Petroleum coke blending method and system

By constructing a recurrent neural network model and quality index weight algorithm, the problems of quality segmentation and economic benefit evaluation in petroleum coke mix are solved, precise management and market value evaluation of petroleum coke batches are realized, and market competitiveness and economy are enhanced. benefit.

CN119067500APending Publication Date: 2024-12-03河北鸿科碳素有限公司
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202411183981.8
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-08-27
Publication Date
2024-12-03

AI Technical Summary

Technical Problem

The existing petroleum coke dosing methods lack a segmentation mechanism and a comprehensive score weighting mechanism, which makes it difficult for enterprises to accurately locate the batches of petroleum cokes of different quality conditions and evaluate the market value, affecting market segmentation, product differentiation and economic benefits.

Method used

By collecting and integrating the physical and chemical index data of petroleum coke, a recurrent neural network model is constructed to predict batch quality, setting threshold quality conditions, and calculating comprehensive scores using the quality index weight algorithm, and determining the optimal doping combination with cost-benefit analysis.

Benefits of technology

It has realized the precise management and market value evaluation of petroleum coke batches, optimized resource allocation, improved product quality and market competitiveness, and ensured the maximization of economic benefits.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN119067500A_ABST
    Figure CN119067500A_ABST
Patent Text Reader

Abstract

The invention discloses a petroleum coke blending method and system, and relates to the technical field of petroleum processing.The petroleum coke blending method comprises the three core strategies of refined quality management, a scientific decision-making system and economic benefit optimization, and the scientific decision-making system represents a scientific blending plan making system; according to the method, quality management refinement, decision process scientization and economic benefit maximization in the petroleum coke blending process are achieved, the product quality and market competitiveness are improved through the strategies, the operation efficiency and economic benefits of enterprises are enhanced, powerful support is provided for winning advantages of the enterprises in fierce competition, and the economic benefit of the enterprises is improved. And through implementation of the steps of the method, enterprises can better adapt to market changes, customer requirements are met, sustainable development is achieved, and a stable market status and core competitiveness are constructed.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The invention relates to the technical field of petroleum processing, and in particular to a petroleum coke blending method and system. Background Art

[0002] Under the background of global energy structure transformation and refined management, petroleum coke, as an important by-product of the petrochemical industry, is increasingly used in carbon material production, power energy supply, and even metallurgical chemical industry. However, the large fluctuation in quality and complex composition of petroleum coke directly restrict the performance stability and production cost control of downstream products. Therefore, how to blend petroleum coke efficiently and accurately to meet the specific requirements of raw material quality in different application scenarios has become a key link in improving the competitiveness of the entire industrial chain. With the rapid development of big data, artificial intelligence and Internet of Things technologies, the petroleum coke blending industry has also tried to introduce intelligent means to improve product quality and production efficiency. However, even with the use of preliminary intelligent technology, the existing methods still have the following shortcomings: First, in the existing methods, even if the prediction model can give the predicted physical and chemical index values, there is no mechanism to classify different petroleum coke batches according to specific thresholds, which makes it difficult for enterprises to accurately locate petroleum coke batches with different quality conditions, affecting the implementation of market segmentation and product differentiation strategies; Second, in the existing methods, the index values ​​obtained by the prediction model are not fully used as weighted conditions for the comprehensive score, which limits the company's ability to make comprehensive and scientific blending decisions based on data, making it difficult to accurately evaluate the market value and economic benefits of each batch, and affecting the optimal allocation of resources; 3. In the existing methods, when conducting cost-benefit analysis, only direct costs and sales revenue are often considered, while the impact of the comprehensive score on the total benefit is ignored. This leads to companies underestimating the economic benefits of certain high-value batches, affecting the determination of the optimal blending combination, and further affecting the overall economic benefits and market competitiveness; Therefore, the prior art is in urgent need of a technical solution for a petroleum coke blending method and system. Summary of the invention

[0003] In order to solve the above technical problems, the present invention provides a petroleum coke blending method, which is characterized by comprising the following steps: Step S1, collecting and integrating the physical and chemical index data of petroleum coke from different source batches to form an initial data set; Step S2, determining the target petroleum coke according to customer demand, and setting the physical and chemical index range of the target petroleum coke according to the initial data set to establish the blending target of the target petroleum coke; Step S3, predicting petroleum coke batches according to the blending target, and analyzing and integrating the predicted petroleum coke batches to obtain a list of candidate petroleum coke batches; Step S31, preprocessing the initial data set; Step S32, formulating the preprocessed initial data set as a sample set; Step S33: using a random sampling method to divide the sample set according to a predetermined ratio, and obtaining a training set according to the ratio division result; Step S34: construct a prediction model, and use the training set to train the prediction model to obtain a trained prediction model; The prediction model adopts a recurrent neural network model; Step S35, determining the physical and chemical index data of the batch of petroleum coke to be predicted, and inputting the data into the prediction model to obtain the predicted physical and chemical index values ​​of the batch of petroleum coke; Step S36, comparing the predicted physical and chemical index values ​​of the batch of petroleum coke with the blending target of the target petroleum coke. If the predicted physical and chemical index values ​​of the batch of petroleum coke meet the blending target of the target petroleum coke, the petroleum coke batch is output; if the predicted physical and chemical index values ​​of the batch of petroleum coke do not meet the blending target of the target petroleum coke, the process returns to step S35, and the prediction model outputs the predicted physical and chemical index values ​​of the next batch of petroleum coke. Step S37, setting at least two thresholds for the output petroleum coke batch according to customer needs, when the output petroleum coke batch is greater than or equal to one of the thresholds, a petroleum coke batch with quality condition A is obtained; when the output petroleum coke batch is less than one of the thresholds and greater than or equal to another threshold, a petroleum coke batch with quality condition B is obtained; when the output petroleum coke batch is less than another threshold, a petroleum coke batch with quality condition C is obtained; Step S38, integrating petroleum coke batches with different quality conditions to obtain a list of candidate petroleum coke batches; Step S4: applying a quality index weight algorithm to the petroleum coke batches in the candidate batch list to obtain a comprehensive score for the petroleum coke batches; Step S41, determining different quality indicators of petroleum coke batches with different quality conditions in the petroleum coke candidate batch list according to customer needs; Step S42: Evaluate different quality indicators based on the empirical method, and assign weights to the different quality indicators according to the evaluation results; Step S43, preprocessing the predicted physical and chemical index value of each petroleum coke that meets the blending target of the target petroleum coke obtained by the prediction model to obtain a standardized predicted physical and chemical index value of each petroleum coke; Step S44, obtaining a comprehensive score of petroleum coke batches with different quality conditions by weighting different quality indicators and standardizing the predicted physical and chemical index value of each petroleum coke; The calculation formula for the comprehensive score of petroleum coke batches with different quality conditions is: ; in, represents the comprehensive score of the petroleum coke batch of the i-th quality condition; represents the weight of the xth quality index under the i-th quality condition of the petroleum coke batch; represents the predicted physicochemical index value of the i-th standardized petroleum coke; Step S5, performing a cost-benefit analysis on the petroleum coke batches in the candidate batch list according to the comprehensive scores of the petroleum coke batches, and determining the optimal blending combination according to the analysis results; Step S51, respectively obtaining the sales revenue and total cost of petroleum coke batches with different quality conditions, and obtaining the expected sales revenue and expected total cost of petroleum coke batches with different quality conditions; Step S52: Obtaining weight coefficients of the impact of comprehensive scores of petroleum coke batches with different quality conditions on expected total benefits by empirical method; Step S53, obtaining the expected total benefits of the petroleum coke batches with different quality conditions through the expected sales revenue of the petroleum coke batches with different quality conditions, the expected total cost of the petroleum coke batches with different quality conditions, the weight coefficient of the impact of the comprehensive score of the petroleum coke batches with different quality conditions on the expected total benefit, and the comprehensive score of the petroleum coke batches with different quality conditions; The calculation formula for the expected total benefit of petroleum coke batches with different quality conditions is: ; in, represents the expected total benefit of the petroleum coke batch of the i-th quality condition; represents the expected sales revenue of the petroleum coke batch of the i-th quality condition; represents the expected total cost of a petroleum coke batch of the ith quality condition; The weight coefficient representing the impact of the comprehensive score of the petroleum coke batch of the i-th quality condition on the expected total benefit; represents the comprehensive score of the petroleum coke batch of the i-th quality condition; The total number of petroleum coke batches representing different quality conditions; Step S54, formulating optimal blending combinations for the petroleum coke batches with different quality conditions according to the expected total benefits of the petroleum coke batches with different quality conditions; Step S6, formulating a blending plan according to the optimal blending combination; Step S7, executing the blending plan and monitoring the blending process; A petroleum coke blending system, using the petroleum coke blending method as described in any one of the above items, comprises the following parts: Data collection module: used to collect and integrate the physical and chemical index data of petroleum coke from different sources to form an initial data set; Blending target establishment module: connected to the data collection module, used to determine the target petroleum coke according to customer needs, and set the physical and chemical index range of the target petroleum coke according to the initial data set to establish the blending target of the target petroleum coke; A petroleum coke candidate batch list acquisition module: connected to the blending target establishment module, used to predict the petroleum coke batches according to the blending target, and analyze and integrate the predicted petroleum coke batches to obtain a petroleum coke candidate batch list; A petroleum coke batch comprehensive score acquisition module: connected to the petroleum coke candidate batch list acquisition module, used to apply a quality index weight algorithm to the petroleum coke batches in the candidate batch list to obtain a comprehensive score for the petroleum coke batches; An optimal blending combination determination module is connected to the petroleum coke batch comprehensive score acquisition module, and is used to perform a cost-benefit analysis on the petroleum coke batches in the candidate batch list with the comprehensive scores, and determine the optimal blending combination according to the analysis results; A blending plan formulation module: connected to the optimal blending combination determination module, used to formulate a blending plan according to the optimal blending combination; Blending plan execution module: connected with the blending plan formulation, used to execute the blending plan and monitor the blending process.

[0004] The embodiments of the present invention have the following technical effects: The present invention first obtains the predicted physical and chemical index value of each petroleum coke that meets the blending target of the target petroleum coke through a prediction model, thereby obtaining different petroleum coke batches that meet the blending target of the target petroleum coke, and then judges the different petroleum coke batches according to at least two thresholds based on customer needs to obtain petroleum coke batches with different quality conditions. This method subdivides the petroleum coke batches into different quality grades by setting different thresholds. This hierarchical management not only helps internal quality control and ensures that each batch can meet the corresponding quality standards, but also facilitates enterprises to carry out targeted product development and sales according to the specific needs of customers, and petroleum coke batches with different quality conditions can meet the needs of different customer groups in the market. In this way, enterprises can expand market coverage and enhance market adaptability and competitiveness by providing diversified petroleum coke products. At the same time, through the division of quality grades, enterprises can better identify and control product quality risks and avoid market reputation losses caused by low-quality batches. At the same time, this hierarchical management also helps to optimize production costs, thereby controlling production costs while ensuring quality; The index values ​​obtained by the prediction model are then used as weighted conditions for the comprehensive scores of petroleum coke batches with different quality conditions. This method constructs a comprehensive evaluation system covering multiple dimensions such as product quality and market acceptance. This system ensures that the subsequent blending plan not only meets the basic physical and chemical index requirements, but also fully considers the actual needs of customers, improves the comprehensiveness and scientificity of blending decisions, and can more accurately evaluate the value of different petroleum coke batches for enterprises, thereby optimizing resource allocation, giving priority to the production of batches with high comprehensive scores, great market potential, and good economic benefits, and improving overall operational efficiency and economic benefits; Finally, the petroleum coke batches with different quality conditions and comprehensive scores are combined with the calculation process of the expected total benefits of the existing petroleum coke batches. This method combines the comprehensive score with the expected total benefits, so that the enterprise can pursue the maximization of economic benefits on the basis of ensuring product quality. This combination method prompts the enterprise to pay more attention to the consideration of economic benefits, realize the dual optimization of cost control and quality assurance, and can prompt the enterprise to pay more attention to the matching of resources and capabilities, that is, in the case of limited resources, give priority to those projects that can generate higher benefits, realize the optimal allocation of resources, and improve the overall operational efficiency and market competitiveness of the enterprise. At the same time, this method also provides senior managers with a more scientific basis for the formulation of blending plans, helping the enterprise leadership to make more accurate strategic decisions in a complex and changing market environment, that is, to formulate more accurate blending plans, reduce the risks of blending plan formulation, and improve the quality and efficiency of blending plan formulation; In summary, the present invention provides a petroleum coke blending method and system, which realizes the refinement of quality management, scientific decision-making system and economic benefit optimization through three core strategies, among which the scientific decision-making system represents a scientific blending plan formulation system, and realizes the refinement of quality management, scientific decision-making process and maximization of economic benefits in the petroleum coke blending process. These strategies not only improve product quality and market competitiveness, but also enhance the operational efficiency and economic benefits of the enterprise, and provide strong support for the enterprise to gain advantages in fierce competition. Moreover, through the implementation of the method steps of the present invention, the enterprise can better adapt to market changes, meet customer needs, achieve sustainable development, and build a solid market position and core competitiveness. BRIEF DESCRIPTION OF THE DRAWINGS

[0005] In order to more clearly illustrate the specific implementation methods of the present invention or the technical solutions in the prior art, the drawings required for use in the specific implementation methods or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are some implementation methods of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0006] Figure 1is a flow chart of a petroleum coke blending method provided by an embodiment of the present invention; Figure 2 It is a framework diagram of the petroleum coke blending method provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0007] In order to make the purpose, technical solution and advantages of the present invention clearer, the technical solution of the present invention will be described clearly and completely below. 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 creative work belong to the scope of protection of the present invention.

[0008] Embodiment 1: Figure 1 As shown, the petroleum coke blending method provided by the present invention comprises the following steps: Step S1, collecting and integrating the physical and chemical index data of petroleum coke from different source batches to form an initial data set; For example, the physical and chemical index data of petroleum coke from different source batches may include, but are not limited to, the following sources and specific collection methods of the physical and chemical index data of petroleum coke from the following source batches: First, we collected data from major refineries across the country, which process a variety of crude oil types. 2. Obtain petroleum coke samples from suppliers around the world; 3. Physical and chemical index reports of petroleum coke obtained from professional third-party testing agencies. These reports usually contain more detailed analysis results; 4. Extract the physical and chemical index data of each batch of petroleum coke from historical transaction records, including the test reports and actual usage feedback provided by both parties to the transaction, to ensure the comprehensiveness and reliability of the data; 5. Feedback information from end users after using petroleum coke, including actual combustion efficiency, emissions, etc. Although this information does not directly reflect physical and chemical indicators, it can indirectly verify the accuracy and applicability of the data; Collect petroleum coke samples directly from the above sources and send them to professional laboratories for physical and chemical index testing to ensure the real-time and accuracy of the data; consult official reports from refineries and testing agencies, as well as historical transaction records, and extract relevant data from them; use authoritative petroleum coke database resources at home and abroad, such as data released by the International Energy Agency and other institutions, as well as experimental data in industry journals and academic papers, for comprehensive compilation; collect feedback from end users on the use of petroleum coke through telephone interviews, online questionnaires, etc., to understand the physical and chemical performance in actual applications as supplementary information for the data set; After completing the above collection and integration process, we entered all the data into a central database to form an initial data set covering the physical and chemical indicators of petroleum coke from batches of different sources. This data set not only includes basic physical and chemical indicators, but also integrates feedback from actual application effects, providing comprehensive and reliable data support for subsequent prediction model training and blending scheme design.

[0009] Step S2, determining the target petroleum coke according to customer demand, and setting the physical and chemical index range of the target petroleum coke according to the initial data set to establish the blending target of the target petroleum coke; For example, after successfully collecting and integrating the physical and chemical index data of petroleum coke batches from different sources to form an initial data set, the next key step is to determine the target petroleum coke based on specific customer needs and set the range of its physical and chemical indexes based on this. This process is crucial for developing an effective petroleum coke blending strategy; Specifically, first, communicate in depth with customers to understand their specific needs, which may include but are not limited to: Intended use: For which industrial process or product production does the customer plan to use petroleum coke, such as power generation, metallurgy, chemical raw materials, etc. Performance requirements: The customer's specific requirements for the calorific value, ash content, sulfur content, volatile matter, true density and other physical and chemical indicators of petroleum coke; Environmental standards: Emission standard restrictions in the customer's region or industry, such as emission limits of sulfur oxides and nitrogen oxides; Cost considerations: The customer's budget range, and the trade-off preference between the price and performance of petroleum coke; Based on the results of the customer demand analysis, petroleum coke batches that meet the preliminary conditions are screened out from the initial data set, and then the physical and chemical index range of the target petroleum coke is further refined. The specific steps include: Use data query and screening techniques to identify the petroleum coke batch that is closest to the customer's performance requirements from the initial data set, which involves multi-dimensional comparison and sorting to ensure that the selected petroleum coke can meet or exceed customer expectations; Consult with customers to jointly determine a more precise target indicator range based on the physical and chemical indicator data of the optional petroleum coke batches. For example, if the customer is concerned about low sulfur emissions, the upper limit of sulfur content may need to be strictly limited; if the customer attaches importance to calorific value, a minimum calorific value standard needs to be set; Once the target range of physical and chemical indicators of petroleum coke is determined, the next step is to determine how to achieve these targets by blending different batches of petroleum coke, which involves the application of mathematical optimization algorithms to find the best blending ratio that can meet all physical and chemical indicators and minimize costs or maximize benefits; After completing the above steps, we will have a detailed list of target petroleum coke blending goals, including: Target petroleum coke physical and chemical index range: clearly defined minimum and maximum values ​​of key indicators such as calorific value, ash content, sulfur content, volatile matter, true density, etc.; Overview of blending targets: List the different batches of petroleum coke involved in the blending and their respective proportions, as well as the comprehensive physical and chemical indicators expected to be achieved.

[0010] Step S3, predicting petroleum coke batches according to the blending target, and analyzing and integrating the predicted petroleum coke batches to obtain a list of candidate petroleum coke batches; After clarifying the range of physical and chemical indicators of the target petroleum coke, the next step is to select qualified petroleum coke batches from existing petroleum coke batches by building and applying a prediction model. This process is divided into multiple sub-steps, covering key links such as data preprocessing, model training, prediction analysis, and result integration; Step S31, preprocessing the initial data set; Preprocess the initial data set, including but not limited to data cleaning and feature engineering; Data cleaning: removing or correcting missing values, outliers and duplicate records in the data set to ensure data quality; Feature engineering: according to the prediction requirements, the original data needs to be transformed, normalized or new feature variables need to be created to improve the prediction accuracy of the model; Step S32, formulating the preprocessed initial data set as a sample set; Construct a sample set and convert the preprocessed initial data set into a format acceptable to the machine learning model, usually including feature vectors and corresponding labels; Step S33: using a random sampling method to divide the sample set according to a predetermined ratio, and obtaining a training set according to the ratio division result; Data partitioning includes using random sampling to allocate the sample set to the training set and the test set according to common predetermined proportions, such as 80% and 20%, respectively; Among them, 80% of the sample sets were selected as training sets to ensure that the training sets contained data with sufficient diversity so that the model could learn the broad patterns of the physical and chemical indicators of petroleum coke; Step S34: construct a prediction model, and use the training set to train the prediction model to obtain a trained prediction model; A recurrent neural network model is selected as the prediction model, and the prediction model parameters are fitted using the above training set. Optionally, 20% of the sample set can be used as a test set to test the trained prediction model, that is, through an iterative optimization process, the prediction model can accurately predict the physical and chemical index values ​​of petroleum coke. The specific reason for only obtaining 80% of the sample set as the training set is that if the subsequent prediction model uses the entire sample set for training, it will match these specific data points too closely, including the noise and anomalies therein, instead of learning the general rules behind the data. In this way, the prediction ability of the prediction model will be greatly reduced when facing new and unseen data. By retaining a part of the sample set as a validation set or test set, the generalization ability of the model can be evaluated, that is, the performance on data that has not participated in the training, so as to determine whether the prediction model is overfitting or underfitting, and make corresponding adjustments accordingly; Step S35, determining the physical and chemical index data of the batch of petroleum coke to be predicted, and inputting the data into the prediction model to obtain the predicted physical and chemical index values ​​of the batch of petroleum coke; The physical and chemical index data of the batch of petroleum coke to be predicted are input into the prediction model in the same format as that used in training. At this time, the prediction model will predict the predicted physical and chemical index values ​​of the batch of petroleum coke based on the input data; Step S36, comparing the predicted physical and chemical index values ​​of the batch of petroleum coke with the blending target of the target petroleum coke. If the predicted physical and chemical index values ​​of the batch of petroleum coke meet the blending target of the target petroleum coke, the petroleum coke batch is output; if the predicted physical and chemical index values ​​of the batch of petroleum coke do not meet the blending target of the target petroleum coke, the process returns to step S35, and the prediction model outputs the predicted physical and chemical index values ​​of the next batch of petroleum coke. Next, a comparative analysis is performed to check whether the predicted physical and chemical index values ​​of the batch of petroleum coke meet the blending target set in step S2. If so, the batch is output; if not, the prediction model continues to process the next batch of data until all batches are evaluated; Step S37, setting at least two thresholds for the output petroleum coke batch according to customer needs, when the output petroleum coke batch is greater than or equal to one of the thresholds, a petroleum coke batch with quality condition A is obtained; when the output petroleum coke batch is less than one of the thresholds and greater than or equal to another threshold, a petroleum coke batch with quality condition B is obtained; when the output petroleum coke batch is less than another threshold, a petroleum coke batch with quality condition C is obtained; At this time, different quality grade thresholds are set according to the customer's sensitivity to the quality of the petroleum coke batches, such as grade A, grade B, and grade C, but not limited to the above-mentioned quality grade thresholds, and the output petroleum coke batches are divided into different quality grades according to the relationship between the output petroleum coke batches and the thresholds; Step S38, integrating petroleum coke batches with different quality conditions to obtain a list of candidate petroleum coke batches; All quality-graded petroleum coke batches are aggregated to form a comprehensive list of candidate petroleum coke batches, which may include: Batch ID: a number used to uniquely identify each petroleum coke batch; predicted physical and chemical index values: including predicted values ​​of key indicators such as calorific value, ash, sulfur content, volatile matter, true density, etc.; Quality grade: the quality level of the petroleum coke batch determined by the threshold, such as grade A, B, and C; and even the degree of recommendation: a blending recommendation index given based on the confidence of the prediction model and the matching degree of customer needs.

[0011] Step S4, applying a quality index weight algorithm to the petroleum coke batches with different quality conditions in the candidate batch list to obtain a comprehensive score of the petroleum coke batches with different quality conditions; After obtaining the list of candidate petroleum coke batches, the next step is to calculate the comprehensive score for each batch by introducing a quality index weighting algorithm, so as to achieve quantitative comparison and optimization of batches with different quality conditions. This process carefully considers the importance of various physical and chemical indicators and the differences between batches to ensure that the petroleum coke batch finally selected best meets the customer's specific needs; Step S41, determining different quality indicators of petroleum coke batches with different quality conditions in the petroleum coke candidate batch list according to customer needs; Communicate with customers to clarify their preferences and priorities for the physical and chemical indicators of petroleum coke, such as calorific value, ash content, sulfur content, volatile matter, true density, etc. Based on customer feedback, identify several key quality indicators that have the greatest impact on blending effects; Step S42: Evaluate different quality indicators based on the empirical method, and assign weights to the different quality indicators according to the evaluation results; Invite industry experts to evaluate the importance of the identified key quality indicators, taking into account their impact on the final product quality and processing costs. Based on the expert evaluation results, assign a weight value between 0 and 1 to each quality indicator, ensuring that the sum of all indicator weights is 1; Step S43, preprocessing the predicted physical and chemical index value of each petroleum coke that meets the blending target of the target petroleum coke obtained by the prediction model to obtain a standardized predicted physical and chemical index value of each petroleum coke; Use z-score standardization or other applicable methods to convert the predicted physical and chemical index values ​​into dimensionless standard scores to eliminate the impact of dimension differences and ensure that all indicators are converted to values ​​on the same scale to facilitate subsequent weighted calculations; Step S44, obtaining a comprehensive score of petroleum coke batches with different quality conditions by weighting different quality indicators and standardizing the predicted physical and chemical index value of each petroleum coke; For petroleum coke batches with different quality conditions, the standardized predicted physical and chemical index values ​​are multiplied by the weights of the corresponding quality indicators, and then summed to obtain the comprehensive score of the batch; The calculation formula for the comprehensive score of petroleum coke batches with different quality conditions is: ; in, represents the comprehensive score of the petroleum coke batch of the i-th quality condition; represents the weight of the xth quality index under the i-th quality condition of the petroleum coke batch; Represents the predicted physicochemical index value of the ith standardized petroleum coke.

[0012] Step S5, performing a cost-benefit analysis on the petroleum coke batches in the candidate batch list according to the comprehensive scores of the petroleum coke batches, and determining the optimal blending combination according to the analysis results; After completing the comprehensive scoring of the petroleum coke batches, the next step is to focus on cost-benefit analysis, aiming to further screen and optimize the blending combination of petroleum coke from the perspective of economic benefits, ensuring that the final selected petroleum coke batches not only meet the quality requirements but also maximize the economic benefits of the enterprise; Step S51, obtaining the sales revenue and total cost of petroleum coke batches with different quality conditions through market research, and obtaining the expected sales revenue and expected total cost of petroleum coke batches with different quality conditions; Conduct market research: collect information on the sales price, sales volume and production costs of similar petroleum coke products in the market, including raw material costs, processing costs, transportation costs, etc.; Financial forecast: Based on market research data, combined with the company's own production capacity, sales channels and other factors, forecast the possible sales price and sales volume of petroleum coke batches with different quality conditions, and then estimate the expected sales revenue; Cost estimation: Consider the cost expenditure of each link such as procurement, production, logistics, etc., and calculate the expected total cost of each batch; Step S52: Obtaining weight coefficients of the impact of comprehensive scores of petroleum coke batches with different quality conditions on expected total benefits by empirical method; Based on historical data and industry experience, analyze how petroleum coke batches with different quality conditions affect sales prices, market demand and customer satisfaction, thereby indirectly affecting the company's total revenue. Then, through expert judgment or regression analysis, determine the weight coefficient of the relationship between the comprehensive score of petroleum coke batches with different quality conditions and the expected total benefit, reflecting the contribution of quality to economic benefits; Step S53, obtaining the expected total benefits of the petroleum coke batches with different quality conditions through the expected sales revenue of the petroleum coke batches with different quality conditions, the expected total cost of the petroleum coke batches with different quality conditions, the weight coefficient of the impact of the comprehensive score of the petroleum coke batches with different quality conditions on the expected total benefit, and the comprehensive score of the petroleum coke batches with different quality conditions; The expected total benefit of each petroleum coke batch is calculated using the formula: expected total benefit = expected sales revenue − expected total cost + (comprehensive score × weight coefficient), where the weight coefficient of the comprehensive score reflects the positive effect of quality on economic benefits. The calculation formula for the expected total benefit of petroleum coke batches with different quality conditions is: ; in, represents the expected total benefit of the petroleum coke batch of the i-th quality condition; represents the expected sales revenue of the petroleum coke batch of the i-th quality condition; represents the expected total cost of a petroleum coke batch of the ith quality condition; The weight coefficient representing the impact of the comprehensive score of the petroleum coke batch of the i-th quality condition on the expected total benefit; represents the comprehensive score of the petroleum coke batch of the i-th quality condition; The total number of petroleum coke batches representing different quality conditions; It is worth noting that adjusting the size of α is actually adjusting the relative importance between quality and cost-effectiveness. If an enterprise believes that high-quality products are the core of its market competitiveness, it may tend to assign a higher value to α, which means that when calculating the expected total benefit, the contribution of the comprehensive score will be magnified. Even if the cost of some batches is higher, if their comprehensive score is high, they may be regarded as a better choice. On the contrary, if the enterprise currently pays more attention to cost control and economic benefits, it may reduce the value of α, so that the cost factor occupies a larger proportion in the total benefit calculation. Among them, the adjustment of α is usually based on the following considerations: the enterprise's understanding of market positioning, For example, whether to pursue the high-end market or focus more on cost advantage; the sensitivity of the target customer group to product quality and the price they are willing to pay for it; the competitive situation in the industry and the relative importance of quality and cost in industry competition; considerations on production cost fluctuations, supply chain stability, and how to balance quality and cost risks; Therefore, in actual operations, enterprises can adjust the value of α in a timely manner according to their own strategic goals and changes in the market environment to ensure that the calculation of expected total benefits can more accurately reflect their business goals and market strategies. This flexibility helps enterprises make decisions that are more in line with their own interests under changing market conditions; Step S54, formulating optimal blending combinations for the petroleum coke batches with different quality conditions according to the expected total benefits of the petroleum coke batches with different quality conditions; An optional implementation method is to sort all petroleum coke batches of different quality conditions according to the calculated expected total benefit. The principle of sorting is very direct, that is, the batches are arranged in descending order according to the numerical value of the expected total benefit of each batch. Batches with higher expected total benefits will be placed in front, and batches with lower benefits will be placed in the back. The top batches with the highest benefits are identified, and then based on the benefit ranking results, considering the complementarity and cost structure between different batches, a variety of blending combination schemes are designed, and the comprehensive benefit of each combination is evaluated. Finally, the blending combination with the best benefit, controllable risk and reasonable cost is selected as the optimal blending combination. At the same time, it is worth considering the stability of the supply chain and the adaptability to market changes.

[0013] Step S6, formulating a blending plan according to the optimal blending combination; After determining the optimal blending combination, the next step is to focus on developing a detailed blending plan to ensure that the optimal blending combination can be effectively implemented in actual production. This step is the key link in transforming theoretical analysis into practical action, and is crucial for ensuring product quality, controlling costs and improving efficiency. Specifically, the optional implementation methods are as follows: Blending plan formulation: 1. Bill of Materials: Based on the optimal blending combination, list all required petroleum coke batches and their proportions, and clarify the quantity requirements for each batch; 2. Resource Allocation: Taking into account factors such as the availability of production facilities, human resources, and storage space, reasonably arrange the various resources required for blending; 3. Schedule Planning: Develop a detailed production schedule, including time nodes for each stage such as pretreatment, blending, and inspection to ensure that the entire process is smooth and efficient; 4. Quality Control Standards: Define quality inspection standards for the blending process and finished products, including but not limited to key indicators such as ash content, sulfur content, and volatile matter, to ensure that the final product meets the established quality requirements; 5. Emergency Plan: Prepare corresponding response measures for possible unexpected situations, such as raw material shortages, equipment failures, etc., to reduce the risk of production interruptions.

[0014] Step S7, executing the blending plan and monitoring the blending process; Entering step S7, the enterprise begins to implement the previously carefully formulated blending plan, and closely monitors the entire blending process to ensure that production activities proceed smoothly according to plan and that possible problems are discovered and resolved in a timely manner; Specifically, the optional implementation modes are as follows: Plan Execution: 1. Material preparation: prepare the required batches of petroleum coke in advance according to the plan to ensure that all raw materials are ready when blending begins; 2. Equipment inspection: conduct a comprehensive inspection of the blending equipment to confirm that it is in good condition to avoid production delays due to equipment failure; 3. Production start-up: start the blending operation according to the schedule and strictly follow the process flow and operating specifications; Process monitoring: 1. Real-time monitoring: Use a combination of automated systems and manual inspections to monitor various parameters in the blending process in real time, such as temperature, pressure, flow, etc., to ensure that operating conditions meet the requirements; 2. Quality inspection: Perform quality inspections at key nodes, check whether the blended materials meet the requirements against the predetermined standards, and make adjustments if necessary; 3. Abnormal response: Once any deviation from the plan is found, take corresponding measures immediately, such as adjusting the ingredient ratio, correcting process parameters or suspending production for investigation.

[0015] Embodiment 2: Figure 2 As shown, the present invention also proposes a petroleum coke blending system, which executes the above-mentioned petroleum coke blending method, including the following modules: Data collection module: used to collect and integrate the physical and chemical index data of petroleum coke from different sources to form an initial data set; Blending target establishment module: connected to the data collection module, used to determine the target petroleum coke according to customer needs, and set the physical and chemical index range of the target petroleum coke according to the initial data set to establish the blending target of the target petroleum coke; A petroleum coke candidate batch list acquisition module: connected to the blending target establishment module, used to predict the petroleum coke batches according to the blending target, and analyze and integrate the predicted petroleum coke batches to obtain a petroleum coke candidate batch list; A petroleum coke batch comprehensive score acquisition module: connected to the petroleum coke candidate batch list acquisition module, used to apply a quality index weight algorithm to the petroleum coke batches in the candidate batch list to obtain a comprehensive score for the petroleum coke batches; An optimal blending combination determination module is connected to the petroleum coke batch comprehensive score acquisition module, and is used to perform a cost-benefit analysis on the petroleum coke batches in the candidate batch list with the comprehensive scores, and determine the optimal blending combination according to the analysis results; A blending plan formulation module: connected to the optimal blending combination determination module, used to formulate a blending plan according to the optimal blending combination; Blending plan execution module: connected with the blending plan formulation, used to execute the blending plan and monitor the blending process.

[0016] It should be noted that the terms used in the present invention are only for describing specific embodiments, rather than limiting the scope of the present application. As shown in the present specification, unless the context clearly indicates an exception, the words "one", "a", "a kind of" and / or "the" do not specifically refer to the singular, but may also include the plural. The terms "include", "comprise" or any other variant thereof are intended to cover non-exclusive inclusion, so that the process, method or device including a series of elements includes not only those elements, but also includes other elements not explicitly listed, or also includes elements inherent to such process, method or device. In the absence of more restrictions, the elements defined by the sentence "include one..." do not exclude the presence of other identical elements in the process, method or device including the elements.

[0017] It should also be noted that the terms "center", "up", "down", "left", "right", "vertical", "horizontal", "inside", "outside", etc., indicating the orientation or positional relationship, are based on the orientation or positional relationship shown in the drawings, and are only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore cannot be understood as a limitation on the present invention. Unless otherwise clearly specified and limited, the terms "installed", "connected", "connected", etc. should be understood in a broad sense, for example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be a direct connection, or it can be an indirect connection through an intermediate medium, or it can be a connection between the two elements. For those of ordinary skill in the art, the specific meanings of the above terms in the present invention can be understood according to specific circumstances.

[0018] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or replace some or all of the technical features therein by equivalents. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the technical solutions of the embodiments of the present invention.

Claims

1. A petroleum coke blending method, characterized in that: The following steps are involved: Step S1, collecting and integrating the physical and chemical index data of petroleum coke from different source batches to form an initial data set; Step S2, determining the target petroleum coke according to customer demand, and setting the physical and chemical index range of the target petroleum coke according to the initial data set to establish the blending target of the target petroleum coke; Step S3, predicting petroleum coke batches according to the blending target, and analyzing and integrating the predicted petroleum coke batches to obtain a list of candidate petroleum coke batches; Step S4: applying a quality index weight algorithm to the petroleum coke batches in the candidate batch list to obtain a comprehensive score for the petroleum coke batches; Step S5, performing a cost-benefit analysis on the petroleum coke batches in the candidate batch list according to the comprehensive scores of the petroleum coke batches, and determining the optimal blending combination according to the analysis results; Step S6, formulating a blending plan according to the optimal blending combination; Step S7: Execute the blending plan and monitor the blending process.

2. The petroleum coke blending method according to claim 1, characterized in that: The step S3 predicts petroleum coke batches according to the blending target, and analyzes and integrates the predicted petroleum coke batches to obtain a list of candidate petroleum coke batches including: Step S31, preprocessing the initial data set; Step S32, formulating the preprocessed initial data set as a sample set; Step S33: using a random sampling method to divide the sample set according to a predetermined ratio, and obtaining a training set according to the ratio division result; Step S34: construct a prediction model, and use the training set to train the prediction model to obtain a trained prediction model; Step S35, determining the physical and chemical index data of the batch of petroleum coke to be predicted, and inputting the data into the prediction model to obtain the predicted physical and chemical index values ​​of the batch of petroleum coke; Step S36, comparing the predicted physical and chemical index values ​​of the batch of petroleum coke with the blending target of the target petroleum coke. If the predicted physical and chemical index values ​​of the batch of petroleum coke meet the blending target of the target petroleum coke, the petroleum coke batch is output; if the predicted physical and chemical index values ​​of the batch of petroleum coke do not meet the blending target of the target petroleum coke, the process returns to step S35, and the predicted physical and chemical index values ​​of the next batch of petroleum coke are output by the prediction model; Step S37, setting at least two thresholds for the output petroleum coke batch according to customer needs, when the output petroleum coke batch is greater than or equal to one of the thresholds, a petroleum coke batch with quality condition A is obtained; when the output petroleum coke batch is less than one of the thresholds and greater than or equal to another threshold, a petroleum coke batch with quality condition B is obtained; when the output petroleum coke batch is less than another threshold, a petroleum coke batch with quality condition C is obtained; Step S38: Integrate petroleum coke batches with different quality conditions to obtain a list of candidate petroleum coke batches.

3. The petroleum coke blending method according to claim 1, characterized in that: The step S4, applying a quality index weight algorithm to the petroleum coke batches with different quality conditions in the candidate batch list, to obtain a comprehensive score of the petroleum coke batches with different quality conditions includes: Step S41, determining different quality indicators of petroleum coke batches with different quality conditions in the petroleum coke candidate batch list according to customer needs; Step S42: Evaluate different quality indicators based on the empirical method, and assign weights to the different quality indicators according to the evaluation results; Step S43, preprocessing the predicted physical and chemical index value of each petroleum coke that meets the blending target of the target petroleum coke obtained by the prediction model to obtain a standardized predicted physical and chemical index value of each petroleum coke; Step S44, obtaining the comprehensive scores of petroleum coke batches with different quality conditions by weighting different quality indicators and standardizing the predicted physical and chemical indicator values ​​of each petroleum coke.

4. The petroleum coke blending method according to claim 3, characterized in that: The calculation formula for the comprehensive score of petroleum coke batches with different quality conditions is: ; in, represents the comprehensive score of the petroleum coke batch of the i-th quality condition; represents the weight of the xth quality index under the i-th quality condition of the petroleum coke batch; Represents the predicted physicochemical index value of the ith standardized petroleum coke.

5. The petroleum coke blending method according to claim 1, characterized in that: The step S5, performing a cost-benefit analysis on the petroleum coke batches with different quality conditions in the candidate batch list with comprehensive scores, and determining the optimal blending combination according to the analysis results includes: Step S51, respectively obtaining the sales revenue and total cost of petroleum coke batches with different quality conditions, and obtaining the expected sales revenue and expected total cost of petroleum coke batches with different quality conditions; Step S52: Obtaining weight coefficients of the impact of comprehensive scores of petroleum coke batches with different quality conditions on expected total benefits by empirical method; Step S53, obtaining the expected total benefits of the petroleum coke batches with different quality conditions through the expected sales revenue of the petroleum coke batches with different quality conditions, the expected total cost of the petroleum coke batches with different quality conditions, the weight coefficient of the impact of the comprehensive score of the petroleum coke batches with different quality conditions on the expected total benefit, and the comprehensive score of the petroleum coke batches with different quality conditions; Step S54: formulate optimal blending combinations for the petroleum coke batches with different quality conditions according to the expected total benefits of the petroleum coke batches with different quality conditions.

6. The petroleum coke blending method according to claim 5, characterized in that: The calculation formula for the expected total benefit of petroleum coke batches with different quality conditions is: ; in, represents the expected total benefit of the petroleum coke batch of the i-th quality condition; represents the expected sales revenue of the petroleum coke batch of the ith quality condition; represents the expected total cost of a petroleum coke batch of the ith quality condition; The weight coefficient representing the impact of the comprehensive score of the petroleum coke batch of the i-th quality condition on the expected total benefit; represents the comprehensive score of the petroleum coke batch of the i-th quality condition; Total number of petroleum coke batches representing different quality conditions.

7. The petroleum coke blending method according to claim 2, characterized in that: The prediction model adopts a recurrent neural network model.

8. A petroleum coke blending system, used to implement the petroleum coke blending method according to any one of claims 1 to 7, characterized in that: The system comprises: Data collection module: used to collect and integrate the physical and chemical index data of petroleum coke from different sources to form an initial data set; Blending target establishment module: connected to the data collection module, used to determine the target petroleum coke according to customer needs, and set the physical and chemical index range of the target petroleum coke according to the initial data set to establish the blending target of the target petroleum coke; A petroleum coke candidate batch list acquisition module: connected to the blending target establishment module, used to predict the petroleum coke batches according to the blending target, and analyze and integrate the predicted petroleum coke batches to obtain a petroleum coke candidate batch list; A petroleum coke batch comprehensive score acquisition module: connected to the petroleum coke candidate batch list acquisition module, used to apply a quality index weight algorithm to the petroleum coke batches in the candidate batch list to obtain a comprehensive score for the petroleum coke batches; An optimal blending combination determination module is connected to the petroleum coke batch comprehensive score acquisition module, and is used to perform a cost-benefit analysis on the petroleum coke batches in the candidate batch list with the comprehensive scores, and determine the optimal blending combination according to the analysis results; A blending plan formulation module: connected to the optimal blending combination determination module, used to formulate a blending plan according to the optimal blending combination; Blending plan execution module: connected with the blending plan formulation, used to execute the blending plan and monitor the blending process.