Oil refining whole process carbon footprint monitoring optimization decision-making platform

Abstract

The invention discloses an oil refining whole process carbon footprint monitoring optimization decision-making platform, and relates to the technical field of petroleum production. A data acquisition end is used for acquiring data of production equipment involved in the oil refining whole process in real time; the cost optimization end is used for acquiring the sum of heat energy required to be provided by a heating furnace according to the total processing amount of raw materials in the production plan and automatically adjusting the use amount of fuel; the emission monitoring linkage end is used for monitoring the carbon dioxide content of the gas exhausted by the heating furnace and generating a corresponding linkage instruction when the concentration value of the exhausted carbon dioxide is abnormal; and the path optimization end obtains the production data corresponding to the production equipment and generates a corresponding optimization decision, thereby solving the defects in the aspects of data utilization, cost emission balance, exception handling, decision optimization and the like in the prior art, and meeting the core requirements of enterprises in the double aspects of low-carbon transformation and production profit.

Description

Technical Field

[0001] The present invention relates to the field of oil production technology, and in particular to a carbon footprint monitoring and optimization decision-making platform for the entire oil refining process. Background Art

[0002] In the field of carbon management in the oil refining industry, existing technologies have achieved the goal of collecting some production data through sensors and calculating carbon emissions using basic models. Some systems can also perform simple energy consumption optimization and emission monitoring tasks. However, as the industry's demand for refined and intelligent carbon management increases, existing technologies face new challenges: insufficient comprehensive collection and efficient fusion processing capabilities for multi-source data make it difficult to meet the needs of complex production scenarios; when adjusting fuel usage, there is a lack of comprehensive dynamic optimization mechanisms for costs, carbon emissions, and production stability; abnormal responses to emission monitoring are not intelligent enough, and combustion anomalies cannot be handled quickly and accurately; production path optimization decisions lack in-depth analysis and dynamic adjustment of real-time data, making it difficult to achieve the optimal balance between energy conservation and emission reduction and corporate production profitability. These issues are prompting the industry to seek more advanced and comprehensive carbon management solutions to meet increasingly stringent environmental protection requirements and the needs of enterprises for refined management. Summary of the Invention

[0003] In response to the shortcomings of existing technologies, the present invention provides a carbon footprint monitoring and optimization decision-making platform for the entire refining process, which solves the shortcomings of existing technologies in data utilization, cost-emission balance, exception handling and decision optimization, and meets the core needs of enterprises in both low-carbon transformation and production profitability.

[0004] To achieve the above objectives, the present invention is implemented through the following technical solutions: a carbon footprint monitoring and optimization decision-making platform for the entire refining process, The data acquisition terminal is used to obtain real-time data on production equipment involved in the entire refining process; On the cost optimization side, based on the total amount of raw material processing in the production plan, the total amount of heat energy required by the heating furnace is obtained and the amount of fuel used is automatically adjusted; The emission monitoring linkage terminal monitors the carbon dioxide content of the exhaust gas from the heating furnace and generates corresponding linkage instructions when the concentration of carbon dioxide emitted is abnormal. The path optimization end obtains the production data corresponding to the production equipment and generates corresponding optimization decisions.

[0005] Preferably, the data of the production equipment includes energy consumption data NHi, raw material flow YLi, and product flow CLi, where 1≤i≤n, representing that there are n production equipment in total, and the value of i represents the specific equipment name.

[0006] Preferably, in the cost optimization end, the usage of different fuels RLj and their corresponding thermal energy conversion rates RZj are obtained, where 1≤j≤m, representing a total of m types of fuels, and the value of j represents the specific name of the fuel; According to the formula 、 Dynamically adjust RLj and generate constraints Constrain the adjustment value; Where G is the total heat energy corresponding to the total processing volume, Q is the fuel production cost, HBj is the real-time purchase price of the fuel, TP is the carbon emission of the heating furnace in the production plan, HLj is the carbon content percentage of the fuel, and PYj is the percentage of fuel burned in the heating furnace.

[0007] Preferably, when the price of clean energy drops, its RLj value is increased and the RLj value of non-clean energy is decreased; when the price of non-clean energy drops, its RLj value is increased and the RLj value of clean energy is decreased.

[0008] Preferably, in the emission monitoring linkage terminal, the carbon dioxide concentration EOd in the exhaust gas of the heating furnace is obtained by a sensor, where 1≤d≤e, which means that the carbon dioxide concentration is collected e times, and the collection frequency is 3 seconds / time; When EOd is abnormal, it is recorded as abnormal collection, and the number of abnormal collections c is counted; it is judged whether c / e≤γ is established, where γ is the abnormal collection probability threshold. If not, a linkage instruction is generated.

[0009] Preferably, the number of consecutive exceptions f is counted to determine whether 3×f≤β holds true, where β is a continuous exception duration threshold. If not, a linkage instruction is generated.

[0010] Preferably, it is determined whether (HLj×PYj)-α≤EOd≤(HLj×PYj)+α holds true, where α is a preset fluctuation value; If EOd<(HLj×PYj)-α, perform sensor self-test. If the sensor is normal, it means that the EOd value is abnormal.

[0011] Preferably, in the path optimization end, the carbon emissions CPi of the production equipment are obtained according to the formula CPi=YLi×YZi, where YZi is the carbon emission factor corresponding to the production equipment, and the total carbon emissions ZH of all production equipment are obtained according to ZH=CP1+CP2+...+CPn.

[0012] Preferred, judge Is it true, where θ is the preset value, CLi / YLi is the raw material conversion ratio, if not true, an energy-saving and emission reduction decision is generated.

[0013] Preferably, the carbon emission factor YZi is obtained by the formula YZi=F(Ti, YLi, CLi, NHi), where Ti is the reaction temperature of the production equipment.

[0014] This invention provides a platform for monitoring and optimizing the carbon footprint of the entire oil refining process. Compared with existing technologies, it has the following advantages: Comprehensive and accurate data processing: The data acquisition end uses a variety of sensors to comprehensively collect production equipment data in real time and performs dimensionless processing to provide the system with an accurate and standardized data foundation, thereby improving the reliability of subsequent analysis and decision-making.

[0015] Fuel Cost and Emission Optimization: The cost optimization platform dynamically adjusts fuel usage by integrating multiple factors, including fuel cost, thermal energy conversion rate, and carbon emission constraints. This approach considers price fluctuations for both clean and dirty energy sources while ensuring that carbon emissions meet production plans through constraints, achieving the dual benefits of lowering production costs and reducing carbon emissions.

[0016] Intelligent processing of emission anomalies: The emission monitoring linkage terminal judges the abnormal carbon dioxide concentration through multiple dimensions (such as the probability of abnormal collection and the duration of continuous anomalies), and promptly generates linkage instructions to adjust the feeding and air intake of the heating furnace to ensure sufficient combustion of the fuel, reduce the emission of harmful substances, and improve the environmental friendliness and stability of production.

[0017] Scientific decision-making on production paths: The path optimization end calculates carbon emissions based on real-time production data and generates optimization decisions (such as adjusting processes, replacing raw materials, etc.) based on the raw material conversion ratio. While saving energy and reducing emissions, it ensures the company's production and profit needs and achieves an organic balance between the two.

[0018] Enhanced overall competitiveness: This technology, through the collaborative work of various modules, comprehensively improves the intelligent and refined level of carbon management in oil refineries, helping companies to be more competitive in environmental compliance and cost control, and creating significant economic and environmental benefits. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] Figure 1 It is a system framework diagram of the present invention; Figure 2 This is a system flow chart of the emission monitoring linkage terminal of the present invention. DETAILED DESCRIPTION

[0020] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0021] See also Figure 1-2 The present invention provides a technical solution: a carbon footprint monitoring and optimization decision-making platform for the entire refining process: As a first embodiment of the present application, it specifically includes: a data acquisition terminal: used to acquire data of production equipment involved in the entire oil refining process in real time; its specific methods include: Sensors are installed on production equipment to obtain real-time energy consumption data NHi, raw material flow YLi, and product flow CLi, where 1≤i≤n, indicating that there are n production equipment, and the value of i represents the specific equipment name. Production equipment includes atmospheric and vacuum towers, catalytic cracking units, hydrocracking units, coking units, etc. Sensor types include concentration sensors, flow sensors, temperature sensors, etc. De-dimensionalize all types of acquired data; Cost optimization: Based on the total amount of raw material processing in the production plan, the total amount of heat energy required by the heating furnace is obtained and the amount of fuel used is automatically adjusted; the specific method is: The amount of different fuels is obtained, recorded as RLj and its corresponding thermal energy conversion rate RZj, where 1≤j≤m, representing a world with m types of fuels (such as coal, liquid fuel, gas fuel, and electricity). The value of j represents the specific name of the fuel. The weight values ​​DTj corresponding to different fuels are obtained simultaneously. The weight values ​​DTj are preset values. After the weight values ​​are optimized through historical data or expert systems, they are set by production personnel. The cleaner the fuel, the larger the corresponding weight value. The cleanliness level of the fuel is divided according to the carbon content of the fuel itself. According to the formula , , dynamically adjust the value of different fuel consumption RLj; the specific method is: When the price of clean energy (such as electricity, hydrogen) drops, that is, when the current fuel price drops by 10% compared to the previous purchase price, the value of the corresponding fuel consumption RLj is increased, and the value of the corresponding fuel consumption RLj of non-clean energy (coal) is simultaneously reduced; Similarly, when the price of non-clean energy (coal) drops, the value of its corresponding fuel consumption RLj increases, and the value of the corresponding fuel consumption RLj of clean energy (such as electricity, hydrogen) decreases simultaneously; For example, when the current coal price drops by 15% compared to the previous purchase price, the amount of coal used in processing is increased, and the use of other solid fuels, gas fuels, liquid fuels, and electricity is reduced simultaneously. However, the actual purchase cost of the adjusted coal use, other solid fuel use, gas fuel use, liquid fuel use, and electricity use cannot exceed the production cost in the production plan. However, the heat energy generated by the adjusted total fuel use must meet the total heat energy corresponding to the total processing volume. According to this constraint method, the use of different fuels is dynamically adjusted, and all fuels are purchased in the same type, concentration, and specification as the previous purchase. Where G is the total heat energy corresponding to the total processing volume, Q is the production cost of fuel in the production plan, and HBj is the real-time purchase price of different fuels; When dynamically adjusting the fuel consumption RLj value, constraints are generated to constrain the carbon emissions corresponding to the value of the total thermal energy G. The specific constraint method is: Get the carbon emission value of the heating furnace in the production plan and record it as TP; according to When the fuel consumption RLj value is dynamically adjusted, the adjusted value is constrained to avoid causing the actual carbon emissions after adjustment to exceed the carbon emissions value in the production plan. Among them, HLj is the carbon content percentage corresponding to different fuels, and PYj is the combustion percentage of different fuels in the heating furnace. The specific value is set by the production staff according to the actual production situation; Under the condition of production cost constraints, the use of fuel is further constrained by carbon emissions. That is, during the dynamic adjustment of fuel usage values, it is also necessary to consider that after the adjustment, the actual carbon emissions corresponding to the use of solid fuel, gas fuel and liquid fuel cannot exceed the carbon emissions in the production plan; In the actual production process, based on the real-time price of fuel, three multivariate linear equations are used to obtain the optimal solution for the RLj values ​​of different fuel consumptions, and synchronize dynamic adjustments are made to achieve energy conservation and emission reduction, reduce production costs, and meet the actual production needs of the enterprise.

[0022] As a second embodiment of the present application, it specifically includes: an emission monitoring linkage terminal: monitoring the carbon dioxide content of the exhaust gas of the heating furnace, and generating a corresponding linkage instruction when an abnormal value of the carbon dioxide concentration of the exhaust occurs, and the specific method is as follows: The carbon dioxide concentration in the exhaust gas from the heating furnace is obtained through the sensor and recorded as EOd, where 1≤d≤e, which means that the carbon dioxide concentration is collected e times in total, and the collection frequency is set to 3 seconds / time; When the value of carbon dioxide concentration EOd is abnormal, the acquisition is recorded as abnormal acquisition, and the number of all abnormal acquisitions is recorded as c; Determine whether the formula c / e≤γ holds true, where γ is the probability threshold for abnormal collection, which is preset by production personnel based on production experience; If established, no linkage instruction is generated; If not, a linkage instruction is generated to synchronously adjust the feed valve and air intake valve of the heating furnace to reduce the feed amount / feed concentration, or increase the feed amount / intake concentration of oxygen, so that the fuel can be fully burned; When counting the number of abnormal collections, count the number of consecutive abnormalities and record it as f; Determine whether the formula 3*f≤β holds true, where β is the threshold for the duration of continuous abnormalities, which is preset by production personnel based on production experience; If established, no linkage instruction is generated; If not, a linkage instruction is generated; The specific method for determining numerical anomalies is as follows: Determine whether the formula (HLj×PYj)-α≤EOd≤(HLj×PYj)+α holds true, where α is the preset fluctuation value; If it is established, it means that the value of the collected carbon dioxide concentration EOd is normal; If EOd < (HLj × PYj) - α, perform self-test of the corresponding sensor; If there is any sensor abnormality, replace the sensor; If there is no sensor abnormality, it means that the collected carbon dioxide concentration EOd value is abnormal; When generating constraints, the complete combustion of the fuel is monitored and processed synchronously to avoid the problem of incomplete combustion producing harmful substances, thereby increasing the production costs of the enterprise.

[0023] As a third embodiment of the present application, it specifically includes: the path optimization end: obtaining production data corresponding to the production equipment and generating corresponding optimization decisions, the specific method of which is as follows: According to the formula CPi=YLi*YZi, the carbon emissions CPi corresponding to the production equipment are obtained, where YZi is the carbon emission factor corresponding to the production equipment; According to the formula ZH=CP1+CP2+...+CPn, the numerical sum ZH of carbon emissions of all production equipment is obtained; judge Is it true, where θ is the preset value and CLi / YLi is the raw material conversion ratio; If established, maintain normal production; If this is not the case, then production energy conservation and emission reduction decisions are made, such as adjusting production processes, reducing equipment loads, changing raw materials (selecting crude oil with lower carbon intensity), and maintaining equipment; The carbon emission factor YZi is obtained as follows: Obtain the carbon emission factor YZi according to the formula YZi=F(Ti, YLi, CLi, NHi), and establish an online carbon emission factor library to facilitate the rapid acquisition of the carbon emission factor YZi under the same production environment; Where Ti is the reaction temperature corresponding to the production equipment, and the carbon emission factor YZi corresponding to the production equipment is obtained by fitting the nonlinear relationship between real-time parameters and carbon emissions through BP neural network; The specific method is: S1, input layer: receives the real-time values ​​of Ti, YLi, CLi, and NHi; S2, hidden layer: The real-time input value is nonlinearly changed through the activation function, and each layer of neurons learns feature combinations at different levels of abstraction; S3, output layer: generates carbon emission factor prediction value; According to the formula (taking a single hidden layer as an example), the hidden layer outputs h=σ(DA1×YLi+b1), and the output layer predicts y=DA2×h+b2, where σ is the nonlinear activation function, DA1 and DA2 are weight matrices, and b1 and b2 are bias vectors. These parameters are optimized through the back propagation algorithm to make the predicted value y close to the true factor YZi; At the same time, when process parameters are adjusted, such as hydrogen consumption rate and catalyst activity are dynamically updated, the carbon emission factor YZi value is calibrated online synchronously; The model is trained using local data to capture regionally specific nonlinear patterns (e.g., the nonlinear relationship between the carbon emission factor of coal-fired power plants and coal ash content). Furthermore, by accessing real-time monitoring data, the predicted factor values ​​can be dynamically adjusted to address the lag problem of static values ​​in traditional factor libraries. By monitoring the raw material conversion ratio during the production process of production equipment and comparing it with the corresponding carbon emissions CPi of the production equipment, and by generating corresponding decision-making methods, online energy conservation and emission reduction can be achieved while maintaining the company's production and profitability needs.

[0024] While embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions, and variations may be made to these embodiments without departing from the principles and spirit of the invention, and that the scope of the invention is defined by the appended claims and their equivalents.

Claims

1. A carbon footprint monitoring and optimization decision-making platform for the entire refining process, characterized by: The data acquisition terminal is used to obtain real-time data on production equipment involved in the entire refining process; On the cost optimization side, based on the total amount of raw material processing in the production plan, the total amount of heat energy required by the heating furnace is obtained and the amount of fuel used is automatically adjusted; The emission monitoring linkage terminal monitors the carbon dioxide content of the exhaust gas from the heating furnace and generates corresponding linkage instructions when the concentration of carbon dioxide emitted is abnormal. The path optimization end obtains the production data corresponding to the production equipment and generates corresponding optimization decisions.

2. The oil refining full-process carbon footprint monitoring and optimization decision-making platform according to claim 1, characterized in that: The data of production equipment includes energy consumption data NHi, raw material flow YLi, and product flow CLi, where 1≤i≤n, which means there are n production equipment in total, and the value of i represents the specific equipment name.

3. The oil refining full-process carbon footprint monitoring and optimization decision-making platform according to claim 2, characterized in that: In the cost optimization end, the usage of different fuels RLj and their corresponding thermal energy conversion rates RZj are obtained, where 1≤j≤m, indicating that there are m types of fuels in total, and the value of j represents the specific name of the fuel; According to the formula 、 Dynamically adjust RLj and generate constraints Constrain the adjustment value; Where G is the total heat energy corresponding to the total processing volume, Q is the fuel production cost, HBj is the real-time purchase price of the fuel, TP is the carbon emission of the heating furnace in the production plan, HLj is the carbon content percentage of the fuel, and PYj is the percentage of fuel burned in the heating furnace.

4. The oil refining full-process carbon footprint monitoring and optimization decision-making platform according to claim 3 is characterized by: When the price of clean energy drops, its RLj value increases and the RLj value of non-clean energy decreases; when the price of non-clean energy drops, its RLj value increases and the RLj value of clean energy decreases.

5. The oil refining full-process carbon footprint monitoring and optimization decision-making platform according to claim 1, characterized in that: In the emission monitoring linkage terminal, the carbon dioxide concentration EOd in the exhaust gas of the heating furnace is obtained through the sensor, where 1≤d≤e, which means that the carbon dioxide concentration is collected e times, and the collection frequency is 3 seconds / time; When EOd is abnormal, it is recorded as abnormal collection, and the number of abnormal collections c is counted; it is judged whether c / e≤γ is established, where γ is the abnormal collection probability threshold. If not, a linkage instruction is generated.

6. The oil refining full-process carbon footprint monitoring and optimization decision-making platform according to claim 5, characterized in that: Count the number of consecutive exceptions f and determine whether 3×f≤β holds true, where β is the continuous exception duration threshold. If not, generate a linkage instruction.

7. The oil refining full-process carbon footprint monitoring and optimization decision-making platform according to claim 5, characterized in that: Determine whether (HLj×PYj)-α≤EOd≤(HLj×PYj)+α holds, where α is a preset fluctuation value; If EOd<(HLj×PYj)-α, perform sensor self-test. If the sensor is normal, it means that the EOd value is abnormal.

8. The oil refining full-process carbon footprint monitoring and optimization decision-making platform according to claim 1, characterized in that: In the path optimization end, the carbon emissions CPi of the production equipment are obtained according to the formula CPi=YLi×YZi, where YZi is the carbon emission factor corresponding to the production equipment, and the total carbon emissions ZH of all production equipment are obtained according to ZH=CP1+CP2+...+CPn.

9. The oil refining full-process carbon footprint monitoring and optimization decision-making platform according to claim 8, characterized in that: judge Is it true, where θ is the preset value, CLi / YLi is the raw material conversion ratio, if not true, an energy-saving and emission reduction decision is generated.

10. The oil refining full-process carbon footprint monitoring and optimization decision-making platform according to claim 8, characterized in that: The carbon emission factor YZi is obtained by the formula YZi=F(Ti, YLi, CLi, NHi), where Ti is the reaction temperature of the production equipment.

Images