A method for judging ash accumulation in the convection section of the flue of an oilfield heating furnace

Abstract

The present invention discloses a method for determining ash accumulation in the convection section of an oilfield heating furnace flue. The method provides a method for determining ash accumulation in the convection section of an oilfield heating furnace flue. By establishing a model of the axial temperature drop mechanism of the convection section pipe and combining it with historical operating data, the heat transfer coefficient of the pipe is calculated. Based on the change in the heat transfer coefficient, the ash accumulation on the exterior of the boiler convection section pipe can be assessed, allowing for timely ash cleaning prompts to be provided. This prevents ash accumulation from seriously impacting heat conduction inside and outside the heating surface of the convection section, leading to problems such as deteriorated heat transfer conditions, increased exhaust temperature of the heating furnace, reduced thermal efficiency of the heating furnace, increased fuel consumption, and increased flow resistance in the air and smoke system.

Description

Technical Field

[0001] The invention relates to the technical field of judging ash accumulation in a convection section of a flue of a heating furnace, and in particular to a method for judging ash accumulation in a convection section of a flue of a heating furnace in an oil field. Background Art

[0002] Oilfield heater combustion produces a large number of tiny ash particles. As they flow along the flue gas toward the flue outlet, they tend to accumulate on the external heat exchange surfaces of the boiler flue's convection section due to various interactions, including surface tension, viscous forces between the particles and the furnace tube wall, molecular adhesion, electrostatic attraction, and chemical affinity. The thermal conductivity of the ash layer on the boiler's heating surface is 400 to 1000 times lower than that of the metal tube wall. Therefore, ash accumulation severely impairs heat transfer across the convection section, leading to deteriorated heat transfer, increased exhaust gas temperatures, reduced furnace thermal efficiency, increased fuel consumption, and increased flow resistance in the air and smoke system. Currently, there is no effective method for assessing ash accumulation in the convection section of oilfield heater flues to assess the extent of ash accumulation and provide prompt cleaning instructions. Therefore, it is necessary to develop a method for assessing ash accumulation in the convection section of oilfield heater flues to address these issues. Summary of the Invention

[0003] The purpose of the present invention is to address the deficiencies of the existing technology and provide a method for determining ash accumulation in the convection section of the flue of an oilfield heating furnace, which can evaluate the ash accumulation outside the convection section pipe of the boiler so as to provide timely cleaning prompts.

[0004] The present invention provides a method for determining ash accumulation in a convection section of a flue of an oilfield heating furnace, comprising:

[0005] The data acquisition module collects the real-time data of each PLC point during the operation of the oilfield heating furnace from the heating furnace distributed control system, and stores the real-time data in the database;

[0006] Determine a sliding window time through a data extraction module, extract historical stable operation data of the oilfield heating furnace from a database according to the sliding window time, and obtain an analysis sample based on the historical stable operation data;

[0007] Performing cluster analysis on the analysis sample by a cluster analysis module to divide the data distribution in the analysis sample into K clusters;

[0008] Through the mechanism analysis module, according to the prototype of the oilfield heating furnace flue convection section pipeline, the corresponding axial temperature drop mechanism analysis model is established to obtain the derivation formula of the corresponding heat transfer coefficient P in each cluster. According to the derivation formula of the heat transfer coefficient P, the lower limit value P of the heat transfer coefficient of K clusters is determined. k ;

[0009] The online reasoning module inputs the field data of each PLC point during the operation of the oil field heating furnace at regular intervals, calculates the heat transfer coefficient P based on the field data, and determines whether the heat transfer coefficient P is lower than the lower limit value P of the heat transfer coefficient corresponding to the cluster in the database. k ; If the heat transfer coefficient P is lower than the lower limit of the heat transfer coefficient P corresponding to the cluster in the database k , it is determined that there is dust formation, otherwise there is no dust formation.

[0010] Furthermore, determining the sliding window time includes:

[0011] Input the pipeline length L and pipe diameter D of the oilfield heating furnace flue convection section, and use the PLC real-time point medium flow μ and pipe diameter D to convert the medium flow rate Obtain the time it takes for the medium to travel from the inlet to the outlet of the convection section The distribution graph of time t is analyzed to determine the fixed size value of the window as the sliding window time.

[0012] Furthermore, historical stable operation data of the oilfield heating furnace is extracted from the database according to the sliding window time, and an analysis sample is obtained according to the historical stable operation data, including:

[0013] Slide the data set in the sliding window time and calculate the stability of specific points in the window. The specific points include inlet temperature, outlet temperature, medium flow rate, and flue gas temperature. The stability of the specific point in the process is judged according to the following criteria: Where yji is the i-th value of the characteristic variable j, j = 1, 2, 3, 4, representing the characteristic variables: inlet temperature, outlet temperature, medium flow rate, flue gas temperature; n is the number of historical data to determine whether the process is stable; is the average value of the jth feature variable selected ε is a pre-specified steady-state judgment threshold with a value range of (0, 0.05); when the threshold condition is met, the data in the time window segment is representative data, and the average value of each feature in the time window is calculated. is the representative value of the point, thus forming an analysis sample, which is entered into the database for subsequent analysis; the window is continuously sliding, and the above process is repeated to obtain multiple analysis samples.

[0014] Furthermore, cluster analysis is performed on the analysis samples to divide the data distribution in the analysis samples into K clusters, including:

[0015] After normalization of the analysis, a Kmeans algorithm cluster analysis is performed, and the number of clusters is determined to be K using the elbow method;

[0016] Add weight factor γ to each feature variablei′ , adjust γ i′ Value, which maximizes the silhouette coefficient of the clustering target; the silhouette coefficient of the entire data set is defined as N is the number of samples in the entire data set, s i is the silhouette coefficient of a sample i: a(i) is called the intra-cluster dissimilarity of sample i, the average distance between sample i and other samples in the same cluster: C k is the cluster to which samples i and j belong, N k Cluster C k b(i) is the number of samples in the cluster; b(i) is called the inter-cluster dissimilarity of sample i, which is the minimum distance among the average distances of this sample to all samples in all other clusters: C k is the cluster to which sample j belongs, i does not belong to cluster C k ;D ij The distance measurement formula between samples i and j is used. For each cluster, the number of samples falling into the cluster is calculated, and the weight value is established according to the number of samples in the cluster. For subsequent analysis.

[0017] Furthermore, based on the prototype of the convection section pipeline of the oilfield heating furnace flue, a corresponding axial temperature drop mechanism analysis model was established, and the derivation formula of the corresponding heat transfer coefficient P in each cluster was obtained, including:

[0018] Take a micro-section dL at a distance of Lx from the entrance of the convection section. Assume that the oil temperature at this section is T. The temperature change of the medium flowing through the dL section is dT. Therefore, the medium temperature at the Lx+dL section is T+dT. If the influence of friction resistance is ignored during the forward flow of the medium along the pipeline, the heat balance equation on the dL section is stable when the heat transfer is stable:

[0019] P·πD·dL·(T 烟 -T)=ρΦ·C·dT

[0020]

[0021]

[0022] Where: P-heat transfer coefficient, unit W / (m 2 ·℃); D, L-pipe diameter, length, unit: m; ρ, C, Φ-medium density, specific heat capacity, medium flow rate, unit: kg / m 3 、J / (kg·℃)、m 3 / s;T 烟 、T 入 、T 出 - Flue gas temperature, inlet temperature, outlet temperature, unit: °C.

[0023] Furthermore, the lower limit value P of the heat transfer coefficient of K clusters is determined according to the derivation formula of the heat transfer coefficient P. k ,include:

[0024] By deriving the heat transfer coefficient P, the cluster C is calculated. k The heat transfer coefficient P of each sample ki According to the box plot, the lower limit of the cluster heat transfer coefficient P is calculated. The calculation formula is: P = Q1-IQR, where the interquartile range IQR = Q3-Q1, Q1 and Q3 represent the lower quartile and upper quartile respectively, thus determining the cluster C k Heat transfer coefficient lower limit P k ; Determine the lower limit value P of the heat transfer coefficient of K clusters in turn k And save the relevant result information into the database for subsequent use.

[0025] Furthermore, the field data of each PLC point during the operation of the oilfield heating furnace is input at regular intervals, and the heat transfer coefficient P is calculated based on the field data, including:

[0026] During the prediction, the field data is input at regular intervals. The stability judgment condition in the data extraction module is used to determine whether the data meets the requirements of stable data. If not, the data of this period is discarded without any analysis and the judgment is waited for the next cycle. If the data is determined to be stable, the average value of the relevant characteristic variables is calculated. Then, through the cluster analysis module, the cluster k to which the data belongs is obtained. According to the formula of the mechanism analysis module, the heat transfer coefficient P of this stable data is calculated.

[0027] Furthermore, the corresponding ash ratio of the kth cluster in the past period of time in the database is statistically analyzed every day. N′ is the total number of records stored in the cluster, N′ k is the number of records marked as ash in the cluster; according to the weight β of each cluster previously k Perform weighted averaging to obtain the ash ratio r=β determined based on the weighted average of all clusters k r k If the ratio value r exceeds the threshold value α, the alarm is for dust accumulation, otherwise it is not for dust accumulation.

[0028] The present invention has the following beneficial effects: the present invention provides a method for judging ash accumulation in the convection section of the flue of an oilfield heating furnace. By establishing an axial temperature drop mechanism model of the convection section pipe, combined with historical operation data, the heat transfer coefficient of the pipe is solved. Through the change of the heat transfer coefficient, the ash accumulation on the outside of the boiler convection section pipe can be evaluated so as to give a ash cleaning prompt in time to avoid the ash accumulation of the boiler seriously affecting the heat conduction inside and outside the heating surface of the convection section, resulting in problems such as deterioration of the heat transfer condition, increase in the exhaust temperature of the heating furnace, decrease in the thermal efficiency of the heating furnace, increase in fuel consumption and increase in the flow resistance of the air and smoke system. BRIEF DESCRIPTION OF THE DRAWINGS

[0029] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0030] Figure 1 A schematic flow chart of a method for determining ash accumulation in a convection section of a flue of an oilfield heating furnace provided by the present invention;

[0031] Figure 2 Schematic diagram of the flue convection section. DETAILED DESCRIPTION

[0032] It should be noted that, unless otherwise specified, the embodiments and features in the embodiments of the present application may be combined with each other. The present invention will be described in detail below with reference to the accompanying drawings and in conjunction with the embodiments. It should be noted that the following detailed descriptions are illustrative and are intended to provide further explanation of the present application. Unless otherwise specified, all technical and scientific terms used herein have the same meaning as commonly understood by those of ordinary skill in the art to which this application belongs.

[0033] See also Figure 1 and Figure 2 The present invention provides a method for determining ash accumulation in the convection section of an oilfield heating furnace flue. The method is implemented based on a data acquisition module, a data extraction module, a cluster analysis module, a mechanism analysis module, and an online reasoning module. The method includes the following steps:

[0034] Step 1: collect real-time data of each PLC point during the operation of the oilfield heating furnace from the heating furnace distributed control system through the data acquisition module, and store the real-time data in the database.

[0035] Specifically, the real-time data of each PLC point during the operation of the heating furnace can be collected from the heating furnace distributed control system through the OPC communication system, and the relevant point data can be stored in the database.

[0036] Step 2: Determine the sliding window time through the data extraction module, extract the historical stable operation data of the oilfield heating furnace from the database according to the sliding window time, and obtain analysis samples based on the historical stable operation data.

[0037] Specifically, determining the sliding window time includes:

[0038] Input the pipeline length L and pipe diameter D of the oilfield heating furnace flue convection section, and use the PLC real-time point medium flow μ and pipe diameter D to convert the medium flow rate Obtain the time it takes for the medium to travel from the inlet to the outlet of the convection section Analyze the distribution of time t to determine a fixed window size as the sliding window time. For example, if T follows a normal distribution, use μ + 3σ (μ is the mean and σ is the standard deviation) as the minimum sliding window time standard.

[0039] Specifically, extracting historical stable operation data of the oilfield heating furnace from the database according to the sliding window time, and obtaining an analysis sample according to the historical stable operation data, including:

[0040] Slide the data set in the sliding window time and calculate the stability of specific points in the window. The specific points include inlet temperature, outlet temperature, medium flow rate, and flue gas temperature. The stability of the specific point in the process is judged according to the following criteria: Where yji is the i-th value of the characteristic variable j, j = 1, 2, 3, 4, representing the characteristic variables: inlet temperature, outlet temperature, medium flow rate, flue gas temperature; n is the number of historical data to determine whether the process is stable; is the average value of the jth feature variable selected ε is a pre-specified steady-state judgment threshold with a value range of (0, 0.05); when the threshold condition is met, the data in the time window segment is representative data, and the average value of each feature in the time window is calculated. is the representative value of the point, thus forming an analysis sample, which is entered into the database for subsequent analysis; the window is continuously sliding, and the above process is repeated to obtain multiple analysis samples.

[0041] Step 3: Perform cluster analysis on the analysis samples through a cluster analysis module, and divide the data distribution in the analysis samples into K clusters.

[0042] Specifically, cluster analysis is performed on the analysis sample to divide the data distribution in the analysis sample into K clusters, including:

[0043] After normalization of the analysis, a Kmeans algorithm cluster analysis is performed, and the number of clusters is determined to be K using the elbow method;

[0044] Add weight factor γ to each feature variable i′ , adjust γ i′ Value, which maximizes the silhouette coefficient of the clustering target; the silhouette coefficient of the entire data set is defined as N is the number of samples in the entire data set, s i is the silhouette coefficient of a sample i: a(i) is called the intra-cluster dissimilarity of sample i, the average distance between sample i and other samples in the same cluster: C k is the cluster to which samples i and j belong, N k Cluster C k b(i) is the number of samples in the cluster; b(i) is called the inter-cluster dissimilarity of sample i, which is the minimum distance among the average distances of this sample to all samples in all other clusters: C k is the cluster to which sample j belongs, i does not belong to cluster C k ;D ij is the distance measurement formula between samples i and j. You can choose the corresponding distance calculation method according to the research object. Take Euclidean distance as an example: p is the sample feature dimension, Represents the value of the i-th and j-th samples in the p-th feature dimension.

[0045] For each cluster, find the number of samples that fall into the cluster, and establish a weight value based on the number of samples in the cluster For subsequent analysis.

[0046] Step 4: Based on the prototype of the oilfield heating furnace flue convection section pipeline, the corresponding axial temperature drop mechanism analysis model is established through the mechanism analysis module to obtain the derivation formula of the corresponding heat transfer coefficient P in each cluster. According to the derivation formula of the heat transfer coefficient P, the lower limit value P of the heat transfer coefficient of K clusters is determined. k .

[0047] Specifically, based on the prototype of the convection section of the oilfield heating furnace flue, a corresponding axial temperature drop mechanism analysis model was established, and the derivation formula of the corresponding heat transfer coefficient P in each cluster was obtained, including:

[0048] See Figure 2 , take a micro-element section dL at a distance of Lx from the entrance of the convection section, and assume that the oil temperature at this section is T. The temperature change of the medium flowing through the dL section is dT, so the medium temperature at the Lx+dL section is T+dT. If the influence of friction resistance is ignored during the forward flow of the medium along the pipeline, the heat balance equation on the dL section is stable when the heat transfer is stable:

[0049] P·πD·dL·(T 烟-T)=ρΦ·C·dT

[0050]

[0051]

[0052] Where: P-heat transfer coefficient, unit W / (m 2 ·℃); D, L-pipe diameter, length, unit: m; ρ, C, Φ-medium density, specific heat capacity, medium flow rate, unit: kg / m 3 、J / (kg·℃)、m 3 / s;T 烟 、T 入 、T 出 - Flue gas temperature, inlet temperature, outlet temperature, unit: °C.

[0053] Specifically, the lower limit value P of the heat transfer coefficient of K clusters is determined according to the derivation formula of the heat transfer coefficient P. k ,include:

[0054] By deriving the heat transfer coefficient P, the cluster C is calculated. k The heat transfer coefficient P of each sample ki According to the box plot, the lower limit of the cluster heat transfer coefficient P is calculated. The calculation formula is: P = Q1-IQR, where the interquartile range IQR = Q3-Q1, Q1 and Q3 represent the lower quartile and upper quartile respectively, thus determining the cluster C k Heat transfer coefficient lower limit P k ; Determine the lower limit value P of the heat transfer coefficient of K clusters in turn k And save the relevant result information into the database for subsequent use.

[0055] Step 5: The online inference module inputs the field data of each PLC point during the operation of the oilfield heating furnace at regular intervals, and calculates the heat transfer coefficient P based on the field data; determines whether the heat transfer coefficient P is lower than the lower limit value P of the heat transfer coefficient corresponding to the cluster in the database. k ; If the heat transfer coefficient P is lower than the lower limit of the heat transfer coefficient P corresponding to the cluster in the database k , it is determined that there is dust formation, otherwise there is no dust formation.

[0056] Specifically, the field data of each PLC point during the operation of the oilfield heating furnace is input at regular intervals, and the heat transfer coefficient P is calculated based on the field data, including:

[0057] During the prediction, the field data is input at regular intervals. The stability judgment condition in the data extraction module is used to determine whether the data meets the requirements of stable data. If not, the data of this period is discarded without any analysis and the judgment is waited for the next cycle. If the data is determined to be stable, the average value of the relevant characteristic variables is calculated. Then, through the cluster analysis module, the cluster k to which the data belongs is obtained. According to the formula of the mechanism analysis module, the heat transfer coefficient P of this stable data is calculated.

[0058] Specifically, the corresponding ash ratio in the kth cluster is statistically analyzed every day in the past period of time, for example, 15 days. N′ is the total number of records stored in the cluster, N′ k is the number of records marked as ash in the cluster; according to the weight β of each cluster previously k Perform weighted averaging to obtain the ash ratio r=β determined based on the weighted average of all clusters k r k If the ratio value r exceeds the threshold value α, for example, α is set to 0.5, an alarm is given for dust accumulation, otherwise there is no dust accumulation.

[0059] As demonstrated in the preceding examples, the method provided by the present invention for determining ash buildup in the convection section of an oilfield heating furnace flue, based on historical operational data analysis, avoids the complex considerations required by purely physical heat transfer simulation models and exhibits a certain degree of robustness. The method can be integrated with a front-end system to visualize the calculated results, facilitating staff's decision-making regarding descaling timing.

[0060] Embodiments of the present invention further provide a storage medium storing a computer program that, when executed by a processor, implements some or all of the steps of each embodiment of the method for determining ash accumulation in the convection section of an oilfield heating furnace flue provided by the present invention. The storage medium may be a magnetic disk, an optical disk, a read-only memory (ROM), or a random access memory (RAM).

[0061] Those skilled in the art will clearly understand that the techniques in the embodiments of the present invention can be implemented using software plus a necessary general-purpose hardware platform. Based on this understanding, the technical solutions in the embodiments of the present invention, or the portion that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a storage medium such as ROM / RAM, a magnetic disk, or an optical disk, and includes a number of instructions for enabling a computer device (which can be a personal computer, server, or network device, etc.) to execute the methods described in various embodiments of the present invention or certain portions of the embodiments.

[0062] The foregoing description is merely a preferred embodiment of the present invention and is not intended to limit the present invention. Those skilled in the art will readily appreciate that various modifications and variations of the present invention are possible. Any modifications, equivalent substitutions, or improvements made within the spirit and principles of the present invention are intended to be within the scope of protection of the present invention.

Claims

1. A method for determining dust accumulation in the convection section of a flue of an oilfield heating furnace, characterized in that: include: The data acquisition module collects the real-time data of each PLC point during the operation of the oilfield heating furnace from the heating furnace distributed control system, and stores the real-time data in the database; Determine a sliding window time through a data extraction module, extract historical stable operation data of the oilfield heating furnace from a database according to the sliding window time, and obtain an analysis sample based on the historical stable operation data; Performing cluster analysis on the analysis sample by a cluster analysis module to divide the data distribution in the analysis sample into K clusters; Through the mechanism analysis module, according to the prototype of the oilfield heating furnace flue convection section pipeline, the corresponding axial temperature drop mechanism analysis model is established to obtain the derivation formula of the corresponding heat transfer coefficient P in each cluster. According to the derivation formula of the heat transfer coefficient P, the lower limit value P of the heat transfer coefficient of K clusters is determined. k ; The online reasoning module inputs the field data of each PLC point during the operation of the oil field heating furnace at regular intervals, and calculates the heat transfer coefficient P based on the field data; Determine whether the heat transfer coefficient P is lower than the lower limit value P of the heat transfer coefficient corresponding to the cluster in the database k ; If the heat transfer coefficient P is lower than the lower limit of the heat transfer coefficient P corresponding to the cluster in the database k , it is determined that there is dust formation, otherwise there is no dust formation.

2. The method for determining ash accumulation in the convection section of the flue of an oilfield heating furnace according to claim 1, wherein: Determine the sliding window time, including: Input the pipeline length L and pipe diameter D of the oilfield heating furnace flue convection section, and use the PLC real-time point medium flow μ and pipe diameter D to convert the medium flow rate Obtain the time it takes for the medium to travel from the inlet to the outlet of the convection section The distribution graph of time t is analyzed to determine the fixed size value of the window as the sliding window time.

3. The method for determining ash accumulation in the convection section of the flue of an oilfield heating furnace according to claim 2, wherein: Extracting historical stable operation data of the oilfield heating furnace from the database according to the sliding window time, and obtaining an analysis sample based on the historical stable operation data, including: Slide the data set in the sliding window time and calculate the stability of specific points in the window. The specific points include inlet temperature, outlet temperature, medium flow rate, and flue gas temperature. The stability of the specific point in the process is judged according to the following criteria: Where yji is the i-th value of the characteristic variable j, j = 1, 2, 3, 4, representing the characteristic variables: inlet temperature, outlet temperature, medium flow rate, flue gas temperature; n is the number of historical data to determine whether the process is stable; is the average value of the jth feature variable selected ε is a pre-specified steady-state judgment threshold with a value range of (0, 0.05); when the threshold condition is met, the data in the time window segment is representative data, and the average value of each feature in the time window is calculated. is the representative value of the point, thus forming an analysis sample, which is entered into the database for subsequent analysis; the window is continuously sliding, and the above process is repeated to obtain multiple analysis samples.

4. The method for determining ash accumulation in the convection section of the flue of an oilfield heating furnace according to claim 3, wherein: Performing cluster analysis on the analysis sample to divide the data distribution in the analysis sample into K clusters, including: After normalization of the analysis, a Kmeans algorithm cluster analysis is performed, and the number of clusters is determined to be K using the elbow method; Add weight factor γ to each feature variable i′ , adjust γ i′ Value, which maximizes the silhouette coefficient of the clustering target; the silhouette coefficient of the entire data set is defined as N is the number of samples in the entire data set, s i is the silhouette coefficient of a sample i: a(i) is called the intra-cluster dissimilarity of sample i, the average distance between sample i and other samples in the same cluster: C k is the cluster to which samples i and j belong, N k Cluster C k b(i) is the number of samples in the cluster; b(i) is called the inter-cluster dissimilarity of sample i, which is the minimum distance among the average distances of this sample to all samples in all other clusters: C k is the cluster to which sample j belongs, i does not belong to cluster C k ;D ij The distance measurement formula between samples i and j is used. For each cluster, the number of samples falling into the cluster is calculated, and the weight value is established according to the number of samples in the cluster. For subsequent analysis.

5. The method for determining ash accumulation in the convection section of the flue of an oilfield heating furnace according to claim 4, wherein: Based on the prototype of the convection section of the oilfield heating furnace flue, a corresponding axial temperature drop mechanism analysis model was established, and the derivation formula of the corresponding heat transfer coefficient P in each cluster was obtained, including: Take a micro-section dL at a distance of Lx from the entrance of the convection section. Assume that the oil temperature at this section is T. The temperature change of the medium flowing through the dL section is dT. Therefore, the medium temperature at the Lx+dL section is T+dT. If the influence of friction resistance is ignored during the forward flow of the medium along the pipeline, the heat balance equation on the dL section is stable when the heat transfer is stable: P·πD·dL·T 烟 -T)=ρΦ·C·dT Where: P-heat transfer coefficient, unit W / (m 2 ·℃); D, L-pipe diameter, length, unit: m; ρ, C, Φ-medium density, specific heat capacity, medium flow rate, unit: kg / m 3 、J / (kg·℃)、m 3 / s;T 烟 、T 入 、T 出 – Flue gas temperature, inlet temperature, outlet temperature, unit: °C.

6. The method for determining ash accumulation in the convection section of the flue of an oilfield heating furnace according to claim 5, wherein: According to the derivation formula of the heat transfer coefficient P, the lower limit value P of the heat transfer coefficient of K clusters is determined. k ,include: By deriving the heat transfer coefficient P, the cluster C is calculated. k The heat transfer coefficient P of each sample ki According to the box plot, the lower limit of the cluster heat transfer coefficient P is calculated. The calculation formula is: P = Q1-IQR, where the interquartile range IQR = Q3-Q1, Q1 and Q3 represent the lower quartile and upper quartile respectively, thus determining the cluster C k Heat transfer coefficient lower limit P k ; Determine the lower limit value P of the heat transfer coefficient of K clusters in turn k And save the relevant result information into the database for subsequent use.

7. The method for determining ash accumulation in the convection section of the flue of an oilfield heating furnace according to claim 6, wherein: The field data of each PLC point during the operation of the oilfield heating furnace is input at regular intervals, and the heat transfer coefficient P is calculated based on the field data, including: During the prediction, the field data is input at regular intervals. The stability judgment condition in the data extraction module is used to determine whether the data meets the requirements of stable data. If not, the data of this period is discarded without any analysis and the judgment is waited for the next cycle. If the data is determined to be stable, the average value of the relevant characteristic variables is calculated. Then, through the cluster analysis module, the cluster k to which the data belongs is obtained. According to the formula of the mechanism analysis module, the heat transfer coefficient P of this stable data is calculated.

8. The method for determining ash accumulation in the convection section of the flue of an oilfield heating furnace according to claim 7, wherein: In units of days, the corresponding ash ratio in the kth cluster in the past period of time in the database is statistically analyzed every day N′ is the total number of records stored in the cluster, N′ k is the number of records marked as ash in the cluster; according to the weight β of each cluster previously k Perform weighted averaging to obtain the ash ratio r=β determined based on the weighted average of all clusters k r k If the ratio value r exceeds the threshold value α, the alarm is for dust accumulation, otherwise it is not for dust accumulation.

Images