Industrial medium scheduling method and device

Abstract

This application discloses an industrial media scheduling method and apparatus. The method includes: receiving media data from a system to be scheduled under abnormal conditions and production data of multiple scheduling objects in the system, wherein the media data includes at least the imbalance quantity of various media; determining a target scheduling sequence for the system to be scheduled based on the media data, the production data of the multiple scheduling objects, and pre-acquired scheduling rules; constructing a scheduling model for the system to be scheduled under abnormal conditions based on the target scheduling sequence; and determining a scheduling strategy for the system to be scheduled under abnormal conditions based on the scheduling model. The method provided in this application at least solves the technical problem of low scheduling accuracy caused by manual decision-making in industrial media scheduling in related technologies.

Description

Technical Field

[0001] This application relates to the field of process control technology, and more specifically, to an industrial media scheduling method and apparatus. Background Technology

[0002] A chemical plant is a highly complex, continuously operating process industry system. Its various production units are interconnected through a complex network of pipelines, forming a tight coupling relationship between materials and energy. In this environment where multiple units and media (such as hydrogen, nitrogen, ammonia, steam, and various distillates) coexist, any abnormality in the feeding, reaction, separation, or output stages of any unit will disrupt the original equilibrium, creating an "imbalance." This imbalance will propagate upstream or downstream along the material chain. If not managed promptly and effectively, it can lead to increased energy consumption, material consumption, and product quality fluctuations, or even force units to reduce load or shut down. Currently, the traditional method for dealing with such imbalances in chemical plants mainly relies on the personal experience of dispatchers. Operators or dispatchers need to determine the source of the abnormality based on alarm information and their personal experience, and manually calculate which units need adjustment and by how much to restore balance. However, this method has significant drawbacks, such as: Lag: There is a long time delay from the occurrence of an anomaly to human discovery, decision-making, and execution, making rapid response impossible; One-sidedness: Human experience cannot fully consider the complex coupling relationships between all units and media in the entire plant (such as the synergistic ratio of hydrogen and nitrogen in ammonia synthesis). Decisions made may be locally optimal, but they disrupt the balance of other areas and may even lead to a chain reaction of anomalies in other media; Non-optimality: Human decision-making usually prioritizes restoring balance, rarely considering "how to achieve balance with the least operational cost (such as adjusting the fewest units and the smallest adjustment range)," thus missing opportunities for energy conservation, emission reduction, and improved operational efficiency. Other approaches employ mathematical programming or intelligent algorithms for process optimization, but these methods mostly have limitations: they either optimize only a single unit without considering the entire plant system; or they rely solely on precise mathematical models, ignoring the numerous, difficult-to-model expert rules and operational logics present in chemical plants (e.g., which unit to adjust first under what operating conditions), resulting in theoretically feasible optimization schemes but poor practical operability. Summary of the Invention

[0003] This application provides an industrial media scheduling method and apparatus to at least solve the technical problem in the related art where the scheduling accuracy of industrial media is low due to the use of manual decision-making.

[0004] According to one aspect of the embodiments of this application, an industrial media scheduling method is provided, comprising: receiving media data of a system to be scheduled under an abnormal state and production data of multiple scheduling objects in the system to be scheduled, wherein the media data includes at least: the imbalance quantity of various media; determining a target scheduling sequence of the system to be scheduled based on the media data of the system to be scheduled, the production data of the multiple scheduling objects in the system to be scheduled, and pre-acquired scheduling rules; constructing a scheduling model of the system to be scheduled under the abnormal state based on the target scheduling sequence; and determining a scheduling strategy of the system to be scheduled under the abnormal state based on the scheduling model.

[0005] Optionally, constructing a scheduling model for the system to be scheduled under the abnormal state based on the target scheduling sequence includes: obtaining multiple decision variables, the objective function of the scheduling model, and the constraints of the scheduling model, wherein the multiple decision variables include at least one of the following: sequential identifier binary variables, continuous adjustment variables, and relaxation variables; the sequential identifier binary variables are used to characterize the adjustment order relationship between the scheduling objects; the continuous adjustment variables are used to characterize the adjustment amount of the scheduling objects as continuous variables; the relaxation variables are used to characterize the relaxation amount of the constraints; and the objective function is constructed with the goal of minimizing the number of scheduling objects adjusted. The scheduling model for the system to be scheduled is constructed based on the multiple decision variables, the objective function of the scheduling model, and the constraints of the scheduling model.

[0006] Optionally, the objective function of the scheduling model is determined by: obtaining the objective parameters of the system to be scheduled, wherein the objective parameters include: the number of scheduling objects, the adjustment amount of the scheduling objects, and the degree of violation of the constraints of the scheduling objects; and constructing the objective function based on the objective parameters of the system to be scheduled.

[0007] Optionally, the constraints of the scheduling model include: material balance constraints, adjustment capability constraints of the scheduling object, mathematical model constraints of the scheduling object, and scheduling logic constraints. The material balance constraints include: the materials in the system to be scheduled remain balanced after the scheduling operation. The adjustment capability constraints of the scheduling object include: all adjustment operations must be within the safe operating range of the equipment. The mathematical model constraints of the scheduling object include: the states of different scheduling objects satisfy a preset correlation. The scheduling logic constraints are quantified using a Big M method.

[0008] Optionally, determining the scheduling strategy of the system to be scheduled under the abnormal state based on the scheduling model includes: converting the attributes of integer variables in the scheduling model into continuous variables to obtain an adjusted scheduling model; solving the adjusted scheduling model using a linear programming solver to obtain an initial solution, wherein the initial solution is a relaxed solution; when the values ​​of the integer variables in the initial solution are non-integer, using a branch and bound algorithm to branch the integer variables in the initial solution to obtain multiple branches until the branch and bound algorithm converges to obtain the optimal solution of the scheduling model, wherein the optimal solution of the scheduling model is used to characterize the scheduling strategy of the system to be scheduled, and the scheduling strategy includes at least: the adjustment order and adjustment amount of multiple scheduling objects of the system to be scheduled.

[0009] Optionally, a branch and bound algorithm is used to branch the integer variables in the initial solution to obtain multiple branches until the branch and bound algorithm converges. This includes: sequentially determining the non-integer integer variables in the initial solution as branch variables; performing branch operations on the branch variables to obtain the corresponding subproblems; wherein the priority of the branch operation on each non-integer integer variable is determined based on the degree of influence of the non-integer integer variable on the objective function value; adding the constraints of the scheduling model to the subproblems; solving the subproblems to obtain the solution results; determining the upper and lower bounds of the subproblems based on the solution results; pruning the subproblems if the upper bound is not less than the global upper bound at the current time; and pruning the subproblems if the solution results do not meet the constraints of the scheduling model.

[0010] Optionally, determining the upper and lower bounds of the subproblem based on the solution results of the subproblem includes: if the solution result of the subproblem is a feasible solution, determining the objective function value corresponding to the feasible solution as the lower bound of the subproblem; if the solution result of the subproblem is an integer solution, determining the objective function value corresponding to the integer solution as the upper bound of the subproblem, wherein the solution process converges in the direction of minimizing the value of the branch variable.

[0011] According to another aspect of the embodiments of this application, an industrial media scheduling device is also provided, comprising: a receiving module, configured to receive media data of a system to be scheduled under an abnormal state and production data of multiple scheduling objects in the system to be scheduled, wherein the media data includes at least: the imbalance quantity of various media; a determining module, configured to determine a target scheduling sequence of the system to be scheduled based on the media data of the system to be scheduled, the production data of the multiple scheduling objects in the system to be scheduled, and pre-acquired scheduling rules; a constructing module, configured to construct a scheduling model of the system to be scheduled under the abnormal state based on the target scheduling sequence; and a scheduling module, configured to determine a scheduling strategy of the system to be scheduled under the abnormal state based on the scheduling model.

[0012] According to another aspect of the embodiments of this application, a computer device is also provided, including: a memory and a processor, wherein the memory is used to store program instructions; and the processor, connected to the memory, is used to execute the above-described industrial media scheduling method.

[0013] According to another aspect of the embodiments of this application, a computer program product is also provided, including computer instructions that, when executed by a processor, implement the above-described industrial media scheduling method.

[0014] In this embodiment, the method involves receiving media data from a system under abnormal conditions and production data of multiple scheduling objects within that system. The media data includes at least the imbalance quantities of various media. A target scheduling sequence for the system under abnormal conditions is determined based on the media data, the production data of the multiple scheduling objects, and pre-acquired scheduling rules. A scheduling model for the system under abnormal conditions is constructed based on the target scheduling sequence. A scheduling strategy for the system under abnormal conditions is determined based on the scheduling model. By adjusting the scheduling sequence through the constructed scheduling model, the scheduling strategy is adjusted, thereby improving scheduling accuracy. This solves the technical problem in related technologies where low scheduling accuracy is caused by manual decision-making in scheduling industrial media. Attached Figure Description

[0015] The accompanying drawings, which are included to provide a further understanding of this application and form part of this application, illustrate exemplary embodiments of this application and are used to explain this application, but do not constitute an undue limitation of this application. In the drawings:

[0016] Figure 1 This is a hardware structure block diagram of a computer terminal for implementing an industrial media scheduling method according to an embodiment of this application;

[0017] Figure 2This is a flowchart of an industrial media scheduling method according to an embodiment of this application;

[0018] Figure 3 This is a flowchart of another industrial media scheduling method according to an embodiment of this application;

[0019] Figure 4 This is a flowchart of a scheduling model construction method according to an embodiment of this application;

[0020] Figure 5 This is a schematic diagram of a hydrogen and steam dispatching system according to an embodiment of this application;

[0021] Figure 6 This is a schematic diagram of the scheduling logic of a hydrogen and steam scheduling system in case of an anomaly, according to an embodiment of this application.

[0022] Figure 7 This is a structural diagram of an industrial media scheduling device according to an embodiment of this application. Detailed Implementation

[0023] To enable those skilled in the art to better understand the present application, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present application, and not all embodiments. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative effort should fall within the scope of protection of the present application.

[0024] It should be noted that the terms "first," "second," etc., in the specification, claims, and accompanying drawings of this application are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of this application described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0025] The information collected in this application embodiment is information and data authorized by the user or fully authorized by all parties. The collection, storage, use, processing, transmission, provision, disclosure and application of the relevant data all comply with the relevant laws, regulations and standards of the relevant regions, and necessary confidentiality measures have been taken. It does not violate public order and good morals, and provides corresponding operation entry points for users to choose to authorize or reject the automated decision results. If the user chooses to reject, the process will proceed to the expert decision-making process.

[0026] To better understand the embodiments of this application, the technical terms used in the embodiments of this application are explained as follows:

[0027] Mixed-integer mathematical programming: Under a set of constraints, find the values ​​of a set of decision variables that include both continuous and integer variables, so that a given objective function is optimal (maximum or minimum).

[0028] Scheduling: This involves comprehensive monitoring and coordination of the production process to ensure its smooth operation and achieve optimal production efficiency and quality. On one hand, production supervisors need to rationally arrange various tasks during production based on production plans and market demands. On the other hand, they need to promptly handle any abnormal situations that arise during production to ensure stability and safety.

[0029] Scheduling object: When an anomaly occurs, the device that can coordinate and balance the fluctuations in the pipeline network is called the scheduling object.

[0030] To address the problems existing in related technologies, this application provides an industrial media scheduling method, which can be run on... Figure 1 The computer terminal shown is explained below.

[0031] The industrial media scheduling method embodiments provided in this application can be executed on mobile terminals, computer terminals, or similar computing devices. Figure 1 A hardware block diagram of a computer terminal for implementing an industrial media scheduling method is shown. Figure 1 As shown, the computer terminal 10 may include one or more processors (shown as 102a, 102b, ..., 102n in the figure) (the processor may include, but is not limited to, a microprocessor MCU or a programmable logic device FPGA, etc.), a memory 104 for storing data, and a transmission module 106 for communication functions connected via wired and / or wireless networks. In addition, it may also include: a display, a keyboard, a cursor control device, an input / output interface (I / O interface), a universal serial bus (USB) port (which may be included as one of the ports of the I / O interface), a network interface, and a BUS bus. Those skilled in the art will understand that... Figure 1The structure shown is for illustrative purposes only and does not limit the structure of the aforementioned electronic device. For example, computer terminal 10 may also include... Figure 1 The more or fewer components shown, or having the same Figure 1 The different configurations shown.

[0032] It should be noted that the aforementioned one or more processors and / or other data processing circuits are generally referred to herein as "data processing circuits". These data processing circuits may be embodied, in whole or in part, in software, hardware, firmware, or any other combination thereof. Furthermore, the data processing circuits may be a single, independent processing module, or may be integrated, in whole or in part, into any other element within the computer terminal 10. As involved in the embodiments of this application, the data processing circuits serve as a processor control mechanism (e.g., selection of a variable resistor termination path connected to an interface).

[0033] The memory 104 can be used to store software programs and modules of application software, such as the program instructions / data storage device corresponding to the industrial media scheduling method in this embodiment. The processor executes various functional applications and data processing by running the software programs and modules stored in the memory 104, thereby realizing the aforementioned industrial media scheduling method. The memory 104 may include high-speed random access memory, and may also include non-volatile memory, such as one or more magnetic storage devices, flash memory, or other non-volatile solid-state memory. In some instances, the memory 104 may further include memory remotely located relative to the processor, and these remote memories can be connected to the computer terminal 10 via a network. Examples of such networks include, but are not limited to, the Internet, corporate intranets, local area networks, mobile communication networks, and combinations thereof.

[0034] The transmission module 106 is used to receive or send data via a network. Specific examples of the network described above may include a wireless network provided by the communication provider of the computer terminal 10. In one example, the transmission module 106 includes a network interface controller (NIC), which can connect to other network devices via a base station to communicate with the Internet. In another example, the transmission module 106 may be a radio frequency (RF) module, used for wireless communication with the Internet.

[0035] The display can be, for example, a touchscreen liquid crystal display (LCD) that allows the user to interact with the user interface of the computer terminal 10.

[0036] It should be noted here that, in some optional embodiments, the above... Figure 1The computer terminal shown may include hardware components (including circuitry), software components (including computer code stored on a computer-readable medium), or a combination of both hardware and software components. It should be noted that... Figure 1 This is only one instance of a specific particular instance, and is intended to illustrate the types of components that may exist in the aforementioned computer terminal.

[0037] In the above operating environment, this application provides an embodiment of an industrial media scheduling method. It should be noted that the steps shown in the flowchart in the accompanying drawings can be executed in a computer system such as a set of computer-executable instructions. Also, although a logical order is shown in the flowchart, in some cases, the steps shown or described can be executed in a different order than that shown here.

[0038] Figure 2 This is a flowchart of an industrial media scheduling method according to an embodiment of this application, such as... Figure 2 As shown, the method includes the following steps:

[0039] Step S202: Receive media data from the system to be scheduled under abnormal conditions and production data of multiple scheduling objects in the system to be scheduled, wherein the media data includes at least: the number of imbalances of various media.

[0040] In step S202, the production data includes at least the following data:

[0041] 1. Media network topology: A complete description of the upstream and downstream connections and production-consumption-storage balance relationships of devices in each media system, forming a directed graph model of media balance;

[0042] 2. Equipment operating characteristic data: including feed and discharge balance models, efficiency curves, operating flexibility range, response characteristic parameters, etc. for each unit;

[0043] 3. Flow parameters: Current real-time values ​​of each flow, upper and lower limits of safe operation, and other constraints;

[0044] 4. System-related constraints: Global restrictions such as coupling relationships between different media systems and material balance constraints.

[0045] Understandably, this data is collected and verified through standardized data interface specifications to ensure data integrity and consistency, providing a reliable data foundation for subsequent optimization and scheduling.

[0046] Step S204: Determine the target scheduling sequence of the system to be scheduled based on the media data in the system to be scheduled, the production data of multiple scheduling objects in the system to be scheduled, and the pre-acquired scheduling rules;

[0047] In step S204, the specific method for determining the target scheduling sequence of the system to be scheduled based on the media data in the system to be scheduled, the production data of multiple scheduling objects in the system to be scheduled, and the pre-acquired scheduling rules is to use an expert scheduling model and a pre-built scheduling logic topology to obtain the scheduling logic under this working condition through a generation platform. This is achieved through modular configuration, rule engine-driven operation, and multi-algorithm collaboration, transforming scattered expert scheduling experience into a structured logic network that can dynamically adapt to the working condition. The generation platform includes the following core functional modules: Calculation module: Provides various mathematical operators, including arithmetic operations, logical operations, conditional judgments, numerical comparisons, etc., for constructing the calculation logic in the scheduling rules. Conditional judgment module: Implements multi-condition combination judgment functions, supports AND, OR, NOT, and other logical operations, forming a hierarchical decision tree. Parallel processing module: Supports multi-branch parallel processing logic, enabling simultaneous scheduling decisions for multiple subsystems. Conditional sorting module: Sorts scheduling objects according to multiple dimensions such as process priority, safety level, and economic indicators, where sorting includes parallel adjustment, serial adjustment, and parallel plus serial adjustment. Scheduling Object Module: Encapsulates the specific control logic and operational constraints of various devices, forming reusable device control knowledge units.

[0048] Through the organic combination of these modules, the system can automatically generate an expert scheduling logic topology network adapted to specific abnormal operating conditions based on real-time operating data and abnormal scenario characteristics. This topology network can uniquely determine the optimal scheduling strategy sequence under the current operating condition, ensuring the systematicity and integrity of scheduling decisions. Specifically, it adopts a visual configuration function. The left side is the module resource area, which presents core modules such as calculation, condition judgment, and parallel processing in the form of standardized icons. Each icon contains a functional description of the module, input and output parameter specifications, and constraint conditions. The middle is the logic drawing area, which provides a canvas-style editing environment, allowing users to intuitively construct the logical flow between modules by dragging and dropping module icons and drawing connecting lines. It also includes auxiliary functions such as grid alignment and automatic routing to ensure the clarity of the logical topology. The right side is the attribute configuration area. When a module icon is selected, the configurable parameters of that module are automatically displayed (such as the operation type of the calculation module, the logical operators of the condition judgment module, and the priority weight of the condition sorting module), allowing users to accurately configure parameters according to specific operating condition requirements. Each functional module is linked by a visual configuration interface, achieving an organic integration of "visual input of requirements - logical structured combination - dynamic generation of topology". This retains the flexibility of expert scheduling experience while ensuring the systematicity and integrity of the scheduling logic.

[0049] It should be noted that the target scheduling sequence includes at least the dynamic adjustment relationship and execution sequence between various scheduling objects. In this embodiment, two adjustment modes are included: parallel collaborative adjustment and serial sequential adjustment. Parallel adjustment allows multiple devices to participate in the adjustment simultaneously to improve response speed; serial sequential adjustment specifies strict start-stop timing and coordination relationships between devices to ensure a smooth transition of the process.

[0050] Step S206: Construct a scheduling model for the system to be scheduled under the abnormal state based on the target scheduling sequence;

[0051] Step S208: Determine the scheduling strategy of the system to be scheduled under the abnormal state according to the scheduling model.

[0052] Through steps S202 to S206 above, the system receives media data from the system to be scheduled under abnormal conditions and production data of multiple scheduling objects in the system to be scheduled. The media data includes at least the imbalance quantity of various media. A target scheduling sequence for the system to be scheduled is determined based on the media data, the production data of the multiple scheduling objects, and pre-acquired scheduling rules. A scheduling model for the system to be scheduled under abnormal conditions is constructed based on the target scheduling sequence. A scheduling strategy for the system to be scheduled under abnormal conditions is determined based on the scheduling model. By adjusting the scheduling sequence through the constructed scheduling model, the scheduling strategy is adjusted, thereby improving the scheduling accuracy and solving the technical problem of low scheduling accuracy caused by manual decision-making in industrial media scheduling in related technologies. A detailed explanation follows.

[0053] Because the scheduling methods provided by related technologies have the technical problems described in the background technology, in order to solve the defects existing in the background technology, the industrial media scheduling method provided in this application embodiment, by integrating expert experience scheduling logic (pre-acquired scheduling rules), accurate mathematical models of devices, and mixed integer optimization algorithms, realizes a systematic solution for quickly balancing the optimal scheduling instructions of materials across the entire plant under abnormal scenarios, achieving the following: 1. Maximizing production stability and continuity, specifically, reducing disturbance propagation: Chemical plants are highly interconnected coupled systems. Adjusting any device, especially core devices (such as reactors and compressors), will produce a chain reaction like "dominoes," and disturbances may propagate to other related units. Reducing the number of devices to be adjusted is essentially minimizing the source of disturbances, controlling changes within a minimum range, thereby maintaining greater stability of the entire production system. Avoiding "solving one problem and creating multiple new problems": Manual adjustments based on experience are prone to falling into the trap of "local optima." Mathematical programming models aimed at minimizing adjustments perform global calculations to ensure that the solution does not introduce new and more serious imbalances elsewhere. This technology seeks a "systematic optimal solution," not a "local optimal solution." 2. Reduce operational and safety risks. Specifically: Simplify operations: The simpler the dispatch instructions, the easier it is for on-site operators to execute them, and the lower the probability of errors. A plan that only requires adjusting two devices is far clearer and less prone to misoperation than a plan that requires adjusting eight devices. Improve safety: Frequent or significant adjustments to device operating conditions are themselves a risk. Devices are safest when operating in a steady state. Any change in operating parameters will temporarily deviate the device from its most familiar and safest operating window, potentially introducing unpredictable risks. Minimizing the number of adjustments is directly equivalent to minimizing safety risks. 3. Improve the acceptability and feasibility of dispatch plans: Reduce organizational coordination costs: Adjusting a device often requires communication and coordination among multiple shifts and positions. The more devices adjusted, the more exponentially the communication costs increase, and the decision-making and execution processes become lengthy. Minimizing the number of devices adjusted can greatly reduce the complexity of internal coordination and speed up response times. 4. Extend equipment life and reduce equipment wear: Reduce equipment movements: Frequent start-ups, shutdowns, or adjustments of equipment such as valves, pumps, and compressors can lead to equipment wear, seal failure, and other problems. By setting the goal of "minimizing adjustments," the number of operations on related equipment is indirectly reduced, which helps to lower maintenance costs and extend equipment lifespan. At the same time, it also achieves multi-objective optimization of the media scheduling process.

[0054] To address the aforementioned issues, this application also provides another industrial media scheduling method, such as... Figure 3As shown, the process includes: Step 1: Obtaining the medium imbalance caused by the anomaly and basic production data; Step 2: Obtaining the scheduling logic (target scheduling sequence) under this working condition based on the expert scheduling model and scheduling logic topology module; Step 3: Constructing a scheduling mathematical model under the abnormal state based on the unique scheduling logic (target scheduling sequence) completed by the topology; Step 4: Solving the scheduling mathematical model in a preset solver to obtain the optimal scheduling method that can eliminate this anomaly.

[0055] In some embodiments of this application, constructing a scheduling model for the system to be scheduled under the abnormal state based on the target scheduling sequence includes: obtaining various decision variables in the scheduling model, the objective function of the scheduling model, and the constraints of the scheduling model, such as... Figure 4 As shown, the multiple decision variables include at least one of the following: sequential identifier binary variables, continuous adjustment variables, and relaxation variables. The sequential identifier binary variables are used to characterize the adjustment order relationship between the scheduling objects. The continuous adjustment variables are used to characterize the adjustment amount of the scheduling objects as continuous variables. The relaxation variables are used to characterize the relaxation amount of the constraints (the degree of violation of the constraints of the scheduling objects). The objective function is constructed with the goal of minimizing the number of scheduling objects to be adjusted. The scheduling model of the system to be scheduled is constructed based on the multiple decision variables in the scheduling model, the objective function of the scheduling model, and the constraints of the scheduling model.

[0056] Specifically, sequential identification of binary variables: Introduction ∈{0,1} (where k = 0, 1, 2, 3, ... are the order indices of the adjustment objects), used to precisely characterize the adjustment order relationship between scheduling objects. The physical meaning of this variable is as follows:

[0057] when When = 0, it means that the scheduling objects in the corresponding order have not yet started to be adjusted;

[0058] when When = 1, it means that the previous scheduled object has been adjusted to its limit, and the current scheduled object officially begins to be adjusted.

[0059] Application example: For a scenario involving the sequential adjustment of scheduling object A and scheduling object B (i.e., A is adjusted to its limit before B is adjusted), introduce binary variables:

[0060] = 0: This indicates that scheduling object B has not started adjustment, and the system is in the stage of adjusting A;

[0061] = 1: This indicates that scheduling object A has reached its limit, and the system begins to adjust scheduling object B.

[0062] Continuous adjustment variable: Define the adjustment amount of the scheduling object as a continuous variable. , (i = 0, 1, 2... represents the medium type index, j = 0, 1, 2... represents the scheduling object index), the physical meaning is the adjustment amount of the j-th scheduling object on the i-th medium.

[0063] Slack variables: To prevent the model from failing to find a feasible solution due to overly strict equality constraints, nonnegative slack variables are introduced. and (Where i = 0, 1, 2… represents the constraint index), and “+” and “-” represent the positive and negative relaxation amounts of the corresponding constraints, respectively. The introduction of relaxation variables significantly enhances the robustness and numerical stability of the model.

[0064] In some embodiments of this application, the objective function of the scheduling model is determined by: obtaining the objective parameters of the system to be scheduled, wherein the objective parameters include: the number of scheduling objects, the adjustment amount of the scheduling objects, and the degree of violation of the constraints of the scheduling objects; and constructing the objective function based on the objective parameters of the system to be scheduled.

[0065] The objective function is shown in the following equation:

[0066] min + +

[0067] In the formula, and Preset weighting coefficients are used to balance primary and secondary objectives. This indicates the number of objects involved in the adjustment. This represents the k-th adjustment object. Indicates the adjustment amount. Indicates the total amount of adjustment. It represents the total amount of relaxation.

[0068] It should also be noted that the constraints of the scheduling model include: material balance constraints, adjustment capability constraints of the scheduling object, mathematical model constraints of the scheduling object, and scheduling logic constraints. Among them, the material balance constraints include: the materials of the system to be scheduled remain balanced after the scheduling operation. The adjustment capability constraints of the scheduling object include: all adjustment operations must be within the safe operating range of the equipment. The mathematical model constraints of the scheduling object include: the states between different scheduling objects satisfy a preset correlation relationship. The scheduling logic constraints are quantified using the Big M method.

[0069] Specifically, material balance constraints are used to indicate that the system's materials remain balanced after any scheduling operation:

[0070] - = Δ j

[0071] In the formula, j represents the index of the scheduling object. Δ j This represents the amount of imbalance in the i-th medium caused by the anomaly.

[0072] The adjustment capability constraint of the scheduling object indicates that all adjustment operations must be performed within the safe operating range of the equipment:

[0073] ≤ ≤

[0074] in, and Let represent the minimum and maximum adjustment amounts of medium i in the j-th scheduling object, respectively.

[0075] The mathematical model constraints of the scheduling objects indicate that the states of different scheduling objects satisfy a preset association relationship:

[0076] = a + b

[0077] In the formula, a is the proportional coefficient and b is a constant term, both of which are determined by the characteristics of the scheduling object.

[0078] Scheduling logic constraints can be quantified using the Big M method:

[0079] ≥ - M (1 - )

[0080] ≤

[0081] ≥

[0082] Where M represents a positive number greater than a preset threshold.

[0083] For example:

[0084] ≥ - M (1 - )

[0085] ≤

[0086] ≥

[0087] The above constraints indicate that: when When it is 1 (meaning) (Already adjusted to the limit) It can be adjusted freely. It must be equal to its upper limit.

[0088] when When it is 0 (meaning) (Not yet adjusted to the limit) The adjustment is restricted and must be 0.

[0089] In some embodiments of this application, the specific steps for determining the scheduling strategy of the system to be scheduled under the abnormal state based on the scheduling model are as follows: The attributes of integer variables in the scheduling model are converted into continuous variables to obtain an adjusted scheduling model; a linear programming solver is used to solve the adjusted scheduling model to obtain an initial solution, which is a relaxed solution; when the values ​​of the integer variables in the initial solution are non-integer, a branch and bound algorithm is used to branch the integer variables in the initial solution to obtain multiple branches until the branch and bound algorithm converges, thus obtaining the optimal solution of the scheduling model. The optimal solution of the scheduling model is used to characterize the scheduling strategy of the system to be scheduled, and the scheduling strategy includes at least the adjustment order and adjustment amount of multiple scheduling objects of the system to be scheduled.

[0090] The branching operation focuses on integer variables and breaks down the solution space into subproblems by selecting integer variables whose values ​​are not integers. For example: The variable acts as a "branch point," splitting the original solution space into two mutually exclusive subproblems. This forces the branch variable to converge towards integer values. The specific rules are based on the scheduling logic design of this patent: Branch variable selection strategy: Prioritize selecting the variable that has the greatest impact on the scheduling order. Variables, namely, those identified through sensitivity analysis. The variable whose tiny changes cause the largest fluctuation in the objective function value. For example, in a serial scenario where scheduling objects A→B→C occur, if the relaxation solution... =0.6 (the switching indicator from A to B), which affects subsequent... The value of this parameter and the total number of adjustment devices have a significant impact; therefore, priority should be given to... For branching variables. Subproblem construction rules: ... Taking 0.6 as an example, the original problem is split into two subproblems: Subproblem 1: Adding constraints (A is already at its limit, B can be started for adjustment); Sub-problem 2: Add constraints =0 (forces A not to be adjusted to its limit, B is prohibited from adjustment). Both subproblems fully inherit the constraints of the original model, and their solution spaces do not overlap, together covering all feasible solutions of the original problem. Linkage with scheduling logic: Branch operations are deeply bound to the scheduling logic constraints in the implementation examples, for example: in subproblem 1... The constraint = 1 will be propagated through the Big M method to... x i,j The value of is determined by scheduling logic constraints. ≥ ,at this time Available in [ , The value can be freely selected within the range, which is completely consistent with the physical meaning of "B start-up adjustment".

[0091] The method employs a branch and bound algorithm to branch the integer variables in the initial solution, resulting in multiple branches until the algorithm converges. This includes: sequentially identifying non-integer integer variables in the initial solution as branch variables; performing branch operations on these branch variables to obtain corresponding subproblems; where the priority of each branch operation on a non-integer integer variable is determined based on its influence on the objective function value; adding the constraints of the scheduling model to the subproblems; solving the subproblems to obtain their solutions; determining the upper and lower bounds of the subproblems based on their solutions; pruning the subproblems if their upper bound is not less than the current global upper bound; and pruning the subproblems if their solutions do not meet the constraints of the scheduling model.

[0092] In one alternative approach, pruning is key to improving the efficiency of the branch and bound method. By comparing the lower bound of a subproblem with the upper bound of the global bound, branches that cannot contain the optimal solution are eliminated. Specifically, in the scheduling method of this application: if the lower bound of a subproblem is greater than or equal to the current upper bound of the global bound, then the branch cannot find a better result than the existing optimal solution, and the subsequent solution of the branch is directly terminated. For example, if the lower bound of a subproblem is 7.5, and the current UB (global lower bound) is 7.0, then the branch is pruned. To perform pruning operations more efficiently, this application also provides an efficient pruning method that combines constraints: the scheduling object's adjustment capability constraint and the material balance constraint provide additional basis for pruning. For example, in a subproblem... The constraint =1 leads to The value exceeds [ , If the subproblem has no feasible solution, it can be pruned directly without further solution. The pruning logic for multi-level branches is as follows: as the branch depth increases, the constraints on the subproblems increase, and the lower bound gradually rises. For example, in... Under the branch where =1, further... z BC Branching, if subproblems =1、 If the lower bound of =1 is 6.5 (less than UB=7.0), then further branching is needed; if the subproblem =1、 If the lower bound of =0 is 7.2 (greater than UB=7.0), then prune directly.

[0093] It should be noted that the specific steps for determining the upper and lower bounds of the subproblem based on the solution results are as follows: If the solution result of the subproblem is a feasible solution, the objective function value corresponding to the feasible solution is determined to be the lower bound of the subproblem; if the solution result of the subproblem is an integer solution, the objective function value corresponding to the integer solution is determined to be the upper bound of the subproblem, wherein the solution process converges in the direction of minimizing the value of the branch variable.

[0094] The bounding calculation involves iteratively solving subproblems and updating the upper and lower bounds. By solving each branch subproblem, the upper and lower bounds of the objective function are obtained, providing a basis for subsequent pruning. The core is to establish the convergence interval of "global upper bound (UB) - global lower bound (LB)". Specific applications are as follows: Lower bound update: After solving each subproblem, if a feasible solution is obtained, its objective function value is the lower bound of that subproblem. The maximum value among the lower bounds of all subproblems is updated to the global lower bound LB, representing the theoretical minimum of the global optimal solution. Upper bound update: If the solution result of a certain subproblem is such that all... All variables are integers (i.e., integer feasible solutions are obtained), and their objective function values ​​will serve as candidate values ​​for the global upper bound UB. Initially, UB is set to infinity. When an integer feasible solution is obtained for the first time (e.g., the objective function value is 7.0), UB is updated to 7.0, representing the currently found optimal feasible solution. The bounding process always revolves around the core optimization objective of the scheduling method implemented in this application (minimizing the number of adjustment devices). In the objective function... The term with the highest weight will be prioritized during branch solving. The convergence direction is minimized by minimizing the number of values ​​of 1, ensuring that the upper bound update more closely reflects actual needs. It's understandable that a feasible solution is one that satisfies all constraints.

[0095] Specifically, before solving the model, the following preprocessing strategies are adopted to improve the solution efficiency: Boundary contraction: Automatically tighten the upper and lower bounds of variables according to the constraint relationship to reduce the search space; Redundant constraint elimination: Identify and remove unnecessary constraints; Large M coefficient optimization: Determine the optimal M value according to the range of variables to avoid numerical problems, and use a preset solver to solve the scheduling mathematical model to obtain the optimal scheduling method that can eliminate this anomaly.

[0096] For example, mathematical programming algorithms can be used to efficiently solve a constructed Mixed Integer Linear Programming (MILP) model (scheduling model). Since the problem simultaneously contains binary variables (number of objects to be adjusted) and continuous variables (adjustment amounts of the scheduled objects), it belongs to the standard MILP problem type. Optionally, a branch and bound algorithm can be used to solve the problem: first, ignore the integer constraint, relax the original MILP problem into a linear programming (LP) problem and solve it. If the integer variables in the relaxed solution are not integers, then select a variable to branch, decomposing the original problem into two subproblems. By continuously updating the current optimal solution and pruning inferior branches, the solution space is searched efficiently. The algorithm terminates when all branches are either pruned or integer solutions are obtained. To improve the solution efficiency, various acceleration techniques can be used, such as: heuristic initial solution generation: using expert rules to quickly generate high-quality initial solutions and shorten the solution time; cutting plane technique: dynamically adding effective inequalities to tighten the solution space of the relaxed problem; parallel computing: using multi-core processors to process multiple branch nodes in parallel. After the solver outputs the optimal solution, the system performs the following post-processing: Feasibility verification: ensuring the obtained solution meets all process constraints and equipment limitations. Sensitivity analysis: assessing the impact of changes in key parameters on the scheduling scheme. Scheme visualization: transforming the mathematical solution into intuitive operation instructions and flowcharts. The final output includes: Optimal scheduling scheme: clearly defining the adjustment order and adjustment amount for each scheduling object.

[0097] To better illustrate the process of the scheduling model outputting the scheduling strategy in the embodiments of this application, the following further explains the process of the scheduling model outputting the scheduling strategy:

[0098] Taking a pipeline network consisting of hydrogen pipelines and steam systems as an example, abnormal operating conditions lead to a shortage of 12,000 Nm³ / h in the hydrogen pipeline network and a shortage of 50 t / h in the steam system. Figure 5 A simplified equilibrium diagram of the system was constructed, in which,

[0099] Hydrogen medium: Involves streams 1 to 5. Streams 1 and 2 are hydrogen-producing streams, while streams 3, 4, and 5 are hydrogen-consuming streams.

[0100] Steam medium: This involves streams 6 to 8. Stream 6 is the steam-producing stream, while streams 7 and 8 are steam-consuming streams.

[0101] When an anomaly occurs, the imbalance needs to be quickly eliminated by coordinating five main production / consumption scheduling objects to ensure system safety. This includes: Step 1: Data acquisition and system modeling.

[0102] First, the precise imbalance caused by the anomaly is obtained (hydrogen: -12000 Nm³ / h; steam: -50 t / h, where a positive sign indicates a surplus and a negative sign indicates a deficit). Simultaneously, basic production data for five scheduling objects is collected from the real-time database, including:

[0103] Real-time flow rates of each stream (e.g., current hydrogen production and consumption), maximum / minimum adjustment capabilities of each scheduling object, and the internal mathematical model of the scheduling object, combined with... Figure 5 The system balance diagram clarifies the production and consumption relationships and mutual coupling relationships of each scheduling object in the hydrogen and steam network, laying the data foundation for constructing a mathematical optimization model. Step Two: Scheduling Logic Topology and Variable Creation:

[0104] based on Figure 6 The expert scheduling logic shown (a proven priority order) topologically identifies the unique adjustment path (target scheduling sequence) under this abnormal operating condition. This logic determines the adjustment order and coupling relationship of the devices, for example: first, increase the load of device 1 to its limit; if it is still insufficient, then increase the load of device 2 according to priority; finally, decrease the load of device 3.

[0105] Based on this logic, three types of decision variables are created:

[0106] Sequential identifier binary variables ( For example, introducing Indicates whether device 1 has reached its load limit and whether device 2 can be activated for adjustment; (Introduction) This indicates whether device 2 has reached its limit and whether device 3 can be activated for adjustment.

[0107] Continuous adjustment variables ( ):For example, The amount of hydrogen production adjustment for device 1; This represents the adjustment amount of steam production due to the adjustment of by-products.

[0108] Slack variables ( ): Introducing slack variables to the material balance constraints allows for small deviations to ensure the model's feasibility under extreme conditions.

[0109] Step 3: Constructing the mixed-integer programming model:

[0110] Based on the above variables, construct a complete MILP model (scheduling model):

[0111] Objective function: Minimize z1 + z2. The core objective is to minimize the number of devices requiring action, and secondly, to minimize the total adjustment (| |) and constraint violation amount ( ).

[0112] Material balance constraints:

[0113] = 12000

[0114] = 50

[0115] in, , , , , These represent the hydrogen production and consumption flags of the scheduling object j (where 1 represents a hydrogen production unit and -1 represents a hydrogen consumption unit), the current value of the hydrogen flow of the scheduling object j, the adjustment amount of the hydrogen flow of the scheduling object j, and the negative and positive relaxation amounts, respectively. , , , , These represent the steam flow production and consumption identifier of the scheduling object j, the current value of the steam flow of the scheduling object j, the adjustment amount of the steam flow of the scheduling object j, and the negative and positive relaxation amounts, respectively.

[0116] Scheduled object capability constraints:

[0117] ≤ + ≤ ,

[0118] in, , , , These are, respectively, the lower limit of hydrogen flow stream for scheduling object j, the current value of hydrogen flow stream for scheduling object j, the adjustment amount of hydrogen flow stream for scheduling object j, and the upper limit of hydrogen flow stream for device j.

[0119] Mathematical model constraints for scheduling objects:

[0120] = a + b

[0121] The linear coupling relationship within the device is transformed into constraints, where a and b are constant coefficients.

[0122] Scheduling logic constraints (Big M method):

[0123] ≥ - M (1 - (1)

[0124] ≤ (2)

[0125] ≥ (3)

[0126] Equation (1) indicates that if =1, then device 1 must be at full load. Equations (2) and (3) indicate that if If the value is 0 (device 1 is not full), then device 2 is prohibited from starting adjustment. Subsequent scheduling logic sequence constraints are similar.

[0127] Step 4: Model Solving and Solution Generation

[0128] Input the constructed MILP model into the solver. The solver automatically performs the mathematical solution using the branch and bound method.

[0129] First, ignoring integer constraints, we solve the linear relaxation problem to obtain a lower bound. Then, we decompose the problem into branches. =0 and =1. The search proceeds through two subproblems, continuously pruning by "bounding" to eliminate branches that are clearly worse than the current solution. Ultimately, the global optimum that minimizes the objective function is found in the solution space satisfying all integer and linear constraints.

[0130] The solver output is as follows: First, increase the load of Unit 1 to its maximum (increase hydrogen production by 8,000 Nm³ / h, and simultaneously produce an additional 30 t / h of steam). Then, increase the load of Unit 2 by 80% (increase hydrogen production by 4,000 Nm³ / h, and simultaneously produce an additional 20 t / h of steam). This scheme utilizes a total of two scheduling objects, perfectly eliminating the hydrogen shortage and synergistically eliminating the 50 t / h steam shortage.

[0131] Figure 7 An industrial media dispatching device is shown, the device comprising:

[0132] The receiving module 70 is used to receive media data in the system to be scheduled under abnormal conditions and production data of multiple scheduling objects in the system to be scheduled, wherein the media data includes at least the imbalance quantity of various media.

[0133] The determining module 72 is used to determine the target scheduling sequence of the system to be scheduled based on the medium data in the system to be scheduled, the production data of multiple scheduling objects in the system to be scheduled, and the pre-acquired scheduling rules;

[0134] Construction module 74 is used to construct a scheduling model of the system to be scheduled under the abnormal state based on the target scheduling sequence;

[0135] The scheduling module 76 is used to determine the scheduling strategy of the system to be scheduled under the abnormal state according to the scheduling model.

[0136] The aforementioned industrial media scheduling device receives media data from a system under abnormal conditions and production data of multiple scheduling objects within that system. The media data includes at least the imbalance quantities of various media. The device then determines a target scheduling sequence for the system under abnormal conditions based on the media data, the production data of the multiple scheduling objects, and pre-acquired scheduling rules. It constructs a scheduling model for the system under abnormal conditions based on the target scheduling sequence. Finally, it determines a scheduling strategy for the system under abnormal conditions based on the scheduling model. By adjusting the scheduling sequence through the constructed scheduling model, the device achieves the goal of adjusting the scheduling strategy, thereby improving scheduling accuracy. This solves the technical problem of low scheduling accuracy caused by manual decision-making in industrial media scheduling in related technologies.

[0137] The receiving module 70 includes: a receiving submodule, used to receive media data from the system to be scheduled under abnormal conditions and production data of multiple scheduling objects in the system to be scheduled, wherein the media data includes at least: the imbalance quantity of various media; determining the target scheduling sequence of the system to be scheduled based on the media data of the system to be scheduled, the production data of multiple scheduling objects in the system to be scheduled, and pre-acquired scheduling rules; constructing a scheduling model of the system to be scheduled under abnormal conditions based on the target scheduling sequence; and determining the scheduling strategy of the system to be scheduled under abnormal conditions based on the scheduling model.

[0138] The determination module 72 includes: a determination submodule, used to construct a scheduling model of the system to be scheduled under the abnormal state based on the target scheduling sequence, including: obtaining multiple decision variables, the objective function of the scheduling model, and the constraints of the scheduling model, wherein the multiple decision variables include at least one of the following: sequential identifier binary variables, continuous adjustment variables, and relaxation variables, wherein the sequential identifier binary variables are used to characterize the adjustment order relationship between the scheduling objects, the continuous adjustment variables are used to characterize the adjustment amount of the scheduling objects as continuous variables, the relaxation variables are used to characterize the relaxation amount of the constraints, and the objective function is constructed with the goal of minimizing the number of scheduling objects; constructing the scheduling model of the system to be scheduled based on the multiple decision variables, the objective function of the scheduling model, and the constraints of the scheduling model, wherein the objective function of the scheduling model... The objective function is determined as follows: The target parameters of the system to be scheduled are obtained, including the number of scheduling objects, the adjustment amount of the scheduling objects, and the degree of violation of the constraints of the scheduling objects. The objective function is constructed based on the target parameters of the system to be scheduled. The constraints of the scheduling model include: material balance constraints, adjustment capability constraints of the scheduling objects, mathematical model constraints of the scheduling objects, and scheduling logic constraints. The material balance constraints include: the materials in the system to be scheduled remain balanced after scheduling operations. The adjustment capability constraints of the scheduling objects include: all adjustment operations must be within the safe operating range of the equipment. The mathematical model constraints of the scheduling objects include: the states of different scheduling objects satisfy a preset correlation. The scheduling logic constraints are quantified using the Big M method.

[0139] The scheduling module 76 includes a scheduling submodule, used to determine the scheduling strategy of the system to be scheduled under the abnormal state according to the scheduling model, including: converting the attributes of integer variables in the scheduling model into continuous variables to obtain an adjusted scheduling model; solving the adjusted scheduling model using a linear programming solver to obtain an initial solution, wherein the initial solution is a relaxed solution; when the values ​​of the integer variables in the initial solution are non-integer, using a branch and bound algorithm to perform branch operations on the integer variables in the initial solution to obtain multiple branches until the branch and bound algorithm converges to obtain the optimal solution of the scheduling model, wherein the optimal solution of the scheduling model is used to characterize the scheduling strategy of the system to be scheduled, wherein the scheduling strategy includes at least: the adjustment order and adjustment amount of multiple scheduling objects of the system to be scheduled, wherein using a branch and bound algorithm to perform branch processing on the integer variables in the initial solution to obtain multiple branches until the branch and bound algorithm converges includes: sequentially determining the non-integer integer variables in the initial solution as branch variables, performing branch operations on the branch variables to obtain the branch subproblems corresponding to the branch variables, wherein each non-integer integer variable is a branch variable. The priority of branching operations on the variable is determined based on the degree of influence of the non-integer integer variable on the objective function value; the constraints of the scheduling model are added to the branch subproblem, the branch subproblem is solved, and the solution result of the branch subproblem is obtained; the upper bound and the lower bound of the branch subproblem are determined based on the solution result of the branch subproblem; if the upper bound of the branch subproblem is not less than the global upper bound at the current time, the branch subproblem is pruned; if the solution result of the branch subproblem does not meet the constraints of the scheduling model, the branch subproblem is pruned, and the upper bound and the lower bound of the branch subproblem are determined based on the solution result of the branch subproblem, including: if the solution result of the branch subproblem is a feasible solution, determining the objective function value corresponding to the feasible solution as the lower bound of the branch subproblem; if the solution result of the branch subproblem is an integer solution, determining the objective function value corresponding to the integer solution as the upper bound of the branch subproblem, wherein, during the solution of the branch subproblem, convergence is directed towards the direction of minimizing the value of the branch variable.

[0140] It should be noted that, Figure 7 The industrial media dispatching device shown is used to perform Figure 2 The industrial media scheduling method shown above also applies to this industrial media scheduling device, and will not be repeated here.

[0141] This application also provides a computer device, including: a memory and a processor, wherein the memory is used to store program instructions; and the processor, connected to the memory, is used to execute the above-described industrial media scheduling method.

[0142] This application also provides a computer program product, including computer instructions that, when executed by a processor, implement the steps of the industrial media scheduling method of this application.

[0143] The sequence numbers of the embodiments in this application are for descriptive purposes only and do not represent the superiority or inferiority of the embodiments.

[0144] In the above embodiments of this application, the descriptions of each embodiment have different focuses. For parts not described in detail in a certain embodiment, please refer to the relevant descriptions of other embodiments.

[0145] In the several embodiments provided in this application, it should be understood that the disclosed technical content can be implemented in other ways. The device embodiments described above are merely illustrative; for example, the division of units can be a logical functional division, and in actual implementation, there may be other division methods. For instance, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the displayed or discussed mutual coupling, direct coupling, or communication connection may be through some interfaces; the indirect coupling or communication connection between units or modules may be electrical or other forms.

[0146] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0147] Furthermore, the functional units in the various embodiments of this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit.

[0148] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of this application. The aforementioned storage medium includes various media capable of storing program code, such as a USB flash drive, read-only memory (ROM), random access memory (RAM), portable hard drive, magnetic disk, or optical disk.

[0149] The above description is only a preferred embodiment of this application. It should be noted that for those skilled in the art, several improvements and modifications can be made without departing from the principle of this application, and these improvements and modifications should also be considered within the scope of protection of this application.

Claims

1. An industrial media scheduling method, characterized in that, include: Receive media data from the system to be scheduled under abnormal conditions and production data of multiple scheduling objects in the system to be scheduled, wherein the media data includes at least: the number of imbalances of various media. The target scheduling sequence of the system to be scheduled is determined based on the media data in the system to be scheduled, the production data of multiple scheduling objects in the system to be scheduled, and the pre-acquired scheduling rules. Construct a scheduling model for the system to be scheduled under the abnormal state based on the target scheduling sequence; The scheduling strategy for the system to be scheduled under the abnormal state is determined based on the scheduling model. Constructing a scheduling model for the system to be scheduled under the abnormal state based on the target scheduling sequence includes: The scheduling model is obtained by acquiring multiple decision variables, the objective function of the scheduling model, and the constraints of the scheduling model. The multiple decision variables include at least one of the following: sequential identifier binary variables, continuous adjustment variables, and relaxation variables. The sequential identifier binary variables are used to characterize the adjustment order relationship between the scheduling objects. The continuous adjustment variables are used to characterize the adjustment amount of the scheduling objects as continuous variables. The relaxation variables are used to characterize the relaxation amount of the constraints. The objective function is constructed with the goal of minimizing the number of scheduling objects to be adjusted. The scheduling model of the system to be scheduled is constructed based on the various decision variables in the scheduling model, the objective function of the scheduling model, and the constraints of the scheduling model. The constraints of the scheduling model include: The system includes material balance constraints, adjustment capability constraints of the scheduling object, mathematical model constraints of the scheduling object, and scheduling logic constraints. The material balance constraints include: the materials in the system to be scheduled remain balanced after a scheduling operation. The adjustment capability constraints of the scheduling object include: all adjustment operations must be within the safe operating range of the equipment. The mathematical model constraints of the scheduling object include: the states of different scheduling objects satisfy a preset correlation. The scheduling logic constraints are quantified using a Big M method. In the initial solution of the adjusted scheduling model, integer variables with non-integer values ​​are sequentially determined as branch variables. Branching operations are performed on the branch variables to obtain the branch subproblems corresponding to the branch variables. The priority of the branching operation for each non-integer integer variable is determined based on the degree of influence of the non-integer integer variable on the objective function value. The constraints of the scheduling model are added to the branch subproblem, the branch subproblem is solved, and the solution result of the branch subproblem is obtained. The upper bound and the lower bound of the subproblem are determined based on the solution results of the subproblem. If the upper bound of the branch subproblem is not less than the global upper bound at the current time, the branch subproblem is pruned. If the solution to the subproblem does not meet the constraints of the scheduling model, the subproblem will be pruned.

2. The method according to claim 1, characterized in that, The objective function of the scheduling model is determined in the following way: The target parameters of the system to be scheduled are obtained respectively, wherein the target parameters include: the number of scheduling objects, the adjustment amount of the scheduling objects, and the degree of violation of the constraints of the scheduling objects; The objective function is constructed based on the target parameters of the system to be scheduled.

3. The method according to claim 1, characterized in that, Determining the scheduling strategy of the system to be scheduled under the abnormal state based on the scheduling model includes: The properties of integer variables in the scheduling model are converted into continuous variables to obtain the adjusted scheduling model. The adjusted scheduling model is solved using a linear programming solver to obtain an initial solution, which is a relaxed solution. When the integer variables in the initial solution have non-integer values, a branch and bound algorithm is used to branch the integer variables in the initial solution to obtain multiple branches until the branch and bound algorithm converges, thus obtaining the optimal solution of the scheduling model. The optimal solution of the scheduling model is used to characterize the scheduling strategy of the system to be scheduled. The scheduling strategy includes at least the adjustment order and adjustment amount of multiple scheduling objects of the system to be scheduled.

4. The method according to claim 1, characterized in that, Determining the upper bound and lower bound of the subproblem based on the solution results of the subproblem includes: If the solution to the subproblem is a feasible solution, the objective function value corresponding to the feasible solution is determined to be a lower bound of the subproblem. If the solution to the subproblem is an integer solution, the objective function value corresponding to the integer solution is determined to be an upper bound of the subproblem, wherein the subproblem is solved by converging in the direction of minimizing the value of the branch variable.

5. An industrial media dispatching device, characterized in that, include: The receiving module is used to receive media data in the system to be scheduled under abnormal conditions and production data of multiple scheduling objects in the system to be scheduled, wherein the media data includes at least the imbalance quantity of various media. The determination module is used to determine the target scheduling sequence of the system to be scheduled based on the media data in the system to be scheduled, the production data of multiple scheduling objects in the system to be scheduled, and the pre-acquired scheduling rules; The construction module is used to construct a scheduling model of the system to be scheduled under the abnormal state based on the target scheduling sequence; The scheduling module is used to determine the scheduling strategy of the system to be scheduled under the abnormal state based on the scheduling model. Constructing a scheduling model for the system to be scheduled under the abnormal state based on the target scheduling sequence includes: The scheduling model is obtained by acquiring multiple decision variables, the objective function of the scheduling model, and the constraints of the scheduling model. The multiple decision variables include at least one of the following: sequential identifier binary variables, continuous adjustment variables, and relaxation variables. The sequential identifier binary variables are used to characterize the adjustment order relationship between the scheduling objects. The continuous adjustment variables are used to characterize the adjustment amount of the scheduling objects as continuous variables. The relaxation variables are used to characterize the relaxation amount of the constraints. The objective function is constructed with the goal of minimizing the number of scheduling objects to be adjusted. The scheduling model of the system to be scheduled is constructed based on the various decision variables in the scheduling model, the objective function of the scheduling model, and the constraints of the scheduling model. The constraints of the scheduling model include: The system includes material balance constraints, adjustment capability constraints of the scheduling object, mathematical model constraints of the scheduling object, and scheduling logic constraints. The material balance constraints include: the materials in the system to be scheduled remain balanced after a scheduling operation. The adjustment capability constraints of the scheduling object include: all adjustment operations must be within the safe operating range of the equipment. The mathematical model constraints of the scheduling object include: the states of different scheduling objects satisfy a preset correlation. The scheduling logic constraints are quantified using a Big M method. In the initial solution of the adjusted scheduling model, integer variables with non-integer values ​​are sequentially determined as branch variables. Branching operations are performed on the branch variables to obtain the branch subproblems corresponding to the branch variables. The priority of the branching operation for each non-integer integer variable is determined based on the degree of influence of the non-integer integer variable on the objective function value. The constraints of the scheduling model are added to the branch subproblem, the branch subproblem is solved, and the solution result of the branch subproblem is obtained. The upper bound and the lower bound of the subproblem are determined based on the solution results of the subproblem. If the upper bound of the branch subproblem is not less than the global upper bound at the current time, the branch subproblem is pruned. If the solution to the subproblem does not meet the constraints of the scheduling model, the subproblem will be pruned.

6. A computer device, characterized in that, include: A memory and a processor, wherein the memory is used to store program instructions; The processor, connected to the memory, is used to execute the industrial media scheduling method according to any one of claims 1 to 4.

7. A computer program product comprising computer instructions, characterized in that, When the computer instructions are executed by the processor, they implement the industrial media scheduling method according to any one of claims 1 to 4.

Images