Oil field steam injection boiler steam quality diagnosis method
By establishing a steam quality diagnostic model through big data analysis and orthogonal experimental design, the problems of low timeliness and poor accuracy in steam quality diagnosis of steam injection boilers have been solved, and efficient and accurate online monitoring and automatic statistics of steam quality have been achieved.
Patent Information
- Application Number
- CN202411135221.X
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-08-19
- Publication Date
- 2026-03-03
AI Technical Summary
Steam quality diagnosis for steam injection boilers suffers from low timeliness, unclear algorithms, and poor accuracy, making it impossible to achieve online continuous monitoring and process control of steam quality.
A boiler steam quality diagnostic model was established using a multiple linear regression algorithm based on big data analysis and an orthogonal analysis experimental method. Through data acquisition and preprocessing, core parameters were determined, steam quality management rules were established, and automatic detection and quality process control of steam quality were realized.
It improves the accuracy and efficiency of steam quality diagnosis, realizes online continuous monitoring and automatic statistics of steam quality, and has a diagnostic efficiency of 350 times that of traditional methods, with an accuracy rate of over 99%.
Smart Images
Figure CN121597966A_ABST
Abstract
Description
Technical Field
[0001] This invention belongs to the field of big data analysis and application of steam injection boilers, and specifically relates to a method for diagnosing the steam quality of oilfield steam injection boilers. Background Technology
[0002] Heavy oil resources have enormous exploitation potential, but their high viscosity and density make extraction difficult. Steam injection thermal extraction technology can reduce the viscosity of heavy oil and improve its mobility ratio, making it one of the effective ways to improve oil displacement efficiency.
[0003] The high-temperature, high-pressure steam used in steam injection thermal extraction of heavy oil is generated by a boiler. The steam generated by the boiler is affected by multiple factors such as the content of various components in the steam, temperature, pressure, and superheat. Inevitably, there will be situations where the quality does not meet the requirements. In order to ensure that the injected steam indicators meet the production requirements, the boiler parameters are manually inspected and the core operating data is analyzed based on experience to judge the quality of the boiler steam, or the steam quality is tested using precision instruments. This method has problems such as low timeliness, unclear algorithms, and poor accuracy. It is also difficult to conduct periodic analysis of data change trends and benchmarking. Summary of the Invention
[0004] This invention aims to address the problems of low timeliness, unclear algorithms, and poor accuracy in steam quality diagnosis of steam injection boilers. It provides a steam quality diagnosis method for oilfield steam injection boilers, based on a multiple linear regression algorithm using big data analysis and combined with orthogonal analysis experiments, to establish a boiler steam quality diagnosis model and a pass rate judgment model. The method has good diagnostic accuracy and can realize online continuous monitoring of steam quality, automatic statistics of pass rates, and quality process control, significantly improving the efficiency of steam quality diagnosis and analysis for steam injection boilers.
[0005] To achieve the above technical objectives, the present invention adopts the following technical solution:
[0006] A method for diagnosing the steam quality of an oilfield steam injection boiler, the method comprising the following steps:
[0007] Step S1: Data acquisition and preprocessing;
[0008] Step S2: Apply the multiple linear regression algorithm to perform data correlation analysis and determine the core parameters for steam quality diagnosis;
[0009] Step S3: Apply orthogonal analysis experiments to determine the optimal operating index values of the core parameters for steam quality diagnosis, establish boiler steam injection quality management rules, and obtain the boiler steam quality diagnosis model.
[0010] Step S4: Based on the boiler steam quality diagnosis model, calculate the duration of boiler operation with substandard steam quality on the same day, establish a boiler steam quality pass rate judgment model through the database, automatically diagnose whether the boiler steam quality is qualified, and calculate the boiler steam quality pass rate.
[0011] Further, step S1 specifically includes:
[0012] Step S101: Extract boiler production and operation data by parsing the boiler PLC controller communication protocol;
[0013] Step S102: For the extracted production operation data, data standardization and normalization algorithms are used to eliminate data dimensions, and normal distribution algorithms are used for classification and identification to remove unqualified data.
[0014] Furthermore, step S102 specifically includes:
[0015] By applying data quality inspection technology, the extracted production operation data is cleaned, scaled, and standardized to transform it into a standard normal distribution with a mean of 0 and a standard deviation of 1. Outliers, missing values, and duplicate values are then identified, filtered, and removed.
[0016] Further, step S2 specifically includes:
[0017] Step S201: Based on the multiple linear regression algorithm, extract the correlation parameters related to steam quality judgment, and select parameters with a correlation coefficient > 0.4 as parameters for steam quality diagnosis;
[0018] Step S202: Among the parameters used for steam quality diagnosis, determine the parameters that are among the top N items in terms of correlation with steam quality as the core parameters for steam quality diagnosis.
[0019] Furthermore, step S202 specifically includes:
[0020] Among the parameters used for diagnosing the steam quality of superheated boilers, the parameters that are most correlated with the superheated steam quality are identified as the core parameters for superheated steam quality diagnosis. These parameters include: superheat, separator level, and flame signal.
[0021] And / or,
[0022] Among the parameters used for steam quality diagnosis of wet steam boilers, the parameters that are most relevant to wet steam quality are identified as the core parameters for wet steam quality diagnosis. These parameters include: flame signal, natural gas flow rate, pump feedwater flow rate, feedwater temperature, dryness fraction, and steam pressure.
[0023] Furthermore, step S3 specifically includes:
[0024] Step S301: Apply orthogonal analysis experiments to determine the optimal operating index values of the core parameters for steam quality diagnosis of superheated boilers and / or wet steam boilers;
[0025] Step S302: Based on the optimal operating index values of the core parameters used for superheated steam quality diagnosis, establish the steam injection quality management rules for the superheated boiler, and the superheated steam quality diagnosis model of the superheated boiler can be obtained.
[0026] And / or,
[0027] Based on the optimal operating index values of the core parameters used for wet steam quality diagnosis, steam injection quality management rules for wet steam boilers can be established, thus obtaining the wet steam quality diagnosis model for wet steam boilers.
[0028] Furthermore, the steam injection quality management rules of the superheated steam quality diagnostic model are as follows:
[0029] Extract the operating data of the superheated boiler, and make the following judgments for the superheated boiler that is in operation with flame signal = 1:
[0030] When 1℃≤superheat≤30℃, the current superheated boiler is judged to be operating in superheated mode, and the quality of the superheated steam produced meets the standards.
[0031] When the superheat is <1℃, the superheat is empty, the superheat is 0℃, or the superheat is >30℃, the current superheat boiler model is identified, and further judgment is made based on the separator liquid level:
[0032] If the separator liquid level is higher than the set liquid level corresponding to the current superheated boiler model, then the current superheated boiler is determined to be in saturation operation.
[0033] If the total saturated operating time of the superheated boiler in the 24 hours from the previous day to the present time is greater than 3 hours, then the quality of the superheated steam produced by the current superheated boiler is judged to be substandard.
[0034] Wherein: the saturated operating time of the superheated boiler in the 24 hours from the previous day to the present time does not include the superheating time after boiler start-up; the definition of superheating time is:
[0035] After ignition, the boiler needs to run for a period of time to reach the superheated operating state; this period is called the superheating time.
[0036] Furthermore, the steam injection quality management rules of the wet steam quality diagnostic model are as follows:
[0037] Extract the operating data of the wet steam boiler and combine it with the orthogonal analysis experimental data under the set operating dryness. For a wet steam boiler with flame signal = 1 and currently in operation, make the following judgment:
[0038] When the steam pressure of the wet steam boiler is 4-12MPa, the feedwater temperature is between 50-110℃, the boiler unit consumption is between 64.7-57.49, and the unit consumption decreases by no more than the set threshold for every 10℃ increase in pump feedwater temperature, the quality of the wet steam produced by the current wet steam boiler is judged to meet the standard.
[0039] Boiler unit consumption = natural gas flow rate / pump feed water flow rate.
[0040] Furthermore, step S4 also includes the following steps:
[0041] The boiler steam quality pass rate is automatically calculated and dynamically released at two levels: workshop and work group.
[0042] Furthermore, step S4 specifically includes:
[0043] Step S401: Based on the boiler operating status data, determine the number of boilers operating in the current day within 24 hours from the previous day to the current time. Each boiler with an operating time of 12 hours or more is considered to be operating in the current day, while boilers with an operating time of less than 12 hours are not counted.
[0044] Step S402: Based on the boiler steam quality diagnosis model, diagnose and statistically analyze the duration of substandard steam quality for each boiler within 24 hours from the previous day to the current time.
[0045] If a boiler operates for more than 3 hours on a given day with substandard steam quality, it will be marked as a boiler with substandard steam quality.
[0046] The substandard operating time does not include the initial superheating time at the start of furnace operation;
[0047] Step S403: Map the basic information of boilers with substandard steam quality to the corresponding steam supply station and operating team, and automatically calculate the boiler steam quality pass rate at the workshop and team levels, and dynamically release the pass rate.
[0048] Compared with the prior art, the beneficial effects of the present invention are:
[0049] This invention provides a steam quality diagnosis method for oilfield steam injection boilers. By analyzing the boiler PLC controller communication protocol, basic parameters such as superheat of the superheated boiler, separator liquid level, and flame signal are obtained. Data preprocessing is used to identify and filter dead values, null values, and abnormally large values. A multiple linear regression algorithm is applied to determine the core parameters related to the steam quality of both superheated and wet steam boilers. Orthogonal analysis experiments are used to determine the optimal operating indicators for each core parameter, establishing boiler steam injection quality management rules. This yields a boiler steam quality diagnosis model. Based on real-time diagnosis results, a boiler steam quality pass rate judgment model is established using a database, and boiler steam quality pass rate is automatically statistically analyzed by shift and station. This diagnostic method provides a simpler alternative to traditional experimental methods that require periodic manual testing of boiler steam quality and the determination of the content requirements and limitations of various components in boiler steam. It achieves automatic detection and process control of boiler steam quality.
[0050] Meanwhile, judging from the quantitatively calculated boiler steam quality qualification rate and the application of the multiple linear regression algorithm to establish a steam quality analysis model for injection boilers, the accuracy of this invention reaches over 99%, which can meet the accuracy requirements of boiler steam quality evaluation. In terms of timeliness, the original manual experience judgment of the steam quality analysis time for a single boiler was 5 minutes. After adopting the method of this invention, the cycle of the algorithm for 140 boilers is 2 minutes (i.e., the steam quality analysis time for a single boiler is 1 / 70 minutes). In comparison, the diagnostic efficiency of the method of this invention is 350 times that of the original, effectively meeting the needs of high-efficiency supervision and evaluation of boiler steam quality. Attached Figure Description
[0051] Figure 1 This is a flowchart of the oilfield steam injection boiler steam quality diagnosis method according to an embodiment of the present invention;
[0052] Figure 2 This is a flowchart illustrating the process of determining the quality of superheated boiler steam according to an embodiment of the present invention. Detailed Implementation
[0053] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.
[0054] Example 1
[0055] This invention provides a method for diagnosing the steam quality of an oilfield steam injection boiler, combined with... Figure 1 As shown, the method includes the following steps:
[0056] Step S1: Data acquisition and preprocessing, specifically including:
[0057] Step S101: Extract boiler production and operation data by parsing the boiler PLC controller communication protocol;
[0058] Step S102: For the extracted production operation data, use data standardization and normalization algorithms to eliminate data dimensions, use normal distribution algorithms for classification and identification, and remove unqualified data with excessive deviation;
[0059] Step S2: Apply the multiple linear regression algorithm to perform data correlation analysis and determine the core parameters for steam quality diagnosis;
[0060] Step S3: Apply orthogonal analysis experiments to determine the optimal operating index values of the core parameters for steam quality diagnosis, establish boiler steam injection quality management rules, and thus obtain the boiler steam quality diagnosis model, which specifically includes:
[0061] Step S301: Apply orthogonal analysis experiments to determine the optimal operating index values of the core parameters for steam quality diagnosis.
[0062] Step S302: Based on the optimal operating index values of the core parameters used for steam quality diagnosis, establish boiler steam injection quality management rules to obtain the boiler steam quality diagnosis model.
[0063] Step S4: Based on the boiler steam quality diagnosis model, the daily operating time of the boiler with substandard steam quality is statistically analyzed. A boiler steam quality pass rate judgment model is established through the database to automatically diagnose whether the boiler steam quality is qualified and to statistically analyze the boiler steam quality pass rate, using data to accurately reflect the boilers with substandard steam quality.
[0064] Furthermore, in some embodiments, step S4 further includes:
[0065] The system automatically calculates and dynamically publishes boiler steam quality pass rates at both the workshop and work group levels, enabling more intuitive and efficient automatic detection and process control of boiler steam quality.
[0066] Example 2
[0067] The method in Example 1 will be further described in detail below with reference to specific application examples, so that the purpose, technical solution and advantages of the present invention will be clearer.
[0068] The boiler in this embodiment includes both a superheated boiler and a wet steam boiler. It is possible that only one type of boiler may be used in other scenarios. If only one type of boiler is used, the implementation process of the steam quality diagnosis method for the corresponding boiler in this embodiment is the same or similar, and should also be included within the scope of this invention.
[0069] The specific implementation steps of this embodiment are as follows:
[0070] Step S1: Data acquisition and preprocessing;
[0071] Step S101: Extract boiler production and operation data by parsing the boiler PLC controller communication protocol;
[0072] Step S102: For the extracted production operation data, data standardization and normalization algorithms are used to eliminate data dimensions. A normal distribution algorithm is then used for classification and identification to remove unqualified data with excessive deviations. Specifically, this includes:
[0073] Data quality control techniques are applied to the extracted production operation data, including data cleaning, data scaling (i.e., adjusting the data scale to be on the same order of magnitude), and standardization transformation. This transforms the data into a standard normal distribution with a mean of 0 and a standard deviation of 1, identifying, filtering, and removing outliers, missing values, and duplicates. The characteristics of the normal distribution easily identify outliers that deviate from the mean. These outliers may be due to data entry errors, measurement errors, or other reasons, requiring further inspection and cleaning. Transforming the data to conform to a standard normal distribution (mean of 0, standard deviation of 1) helps eliminate dimensional differences between different variables, making data analysis more accurate.
[0074] In this step, the data standardization formula, normalization formula, standard normal distribution probability density function (PDF), and standard normal distribution cumulative distribution function (CDF) used are all existing technologies and will not be elaborated here.
[0075] When removing outliers, statistical test methods are applied. The significance level for detecting outliers is specified as α = 0.05, which is called the detection level. The significance level for detecting highly abnormal outliers is specified as α = 0.01, which is called the rejection level.
[0076] Step S2: Apply the multiple linear regression algorithm to perform data correlation analysis and determine the core parameters for steam quality diagnosis;
[0077] The multiple linear regression algorithm is as follows:
[0078] y = β0+β1x1+β2x2+... + β k xk +ε (1)
[0079] In equation (1) above: y is the dependent variable; β0, β1, β2, ..., β k These are regression coefficients; x1, x2, ..., x k ε is the independent variable; ε is the error term. The best-fitting line is found by minimizing the squared error function.
[0080] The squared error function is defined as:
[0081] J(β0, β1, β2,...,β k )=∑(y i -(β0+β1x 1i +β2x 2i +...+β k x ki )) 2 (2)
[0082] In the above formula (2): J(β0,β1,β2,...,β k y is the loss function (also known as the residual sum of squares), used to measure the difference between the model's predictions and the actual observed values; i It is the actual observed value of the i-th observation point; (β0+β1x) 1i +β2x 2i +...+β k x ki ) is the model prediction; x 1i x 2i , ..., x ki It is the independent variable.
[0083] In Python, multiple linear regression can be implemented using libraries such as statsmodels or sklearn. An example Python script for the multiple linear regression algorithm is shown below:
[0084] "from sklearn.model_selection import train_test_split
[0085] from sklearn.linear_model import LinearRegression
[0086] from sklearn.metrics import mean_squared_error
[0087] import pandas as pd
[0088] #Suppose we have a DataFrame named df, which contains independent and dependent variables.
[0089] #X is the independent variable DataFrame, and y is the dependent variable Series.
[0090] X = df[['x1','x2','x3']] # Assume there are three independent variables x1, x2, x3
[0091] y = df['y'] # Assume y is the dependent variable
[0092] # Divide the dataset into training and test sets
[0093] X_train,X_test,y_train,y_test=train_test_split(X,y,test_size=0.2,random_state=42)
[0094] #Creating and training the model
[0095] model = LinearRegression()
[0096] model.fit(X_train, y_train)
[0097] #Prediction Test Set
[0098] y_pred=model.predict(X_test)
[0099] #Evaluation Model
[0100] mse=mean_squared_error(y_test,y_pred)
[0101] print(f'Mean Squared Error:{mse}')
[0102] # Output model coefficients and intercept
[0103] print(f'Coefficients:{model.coef_}')
[0104] print(f'Intercept:{model.intercept_}')".
[0105] For operating scenarios that simultaneously include superheated boilers and wet steam boilers, step S2 specifically includes:
[0106] Step S201: Based on the multiple linear regression algorithm, extract the correlation parameters related to steam quality judgment, and use the parameters with a correlation coefficient > 0.4 as the core parameters for steam quality diagnosis;
[0107] Step S202: Among the 23 parameters for superheated boiler steam quality diagnosis, determine the top 3 parameters that are most relevant to superheated steam quality as the core parameters for superheated steam quality diagnosis, specifically including: superheat, separator level, and flame signal.
[0108] Among the 20 parameters used for steam quality diagnosis of wet steam boilers, the parameters that are most relevant to wet steam quality are identified as the core parameters for wet steam quality diagnosis. These parameters include: flame signal, natural gas flow rate, pump feedwater flow rate, feedwater temperature, dryness, and steam pressure.
[0109] Step S3: Apply orthogonal analysis experiments to determine the optimal values for each index of the algorithm;
[0110] Step S301: Apply orthogonal analysis experiments to determine the optimal operating index values of the core parameters for superheated steam quality diagnosis and the optimal operating index values of the core parameters for wet steam quality diagnosis, respectively.
[0111] The experimental data for orthogonal analysis are shown in Table 1-5:
[0112] Table 1. Orthogonal Verification Table for Superheated Steam Quality Compliance of Superheated Boiler with Separator Liquid Level Upper Limit of 800mm
[0113]
[0114] Table 2. Orthogonal Verification Table for Superheated Steam Quality Compliance of Superheated Boiler with Separator Liquid Level Upper Limit of 800mm
[0115]
[0116]
[0117] Table 3. Orthogonal Verification Table for Superheated Steam Quality Compliance of Superheated Boiler with Separator Liquid Level of 1850mm (Table 1)
[0118]
[0119] Table 4. Orthogonal Verification Table 2 for Superheated Steam Quality Compliance of Superheated Boiler with Separator Liquid Level of 1850mm
[0120]
[0121] Table 5. Orthogonal experimental table for verifying the unit consumption of wet steam boilers.
[0122]
[0123] Step S302: Based on the optimal operating index values of the core parameters for superheated steam quality diagnosis and the optimal operating index values of the core parameters for wet steam quality diagnosis, establish steam injection quality management rules for superheated boilers and wet steam boilers respectively. This will yield the superheated steam quality diagnosis model for superheated boilers and the wet steam quality diagnosis model for wet steam boilers, respectively. Wherein:
[0124] A. The steam injection quality management rules for the superheated steam quality diagnostic model of superheated boilers are as follows (see...). Figure 2 ):
[0125] Extract the operating data of the superheated boiler and, in conjunction with Table 1-4, make the following judgments for superheated boilers that are currently in operation with a flame signal of 1 (if the flame signal is 0, the boiler is shut down and no judgment is made):
[0126] When 1℃≤superheat≤30℃, the current superheated boiler is directly judged to be operating in superheated mode, and the quality of the superheated steam produced meets the standards.
[0127] When the superheat is <1℃, the superheat is empty, the superheat is 0℃, or the superheat is >30℃, the current superheat boiler model is identified, and further judgment is made based on the separator liquid level:
[0128] If the separator liquid level is higher than the set liquid level corresponding to the current superheated boiler model, then the current superheated boiler is determined to be in saturation operation.
[0129] If the current superheated boiler has accumulated more than 3 hours of saturated operation time in the 24 hours from the previous day to the current time, then the quality of the superheated steam produced by the current superheated boiler is judged to be substandard.
[0130] It should be noted that the saturated operating time of the superheated boiler within 24 hours from the previous day to the present does not include the 2-hour superheating time after ignition; after ignition, the boiler needs to run for a period of time to reach the superheated operating state, and this period of time is called the superheating time.
[0131] For example: For a superheated boiler with a separator level gauge upper limit of 800mm: if the separator level is >620mm (see Table 1 and Table 2), the superheated boiler is judged to be operating at saturation; if the superheated boiler has accumulated more than 3 hours of saturation operation time in the 24 hours from the previous day to the current time, the quality of the superheated steam produced by the current superheated boiler is judged to be substandard.
[0132] For superheated boilers with a separator level gauge upper limit of 1850mm: if the separator level is >800mm (see Tables 3 and 4), the superheated boiler is considered to be operating at saturation; if the superheated boiler has been operating at saturation for more than 3 hours in the 24 hours from the previous day to the current time (excluding the 2 hours of superheating after ignition), the quality of the superheated steam produced by the superheated boiler is not up to standard.
[0133] B. The steam injection quality management rules for the wet steam quality diagnostic model of wet steam boilers are as follows:
[0134] Extract the operating data of the wet steam boiler and, in conjunction with the orthogonal analysis experimental data under the set operating dryness in Table 5, make the following judgments for wet steam boilers that are in operation with flame signal = 1 (flame signal = 0 indicates shutdown and no judgment is made):
[0135] When the steam pressure of the wet steam boiler is 7MPa, the feedwater temperature is between 50-110℃, the boiler unit consumption is between 64.7-57.49, and the unit consumption decreases by no more than 1.2 for every 10℃ increase in pump feedwater temperature, the quality of the wet steam produced by the wet steam boiler is deemed to meet the standards.
[0136] Boiler unit consumption = natural gas flow rate / pump feed water flow rate, and boiler unit consumption is dimensionless.
[0137] Step S4: Based on the boiler steam quality diagnosis model, the boiler operating time under non-compliance is statistically analyzed. A boiler steam quality qualification rate judgment model is established through the database to automatically diagnose whether the boiler steam quality is qualified and to statistically analyze the boiler steam quality qualification rate, so as to accurately reflect the boilers with unqualified steam quality using data.
[0138] At the same time, the boiler steam quality qualification rate is automatically counted and dynamically released at the workshop and work group levels respectively.
[0139] Step S4 specifically includes:
[0140] Step S401: Based on the boiler operating status data, determine the number of boilers operating in the current day within 24 hours from the previous day to the current time. Each boiler with an operating time of 12 hours or more is considered to be operating in the current day, while boilers with an operating time of less than 12 hours are not counted.
[0141] Step S402: Based on the boiler steam quality diagnostic model, diagnose and statistically analyze the duration of substandard steam quality for each boiler within 24 hours from the previous day to the current time (e.g., from 8:00 AM the previous day to 8:00 AM the current day); it should be noted that the duration of substandard operation is calculated from the time when the boiler steam quality diagnostic model diagnoses the substandard boiler steam quality.
[0142] If the boiler operates for more than 3 hours with substandard steam quality on the same day, it will be marked as a boiler with substandard steam quality.
[0143] The following applies to the operating time when the steam quality is substandard on the same day: the initial superheating time at the start of the furnace (2 hours in this example).
[0144] Step S403: Map the basic information of boilers with substandard steam quality to the corresponding steam supply station and operating team, and automatically calculate the boiler steam quality pass rate at the workshop and team levels respectively, and dynamically release the pass rate.
[0145] The above description is merely an embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, and improvements made within the scope of the present invention should be included within the protection scope of the present invention.
Claims
1. A method for diagnosing the steam quality of an oilfield steam injection boiler, characterized in that, The method includes the following steps: Step S1: Data acquisition and preprocessing; Step S2: Apply the multiple linear regression algorithm to perform data correlation analysis and determine the core parameters for steam quality diagnosis; Step S3: Apply orthogonal analysis experiments to determine the optimal operating index values of the core parameters for steam quality diagnosis, establish boiler steam injection quality management rules, and obtain the boiler steam quality diagnosis model. Step S4: Based on the boiler steam quality diagnosis model, calculate the duration of boiler operation with substandard steam quality on the same day, establish a boiler steam quality pass rate judgment model through the database, automatically diagnose whether the boiler steam quality is qualified, and calculate the boiler steam quality pass rate.
2. The method for diagnosing the steam quality of an oilfield steam injection boiler according to claim 1, characterized in that, Step S1 specifically includes: Step S101: Extract boiler production and operation data by parsing the boiler PLC controller communication protocol; Step S102: For the extracted production operation data, data standardization and normalization algorithms are used to eliminate data dimensions, and normal distribution algorithms are used for classification and identification to remove unqualified data.
3. The method for diagnosing the steam quality of an oilfield steam injection boiler according to claim 2, characterized in that, Step S102 specifically includes: By applying data quality inspection technology, the extracted production operation data is cleaned, scaled, and standardized to transform it into a standard normal distribution with a mean of 0 and a standard deviation of 1. Outliers, missing values, and duplicate values are then identified, filtered, and removed.
4. The method for diagnosing the steam quality of an oilfield steam injection boiler according to claim 1, characterized in that, Step S2 specifically includes: Step S201: Based on the multiple linear regression algorithm, extract the correlation parameters related to steam quality judgment, and select parameters with a correlation coefficient > 0.4 as parameters for steam quality diagnosis; Step S202: Among the parameters used for steam quality diagnosis, determine the parameters that are among the top N items in terms of correlation with steam quality as the core parameters for steam quality diagnosis.
5. The method for diagnosing the steam quality of an oilfield steam injection boiler according to claim 4, characterized in that, Step S202 specifically includes: Among the parameters used for diagnosing the steam quality of superheated boilers, the parameters that are most correlated with the superheated steam quality are identified as the core parameters for superheated steam quality diagnosis. These parameters include: superheat, separator level, and flame signal. And / or, Among the parameters used for steam quality diagnosis of wet steam boilers, the parameters that are most relevant to wet steam quality are identified as the core parameters for wet steam quality diagnosis. These parameters include: flame signal, natural gas flow rate, pump feedwater flow rate, feedwater temperature, dryness fraction, and steam pressure.
6. The method for diagnosing the steam quality of an oilfield steam injection boiler according to claim 5, characterized in that, Step S3 specifically includes: Step S301: Apply orthogonal analysis experiments to determine the optimal operating index values of the core parameters for steam quality diagnosis of superheated boilers and / or wet steam boilers; Step S302: Based on the optimal operating index values of the core parameters used for superheated steam quality diagnosis, establish the steam injection quality management rules for the superheated boiler, and the superheated steam quality diagnosis model of the superheated boiler can be obtained. And / or, Based on the optimal operating index values of the core parameters used for wet steam quality diagnosis, steam injection quality management rules for wet steam boilers can be established, thus obtaining the wet steam quality diagnosis model for wet steam boilers.
7. The method for diagnosing the steam quality of an oilfield steam injection boiler according to claim 6, characterized in that, The steam injection quality management rules of the superheated steam quality diagnostic model are as follows: Extract the operating data of the superheated boiler, and make the following judgments for the superheated boiler that is in operation with flame signal = 1: When 1℃≤superheat≤30℃, the current superheated boiler is judged to be operating in superheated mode, and the quality of the superheated steam produced meets the standards. When the superheat is <1℃, the superheat is empty, the superheat is 0℃, or the superheat is >30℃, the current superheat boiler model is identified, and further judgment is made based on the separator liquid level: If the separator liquid level is higher than the set liquid level corresponding to the current superheated boiler model, then the current superheated boiler is determined to be in saturation operation. If the total saturated operating time of the superheated boiler in the 24 hours from the previous day to the present time is greater than 3 hours, then the quality of the superheated steam produced by the current superheated boiler is judged to be substandard. Wherein: the saturated operating time of the superheated boiler in the 24 hours from the previous day to the present time does not include the superheating time after boiler start-up; the definition of superheating time is: After ignition, the boiler needs to run for a period of time to reach the superheated operating state; this period is called the superheating time.
8. The method for diagnosing the steam quality of an oilfield steam injection boiler according to claim 6, characterized in that, The steam injection quality management rules of the wet steam quality diagnostic model are as follows: Extract the operating data of the wet steam boiler and combine it with the orthogonal analysis experimental data under the set operating dryness. For a wet steam boiler with flame signal = 1 and currently in operation, make the following judgment: When the steam pressure of the wet steam boiler is 4-12MPa, the feedwater temperature is between 50-110℃, the boiler unit consumption is between 64.7-57.49, and the unit consumption decreases by no more than the set threshold for every 10℃ increase in pump feedwater temperature, the quality of the wet steam produced by the current wet steam boiler is judged to meet the standard. Boiler unit consumption = natural gas flow rate / pump feed water flow rate.
9. The method for diagnosing the steam quality of an oilfield steam injection boiler according to any one of claims 1-8, characterized in that, Step S4 further includes the following steps: The boiler steam quality pass rate is automatically calculated and dynamically released at two levels: workshop and work group.
10. The method for diagnosing the steam quality of an oilfield steam injection boiler according to claim 9, characterized in that, Step S4 specifically includes: Step S401: Based on the boiler operating status data, determine the number of boilers operating in the current day within 24 hours from the previous day to the current time. Each boiler with an operating time of 12 hours or more is considered to be operating in the current day, while boilers with an operating time of less than 12 hours are not counted. Step S402: Based on the boiler steam quality diagnosis model, diagnose and statistically analyze the duration of substandard steam quality for each boiler within 24 hours from the previous day to the current time. If a boiler operates for more than 3 hours on a given day with substandard steam quality, it will be marked as a boiler with substandard steam quality. The substandard operating time does not include the initial superheating time at the start of furnace operation; Step S403: Map the basic information of boilers with substandard steam quality to the corresponding steam supply station and operating team, and automatically calculate the boiler steam quality pass rate at the workshop and team levels, and dynamically release the pass rate.