Oil mixing cut point identification method and system

Abstract

This disclosure relates to the field of mixed oil cutting in long-distance pipelines for refined oil products in the oil and gas storage and transportation industry. It provides a method and system for identifying mixed oil cutting points. By acquiring the time point of mixed oil cutting and obtaining real-time data from the SCADA system, the corresponding mixed oil cutting point judgment results are obtained based on density calculation decision models, sound velocity calculation decision models, and optical refractive index calculation decision models using this real-time data. The weights corresponding to the judgment results of each mixed oil cutting point are added together. When the sum of the weights exceeds a preset judgment threshold, it is determined that the mixed oil cutting time point has actually been reached. This achieves accurate identification of the mixed oil cutting time point when different types of refined oil are transported sequentially through pipelines. It can reduce mixed oil losses while ensuring oil quality, improve the economic efficiency of pipeline operation, and also ensure the real-time performance of the identification. It has a wide range of applications and high stability. It can be integrated with the SCADA system or run independently to guide mixed oil cutting and reduce the workload of mixed oil cutting dispatchers.

Description

Technical Field

[0001] This disclosure relates to the field of oil mixing and cutting technology in long-distance pipelines for refined oil in the oil and gas storage and transportation industry, and in particular to a method and system for identifying oil mixing and cutting points. Background Technology

[0002] During the sequential transport of refined oil products through pipelines, oil mixing can occur, meaning that different types or properties of oil mix within the pipeline. This mixing phenomenon leads to the formation of a distinct mixing interface within the pipeline. Therefore, accurately identifying the location of this interface is crucial for ensuring the quality of the transported oil products.

[0003] Traditional methods for identifying mixed oil interfaces primarily rely on single interface detection devices, such as densitometers and optical interface detectors. However, the complexity of operating conditions can affect the performance and accuracy of these devices, limiting their application under specific conditions. For example, densitometers are effective at identifying interfaces between mixed oils with significant density differences, but their performance is less than ideal when identifying interfaces between similar oils due to their similar densities. Optical interface detectors are more effective at identifying interfaces between oils with similar densities, but their data are relative, leading to inconsistent responses and poor stability across different batches of oil.

[0004] Therefore, how to more comprehensively and accurately identify the oil mixing cutting point to reduce the workload of dispatchers, reduce oil mixing losses, improve the economic benefits of pipeline transportation, and lay the foundation for automated oil mixing cutting has become an urgent problem to be solved in this field. Summary of the Invention

[0005] This disclosure aims to address at least one of the problems existing in the prior art by providing a method and system for identifying oil-mixed cutting points.

[0006] One aspect of this disclosure provides a method for identifying oil-mixed cutting points, the method comprising:

[0007] Obtain the cutting time point for mixed oil;

[0008] Acquire real-time data from the SCADA system; the real-time data includes density detection values, sound velocity detection values, optical refractive index detection values, and their respective trend change values.

[0009] The density detection value, the sound velocity detection value, the optical refractive index detection value, and their respective trend change values ​​are input into the density calculation decision model, the sound velocity calculation decision model, and the optical refractive index calculation decision model, respectively, to obtain the corresponding oil mixing cutting point determination results.

[0010] According to the preset weighting scheme for the judgment results, the weights corresponding to the judgment results of each mixed oil cutting point are determined respectively, and the weights are added together to obtain the corresponding weight sum value;

[0011] When the weight and value exceed the preset judgment threshold, it is determined that the oil mixing cutting time point has actually been reached, and an oil mixing cutting point identification signal is output.

[0012] Optionally, obtaining the cutting time point of the mixed oil includes:

[0013] Select the corresponding blending potential calculation model based on the blending type;

[0014] Calculate the potential volume of the oil mixture corresponding to the oil mixture type based on the oil mixture potential calculation model.

[0015] The oil mixing cut-off time point is determined based on the oil mixing potential volume.

[0016] Optionally, the blending potential calculation model includes a blending potential calculation model for gasoline blended with diesel, a blending potential calculation model for diesel blended with gasoline, a blending potential calculation model for two types of gasoline transported sequentially, and a blending potential calculation model for two types of diesel transported sequentially.

[0017] The step of selecting the corresponding blending potential calculation model based on the blending type includes:

[0018] The type of mixed oil is determined based on the oil delivery sequence information input by the user and the data parameters of the upstream station's detection instruments;

[0019] When the blending type is gasoline mixed with diesel, the blending potential calculation model for gasoline mixed with diesel is selected; when the blending type is diesel mixed with gasoline, the blending potential calculation model for diesel mixed with gasoline is selected; when the blending type is two types of gasoline transported sequentially, the blending potential calculation model for the sequential transport of the two types of gasoline is selected; when the blending type is two types of diesel transported sequentially, the blending potential calculation model for the sequential transport of the two types of diesel is selected.

[0020] The calculation model for the blending potential of gasoline with diesel is expressed as follows:

[0021]

[0022] Among them, K C This indicates the permissible concentration of diesel fuel mixed into pure gasoline; [t] a ] indicates the permissible final boiling point of gasoline as specified in the preset standard; t a Indicates the actual final boiling point of gasoline with a certain quality potential;

[0023] The calculation model for the blending potential of diesel fuel with gasoline is expressed as follows:

[0024]

[0025] Among them, K Q This indicates the permissible concentration of gasoline blended into pure diesel fuel; [t] b ] indicates the permissible flash point of diesel fuel as specified in the preset standard; t b Indicates the actual flash point of diesel fuel with a certain quality potential;

[0026] The calculation model for the blending potential of the two types of gasoline delivered sequentially is expressed as follows:

[0027]

[0028] Among them, K X This indicates the permissible volume percentage of a second grade of gasoline blended into a first grade of gasoline, where the first grade is greater than the second grade; N represents the minimum octane rating of the first grade of gasoline as specified in a preset standard; N1 represents the actual octane rating of the first grade of gasoline; and N2 represents the actual octane rating of the second grade of gasoline.

[0029] The calculation model for the blending potential of the two types of diesel fuel delivered sequentially is expressed as follows:

[0030]

[0031] Among them, K S The percentage by volume of a third-grade diesel fuel blended with a fourth-grade diesel fuel is indicated, where the third-grade diesel fuel is less than the fourth-grade diesel fuel; I represents the flash point index corresponding to the permissible flash point of the third-grade diesel fuel specified in the preset standard; I1 represents the flash point index corresponding to the actual flash point of the fourth-grade diesel fuel; and I2 represents the flash point index corresponding to the actual flash point of the third-grade diesel fuel.

[0032] Optionally, calculating the mixing potential volume corresponding to the mixing type based on the mixing potential calculation model includes:

[0033] When the blending type is gasoline mixed with diesel, the potential volume of the blending head is calculated according to the following formula:

[0034] Q M =Q*K C ;

[0035] Among them, Q M Q represents the potential volume of the blended oil; Q represents the volume of oil entering the pure oil tank.

[0036] When the blending type is gasoline mixed with diesel, the potential volume of the blend at the tail end is calculated according to the following formula:

[0037] Q M =Qt *K Q ;

[0038] Among them, Q t Indicates the capacity of the receiving storage tank;

[0039] When the blending type is diesel mixed with gasoline, the mixing potential volume at the mixing head is calculated according to the following formula:

[0040] Q M =Q*K Q ;

[0041] When the blending type is diesel mixed with gasoline, the potential volume of the blend at the tail end is calculated according to the following formula:

[0042] Q M =Q t *K C ;

[0043] When the blending type involves sequential delivery of two types of gasoline, with the first grade gasoline preceding the second grade gasoline, the blending potential volume is calculated using the following formula:

[0044] Q M =Q*K X ;

[0045] When the blending type involves sequential delivery of two types of gasoline, with the second grade gasoline preceding the first grade gasoline, the blending potential volume is calculated using the following formula:

[0046] Q M =Q t *K X ;

[0047] Among them, Q t This indicates the capacity of the receiving storage tank for the first grade of gasoline;

[0048] When the blending type involves the sequential delivery of two types of diesel fuel, with the preceding fuel being the fourth grade diesel fuel and the following fuel being the third grade diesel fuel, the blending potential volume is calculated according to the following formula:

[0049] Q M =Q*Ks;

[0050] When the blending type involves the sequential delivery of two types of diesel fuel, with the preceding fuel being the third grade diesel fuel and the following fuel being the fourth grade diesel fuel, the blending potential volume is calculated according to the following formula:

[0051] Q M =Q t *Ks;

[0052] Among them, Q tThis indicates the capacity of the receiving storage tank for the fourth grade of diesel oil.

[0053] Optionally, determining the oil mixing cut-off time point based on the oil mixing potential volume includes:

[0054] Calculate the real-time volume of the blended oil in the mixed oil using the following formula:

[0055] Q x =(ρ 实 -ρ2)*Q 测 / (ρ1-ρ2);

[0056] Among them, Q x Q represents the real-time volume of the blended oil; 测 ρ represents the total volume of the mixed oil; 实 ρ1 represents the density of the mixed oil; ρ2 represents the density of the mixed oil as determined by laboratory tests; ρ3 represents the density of the pure oil as determined by laboratory tests.

[0057] The real-time volume of the blended oil is integrated over time to determine the moment when the real-time volume of the blended oil equals the potential volume of the blended oil, and this moment is taken as the blended oil cutting time point.

[0058] Optionally, obtaining the cutting time point of the mixed oil also includes:

[0059] Based on pipeline operation experience, the timing for cutting off the mixed oil was determined directly.

[0060] Optionally, the density calculation decision model, the sound speed calculation decision model, and the optical refractive index calculation decision model are constructed according to the following steps:

[0061] The SCADA system is used to obtain historical density data, historical sound velocity data, and historical optical refractive index data for different types of oil mixtures.

[0062] Determine the density change characteristic curve equation, sound speed change characteristic curve equation, and optical refractive index change characteristic curve equation corresponding to the density history data, the sound speed history data, and the optical refractive index history data, respectively.

[0063] Based on the density change characteristic curve equation, the sound speed change characteristic curve equation, and the optical refractive index change characteristic curve equation, the density calculation decision model, the sound speed calculation decision model, and the optical refractive index calculation decision model are constructed respectively using a black-box model based on the time clustering decision tree algorithm.

[0064] Optionally, determining the density change characteristic curve equation, sound speed change characteristic curve equation, and optical refractive index change characteristic curve equation corresponding to the density history data, the sound speed history data, and the optical refractive index history data, respectively, includes:

[0065] Based on the historical density data, the historical sound velocity data, and the historical optical refractive index data, the historical density values, historical sound velocity values, and historical optical refractive index values ​​corresponding to the oil mixing segments of each type of oil mixing are extracted in chronological order.

[0066] Based on the historical density values, historical sound velocity values, and historical optical refractive index values, the density difference, sound velocity difference, and optical refractive index difference values ​​between the next moment and the previous moment are determined using a difference algorithm. Then, the first derivatives of the historical density values, historical sound velocity values, and historical optical refractive index values ​​over a certain period of time are calculated to obtain the corresponding density change trend characteristics, sound velocity change trend characteristics, and optical refractive index change trend characteristics.

[0067] Based on the density difference value and the density change trend characteristics, the sound velocity difference value and the sound velocity change trend characteristics, and the optical refractive index difference value and the optical refractive index change trend characteristics, the change characteristic curve equations corresponding to the density value, sound velocity value, and optical refractive index value in the mixing segment of each type of oil mixture are fitted respectively, so as to obtain the corresponding density change characteristic curve equation, sound velocity change characteristic curve equation, and optical refractive index change characteristic curve equation.

[0068] Optionally, the step of constructing the density calculation decision model, the sound speed calculation decision model, and the optical refractive index calculation decision model based on the density change characteristic curve equation, the sound speed change characteristic curve equation, and the optical refractive index change characteristic curve equation, respectively, using a black-box model based on a time-clustering decision tree algorithm, includes:

[0069] The density change characteristic curve equation, the sound speed change characteristic curve equation, and the optical refractive index change characteristic curve equation are used as white-box models and respectively input into the black-box model based on the time clustering decision tree algorithm to obtain the probability of each cluster and the predicted label of each data point output by the black-box model based on the time clustering decision tree algorithm.

[0070] The black-box model based on the time-clustering decision tree algorithm is optimized using the following formula to construct the density calculation decision model, the sound speed calculation decision model, and the optical refractive index calculation decision model:

[0071]

[0072] Where, minimize J means minimizing the value of J; J represents the objective function value of clustering in the black-box model based on the time-based clustering decision tree algorithm, used to measure the quality of clustering; α represents a coefficient used to control the compactness within the clusters; β represents a coefficient used to control the temporal smoothness of the clusters; x tThis represents the t-th data point among historical density values, historical sound velocity values, or historical optical refractive index values ​​arranged in chronological order; μ c Let d(x) represent the c-th cluster center, where c is the cluster center number and its value ranges from 1 to C, and C represents the total number of cluster centers; t μ c ) represents data point x t With cluster center μ c The distance between them; y t Represents data point x t Corresponding real label; y t′ The data point x represents the output of the black-box model based on the time-clustering decision tree algorithm. t′ The corresponding predicted label; Q(y) t y t′ ) represents data point x t Corresponding label y t With data point x t′ Corresponding label y t′ The similarity measure between them; t and t′ both represent data point numbers and their values ​​range from 1 to T, where T represents the total number of data points.

[0073] Another aspect of this disclosure provides a system for identifying oil-mixed cutting points, the system comprising:

[0074] The first acquisition module is used to acquire the cutting time point of the mixed oil;

[0075] The second acquisition module is used to acquire real-time data from the SCADA system; the real-time data includes density detection values, sound velocity detection values, optical refractive index detection values, and their respective trend change values.

[0076] The decision module is used to input the density detection value, the sound velocity detection value, the optical refractive index detection value and their respective trend change values ​​into the density calculation decision model, the sound velocity calculation decision model and the optical refractive index calculation decision model, respectively, to obtain the corresponding oil mixing cutting point determination result;

[0077] The determination module is used to determine the weight corresponding to each of the mixed oil cutting points according to the preset judgment result weighting scheme, and add the weights to obtain the corresponding weight sum value;

[0078] The determination module is used to determine that the oil mixing cutting time point has actually been reached when the weight and value exceed the preset determination threshold, and outputs an oil mixing cutting point identification signal.

[0079] The oil mixing cut-off point identification method and system disclosed herein, compared with existing technologies, obtains the oil mixing cut-off time point and real-time data from the SCADA system. Using this real-time data, based on density calculation decision models, sound velocity calculation decision models, and optical refractive index calculation decision models, the corresponding oil mixing cut-off point determination results are obtained. Then, the weights corresponding to each oil mixing cut-off point determination result are added together. When the sum of the weights exceeds the corresponding preset determination threshold, the oil mixing cut-off time point is determined to have actually arrived. This achieves accurate identification of the oil mixing cut-off time point during the sequential transportation of different types of refined oil products through pipelines. It can reduce oil mixing losses while ensuring oil quality and improve the economic efficiency of pipeline operation. Furthermore, the oil mixing cut-off point identification method and system disclosed herein can automatically complete the oil mixing cut-off point identification, with a determination cycle of less than 5 seconds, ensuring both accurate identification and real-time performance. The oil mixing cut-off point identification method and system disclosed herein can also be applied to different pipeline sequential transportation situations, with a wide range of applications and high stability. It can be integrated with the SCADA system or operate independently to guide refined oil pipelines in oil mixing cut-off, thereby reducing the workload of oil mixing cut-off dispatchers. Attached Figure Description

[0080] One or more embodiments are illustrated by way of example with the corresponding pictures in the accompanying drawings. These illustrations do not constitute a limitation on the embodiments. Elements with the same reference numerals in the drawings are denoted as similar elements. Unless otherwise stated, the figures in the drawings are not to be limited by scale.

[0081] Figure 1 A flowchart illustrating a method for identifying oil-mixed cutting points according to an embodiment of this disclosure;

[0082] Figure 2 A density variation trend characteristic curve of the diesel-gasoline mixture section provided for another embodiment of this disclosure;

[0083] Figure 3 A characteristic curve of sound velocity variation in a diesel-gasoline mixture section provided for another embodiment of this disclosure;

[0084] Figure 4 A characteristic curve of the optical refractive index variation trend of a diesel-gasoline mixture section provided in another embodiment of this disclosure;

[0085] Figure 5 A flowchart illustrating a method for identifying oil-mixed cutting points according to another embodiment of this disclosure;

[0086] Figure 6 This is a schematic diagram of the structure of a mixed oil cutting point identification system provided for another embodiment of this disclosure. Detailed Implementation

[0087] To make the objectives, technical solutions, and advantages of the embodiments of this disclosure clearer, the various embodiments of this disclosure will be described in detail below with reference to the accompanying drawings. However, those skilled in the art will understand that many technical details have been provided in the various embodiments of this disclosure to facilitate a better understanding of the disclosure. However, the technical solutions claimed in this disclosure can be implemented even without these technical details and with various variations and modifications based on the following embodiments. The division of the various embodiments below is for ease of description and should not constitute any limitation on the specific implementation of this disclosure. The various embodiments can be combined with and referenced by each other without contradiction.

[0088] One embodiment of this disclosure relates to a method for identifying oil-mixed cutting points, the process of which is as follows: Figure 1 As shown, it includes steps S110 to S160.

[0089] Step S110: Obtain the cutting time point of the mixed oil.

[0090] Specifically, step S110 can obtain the oil mixing cutting time point in the following two ways.

[0091] Method 1: Obtain the cutting time point for mixed oil, including: directly determining the cutting time point for mixed oil based on pipeline operation experience.

[0092] Specifically, pipeline operation experience refers to the various knowledge and skills accumulated during pipeline operation and management. Method 1 can directly determine the oil mixing cut-off time based on this accumulated knowledge and skills, which is fast and efficient.

[0093] Method 2: Obtain the oil mixing cut-off time point, including: selecting the corresponding oil mixing potential calculation model based on the oil mixing type; calculating the oil mixing potential volume corresponding to the oil mixing type based on the oil mixing potential calculation model; and determining the oil mixing cut-off time point based on the oil mixing potential volume.

[0094] Specifically, Method 2 combines oil-related parameters and pipeline operation-related parameters to automatically calculate the oil mixing and cutting time, offering greater accuracy, speed, and efficiency. Oil-related parameters may include, but are not limited to, oil delivery sequence, oil quantity, and oil physical properties. Pipeline operation-related parameters may include, but are not limited to, preset pipeline parameters such as pipeline elevation, pipeline mileage, pipeline diameter, and pure oil tank capacity, as well as pipeline operation parameters obtained through pipeline monitoring instruments, such as parameters collected by temperature transmitters, pressure transmitters, densitometers, optical interface detectors, ultrasonic flow meters, and pump units at various pipeline stations.

[0095] Since different types of blended oils typically have different physical and chemical properties, in order to make the calculation results of the blended oil potential volume more accurate, the blended oil potential calculation model can be subdivided according to the different types of blended oils: gasoline blended with diesel blended with ...

[0096] Based on the above, the corresponding blending potential calculation model is selected according to the blending type, including: determining the blending type based on the user-input oil delivery sequence information and upstream station monitoring instrument data parameters. When the blending type is gasoline mixed with diesel, the gasoline-diesel blending potential calculation model is selected. When the blending type is diesel mixed with gasoline, the diesel-gasoline blending potential calculation model is selected. When the blending type involves the sequential delivery of two types of gasoline, the blending potential calculation model for the sequential delivery of two types of gasoline is selected. When the blending type involves the sequential delivery of two types of diesel, the blending potential calculation model for the sequential delivery of two types of diesel is selected.

[0097] Specifically, the user-inputted oil delivery sequence information indicates the order in which oil products should be delivered. Upstream station monitoring instrument data parameters can include information collected by equipment such as temperature transmitters, pressure transmitters, densitometers, optical interface detectors, ultrasonic flow meters, and pump units. This information indicates the type of oil being delivered upstream. Therefore, by combining the user-inputted oil delivery sequence information with the upstream station monitoring instrument data parameters, the current type of oil mixing can be determined.

[0098] The calculation model for the blending potential of gasoline with diesel is expressed as follows: Among them, K C This indicates the permissible concentration of diesel fuel blended into pure gasoline, expressed as a percentage. [t] a The value [] indicates the permissible final boiling point of gasoline as specified in the pre-defined standard, in °C. t a The actual final boiling point of gasoline with a certain quality potential is indicated by the unit °C.

[0099] The calculation model for the blending potential of diesel fuel with gasoline is expressed as follows:

[0100]

[0101] Among them, K Q This indicates the permissible concentration of gasoline blended into pure diesel fuel, expressed as a percentage. [t] b The value [] indicates the permissible flash point of diesel fuel as specified in the pre-defined standard, in °C. b The actual flash point of diesel fuel with a certain quality potential is expressed in °C.

[0102] The model for calculating the blending potential of two types of gasoline delivered sequentially is expressed as follows:

[0103]

[0104] Among them, K X This indicates the permissible volume percentage of Grade 1 gasoline blended with Grade 2 gasoline, expressed as a percentage. N represents the minimum octane rating of Grade 1 gasoline as specified in the predefined standard. N1 represents the actual octane rating of Grade 1 gasoline. N2 represents the actual octane rating of Grade 2 gasoline.

[0105] It should be noted that the first grade is higher than the second grade. The first grade gasoline can also be called higher octane gasoline or high-grade gasoline. The second grade gasoline can also be called lower octane gasoline or low-grade gasoline.

[0106] The calculation model for the blending potential of two types of diesel fuel delivered sequentially is expressed as follows:

[0107] Among them, K s This indicates the permissible volume percentage of fourth-grade diesel fuel blended into third-grade diesel fuel, expressed as a percentage. I represents the flash point index corresponding to the permissible flash point of third-grade diesel fuel specified in the pre-defined standard. I1 represents the flash point index corresponding to the actual flash point of fourth-grade diesel fuel. I2 represents the flash point index corresponding to the actual flash point of third-grade diesel fuel.

[0108] It should be noted that the third grade is lower than the fourth grade. The third grade diesel fuel can also be referred to as a lower grade diesel fuel or a low-octane diesel fuel. The fourth grade diesel fuel can also be referred to as a higher grade diesel fuel or a high-octane diesel fuel.

[0109] For example, based on the oil mixing potential calculation model, the oil mixing potential volume corresponding to the oil mixing type is calculated, including:

[0110] When the fuel mixture is gasoline mixed with diesel (i.e., gasoline-diesel), the preceding fuel is gasoline and the following fuel is diesel. The mixture is cut off from the pipeline. The potential volume of the mixture at the mixing head is calculated using the following formula: Q M =Q*K C And calculate the mixing potential volume at the tail end of the mixture according to the following formula: Q M =Q t *K Q Among them, Q M This indicates the potential volume of the mixed oil, in cubic meters (m³). 3 Q represents the volume of oil entering the pure oil tank, in cubic meters (m³). 3 This can be obtained by measuring the flow rate on site. Q t This indicates the capacity of the receiving storage tank, which is a known quantity.

[0111] When the fuel mixture is diesel mixed with gasoline (i.e., diesel-gasoline), the preceding fuel is diesel and the following fuel is gasoline. The mixture is cut off from the pipeline. The potential volume of the mixture at the mixing head is calculated using the following formula: Q M =Q*K Q And calculate the mixing potential volume at the tail end of the mixture according to the following formula: Q M =Q t *K C Where Q represents the volume of oil entering the pure oil tank, in cubic meters (m³). 3 It can be obtained by measuring the flow rate on site.

[0112] When the blending type involves sequential delivery of two types of gasoline, with the first type (high-octane) gasoline being delivered first and the second type (low-octane) gasoline being delivered second, the blending type can be expressed as high-octane gasoline - low-octane gasoline. In this case, the blended gasoline is not cut off, and the potential volume of the blend is calculated using the following formula: Q M =Q*K X .

[0113] When the blending type involves sequential delivery of two types of gasoline, with the first type being low-octane gasoline and the second type being high-octane gasoline, the blending type can be expressed as low-octane gasoline - high-octane gasoline. In this case, the blend is not cut off, and the potential volume of the blend is calculated using the following formula: Q M =Q t *K X Among them, Q t This indicates the tank capacity of the receiving storage tank for first-grade gasoline, i.e., high-octane gasoline, in cubic meters (m³). 3 .

[0114] When the blending type involves sequential delivery of two types of diesel fuel, with the preceding fuel being the fourth grade (high-grade) and the following fuel being the third grade (low-grade), the blending type can be expressed as high-grade diesel - low-grade diesel. In this case, the blended fuel is not cut off, and the potential volume of the blend is calculated using the following formula: Q M =Q*Ks.

[0115] When the blending type involves sequential delivery of two types of diesel fuel, with the preceding fuel being the third grade (low-grade) and the following fuel being the fourth grade (high-grade), the blending type can be expressed as low-grade diesel - high-grade diesel. In this case, the blended fuel is not cut off, and the potential volume of the blend is calculated using the following formula: Q M =Q t *Ks. Where Q t This indicates the tank capacity of the receiving storage tank for the fourth grade diesel oil, i.e., high-grade diesel oil.

[0116] For example, determining the cut-off time point for the mixed oil based on its potential volume includes: calculating the real-time volume of the blended oil in the mixed oil according to the following formula: Q x=(ρ 实 -ρ2)*Q 测 / (ρ1-ρ2). The real-time volume of the blended oil is integrated over time to determine the moment when the real-time volume of the blended oil equals its potential volume. This moment is taken as the blending cut-off time point. Where Q x This indicates the real-time volume of the blended oil. Q 测 ρ represents the total volume of the mixed oil. 实 ρ1 represents the density of the mixed oil. ρ2 represents the density of the pure oil.

[0117] It should be noted that "blended oil" here refers to the oil type that is subsequently used. For example, if the current oil type is gasoline and the subsequent oil type is diesel, the blended oil refers to diesel. If the current oil type is diesel and the subsequent oil type is gasoline, the blended oil refers to gasoline. If the current oil type is high-octane gasoline and the subsequent oil type is low-octane gasoline, the blended oil refers to low-octane gasoline. If the current oil type is low-octane gasoline and the subsequent oil type is high-octane gasoline, the blended oil refers to high-octane gasoline. If the current oil type is high-octane diesel and the subsequent oil type is low-octane diesel, the blended oil refers to low-octane gasoline. If the current oil type is low-octane diesel and the subsequent oil type is high-octane diesel, the blended oil refers to high-octane diesel.

[0118] In other words, the real-time volume Q of the blended oil is calculated. x After that, regarding Q x Integrating over time, when Q... x These are respectively equal to Q under the different types of oil blending mentioned above. M The corresponding time is the oil mixing cutting time point X under each type of oil mixing.

[0119] It should be noted that when calculating the potential volume of different types of oil blends, the oil physical property parameters such as final boiling point, flash point, and octane number in the relevant calculation formulas are the actual indicators within the blending section before oil blending is cut off. These parameters need to be obtained by combining the oil physical property indicators of the first station with the loss coefficient. The loss coefficient is affected by parameters such as pipeline mileage, pipeline elevation, and temperature, and needs to be obtained after the pipeline has been running for a certain period of time.

[0120] Step S120: Obtain real-time data from the SCADA system. Real-time data includes density detection values, sound velocity detection values, optical refractive index detection values, and their respective trend change values.

[0121] Specifically, SCADA system refers to Supervisory Control and Data Acquisition system. Since the densitometer, ultrasonic flow meter, and optical interface detector in the oil mixing cutting station are usually connected to the SCADA system, step S120 can directly acquire real-time data from the SCADA system, thereby obtaining the density detection value, sound velocity detection value, and optical refractive index detection value detected by the densitometer, ultrasonic flow meter, and optical interface detector, respectively, as well as their corresponding trend change values.

[0122] Step S130: Input the density detection value, sound velocity detection value, optical refractive index detection value and their corresponding trend change values ​​into the density calculation decision model, sound velocity calculation decision model and optical refractive index calculation decision model respectively to obtain the corresponding oil mixing cutting point determination results.

[0123] Specifically, the density calculation decision model, the sound velocity calculation decision model, and the optical refractive index calculation decision model can all be pre-built based on historical data in the historical database of the SCADA system. The historical data here can include historical density data, historical sound velocity data, and historical optical refractive index data obtained by density meters, ultrasonic flow meters, and optical interface detectors for different types of mixed oil sections.

[0124] For example, the density calculation decision model, sound velocity calculation decision model, and optical refractive index calculation decision model are constructed according to the following steps: Using a SCADA system, acquire historical density data, historical sound velocity data, and historical optical refractive index data corresponding to different types of oil mixtures. Determine the density change characteristic curve equations, sound velocity change characteristic curve equations, and optical refractive index change characteristic curve equations corresponding to the historical density data, historical sound velocity data, and historical optical refractive index data, respectively. Based on the density change characteristic curve equations, sound velocity change characteristic curve equations, and optical refractive index change characteristic curve equations, construct the density calculation decision model, sound velocity calculation decision model, and optical refractive index calculation decision model respectively using a black box (BB) model based on a time-clustered decision tree algorithm.

[0125] Specifically, when the oil medium in the pipeline changes, the corresponding density, sound velocity, and optical refractive index values ​​will also change accordingly, and the trends of change will differ at different lengths of the mixed oil section. Therefore, this implementation method uses historical density data, historical sound velocity data, and historical optical refractive index data to construct density change characteristic curve equations, sound velocity change characteristic curve equations, and optical refractive index change characteristic curve equations, respectively. These equations are then used as white box (WB) models and input into a black box model based on a time-based clustering decision tree algorithm, thereby constructing density calculation decision models, sound velocity calculation decision models, and optical refractive index calculation decision models as gray box (GB) models to improve the decision accuracy of each calculation decision model.

[0126] For example, determining the density change characteristic curve equation, sound velocity change characteristic curve equation, and optical refractive index change characteristic curve equation corresponding to historical density data, historical sound velocity data, and historical optical refractive index data respectively includes: based on the historical density data, historical sound velocity data, and historical optical refractive index data, extracting the historical density values, historical sound velocity values, and historical optical refractive index values ​​arranged in chronological order for each type of oil mixture segment. Based on the historical density values, historical sound velocity values, and historical optical refractive index values, using a difference algorithm, determining the density difference value, sound velocity difference value, and optical refractive index difference value between the next moment and the previous moment respectively, and then calculating the first derivative of each historical density value, historical sound velocity value, and historical optical refractive index value over a certain period of time to obtain the corresponding density change trend characteristics, sound velocity change trend characteristics, and optical refractive index change trend characteristics. Based on the density difference value and density change trend characteristics, sound velocity difference value and sound velocity change trend characteristics, and optical refractive index difference value and optical refractive index change trend characteristics, the change characteristic curve equations corresponding to the density value, sound velocity value, and optical refractive index value in the mixing segment of each type of oil mixture are fitted respectively, and the corresponding density change characteristic curve equation, sound velocity change characteristic curve equation, and optical refractive index change characteristic curve equation are obtained.

[0127] For example, when the fuel blend is diesel-gasoline, the density variation trend, sound velocity variation trend, and optical refractive index variation trend characteristics of the blend segment can be respectively as follows: Figure 2 , Figure 3 , Figure 4 As shown.

[0128] It should be noted that, in order to make the equations of each change characteristic curve closer to the actual change trend characteristics, the equations of density change characteristic curve, sound speed change characteristic curve, and optical refractive index change characteristic curve can be modified by combining parameters such as temperature and pipeline mileage. The modified equations of density change characteristic curve, sound speed change characteristic curve, and optical refractive index change characteristic curve can be used to construct the corresponding density calculation decision model, sound speed calculation decision model, and optical refractive index calculation decision model, thereby further improving the decision accuracy of the density calculation decision model, sound speed calculation decision model, and optical refractive index calculation decision model.

[0129] For example, based on the equations of density change characteristic curves, sound speed change characteristic curves, and optical refractive index change characteristic curves, and using a black-box model based on the time-based clustering decision tree algorithm, density calculation decision models, sound speed calculation decision models, and optical refractive index calculation decision models are constructed respectively. This includes: using the density change characteristic curve equations, sound speed change characteristic curve equations, and optical refractive index change characteristic curve equations as white-box models, and inputting them respectively into the black-box model based on the time-based clustering decision tree algorithm, to obtain the probability and predicted label of each cluster corresponding to each data point output by the black-box model based on the time-based clustering decision tree algorithm. The black-box model based on the time-based clustering decision tree algorithm is then optimized using the following formula to construct the density calculation decision model, sound speed calculation decision model, and optical refractive index calculation decision model:

[0130]

[0131] Where `minimize J` means minimizing the value of J. J represents the objective function value of clustering in the black-box model of the time-based clustering decision tree algorithm, used to measure the quality of clustering. α represents a coefficient used to control the compactness within clusters. β represents a coefficient used to control the temporal smoothness of clustering. t This represents the t-th data point among historical density values, historical sound velocity values, or historical optical refractive index values ​​arranged in chronological order. μ c Let d(x) represent the c-th cluster center, where c is the cluster center number and its value ranges from 1 to C, and C represents the total number of cluster centers. t μ c ) represents data point x t With cluster center μ c The distance between them; for example, this distance could be Euclidean distance. t Represents data point x t The corresponding real label. y t′ This represents the data point x output by the black-box model based on the time-based clustering decision tree algorithm. t′ The corresponding predicted label. Q(y) t y t′) represents data point x t Corresponding label y t With data point x t′ Corresponding label y t′ The similarity measure between them, for example, can use a 0-1 loss function. t and t′ both represent data point numbers and range from 1 to T, where T represents the total number of data points.

[0132] By optimizing the black-box model based on the time-clustering decision tree algorithm, the decision accuracy of the constructed density calculation decision model, sound speed calculation decision model, and optical refractive index calculation decision model can be further improved.

[0133] After constructing the density calculation decision model, sound velocity calculation decision model, and optical refractive index calculation decision model, the density detection value, sound velocity detection value, optical refractive index detection value, and their respective trend change values ​​are input into the density calculation decision model, sound velocity calculation decision model, and optical refractive index calculation decision model, respectively. Then, the determination results of the oil mixing cutting point based on density, sound velocity, and optical refractive index by the three calculation decision models can be obtained.

[0134] Step S140: According to the preset judgment result weighting scheme, determine the weight corresponding to each mixed oil cutting point judgment result, and add the weights together to obtain the corresponding weight sum value.

[0135] Specifically, the preset weighting scheme for the judgment results can be set according to actual needs, and corresponding weight values ​​can be set for the judgment results of the oil mixing cut point obtained by different calculation decision models under different oil mixing types, so as to indicate the importance of the judgment results of the oil mixing cut point obtained by different calculation decision models under different oil mixing types.

[0136] Step S150: When the weight sum exceeds the preset judgment threshold, it is determined that the oil mixing cutting time point has actually been reached, and the oil mixing cutting point identification signal is output.

[0137] Specifically, the preset weighting scheme for the judgment results and the preset judgment thresholds corresponding to different oil mixture types can be represented in Table 1 below. Here, a1 to a6 represent the weights corresponding to the oil mixture cutting point judgment results derived by the density calculation decision model for each oil mixture type based on the input density detection value and its corresponding trend change value. b1 to b6 represent the weights corresponding to the oil mixture cutting point judgment results derived by the sound velocity calculation decision model for each oil mixture type based on the input sound velocity detection value and its corresponding trend change value. c1 to c6 represent the weights corresponding to the oil mixture cutting point judgment results derived by the optical refractive index calculation decision model for each oil mixture type based on the input optical refractive index detection value and its corresponding trend change value. d1 to d6 represent the preset judgment thresholds for each oil mixture type, i.e., the threshold values ​​of the weights and values ​​corresponding to the oil mixture cutting point judgment results obtained by the three calculation decision models for different oil mixture types.

[0138] Table 1 Preset weighting scheme for judgment results

[0139]

[0140] For example, if the blending type is gasoline-diesel, and the weight sum a1+b1+c1 corresponding to the blending cut point determination results obtained by the three calculation decision models exceeds the corresponding preset determination threshold d1, then it is determined that the blending cut time point has actually been reached, and the blending cut point identification signal is output.

[0141] If the blending type is diesel-gasoline, and the weight sum a2+b2+c2 of the blending cut point determination results obtained by the three calculation decision models exceeds the corresponding preset determination threshold d2, then it is determined that the blending cut time point has actually been reached, and the blending cut point identification signal is output.

[0142] If the blending type is high-octane gasoline and low-octane gasoline, and the weight sum a3+b3+c3 of the blending cut-off point determination results obtained by the three calculation decision models exceeds the corresponding preset determination threshold d3, then it is determined that the blending cut-off time point has actually been reached, and the blending cut-off point identification signal is output.

[0143] If the blending type is low-octane gasoline and high-octane gasoline, and the weight sum a4+b4+c4 of the blending cut-off point determination results obtained by the three calculation decision models exceeds the corresponding preset determination threshold d4, then it is determined that the blending cut-off time point has actually been reached, and the blending cut-off point identification signal is output.

[0144] If the blending type is high-grade diesel and low-grade diesel, and the weight sum a5+b5+c5 of the blending cut-off point determination results obtained by the three calculation decision models exceeds the corresponding preset determination threshold d5, then it is determined that the blending cut-off time point has actually been reached, and the blending cut-off point identification signal is output.

[0145] If the blending type is low-grade diesel - high-grade diesel, and the weight sum a6+b6+c6 of the blending cut-off point determination results obtained by the three calculation decision models exceeds the corresponding preset determination threshold d6, then it is determined that the blending cut-off time point has actually been reached, and the blending cut-off point identification signal is output.

[0146] It should be noted that the mixed oil cutting point identification signal can be output to the SCADA system for prompting and alarm, so as to guide the dispatcher or the SCADA system to complete the mixed oil cutting, thereby reducing the workload of the mixed oil cutting dispatcher.

[0147] The oil mixing cut-off point identification method provided in this disclosure, compared with the prior art, obtains the oil mixing cut-off time point and real-time data from the SCADA system. Using this real-time data, it obtains the corresponding oil mixing cut-off point determination result based on density calculation decision models, sound velocity calculation decision models, and optical refractive index calculation decision models. Then, it adds the weights corresponding to each oil mixing cut-off point determination result. When the sum of the weights exceeds the corresponding preset determination threshold, it is determined that the oil mixing cut-off time point has actually arrived. This achieves accurate identification of the oil mixing cut-off time point during the sequential transportation of different types of refined oil products through pipelines. It can reduce oil mixing losses while ensuring oil quality and improve the economic efficiency of pipeline operation. Furthermore, the oil mixing cut-off point identification method provided in this disclosure can be completed automatically, with a determination cycle of less than 5 seconds, ensuring both accurate identification of the oil mixing cut-off point and real-time identification. The oil mixing cut-off point identification method provided in this disclosure can also be applied to different pipeline sequential transportation situations, with a wide range of applications and high stability. It can be integrated with the SCADA system or operate independently to guide refined oil pipelines in oil mixing cut-off, thereby reducing the workload of oil mixing cut-off dispatchers.

[0148] To enable those skilled in the art to better understand the above embodiments, a specific example is provided below.

[0149] Combined Figure 5 A method for identifying oil-mixed cutting points includes the following steps:

[0150] Select a blending potential calculation model: Select the corresponding blending potential calculation model according to the blending type. The blending type can be any one of diesel-gasoline, gasoline-diesel, high-octane gasoline-low-octane gasoline, low-octane gasoline-high-octane gasoline, high-octane diesel-low-octane diesel, or low-octane diesel-high-octane diesel.

[0151] Calculate blending potential: Based on the selected blending calculation model, calculate the blending potential volume corresponding to the blending type.

[0152] Calculate the mixing cut point: Determine the corresponding mixing cut time point based on the calculated mixing potential volume.

[0153] Select the calculation decision model, model decision weights, and thresholds corresponding to the oil mixing cutting point: select the density calculation decision model, namely the density change GB (gray box) model, the sound velocity calculation decision model, namely the sound velocity change GB (gray box) model, and the optical refractive index calculation decision model, namely the optical refractive index change GB (gray box) model, and set the preset judgment result weight ratio scheme and preset judgment thresholds.

[0154] The real-time data from the oil mixing interface detection equipment is used for decision-making through models: real-time data from the SCADA system is acquired, including density detection values, sound velocity detection values, optical refractive index detection values, and their corresponding trend change values. The density detection values, sound velocity detection values, optical refractive index detection values, and their corresponding trend change values ​​are input into the density calculation decision model, the sound velocity calculation decision model, and the optical refractive index calculation decision model, respectively, to obtain the corresponding oil mixing cut point determination results.

[0155] Total weight greater than threshold weight: Based on the preset weight allocation scheme for the judgment results, the weights corresponding to the oil mixing cutting point judgment results output by the density calculation decision model, sound velocity calculation decision model, and optical refractive index calculation decision model are determined respectively, and the weights are added together to obtain the corresponding weight sum. When the weight sum is greater than the preset judgment threshold for the corresponding oil mixing type, the oil mixing cutting time point is determined to have actually arrived, and an oil mixing cutting point identification signal is output. When the weight sum is not greater than the preset judgment threshold for the corresponding oil mixing type, the process returns to the step of making a decision based on the real-time data of the oil mixing interface detection device through the model.

[0156] In a practical application scenario, the fuel mixture is diesel-gasoline, with -20 diesel as the initial fuel and 92 gasoline as the subsequent fuel. The physical properties of the -20 diesel are shown in Table 2. In this scenario, the pipeline diameter is D323.9×11.1, the liquid flow velocity in the pipeline is approximately 0.75 m / s, the operation time for the fuel mixing valve to fully open / close is 10 seconds, the length of the fuel mixing section detected at the upstream station is 2873 m, and the volume of the fuel mixing is approximately 203 m³. 3 .

[0157] Table 2-20 Physical property parameters of diesel oil

[0158] project Quality Specifications (-20 Diesel) <![CDATA[Kinematic viscosity (20 °C) (mm 2 / s)]]> 2.74 Pour point (°C) -28 Cold filter plugging point (°C) -18 Flash point (closed cup) (°C) 66 cetane number 49 cetane number 50 <![CDATA[Density (20 °C) (kg / m 3 )]]> 814.7 Sulfur content (mg / kg) 6.3 Saturated vapor pressure / kPa 12.2 Boiling range (°C) 150-320 Melting point (°C) 1 Autoignition temperature (°C) 214 Flash point (°C) 54 Explosion limits (V%) 1.4-6.0

[0159] The flash point of automotive diesel fuel is required to be no less than 50℃. Based on the calculation model for the blending potential of diesel fuel with gasoline, and taking the mass loss system as 0.93, K is calculated. QThe value is 1.2%, with t1 being the time when the oil enters the new pure oil storage tank and t2 being the time when the mixed oil head arrives. The volumetric flow rate of the flow meter between t1 and t2 is 946 m³. 3 The blending potential is 11.35m. 3 .

[0160] The operation and scheduling time and valve action time are considered to be 60 seconds. Based on the liquid flow velocity in the pipeline, combined with the oil concentration distribution curve in the upstream station's mixing section, i.e., the oil density change characteristic curve (such as... Figure 3 As shown in the figure, the cutting point X1 of the mixed oil is 5.2%.

[0161] Select the preset weighting scheme for the diesel-gasoline blending type from the judgment result weighting scheme. Assume the weight corresponding to the density calculation decision model is 5, the weight corresponding to the optical refractive index calculation decision model is 2, and the weight corresponding to the sound velocity calculation decision model is 3. Read the sound velocity data from the density meter, optical interface detector, and ultrasonic flow meter collected by the SCADA system to obtain the corresponding density detection value, optical refractive index detection value, sound velocity detection value, and their corresponding trend change value. Input the density detection value, optical refractive index detection value, sound velocity detection value, and their corresponding trend change value into the three decision models corresponding to the blending cut point of 5.2%. Assuming the preset judgment threshold is 6, since the weight corresponding to the density calculation decision model is 5, the weight corresponding to the optical refractive index calculation decision model is 2, and the weight corresponding to the sound velocity calculation decision model is 3, the sum of the three weights (10) is greater than 6. Therefore, it is determined that the blending cut point has actually been reached, and a blending cut point identification signal is output to the SCADA system's host computer to issue a blending cut point arrival alarm.

[0162] In another practical application scenario, the fuel mixture type is low-octane gasoline - high-octane gasoline, with the preceding fuel being 92-octane gasoline and the following fuel being 95-octane gasoline, and the system receives an input fuel mixture cut-off point of 40%.

[0163] Select the preset weighting scheme for the blending type (low-octane gasoline - high-octane gasoline). Assume the weight corresponding to the density calculation decision model is 3, the weight corresponding to the optical refractive index calculation decision model is 4, and the weight corresponding to the sound velocity calculation decision model is 3. Read the sound velocity data from the density meter, optical interface detector, and ultrasonic flow meter collected by the SCADA system to obtain the corresponding density detection value, optical refractive index detection value, sound velocity detection value, and their corresponding trend change value. Input the density detection value, optical refractive index detection value, sound velocity detection value, and their corresponding trend change value into the three decision models corresponding to the blending cut point of 40%. Assuming the preset judgment threshold is 6.5, since the weight corresponding to the density calculation decision model is 3, the weight corresponding to the optical refractive index calculation decision model is 4, and the weight corresponding to the sound velocity calculation decision model is 3, the sum of the three weights (10) is greater than 6.5. Therefore, it is determined that the blending cut point has actually been reached, and the blending cut point identification signal is output to the SCADA system host computer to provide an alarm prompt that the blending cut point has been reached.

[0164] Another embodiment of this disclosure relates to a system for identifying oil-mixed cutting points, such as... Figure 6 As shown, the mixed oil cutting point identification system includes a first acquisition module 610, a second acquisition module 620, a decision module 630, a determination module 640, and a judgment module 650.

[0165] The first acquisition module 610 is used to acquire the cutting time point of the mixed oil.

[0166] The second acquisition module 620 is used to acquire real-time data from the SCADA system; the real-time data includes density detection values, sound velocity detection values, optical refractive index detection values, and their respective trend change values.

[0167] The decision module 630 is used to input the density detection value, sound velocity detection value, optical refractive index detection value and their respective trend change values ​​into the density calculation decision model, the sound velocity calculation decision model and the optical refractive index calculation decision model, respectively, to obtain the corresponding oil mixing cutting point determination result.

[0168] The determination module 640 is used to determine the weight corresponding to each mixed oil cutting point judgment result according to the preset judgment result weighting scheme, and add the weights to obtain the corresponding weight sum value.

[0169] The judgment module 650 is used to determine that the oil mixing cutting time point has actually been reached when the weight and value exceed the preset judgment threshold, and outputs the oil mixing cutting point identification signal.

[0170] The specific implementation method of the mixed oil cutting point identification system provided in this disclosure can be found in the mixed oil cutting point identification method provided in this disclosure, and will not be repeated here.

[0171] Compared to existing technologies, this disclosure provides a method for identifying mixed oil cutting points by acquiring the oil mixing cutting time point and real-time data from the SCADA system. Using this real-time data, based on density calculation decision models, sound velocity calculation decision models, and optical refractive index calculation decision models, the corresponding oil mixing cutting point determination results are obtained. Then, the weights corresponding to each oil mixing cutting point determination result are summed. When the sum of the weights exceeds the corresponding preset determination threshold, the oil mixing cutting time point is determined to have actually arrived. This achieves accurate identification of oil mixing cutting time points during the sequential transportation of different types of refined oil products through pipelines. It can reduce oil mixing losses while ensuring oil quality and improve the economic efficiency of pipeline operation. Furthermore, the oil mixing cutting point identification method provided by this disclosure can be completed automatically, with a determination cycle of less than 5 seconds, ensuring both accurate identification of oil mixing cutting points and real-time identification. The oil mixing cutting point identification method provided by this disclosure can also be applied to different pipeline sequential transportation situations, with a wide range of applications and high stability. It can be integrated with the SCADA system or operate independently to guide refined oil pipelines in oil mixing cutting, thereby reducing the workload of oil mixing cutting dispatchers.

[0172] Those skilled in the art will understand that the above embodiments are specific implementations of this disclosure, and in practical applications, various changes can be made to them in form and detail without departing from the spirit and scope of this disclosure.

Claims

1. A method for identifying oil-mixed cutting points, characterized in that, The method for identifying oil-mixed cutting points includes: Obtain the cutting time point for mixed oil; Acquire real-time data from the SCADA system; the real-time data includes density detection values, sound velocity detection values, optical refractive index detection values, and their respective trend change values. The density detection value, the sound velocity detection value, the optical refractive index detection value, and their respective trend change values ​​are input into the density calculation decision model, the sound velocity calculation decision model, and the optical refractive index calculation decision model, respectively, to obtain the corresponding oil mixing cutting point determination results. According to the preset weighting scheme for the judgment results, the weights corresponding to the judgment results of each mixed oil cutting point are determined respectively, and the weights are added together to obtain the corresponding weight sum value; When the weight and value exceed the preset judgment threshold, it is determined that the oil mixing cutting time point has actually been reached, and an oil mixing cutting point identification signal is output.

2. The method for identifying mixed oil cutting points according to claim 1, characterized in that, The process of obtaining the cutting time point of the mixed oil includes: Select the corresponding blending potential calculation model based on the blending type; Calculate the potential volume of the oil mixture corresponding to the oil mixture type based on the oil mixture potential calculation model. The oil mixing cut-off time point is determined based on the oil mixing potential volume.

3. The method for identifying oil-mixed cutting points according to claim 2, characterized in that, The blending potential calculation model includes a blending potential calculation model for gasoline mixed with diesel, a blending potential calculation model for diesel mixed with gasoline, a blending potential calculation model for two types of gasoline transported sequentially, and a blending potential calculation model for two types of diesel transported sequentially. The step of selecting the corresponding blending potential calculation model based on the blending type includes: The type of mixed oil is determined based on the oil delivery sequence information input by the user and the data parameters of the upstream station's detection instruments; When the blending type is gasoline mixed with diesel, the blending potential calculation model for gasoline mixed with diesel is selected; when the blending type is diesel mixed with gasoline, the blending potential calculation model for diesel mixed with gasoline is selected; when the blending type is two types of gasoline transported sequentially, the blending potential calculation model for the sequential transport of the two types of gasoline is selected; when the blending type is two types of diesel transported sequentially, the blending potential calculation model for the sequential transport of the two types of diesel is selected. The calculation model for the blending potential of gasoline with diesel is expressed as follows: Among them, K C This indicates the permissible concentration of diesel fuel mixed into pure gasoline; [t] a ] indicates the permissible final boiling point of gasoline as specified in the preset standard; t a Indicates the actual final boiling point of gasoline with a certain quality potential; The calculation model for the blending potential of diesel fuel with gasoline is expressed as follows: Among them, K Q This indicates the permissible concentration of gasoline blended into pure diesel fuel; [t] b ] indicates the permissible flash point of diesel fuel as specified in the preset standard; t b Indicates the actual flash point of diesel fuel with a certain quality potential; The calculation model for the blending potential of the two types of gasoline delivered sequentially is expressed as follows: Among them, K X This indicates the permissible volume percentage of a second grade of gasoline blended into a first grade of gasoline, where the first grade is greater than the second grade; N represents the minimum octane rating of the first grade of gasoline as specified in a preset standard; N1 represents the actual octane rating of the first grade of gasoline; and N2 represents the actual octane rating of the second grade of gasoline. The calculation model for the blending potential of the two types of diesel fuel delivered sequentially is expressed as follows: Among them, K s The percentage by volume of a third-grade diesel fuel blended with a fourth-grade diesel fuel is indicated, where the third-grade diesel fuel is less than the fourth-grade diesel fuel; I represents the flash point index corresponding to the permissible flash point of the third-grade diesel fuel specified in the preset standard; I1 represents the flash point index corresponding to the actual flash point of the fourth-grade diesel fuel; and I2 represents the flash point index corresponding to the actual flash point of the third-grade diesel fuel.

4. The method for identifying mixed oil cutting points according to claim 3, characterized in that, The step of calculating the potential volume of the oil mixture corresponding to the oil mixture type according to the oil mixture potential calculation model includes: When the blending type is gasoline mixed with diesel, the potential volume of the blending head is calculated according to the following formula: Q M =Q*K C ; Among them, Q M Q represents the potential volume of the blended oil; Q represents the volume of oil entering the pure oil tank. When the blending type is gasoline mixed with diesel, the potential volume of the blend at the tail end is calculated according to the following formula: Q M =Q t *K Q ; Among them, Q t Indicates the capacity of the receiving storage tank; When the blending type is diesel mixed with gasoline, the mixing potential volume at the mixing head is calculated according to the following formula: Q M =Q*K Q ; When the blending type is diesel mixed with gasoline, the potential volume of the blend at the tail end is calculated according to the following formula: Q M =Q t *K C ; When the blending type involves sequential delivery of two types of gasoline, with the first grade gasoline preceding the second grade gasoline, the blending potential volume is calculated using the following formula: Q M =Q*K X ; When the blending type involves sequential delivery of two types of gasoline, with the second grade gasoline preceding the first grade gasoline, the blending potential volume is calculated using the following formula: Q M =Q t *K X ; Among them, Q t This indicates the capacity of the receiving storage tank for the first grade of gasoline; When the blending type involves the sequential delivery of two types of diesel fuel, with the preceding fuel being the fourth grade diesel fuel and the following fuel being the third grade diesel fuel, the blending potential volume is calculated according to the following formula: Q M =Q*Ks; When the blending type involves the sequential delivery of two types of diesel fuel, with the preceding fuel being the third grade diesel fuel and the following fuel being the fourth grade diesel fuel, the blending potential volume is calculated according to the following formula: Q M =Q t *Ks; Among them, Q t This indicates the capacity of the receiving storage tank for the fourth grade of diesel oil.

5. The method for identifying mixed oil cutting points according to claim 4, characterized in that, Determining the oil mixing cut-off time point based on the oil mixing potential volume includes: Calculate the real-time volume of the blended oil in the mixed oil using the following formula: Q x =(ρ 实 -p2)*Q 测 / (ρ1-ρ2); Among them, Q x Q represents the real-time volume of the blended oil; 测 ρ represents the total volume of the mixed oil; 实 ρ1 represents the density of the mixed oil; ρ2 represents the density of the mixed oil as determined by laboratory tests; ρ3 represents the density of the pure oil as determined by laboratory tests. The real-time volume of the blended oil is integrated over time to determine the moment when the real-time volume of the blended oil equals the potential volume of the blended oil, and this moment is taken as the blended oil cutting time point.

6. The method for identifying oil-mixed cutting points according to claim 1, characterized in that, The method of obtaining the cutting time point of the mixed oil also includes: Based on pipeline operation experience, the timing for cutting off the mixed oil was determined directly.

7. The method for identifying mixed oil cutting points according to any one of claims 1 to 6, characterized in that, The density calculation decision model, the sound speed calculation decision model, and the optical refractive index calculation decision model are constructed according to the following steps: The SCADA system is used to obtain historical density data, historical sound velocity data, and historical optical refractive index data for different types of oil mixtures. Determine the density change characteristic curve equation, sound speed change characteristic curve equation, and optical refractive index change characteristic curve equation corresponding to the density history data, the sound speed history data, and the optical refractive index history data, respectively. Based on the density change characteristic curve equation, the sound speed change characteristic curve equation, and the optical refractive index change characteristic curve equation, the density calculation decision model, the sound speed calculation decision model, and the optical refractive index calculation decision model are constructed respectively using a black-box model based on the time clustering decision tree algorithm.

8. The method for identifying mixed oil cutting points according to claim 7, characterized in that, The process of determining the density change characteristic curve equation, sound speed change characteristic curve equation, and optical refractive index change characteristic curve equation corresponding to the density history data, the sound speed history data, and the optical refractive index history data, respectively, includes: Based on the historical density data, the historical sound velocity data, and the historical optical refractive index data, the historical density values, historical sound velocity values, and historical optical refractive index values ​​corresponding to the oil mixing segments of each type of oil mixing are extracted in chronological order. Based on the historical density values, historical sound velocity values, and historical optical refractive index values, the density difference, sound velocity difference, and optical refractive index difference values ​​between the next moment and the previous moment are determined using a difference algorithm. Then, the first derivatives of the historical density values, historical sound velocity values, and historical optical refractive index values ​​over a certain period of time are calculated to obtain the corresponding density change trend characteristics, sound velocity change trend characteristics, and optical refractive index change trend characteristics. Based on the density difference value and the density change trend characteristics, the sound velocity difference value and the sound velocity change trend characteristics, and the optical refractive index difference value and the optical refractive index change trend characteristics, the change characteristic curve equations corresponding to the density value, sound velocity value, and optical refractive index value in the mixing segment of each type of oil mixture are fitted respectively, so as to obtain the corresponding density change characteristic curve equation, sound velocity change characteristic curve equation, and optical refractive index change characteristic curve equation.

9. The method for identifying mixed oil cutting points according to claim 8, characterized in that, Based on the density change characteristic curve equation, the sound speed change characteristic curve equation, and the optical refractive index change characteristic curve equation, and using a black-box model based on a time-based clustering decision tree algorithm, the density calculation decision model, the sound speed calculation decision model, and the optical refractive index calculation decision model are constructed respectively, including: The density change characteristic curve equation, the sound speed change characteristic curve equation, and the optical refractive index change characteristic curve equation are used as white-box models and respectively input into the black-box model based on the time clustering decision tree algorithm to obtain the probability of each cluster and the predicted label of each data point output by the black-box model based on the time clustering decision tree algorithm. The black-box model based on the time-clustering decision tree algorithm is optimized using the following formula to construct the density calculation decision model, the sound speed calculation decision model, and the optical refractive index calculation decision model: Where, minimize J means minimizing the value of J; J represents the objective function value of clustering in the black-box model based on the time-based clustering decision tree algorithm, used to measure the quality of clustering; α represents a coefficient used to control the compactness within the clusters; β represents a coefficient used to control the temporal smoothness of the clusters; x t This represents the t-th data point among historical density values, historical sound velocity values, or historical optical refractive index values ​​arranged in chronological order; μ c Let d(x) represent the c-th cluster center, where c is the cluster center number and its value ranges from 1 to C, and C represents the total number of cluster centers; t μ c ) represents data point x t With cluster center μ c The distance between them; y t Represents data point x t Corresponding real label; y t′ The data point x represents the output of the black-box model based on the time-clustering decision tree algorithm. t′ The corresponding predicted label; Q(y) t y t′ ) represents data point x t Corresponding label y t With data point x t′ Corresponding label y t′ The similarity measure between them; t and t′ both represent data point numbers and their values ​​range from 1 to T, where T represents the total number of data points.

10. A system for identifying oil-mixed cutting points, characterized in that, The mixed oil cutting point identification system includes: The first acquisition module is used to acquire the cutting time point of the mixed oil; The second acquisition module is used to acquire real-time data from the SCADA system; the real-time data includes density detection values, sound velocity detection values, optical refractive index detection values, and their respective trend change values. The decision module is used to input the density detection value, the sound velocity detection value, the optical refractive index detection value and their respective trend change values ​​into the density calculation decision model, the sound velocity calculation decision model and the optical refractive index calculation decision model, respectively, to obtain the corresponding oil mixing cutting point determination result; The determination module is used to determine the weight corresponding to each of the mixed oil cutting points according to the preset judgment result weighting scheme, and add the weights to obtain the corresponding weight sum value; The determination module is used to determine that the oil mixing cutting time point has actually been reached when the weight and value exceed the preset determination threshold, and outputs an oil mixing cutting point identification signal.

Images