Coupling mechanism and data product oil pipeline mixed oil length calculation method and system

Abstract

The application relates to a product oil pipeline mixed oil length calculation method and system of a coupling mechanism and data, and the method is characterized in that the method comprises the following steps: acquiring field real-time operation data and pipeline basic parameters of a to-be-tested product oil pipeline, and calculating pipeline characteristic parameters, water thermal parameters, oil property and initial mixed oil information; acquiring a mixed oil length mechanism formula calculation result of the to-be-tested product oil pipeline; inputting the obtained pipeline characteristic parameters, water thermal parameters, oil property and initial mixed oil information into a pre-trained mixed oil length mixed prediction model, and fusing the mixed oil length mechanism formula calculation result to obtain a mixed oil length prediction value of the to-be-tested product oil pipeline, and the application can be widely applied to the field of mixed oil length calculation of product oil sequential transportation.

Description

Technical Field

[0001] This invention relates to the field of calculating the mixed length of refined oil during sequential transportation, and in particular to a method and system for calculating the mixed length of refined oil pipelines that couples mechanisms and data. Background Technology

[0002] Refined oil pipelines employ a sequential transport process. During this process, a significant amount of mixed oil inevitably occurs at the interface between the two types of oil. This mixing degrades the quality of the oil. Therefore, appropriate methods for receiving and handling mixed oil are crucial for improving the economic efficiency of pipeline-transported refined oil. Operators pay close attention to information related to mixed oil, with the length of the mixed oil being of paramount importance in informing decisions regarding its acceptance and handling.

[0003] Currently, the main methods for calculating the mixed oil length in refined oil pipelines are empirical formulas and data-driven methods. Empirical formulas, after years of development, have progressed through stages: Birge formula, Smith & Schulze mixing model, Sjenitzer F formula, and Austin empirical formula. The most commonly used Austin empirical formula has been developed, which determines the relationship between the critical Reynolds number and the pipeline inner diameter. This formula is based on a large database containing both experimental and real-world data and has wide engineering applications; many steady-state calculation software programs use this formula to calculate mixed oil length. Data-driven methods utilize real-world data, process it, and then use a regression model to obtain predicted mixed oil length values. Existing technology discloses a hybrid model combining data-driven and empirical formulas, using the Austin formula as input to the regression model to compensate for errors between the calculated and actual values. The resulting calculations show improved accuracy compared to the Austin formula and are among the most accurate methods currently being researched.

[0004] However, while empirical formulas can quickly and easily estimate the mixing length, the nonlinearity between the mixing length and operational data is strong. Empirical methods are often based on years of operational patterns in a specific region, resulting in limitations, insufficient universality and nonlinear fitting capabilities, and prediction accuracy that fails to meet field requirements. Parameter adjustments are often necessary when used in different pipelines. For example, the Austin formula, when applied in practice, often yields conservative values, making it difficult to accurately guide the formulation of mixing receiving schemes. While data-driven methods can fit the complex nonlinear relationship between input and output, the results lack mechanistic interpretability. The development of mixing is influenced by a variety of factors, making the model input crucial. Furthermore, current technologies do not comprehensively consider the model input, failing to account for the varying importance of different factors in predicting mixing length. Summary of the Invention

[0005] To address the aforementioned issues, the purpose of this invention is to provide a method and system for calculating the mixed oil length of refined oil pipelines that couples the mechanism and data. The accuracy of the prediction results can meet the needs of the field and has interpretable mechanism.

[0006] To achieve the above objectives, the present invention adopts the following technical solution: Firstly, it provides a method for calculating the mixed oil length of a refined oil pipeline, which couples the mechanism and data, comprising:

[0007] The system acquires real-time on-site operating data and basic parameters of the pipeline to be tested, and calculates pipeline characteristic parameters, hydrothermal parameters, oil properties and initial mixing information.

[0008] Obtain the calculation results of the mixing length mechanism formula of the pipeline to be tested;

[0009] The obtained pipeline characteristic parameters, hydrothermal parameters, oil properties and initial oil mixing information are input into the pre-trained oil mixing length prediction model and fused with the calculation results of the oil mixing length mechanism formula to obtain the predicted value of the oil mixing length of the pipeline to be tested.

[0010] Furthermore, the construction process of the oil mixing length prediction model is as follows:

[0011] A deep learning algorithm is used to pre-build a prediction model for the mixing length of mixed oil.

[0012] Using a pre-built oil mixing database, the oil mixing length prediction model is trained and validated to obtain a well-trained oil mixing length prediction model.

[0013] Furthermore, the oil blending database includes initial oil blending information, final oil blending information, oil properties, pipeline characteristic parameters, and hydrothermal parameters.

[0014] Furthermore, the mixed oil length prediction model includes an input layer, a hidden layer, a decoding layer, and an output layer. The input layer is equipped with a feature adaptive weighting module, and the output layer is equipped with a mechanism formula result fusion module.

[0015] The input layer is used to input pipeline characteristic parameters, hydrothermal parameters, oil properties, and initial oil mixing information;

[0016] The feature adaptive weighting module is used to perform weighted transformation on the data input to the input layer to capture the correlation between variables, output the weight of each input variable on the importance of the mixed oil length, and use the weighted input variables as the input of the mixed oil length prediction model.

[0017] The hidden layer is used to capture the nonlinear correlation between different variables and the change in the length of the mixed oil;

[0018] The decoding layer is used to weight the input features and restore them to the same dimension as the input. The weighted input features are used for oil mixing length prediction.

[0019] The mechanism formula result fusion module is used to couple the output result of the deep learning model with the pre-acquired calculation result of the mixing length mechanism formula to obtain the predicted value of the mixing length.

[0020] The output layer is used to output the predicted value of the oil mixing length.

[0021] Furthermore, the input xq of the oil mixing length prediction model is:

[0022] xq=[x1q1,x2q2,x3q3,...,x n q n ]

[0023] Where q is the weight; x n For each different input feature in the input layer; q n These are the weighted hidden layer features;

[0024] The output y of the oil mixing length prediction model is:

[0025] y = y1q1 + 2q2

[0026] Where y1 is the output variable of the mixed oil length prediction model; y2 is the calculation result of the mixed oil length mechanism formula; q1 and q2 are the weights of the two input variables in the mechanism formula result fusion module.

[0027] Furthermore, the activation function of the mixed oil length prediction model is the ReLU function, and the model is trained using the Adam optimizer.

[0028] Furthermore, the acquisition of real-time on-site operating data and basic parameters of the pipeline to be tested, and the calculation of pipeline characteristic parameters, hydrothermal parameters, oil properties, and initial blending information, includes:

[0029] The real-time on-site operating data and basic parameters of the pipeline to be tested are obtained through the SCADA system.

[0030] Based on the density data changes in the real-time on-site operation data, determine the start and end times of the mixed oil migration in a certain section of the pipeline, as well as the initial and final mixed oil information;

[0031] Based on the start and end times of the mixed oil's movement in a certain pipeline segment, as well as the initial and final mixed oil information, the temperature, pressure, flow rate, and oil properties of the mixed oil during the pipeline's movement time are determined, and the Reynolds number is calculated.

[0032] Based on Austin's formula, characteristic transformations are performed on pipe diameter, pipe length, and Reynolds number to obtain the corresponding pipe characteristic parameters and hydrothermal parameters.

[0033] Secondly, a system for calculating the mixed oil length of refined oil pipelines, which couples the mechanism and data, is provided, including:

[0034] The parameter acquisition and processing module is used to acquire the real-time on-site operating data and basic parameters of the pipeline to be tested, and to calculate the pipeline characteristic parameters, hydrothermal parameters, oil properties and initial mixing information.

[0035] The calculation result acquisition module is used to obtain the calculation results of the mixing length mechanism formula of the pipeline to be tested refined oil;

[0036] The oil mixing length prediction module is used to input the obtained pipeline characteristic parameters, hydrothermal parameters, oil properties and initial oil mixing information into the pre-trained oil mixing length prediction model, and integrate it with the calculation results of the oil mixing length mechanism formula to obtain the predicted value of the oil mixing length of the pipeline to be tested.

[0037] Thirdly, a processing device is provided, including computer program instructions, wherein when the computer program instructions are executed by the processing device, they are used to implement the steps corresponding to the above-mentioned method for calculating the mixed oil length of refined oil pipelines based on the coupling mechanism and data.

[0038] Fourthly, a computer-readable storage medium is provided, wherein computer program instructions are stored on the computer-readable storage medium, wherein when the computer program instructions are executed by a processor, they are used to implement the steps corresponding to the above-mentioned method for calculating the mixed oil length of a refined oil pipeline with coupling mechanism and data.

[0039] The present invention has the following advantages due to the adoption of the above technical solutions:

[0040] 1. On-site personnel use empirical methods to obtain the length of oil mixing in pipelines, but the results are not very accurate. This invention uses pipeline operation data and related pipeline parameters to comprehensively introduce many factors that can affect the development of oil mixing, and assigns weights to each factor according to their different importance to the prediction results.

[0041] 2. This invention constructs a hybrid model to couple the prediction results with the mechanistic formula results, further improving the interpretability and accuracy of the mechanism, realizing real-time and accurate prediction of the mixed oil length, thereby guiding the mixed oil download and processing, reducing unnecessary processing costs, and ensuring the quality of transported oil products.

[0042] 3. This invention utilizes pipeline SCADA system operation data, comprehensively considers multiple characteristic variables from a mechanistic perspective, reconstructs input features, and extracts features according to their importance. By constructing a hybrid model, it couples data-driven approaches with prior mechanistic knowledge to achieve real-time and accurate calculation of the mixed oil length in refined oil pipelines. This can assist pipeline operators in accurately grasping mixed oil information, formulating mixed oil acceptance and processing methods, reducing mixed oil processing costs, ensuring the quality of sequentially transported oil products, and improving economic efficiency.

[0043] In summary, this invention can be widely applied in the field of calculating the mixing length of finished oil products during sequential transportation. Attached Figure Description

[0044] Various other advantages and benefits will become apparent to those skilled in the art upon reading the following detailed description of preferred embodiments. The accompanying drawings are for illustrative purposes only and are not intended to limit the invention. Throughout the drawings, the same reference numerals denote the same parts. In the drawings:

[0045] Figure 1 This is a schematic diagram of the process for constructing a mixed oil database according to an embodiment of the present invention;

[0046] Figure 2 This is a schematic diagram of the construction process of the mixed oil length mixing prediction model provided in an embodiment of the present invention. Detailed Implementation

[0047] Exemplary embodiments of the invention will now be described in more detail with reference to the accompanying drawings. While exemplary embodiments of the invention are shown in the drawings, it should be understood that the invention can be implemented in various forms and should not be limited to the embodiments set forth herein. Rather, these embodiments are provided to enable a more thorough understanding of the invention and to fully convey the scope of the invention to those skilled in the art.

[0048] It should be understood that the terminology used herein is for the purpose of describing particular exemplary embodiments only and is not intended to be limiting. Unless the context clearly indicates otherwise, the singular forms “a,” “an,” and “described” as used herein may also include the plural forms. The terms “comprising,” “including,” “containing,” and “having” are inclusive and therefore indicate the presence of the stated features, steps, operations, elements, and / or components, but do not exclude the presence or addition of one or more other features, steps, operations, elements, components, and / or combinations thereof. The method steps, processes, and operations described herein are not construed as requiring them to be performed in a particular order described or illustrated unless the order of performance is explicitly indicated. It should also be understood that additional or alternative steps may be used.

[0049] Although terms such as first, second, third, etc., may be used in this document to describe multiple elements, components, regions, layers, and / or segments, these elements, components, regions, layers, and / or segments should not be limited by these terms. These terms may be used only to distinguish one element, component, region, layer, or segment from another. Unless the context clearly indicates otherwise, terms such as "first," "second," and other numerical terms used herein do not imply order or sequence. Therefore, the first element, component, region, layer, or segment discussed below may be referred to as the second element, component, region, layer, or segment without departing from the teachings of the exemplary embodiments.

[0050] Oil pipelines have the need to distribute and inject various oil products, currently mainly using sequential transportation methods. This inevitably leads to significant mixing of the two types of oil at the interface. Mixing often results in a decline in oil quality; therefore, stations along the pipeline need to promptly and accurately remove mixed oil for subsequent processing. However, the development of mixing is influenced by multiple factors. Current empirical formulas are insufficient to obtain accurate mixing length calculations, while purely data-driven methods lack mechanistic interpretability and often do not consider all factors comprehensively. Therefore, the oil pipeline mixing length calculation method and system provided in this invention, which couples mechanism and data, utilizes pipeline operation data and related parameters to comprehensively consider factors affecting mixing development. By coupling data-driven and mechanistic models, it obtains timely and accurate mixing length information, thereby enabling accurate removal and processing of mixed oil, reducing unnecessary processing costs, and ensuring the quality of transported oil.

[0051] Example 1

[0052] This embodiment provides a method for calculating the mixed oil length of a refined oil pipeline, which combines coupling mechanisms and data, and includes the following steps:

[0053] 1) such as Figure 1 As shown, a mixed oil database is pre-built based on real-time on-site operating data and basic pipeline parameters for subsequent training and validation of the mixed oil length prediction model.

[0054] Specifically, oil blending is influenced by multiple factors, therefore, the dataset construction process must comprehensively consider these influencing factors. Mechanistically, oil blending primarily results from diffusion and convection transfer between different grades and types of oil. Ambient temperature affects the temperature of the oil within the pipeline. Oil temperature is directly related to oil density and viscosity; differences in density lead to different concentration gradients between the two oils, thus affecting the diffusion coefficient. Differences in viscosity affect the convective diffusion coefficient; therefore, oil temperature, density, and viscosity must be included as characteristic variables. Oil has a certain degree of compressibility; different pressures cause changes in oil viscosity and density, thus affecting oil blending. Pipeline flow rate affects the flow regime and velocity of the transported oil. In laminar flow, a wedge-shaped oil head is generated. Increasing the blending volume results in a shorter blending length in turbulent flow compared to laminar flow. Therefore, flow rate and pressure during pipeline operation must be included as characteristic variables. As oil mixes flow through a pipe segment, they evolve from one state to another. Therefore, the initial state of the mixed oil and the state of the flow through the pipe segment inevitably affect the final state of the mixed oil. The initial mixed oil length affects the rate of subsequent mixed oil development; the more complete the initial mixed oil development, the slower the subsequent mixed oil development. Empirical models reveal the relationship between pipe diameter, pipe length, Reynolds number, and mixed oil length from a mechanistic perspective. According to Austin's formula, the mixed oil length is proportional to the 0.5 power of both pipe diameter and pipe length, and proportional to the -0.1 power of the Reynolds number. Therefore, the initial mixed oil length and the pipe diameter, pipe length, and Reynolds number after characteristic transformation are used as characteristic variables. The specific process of this step is as follows:

[0055] 1.1) Collect real-time on-site operating data and basic pipeline parameters through the SCADA system.

[0056] Specifically, real-time operational data includes temperature, pressure, flow rate, density, and viscosity, while basic pipeline parameters include pipe diameter and length.

[0057] 1.2) Based on the density data changes in the real-time on-site operation data, determine the start and end times of the mixed oil movement in a certain section of the pipeline, as well as the initial and final mixed oil information.

[0058] 1.3) Based on the start and end times of the mixed oil's movement in a certain section of the pipeline, as well as the initial and final mixed oil information, determine the temperature, pressure, flow rate, and oil properties (density, viscosity) during the mixed oil's movement in the pipeline, and calculate the Reynolds number.

[0059] 1.4) Based on Austin's formula, characteristic transformations are performed on the pipe diameter, pipe length, and Reynolds number to obtain the corresponding pipeline characteristic parameters and hydrothermal parameters (inlet and outlet pressure, inlet and outlet flow rate, and inlet and outlet temperature at each station of the pipeline).

[0060] 1.5) Construct a blending database based on initial blending information, final blending information, oil properties, pipeline characteristic parameters, and hydrothermal parameters.

[0061] 2) such as Figure 2 As shown, a pre-constructed prediction model for the mixing length of oil is constructed, specifically as follows:

[0062] 2.1) A deep learning algorithm is used to pre-build a mixed oil length prediction model.

[0063] Specifically, the mixed oil length prediction model includes an input layer, a hidden layer, a decoding layer, and an output layer. The input layer incorporates a feature adaptive weighting module, while the output layer includes a mechanism formula result fusion module. To characterize the strong nonlinear relationship between mixed oil length and input variables, the ReLU function is used as the activation function. To optimize the parameters of the mixed oil length prediction model and improve prediction accuracy, the Adam optimizer is used for model training.

[0064] The input layer is used to input pipeline characteristic parameters, hydrothermal parameters, oil properties, and initial oil mixing information.

[0065] The feature adaptive weighting module is used to perform weighted transformation on the input data of the input layer to capture the correlation between variables, output the weight of each input variable on the importance of the mixed oil length, and use the weighted input variables as the input of the mixed oil length prediction model.

[0066] Hidden layers are used to capture the nonlinear relationships between different variables and changes in the length of the mixed oil.

[0067] The decoding layer is used to weight the input features and restore them to the same dimension as the input. The weighted input features are used for oil mixing length prediction.

[0068] The mechanism formula result fusion module is used to couple the output results of the deep learning model with the pre-acquired calculation results of the mixing length mechanism formula to obtain the predicted value of the mixing length.

[0069] The output layer is used to output the predicted value of the oil mixing length.

[0070] Specifically, for the feature adaptive weighting module:

[0071] After obtaining the mixed oil database, there are many input feature categories, and each parameter has a different impact on the final output mixed oil length. Therefore, it is necessary to customize a network layer to extract input features based on the importance of each feature. Simply put, it involves weighting each input to the model, and the final determination of the weights adopts an adaptive strategy, aiming to achieve the best final prediction result. The construction process of the feature adaptive weighting module is as follows: a hidden layer is customized to perform a weighted transformation on the input variables of the input layer, capturing the correlation between variables, outputting the weight of each input variable's importance to the mixed oil length, and then using this weight to weight the input variables as the input to the mixed oil length prediction model.

[0072] Assume the input variables for the mixed oil length prediction model are:

[0073] X = [X1, X2, X3, ..., X...] N (1)

[0074] Where N is the dimension of the input variable X, and the input is mapped through the hidden layer as follows:

[0075] h = [h1, h2, h3, ..., h m (2)

[0076] The mapping process can be represented as:

[0077] h = f (1) (w (1) x+b (1) (3)

[0078] Where h represents the hidden layer features, f (1) For the activated ReLU function, w (1) Let x be the weight matrix corresponding to the input in the hidden layer, and b be the input feature matrix of each hidden layer. (1) Given the corresponding bias vector, the hidden layer features h are then decoded to construct weights q with the same dimension as the input variable X:

[0079] q = [q1, q2, q3, ..., q n (4)

[0080] The process can be expressed as follows:

[0081] q = f (2) (w (2) h+b (2) (5)

[0082] Where, q n The input features are the weighted hidden layer features, f. (2) For the activated sigmoid function, w (2)Let b be the weight matrix in the decoding layer corresponding to the hidden layer. (2) This is the corresponding bias vector, where the weight q reflects the correlation between the input and output variables, and has already been passed through f. (2) By constraining the output to the range [0,1], a new input can be obtained by weighting q and x. The final input xq for the mixed oil length prediction model is:

[0083] xq=[x1q1,x2q2,x3q3,...,x n q n (6)

[0084] Where, x n These are the different input features in the input layer.

[0085] Specifically, for the mechanism formula result fusion module:

[0086] While empirical mechanistic models lack accuracy, they reveal certain patterns in oil mixing at the mechanistic level and possess mechanistic interpretability. In contrast, purely data-driven methods, although capable of fitting complex nonlinear relationships between oil mixing inputs and outputs, lack mechanistic interpretability in their results. Therefore, this invention couples the two approaches to improve the model's mechanistic interpretability and nonlinear fitting capability. The calculation result of the oil mixing length mechanism formula is used as input in the layer preceding the final output layer, and adaptive weights are employed to obtain the final prediction result.

[0087] If the output variable of the mixed oil length prediction model is y1, the calculation result of the mixed oil length mechanism formula is y2, and q1 and q2 are the weights corresponding to the two input variables in the coupling layer (mechanism formula result fusion module), then the new output y after coupling is:

[0088] y = y1q1 + y2q2

[0089] Among them, q1 and q2 are adaptively adjusted according to the error between the coupled output value and the true value.

[0090] 2.2) Using a pre-built mixed oil database, the mixed oil length prediction model is trained and validated to obtain a trained mixed oil length prediction model.

[0091] 3) Obtain real-time on-site operating data and basic parameters of the pipeline to be tested, and calculate pipeline characteristic parameters, hydrothermal parameters, oil properties, and initial blending information, specifically:

[0092] 3.1) Obtain real-time on-site operating data and basic parameters of the pipeline to be tested through the SCADA system.

[0093] 3.2) Based on the density data changes in the real-time on-site operation data, determine the start and end times of the mixed oil movement in a certain section of the pipeline, as well as the initial and final mixed oil information.

[0094] 3.3) Based on the start and end times of the mixed oil's movement in a certain section of the pipeline, as well as the initial and final mixed oil information, determine the temperature, pressure, flow rate, and oil properties of the mixed oil during the pipeline's movement time, and calculate the Reynolds number.

[0095] 3.4) Based on Austin's formula, characteristic transformations are performed on the pipe diameter, pipe length, and Reynolds number to obtain the corresponding pipe characteristic parameters and hydrothermal parameters.

[0096] 4) Obtain the calculation results of the mixing length mechanism formula of the pipeline to be tested.

[0097] 5) Input the obtained pipeline characteristic parameters, hydrothermal parameters, oil properties and initial oil mixing information into the pre-trained oil mixing length prediction model, and fuse it with the calculation results of the oil mixing length mechanism formula to obtain the predicted value of the oil mixing length of the pipeline to be tested.

[0098] The following detailed embodiments illustrate the method for calculating the mixed oil length of a finished oil pipeline according to the present invention:

[0099] A test was conducted on a refined oil pipeline network. The feature adaptive weighting module and the mechanism formula result fusion module from the mixed oil length prediction model constructed in this invention were not included in Test 1; only the feature adaptive weighting module was included in Test 2; and both modules were included in Test 3. The prediction results are shown in Table 1 below, with RMSE, MAE, and MAPE used as evaluation indicators. RMSE is the root mean square error, MAE is the mean absolute error, and MAPE is the mean absolute percentage error. It can be seen that after introducing the feature adaptive weighting module and the mechanism formula result fusion module, the mixed oil length prediction error is reduced to varying degrees, proving the effectiveness of the input adaptive weighting and the coupling of mechanism formula results.

[0100] Table 1: Comparison of Prediction Results for Each Module

[0101]

[0102] To verify the superiority of the method of this invention, its calculation results were compared with those of Austin's formula, a hybrid model combining data-driven and empirical formulas, and commonly used data-driven algorithms such as Random Forest (RFR) and Decision Tree (XGB), as shown in Table 2 below. Using RMSE, MAE, and MAPE as metrics, it can be seen that the method of this invention significantly improves upon other methods in all three metrics. This demonstrates that the method of this invention comprehensively considers the influencing factors at the mechanistic level, and that the scheme of introducing adaptive feature weighting and fusion of mechanistic model results is effective and has engineering application value.

[0103] Table 2: Comparison with Predictions Based on Existing Technologies

[0104]

[0105] As can be seen, compared with traditional data-driven algorithms such as XGB, DNN, and RFR, the method of this invention has lower prediction errors for the oil mixing length, with RFR exhibiting the worst prediction performance among the three methods. Compared with the traditional Austin formula method, the method of this invention achieves better performance than empirical models. Furthermore, compared with existing research, the method of this invention, by coupling three different modules, can predict a more accurate oil mixing length.

[0106] After the oil batch is injected, it mixes with the preceding oil in the pipeline. The length of the mixture is detected at the batch interface upon arrival at the station and compared with the corresponding prediction result. Table 3 shows a comparison of the prediction errors of the Austin formula method, the method of this invention, and the hybrid model combining data-driven and empirical formulas.

[0107] Table 3: Prediction Results Presentation and Comparison

[0108]

[0109] Example 2

[0110] This embodiment provides a system for calculating the mixed oil length of refined oil pipelines, which combines coupling mechanisms and data, including:

[0111] The parameter acquisition and processing module is used to acquire real-time on-site operating data and basic parameters of the pipeline to be tested, and to calculate pipeline characteristic parameters, hydrothermal parameters, oil properties and initial oil mixing information.

[0112] The calculation result acquisition module is used to obtain the calculation results of the mixing length mechanism formula of the pipeline to be tested finished oil.

[0113] The oil mixing length prediction module is used to input the obtained pipeline characteristic parameters, hydrothermal parameters, oil properties and initial oil mixing information into the pre-trained oil mixing length prediction model, and integrate it with the calculation results of the oil mixing length mechanism formula to obtain the predicted value of the oil mixing length of the pipeline to be tested.

[0114] In a preferred embodiment, the parameter acquisition processing module includes:

[0115] The basic parameter acquisition unit is used to acquire real-time on-site operating data and basic parameters of the pipeline to be tested through the SCADA system.

[0116] The oil mixing information determination unit is used to determine the start and end times of oil mixing in a certain section of pipeline, as well as the initial and final oil mixing information, based on the density data changes in the real-time on-site operation data.

[0117] The calculation unit is used to determine the temperature, pressure, flow rate and oil properties of the mixed oil during the pipeline transportation time based on the start and end times of the mixed oil transportation in a certain pipeline section, as well as the initial and final mixed oil information, and to calculate the Reynolds number.

[0118] The characteristic transformation unit is used to perform characteristic transformation on pipe diameter, pipe length and Reynolds number according to Austin's formula to obtain the corresponding pipe characteristic parameters and hydrothermal parameters.

[0119] The system provided in this embodiment is used to execute the above-described method embodiments. For specific processes and details, please refer to the above embodiments, which will not be repeated here.

[0120] Example 3

[0121] This embodiment provides a processing device corresponding to the method for calculating the mixed oil length of a refined oil pipeline, which is based on the coupling mechanism and data provided in Embodiment 1. The processing device can be applied to client-side processing devices, such as mobile phones, laptops, tablets, and desktop computers, to execute the method of Embodiment 1.

[0122] The processing device includes a processor, a memory, a communication interface, and a bus. The processor, memory, and communication interface are connected via the bus to enable communication between them. The memory stores a computer program that can run on the processing device. When the processing device runs the computer program, it executes the method for calculating the mixed oil length of refined oil pipelines based on the coupling mechanism and data provided in Embodiment 1.

[0123] In some implementations, the memory may be high-speed random access memory (RAM), and may also include non-volatile memory, such as at least one disk storage device.

[0124] In other implementations, the processor can be any type of general-purpose processor, such as a central processing unit (CPU) or a digital signal processor (DSP), and there is no limitation here.

[0125] Furthermore, the logical instructions in the aforementioned memory can be implemented as software functional units and sold or used as independent products, and can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, or a part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of the present invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0126] Those skilled in the art will understand that the structure of the above-described computing device is only a partial structure related to the solution of this application and does not constitute a limitation on the computing device to which the solution of this application is applied. A specific computing device may include more or fewer components, or combine certain components, or have different component arrangements.

[0127] Example 4

[0128] This embodiment provides a computer program product corresponding to the method for calculating the mixed oil length of a refined oil pipeline based on the coupling mechanism and data provided in Embodiment 1. The computer program product may include a computer-readable storage medium on which computer-readable program instructions are loaded for executing the method for calculating the mixed oil length of a refined oil pipeline based on the coupling mechanism and data described in Embodiment 1.

[0129] A computer-readable storage medium can be a tangible device that holds and stores instructions for use by an instruction execution device. A computer-readable storage medium can be, for example, but not limited to, an electrical storage device, a magnetic storage device, an optical storage device, an electromagnetic storage device, a semiconductor storage device, or any combination thereof.

[0130] The computer-readable storage medium provided in the above embodiments has a similar implementation principle and technical effect to the above method embodiments, and will not be described again here.

[0131] This application is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of this application. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart... Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0132] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0133] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0134] The above embodiments are only used to illustrate the present invention. The structure, connection method and manufacturing process of each component can be varied. All equivalent transformations and improvements made on the basis of the technical solution of the present invention should not be excluded from the protection scope of the present invention.

Claims

1. A method for calculating the mixed oil length of a refined oil pipeline, characterized in that, include: The system acquires real-time on-site operating data and basic parameters of the pipeline to be tested, and calculates pipeline characteristic parameters, hydrothermal parameters, oil properties and initial mixing information. Obtain the calculation results of the mixing length mechanism formula of the pipeline to be tested; The obtained pipeline characteristic parameters, hydrothermal parameters, oil properties and initial oil mixing information are input into the pre-trained oil mixing length prediction model and fused with the calculation results of the oil mixing length mechanism formula to obtain the predicted value of the oil mixing length of the pipeline to be tested. The mixed oil length prediction model includes an input layer, a hidden layer, a decoding layer, and an output layer. The input layer is equipped with a feature adaptive weighting module, and the output layer is equipped with a mechanism formula result fusion module. The input layer is used to input pipeline characteristic parameters, hydrothermal parameters, oil properties, and initial oil mixing information; The feature adaptive weighting module is used to perform weighted transformation on the data input to the input layer to capture the correlation between variables, output the weight of each input variable on the importance of the mixed oil length, and use the weighted input variables as the input of the mixed oil length prediction model. The hidden layer is used to capture the nonlinear correlation between different variables and the change in the length of the mixed oil; The decoding layer is used to weight the input features and restore them to the same dimension as the input. The weighted input features are used for oil mixing length prediction. The mechanism formula result fusion module is used to couple the output result of the deep learning model with the pre-acquired calculation result of the oil mixing length mechanism formula to obtain the predicted value of the oil mixing length. The output layer is used to output the predicted value of the oil mixing length.

2. The method for calculating the mixed oil length of a refined oil pipeline based on coupling mechanism and data as described in claim 1, characterized in that, The construction process of the oil mixing length prediction model is as follows: A deep learning algorithm is used to pre-build a prediction model for the mixing length of mixed oil. Using a pre-built oil mixing database, the oil mixing length prediction model is trained and validated to obtain a well-trained oil mixing length prediction model.

3. The method for calculating the mixed oil length of a refined oil pipeline based on coupling mechanism and data as described in claim 2, characterized in that, The oil blending database includes initial oil blending information, final oil blending information, oil properties, pipeline characteristic parameters, and hydrothermal parameters.

4. The method for calculating the mixed oil length of a refined oil pipeline based on coupling mechanism and data as described in claim 1, characterized in that, Input to the oil mixing length prediction model for: in, For weights; For each different input feature in the input layer; These are the weighted hidden layer features; The output of the oil mixing length prediction model for: in, For the mixed oil length mixing prediction model, these are the output variables; The result is the calculation result of the oil mixing length mechanism formula; , The weights are the two input variables in the mechanism formula result fusion module.

5. The method for calculating the mixed oil length of a refined oil pipeline based on coupling mechanism and data as described in claim 1, characterized in that, The activation function of the mixed oil length prediction model is the ReLU function, and the model is trained using the Adam optimizer.

6. The method for calculating the mixed oil length of a refined oil pipeline based on coupling mechanism and data as described in claim 1, characterized in that, The process of acquiring real-time on-site operating data and basic parameters of the pipeline to be tested, and calculating pipeline characteristic parameters, hydrothermal parameters, oil properties, and initial blending information, includes: The real-time on-site operating data and basic parameters of the pipeline to be tested are obtained through the SCADA system. Based on the density data changes in the real-time on-site operation data, determine the start and end times of the mixed oil migration in a certain section of the pipeline, as well as the initial and final mixed oil information; Based on the start and end times of the mixed oil's movement in a certain pipeline segment, as well as the initial and final mixed oil information, the temperature, pressure, flow rate, and oil properties of the mixed oil during the pipeline's movement time are determined, and the Reynolds number is calculated. Based on Austin's formula, characteristic transformations are performed on pipe diameter, pipe length, and Reynolds number to obtain the corresponding pipe characteristic parameters and hydrothermal parameters.

7. A system for calculating the mixed oil length of a refined oil pipeline, characterized in that, include: The parameter acquisition and processing module is used to acquire the real-time on-site operating data and basic parameters of the pipeline to be tested, and to calculate the pipeline characteristic parameters, hydrothermal parameters, oil properties and initial mixing information. The calculation result acquisition module is used to obtain the calculation results of the mixing length mechanism formula of the pipeline to be tested refined oil; The oil mixing length prediction module is used to input the obtained pipeline characteristic parameters, hydrothermal parameters, oil properties and initial oil mixing information into the pre-trained oil mixing length prediction model, and integrate it with the calculation results of the oil mixing length mechanism formula to obtain the predicted value of the oil mixing length of the pipeline to be tested. The mixed oil length prediction model includes an input layer, a hidden layer, a decoding layer, and an output layer. The input layer is equipped with a feature adaptive weighting module, and the output layer is equipped with a mechanism formula result fusion module. The input layer is used to input pipeline characteristic parameters, hydrothermal parameters, oil properties, and initial oil mixing information; The feature adaptive weighting module is used to perform weighted transformation on the data input to the input layer to capture the correlation between variables, output the weight of each input variable on the importance of the mixed oil length, and use the weighted input variables as the input of the mixed oil length prediction model. The hidden layer is used to capture the nonlinear correlation between different variables and the change in the length of the mixed oil; The decoding layer is used to weight the input features and restore them to the same dimension as the input. The weighted input features are used for oil mixing length prediction. The mechanism formula result fusion module is used to couple the output result of the deep learning model with the pre-acquired calculation result of the oil mixing length mechanism formula to obtain the predicted value of the oil mixing length. The output layer is used to output the predicted value of the oil mixing length.

8. A processing apparatus, characterized in that, It includes computer program instructions, wherein when the computer program instructions are executed by a processing device, they are used to implement the steps corresponding to the method for calculating the mixed oil length of a refined oil pipeline according to any one of claims 1-6, which involves coupling mechanism and data.

9. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer program instructions, wherein when the computer program instructions are executed by a processor, they are used to implement the steps corresponding to the method for calculating the mixed oil length of a refined oil pipeline based on the coupling mechanism and data as described in any one of claims 1-6.

Images