A crude oil scheduling fault-tolerant control method considering key properties of kerosene

Abstract

The application provides a crude oil scheduling fault-tolerant control method considering key properties of kerosene, an optimization scheduling scheme is generated by constructing a blending ratio and scheduling production model and dynamically correcting a model boundary and an operation constraint according to a real-time diagnosis result. When the system identifies that the crude oil property seriously deviates from the expectation, the blending ratio is re-optimized based on a safety constraint boundary, the optimal proportion of each component crude oil is determined, and then the scheduling production scheme is dynamically adjusted in combination with a storage tank state, so that the blended crude oil with stable properties is continuously provided under the abnormal working condition of the crude oil property. The method effectively guarantees the stable controllability of the kerosene product quality and improves the adaptability and reliability of production scheduling of a refining enterprise.

Description

Technical Field

[0001] This invention relates to the field of production scheduling in refining and chemical enterprises, specifically to a fault-tolerant control method for crude oil scheduling that takes into account the key properties of kerosene. Background Technology

[0002] Kerosene, as a crucial fuel in industrial production and energy supply, is primarily refined using atmospheric and vacuum distillation units. Key kerosene properties, such as the narrow smoke point of jet fuel, are directly related to combustion performance and operational safety, and these parameters are essentially determined by the properties of the blended crude oil entering the unit. Therefore, ensuring the stability of kerosene's key properties hinges on achieving stable and controllable properties of the blended crude oil.

[0003] To achieve the above objectives, refineries need to optimize the scheduling of various crude oils stored in crude oil tank farms. However, in actual operation, to reduce procurement costs, companies often purchase opportunistic crude oils at lower prices. The key property data obtained during the procurement of such crude oils is often incomplete or inaccurate. If production plans are formulated directly based on this data, the properties of the blended crude oils can easily deviate from expectations, thus affecting the quality of kerosene products. In particular, traditional scheduling methods often neglect effective constraints on key properties affecting kerosene product quality. Once these key property indicators experience abnormal fluctuations, the original scheduling plan will fail, requiring re-optimization of the crude oil processing sequence and blending ratio in the tank farm, increasing the volatility and risk of production operations.

[0004] Therefore, it is necessary to monitor the properties of blended crude oil in real time and diagnose any faults to address the aforementioned risks.

[0005] In summary, this invention proposes a fault-tolerant control method for crude oil scheduling that considers the key properties of kerosene. This method aims to overcome the shortcomings of existing manual scheduling methods, such as coarse calculations and reliance on experience, by achieving online solution of the optimal blending ratio and generating production scheduling plans. This results in providing blended crude oil with stable properties, ensuring the stable and controllable quality of kerosene products even under abnormal crude oil property conditions. Summary of the Invention

[0006] This invention discloses a fault-tolerant control method for crude oil scheduling that considers the key properties of kerosene, which is used to generate the optimal production scheduling plan online when the properties of crude oil are abnormal, so as to ensure the stable and controllable quality of kerosene products.

[0007] This method includes the following steps:

[0008] (1) Obtain production process data and perform online fault diagnosis based on the preprocessed data;

[0009] (2) If the diagnosis is that the crude oil properties are abnormal, proceed to step (3); otherwise, continue with the original scheduling and production plan and proceed to step (9).

[0010] (3) Calculate the properties of blended crude oil according to the following formula. Species Indicators With target value relative deviation If any If the deviation is greater than the deviation threshold D, proceed to step (5); otherwise, proceed to step (4).

[0011]

[0012] in, Represents a set of properties. ,serial number These correspond to sulfur content, acid value, density, and narrow fraction smoke point of aviation kerosene, respectively.

[0013] (4) Adjust the operating parameters in the control closed loop and continue using the original scheduling plan, then jump to step (9).

[0014] (5) Based on the diagnostic conclusions, correct the crude oil properties of the tanks in the plant's tank area, and combine the corrected crude oil properties, inventory and production and processing requirements to design a crude oil blending ratio model and a scheduling model.

[0015] (5-1) Establish a crude oil blending ratio model and solve for the optimal blending ratio of existing crude oil components in the plant's tank area; to minimize the overall deviation of blended crude oil properties and production costs, minimize the objective function. To obtain the optimal blending ratio The calculation is as follows:

[0016]

[0017] Wherein, objective function It is expressed as follows:

[0018]

[0019] In the formula, Let be the crude oil blending ratio of the components to be solved. Indicates the composition of crude oil in the plant's tank area The blending ratio, This indicates the total number of component crude oils that can be dispensed in the plant's tank area. and These are the weighting coefficients. Indicates the composition of crude oil in the plant's tank area production costs, Indicates crude oil The weight of the property indicators;

[0020] (5-2) Establish a scheduling model by minimizing the objective function. Obtain the optimal tank switching scheme within the scheduling period. The calculation is as follows:

[0021]

[0022] Wherein, objective function It is expressed as follows:

[0023]

[0024] In the formula, The solution to be found is the tank switching scheme. Indicates in From the storage tank at all times Switch to storage tank The discrete variable has a value of 0 indicating no switching and a value of 1 indicating switching. In order to schedule production cycles, This refers to the number of crude oil tanks. For any scheduling time, and For different crude oil storage tanks;

[0025] (6) Generate a crude oil scheduling and production plan report based on the optimal blending ratio and scheduling plan;

[0026] (7) The crude oil scheduling and production plan report is sent to the workshop for execution.

[0027] In this method, the range of the deviation threshold D is: .

[0028] In this method, the constraints of the crude oil blending ratio model are the crude oil blending ratio constraint, the upper and lower limits of the crude oil blending ratio constraint, and the upper and lower limits of the blended crude oil properties constraint. The expressions for each constraint are as follows:

[0029] (1) Crude oil blending ratio constraints:

[0030]

[0031] (2) Upper and lower limits of crude oil blending ratio constraints:

[0032]

[0033] in, and This indicates that crude oil is composed of components in the tank area within the plant. The lower and upper limits of the blending ratio;

[0034] (3) Upper and lower limits of blended crude oil properties constraints:

[0035]

[0036] in, and This indicates that during feeding, the first... The upper and lower limits of the permissible properties.

[0037] In this method, sulfur content, acid value, crude oil density, and narrow fraction smoke point of jet fuel are selected as indicators of the properties of blended crude oil. The properties are expressed as follows:

[0038] (1) Sulfur content and acid value show a linear additive relationship:

[0039]

[0040] in, Indicates the composition of crude oil in the plant's tank area The a kind of property, It is a set of linearly weighted crude oil properties. ,serial number These correspond to sulfur content and acid value, respectively.

[0041] (2) The crude oil density adopts a nonlinear harmonic model:

[0042]

[0043] in, To adjust the density of crude oil, Indicates the composition of crude oil in the plant's tank area The density;

[0044] (3) The narrow fraction smoke point of aviation kerosene is calculated using the following empirical formula model:

[0045] First, based on the aniline point of each component of the crude oil Calculate the aniline point of blended crude oil :

[0046]

[0047] Furthermore, the obtained aniline point Density of blended crude oil Substituting into the empirical formula, we can solve for the narrow fraction smoke point of jet fuel in blended crude oil. :

[0048]

[0049] in, , and is the regression coefficient.

[0050] In this method, the constraints of the crude oil scheduling and production model are the oil quantity limit constraint in the storage tank, the material balance constraint in the storage tank, and the processing load constraint. The expressions for each constraint are as follows:

[0051] (1) Limitation constraints on oil volume in storage tanks:

[0052]

[0053] in, for Time storage tank The amount of oil in it, and Storage tanks The lower and upper limits of oil quantity;

[0054] (2) Material balance constraints in storage tanks:

[0055]

[0056] in, for Time storage tank Processing volume, for Time and The time interval between moments;

[0057] (3) Processing load constraints:

[0058]

[0059] in, for Constant processing load, and These represent the lower and upper limits of crude oil processing load, respectively.

[0060] In this method, to ensure the safety of downstream critical equipment, the safety constraint boundary is modified for abnormal operating conditions that are not related to crude oil properties as follows:

[0061] (1) Reduce the blending ratio of high-risk crude oil components; the upper limit of the blending ratio of crude oil component c after revision. The expression is as follows:

[0062]

[0063] in, This is a correction factor for the proportion of high-risk crude oil components. ;

[0064] (2) Narrow the constraint range of abnormal properties in blended crude oil. The corrected constraint range can be expressed as:

[0065]

[0066]

[0067] in, and This indicates the corrected blended crude oil. Lower and upper limits of property constraints, To narrow the range, The narrowing coefficient is the constraint for anomalous properties;

[0068] (3) Increase the weight coefficients in the objective function of the crude oil blending ratio model. Corrected weighting coefficients The calculation is as follows:

[0069]

[0070] in, This is the property deviation weighting correction coefficient. ;

[0071] (4) Reduce the upper limit of crude oil processing load during the scheduling cycle. The revised upper limit of processing load is as follows: The expression is as follows:

[0072]

[0073] in, This is a correction factor for processing load. .

[0074] Beneficial effects:

[0075] This invention provides a fault-tolerant control method for crude oil scheduling that considers the key properties of kerosene. By constructing a blending ratio and scheduling model, and dynamically correcting the model boundaries and operational constraints based on real-time diagnostic results, an optimal scheduling plan is generated. When the system identifies that the crude oil properties significantly deviate from expectations, it first re-optimizes the blending ratio based on safety constraint boundaries to determine the optimal proportions of each component crude oil. Then, combined with the storage tank status, it dynamically adjusts the scheduling plan to provide blended crude oil with stable properties, ensuring the stable and controllable quality of kerosene products under abnormal crude oil property conditions. Attached Figure Description

[0076] Figure 1 This is a flowchart of a fault-tolerant control method for crude oil dispatching that takes into account the key properties of kerosene.

[0077] Figure 2 This is the scheduling and production plan obtained in the implementation examples of this invention. Detailed Implementation

[0078] The following detailed calculation process and specific operation procedures are provided with reference to the accompanying drawings and specific examples to further illustrate the present invention. This embodiment is implemented based on the technical solution of the present invention, but the scope of protection of the present invention is not limited to the following embodiment.

[0079] This case study focuses on the crude oil scheduling process of an atmospheric and vacuum distillation unit in a refining and chemical enterprise. The enterprise determines its blending formula based on processing plans and crude oil property data, blending the various crude oil components in proportion before supplying them to the atmospheric and vacuum distillation unit. To reduce procurement costs, the enterprise purchases a batch of lower-priced opportunistic crude oil and incorporates it into the production plan for blending. This case study details the implementation process and effectiveness of the described scheduling fault-tolerant control method in response to the kerosene smoke point anomaly that occurred on May 22, 2024. At the time of the fault, the key property indicators of each component of the crude oil stored in the plant's tank area are shown in Table 1.

[0080] Table 1. Properties of crude oil components in the plant's tank farm

[0081] The specific implementation process of this case is as follows: Figure 1 As shown, the detailed steps are as follows:

[0082] 1) Acquire production process data, perform preprocessing such as outlier removal and missing value interpolation, and analyze the data using an online fault diagnosis system based on fault tree analysis (FTA).

[0083] 2) At 22:00 on May 22, 2024, the system diagnosed an abnormal smoke point for the narrow fraction of kerosene-jet fuel. Calculations showed the relative deviation of the smoke point for the narrow fraction of blended crude oil-jet fuel. The value is 0.15, which exceeds the deviation threshold (D = 0.1) set in this case.

[0084] 3) Analysis revealed that the anomaly was caused by inaccurate property data for the opportunistic crude oil CRUDE04, leading to unreasonable blending ratios and production scheduling plans calculated based on this data. Subsequently, the property data of CRUDE04 was corrected based on the diagnostic conclusions. Combined with the corrected crude oil properties, inventory, and production processing requirements of the storage tanks, safety constraint boundaries and parameters for the crude oil blending ratio model and the production scheduling model were designed.

[0085] 4) Establish a crude oil blending ratio model to determine the optimal blending ratio for the existing crude oil components in the plant's tank farm. To minimize both the physical property deviation of the blended crude oil and production costs, the objective function is minimized. To obtain the optimal blending ratio The calculation is as follows:

[0086]

[0087] Wherein, objective function It is expressed as follows:

[0088]

[0089] In the formula, Let be the crude oil blending ratio of the components to be solved. Indicates the composition of crude oil in the plant's tank area The blending ratio, Represents a set of properties. ,serial number These correspond to sulfur content, acid value, density, and narrow fraction smoke point of aviation kerosene, respectively. Indicates the blending of crude oil Type of property index, Indicates the blending of crude oil The target values ​​for each property indicator; in this case, the weight coefficients for each property are set as follows: Weighting coefficient Weighting coefficient Total number of crude oil components It is 5. The prices are the component crude oil prices; specific values ​​are shown in Table 1.

[0090] The sum of the blending ratios of the various crude oil components should satisfy the following:

[0091]

[0092] in, Indicates the composition of crude oil in the plant's tank area The blending ratio.

[0093] The upper and lower limits of crude oil blending ratio constraints are expressed as follows:

[0094]

[0095] in, , , , , These represent the blending ratios of crude oil components CRUDE01, CRUDE02, CRUDE03, CRUDE04, and CRUDE05, respectively.

[0096] The constraints on the properties of blended crude oil are shown in Table 2.

[0097] Table 2. Constraints on the Properties of Blended Crude Oil

[0098] The optimal blending ratio for each component crude oil was obtained by solving the blending ratio model. The comparison of blending formulas for new and old crude oil is shown in Table 3.

[0099] Table 3 Comparison of blending formulas for new and old crude oil

[0100] 5) Establish a scheduling model by minimizing the objective function. Obtain the optimal tank switching scheme within the scheduling period. The calculation is as follows:

[0101]

[0102] Wherein, objective function It is expressed as follows:

[0103]

[0104] In the formula, Let the set of tank switching decision variables be the one to be solved. Indicates in From the storage tank at all times Switch to storage tank The discrete variable has a value of 0 indicating no switching and a value of 1 indicating switching. In order to schedule production cycles, This refers to the number of crude oil tanks. For any scheduling time, and For different crude oil storage tanks;

[0105] Storage tank oil volume limits:

[0106]

[0107] in, for Time storage tank The amount of oil in it, ;

[0108] Material balance constraints in storage tanks:

[0109] in, for Time storage tank Processing volume, for Time and The time interval between moments;

[0110] Processing load constraints:

[0111]

[0112] Among these measures, the upper limit of processing load will be increased to cope with abnormal operating conditions. The processing load limit has been reduced from 1200t / h to 1000t / h. Set to 600t / h for Constant processing load.

[0113] 6) Based on the optimal blending ratio and scheduling plan, generate a crude oil scheduling and production plan report, such as... Figure 2 As shown.

[0114] 7) The generated crude oil scheduling and production plan report was sent to the workshop for execution. The system executed the new formula switch as planned at 02:00 on May 23, 2024, and controlled the oil pipeline flow by adjusting the oil pump set and valves. By 09:00 on the same day, the properties of the blended crude oil met the unit's feed requirements. A comparison of the specific properties is shown in the table below.

[0115] Table 4 Comparison of Properties of Blended Crude Oil

[0116] Table 4 shows that after adopting the new formula, the smoke point of the narrow fraction of jet fuel in the blended crude oil has increased from the abnormal value of 22.4 mm under the original scheme to a reasonable range of 24.2 mm, and other key properties also meet the target requirements. The results indicate that this method achieves rapid online optimization of production scheduling schemes under abnormal crude oil property conditions, and can continuously provide stable blended crude oil to the atmospheric and vacuum distillation unit, ensuring the stable and controllable quality of kerosene products under abnormal crude oil property conditions.

Claims

1. A fault-tolerant control method for crude oil dispatch considering key properties of kerosene, characterized in that... A crude oil blending ratio model and a production scheduling model are established. Based on the diagnostic results, the model boundaries and operational constraints are corrected, and the optimal production scheduling scheme is generated by solving the problem. The specific steps are as follows: (1) Obtain production process data and perform online fault diagnosis based on the preprocessed data; (2) If the diagnosis is that the blended crude oil has abnormal properties, proceed to step (3); otherwise, continue with the original production scheduling plan and proceed to step (7). (3) Calculate the properties of blended crude oil according to the following formula. Species Indicators With target value relative deviation If any If the deviation is greater than the deviation threshold D, proceed to step (5); otherwise, proceed to step (4). in, Represents a set of properties. ,serial number These correspond to sulfur content, acid value, density, and narrow fraction smoke point of aviation kerosene, respectively. (4) Adjust the operating parameters in the control closed loop and continue using the original scheduling plan, then jump to step (7). (5) Based on the diagnostic conclusions, correct the crude oil properties of the tanks in the plant's tank area, and combine the corrected crude oil properties, inventory and production and processing requirements to design a crude oil blending ratio model and a scheduling model. (5-1) Establish a crude oil blending ratio model and solve for the optimal blending ratio of existing crude oil components in the plant's tank area; to minimize the overall deviation of blended crude oil properties and production costs, minimize the objective function. To obtain the optimal blending ratio The calculation is as follows: Wherein, objective function It is expressed as follows: In the formula, Let be the crude oil blending ratio of the components to be solved. Indicates the composition of crude oil in the plant's tank area The blending ratio, This indicates the total number of component crude oils that can be dispensed in the plant's tank area. and These are the weighting coefficients. Indicates the composition of crude oil in the plant's tank area production costs, Indicates crude oil The weight of the various property indicators; (5-2) Establish a scheduling model by minimizing the objective function. Obtain the optimal tank switching scheme within the scheduling period. The calculation is as follows: Wherein, objective function It is expressed as follows: In the formula, The solution to be found is the tank switching scheme. Indicates in From the storage tank at all times Switch to storage tank The discrete variable has a value of 0 indicating no switching and a value of 1 indicating switching. In order to schedule production cycles, This refers to the number of crude oil tanks. For any scheduling time, and For different crude oil storage tanks; (6) Generate a crude oil scheduling and production plan report based on the optimal blending ratio and scheduling plan; (7) The crude oil scheduling and production plan report is sent to the workshop for execution.

2. The crude oil dispatching fault-tolerant control method considering the key properties of kerosene according to claim 1, characterized in that... Deviation threshold D While ensuring that the properties of the blended crude oil meet the requirements, it is important to avoid relative deviations. Solving the crude oil blending ratio model and scheduling production model when the value is too small and triggers too frequently.

3. The crude oil dispatching fault-tolerant control method considering the key properties of kerosene according to claim 1, characterized in that... The design constraints for the crude oil blending ratio model are the crude oil blending ratio constraint, the upper and lower limits of the crude oil blending ratio constraint, and the upper and lower limits of the properties of the blended crude oil constraint. The expressions for each constraint are as follows: (1) Crude oil blending ratio constraints: (2) Upper and lower limits of crude oil blending ratio constraints: in, and This indicates that crude oil is composed of components in the tank area within the plant. The lower and upper limits of the blending ratio; (3) Upper and lower limits of blended crude oil properties constraints: in, and This indicates that during feeding, the first... The upper and lower limits of the permissible properties.

4. The crude oil dispatching fault-tolerant control method considering the key properties of kerosene according to claim 3, characterized in that... Sulfur content, acid value, crude oil density, and narrow fraction smoke point of jet fuel were selected as the property indicators for blended crude oil. The calculation of each property indicator is as follows: (1) Sulfur content and acid value show a linear additive relationship: in, Indicates the composition of crude oil in the plant's tank area The a kind of property, It is a set of linearly weighted crude oil properties. ,serial number These correspond to sulfur content and acid value, respectively. (2) The crude oil density adopts a nonlinear harmonic model: in, To adjust the density of crude oil, Indicates the composition of crude oil in the plant's tank area The density; (3) The narrow fraction smoke point of aviation kerosene is calculated using the following empirical formula model: First, based on the aniline point of each component of the crude oil Calculate the aniline point of blended crude oil : Furthermore, the obtained aniline point Density of blended crude oil Substituting into the empirical formula, we can solve for the narrow fraction smoke point of jet fuel in blended crude oil. : in, , and is the regression coefficient.

5. A crude oil dispatching fault-tolerant control method considering the key properties of kerosene according to claim 1, characterized in that... The constraints for the crude oil scheduling and production model are: tank oil quantity limit constraint, tank material balance constraint, and processing load constraint. The expressions for each constraint are as follows: (1) Limitation constraints on oil volume in storage tanks: in, for Time storage tank The amount of oil in it, and Storage tanks The lower and upper limits of oil quantity; (2) Material balance constraints in storage tanks: in, for Time storage tank Processing volume, for Time and The time interval between moments; (3) Crude oil processing load constraints: in, for Constant processing load, and These represent the lower and upper limits of crude oil processing load, respectively.

6. The crude oil dispatching fault-tolerant control method considering the key properties of kerosene according to claim 1, characterized in that, To ensure the safety of downstream critical equipment, the safety constraint boundary is modified as follows for non-crude oil properties that deviate significantly from the operating conditions: (1) Reduce the blending ratio of high-risk crude oil components; the upper limit of the blending ratio of crude oil component c after revision. The expression is as follows: in, This is a correction factor for the proportion of high-risk crude oil components. ; (2) Narrow the constraint range of abnormal properties in blended crude oil. The corrected constraint range can be expressed as: in, and This indicates the corrected blended crude oil. Lower and upper limits of property constraints, To narrow the range, The narrowing coefficient is the constraint for anomalous properties; (3) Increase the weight coefficients in the objective function of the crude oil blending ratio model. Corrected weighting coefficients The calculation is as follows: in, This is the property deviation weighting correction coefficient. ; (4) Reduce the upper limit of crude oil processing load during the scheduling cycle. The revised upper limit of processing load is as follows: The expression is as follows: in, This is a correction factor for processing load. .

Images