Petrochemical device multi-level production plan collaboration method and system
By constructing a dynamic and flexible capacity model and collaborative optimization, the problems of resource waste and slow feedback caused by the fixed capacity assumption in traditional production planning have been solved. Dynamic collaboration and closed-loop optimization of petrochemical plant production planning have been achieved, improving resource utilization efficiency and the stability of the planning system.
Patent Information
- Application Number
- CN202511358320.9
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-09-23
- Publication Date
- 2025-12-30
AI Technical Summary
Traditional production planning methods are based on fixed capacity, ignoring the impact of actual operating conditions on capacity. This leads to overly conservative planning or failure to fully utilize capacity, and the feedback mechanism is slow to respond and difficult to adjust quickly, resulting in serious waste of resources.
A dynamic and flexible production capacity model is constructed. Based on the historical operating data and current operating parameters of petrochemical plants, an adjustable production capacity boundary is output and used as a constraint in long-term, medium-term and short-term production plans. By calculating the synergy index and introducing a synergy penalty term, the load allocation at each level is optimized to achieve synergy and flexibility in multi-level production plans.
It has achieved a precise depiction of the actual processing capacity of petrochemical plants, improved the overall consistency and flexibility of production planning, enhanced resource utilization efficiency and operational economy, and has the ability to adapt to continuous disturbances, realizing the transformation from static extensive to dynamic collaborative.
Smart Images
Figure CN121235489A_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of production planning coordination methods, and in particular to a multi-level production planning coordination method and system for petrochemical plants. Background Technology
[0002] Production planning coordination methods are a technique that constructs dynamic, flexible capacity models and applies them to long-term, medium-term, and short-term production plans to achieve close coordination among multi-level plans. Therefore, how to utilize advanced technologies to improve the intelligence and security of production planning coordination methods has become one of the most pressing issues to be addressed.
[0003] In the field of production planning coordination methods, traditional production planning methods are usually based on the fixed capacity of the equipment, ignoring the impact of changes in actual operating conditions on capacity. This can easily lead to overly conservative planning or failure to fully utilize capacity when operating conditions improve, resulting in resource waste. Furthermore, traditional feedback mechanisms are slow to respond to deviations between actual load and planned values during the production process, making it difficult to make reasonable adjustments quickly. Summary of the Invention
[0004] In view of the aforementioned existing problems, the present invention is proposed.
[0005] Therefore, this invention provides a multi-level production planning coordination method for petrochemical plants to solve the problems of traditional production planning methods, which are usually based on the fixed capacity of the plant and ignore the impact of changes in actual operating conditions on capacity. This can easily lead to overly conservative planning or failure to fully utilize capacity when operating conditions improve, resulting in resource waste. Furthermore, traditional feedback mechanisms are slow to respond to deviations between actual load and planned values during production, making it difficult to make reasonable adjustments quickly.
[0006] To solve the above-mentioned technical problems, the present invention provides the following technical solution: In a first aspect, the present invention provides a multi-level production planning coordination method for petrochemical plants, comprising: Based on historical operating data and current operating parameters of petrochemical plants, a dynamic elastic capacity model is constructed to output the adjustable capacity boundary of petrochemical plants. The adjustable capacity boundary is used as a constraint input into long-term, medium-term and short-term production plans, and the load plans at each level are output. Calculate the load fluctuation differences between load plans at different levels and output the coordination index; The synergy index is compared with a preset threshold. If the threshold is not met, synergy constraints are introduced into the multi-objective optimization, the load allocation at each level is readjusted, and a production plan is output. The production plan is executed, actual load data is collected, the deviation between the plan and the actual load is calculated, the execution deviation result is output, and it is determined whether the deviation exceeds the allowable range based on the execution deviation result. If it does, the corresponding correction amount is generated and fed back to the medium-term planning model. When the frequency of correction triggers exceeds the preset number, optimize the long-term plan and output the updated long-term load guidance target.
[0007] As a preferred embodiment of the multi-level production planning coordination method for petrochemical plants described in this invention, the step of constructing a dynamic elastic capacity model based on historical operating data and current operating parameters of the petrochemical plant, and outputting the adjustable capacity boundary of the petrochemical plant, specifically includes the following steps: Collect operational data of petrochemical units during historical operating cycles. The operational data includes feedstock properties, reaction conditions, catalyst status, and equipment operation indicators. Simultaneously collect the maximum load value actually maintained by the device for each time period; Using operating parameters as input variables and actual maximum load as output variables, a set of sample pairs is formed. The sample pairs are cleaned and normalized to form a training dataset for modeling. A nonlinear regression method is used to establish a mapping relationship model between input and output variables, expressed as follows: ; in, This represents the adjustable capacity value predicted by the model. For model parameters, For the operating condition parameter vector, Nonlinear mapping function The model's prediction accuracy is evaluated using a validation set. If it meets the preset criteria, the model is then used as a dynamic flexible capacity model. When new operating condition parameters are received, the model calculates and outputs the adjustable capacity boundary value under the corresponding operating condition.
[0008] As a preferred embodiment of the multi-level production planning coordination method for petrochemical plants described in this invention, the specific steps of inputting the adjustable capacity boundary as a constraint condition into the long-term, medium-term, and short-term production plans, and outputting the load plans for each level, are as follows: During the long-term planning optimization process, for each planned month, the expected operating parameters for that month are obtained and input into the dynamic elastic capacity model to obtain the upper limit of capacity for that month. The monthly capacity limit is used as a hard constraint on the unit load. Combined with the annual total target and economic benefit target, the monthly load allocation scheme is generated at the monthly time granularity. During the mid-term planning optimization process, for each planning week, the operating parameters for that week are obtained and input into the dynamic elastic capacity model to obtain the upper limit of capacity for that week. Using the weekly capacity ceiling as a constraint and the total load of the corresponding month in the long-term plan as the boundary input, the detailed load scheme for each week is generated by solving at the weekly time granularity. In the process of optimizing short-term plans, for each planned hour, the current operating parameters are obtained and input into the dynamic elastic capacity model to obtain the upper limit of capacity for that hour. The hourly capacity limit is used as an operational constraint, and the total load of the corresponding week in the medium-term plan is used as the boundary input. The hourly load sequence is generated by solving at the hourly time granularity. After optimizing plans at all levels, the load planning results at three levels—long-term, medium-term, and short-term—are output.
[0009] As a preferred embodiment of the multi-level production planning coordination method for petrochemical plants described in this invention, the specific steps for calculating the load fluctuation differences between load plans at each level and outputting a coordination degree index are as follows: To quantify the consistency level of load arrangements among the long-term, medium-term and short-term production plans, the petrochemical unit to be evaluated and the target analysis month were selected. Extract the overall load setpoint for the month from the long-term production plan and record it as the first load value. ; The planned load values for each week within the month are extracted from the medium-term production plan, and a weighted average is calculated based on the length of each week to obtain the second load value. ; Extract all hourly load plan data for the month from the short-term production plan, calculate their arithmetic mean, and use it as the third load value. ; By the first load value Second load value With the third load value By comparing each pairwise, the load fluctuation parameters among the three levels are calculated, including: Calculate the absolute deviation between the first load value and the second load value; ; Calculate the absolute deviation between the second load value and the third load value; ; Calculate the absolute deviation between the third load value and the first load value; ; Using the deviation vector formed by the three fluctuation parameters, a ternary probability distribution is generated through normalization to characterize the relative weights of the inconsistency contributions between different levels. An information entropy model is constructed based on this probability distribution to measure the degree of disorder or dispersion in multi-level planned load allocation; The information entropy is normalized and compressed to eliminate the influence of dimensions, and a coordination index is output. A higher value indicates better coordination between planning levels, as detailed below: Normalizing the three deviations yields three non-negative values, which form a probability distribution, expressed as: ; in, This represents the normalized weight of the load deviation between the long-term and medium-term plans in the total deviation, reflecting the relative degree of inconsistency between the two. This represents the normalized weight of the load deviation between the medium-term and short-term plans in the total deviation, reflecting the level of fluctuation during the operational refinement process. This represents the normalized weight of the load deviation between short-term and long-term plans in the total deviation, reflecting the deviation trend between long-term goals and recent implementation. The information entropy is calculated based on this probability distribution, and the expression is: ; in, Let represent the logarithmic function with base to the natural logarithm. Indicates by , , The information entropy corresponding to the probability distribution is used to quantify the dispersion or disorder of load arrangements among the three levels of plans. The larger the value, the more dispersed the differences between plans. Substituting information entropy into the synergy expression, the synergy index is output. The expression is: ; in, is a constant term, representing the maximum possible entropy value when the three-level plans deviate completely uniformly.
[0010] As a preferred embodiment of the multi-level production planning coordination method for petrochemical plants described in this invention, the following steps are taken: When the coordination index is compared with a preset threshold and fails to meet the threshold, a coordination constraint is introduced into the multi-objective optimization to readjust the load allocation at each level and output a production plan scheme. Obtain the synergy index corresponding to the current optimization result. ; Coordination index With the preset synergy threshold Compare; like < If so, it is determined that the coordination of multi-level plans is insufficient; A cooperative penalty term is introduced into the original multi-objective optimization objective function. This is used to improve the coordination of multi-level production planning, among which These are the weighting coefficients; The collaborative penalty item Based on the current coordination index and preset threshold It is obtained through dynamic calculation, where, when Approaching or equal to hour, The value gradually decreases, thus reducing its impact on the optimization results; conversely, when... Significantly smaller than hour, The value increases; Will It is added to the original multi-objective optimization objective function to form a new comprehensive objective function; Under the premise of keeping the material balance, energy balance and safe operation constraints of the equipment unchanged, the load distribution of each level is re-solved. Output a new production plan and ensure its synergy meets requirements. .
[0011] As a preferred embodiment of the multi-level production planning coordination method for petrochemical plants described in this invention, the following steps are involved: executing the production plan involves collecting actual load data, calculating the deviation between the plan and the actual load, outputting the execution deviation result, determining whether the deviation exceeds the allowable range based on the result, and if so, generating a corresponding correction amount and feeding it back to the medium-term planning model. During the execution of short-term plans, the actual load values of the periodic acquisition devices are collected. The actual load value is compared point by point with the planned value at the corresponding time point in the short-term plan to obtain the deviation at each point. Set a continuous time window and count the number of times the deviation exceeds the allowable threshold within the window; If the number of occurrences exceeds the preset frequency, it is determined to be a systemic execution deviation; Calculate the average deviation within the time window and output it as the execution deviation result. Multiply the average deviation by the attenuation coefficient to generate the feedback correction amount; The feedback correction amount is input as incremental information into the initial load setting of the next cycle of the medium-term planning model, triggering rolling optimization.
[0012] As a preferred embodiment of the multi-level production planning coordination method for petrochemical plants described in this invention, the step of optimizing the long-term plan and outputting an updated long-term load guidance target when the correction trigger frequency exceeds a preset number includes the following specific steps: Set a statistical period and record the number of times feedback corrections are generated within that period; If the number of generation exceeds the preset limit, the short-term disturbance is determined to have a lasting impact. Initiate a mechanism for further optimization of long-term plans; The dynamic elastic capacity model is re-initiated, and the latest operating conditions are combined to update the capacity boundary forecasts for the next few months. The latest optimized medium-term plan results are used as boundary conditions to constrain the total load for the corresponding month in the long-term plan; Under the premise of meeting the annual targets and capacity constraints, the load distribution for future months is recalculated; Output the updated long-term load guidance scheme.
[0013] Secondly, the present invention provides a multi-level production planning and coordination system for petrochemical plants, comprising: The module includes a dynamic modeling module, a multi-level optimization module, a collaborative evaluation module, a plan correction module, an execution feedback module, and a closed-loop control module. The dynamic modeling module is used to construct a dynamic elastic capacity model based on the historical operating data and current operating parameters of the petrochemical unit, and output the adjustable capacity boundary of the unit under different operating conditions. The multi-level optimization module is used to input the adjustable capacity boundary as a unified constraint into the long-term, medium-term and short-term production plans, and generate monthly, weekly and hourly load plans by combining the objectives and boundary conditions at each level. The collaborative evaluation module is used to extract the load values of each level of plan within the same time period, calculate their pairwise deviations and normalize them, quantify the disorder of the deviation distribution based on the information entropy method, and output a collaborative index that characterizes the consistency of the plan. The plan correction module is used to compare the coordination index with a preset threshold. When the target is not met, a coordination penalty term is introduced in the multi-objective optimization, and a new production plan scheme that meets the coordination requirements is solved. The execution feedback module is used to collect actual load data during the execution of the plan, identify systematic deviations between the plan and the actual load, generate an average deviation amount and construct a feedback correction amount accordingly, and transmit it to the medium-term planning model to trigger rolling optimization. The closed-loop control module is used to count the trigger frequency of feedback correction. When the frequency exceeds the preset upper limit, it is determined that there is a continuous disturbance, and the long-term plan is re-optimized to update the long-term load guidance target, so as to realize the reverse adjustment and closed-loop control from the execution layer to the strategic layer.
[0014] Thirdly, the present invention provides a computer device including a memory and a processor, wherein the memory stores a computer program, wherein when the computer program is executed by the processor, it implements any step of the multi-level production planning coordination method for petrochemical plants as described in the first aspect of the present invention.
[0015] Fourthly, the present invention provides a computer-readable storage medium having a computer program stored thereon, wherein, when the computer program is executed by a processor, it implements any step of the multi-level production planning coordination method for petrochemical plants as described in the first aspect of the present invention.
[0016] The beneficial effects of this invention are as follows: By constructing a dynamic and flexible capacity model based on actual operating conditions, it achieves an accurate characterization of the actual processing capacity of petrochemical plants, and uses this to unify the constraint benchmark of multi-level production plans, solving the problems of disconnect between upper and lower levels and inconsistent capacity assumptions in traditional planning systems. By introducing a synergy index based on information entropy, it quantitatively evaluates the load fluctuation differences between long-term, medium-term, and short-term plans, transforming plan synergy into a calculable and optimizable goal, thus improving the overall consistency of the planning system. Furthermore, it designs a closed-loop control mechanism from execution feedback to medium-term rolling optimization and then to long-term goal adjustment, enabling production plans to have the adaptive ability to cope with continuous disturbances, significantly enhancing the stability and flexibility of the planning system, realizing the transformation of petrochemical enterprise production planning from static and extensive to dynamic, collaborative, and closed-loop optimization, and effectively improving resource utilization efficiency, operational economy, and production execution rate. Attached Figure Description
[0017] To more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings used in the following description of the embodiments will be briefly introduced. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.
[0018] Figure 1 This is a flowchart of the multi-level production planning coordination method for petrochemical plants in Example 1.
[0019] Figure 2 This is a schematic diagram of the multi-level production planning and coordination system of the petrochemical plant in Example 1. Detailed Implementation
[0020] To make the above-mentioned objects, features and advantages of the present invention more apparent and understandable, the specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings.
[0021] Many specific details are set forth in the following description in order to provide a full understanding of the invention. However, the invention may also be practiced in other ways different from those described herein, and those skilled in the art can make similar extensions without departing from the spirit of the invention. Therefore, the invention is not limited to the specific embodiments disclosed below.
[0022] Secondly, the term "one embodiment" or "embodiment" as used herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation of the present invention. The phrase "in one embodiment" appearing in different places in this specification does not necessarily refer to the same embodiment, nor is it a single or selective embodiment that is mutually exclusive with other embodiments.
[0023] Example 1, referring to Figure 1 and Figure 2 This embodiment of the invention provides a multi-level production planning coordination method for petrochemical plants, comprising the following steps: S1. Based on the historical operating data and current operating parameters of the petrochemical unit, construct a dynamic elastic capacity model and output the adjustable capacity boundary of the petrochemical unit. Furthermore, the operation data of the petrochemical unit during its historical operating cycle is collected. The operation data includes the properties of raw materials, reaction conditions, catalyst status, and equipment operation indicators. Simultaneously collect the maximum load value actually maintained by the device for each time period; Using operating parameters as input variables and actual maximum load as output variables, a set of sample pairs is formed. The sample pairs are cleaned and normalized to form a training dataset for modeling. A nonlinear regression method is used to establish a mapping relationship model between input and output variables, expressed as follows: ; in, This represents the adjustable capacity value predicted by the model. For model parameters, For the operating condition parameter vector, Nonlinear mapping function The model's prediction accuracy is evaluated using a validation set. If it meets the preset criteria, the model is then used as a dynamic flexible capacity model. When new operating condition parameters are received, the model calculates and outputs the adjustable capacity boundary value under the corresponding operating condition. It should be noted that the establishment of the dynamic flexible capacity model is based on the objective law that the actual operating boundary of the unit changes dynamically with the operating conditions. The nonlinear regression method captures the coupled influence of multi-dimensional variables such as raw material properties, reaction conditions, and catalyst state on the maximum processing capacity of the unit, avoiding the problem of overly conservative or infeasible planning caused by the traditional fixed capacity assumption, and ensuring that the output adjustable capacity boundary can truly reflect the actual operating capacity of the unit under the current operating conditions.
[0024] S2. Input the adjustable capacity boundary as a constraint into the long-term, medium-term and short-term production plans, and output the load plans at each level. Furthermore, during the long-term planning optimization process, for each planned month, the expected operating parameters for that month are obtained and input into the dynamic elastic capacity model to obtain the upper limit of capacity for that month. The monthly capacity limit is used as a hard constraint on the unit load. Combined with the annual total target and economic benefit target, the monthly load allocation scheme is generated at the monthly time granularity. During the mid-term planning optimization process, for each planning week, the operating parameters for that week are obtained and input into the dynamic elastic capacity model to obtain the upper limit of capacity for that week. Using the weekly capacity ceiling as a constraint and the total load of the corresponding month in the long-term plan as the boundary input, the detailed load scheme for each week is generated by solving at the weekly time granularity. In the process of optimizing short-term plans, for each planned hour, the current operating parameters are obtained and input into the dynamic elastic capacity model to obtain the upper limit of capacity for that hour. The hourly capacity limit is used as an operational constraint, and the total load of the corresponding week in the medium-term plan is used as the boundary input. The hourly load sequence is generated by solving at the hourly time granularity. After optimizing plans at all levels, the load planning results at three levels—long-term, medium-term, and short-term—are output. It should be noted that during the optimization process, all production plans at each level are constrained by the capacity boundary output by the same dynamic elastic capacity model. This achieves consistency in capacity assumptions from long-term strategic planning to short-term operational execution, effectively solving problems such as disconnect between upper and lower levels and difficulty in implementing plans caused by inconsistent capacity settings in traditional multi-level planning. This improves the overall coordination and executability of the planning system.
[0025] S3. Calculate the load fluctuation differences between load plans at each level and output the coordination index; Furthermore, in order to quantify the consistency level of load arrangements among the long-term, medium-term and short-term production plans, the petrochemical unit to be evaluated and the target analysis month were selected. Extract the overall load setpoint for the month from the long-term production plan and record it as the first load value. ; The planned load values for each week within the month are extracted from the medium-term production plan, and a weighted average is calculated based on the length of each week to obtain the second load value. ; Extract all hourly load plan data for the month from the short-term production plan, calculate their arithmetic mean, and use it as the third load value. ; By the first load value Second load value With the third load value By comparing each pairwise, the load fluctuation parameters among the three levels are calculated, including: Calculate the absolute deviation between the first load value and the second load value; ; Calculate the absolute deviation between the second load value and the third load value; ; Calculate the absolute deviation between the third load value and the first load value; ; Using the deviation vector formed by the three fluctuation parameters, a ternary probability distribution is generated through normalization to characterize the relative weights of the inconsistency contributions between different levels. An information entropy model is constructed based on this probability distribution to measure the degree of disorder or dispersion in multi-level planned load allocation; The information entropy is normalized and compressed to eliminate the influence of dimensions, and a coordination index is output. A higher value indicates better coordination between planning levels, as detailed below: Normalizing the three deviations yields three non-negative values, which form a probability distribution, expressed as: ; in, This represents the normalized weight of the load deviation between the long-term and medium-term plans in the total deviation, reflecting the relative degree of inconsistency between the two. This represents the normalized weight of the load deviation between the medium-term and short-term plans in the total deviation, reflecting the level of fluctuation during the operational refinement process. This represents the normalized weight of the load deviation between short-term and long-term plans in the total deviation, reflecting the deviation trend between long-term goals and recent implementation. The information entropy is calculated based on this probability distribution, and the expression is: ; in, Let represent the logarithmic function with base to the natural logarithm. Indicates by , , The information entropy corresponding to the probability distribution is used to quantify the dispersion or disorder of load arrangements among the three levels of plans. The larger the value, the more dispersed the differences between plans. Substituting information entropy into the synergy expression, the synergy index is output. The expression is: ; in, For constant terms, it represents the maximum possible entropy value when the three-level plans deviate completely uniformly; It should be noted that the coordination index quantifies the deviation distribution characteristics among the three types of load plans—long-term, medium-term, and short-term—using the information entropy method. It transforms the connection quality of multi-level plans into a measurable and comparable numerical indicator, reflecting not only the magnitude of the deviation but also the distribution pattern of the deviation among different levels. This provides an objective basis for judging whether the planning system is in a coordinated state.
[0026] S4. Compare the synergy index with the preset threshold. If the index is not met, introduce synergy constraints in the multi-objective optimization, readjust the load allocation at each level, and output the production plan scheme. Furthermore, obtain the synergy index corresponding to the current optimization result. ; Coordination index With the preset synergy threshold Compare; like < If so, it is determined that the coordination of multi-level plans is insufficient; A cooperative penalty term is introduced into the original multi-objective optimization objective function. This is used to improve the coordination of multi-level production planning, among which These are the weighting coefficients; The collaborative penalty item Based on the current coordination index and preset threshold It is obtained through dynamic calculation, where, when Approaching or equal to hour, The value gradually decreases, thus reducing its impact on the optimization results; conversely, when... Significantly smaller than hour, The value increases; Will It is added to the original multi-objective optimization objective function to form a new comprehensive objective function; Under the premise of keeping the material balance, energy balance and safe operation constraints of the equipment unchanged, the load distribution of each level is re-solved. Output a new production plan and ensure its synergy meets requirements. ; It should be noted that the mechanism of introducing a synergy penalty term enables the planning optimization process to actively consider the consistency of load arrangements at all levels while pursuing economic benefits, avoiding planning gaps caused by unilaterally pursuing local optima. This penalty term is only activated when the synergy is not up to standard, taking into account both optimization flexibility and system stability, and ensuring the overall continuity of the production plan.
[0027] S5. Execute the production plan, collect actual load data, calculate the deviation between the plan and the actual load, output the execution deviation results, determine whether the deviation exceeds the allowable range based on the execution deviation results, and if it does, generate the corresponding correction amount and feed it back to the medium-term planning model. Furthermore, during the execution of short-term plans, the actual load values of the devices are periodically collected; The actual load value is compared point by point with the planned value at the corresponding time point in the short-term plan to obtain the deviation at each point. Set a continuous time window and count the number of times the deviation exceeds the allowable threshold within the window; If the number of occurrences exceeds the preset frequency, it is determined to be a systemic execution deviation; Calculate the average deviation within the time window and output it as the execution deviation result. Multiply the average deviation by the attenuation coefficient to generate the feedback correction amount; The feedback correction amount is input as incremental information into the initial load setting of the next cycle of the medium-term planning model to trigger rolling optimization; It should be noted that the execution feedback mechanism identifies systematic deviations that exceed the range of random fluctuations by continuously monitoring the short-term actual load and planned values, and generates feedback correction quantities with decay characteristics. This not only responds to operational disturbances in a timely manner, but also avoids over-adjustment caused by instantaneous fluctuations, thereby improving the robustness and adaptability of medium-term plan rolling optimization.
[0028] S6. When the frequency of correction triggering exceeds the preset number, optimize the long-term plan and output the updated long-term load guidance target. Furthermore, a statistical period is set to record the number of times feedback corrections are generated within that period; If the number of generation exceeds the preset limit, the short-term disturbance is determined to have a lasting impact. Initiate a mechanism for further optimization of long-term plans; The dynamic elastic capacity model is re-initiated, and the latest operating conditions are combined to update the capacity boundary forecasts for the next few months. The latest optimized medium-term plan results are used as boundary conditions to constrain the total load for the corresponding month in the long-term plan; Under the premise of meeting the annual targets and capacity constraints, the load distribution for future months is recalculated; Output the updated long-term load guidance scheme; It should be noted that when feedback corrections are frequently triggered, it indicates that short-term disturbances have evolved into a persistent trend. At this point, long-term plan re-optimization is initiated, and the long-term load target is dynamically adjusted by combining the latest operating condition forecasts and mid-term optimization results. This achieves reverse feedback from execution-level anomalies to strategic-level targets, enabling the long-term plan to have dynamic evolution capabilities and enhancing the overall planning system's adaptability to complex operating environments.
[0029] This embodiment also provides a multi-level production planning and coordination system for petrochemical plants, including: The module includes a dynamic modeling module, a multi-level optimization module, a collaborative evaluation module, a plan correction module, an execution feedback module, and a closed-loop control module. The dynamic modeling module is used to build a dynamic elastic capacity model based on the historical operating data and current operating parameters of the petrochemical unit, and output the adjustable capacity boundary of the unit under different operating conditions. The multi-level optimization module is used to input the adjustable capacity boundary as a unified constraint into the long-term, medium-term and short-term production plans, and generate monthly, weekly and hourly load plans by combining the objectives and boundary conditions of each level. The collaborative evaluation module is used to extract the load values of plans at each level within the same time period, calculate their pairwise deviations and normalize them, quantify the disorder of the deviation distribution based on the information entropy method, and output a collaborative index that characterizes the consistency of the plans. The plan correction module is used to compare the coordination index with the preset threshold. When the index is not met, a coordination penalty term is introduced into the multi-objective optimization to resolve the production plan scheme that meets the coordination requirements. The execution feedback module is used to collect actual load data during the execution of the plan, identify systematic deviations between the plan and the actual load, generate the average deviation amount and construct the feedback correction amount accordingly, and pass it to the medium-term planning model to trigger rolling optimization. The closed-loop control module is used to count the frequency of feedback corrections. When the frequency exceeds the preset upper limit, it is determined that there is a continuous disturbance, and the long-term plan is re-optimized to update the long-term load guidance target, so as to realize reverse adjustment and closed-loop control from the execution layer to the strategic layer.
[0030] This embodiment also provides a computer device applicable to the multi-level production planning coordination method for petrochemical plants, comprising: a memory and a processor; the memory is used to store computer-executable instructions, and the processor is used to execute the computer-executable instructions to implement the multi-level production planning coordination method for petrochemical plants as proposed in the above embodiment.
[0031] The computer device can be a terminal, comprising a processor, memory, communication interface, display screen, and input devices connected via a system bus. The processor provides computing and control capabilities. The memory includes non-volatile storage media and internal memory. The non-volatile storage media stores the operating system and computer programs. The internal memory provides an environment for the operation of the operating system and computer programs stored in the non-volatile storage media. The communication interface is used for wired or wireless communication with external terminals; wireless communication can be achieved through Wi-Fi, carrier networks, NFC (Near Field Communication), or other technologies. The display screen can be an LCD screen or an e-ink screen. The input devices can be a touch layer covering the display screen, buttons, a trackball, or a touchpad on the computer device's casing, or an external keyboard, touchpad, or mouse.
[0032] This embodiment also provides a storage medium storing a computer program that, when executed by a processor, implements the multi-level production planning coordination method for petrochemical plants as proposed in the above embodiments. The storage medium can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as Static Random Access Memory (SRAM), Electrically Erasable Programmable Read-Only Memory (EEPROM), Erasable Programmable Read Only Memory (EPROM), Programmable Red-Only Memory (PROM), Read-Only Memory (ROM), magnetic storage, flash memory, magnetic disk, or optical disk.
[0033] In summary, this invention, by constructing a dynamic and flexible capacity model based on actual operating conditions, achieves an accurate depiction of the real processing capacity of petrochemical plants. This model unifies the constraint benchmarks of multi-level production plans, solving the problems of disconnect between upper and lower levels and inconsistent capacity assumptions in traditional planning systems. By introducing a synergy index based on information entropy, it quantitatively assesses the load fluctuation differences between long-term, medium-term, and short-term plans, transforming plan synergy into a calculable and optimizable objective, thus improving the overall consistency of the planning system. Furthermore, it designs a closed-loop control mechanism from execution feedback to medium-term rolling optimization and then to long-term target adjustment, enabling production plans to have adaptive capabilities to cope with continuous disturbances. This significantly enhances the stability and flexibility of the planning system, realizing the transformation of petrochemical enterprise production planning from static and extensive to dynamic, collaborative, and closed-loop optimized, effectively improving resource utilization efficiency, operational economy, and production execution rate.
[0034] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention, and all such modifications or substitutions should be covered within the scope of the claims of the present invention.
Claims
1. A petrochemical plant multi-level production planning coordination method, characterized by: The application relates to a method for optimizing production plan of petrochemical device, comprising the following steps: Based on historical operation data and current working condition parameters of the petrochemical device, a dynamic elastic production capacity model is constructed, and the adjustable production capacity boundary of the petrochemical device is outputted; The adjustable production capacity boundary is inputted as a constraint condition into long-term, medium-term and short-term production plans, and each level load plan is outputted; The load fluctuation difference between each level load plan is calculated, and a coordination degree index is outputted; The coordination degree index is compared with a preset threshold value, when the coordination degree index does not reach the threshold value, a coordination constraint is introduced in multi-objective optimization, each level load distribution is readjusted, and a production plan scheme is outputted; The production plan scheme is executed, actual load data is collected, a deviation between the plan and the actual is calculated, an execution deviation result is outputted, whether the execution deviation result exceeds an allowable range is judged, if the execution deviation result exceeds the allowable range, a corresponding correction amount is generated and is fed back to the medium-term plan model; When the correction amount triggering frequency exceeds a preset number of times, the long-term plan is optimized, and an updated long-term load guidance target is outputted.
2. The petrochemical plant multi-level production planning coordination method of claim 1, wherein: The dynamic elastic production capacity model is constructed based on the historical operation data and the current working condition parameters of the petrochemical device, and the adjustable production capacity boundary of the petrochemical device is outputted, and the specific steps are as follows: Operation data of the petrochemical device in a historical operation period are collected, the operation data comprising raw material properties, reaction conditions, catalyst states and equipment operation identifiers; The maximum load values actually maintained by the device in each period are synchronously collected; The working condition parameters are taken as input variables, and the actual maximum load is taken as an output variable, thereby forming a sample pair set; The sample pair set is subjected to data cleaning and normalization treatment, thereby forming a training data set for modeling; A mapping relationship model between the input variables and the output variable is established by adopting a nonlinear regression method, and the expression is as follows: ； wherein, is a model predicted adjustable capacity value, is a model parameter, is a working condition parameter vector, is a nonlinear mapping function The model prediction precision is evaluated through a verification set, if the model prediction precision meets a preset standard, the model is taken as the dynamic elastic production capacity model and is put into use; When new working condition parameters are inputted, the model is used to calculate and output the adjustable production capacity boundary value under the corresponding working condition.
3. The petrochemical plant multi-level production planning coordination method of claim 2, wherein: The adjustable production capacity boundary is inputted as a constraint condition into long-term, medium-term and short-term production plans, and each level load plan is outputted, and the specific steps are as follows: In the long-term plan optimization process, for each plan month, the expected working condition parameters of the plan month are obtained and are inputted into the dynamic elastic production capacity model, thereby obtaining the production capacity upper limit value of the month; The production capacity upper limit value of the month is taken as a hard constraint of the device load, and the annual total amount target and the economic benefit target are combined to solve and generate a monthly load distribution scheme in a monthly time granularity; In the medium-term plan optimization process, for each plan week, the working condition parameters of the plan week are obtained and are inputted into the dynamic elastic production capacity model, thereby obtaining the production capacity upper limit value of the week; The production capacity upper limit value of the week is taken as a constraint condition, and the load total amount of the corresponding month in the long-term plan is taken as a boundary input, thereby solving and generating a weekly load refinement scheme in a weekly time granularity; In the short-term plan optimization process, for each plan hour, the current working condition parameters are obtained and are inputted into the dynamic elastic production capacity model, thereby obtaining the production capacity upper limit value of the hour; The production capacity upper limit value of the hour is taken as an operation constraint, and the load total amount of the corresponding week in the medium-term plan is taken as a boundary input, thereby solving and generating a hourly load sequence in a hourly time granularity; After the optimization of each level of the plan is completed, long-term, medium-term and short-term load plan results are output.
4. The petrochemical plant multi-level production planning coordination method of claim 3, wherein: The load fluctuation difference between the load plans of each level is calculated, and a coordination degree index is output, and the specific steps are as follows: In order to quantify the consistency level of load arrangement between the long-term, medium-term and short-term production plans, a petrochemical device to be evaluated and a target analysis month are selected; extracting an overall load set value for this month from a long-term production plan, denoted as a first load value ; The second load value is obtained by extracting the load plan value of each week in the month from the medium-term production plan and performing a weighted average according to the length of each week ; Extract all hourly load plan data for this month from the short term production plan, calculate the arithmetic average of it as the third load value ; By pairwise comparison of the first load value , the second load value and the third load value , the load fluctuation parameters between the three levels are calculated, including: The absolute deviation between the first load value and the second load value is calculated; ; The absolute deviation between the second load value and the third load value is calculated; ; The absolute deviation between the third load value and the first load value is calculated; ; A three-element probability distribution is generated by normalization processing using the deviation vector composed of the three fluctuation parameters, representing the relative weight of the inconsistency contribution between different levels; Based on the probability distribution, an information entropy model is constructed to measure the degree of disorder or dispersion of the load distribution of the multi-level plan; The information entropy is normalized and compressed to eliminate the dimension influence, and a coordination degree index is output, wherein the higher the value, the better the coordination between the plan levels, and the specific steps are as follows: The three deviation quantities are normalized to obtain three non-negative values to form a probability distribution, and the expression is as follows: ; wherein, represents the normalized weight of the load deviation between the medium-term plan and the short-term plan in the total deviation, reflecting the fluctuation level in the operation refinement process, represents the normalized weight of the load deviation between the medium-term plan and the short-term plan in the total deviation, reflecting the fluctuation level in the operation refinement process, represents the normalized weight of the load deviation between the medium-term plan and the short-term plan in the total deviation, reflecting the fluctuation level in the operation refinement process, Based on the probability distribution, the information entropy is calculated, and the expression is as follows: ; wherein, denotes a logarithmic function with base of natural logarithm, denotes a probability distribution of the load arrangement of the three levels of plans, , , information entropy corresponding to the probability distribution, for quantifying the dispersion or disorder of the load arrangement among the three levels of plans, the greater the value, the more dispersed the differences among the plans The information entropy is substituted into the expression of the coordination degree to output a coordination degree index , and the expression is ; where, is the constant term, representing the maximum possible entropy value when the three-level plan is perfectly uniform deviated.
5. The petrochemical plant multi-level production planning coordination method of claim 4, wherein: The coordination degree index is compared with a preset threshold, and when it does not meet the standard, a coordination constraint is introduced in the multi-objective optimization to adjust the load distribution of each level, and a production plan scheme is output, and the specific steps are as follows: obtaining a coordination degree index corresponding to the current optimization result ; comparing the synergy index with a preset synergy threshold value; If then the multi-level plan coordination is determined to be insufficient; Introducing a coordination penalty term in the original multi-objective optimization objective function for improving the coordination of multi-level production planning, wherein is a weight coefficient; The synergistic penalty term Based on the current synergy index And a preset threshold Is dynamically calculated, wherein, when Approaches or equals , The value gradually decreases, thereby reducing the influence on the optimization result, on the contrary, when Significantly less than , The value increases; Will It is added to the original multi-objective optimization objective function to form a new comprehensive objective function; While keeping the material balance, energy balance and device safe operation constraints unchanged, the load distribution of each level is recalculated; output a new production plan scheme and ensure that its coordination degree meets .
6. The petrochemical plant multi-level production planning coordination method of claim 5, wherein: The production plan scheme is executed, actual load data is collected, the deviation between the plan and the actual value is calculated, and an execution deviation result is output, and whether the execution deviation result exceeds the allowed range is determined, and if it exceeds, a corresponding correction amount is generated and fed back to the medium-term plan model, and the specific steps are as follows: During the execution of the short-term plan, the actual load value of the device is periodically collected; The actual load value is compared with the planned value at the corresponding time point in the short-term plan point by point to obtain the deviation quantity of each point; A continuous time window is set to count the number of times the deviation exceeds the allowed threshold in the window; If the number of times exceeds the preset frequency, it is determined that there is a systematic execution deviation; The average deviation quantity in the time window is calculated as the execution deviation result output; The average deviation quantity is multiplied by a decay coefficient to generate a feedback correction amount; The feedback correction amount is input as incremental information into the initial load setting of the next cycle of the medium-term plan model to trigger rolling optimization; The incremental information is used as a dynamic disturbance compensation signal to participate in the re-optimization process of the medium-term plan; In the optimization process of the medium-term plan model, the incremental information is introduced in the form of an additional term to the load total quantity constraint boundary, and the expression is as follows: ； wherein, is the feedback correction; By introducing the incremental information, the medium-term plan can respond to the persistent deviation of the short-term execution layer, regenerate a weekly load distribution scheme suitable for the current working condition, and pass the updated boundary guidance to the short-term plan layer.
7. The petrochemical plant multi-level production planning coordination method of claim 6, wherein: When the correction amount triggers the frequency more than the preset number of times, the long-term plan is optimized, and an updated long-term load guidance target is output, and the specific steps are as follows: A statistical period is set to record the number of times the feedback correction amount is generated in the period; If the number of generations exceeds the preset upper limit, it is determined that the short-term disturbance has a persistent impact; Start the long-term plan re-optimization mechanism; Recall the dynamic flexible capacity model, combine the latest operating condition trend, and update the capacity boundary prediction value for future months; Take the latest optimized medium-term plan results as boundary conditions to constrain the total load in the corresponding months of the long-term plan; Resolving the load distribution of future months under the premise of meeting the annual target and capacity constraints; Output the updated long-term load guidance scheme.
8. A petrochemical plant multi-level production planning coordination system based on the petrochemical plant multi-level production planning coordination method of any one of claims 1-7, characterized in that: Comprise: Dynamic modeling module, multi-level optimization module, collaborative evaluation module, plan correction module, execution feedback module and closed-loop control module; The dynamic modeling module is used to construct a dynamic flexible capacity model based on the historical operation data and current operating condition parameters of the petrochemical device, and output the adjustable capacity boundary of the device under different operating conditions; The multi-level optimization module is used to input the adjustable capacity boundary into the long-term, medium-term and short-term production plans as unified constraints, and generate monthly, weekly and hourly load plans based on the target and boundary conditions of each level; The collaborative evaluation module is used to extract the load values of each level plan in the same time period, calculate the deviation between them and normalize, quantify the disorder of the deviation distribution based on the information entropy method, and output the collaborative degree index representing the consistency of the plan; The plan correction module is used to compare the collaborative degree index with the preset threshold, and when it does not meet the standard, introduce a collaboration penalty term in the multi-objective optimization to re-solve the production plan scheme that meets the collaboration requirement; The execution feedback module is used to collect actual load data during plan execution, identify systematic deviations between the plan and the actual, generate an average deviation amount and construct a feedback correction amount based on it, and pass it to the medium-term plan model to trigger rolling optimization; The closed-loop control module is used to count the trigger frequency of the feedback correction amount, and when the frequency exceeds the preset upper limit, it is determined that there is a persistent disturbance, the long-term plan re-optimization is started, the long-term load guidance target is updated, and the reverse adjustment and closed-loop control from the execution layer to the strategic layer are realized. 9.A computer device, comprising a memory and a processor, wherein the memory stores a computer program, and the computer device is characterized in that: The processor executes the computer program to realize the steps of the petrochemical device multi-level production plan collaboration method of any one of claims 1-7.
10. A computer readable storage medium having stored thereon a computer program, characterized in that: The computer program is executed by the processor to realize the steps of the petrochemical device multi-level production plan collaboration method of any one of claims 1-7.