Coking plant monitoring point layout method and device, electronic equipment and storage medium

Abstract

The invention relates to the technical field of coking plant soil pollution monitoring, in particular to a coking plant monitoring point layout method and device, electronic equipment and a storage medium, and the method comprises the steps: firstly obtaining a plurality of first point data sets; then generating a dimension conversion matrix according to first volatile matter vectors extracted from the plurality of first point location data sets, and performing dimension conversion on each first volatile matter vector according to the dimension conversion matrix; constructing the plurality of obtained second volatile matter vectors into a plurality of second point location data sets according to source point locations of the second volatile matter vectors; clustering the plurality of second volatile matter vectors, and taking the center of each obtained vector class as a first reference vector; and finally, according to the plurality of first reference vectors and the plurality of second point location data sets, determining a plurality of target monitoring points from the plurality of point locations. According to the method, repeated construction of monitoring points is avoided, and reliability and stability of monitoring data are guaranteed.

Description

Technical Field

[0001] This invention relates to the field of soil pollution monitoring technology in coking plants, and particularly to a method, apparatus, electronic equipment, and storage medium for setting up monitoring points in coking plants. Background Technology

[0002] Soil pollution monitoring points in coking plants are the core "sensing nodes" for assessing, controlling, and remediating soil environmental risks in the plant area. Their role runs through the entire process of pollution prevention, current status assessment, risk control, remediation, and long-term monitoring. Specifically, a well-placed monitoring point can: accurately identify the current pollution status, trace the source of pollution, support risk assessment and control, evaluate the effectiveness of remediation, and provide long-term monitoring and early warning.

[0003] Currently, the main methods for setting up monitoring points are divided into two categories: professional judgment-based point placement and zoned point placement. The professional sampling method is used when the potential soil pollution situation at a coking plant site is relatively clear. Sampling points are usually set up outside or at the boundary of units such as coke ovens, quenching towers and waste coke yards, crude benzene areas, and ammonium sulfate areas, while control points are set up in the green area on the south side upwind.

[0004] The zoned sampling method is used for coking plant sites with uneven pollution distribution. Based on the potential pollution areas, the site is divided into production areas, dormitory areas, kiln areas, coal storage areas and their surroundings, etc. Sampling points are set up in each area, along with control points.

[0005] However, regardless of the method mentioned above, the specific location of monitoring points is usually determined based on experience, which is somewhat arbitrary. In order to thoroughly understand the soil pollution situation of coking plants, a large number of monitoring points are often set up, resulting in a high degree of duplication in the setting of monitoring points.

[0006] Therefore, it is necessary to develop and design a method for the layout of monitoring points in coking plants. Summary of the Invention

[0007] The present invention provides a method, apparatus, electronic device and storage medium for setting up monitoring points in a coking plant, which solves the problem of redundant construction of monitoring points in the prior art.

[0008] In a first aspect, embodiments of the present invention provide a method for the layout of monitoring points in a coking plant, comprising: Multiple first point datasets are obtained, where each first point dataset corresponds to a point in the target area. The first point dataset includes multiple first volatile vectors, which are constructed based on O2 concentration, CO2 concentration and multiple light non-aqueous liquid volatile concentrations. A dimension transformation matrix is ​​generated based on the first volatile vectors extracted from the plurality of first point datasets. The dimension transformation is performed on each first volatile vector based on the dimension transformation matrix. The plurality of second volatile vectors obtained are then used to construct a plurality of second point datasets based on the source points of the second volatile vectors. Cluster the plurality of second volatile vectors and use the center of each vector class as the first reference vector; Based on multiple first reference vectors and multiple second point datasets, multiple target monitoring points are determined from multiple points.

[0009] In one possible implementation, generating the dimension transformation matrix based on the first volatile vector extracted from the plurality of first location datasets includes: The multiple first volatile vectors are standardized to obtain multiple standard vectors; Based on the multiple standard vectors, multiple correlation values ​​characterizing the correlation of volatile concentrations are determined, and the multiple correlation values ​​are constructed into a correlation matrix in a preset order; Extract multiple eigenvalues ​​and multiple eigenvectors from the correlation matrix, wherein the eigenvalues ​​and eigenvectors correspond one-to-one; Select multiple principal feature values ​​from the plurality of feature values; The eigenvectors corresponding to the multiple principal eigenvalues ​​are used to construct a dimension transformation matrix.

[0010] In one possible implementation, the standardization of the plurality of first volatile vectors to obtain a plurality of standard vectors includes: For each vector dimension, vector elements are extracted sequentially from the plurality of first volatile vectors to construct a process vector; Each of the multiple process vectors is standardized according to the first formula to obtain the multiple standard vectors, wherein the first formula is:

[0011] In the formula, The first standard vector One element, The process vector of the first One element, The mean of the elements of the process vector. The standard deviation of the process vector elements; The step involves determining multiple correlation values ​​characterizing the correlation of volatile concentrations based on the multiple standard vectors, and constructing a correlation matrix from these multiple correlation values ​​in a preset order, including: Sort the multiple standard vectors to obtain a standard vector queue; Vectors are sequentially retrieved from the standard vector queue as vectors to be processed; The correlation coefficient between the vector to be processed and each vector in the standard vector queue is calculated using the second formula, and the obtained correlation coefficient is added to the correlation matrix. The second formula is:

[0012] In the formula, The first in the correlation matrix Line 1 Column elements, The first vector to be processed One element, This represents the position of the vector to be processed in the standard vector queue. For the standard vector queue The first vector One element, This represents the total number of elements in the standard vector; If the traversal of the standard vector queue is not completed, then proceed to the step of sequentially retrieving vectors from the standard vector queue as vectors to be processed. The step of performing dimensional transformation on each first volatile vector according to the dimensional transformation matrix includes: The first volatile vector is dimensionally transformed according to the third formula and the aforementioned dimensional transformation matrix, wherein the third formula is:

[0013] In the formula, This is the second volatile vector. The first volatile vector, This is the dimension transformation matrix.

[0014] In one possible implementation, the clustering of the plurality of second volatile vectors, and the use of the center of each obtained vector class as a first reference vector, includes: Obtain the reference number and cluster radius; For each second volatile vector, a second reference vector is found from the plurality of second volatile vectors, and the distance between the second volatile vector and the second reference vector is taken as the first typical distance of the second volatile vector, wherein the second reference vector is the vector whose median number among the plurality of vectors closest to the second volatile vector is the reference number; The second volatile vector whose first typical distance is smaller than the cluster radius is used as the third reference vector; Vectors are extracted iteratively from multiple third reference vectors and used as reference vectors to be processed; Among the multiple second volatile vectors that have not been clustered, a target vector is searched, wherein the target vector is a vector whose distance from the reference vector to be processed is less than the clustering radius; If a target vector exists, the target vector is added to the class of the reference vector to be processed, the target vector is used as the reference vector to be processed, and the process jumps to the multiple second volatile vectors that have not been clustered to search for the target vector. If the traversal of the plurality of third reference vectors is not completed, then proceed to the step of extracting vectors from the plurality of third reference vectors as reference vectors to be processed. The unclustered second volatile vector is marked as an outlier vector; For each class, the center of the vector class is extracted as the first reference vector.

[0015] In one possible implementation, the cluster radius is determined based on the plurality of second volatile vectors, including: Arrange multiple primary typical distances in ascending order to obtain the primary distance queue; For each typical distance in the first distance queue, calculate the difference with the preceding typical distance, where the preceding typical distance is the typical distance in the first distance queue that is adjacent to the typical distance before the typical distance. The maximum value among multiple difference values ​​is taken as the target difference value; The typical distance corresponding to the target difference value is used as the cluster radius.

[0016] In one possible implementation, determining multiple monitoring points from multiple locations based on multiple first reference vectors and the multiple second location datasets includes: Multiple anomaly vectors are obtained, wherein the anomaly vectors are obtained based on vectors that failed to be clustered after the multiple second volatile vectors are clustered; The second point dataset containing the abnormal vectors is deleted, and the remaining second point dataset is used as the first candidate point dataset. For each first candidate point dataset, select multiple closest reference vectors from the multiple first reference vectors, and determine the comprehensive distance index of the first candidate point dataset based on the multiple closest reference vectors. Select the index with the smallest value from multiple comprehensive distance indices as multiple target indices, and use the first candidate point dataset corresponding to the target index as the target point dataset; Monitoring points are set up by selecting points from the source points of the target point dataset.

[0017] In one possible implementation, for each first candidate point dataset, selecting multiple closest reference vectors from the plurality of first reference vectors, and determining the comprehensive distance index of the first candidate point dataset based on the plurality of closest reference vectors, includes: For each dataset of first candidate points, perform the following steps: For each second volatile vector in the first candidate site dataset, the vector that is closest to the first reference vector is found from the plurality of first reference vectors as the closest reference vector, and the distance to the closest reference vector is used as the reference distance. The sum of multiple reference distances is used as the comprehensive distance index.

[0018] Secondly, embodiments of the present invention provide a coking plant monitoring point deployment device for implementing the coking plant monitoring point deployment method as described in the first aspect or any possible implementation thereof, the coking plant monitoring point deployment device comprising: The monitoring data acquisition module is used to acquire multiple first point datasets. Each first point dataset corresponds to a point in the target area. The first point dataset includes multiple first volatile vectors, which are constructed based on the concentration, concentration, and concentrations of multiple light non-aqueous liquid volatiles. The data dimension transformation module is used to generate a dimension transformation matrix based on the first volatile vectors extracted from the plurality of first point datasets, perform dimension transformation on each first volatile vector based on the dimension transformation matrix, and construct a plurality of second point datasets based on the source points of the obtained multiple second volatile vectors. The reference vector acquisition module is used to cluster the plurality of second volatile vectors and take the center of each vector class as the first reference vector. as well as, The target monitoring point localization module is used to determine multiple target monitoring points from multiple points based on multiple first reference vectors and the multiple second point datasets.

[0019] Fourthly, embodiments of the present invention provide a computer-readable storage medium storing a computer program that, when executed by a processor, implements the steps of the method as described in the first aspect or any possible implementation thereof.

[0020] The beneficial effects of the embodiments of the present invention compared with the prior art are as follows: This invention discloses a method for setting up monitoring points in a coking plant. First, multiple first-point datasets are acquired, each corresponding to a point in a target area. Each first-point dataset includes multiple first volatile matter vectors, constructed based on O2 concentration, CO2 concentration, and multiple light non-aqueous liquid volatile matter concentrations. Then, a dimension transformation matrix is ​​generated based on the first volatile matter vectors extracted from the multiple first-point datasets. The dimension transformation matrix is ​​used to transform the dimension of each first volatile matter vector, and the resulting multiple second volatile matter vectors are constructed into multiple second-point datasets based on their source points. Next, the multiple second volatile matter vectors are clustered, and the center of each vector cluster is used as a first reference vector. Finally, multiple target monitoring points are determined from the multiple points based on the multiple first reference vectors and the multiple second-point datasets. The method of this invention obtains multiple reference vectors through data dimension transformation and clustering, and then selects typical datasets from multiple point datasets based on the reference vectors. The monitoring points deployed based on the typical datasets can reflect the overall coking pollution situation, avoid the duplication of monitoring point construction, and ensure the reliability and stability of monitoring data. Attached Figure Description

[0021] To more clearly illustrate the technical solutions in the embodiments of the present invention, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0022] Figure 1 This is a flowchart of the coking plant monitoring point layout method provided by the embodiments of the present invention; Figure 2 This is a schematic diagram of the point monitoring data acquisition process provided by the embodiments of the present invention; Figure 3 This is a functional block diagram of the coking plant monitoring point deployment device provided in the embodiments of the present invention; Figure 4 This is a functional block diagram of an electronic device provided in an embodiment of the present invention. Detailed Implementation

[0023] In the following description, specific details such as particular system structures and techniques are set forth for illustrative purposes and not for limitation, so as to provide a thorough understanding of embodiments of the invention. However, those skilled in the art will understand that the invention can be implemented in other embodiments without these specific details. In other instances, detailed descriptions of well-known systems, apparatuses, and methods are omitted so as not to obscure the description of the invention with unnecessary detail.

[0024] To make the objectives, technical solutions, and advantages of the present invention clearer, specific embodiments will be described below in conjunction with the accompanying drawings.

[0025] The embodiments of the present invention will be described in detail below. This example is implemented based on the technical solution of the present invention, and provides detailed implementation methods and specific operation processes. However, the protection scope of the present invention is not limited to the following embodiments.

[0026] Figure 1 A flowchart illustrating the method for setting up monitoring points in a coking plant, as provided in an embodiment of the present invention.

[0027] like Figure 1 As shown, a flowchart illustrating the implementation of the coking plant monitoring point layout method provided by the embodiments of the present invention is presented below: In step 101, multiple first point datasets are obtained, wherein each first point dataset corresponds to a point in the target area. The first point dataset includes multiple first volatile vectors, which are constructed based on O2 concentration, CO2 concentration and multiple light non-aqueous liquid volatile concentrations.

[0028] For example, this invention proposes a monitoring point deployment method based on micro-disturbance. Micro-disturbance technology refers to a survey technique that, without affecting the normal production of operating enterprises or causing safety hazards, rapidly detects more than 10 indicator indicators produced by pollutants and microbial metabolism in shallow topsoil at depths of 0.7 meters or more to determine the pollution status of soil and groundwater. This invention uses micro-disturbance technology to acquire data from multiple locations, constructs a dataset, and selects the deployment locations of monitoring points based on the indicator data.

[0029] like Figure 2 As shown, before setting up monitoring points around pollution source 201, the present invention will set up multiple points 202 around pollution source 201, obtain volatile concentration data based on these points 202, and organize these concentration data into a first point dataset.

[0030] Specifically, for each point 202, a sampling pit is formed at that point 202 by gradually deepening it. Each time it is deepened, O2 concentration, CO2 concentration, and multiple light non-aqueous liquid volatile concentration data are collected from that depth 203 using a micro-perturbation method. The concentration data at that depth 203 is used to construct a first volatile vector 204. It can be seen that multiple depths 203 will result in multiple first volatile vectors 204. These first volatile vectors 204 will be aggregated together to form the first point dataset.

[0031] Based on the first location dataset obtained through the above process, this invention determines the location of monitoring points through methods such as dimension transformation, clustering, and data filtering.

[0032] In step 102, a dimension transformation matrix is ​​generated based on the first volatile vectors extracted from the plurality of first point datasets. The dimension transformation is performed on each first volatile vector based on the dimension transformation matrix. The plurality of second volatile vectors obtained are then used to construct a plurality of second point datasets based on the source points of the second volatile vectors.

[0033] In some implementations, generating a dimension transformation matrix based on the first volatile vector extracted from the plurality of first location datasets includes: The multiple first volatile vectors are standardized to obtain multiple standard vectors; Based on the multiple standard vectors, multiple correlation values ​​characterizing the correlation of volatile concentrations are determined, and the multiple correlation values ​​are constructed into a correlation matrix in a preset order; Extract multiple eigenvalues ​​and multiple eigenvectors from the correlation matrix, wherein the eigenvalues ​​and eigenvectors correspond one-to-one; Select multiple principal feature values ​​from the plurality of feature values; The eigenvectors corresponding to the multiple principal eigenvalues ​​are used to construct a dimension transformation matrix.

[0034] In some implementations, the standardization of the plurality of first volatile vectors to obtain a plurality of standard vectors includes: For each vector dimension, vector elements are extracted sequentially from the plurality of first volatile vectors to construct a process vector; Each of the multiple process vectors is standardized according to the first formula to obtain the multiple standard vectors, wherein the first formula is:

[0035] In the formula, The first standard vector One element, The process vector of the first One element, The mean of the elements of the process vector. The standard deviation of the process vector elements; The step involves determining multiple correlation values ​​characterizing the correlation of volatile concentrations based on the multiple standard vectors, and constructing a correlation matrix from these multiple correlation values ​​in a preset order, including: Sort the multiple standard vectors to obtain a standard vector queue; Vectors are sequentially retrieved from the standard vector queue as vectors to be processed; The correlation coefficient between the vector to be processed and each vector in the standard vector queue is calculated using the second formula, and the obtained correlation coefficient is added to the correlation matrix. The second formula is:

[0036] In the formula, The first in the correlation matrix Line 1 Column elements, The first vector to be processed One element, This represents the position of the vector to be processed in the standard vector queue. For the standard vector queue The first vector One element, This represents the total number of elements in the standard vector; If the traversal of the standard vector queue is not completed, then proceed to the step of sequentially retrieving vectors from the standard vector queue as vectors to be processed. The step of performing dimensional transformation on each first volatile vector according to the dimensional transformation matrix includes: The first volatile vector is dimensionally transformed according to the third formula and the aforementioned dimensional transformation matrix, wherein the third formula is:

[0037] In the formula, This is the second volatile vector. The first volatile vector, This is the dimension transformation matrix.

[0038] For example, the first point dataset obtained by the aforementioned steps has multi-dimensional characteristics, which increases the difficulty of data processing. At the same time, the vectors in the first point dataset have the property of data redundancy, which affects the results of subsequent clustering and increases the difficulty of clustering.

[0039] This invention reduces the dimensionality of vectors in the first point dataset by using a dimensionality transformation matrix, and then reassembles the resulting vectors into a second point dataset.

[0040] In this process, the dimension transformation matrix is ​​constructed by extracting the first volatile vector from the first point dataset, constructing an association matrix, extracting eigenvalues ​​from the association matrix, and finally constructing the matrix based on the vectors corresponding to the eigenvalues.

[0041] Specifically, a standard vector is first constructed using all the first volatile matter vectors. For each dimension of the vector, a process vector is constructed from the data extracted from each first volatile matter vector. These process vectors are then standardized using the first formula to obtain the standard vector:

[0042] In the formula, The first standard vector One element, The process vector of the first One element, The mean of the elements of the process vector. represents the standard deviation of the process vector elements.

[0043] Then, multiple standard vectors are arranged in the order corresponding to the elements in the first volatile vector to form a standard vector queue. Vectors are then extracted sequentially from the standard vector queue, and the correlation coefficient is calculated using the second formula to fill the correlation matrix.

[0044] In the formula, The first in the correlation matrix Line 1 Column elements, The first vector to be processed One element, This represents the position of the vector to be processed in the standard vector queue. For the standard vector queue The first vector One element, This represents the total number of elements in the standard vector.

[0045] Once the standard vector queue has been traversed, a complete incidence matrix is ​​obtained.

[0046] Next, the eigenvalues ​​and eigenvectors of the correlation matrix are calculated.

[0047] Next, sort the feature values ​​according to their size, and select the feature values ​​with the largest values ​​as the principal feature values. Note that a ratio threshold is usually set at this time, for example, 0.92. The ratio of the sum of the selected principal feature values ​​to the sum of all feature values ​​should be greater than the above ratio threshold.

[0048] Finally, the eigenvectors corresponding to these principal eigenvalues ​​are used to construct a dimension transformation matrix.

[0049] If the first volatile vector is used as a row vector for dimensionality transformation, then the eigenvectors will be used as column vectors to construct the dimensionality transformation matrix. The first volatile vector is then transformed using the third formula to obtain the second volatile vector:

[0050] In the formula, This is the second volatile vector. The first volatile vector, This is the dimension transformation matrix.

[0051] In step 103, the plurality of second volatile vectors are clustered, and the center of each vector class is used as the first reference vector.

[0052] In some implementations, the step of clustering the plurality of second volatile vectors and using the center of each obtained vector class as a first reference vector includes: Obtain the reference number and cluster radius; For each second volatile vector, a second reference vector is found from the plurality of second volatile vectors, and the distance between the second volatile vector and the second reference vector is taken as the first typical distance of the second volatile vector, wherein the second reference vector is the vector whose median number among the plurality of vectors closest to the second volatile vector is the reference number; The second volatile vector whose first typical distance is smaller than the cluster radius is used as the third reference vector; Vectors are extracted iteratively from multiple third reference vectors and used as reference vectors to be processed; Among the multiple second volatile vectors that have not been clustered, a target vector is searched, wherein the target vector is a vector whose distance from the reference vector to be processed is less than the clustering radius; If a target vector exists, the target vector is added to the class of the reference vector to be processed, the target vector is used as the reference vector to be processed, and the process jumps to the multiple second volatile vectors that have not been clustered to search for the target vector. If the traversal of the plurality of third reference vectors is not completed, then proceed to the step of extracting vectors from the plurality of third reference vectors as reference vectors to be processed. The unclustered second volatile vector is marked as an outlier vector; For each class, the center of the vector class is extracted as the first reference vector.

[0053] In some implementations, the cluster radius is determined based on the plurality of second volatile vectors, including: Arrange multiple primary typical distances in ascending order to obtain the primary distance queue; For each typical distance in the first distance queue, calculate the difference with the preceding typical distance, where the preceding typical distance is the typical distance in the first distance queue that is adjacent to the typical distance before the typical distance. The maximum value among multiple difference values ​​is taken as the target difference value; The typical distance corresponding to the target difference value is used as the cluster radius.

[0054] For example, in clustering the second volatile vectors, the present invention first determines a first typical distance for each second volatile vector, which is the distance between the second volatile vector and the nth nearest second volatile vector, where n is the reference number.

[0055] Then, the first typical distances are arranged from smallest to largest to obtain the first distance queue. The first distance queue is then subjected to difference calculation: that is, the value of the next item in the queue is subtracted from the value of the previous item to obtain multiple difference values. The first distance corresponding to the maximum value among the multiple difference values ​​(the minuend when obtaining the difference value) is used as the cluster radius.

[0056] Next, the second volatile vector corresponding to the first canonical distance, which is smaller than the cluster radius, is used as the third reference vector.

[0057] Next, from multiple third reference vectors, a vector is randomly selected. Using this vector as the starting point, a vector with a distance smaller than the cluster radius is searched in the unclustered second volatile vectors. If a vector with a distance smaller than the cluster radius can be found, it is added to the class and used as a new starting point. The search continues to find vectors with a distance smaller than the cluster radius until no vector with a distance smaller than the cluster radius can be found.

[0058] If there are unclustered third reference vectors, then randomly select a vector from the unclustered third reference vectors and repeat the above steps until all third reference vectors are clustered.

[0059] After all the third reference vectors are clustered, some of the second volatile vectors may not be clustered, and these will be treated as outliers.

[0060] Finally, the class center is extracted from each class and used as the first reference vector.

[0061] In step 104, multiple target monitoring points are determined from multiple points based on multiple first reference vectors and the multiple second point datasets.

[0062] In some implementations, determining multiple monitoring points from multiple locations based on multiple first reference vectors and the multiple second location datasets includes: Multiple anomaly vectors are obtained, wherein the anomaly vectors are obtained based on vectors that failed to be clustered after the multiple second volatile vectors are clustered; The second point dataset containing the abnormal vectors is deleted, and the remaining second point dataset is used as the first candidate point dataset. For each first candidate point dataset, select multiple closest reference vectors from the multiple first reference vectors, and determine the comprehensive distance index of the first candidate point dataset based on the multiple closest reference vectors. Select the index with the smallest value from multiple comprehensive distance indices as multiple target indices, and use the first candidate point dataset corresponding to the target index as the target point dataset; Monitoring points are set up by selecting points from the source points of the target point dataset.

[0063] In some implementations, the step of selecting multiple closest reference vectors from the plurality of first reference vectors for each first candidate point dataset, and determining the comprehensive distance index of the first candidate point dataset based on the plurality of closest reference vectors, includes: For each dataset of first candidate points, perform the following steps: For each second volatile vector in the first candidate site dataset, the vector that is closest to the first reference vector is found from the plurality of first reference vectors as the closest reference vector, and the distance to the closest reference vector is used as the reference distance. The sum of multiple reference distances is used as the comprehensive distance index.

[0064] For example, in deploying monitoring points based on a first reference vector, the present invention first deletes the second point dataset that includes anomaly vectors.

[0065] Since each second location dataset includes multiple second volatile vectors, based on each second volatile vector, the nearest reference vector is found from multiple first reference vectors, and the distance to the nearest reference vector is calculated; this distance is used as the reference distance. In this way, each second location dataset will have multiple reference distances, and the sum of these reference distances is used as the comprehensive distance index.

[0066] The dataset of second-point locations with the smallest overall distance index is more representative. This invention selects points from multiple datasets of second-point locations with the smallest overall distance and deploys target monitoring points based on these points.

[0067] The present invention discloses a method for deploying monitoring points in a coking plant. First, multiple first-point datasets are acquired, each corresponding to a point in a target area. Each first-point dataset includes multiple first volatile matter vectors, constructed based on O2 concentration, CO2 concentration, and multiple light non-aqueous liquid volatile matter concentrations. Then, a dimension transformation matrix is ​​generated based on the first volatile matter vectors extracted from the multiple first-point datasets. Each first volatile matter vector is then dimension-transformed according to the dimension transformation matrix, and the resulting multiple second volatile matter vectors are constructed into multiple second-point datasets based on their source points. Next, the multiple second volatile matter vectors are clustered, and the center of each vector cluster is used as a first reference vector. Finally, multiple target monitoring points are determined from the multiple points based on the multiple first reference vectors and the multiple second-point datasets. The method of this invention obtains multiple reference vectors through data dimension transformation and clustering, and then selects typical datasets from multiple point datasets based on the reference vectors. The monitoring points deployed based on the typical datasets can reflect the overall coking pollution situation, avoid the duplication of monitoring point construction, and ensure the reliability and stability of monitoring data.

[0068] It should be understood that the sequence number of each step in the above embodiments does not imply the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present invention.

[0069] The following are embodiments of the apparatus of the present invention. For details not described in detail, please refer to the corresponding method embodiments described above.

[0070] Figure 3 This is a functional block diagram of the coking plant monitoring point deployment device provided in the embodiments of the present invention, with reference to... Figure 3 The coking plant monitoring point deployment device includes: a monitoring data acquisition module 301, a data dimension conversion module 302, a reference vector acquisition module 303, and a target monitoring point positioning module 304, wherein: The monitoring data acquisition module 301 is used to acquire multiple first point datasets, wherein each first point dataset corresponds to a point in the target area. The first point dataset includes multiple first volatile vectors, which are constructed based on the concentration, concentration and multiple light non-aqueous liquid volatile concentrations. The data dimension transformation module 302 is used to generate a dimension transformation matrix based on the first volatile vectors extracted from the plurality of first point datasets, perform dimension transformation on each first volatile vector based on the dimension transformation matrix, and construct a plurality of second point datasets based on the source points of the second volatile vectors. The reference vector acquisition module 303 is used to cluster the plurality of second volatile vectors and take the center of each vector class as the first reference vector. The target monitoring point localization module 304 is used to determine multiple target monitoring points from multiple points based on multiple first reference vectors and the multiple second point datasets.

[0071] Figure 4 This is a functional block diagram of the electronic device provided in an embodiment of the present invention. For example... Figure 4 As shown, the electronic device 4 in this embodiment includes a processor 400 and a memory 401, wherein the memory 401 stores a computer program 402 that can run on the processor 400. When the processor 400 executes the computer program 402, it implements the steps of the above-described methods and embodiments for the deployment of monitoring points in coking plants, for example... Figure 1 Steps 101 to 104 are shown.

[0072] For example, the computer program 402 may be divided into one or more modules / units, which are stored in the memory 401 and executed by the processor 400 to complete the present invention.

[0073] The electronic device 4 can be a desktop computer, laptop, handheld computer, cloud server, or other computing device. The electronic device 4 may include, but is not limited to, a processor 400 and a memory 401. Those skilled in the art will understand that... Figure 4 This is merely an example of electronic device 4 and does not constitute a limitation on electronic device 4. It may include more or fewer components than shown, or combine certain components, or different components. For example, electronic device 4 may also include input / output devices, network access devices, buses, etc.

[0074] The processor 400 may be a Central Processing Unit (CPU), or other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. A general-purpose processor may be a microprocessor or any conventional processor.

[0075] The memory 401 can be an internal storage unit of the electronic device 4, such as a hard disk or memory. The memory 401 can also be an external storage device of the electronic device 4, such as a plug-in hard disk, Smart Media Card (SMC), Secure Digital (SD) card, or Flash Card. Furthermore, the memory 401 can include both internal and external storage units of the electronic device 4. The memory 401 is used to store the computer program 402 and other programs and data required by the electronic device 4. The memory 401 can also be used to temporarily store data that has been output or will be output.

[0076] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the above-described division of functional units and modules is merely an example. In practical applications, the above functions can be assigned to different functional units and modules as needed, that is, the internal structure of the device can be divided into different functional units or modules to complete all or part of the functions described above. The functional units and modules in the embodiments can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit. Furthermore, the specific names of the functional units and modules are only for easy differentiation and are not intended to limit the scope of protection of this application. The specific working process of the units and modules in the above system can be referred to the corresponding process in the aforementioned method embodiments, and will not be repeated here.

[0077] In the above embodiments, the descriptions of each embodiment have their own emphasis. For parts that are not described in detail or recorded in a certain embodiment, please refer to the relevant descriptions of other embodiments.

[0078] Those skilled in the art will recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementations should not be considered beyond the scope of this invention.

[0079] In the embodiments provided by this invention, it should be understood that the disclosed devices / electronic devices and methods can be implemented in other ways. For example, the device / electronic device embodiments described above are merely illustrative. For instance, the division of modules or units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the mutual coupling or direct coupling or communication connection shown or discussed may be through some interfaces; the indirect coupling or communication connection between devices or units may be electrical, mechanical, or other forms.

[0080] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment, depending on actual needs.

[0081] Furthermore, the functional units in the various embodiments of the present invention can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit.

[0082] If the integrated module / unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, all or part of the processes in the above-described embodiments can also be implemented by a computer program instructing related hardware. The computer program can be stored in a computer-readable storage medium, and when executed by a processor, it can implement the steps of the various methods and apparatus embodiments described above. The computer program includes computer program code, which can be in the form of source code, object code, executable files, or certain intermediate forms. The computer-readable medium can include: any entity or device capable of carrying the computer program code, a recording medium, a USB flash drive, a portable hard drive, a magnetic disk, an optical disk, a computer memory, a read-only memory (ROM), a random access memory (RAM), an electrical carrier signal, a telecommunication signal, and a software distribution medium, etc.

[0083] The above-described embodiments are only used to illustrate the technical solutions of the present invention, and are not intended to limit them. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention, and should all be included within the protection scope of the present invention.

Claims

1. A method for setting up monitoring points in a coking plant, characterized in that, include: Multiple first point datasets are obtained, where each first point dataset corresponds to a point in the target area. The first point dataset includes multiple first volatile vectors, which are constructed based on O2 concentration, CO2 concentration and multiple light non-aqueous liquid volatile concentrations. A dimension transformation matrix is ​​generated based on the first volatile vectors extracted from the plurality of first point datasets. The dimension transformation is performed on each first volatile vector based on the dimension transformation matrix. The plurality of second volatile vectors obtained are then used to construct a plurality of second point datasets based on the source points of the second volatile vectors. Cluster the plurality of second volatile vectors and use the center of each vector class as the first reference vector; Based on multiple first reference vectors and multiple second point datasets, multiple target monitoring points are determined from multiple points.

2. The method for setting up monitoring points in a coking plant according to claim 1, characterized in that, The step of generating a dimension transformation matrix based on the first volatile vector extracted from the plurality of first location datasets includes: The multiple first volatile vectors are standardized to obtain multiple standard vectors; Based on the multiple standard vectors, multiple correlation values ​​characterizing the correlation of volatile concentrations are determined, and the multiple correlation values ​​are constructed into a correlation matrix in a preset order; Extract multiple eigenvalues ​​and multiple eigenvectors from the correlation matrix, wherein the eigenvalues ​​and eigenvectors correspond one-to-one; Select multiple principal feature values ​​from the plurality of feature values; The eigenvectors corresponding to the multiple principal eigenvalues ​​are used to construct a dimension transformation matrix.

3. The method for setting up monitoring points in a coking plant according to claim 2, characterized in that, The standardization of the plurality of first volatile vectors yields a plurality of standard vectors, including: For each vector dimension, vector elements are extracted sequentially from the plurality of first volatile vectors to construct a process vector; Each of the multiple process vectors is standardized according to the first formula to obtain the multiple standard vectors, wherein the first formula is: In the formula, The first standard vector One element, The process vector of the first One element, The mean of the elements of the process vector. The standard deviation of the process vector elements; The step involves determining multiple correlation values ​​characterizing the correlation of volatile concentrations based on the multiple standard vectors, and constructing a correlation matrix from these multiple correlation values ​​in a preset order, including: Sort the multiple standard vectors to obtain a standard vector queue; Vectors are sequentially retrieved from the standard vector queue as vectors to be processed; The correlation coefficient between the vector to be processed and each vector in the standard vector queue is calculated using the second formula, and the obtained correlation coefficient is added to the correlation matrix. The second formula is: In the formula, The first in the correlation matrix Line 1 Column elements, The first vector to be processed One element, This represents the position of the vector to be processed in the standard vector queue. For the standard vector queue The first vector One element, This represents the total number of elements in the standard vector; If the traversal of the standard vector queue is not completed, then proceed to the step of sequentially retrieving vectors from the standard vector queue as vectors to be processed. The step of performing dimensional transformation on each first volatile vector according to the dimensional transformation matrix includes: The first volatile vector is dimensionally transformed according to the third formula and the aforementioned dimensional transformation matrix, wherein the third formula is: In the formula, This is the second volatile vector. The first volatile vector, This is the dimension transformation matrix.

4. The method for setting up monitoring points in a coking plant according to claim 1, characterized in that, The step of clustering the plurality of second volatile vectors and using the center of each vector class as a first reference vector includes: Obtain the reference number and cluster radius; For each second volatile vector, a second reference vector is found from the plurality of second volatile vectors, and the distance between the second volatile vector and the second reference vector is taken as the first typical distance of the second volatile vector, wherein the second reference vector is the vector whose median number among the plurality of vectors closest to the second volatile vector is the reference number; The second volatile vector whose first typical distance is smaller than the cluster radius is used as the third reference vector; Vectors are extracted iteratively from multiple third reference vectors and used as reference vectors to be processed; Among the multiple second volatile vectors that have not been clustered, a target vector is searched, wherein the target vector is a vector whose distance from the reference vector to be processed is less than the clustering radius; If a target vector exists, the target vector is added to the class of the reference vector to be processed, the target vector is used as the reference vector to be processed, and the process jumps to the multiple second volatile vectors that have not been clustered to search for the target vector. If the traversal of the plurality of third reference vectors is not completed, then proceed to the step of extracting vectors from the plurality of third reference vectors as reference vectors to be processed. The unclustered second volatile vector is marked as an outlier vector; For each class, the center of the vector class is extracted as the first reference vector.

5. The method for setting up monitoring points in a coking plant according to claim 4, characterized in that, The cluster radius is determined based on the plurality of second volatile vectors, including: Arrange multiple primary typical distances in ascending order to obtain the primary distance queue; For each typical distance in the first distance queue, calculate the difference with the preceding typical distance, where the preceding typical distance is the typical distance in the first distance queue that is adjacent to the typical distance before the typical distance. The maximum value among multiple difference values ​​is taken as the target difference value; The typical distance corresponding to the target difference value is used as the cluster radius.

6. The method for setting up monitoring points in a coking plant according to any one of claims 1-5, characterized in that, The step of determining multiple monitoring points from multiple points based on multiple first reference vectors and multiple second point datasets includes: Multiple anomaly vectors are obtained, wherein the anomaly vectors are obtained based on vectors that failed to be clustered after the multiple second volatile vectors are clustered; The second point dataset containing the abnormal vectors is deleted, and the remaining second point dataset is used as the first candidate point dataset. For each first candidate point dataset, select multiple closest reference vectors from the multiple first reference vectors, and determine the comprehensive distance index of the first candidate point dataset based on the multiple closest reference vectors. Select the index with the smallest value from multiple comprehensive distance indices as multiple target indices, and use the first candidate point dataset corresponding to the target index as the target point dataset; Monitoring points are set up by selecting points from the source points of the target point dataset.

7. The method for setting up monitoring points in a coking plant according to claim 6, characterized in that, For each first candidate point dataset, selecting multiple closest reference vectors from the plurality of first reference vectors, and determining the comprehensive distance index of the first candidate point dataset based on the plurality of closest reference vectors, includes: For each dataset of first candidate points, perform the following steps: For each second volatile vector in the first candidate site dataset, the vector that is closest to the first reference vector is found from the plurality of first reference vectors as the closest reference vector, and the distance to the closest reference vector is used as the reference distance. The sum of multiple reference distances is used as the comprehensive distance index.

8. A device for deploying monitoring points in a coking plant, characterized in that, For implementing the coking plant monitoring point layout method as described in any one of claims 1-7, the coking plant monitoring point layout device includes: The monitoring data acquisition module is used to acquire multiple first point datasets. Each first point dataset corresponds to a point in the target area. The first point dataset includes multiple first volatile vectors, which are constructed based on the concentration, concentration, and concentrations of multiple light non-aqueous liquid volatiles. The data dimension transformation module is used to generate a dimension transformation matrix based on the first volatile vectors extracted from the plurality of first point datasets, perform dimension transformation on each first volatile vector based on the dimension transformation matrix, and construct a plurality of second point datasets based on the source points of the obtained multiple second volatile vectors. The reference vector acquisition module is used to cluster the plurality of second volatile vectors and take the center of each vector class as the first reference vector. as well as, The target monitoring point localization module is used to determine multiple target monitoring points from multiple points based on multiple first reference vectors and the multiple second point datasets.

9. An electronic device comprising a memory and a processor, wherein the memory stores a computer program executable on the processor, characterized in that, When the processor executes the computer program, it implements the steps of the method as described in any one of claims 1 to 7 above.

10. A computer-readable storage medium storing a computer program, characterized in that, When the computer program is executed by a processor, it implements the steps of the method as described in any one of claims 1 to 7 above.

Images