A method and system for predicting components of a mixed sample

Abstract

The present invention relates to the technical field of petrochemical production, and more specifically, to a method and system for predicting the components of a mixed sample. The method includes: collecting m samples of different varieties of the mixed sample to be tested, each sample of a variety being regarded as a sample node, and obtaining the physical and chemical properties of each sample of a variety; generating a graph data structure from the collected sample nodes; mixing the m samples of different varieties to obtain a mixed sample, and obtaining the physical and chemical properties of the mixed sample; enumerating all paths passing through a new column of nodes based on the existing node paths, and calculating the physical and chemical properties under this path according to the physical and chemical properties and proportional parameters of the sample nodes on the path; using a similarity criterion method based on distance to compare the obtained physical and chemical properties, and the path with the closest similarity is the corresponding sample type and proportion. The present invention uses a similarity judgment method based on distance to solve the problem of incomplete collection of specific properties of samples, and provides reliable support for comprehensively and quickly analyzing the physical and chemical properties of samples.

Description

Technical Field

[0001] The present invention relates to the technical field of petrochemical production, and more specifically, to a method and system for predicting the components of a mixed sample. Background Art

[0002] Mixed samples are widely used in fields such as petrochemical industry, scientific research, medical diagnosis, environmental monitoring, and food safety. A mixed sample is a sample formed by mixing samples from different sources or of different types. Such mixing can be intentional or accidental.

[0003] In the petrochemical field, various different compounds have different effects on the performance and uses of oil products. Analyzing the components of oil products can help determine their quality. In addition, the components of oil products will profoundly affect the production and processing processes of oil products. Different types of crude oils have different compositions, and different processing technologies can separate different components. The analysis of the composition of oil products can help refineries determine the optimal production and processing processes.

[0004] Since near-infrared spectroscopy can be used to analyze compounds containing hydrocarbon groups, it is widely used in analyzing the chemical components and properties of petrochemical products such as petroleum, natural gas, refinery by-products, plastics, etc., so as to help determine the quality and consistency of products. However, due to the situation that some physical and chemical properties cannot be measured in the actual analysis process, near-infrared spectroscopy analysis cannot meet the requirements of optimizing the operation of refinery units.

[0005] Therefore, there is an urgent need for a method for predicting the components of a mixed sample at present, which combines the comprehensive evaluation and rapid evaluation of the sample, and has important practical significance for ensuring the stable operation of equipment and improving the level of operation optimization. Summary of the Invention

[0006] The purpose of the present invention is to provide a rapid calculation method and system for the components of a mixed sample, so as to solve the problems in the prior art that the physical and chemical properties of the mixed sample are not comprehensively obtained and time-consuming and laborious.

[0007] To achieve the above purpose, the present invention provides a method for predicting the components of a mixed sample, including the following steps:

[0008] Step S1: Collect m varieties of samples of the mixed sample to be measured. Each variety of sample is regarded as a sample node, and obtain the physical and chemical properties of each variety of sample;

[0009] Step S2: Generate a graph data structure from the sample nodes collected in Step S1. The graph data structure is a graph data of M rows and N columns, where M is the number of sample nodes and N is an arbitrarily specified value;

[0010] Step S3: Mix m variety samples to obtain a mixed sample, and acquire the physical and chemical properties of the mixed sample;

[0011] Step S4: Enumerate all paths passing through the new column of nodes based on the existing node paths, and calculate the physical and chemical properties under this path according to the physical and chemical properties and proportion parameters of the sample nodes on the path;

[0012] Step S5: Adopt a similarity criterion method based on distance to compare the physical and chemical properties of Step S3 and Step S4, and the path with the closest physical and chemical properties is the corresponding sample type and proportion.

[0013] In one embodiment, the samples collected in Step S1 are single-variety oil products;

[0014] The physical and chemical properties of the variety samples include: density, sulfur content, carbon residue, naphtha yield, diesel yield, wax oil yield, residue oil yield.

[0015] In one embodiment, the acquisition methods of the physical and chemical properties in Step S1 include near-infrared sample rapid evaluation, nuclear magnetic sample rapid evaluation, and sample evaluation standard test methods.

[0016] In one embodiment, Step S2 further includes:

[0017] Step S21: Arrange the sample nodes collected in Step S1 into a graph data structure with M rows and N columns, where M is the number of sample nodes and N can be arbitrarily specified;

[0018] Step S22: Assign a proportion parameter p to each sample node, and the corresponding calculation formula is as follows:

[0019] p = w / u

[0020] where p is the proportion parameter of the sample nodes in each column under the current path;

[0021] u is the column number where the new column of sample nodes to be reached is located;

[0022] w is the weight coefficient.

[0023] In one embodiment, Step S4 further includes:

[0024] Step S41: Set the parameter n i as the column number where the reached sample node is located, n i+1 as the column number where the new column of sample nodes to be reached is located, initialize the parameter n i = 1, n i+1 = 2, and calculate the proportion parameter p of each sample node;

[0025] Step S42: Standardize or normalize the m variety sample data in Step S1 and the mixed sample data in Step S3 to obtain the processed sample data;

[0026] Step S43: Based on the existing node paths, enumerate all paths between the n i and n i+1 columns, and cooperate with the calculation model of the sample physical and chemical properties to calculate the physical and chemical properties under all paths.

[0027] In one embodiment, the calculation formula for standardization in Step S42 is as follows:

[0028] a′ = (a - a avg ) / σ

[0029] where a′ is the value of the sample physical and chemical properties after standardization, a is the original value of the sample physical and chemical properties, a avg is the average value of the sample physical and chemical properties in all samples, and σ is the standard deviation of the sample physical and chemical properties in all samples.

[0030] In one embodiment, the normalization process in Step S42 includes linear normalization and mean normalization;

[0031] The calculation formula for linear normalization is as follows:

[0032] a′ = (a - a min ) / (a max - a min )

[0033] where a′ is the value of the sample physical and chemical properties after linear normalization, a is the original value of the sample physical and chemical properties, a max is the maximum value of the sample physical and chemical properties in all samples, and a min is the minimum value of the sample physical and chemical properties in all samples;

[0034] The calculation formula for mean normalization is as follows:

[0035] a′ = (a - a avg ) / (a max - a min )

[0036] where a′ is the value of the sample physical and chemical properties after mean normalization, a is the original value of the sample physical and chemical properties, a avg is the average value of the sample physical and chemical properties in all samples, a max is the maximum value of the sample physical and chemical properties in all samples, and a min is the minimum value of the sample physical and chemical properties in all samples.

[0037] In one embodiment, step S5 further includes:

[0038] Step S51: Using a distance-based similarity criterion method, screen out K paths that pass through the node and are closest to the mixed sample in step S3 among the n i+1 nodes, and eliminate the remaining paths that pass through the node;

[0039] Step S52: Determine whether n i < n i+1 < N is satisfied. If it is satisfied, then n i and n i+1 are each incremented by 1, and step S43, step S51, and step S52 are sequentially executed;

[0040] If it is not satisfied, then obtain the final K*M paths, and count the oil product nodes and proportional parameters under the above paths to realize the prediction of the components of the mixed sample.

[0041] In one embodiment, the distance-based similarity criterion method further includes Euclidean distance, cosine of the angle distance, and Mahalanobis distance.

[0042] To achieve the above object, the present invention provides a prediction system for the components of a mixed sample, including:

[0043] A memory for storing instructions executable by a processor;

[0044] A processor for executing the instructions to implement the method as described in any one of the above.

[0045] To achieve the above object, the present invention provides a computer-readable medium having computer instructions stored thereon, wherein when the computer instructions are executed by a processor, the method as described in any one of the above is executed.

[0046] The prediction method and system for the components of a mixed sample provided by the present invention adopt a distance-based similarity judgment method to solve the problem of incomplete acquisition of specific properties of samples, provide reliable support for comprehensively and rapidly analyzing the physical and chemical properties of samples, and effectively promote the improvement of refinery production efficiency and the growth of economic benefits. BRIEF DESCRIPTION OF THE DRAWINGS

[0047] The above and other features, properties, and advantages of the present invention will become more apparent from the following description in conjunction with the drawings and embodiments, in which like reference numerals always represent the same features, where:

[0048] Figure 1 Discloses a flowchart of a prediction method for the components of a mixed sample according to an embodiment of the present invention;

[0049] Figure 2Disclosed is a principle block diagram of a prediction system for components of a mixed sample according to an embodiment of the present invention. Detailed implementation manners

[0050] In order to make the objectives, technical solutions and advantages of the present invention clearer and more understandable, the present invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the invention and are not used to limit the invention.

[0051] Aiming at the problem that the collection of specific properties of samples in the prior art is not comprehensive, the present invention provides a calculation method and system for quickly calculating the components of a mixed sample, which can be widely applied to the field of intelligent sample scheduling.

[0052] Figure 1 Disclosed is a flowchart of a calculation method for quickly calculating the components of a mixed sample according to an embodiment of the present invention. As Figure 1 shown, a prediction method for components of a combined sample proposed by the present invention specifically includes the following steps:

[0053] Step S1: Collect m samples of different varieties of the mixed sample to be measured. Each sample of a variety is regarded as a sample node, and the physical and chemical properties of each sample of a variety are obtained.

[0054] Step S2: Generate a graph data structure from the sample nodes collected in Step S1. The graph data structure is a graph data of M rows and N columns, where M is the number of sample nodes and N is an arbitrarily specified value.

[0055] Step S3: Mix the m samples of different varieties to obtain a mixed sample, and obtain the physical and chemical properties of the mixed sample.

[0056] Step S4: Based on the existing node paths, enumerate all paths passing through the nodes of a new column, and calculate the physical and chemical properties under this path according to the physical and chemical properties and proportional parameters of the sample nodes on the path.

[0057] Step S5: Adopt a similarity criterion method based on distance to compare the physical and chemical properties in Step S3 and Step S4. The path with the closest similarity is the corresponding sample type and proportion.

[0058] The prediction method and system for components of a mixed sample proposed by the present invention can accurately predict the internal component ratio in a very short time by using the known physical and chemical properties of the samples. Further combined with the calculation model of the physical and chemical properties of the samples, it can quickly conduct a comprehensive evaluation, significantly reduce the complexity of the sampling and experimental analysis processes of detailed sample evaluation, greatly shorten the analysis time, and help improve the efficiency of sample analysis.

[0059] These steps will be described in detail below. It should be understood that within the scope of the present invention, the above-mentioned technical features of the present invention and the technical features specifically described below (such as in the embodiments) can be combined with each other and are interrelated, so as to constitute a preferred technical solution.

[0060] Step S1: Collect m variety samples of the mixed sample to be tested. Each variety sample is regarded as a sample node, and obtain the physical and chemical properties of each variety sample.

[0061] In this embodiment, the collected samples are single-variety oil products.

[0062] The physical and chemical properties of the sample include but are not limited to density, sulfur content, residual carbon, naphtha yield, diesel yield, wax oil yield, and residue yield.

[0063] The acquisition methods of physical and chemical properties include but are not limited to near-infrared sample rapid evaluation, nuclear magnetic sample rapid evaluation, sample evaluation standard test methods, or other sample evaluation methods;

[0064] In this embodiment, for different samples collected, the corresponding sample nodes can be represented by a vector as follows:

[0065]

[0066] where a1 is the value of the first physical and chemical property of the sample, a2 is the value of the second physical and chemical property of the sample, and a n is the value of the nth physical and chemical property of the sample.

[0067] Step S2: Generate a graph data structure from the sample nodes collected in step S1. The graph data structure is a graph data of M rows and N columns, where M is the number of sample nodes and N is an arbitrarily specified value.

[0068] A graph is a non-linear data structure that represents a many-to-many relationship. The graph data structure in this embodiment is used to represent the relationship between objects. It consists of a set of nodes and edges connecting these nodes.

[0069] The starting point of this path is a certain node in the first column. When the path starts from the starting point and continuously expands to the right, the formed path is called the "current path".

[0070] In the process of expanding to the right to generate a new path, it will continuously calculate whether the oil product nodes formed under the current path are similar to the oil product to be tested. If they are similar, the path is retained and the path can continue to expand to the right; if they are not similar, the path is removed.

[0071] In this embodiment, step S2 further includes:

[0072] Step S21: Arrange the sample nodes collected in Step S1 into a graph data structure with M rows and N columns, where M is the number of sample nodes, N can be arbitrarily specified, and each sample node appears only once in each column;

[0073] Step S22: Assign a proportion parameter p to each sample node. The calculation formula for the proportion parameter p is as follows:

[0074] p = w / u (1)

[0075] where p is the proportion parameter of the sample nodes in each column under the current path;

[0076] u is the column number where the new column of sample nodes to be reached is located, that is, u ≤ N;

[0077] w is the weight coefficient of this column;

[0078] The sum of the proportion parameters of all columns should be 1.

[0079] Step S3: For the mixed sample obtained by mixing the above m kinds of variety samples, measure the physical and chemical properties of the same kind as each variety sample.

[0080] Step S4: Based on the existing node path, enumerate all paths passing through the new column of nodes, and calculate the physical and chemical properties under this path according to the physical and chemical properties and proportion parameters of the sample nodes on the path.

[0081] Generate a path on the graph and calculate the physical and chemical properties under this path.

[0082] In this embodiment, Step S4 further includes:

[0083] Step S41: Set the parameter n i as the column number where the reached sample node is located, n i+1 as the column number where the new column of sample nodes to be reached is located, initialize the parameter n i = 1, n i+1 = 2, and calculate the proportion parameter p of each sample node;

[0084] Step S42: Perform standardization or normalization processing on the m variety sample data in Step S1 and the mixed sample data in Step S3 to eliminate the magnitude gap or dimension gap between different physical and chemical properties of the samples, and obtain the processed sample data.

[0085] To obtain better calculation results, the following are several standardization or normalization formulas applicable to this method.

[0086] The calculation formula for standardization is as follows:

[0087] a′ = (a - a avg) / σ (2)

[0088] Among them, a' is the value of the physical and chemical properties of the sample after standardization, a is the original value of the physical and chemical properties of the sample, a avg is the average value of the physical and chemical properties of this sample among all samples, and σ is the standard deviation of the physical and chemical properties of this sample among all samples.

[0089] The normalization process includes but is not limited to linear normalization and mean normalization.

[0090] The calculation formula for the linear normalization is as follows:

[0091] a' = (a - a min ) / (a max - a min ) (3)

[0092] Among them, a' is the value of the physical and chemical properties of the sample after linear normalization, a is the original value of the physical and chemical properties of the sample, a max is the maximum value of the physical and chemical properties of this sample among all samples, a min is the minimum value of the physical and chemical properties of this sample among all samples.

[0093] The calculation formula for the mean normalization is as follows:

[0094] a' = (a - a avg ) / (a max - a min ) (4)

[0095] Among them, a' is the value of the physical and chemical properties of the sample after mean normalization, a is the original value of the physical and chemical properties of the sample, a avg is the average value of the physical and chemical properties of this sample among all samples, a max is the maximum value of the physical and chemical properties of this sample among all samples, a min is the minimum value of the physical and chemical properties of this sample among all samples.

[0096] Step S43: On the basis of the existing path, enumerate all paths between the two columns of n i and n i+1 . Combining with the calculation model of the physical and chemical properties of the sample, the physical and chemical properties under all paths can be calculated.

[0097] Step S5: Adopt the similarity criterion method based on distance to compare the physical and chemical properties in Step S3 and Step S4. The path with the closest physical and chemical properties is the corresponding sample type and proportion.

[0098] Compare the physical and chemical properties obtained in Step S3 and Step S4. The path with the closest physical and chemical properties is the corresponding sample type and proportion.

[0099] The distance - based similarity criterion method is adopted to select and calculate the similarity degree. The distance - based similarity criterion method includes, but is not limited to, Euclidean distance, cosine - angle distance, Mahalanobis distance, etc.

[0100] The Euclidean distance, also known as the Euclid distance, is used to find the distance between two points in an N - dimensional space. This distance is definitely a number greater than or equal to zero. To calculate this distance, the coordinates of the two points in their respective dimensions are subtracted, squared, summed, and then square - rooted.

[0101] The cosine - angle distance, also called cosine similarity, uses the cosine value of the angle between two vectors in a vector space as a measure of the difference between two individuals. If the directions of two vectors are the same, that is, the included angle is close to zero, then the two vectors are more similar.

[0102] The Mahalanobis distance is used to represent the covariance distance of data. Different from the Euclidean distance, it takes into account the relationships between various characteristics and is scale - invariant, that is, independent of the measurement scale.

[0103] In this embodiment, in step S5, it further includes:

[0104] Step S51: Adopt the distance - based similarity criterion method to screen out K paths that pass through the node and are closest to the mixed sample in step S3 among the n i+1 nodes, and eliminate the remaining paths passing through this node, where K ≤ M;

[0105] Step S52: Judge whether it satisfies n i <n i+1 <N. If it is satisfied, then increment n i and n i+1 , add 1 to each respectively and sequentially execute step S43, step S52, and step S53;

[0106] If it is not satisfied, then the final K * M paths can be obtained, and the oil product nodes and proportional parameters under these paths are statistically analyzed, that is, the prediction of the composition of the mixed sample is realized.

[0107] Although the above - mentioned methods are illustrated and described as a series of actions to simplify the explanation, it should be understood and appreciated that these methods are not limited by the order of the actions. Because according to one or more embodiments, some actions may occur in a different order and / or occur concurrently with other actions that are illustrated and described in this document or not illustrated and described in this document but can be understood by those skilled in the art.

[0108] The present invention proposes a method and system for predicting the components of a mixed sample, including but not limited to Euclidean distance, cosine distance and Mahalanobis distance as distance calculation methods, which solves the problem that the prior art is not comprehensive in obtaining specific physical and chemical properties of samples and provides a method for rapid detection of specific physical and chemical properties of samples.

[0109] The following describes in detail a method for predicting the components of a mixed sample proposed in the present invention, taking the collected samples and the corresponding physical and chemical property data as experimental objects.

[0110] Step S1, collecting a batch of 17 crude oil samples and obtaining several physical and chemical properties thereof.

[0111] In this embodiment, the physical and chemical properties of the sample include density, sulfur content, carbon residue, naphtha yield, diesel yield, wax oil yield, and residual oil yield.

[0112] Step S2: Arrange the 17 sample nodes collected in step S1 into a graph data structure with 17 rows and 100 columns. Each sample node appears once and only once in each column. Set the weight coefficient w to 1 and calculate the proportion parameter p for each sample node.

[0113] Step S3: For the sample obtained by mixing the above 17 samples, the same physical and chemical properties are measured.

[0114] Step S4: Initialize parameter n i =1,n i+1 =2, the calculated initialization scale parameter p=0.5;

[0115] Performing linear normalization processing on the 17 sample data in step S1 and the mixed sample data in step S3;

[0116] Enumerate n based on existing paths i and n i+1 All paths between the two columns, combined with the physicochemical property values and proportional parameters of each sample node, are substituted into the calculation model of the corresponding physicochemical properties to calculate the physicochemical properties under this path.

[0117] Step S5: Select Euclidean distance as the method for determining the similarity. i+1 In each node, select one, five or ten paths that pass through the node and are closest to the mixed sample in step S3, and remove the remaining paths that pass through the node, and then determine whether n is satisfied. i <n i+1 <100;

[0118] If satisfied, the loop will execute n i 、n i+1 Increment by 1, enumerate n based on the existing pathi and n i+1 All paths between two columns are calculated, the closest path is computed, and the remaining paths are eliminated until n i = 99, n i+1 = 100. Finally, 17 or 85 or 170 paths can be obtained. Then, the optimal path is selected from the above paths, the oil product nodes under this path are counted, and combined with the proportional parameter p = 0.1 at this time, and the calculation models of the physical and chemical properties of the oil products, including linear calculation and non-linear calculation, the calculation of the components of the mixed oil products is realized.

[0119] Based on the above method, the prediction results of the components of the mixed oil products when K = 1, K = 5, and K = 10 are calculated.

[0120] Table 1. Reference data of the components of the mixed oil products

[0121]

[0122] The prediction results of the main components of the mixed oil products when K = 1 are shown in Table 2.

[0123] Table 2. Prediction results of the main components of the mixed oil products when K = 1

[0124]

[0125] The prediction results of the main components of the mixed oil products when K = 5 are shown in Table 3.

[0126] Table 3. Prediction results of the main components of the mixed oil products when K = 5

[0127]

[0128]

[0129] The prediction results of the main components of the mixed oil products when K = 10 are shown in Table 4.

[0130] Table 4. Prediction results of the main components of the mixed oil products when K = 10

[0131] Sample Serial Number Predicted Component Predicted Ratio 1 WTI: Oman: Forties: Kuwait 0.42：0.26：0.19：0.04 2 Medium in Sand: Kuwait: Basrah: Saudi Ultra Light 0.63：0.18：0.05：0.04 3 Cold Lake: Napo: Kuwait: Bahrain Medium in Sand 0.27：0.21：0.11：0.1 4 Kufji: Light in Sand: El Shaheen: Kuwait 0.22：0.14：0.14：0.11 5 Hout: Kufji: El Shaheen: Bahrain Medium in Sand 0.27：0.21：0.18：0.18 6 Napo: Oman: Light in Sand: Kuwait 0.69：0.22：0.06：0.02 7 Hout: Forties: Kufji: Kuwait 0.69：0.22：0.06：0.02 8 Basrah Heavy: South Ghawar Oil: Oman: Kuwait 0.28：0.28：0.16：0.11 9 Basrah Heavy: El Shaheen: Light in Sand: Medium in Sand 0.63：0.12：0.11：0.09 10 Qatar Land: Saudi Ultra Light: Kuwait: Medium in Sand 0.33：0.2：0.15：0.14

[0132] As can be seen from Table 1, Table 2, Table 3, and Table 4, when calculating the components of the mixed sample using the graph data structure, results similar to the reference data can be obtained. As K increases, the prediction results are more accurate, verifying that this method can be used to predict the components and their proportions of the mixed sample.

[0133] Those skilled in the art should be aware that an increase in K will directly affect the calculation process and results. The larger K is, the higher the accuracy of the calculation results, but the longer the time-consuming. When those skilled in the art use the present invention to calculate the components of a mixed sample, they can select an appropriate K value according to the actual situation, so as to balance the accuracy and rapidity of the prediction results.

[0134] Figure 2 The principle block diagram of a prediction system for the components of a mixed sample according to an embodiment of the present invention is disclosed. The prediction system for the components of a mixed sample may include an internal communication bus 201, a processor 202, a read-only memory (ROM) 203, a random access memory (RAM) 204, a communication port 205, and a hard disk 207. The internal communication bus 201 can enable data communication between the components of the prediction system for the components of a mixed sample. The processor 202 can make judgments and issue prompts. In some embodiments, the processor 202 may be composed of one or more processors.

[0135] The communication port 205 can enable data transmission and communication between the prediction system for the components of a mixed sample and external input / output devices. In some embodiments, the prediction system for the components of a mixed sample can send and receive information and data from a network through the communication port 205. In some embodiments, the prediction system for the components of a mixed sample can perform data transmission and communication with external input / output devices in a wired form through the input / output terminal 206.

[0136] The prediction system for the components of a mixed sample may further include program storage units and data storage units in different forms, such as a hard disk 207, a read-only memory (ROM) 203, and a random access memory (RAM) 204, which can store various data files used for computer processing and / or communication, as well as possible program instructions executed by the processor 202. The processor 202 executes these instructions to implement the main part of the method. The results processed by the processor 202 are transmitted to an external output device through the communication port 205 and displayed on the user interface of the output device.

[0137] For example, the implementation process file of the above prediction method for the components of a mixed sample can be a computer program, stored in the hard disk 207 and can be recorded in the processor 202 for execution to implement the method of the present application.

[0138] When the implementation process document of the prediction method for the components of a mixed sample is a computer program, it can also be stored in a computer-readable storage medium as an article of manufacture. For example, computer-readable storage media can include, but are not limited to, magnetic storage devices (e.g., hard disks, floppy disks, magnetic strips), optical disks (e.g., compact discs (CDs), digital versatile discs (DVDs)), smart cards, and flash memory devices (e.g., electrically erasable programmable read-only memories (EPROMs), cards, sticks, key drives). In addition, the various storage media described herein can represent one or more devices and / or other machine-readable media for storing information. The term "machine-readable medium" can include, but is not limited to, wireless channels and various other media (and / or storage media) that can store, contain, and / or carry code and / or instructions and / or data.

[0139] Compared with the prior art, the present invention provides a prediction method and system for the components of a mixed sample, and specifically has the following beneficial effects:

[0140] 1) It is not necessary to speculate on the physical and chemical properties of the sample based on expert experience, which reduces the difficulty of technical application and provides a method for rapid detection of the physical and chemical properties of the sample;

[0141] 2) It has multiple similarity criterion methods, which ensures the accuracy of the calculation results;

[0142] 3) It is not restricted by the units of the physical and chemical properties of the sample, which improves the universality of the calculation method.

[0143] Although the above methods are illustrated and described as a series of actions for simplicity of explanation, it should be understood and appreciated that these methods are not limited by the order of the actions, because according to one or more embodiments, some actions may occur in a different order and / or concurrently with other actions that are illustrated and described herein or that are not illustrated and described herein but are understandable to those skilled in the art.

[0144] As shown in this application and the claims, unless the context clearly indicates otherwise, words such as "a", "an", "one", and / or "the" are not specifically singular and may also include the plural. Generally speaking, the terms "comprising" and "including" only indicate the inclusion of the steps and elements that have been clearly identified, and these steps and elements do not constitute an exclusive list, and the method or device may also include other steps or elements.

[0145] Those skilled in the art will understand that information, signals, and data can be represented using any of a variety of different technologies and techniques. For example, the data, instructions, commands, information, signals, bits, symbols, and chips described throughout the above description can be represented by voltages, currents, electromagnetic waves, magnetic fields or magnetic particles, optical fields or optical particles, or any combination thereof.

[0146] Those skilled in the art will further appreciate that the various illustrative logical blocks, modules, circuits, and algorithm steps described in connection with the embodiments disclosed herein can be implemented as electronic hardware, computer software, or a combination of both. To clearly illustrate this interchangeability of hardware and software, the various illustrative components, blocks, modules, circuits, and steps are described above in terms of their functionality. Whether such functionality is implemented as hardware or software depends upon the particular application and design constraints imposed on the overall system. Skilled artisans may implement the described functionality in varying ways for each particular application, but such implementation decisions should not be interpreted as causing a departure from the scope of the present invention.

[0147] The various illustrative logical modules and circuits described in connection with the embodiments disclosed herein can be implemented using a general purpose processor, a digital signal processor (DSP), an application specific integrated circuit (ASIC), a field programmable gate array (FPGA) or other programmable logic device, discrete gate or transistor logic, discrete hardware components, or any combination thereof designed to perform the functions described herein. A general purpose processor may be a microprocessor, but in the alternative, the processor may be any conventional processor, controller, microcontroller, or state machine. The processor may also be implemented as a combination of computing devices, such as a combination of a DSP and a microprocessor, multiple microprocessors, one or more microprocessors cooperating with a DSP core, or any other such configuration.

[0148] The steps of a method or algorithm described in connection with the embodiments disclosed herein can be embodied directly in hardware, in a software module executed by a processor, or in a combination of both. The software module may reside in RAM memory, flash memory, ROM memory, EPROM memory, EEPROM memory, registers, a hard disk, a removable disk, a CD-ROM, or any other form of storage medium known in the art. An exemplary storage medium is coupled to the processor such that the processor can read from, and write to, the storage medium. In the alternative, the storage medium may be integral to the processor. The processor and the storage medium may reside in an ASIC. The ASIC may reside in a user terminal. In the alternative, the processor and the storage medium may reside as discrete components in a user terminal.

[0149] The above embodiments are provided to those skilled in the art to implement or use the present invention. Those skilled in the art can make various modifications or variations to the above embodiments without departing from the inventive concept of the present invention. Therefore, the protection scope of the present invention is not limited by the above embodiments, but should be the maximum scope that conforms to the innovative features mentioned in the claims.

Claims

1. A method for predicting components of a mixed sample, characterized in that, It includes the following steps: Step S1: Collect m variety samples of the mixed sample to be measured. Each variety sample is regarded as a sample node, and obtain the physical and chemical properties of each variety sample; Step S2: Generate a graph data structure from the sample nodes collected in Step S1. The graph data structure is a graph data of M rows and N columns, where M is the number of sample nodes and N is an arbitrarily specified value; Step S3: Mix the m variety samples to obtain a mixed sample, and obtain the physical and chemical properties of the mixed sample; Step S4: Based on the existing node path, enumerate all paths passing through the new column of nodes, and calculate the physical and chemical properties under this path according to the physical and chemical properties and proportion parameters of the sample nodes on the path; Step S5: Adopt a similarity criterion method based on distance to compare the physical and chemical properties of Step S3 and Step S4. The closest path is the corresponding variety type and proportion; Among them, Step S2 further includes: Step S21: Arrange the sample nodes collected in Step S1 into a graph data structure of M rows and N columns, where M is the number of sample nodes and N can be arbitrarily specified; Step S22: Assign a proportion parameter p to each sample node. The corresponding calculation formula is as follows: p = w / u where p is the proportion parameter of the sample nodes in each column under the current path; u is the column number where the new column of sample nodes to be reached is located; w is the weight coefficient; Step S4 further includes: Step S41: Set parameter n i is the column number where the arrived sample node is located, and n i+1 is the column number where the upcoming new column of sample nodes is located. Initialize parameter n i = 1, and n i+1 = 2, and calculate the proportion parameter p of each sample node; Step S42: Perform standardization processing or normalization processing on the data of the m variety samples in Step S1 and the data of the mixed sample in Step S3 to obtain the processed sample data; Step S43. Enumerate n based on the existing node paths i and n i+1 All paths between the two columns, combined with the calculation model of the physicochemical properties of the sample, to calculate the physicochemical properties under all paths; Step S5 further includes: Step S51: Using the distance-based similarity criterion method, among the n i+1 nodes, screen out K paths that pass through the node and are closest to the mixed sample in step S3, and eliminate the remaining paths that pass through the node; Step S52: Determine whether n i <n i+1 <N is satisfied. If so, increment n i and n i+1 by 1 respectively, and sequentially execute Step S43, Step S51, and Step S52; If not satisfied, obtain the final K*M paths, and count the oil product nodes and proportion parameters under the above paths to realize the prediction of the components of the mixed sample.

2. The prediction method for the composition part of the mixed sample according to claim 1, wherein The samples collected in Step S1 are single-variety oil products; The physical and chemical properties of the variety samples include: density, sulfur content, carbon residue, naphtha yield, diesel yield, wax oil yield, residue oil yield.

3. The prediction method for the composition part of the mixed sample according to claim 1, wherein The acquisition methods of the physical and chemical properties in Step S1 include near-infrared sample rapid evaluation, nuclear magnetic sample rapid evaluation, and sample evaluation standard test methods.

4. The prediction method for the composition part of the mixed sample according to claim 1, characterized in that The calculation formula for standardization in Step S42 is as follows: a′ = (a - a avg ) / σ Among them, a' is the value of the physical and chemical properties of the standardized sample, a is the value of the physical and chemical properties of the original sample, and a avg is the average value of the physical and chemical properties of this sample among all samples, and σ is the standard deviation of the physical and chemical properties of this sample among all samples.

5. The prediction method for the composition part of the mixed sample according to claim 1, characterized in that, The normalization processing in Step S42 includes linear normalization and mean normalization; The calculation formula for linear normalization is as follows: a′=(a - a min ) / (a max - a min ) Among them, a' is the value of the physical and chemical properties of the sample after linear normalization, a is the original value of the physical and chemical properties of the sample, and a max is the maximum value of the physical and chemical properties of this sample among all samples, and a min is the minimum value of the physical and chemical properties of this sample among all samples; The calculation formula for mean normalization is as follows: a′=(a - a avg ) / (a max - a min ) Among them, a' is the value of the physical and chemical properties of the sample after mean normalization, a is the original value of the physical and chemical properties of the sample, and a avg is the average value of the physical and chemical properties of this sample among all samples, and a max is the maximum value of the physical and chemical properties of this sample among all samples, and a min is the minimum value of the physical and chemical properties of this sample among all samples.

6. The prediction method for the composition part of the mixed sample according to claim 1, characterized in that, The similarity criterion method based on distance further includes Euclidean distance, cosine of the angle distance, and Mahalanobis distance.

7. A prediction system for the components of a mixed sample, including: A memory for storing instructions executable by a processor; A processor for executing the instructions to implement the method according to any one of claims 1-6.

8. A computer-readable medium having computer instructions stored thereon, wherein when the computer instructions are executed by a processor, the method according to any one of claims 1-6 is executed.

Images