the sulfur element constituting the acid gas, the sulfur element constituting the dry gas, the sulfur element and metal elements constituting the naphtha, the sulfur element and metal elements constituting the diesel, and the sulfur element and metal elements constituting the tail oil.

[0012] The present invention also provides a device for determining the yield of residue oil hydrogenation products, the device comprising: an acquisition module, used to acquire feed information, wherein the feed information at least includes feed sulfur content, feed residual carbon content, feed nickel content, feed vanadium content, feed temperature and feed pressure; a processing module, used to input the feed information into a pre-trained autoencoder model to obtain multiple initial yields of multiple residue oil hydrogenation products output by the autoencoder model, wherein the residue oil hydrogenation products correspond to the feed information; and a generation module, used to obtain the target yield of each residue oil hydrogenation product based on the initial yield of each residue oil hydrogenation product.

[0013] The present invention also provides an electronic device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the program, the method for determining the yield of residue oil hydrogenation products as described in any one of the above is implemented.

[0014] The present invention also provides a non-transitory computer-readable storage medium having a computer program stored thereon, and when the computer program is executed by a processor, the method for determining the yield of residue oil hydrogenation products as described in any one of the above is implemented.

[0015] The present invention also provides a computer program product, comprising a computer program, wherein when the computer program is executed by a processor, the method for determining the yield of residue oil hydrogenation products as described in any one of the above is implemented.

[0016] The method, device, electronic device and storage medium for determining the yield of residue oil hydrogenation products provided by the present invention obtain feed information, and input the feed information into a pre-trained autoencoder model to obtain multiple initial yields of multiple residue oil hydrogenation products output by the autoencoder model, and then obtain the target yield of each residue oil hydrogenation product based on the initial yield of each residue oil hydrogenation product, thereby accurately and objectively determining the target yield of each residue oil hydrogenation product, and providing a theoretical basis for subsequent process or production demand adjustment. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] In order to more clearly illustrate the technical solutions in the present invention or the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0018] Figure 1 This is one of the flow diagrams of the method for determining the yield of the residue oil hydrogenation product provided by the present invention;

[0019] Figure 2 It is a schematic diagram of the process of the pre-trained autoencoder model provided by the present invention;

[0020] Figure 3 This is the second flow diagram of the method for determining the yield of the residue oil hydrogenation product provided by the present invention;

[0021] Figure 4 It is a schematic diagram of the process of pre-training the coupled autoencoder model provided by the present invention;

[0022] Figure 5 It is a structural schematic diagram of a device for determining the yield of residue oil hydrogenation products provided by the present invention;

[0023] Figure 6 It is a structural schematic diagram of the electronic device provided by the present invention. DETAILED DESCRIPTION

[0024] In order to make the purpose, technical solution and advantages of the present invention clearer, the technical solution of the present invention will be clearly and completely described below in conjunction with the drawings of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0025] Figure 1 This is one of the flow diagrams of the method for determining the yield of residue oil hydrogenation products provided by the present invention.

[0026] The following will be combined Figure 1 The process of the method for determining the yield of residue oil hydrogenation products provided by the present invention is described.

[0027] In an exemplary embodiment of the present invention, Figure 1 It can be seen that the method for determining the yield of the residue oil hydrogenation product may include steps 110 to 130, and each step will be introduced below.

[0028] In step 110, feed information is obtained.

[0029] In one embodiment, the oil hydrogenation product yield determination method provided by the present invention can be applied to a residual oil hydrogenation unit. During the application process, feed information can be obtained. Among them, the feed information can be understood as the relevant information of the feed sent into the residual oil hydrogenation unit. In one example, the feed information can at least include feed sulfur content, feed residual carbon content, feed nickel content, feed vanadium content, feed temperature and feed pressure. Feed sulfur content can be understood as the content of sulfur element in the feed. Feed residual carbon content can be understood as the content of residual carbon element in the feed. Residual carbon element can be understood as the remaining carbon element. Feed nickel content can be understood as the content of nickel-carbon element in the feed. Feed temperature can be understood as the temperature in the feed furnace. Feed pressure can be understood as the pressure in the feed furnace.

[0030] It should be noted that the yield of the residue oil hydrogenation product that can be produced will be different depending on the specific feed sulfur content, specific feed residual carbon content, specific feed nickel content, specific feed vanadium content, specific feed temperature and specific feed pressure.

[0031] In step 120, the feed information is input into the pre-trained autoencoder model to obtain multiple initial yields of multiple residue hydrogenation products output by the autoencoder model.

[0032] In one embodiment, the aforementioned feed information can be input into a pre-trained autoencoder model, and the autoencoder model outputs multiple initial yields of multiple residue hydrogenation products. It can be understood that, due to the determination of the input feed information, multiple initial yields of corresponding residue hydrogenation products can be generated.

[0033] It should be noted that a change in the type of feed information may result in a change in the residue oil hydrogenation product. The residue oil hydrogenation product corresponds to the feed information. In one example, the residue oil hydrogenation product may include any one or more of acid gas, dry gas, naphtha, diesel, and tail oil.

[0034] For the sake of convenience, the present invention will be described by taking the residual oil hydrogenation products including acid gas, dry gas, naphtha, diesel and tail oil as an example.

[0035] In step 130, a target yield of each residue hydrogenation product is obtained based on the initial yield of each residue hydrogenation product.

[0036] In one embodiment, the number of autoencoder models can match the number of residual oil hydrogenation products. In one example, the initial yield of acid gas can be output based on the autoencoder model for acid gas. The initial yield of naphtha can be output based on the autoencoder model for naphtha. The initial yield of dry gas can be output based on the autoencoder model for dry gas. The initial yield of diesel can be output based on the autoencoder model for diesel. The initial yield of tail oil can be output based on the autoencoder model for tail oil.

[0037] Further, based on the initial yield of each residue hydrogenation product, the target yield of each residue hydrogenation product is obtained. It can be understood that after the above treatment, the sum of the target yields of all residue hydrogenation products is 1. Through this embodiment, the target yield of each residue hydrogenation product can be accurately and objectively determined, providing a theoretical basis for subsequent process or production demand adjustment.

[0038] In an exemplary embodiment of the present invention, based on the initial yield of each residue hydrogenation product, the target yield of each residue hydrogenation product (corresponding to step 130) can be achieved by the following method:

[0039] Based on the initial yields of the various residue hydrogenation products, the sum of the initial yields is obtained;

[0040] The initial yield of each residue oil hydrogenation product and the sum of the initial yields are respectively taken as quotients to obtain the target yield of each residue oil hydrogenation product.

[0041] Continuing to use the embodiment described above as an example for explanation, the initial yield of acid gas, the initial yield of naphtha, the initial yield of dry gas, the initial yield of diesel and the initial yield of tail oil can be summed to obtain the sum of the initial yields. Further, for the target yield of acid gas, it can be obtained by taking the quotient of the initial yield of acid gas and the sum of the initial yields. For the target yield of dry gas, it can be obtained by taking the quotient of the initial yield of dry gas and the sum of the initial yields. For the target yield of naphtha, it can be obtained by taking the quotient of the initial yield of naphtha and the sum of the initial yields. For the target yield of diesel, it can be obtained by taking the quotient of the initial yield of diesel and the sum of the initial yields. For the target yield of tail oil, it can be obtained by taking the quotient of the initial yield of tail oil and the sum of the initial yields. Through this embodiment, it can be ensured that the sum of the target yields of all residue hydrogenation products is 1, thereby ensuring the rationality and referenceability of the obtained target yields.

[0042] Figure 2 It is a flow chart of the pre-trained autoencoder model provided by the present invention.

[0043] In order to further introduce the method for determining the yield of the residue oil hydrogenation product provided by the present invention, Figure 2 Explain the process of pre-training an autoencoder model.

[0044] In an exemplary embodiment of the present invention, Figure 2 It can be seen that the pre-trained autoencoder model may include steps 210 to 230, and each step will be introduced below.

[0045] In step 210, a first training data set is obtained.

[0046] In one embodiment, the source of the first training data set may be historical industrial data, or may be simulated industrial data obtained by using simulation software.

[0047] The first training data set may include multiple sets of first input data and multiple sets of first output data. The first input data may include feed sulfur content training data, feed residual carbon content training data, feed nickel content training data, feed vanadium content training data, feed temperature training data, and feed pressure training data; the first output data may include initial yield training data of residual oil hydrogenation product training data corresponding to the first input data.

[0048] In one example, the first input data can be represented by X, where X={X1, X2, X3, X4, X5, X6}. X1 can represent feed sulfur content training data; X2 can represent feed residual carbon content training data; X3 can represent feed nickel content training data; X4 can represent feed vanadium content training data; X5 can represent feed temperature training data; X6 can represent feed pressure training data. The first output data can be represented by Y, where Y={Y1, Y2, Y3, Y4, Y5}. Among them, Y1 can represent initial yield training data of acid gas training data; Y2 can represent initial yield training data of dry gas training data; Y3 can represent initial yield training data of naphtha training data; Y4 can represent initial yield training data of diesel training data; Y5 can represent initial yield training data of tail oil training data.

[0049] In yet another example, the plurality of groups of first input data and the plurality of groups of first output data may be data at different process times.

[0050] In step 220, a plurality of groups of first input data and a plurality of groups of first output data in the first training data set are normalized to obtain a normalized first training data set.

[0051] In step 230, unsupervised training is performed on the autoencoder model based on the normalized first training data set to obtain a pre-trained autoencoder model.

[0052] In another embodiment, the multiple groups of first input data and the multiple groups of first output data in the obtained first training data set can be normalized to obtain the normalized first training data set; then the autoencoder model is unsupervisedly trained based on the normalized first training data set to obtain the pre-trained autoencoder model. In the application process, the model training can be performed based on the back propagation algorithm and the optimization method. In this embodiment, the model is trained based on the normalized first training data set, which can avoid large errors in the input model and avoid problems such as the model not being easy to converge.

[0053] Figure 3 This is the second flow diagram of the method for determining the yield of residue oil hydrogenation products provided by the present invention.

[0054] In order to further introduce the method for determining the yield of the residue oil hydrogenation product provided by the present invention, Figure 3 Provide explanation.

[0055] In an exemplary embodiment of the present invention, Figure 3It can be seen that the method for determining the yield of the residue oil hydrogenation product may include steps 310 to 360, wherein step 310 is the same as or similar to step 110, and step 360 is the same as or similar to step 130. For its specific implementation and beneficial effects, please refer to the previous description, which will not be repeated in this embodiment. Steps 320 to 350 will be introduced separately below.

[0056] In step 320, a plurality of pre-trained autoencoder models to be coupled are obtained.

[0057] In step 330, a plurality of pre-trained autoencoder models to be coupled are set in parallel with the pre-trained autoencoder model to obtain a stacked autoencoder model.

[0058] In one embodiment, multiple pre-trained autoencoder models to be coupled can also be obtained. Multiple pre-trained autoencoder models to be coupled are set in parallel with the pre-trained autoencoder model to obtain a stacked autoencoder model. It can be understood that the autoencoder model can simultaneously output multiple initial yields of multiple residue hydrogenation products corresponding to the feed information, as well as the element content of the constituent elements corresponding to each residue hydrogenation product. This is to more accurately and objectively determine the production of the residue hydrogenation product, for example, including the target yield and the element content of the constituent elements.

[0059] It should be noted that the pre-trained autoencoder model to be coupled can be coupled with the pre-trained autoencoder model, wherein the pre-training process of the autoencoder model to be coupled is similar to the pre-training process of the autoencoder model, and during the pre-training process, the input data in the training set used by the two can be the same, but the output data correspondence is different, wherein the output data of the autoencoder model to be coupled is about the element content of the constituent elements corresponding to the residual oil hydrogenation product. The output data of the autoencoder model is about the initial yield of the residual oil hydrogenation product.

[0060] In step 340, the feed information is input into the pre-trained autoencoder model in the stacked autoencoder model to obtain a plurality of initial yields of a plurality of residue hydrogenation products output by the autoencoder model.

[0061] In step 350, the feed information is input into a plurality of pre-trained autoencoder models to be coupled in the stacked autoencoder model to obtain the element content of the constituent elements corresponding to each residual oil hydrogenation product output by each pre-trained autoencoder model to be coupled.

[0062] In one embodiment, the feed information can be input into the pre-trained autoencoder model in the stacked autoencoder model to obtain multiple initial yields of multiple residue hydrogenation products output by the autoencoder model. In one example, when there are multiple residue hydrogenation products, there are also multiple corresponding autoencoder models.

[0063] In another example, the feed information can also be input into a plurality of pre-trained autoencoder models to be coupled in the stacked autoencoder model, so as to obtain the element content of the constituent elements corresponding to each residual oil hydrogenation product output by each pre-trained autoencoder model to be coupled. The element content of the constituent elements corresponding to the residual oil hydrogenation product determines the physical properties of the residual oil hydrogenation product.

[0064] In one example, when the residue oil hydrogenation product is acid gas, the constituent elements corresponding to the residue oil hydrogenation product may be sulfur elements constituting acid gas. When the residue oil hydrogenation product is dry gas, the constituent elements corresponding to the residue oil hydrogenation product may be sulfur elements constituting dry gas. When the residue oil hydrogenation product is naphtha, the constituent elements corresponding to the residue oil hydrogenation product may be sulfur elements and metal elements constituting naphtha. When the residue oil hydrogenation product is diesel, the constituent elements corresponding to the residue oil hydrogenation product may be sulfur elements and metal elements constituting diesel. When the residue oil hydrogenation product is tail oil, the constituent elements corresponding to the residue oil hydrogenation product may be sulfur elements and metal elements constituting tail oil. Among them, the aforementioned metal elements may be nickel elements and vanadium elements.

[0065] Figure 4 It is a schematic diagram of the process of pre-training the coupled autoencoder model provided by the present invention.

[0066] The following will be combined Figure 4 Describe the process of pre-training the coupled autoencoder model.

[0067] In an exemplary embodiment of the present invention, Figure 4 It can be seen that the pre-training of the coupled autoencoder model may include steps 410 to 430, and each step will be introduced below.

[0068] In step 410, a second training data set is obtained.

[0069] In one embodiment, the source of the second training data set may be historical industrial data, or may be simulated industrial data obtained by using simulation software.

[0070] The second training data set may include multiple sets of second input data and multiple sets of second output data. The second input data includes feed sulfur content training data, feed residual carbon content training data, feed nickel content training data, feed vanadium content training data, feed temperature training data, and feed pressure training data. The second output data may include element content training data of constituent elements corresponding to the residual oil hydrogenation product training data, wherein the residual oil hydrogenation product training data corresponds to the second input data.

[0071] In one example, the second input data may also be represented by X, where X={X1, X2, X3, X4, X5, X6}. X1 may represent feed sulfur content training data; X2 may represent feed residual carbon content training data; X3 may represent feed nickel content training data; X4 may represent feed vanadium content training data; X5 may represent feed temperature training data; and X6 may represent feed pressure training data.

[0072] The second output data may be represented by Q, where Q={Q1, Q2, Q3, Q4, Q5, Q6, Q7, Q8}. Q1 may represent the element content training data of sulfur element in acid gas training data; Q2 may represent the element content training data of sulfur element in dry gas training data; Q3 may represent the element content training data of sulfur element in naphtha training data; Q4 may represent the element content training data of metal element in naphtha training data; Q5 may represent the element content training data of sulfur element in diesel training data; Q6 may represent the element content training data of metal element in diesel training data; Q7 may represent the element content training data of sulfur element in tail oil training data; and Q8 may represent the element content training data of metal element in tail oil training data.

[0073] In yet another example, the plurality of groups of second input data and the plurality of groups of second output data may be data at different process times.

[0074] In step 420, a plurality of groups of second input data and a plurality of groups of second output data in the second training data set are normalized to obtain a normalized second training data set.

[0075] In step 430, unsupervised training is performed on the coupled autoencoder model based on the normalized second training data set to obtain a pre-trained coupled autoencoder model.

[0076] In another embodiment, the plurality of groups of second input data and the plurality of groups of second output data in the obtained second training data set can be normalized to obtain a normalized second training data set; then, the coupled autoencoder model can be unsupervisedly trained based on the normalized second training data set to obtain a pre-trained coupled autoencoder model. In the application process, the model can be trained based on the back propagation algorithm and the optimization method. In this embodiment, the model is trained based on the normalized second training data set to avoid large errors in the input model and problems such as the model not being easy to converge.

[0077] The method for determining the yield of residue oil hydrogenation products provided by the present invention obtains feed information, and inputs the feed information into a pre-trained autoencoder model to obtain multiple initial yields of multiple residue oil hydrogenation products output by the autoencoder model, and then obtains the target yield of each residue oil hydrogenation product based on the initial yield of each residue oil hydrogenation product, thereby accurately and objectively determining the target yield of each residue oil hydrogenation product, and providing a theoretical basis for subsequent process or production demand adjustment.

[0078] According to the above description, the method for determining the yield of residue oil hydrogenation products provided by the present invention obtains feed information, and inputs the feed information into a pre-trained autoencoder model to obtain multiple initial yields of multiple residue oil hydrogenation products output by the autoencoder model, and then obtains the target yield of each residue oil hydrogenation product based on the initial yield of each residue oil hydrogenation product, thereby accurately and objectively determining the target yield of each residue oil hydrogenation product, providing a theoretical basis for subsequent process or production demand adjustments.

[0079] Based on the same concept, the present invention also provides a device for determining the yield of residual oil hydrogenation products.

[0080] The device for determining the yield of residue oil hydrogenation products provided by the present invention is described below. The device for determining the yield of residue oil hydrogenation products described below and the method for determining the yield of residue oil hydrogenation products described above can be referred to each other.

[0081] Figure 5 It is a structural schematic diagram of a device for determining the yield of residue oil hydrogenation products provided by the present invention.

[0082] In an exemplary embodiment of the present invention, Figure 5 It can be seen that the device for determining the yield of residue oil hydrogenation products may include an acquisition module 510, a processing module 520 and a generation module 530, and each module will be introduced below.

[0083] The acquisition module 510 may be configured to acquire feed information, wherein the feed information includes at least feed sulfur content, feed residual carbon content, feed nickel content, feed vanadium content, feed temperature and feed pressure;

[0084] The processing module 520 may be configured to input the feed information into the pre-trained autoencoder model to obtain a plurality of initial yields of a plurality of residue hydrogenation products output by the autoencoder model, wherein the residue hydrogenation products correspond to the feed information;

[0085] The generation module 530 may be configured to obtain a target yield of each residue hydrogenation product based on an initial yield of each residue hydrogenation product.

[0086] In an exemplary embodiment of the present invention, the generation module 530 can obtain the target yield of each residue hydrogenation product based on the initial yield of each residue hydrogenation product in the following manner:

[0087] Based on the initial yields of the various residue hydrogenation products, the sum of the initial yields is obtained;

[0088] The initial yield of each residue oil hydrogenation product and the sum of the initial yields are respectively taken as quotients to obtain the target yield of each residue oil hydrogenation product.

[0089] In an exemplary embodiment of the present invention, the processing module 520 may implement the pre-trained autoencoder model in the following manner:

[0090] Obtaining a first training data set, wherein the first training data set includes a plurality of groups of first input data and a plurality of groups of first output data, wherein the first input data includes feed sulfur content training data, feed residual carbon content training data, feed nickel content training data, feed vanadium content training data, feed temperature training data, and feed pressure training data; the first output data includes initial yield training data of residual oil hydrogenation product training data corresponding to the first input data;

[0091] Normalizing the plurality of groups of first input data and the plurality of groups of first output data in the first training data set to obtain a normalized first training data set;

[0092] The autoencoder model is unsupervisedly trained based on the normalized first training data set to obtain a pre-trained autoencoder model.

[0093] In an exemplary embodiment of the present invention, the processing module 520 may also be configured to:

[0094] Obtain multiple pre-trained autoencoder models to be coupled;

[0095] The plurality of pre-trained autoencoder models to be coupled are set in parallel with the pre-trained autoencoder model to obtain a stacked autoencoder model;

[0096] The processing module 520 can implement inputting the feed information into the pre-trained autoencoder model in the following manner:

[0097] Inputting the feed information into the pre-trained autoencoder model in the stacked autoencoder model;

[0098] The processing module 520 may also be configured to:

[0099] The feed information is input into a plurality of pre-trained autoencoder models to be coupled in the stacked autoencoder model, and the element contents of the constituent elements corresponding to the respective residue oil hydrogenation products output by the pre-trained autoencoder models to be coupled are obtained, wherein the element contents of the constituent elements corresponding to the residue oil hydrogenation products determine the physical properties of the residue oil hydrogenation products.

[0100] In an exemplary embodiment of the present invention, the processing module 520 may implement the pre-trained coupled autoencoder model in the following manner:

[0101] Obtaining a second training data set, wherein the second training data set includes a plurality of sets of second input data and a plurality of sets of second output data, wherein the second input data includes feed sulfur content training data, feed residual carbon content training data, feed nickel content training data, feed vanadium content training data, feed temperature training data, and feed pressure training data; the second output data includes element content training data of constituent elements corresponding to residual oil hydrogenation product training data, wherein the residual oil hydrogenation product training data corresponds to the second input data;

[0102] Normalizing the plurality of groups of second input data and the plurality of groups of second output data in the second training data set to obtain a normalized second training data set;

[0103] The coupled autoencoder model is unsupervisedly trained based on the normalized second training data set to obtain a pre-trained coupled autoencoder model.

[0104] In an exemplary embodiment of the present invention, the residue oil hydrogenation product may include any one or more of acid gas, dry gas, naphtha, diesel, and tail oil; the constituent elements corresponding to the residue oil hydrogenation product may include any one or more of the sulfur element constituting acid gas, the sulfur element constituting dry gas, the sulfur element and metal elements constituting naphtha, the sulfur element and metal elements constituting diesel, and the sulfur element and metal elements constituting tail oil.

[0105] Figure 6 An example of a physical structure diagram of an electronic device is shown in FIG. Figure 6As shown, the electronic device may include: a processor 610, a communication interface 620, a memory 630 and a communication bus 640, wherein the processor 610, the communication interface 620 and the memory 630 communicate with each other through the communication bus 640. The processor 610 may call the logic instructions in the memory 630 to execute the residual oil hydrogenation product yield determination method, the method comprising: obtaining feed information, wherein the feed information at least includes feed sulfur content, feed residual carbon content, feed nickel content, feed vanadium content, feed temperature and feed pressure; inputting the feed information into the pre-trained autoencoder model to obtain multiple initial yields of multiple residual oil hydrogenation products output by the autoencoder model, wherein the residual oil hydrogenation product corresponds to the feed information; based on the initial yield of each of the residual oil hydrogenation products, obtaining the target yield of each of the residual oil hydrogenation products.

[0106] In addition, the logic instructions in the above-mentioned memory 630 can be implemented in the form of a software functional unit and can be stored in a computer-readable storage medium when it is sold or used as an independent product. Based on such an understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art or the part of the technical solution, can be embodied in the form of a software product, and the computer software product is stored in a storage medium, including a number of instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), disk or optical disk and other media that can store program codes.

[0107] On the other hand, the present invention also provides a computer program product, which includes a computer program, which can be stored on a non-transitory computer-readable storage medium. When the computer program is executed by a processor, the computer can execute the residue oil hydrogenation product yield determination method provided by the above-mentioned methods, the method comprising: obtaining feed information, wherein the feed information at least includes feed sulfur content, feed residual carbon content, feed nickel content, feed vanadium content, feed temperature and feed pressure; inputting the feed information into a pre-trained autoencoder model to obtain multiple initial yields of multiple residue oil hydrogenation products output by the autoencoder model, wherein the residue oil hydrogenation product corresponds to the feed information; based on the initial yield of each of the residue oil hydrogenation products, obtaining the target yield of each of the residue oil hydrogenation products.

[0108] On the other hand, the present invention also provides a non-transitory computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, is implemented to execute the residue oil hydrogenation product yield determination method provided by the above-mentioned methods, the method comprising: obtaining feed information, wherein the feed information at least includes feed sulfur content, feed residual carbon content, feed nickel content, feed vanadium content, feed temperature and feed pressure; inputting the feed information into a pre-trained autoencoder model to obtain multiple initial yields of multiple residue oil hydrogenation products output by the autoencoder model, wherein the residue oil hydrogenation product corresponds to the feed information; and obtaining a target yield of each of the residue oil hydrogenation products based on the initial yield of each of the residue oil hydrogenation products.

[0109] The device embodiments described above are merely illustrative, wherein the units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, they may be located in one place, or they may be distributed on multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the scheme of this embodiment. Ordinary technicians in this field can understand and implement it without paying creative labor.

[0110] Through the description of the above implementation methods, those skilled in the art can clearly understand that each implementation method can be implemented by means of software plus a necessary general hardware platform, and of course, it can also be implemented by hardware. Based on this understanding, the above technical solution is essentially or the part that contributes to the prior art can be embodied in the form of a software product, and the computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, a disk, an optical disk, etc., including a number of instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods described in each embodiment or some parts of the embodiments.

[0111] It is further understood that, although the operations are described in a specific order in the accompanying drawings in the embodiments of the present invention, it should not be understood as requiring the operations to be performed in the specific order or serial order shown, or requiring the execution of all the operations shown to obtain the desired results. In certain environments, multitasking and parallel processing may be advantageous.

[0112] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for determining the yield of residue oil hydrogenation products, characterized in that: The method comprises: Obtaining feed information, wherein the feed information at least includes feed sulfur content, feed residual carbon content, feed nickel content, feed vanadium content, feed temperature and feed pressure; Inputting the feed information into a pre-trained autoencoder model to obtain a plurality of initial yields of a plurality of residue hydrogenation products output by the autoencoder model, wherein the residue hydrogenation products correspond to the feed information; Based on the initial yield of each of the residue oil hydrogenation products, the target yield of each of the residue oil hydrogenation products is obtained.

2. The method for determining the yield of residue hydrogenation products according to claim 1, characterized in that: The target yield of each of the residue oil hydrogenation products is obtained based on the initial yield of each of the residue oil hydrogenation products, specifically comprising: Based on the initial yields of the residue hydrogenation products, a sum of the initial yields is obtained; The initial yield of each of the residue oil hydrogenation products is divided by the sum of the initial yields to obtain the target yield of each of the residue oil hydrogenation products.

3. The method for determining the yield of residue hydrogenation products according to claim 1, characterized in that: The pre-trained autoencoder model is obtained in the following way: Obtaining a first training data set, wherein the first training data set includes a plurality of groups of first input data and a plurality of groups of first output data, wherein the first input data includes feed sulfur content training data, feed residual carbon content training data, feed nickel content training data, feed vanadium content training data, feed temperature training data, and feed pressure training data; the first output data includes initial yield training data of residual oil hydrogenation product training data corresponding to the first input data; Normalizing the plurality of groups of first input data and the plurality of groups of first output data in the first training data set to obtain a normalized first training data set; The autoencoder model is unsupervisedly trained based on the normalized first training data set to obtain a pre-trained autoencoder model.

4. The method for determining the yield of residue hydrogenation products according to claim 1, characterized in that: Before inputting the feed information into the pre-trained autoencoder model, the method further includes: Obtain multiple pre-trained autoencoder models to be coupled; Setting a plurality of the pre-trained autoencoder models to be coupled in parallel with the pre-trained autoencoder model to obtain a stacked autoencoder model; The inputting the feed information into the pre-trained autoencoder model specifically includes: Inputting the feed information into the pre-trained autoencoder model in the stacked autoencoder model; Before obtaining the target yield of each of the residue hydrogenation products based on the initial yield of each of the residue hydrogenation products, the method further comprises: The feed information is input into a plurality of the pre-trained autoencoder models to be coupled in the stacked autoencoder model to obtain the elemental content of the constituent elements corresponding to each of the residue oil hydrogenation products output by each of the pre-trained autoencoder models to be coupled, wherein the elemental content of the constituent elements corresponding to the residue oil hydrogenation product determines the physical properties of the residue oil hydrogenation product.

5. The method for determining the yield of residue hydrogenation products according to claim 4, characterized in that: The pre-trained coupled autoencoder model is obtained in the following manner: Acquire a second training data set, wherein the second training data set includes a plurality of sets of second input data and a plurality of sets of second output data, wherein the second input data includes feed sulfur content training data, feed residual carbon content training data, feed nickel content training data, feed vanadium content training data, feed temperature training data, and feed pressure training data; the second output data includes element content training data of constituent elements corresponding to residual oil hydrogenation product training data, wherein the residual oil hydrogenation product training data corresponds to the second input data; Normalizing the plurality of groups of second input data and the plurality of groups of second output data in the second training data set to obtain a normalized second training data set; The coupled autoencoder model is unsupervisedly trained based on the normalized second training data set to obtain a pre-trained coupled autoencoder model.

6. The method for determining the yield of residue hydrogenation products according to claim 4, characterized in that: The residue oil hydrogenation product includes any one or more of acid gas, dry gas, naphtha, diesel, and tail oil; The constituent elements corresponding to the residual oil hydrogenation product include any one or more of the sulfur element constituting the acid gas, the sulfur element constituting the dry gas, the sulfur element and metal elements constituting the naphtha, the sulfur element and metal elements constituting the diesel, and the sulfur element and metal elements constituting the tail oil.

7. A device for determining the yield of residue oil hydrogenation products, characterized in that: The device comprises: An acquisition module, used to acquire feed information, wherein the feed information at least includes feed sulfur content, feed residual carbon content, feed nickel content, feed vanadium content, feed temperature and feed pressure; A processing module, used for inputting the feed information into a pre-trained autoencoder model to obtain a plurality of initial yields of a plurality of residue hydrogenation products output by the autoencoder model, wherein the residue hydrogenation products correspond to the feed information; A generation module is used to obtain a target yield of each of the residue oil hydrogenation products based on the initial yield of each of the residue oil hydrogenation products.

8. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein: When the processor executes the program, the method for determining the yield of residue oil hydrogenation products according to any one of claims 1 to 6 is implemented.

9. A non-transitory computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the method for determining the residue hydrogenation product yield according to any one of claims 1 to 6 is implemented.

10. A computer program product, comprising a computer program, characterized in that When the computer program is executed by a processor, the method for determining the residue hydrogenation product yield according to any one of claims 1 to 6 is implemented.

Images