A global energy consumption optimization method and device for an industrial production process

Abstract

This invention provides a method, apparatus, and computer-readable storage medium for global energy consumption optimization in industrial production processes. The global energy consumption optimization method includes the following steps: identifying multiple production units involved in energy consumption during the industrial production process; defining each production unit as an energy consumption node and determining an energy consumption model for each production unit; defining production units subject to constraints as constraint nodes and determining constraint models for the constraint-constrained production units; and performing distributed optimization of each production unit through the transmission of energy consumption and constraint information between the nodes to achieve global optimization of the total energy consumption of the industrial production process. By performing these steps, the global energy consumption optimization method simplifies the structure of the optimization model, reduces the modeling difficulty, and improves the optimization efficiency and real-time performance of the industrial production process.

Description

Technical Field

[0001] This invention relates to optimization techniques for industrial production processes, and more particularly to a global energy consumption optimization method for industrial production processes, a global energy consumption optimization device for industrial production processes, and a computer-readable storage medium. Background Technology

[0002] Modern industry is gradually developing towards large-scale and refined production, placing higher demands on efficient, energy-saving, and high-quality production. In accordance with relevant strategic requirements, improving the optimized operation level of my country's continuous industrial manufacturing and achieving full-process production optimization is urgently needed.

[0003] Continuous industrial production, such as in petroleum and chemical industries, typically involves multiple industrial units that process raw materials into intermediate products needed for the next stage of production. These multiple units form a complete production line. However, continuous industries often involve multiple production units, resulting in long processes, large scale, complex production conditions, and coupled operational variables. Traditional centralized optimization methods face challenges in optimizing the production processes of continuous industries, including model complexity and convergence difficulties.

[0004] In order to overcome the above-mentioned problems of existing technologies, there is an urgent need in this field for an optimization technology for industrial production processes, which can simplify the structure of optimization models, reduce the modeling difficulty of optimization, and improve the optimization efficiency and real-time performance of industrial production processes. Summary of the Invention

[0005] The following provides a brief overview of one or more aspects to offer a basic understanding of them. This overview is not an exhaustive summary of all conceived aspects, nor is it intended to identify key or decisive elements of all aspects, nor to define the scope of any or all aspects. Its sole purpose is to present some concepts of one or more aspects in a simplified form as a prelude to the more detailed descriptions that follow.

[0006] To overcome the aforementioned problems in the existing technology, the present invention provides a global energy consumption optimization method for industrial production processes, a global energy consumption optimization device for industrial production processes, and a computer-readable storage medium, which can simplify the structure of the optimization model, reduce the modeling difficulty of optimization, and improve the optimization efficiency and real-time performance of industrial production processes.

[0007] Specifically, the global energy consumption optimization method for the above-mentioned industrial production process provided by the first aspect of the present invention includes the following steps: identifying multiple production units involved in energy consumption in the industrial production process; identifying each production unit as an energy consumption node and determining the energy consumption model of each production unit; identifying production units involving constraints as constraint nodes and determining the constraint model of the production units involving constraints; and performing distributed optimization of each production unit through the transmission of energy consumption information and constraint information between the nodes, so as to achieve global optimization of the total energy consumption of the industrial production process.

[0008] Furthermore, in some embodiments of the present invention, the step of performing distributed optimization of each production unit via global optimization among the nodes to reduce the total energy consumption of the industrial production process includes: based on the local decision variable x of each production unit. i Determine the energy consumption objective function f for each of the production units. i (x i According to the energy consumption objective function f of each production unit. i (x i Determine the distributed optimization function for the total energy consumption. And through communication between the nodes, distributed optimization is performed on each production unit under the constraints based on the distributed optimization function to reduce the total energy consumption of the industrial production process.

[0009] Furthermore, in some embodiments of the present invention, before performing distributed optimization of each of the production units, the global energy consumption optimization method further includes the following step: performing sensitivity analysis on each of the production units to determine the local decision variable x of each of the production units. i .

[0010] Furthermore, in some embodiments of the present invention, the local decision variable x i This includes the process operation parameters and / or raw material flow rate of the production unit.

[0011] Furthermore, in some embodiments of the present invention, the constraints include local variable range constraints, separable coupling equation constraints between production units, and / or rated operating constraints of production units. The local variable range constraints include... in, It is a local decision variable x i The minimum value, It is a local decision variable x i The maximum value of . The separable coupling equality constraint includes . Among them, h iIt is the coupling equality constraint function between production unit i and its neighboring nodes. The rated working constraints include g. c (x c )≤W c , where g c (x c ) is the rated working constraint function of production unit c, x c These are the local decision variables of the nominal working constraint function.

[0012] Furthermore, in some embodiments of the present invention, the step of performing distributed optimization on each of the production units under the constraints based on the distributed optimization function via communication between the nodes to reduce the total energy consumption of the industrial production process includes: before the start of each iteration, obtaining the indivisible constraint information of the neighboring constraint nodes via communication between each energy-consuming node and its neighboring constraint nodes. And through communication between each energy-consuming node and its neighboring energy-consuming nodes, the node information of the neighboring energy-consuming nodes is obtained. To determine the equality constraint information for each of the energy consumption nodes. The dual variables of each energy-consuming node are updated using the dual ascent method. Furthermore, through mutual communication between the energy-consuming nodes, the dual variables of the neighboring nodes of each node are obtained, and the dual variables are updated within each node using gradient projection and gradient descent methods. in, It is a primitive variable. It is an auxiliary variable. It is a dual variable.

[0013] Furthermore, in some embodiments of the present invention, after updating the dual variables, the global energy consumption optimization method further includes the following step: summarizing the local variables of the energy consumption nodes of the neighboring nodes of the constrained node. To determine the local decision variable x of the constraint node. c And calculate the rated working constraint function g. c (x c )-W c The value of the constraint; and the result of returning to the neighboring constraint nodes of the constraint node whether they satisfy the constraint restriction, and the gradient of the nominal working constraint function.

[0014] Furthermore, in some embodiments of the present invention, after performing distributed optimization on each of the production units under the constraints based on the distributed optimization function via communication between the nodes, the global energy consumption optimization method further includes the following step: based on the local decision variable x after the distributed optimization... i The process calculates the total energy consumption of the industrial production process; determines whether the distributed optimization has converged based on the trend of the total energy consumption; proceeds to the next iteration in response to the determination that the distributed optimization has not converged; and, in response to the determination that the distributed optimization has converged, determines the local decision variable x based on the following: i The value of is used to determine the optimal operating conditions for the industrial production process.

[0015] Furthermore, in some embodiments of the present invention, the communication between the nodes includes directed graphs and undirected graphs.

[0016] Furthermore, the global energy consumption optimization apparatus for the industrial production process provided in the second aspect of the present invention includes a memory and a processor. The processor is connected to the memory and is configured to implement the global energy consumption optimization method for the industrial production process provided in the first aspect of the present invention.

[0017] Furthermore, the computer-readable storage medium provided in the third aspect of the present invention stores computer instructions thereon. When the computer instructions are executed by a processor, the global energy consumption optimization method for the industrial production process provided in the first aspect of the present invention is implemented. Attached Figure Description

[0018] The above-described features and advantages of the present invention will be better understood after reading the following detailed description of embodiments of the present disclosure in conjunction with the accompanying drawings. In the drawings, components are not necessarily drawn to scale, and components having similar related characteristics or features may have the same or similar reference numerals.

[0019] Figure 1 A flowchart illustrating a global energy consumption optimization method for an industrial production process according to some embodiments of the present invention is shown.

[0020] Figure 2 A schematic diagram of an ethylene industrial production process according to some embodiments of the present invention is shown.

[0021] Figure 3 A schematic diagram of a distributed optimization framework provided according to some embodiments of the present invention is shown.

[0022] Figure 4 A schematic diagram of the communication topology of an ethylene industrial production process according to some embodiments of the present invention is shown.

[0023] Figure 5 A schematic diagram of the iterative results of the full-process energy consumption minimization solution provided according to some embodiments of the present invention is shown.

[0024] Figures 6A to 6D A schematic diagram of the convergence results of multiple constraints in an ethylene industrial production process provided by some embodiments of the present invention is shown. Detailed Implementation

[0025] The following specific embodiments illustrate the implementation of the present invention. Those skilled in the art can easily understand other advantages and effects of the present invention from the content disclosed in this specification. Although the description of the present invention is presented in conjunction with preferred embodiments, this does not mean that the features of the invention are limited to these embodiments. On the contrary, the purpose of describing the invention in conjunction with embodiments is to cover other options or modifications that may be derived based on the claims of the present invention. To provide a thorough understanding of the invention, many specific details will be included in the following description. The invention may also be implemented without using these details. Furthermore, to avoid confusion or obscuring the focus of the invention, some specific details will be omitted in the description.

[0026] It is understood that although terms such as "first," "second," and "third" may be used herein to describe various components, regions, layers, and / or parts, these components, regions, layers, and / or parts should not be limited by these terms, and these terms are only used to distinguish different components, regions, layers, and / or parts. Therefore, the first components, regions, layers, and / or parts discussed below may be referred to as second components, regions, layers, and / or parts without departing from some embodiments of the present invention.

[0027] As mentioned above, continuous industries typically involve multiple production units, characterized by long processes, large scale, complex production conditions, and coupled operational variables. When optimizing the production process of continuous industries, traditional centralized optimization methods face problems such as model complexity and convergence difficulties.

[0028] To overcome the aforementioned problems in the existing technology, the present invention provides a global energy consumption optimization method for industrial production processes, a global energy consumption optimization device for industrial production processes, and a computer-readable storage medium, which can simplify the structure of the optimization model, reduce the modeling difficulty of optimization, and improve the optimization efficiency and real-time performance of industrial production processes.

[0029] In some non-limiting embodiments, the global energy consumption optimization method for an industrial production process provided in the first aspect of the present invention can be implemented by the global energy consumption optimization device for an industrial production process provided in the second aspect of the present invention. Specifically, the global energy consumption optimization device may be configured with a memory and a processor. The memory includes, but is not limited to, the computer-readable storage medium provided in the third aspect of the present invention, on which computer instructions are stored. The processor is connected to the memory and is configured to implement the global energy consumption optimization method for an industrial production process provided in the first aspect of the present invention.

[0030] The working principle of the aforementioned global energy consumption optimization device will be described below with reference to an embodiment of a global energy consumption optimization method for an ethylene industrial production process. Those skilled in the art will understand that these embodiments of the global energy consumption optimization method are merely non-limiting implementations provided by the present invention, intended to clearly demonstrate the main concepts of the invention and provide specific solutions convenient for public implementation, rather than limiting all functions or operating modes of the global energy consumption optimization device. Similarly, the global energy consumption optimization device is also only a non-limiting implementation provided by the present invention and does not constitute a limitation on the entities performing the steps in these global energy consumption optimization methods.

[0031] Please refer to Figure 1 , Figure 1 A flowchart illustrating a global energy consumption optimization method for an industrial production process according to some embodiments of the present invention is shown.

[0032] like Figure 1 As shown, in the process of global energy consumption optimization in industrial production, the global energy consumption optimization device can first identify multiple production units involved in energy consumption in the industrial production process.

[0033] Specifically, the global energy consumption optimization device can first establish a mechanism model of the industrial production process based on the characteristics of the continuous industrial process. It can then identify one or more industrial devices that involve energy consumption under the same production process as a production unit in the industrial production process, and determine multiple production units that involve energy consumption in the industrial production process based on the multiple production processes involved in the continuous industrial process.

[0034] by Figure 2Taking the entire process of an ethylene industrial production as an example, in the feedstock cracking section of this ethylene industrial production process, the cracking furnace first processes feedstocks such as liquefied petroleum gas (LPG) and naphtha (NAP). Then, the cracked gas discharged from the furnace passes sequentially through a cooling tower, compressor CP1, and heat exchanger HE1, before being sent to the propane de-propanizer C1. The top of the propane de-propanizer mainly consists of C3 and light components, including hydrogen, methane, ethane, ethylene, acetylene, propane, propylene, and propane. This stream passes through a hydrogenation reactor and a series of heat exchangers HE2–HE6, before being sent to the pre-methanizer C4. Simultaneously, the top product of the pre-methanizer is processed by the demethanizer C6 to produce methane and hydrogen, while the bottom product flows to the de-ethaner C8. To separate ethane and ethylene, the upper stream of the de-ethaner is sent to the ethylene distillation tower C7, while the bottom stream passes through the propylene distillation tower to produce propane and propylene. The bottom stream of the propane de-propanizer is separated into mixed C4 and cracked gasoline by the butane de-propanizer.

[0035] For this production process, the global energy consumption optimization device can first establish a mechanism model of the ethylene industrial production process, set the production units i that need energy consumption optimization, such as ethylene distillation tower C7 and heat exchangers HE1 to HE6, as energy consumption nodes of the entire ethylene industrial production process, and select the feed flow rate of each production unit i and the temperature difference between each heat exchanger HE1 to HE6 as global decision variables x of the entire ethylene industrial production process.

[0036] Subsequently, the global energy consumption optimization device can perform sensitivity analysis on each production unit i, changing one variable for each production unit i while keeping the other variables of that production unit i unchanged, and observing whether the variable has an impact on the target value (i.e., the energy consumption of that production unit i). If the variable has an impact on the target value, the global energy consumption optimization device can determine the variable as the local decision variable x for that production unit i. i Conversely, if the variable has no impact on the target value, the global energy consumption optimization device can determine that the variable is not a local decision variable x for production unit i. i In some embodiments, the local decision variable x i This includes, but is not limited to, the process operation parameters and / or raw material flow rate of the corresponding production unit i.

[0037] Sensitivity analysis was used to determine the local decision variables x for each production unit i. i Furthermore, by constructing mathematical models for each production unit i, this invention can accurately reflect the impact of changes in raw material composition and operating parameters in the industrial production process on the energy consumption of each production unit i, thereby providing a good model foundation for global energy consumption optimization.

[0038] Please refer to the reference. Figures 1-3 , Figure 3A schematic diagram of a distributed optimization framework provided according to some embodiments of the present invention is shown.

[0039] like Figure 3 As shown, the local decision variable x for each production unit i is determined. i Subsequently, the global energy consumption optimization device can identify each production unit i as an energy consumption node in the industrial production process, and determine the decision variable x of each production unit i. i Based on the target energy consumption, a proxy model is established that reflects the actual process of each production unit i, thereby obtaining the local objective function f for each energy consumption node. i (x i In some embodiments, the global energy consumption optimization device can obtain the local objective function f of each energy consumption node through Latin hypercube sampling. i (x i Input and output data.

[0040] Subsequently, the global energy consumption optimization device can take minimizing the overall process energy consumption as the optimization objective, and optimize the local objective function f for each production unit i. i (x i This process is integrated to determine the optimal mathematical model for the entire industrial production process and to determine the distributed optimization function for its total energy consumption.

[0041] Furthermore, considering the local decision variables x of each production unit i in a continuous industry i Given the characteristics of mutual coupling and interaction among them, the global energy consumption optimization device can also identify one or more production units c involved in the constraints of the industrial production process as constraint nodes of the industrial production process, and establish a proxy model for each production unit c regarding its constraints to describe the inseparable coupling constraints between nodes, thereby eliminating the limitation of traditional distributed algorithms that cannot characterize the characteristics of mutual coupling and interaction among production units i.

[0042] Furthermore, in some embodiments of the present invention, the constraints on each production unit c in the ethylene industrial production process include, but are not limited to, local variable range constraints, separable coupling equation constraints between production units, and / or rated operating constraints of the production units.

[0043] Specifically, the local variable range constraint includes in, It is a local decision variable x i The minimum value, It is a local decision variable x i The maximum value. By configuring the above local variable range constraints, the present invention can ensure that the operation variables of each production unit i are iteratively optimized only within the preset constraint range.

[0044] The separable coupling equality constraint includes Among them, h i This is the coupling equality constraint function between production unit i and its neighboring nodes. By configuring the above separable coupling equality constraint, the present invention can ensure that the coupling relationship between each production unit i and its neighboring nodes remains unchanged.

[0045] The rated operating constraint includes g c (x c )≤W c , where g c (x c ) is the rated working constraint function of production unit c, x c These are the local decision variables of the nominal working constraint function, which are composed of the local variables of each of its connected neighboring nodes, i.e. By configuring the above-mentioned rated working constraints, the present invention can ensure that each production unit c is iteratively optimized only within the preset rated constraint range.

[0046] Those skilled in the art will understand that the above-described schemes for constructing a full-process mechanism model of industrial production online, performing sensitivity analysis online, constructing an energy consumption model online, and constructing a constraint model online are merely non-limiting embodiments provided by this invention. They are intended to clearly demonstrate the main concept of this invention and provide some specific solutions that are easy for the public to implement, rather than to limit the scope of protection of this invention.

[0047] Alternatively, in other embodiments, technicians may pre-construct full-process mechanism models of various industrial production processes and pre-construct sensitivity analyses to determine the local decision variables x of each production unit i. i The system pre-builds energy consumption models for each production unit i and / or constraint models for each production unit c. Thus, during the global energy consumption optimization of the industrial production process, the user only needs to operate the global energy consumption optimization device to select the industrial production process to be optimized from the pre-built database. The global energy consumption optimization device will then automatically determine the multiple production units i involved in energy consumption during the industrial production process, and the local decision variables x of each production unit i. i Energy consumption model and energy consumption objective function f i (x i ), and the local decision variables x of each production unit c involving constraints. c The system incorporates constraint models and constraint functions, thereby effectively reducing the technical skill requirements for users and improving the efficiency and real-time performance of global energy consumption optimization.

[0048] Please refer to further information. Figure 4 , Figure 4A schematic diagram of the communication topology of an ethylene industrial production process according to some embodiments of the present invention is shown.

[0049] like Figure 4 As shown, in some embodiments of the present invention, each production unit i in the entire industrial production process can maintain communication connections with its adjacent energy consumption node i and constraint node c. The communication network between the nodes can be either a directed graph (one-way communication) or an undirected graph (two-way communication).

[0050] like Figure 1 As shown, after determining the total energy consumption optimization function min F(x) for the entire industrial production process and the constraint function of each constraint node c, the global energy consumption optimization device can determine the distributed optimization strategy based on the total energy consumption optimization function min F(x) and each constraint. Through communication between each node i, the distributed optimization of each production unit i is carried out under the constraint to reduce the total energy consumption of the entire industrial production process.

[0051] Specifically, the mathematical model for optimizing the entire industrial production process is as follows:

[0052]

[0053]

[0054] Among them, f i (x i ) is the energy consumption objective function for each production unit i; x i These are the decision variables for production unit i, including raw material flow parameters and operational parameters during the process; and The decision variable x i The minimum and maximum values ​​of N; i It is a neighboring node of node i; g c (x c ) is the rated operating constraint function of production unit c; x c It is the constraint function g c (x c The local decision variables of ) are composed of the local variables of each production unit i connected to it.

[0055] In the process of distributed optimization of each production unit i, the energy consumption optimization device can first set the original variables of each energy consumption node i before the start of each iteration. and auxiliary variables Set initial values ​​within the feasible region and define dual variables. It is a zero vector.

[0056] Subsequently, for each energy consumption node i in the entire industrial production process, the global energy consumption optimization device can communicate between each energy consumption node i and its neighboring constraint nodes to obtain the indivisible constraint information of each neighboring constraint node. Furthermore, the global energy consumption optimization device can also communicate between each energy consumption node i and its neighboring energy consumption nodes to obtain node information of the neighboring energy consumption nodes. This determines the equality constraint information for each energy consumption node i. Subsequently, the global energy consumption optimization device can utilize Lagrange duality and use the dual ascent method to update the dual variables of each energy consumption node i:

[0057]

[0058] Next, the global energy consumption optimization device can obtain the dual variables of the neighboring nodes of each energy consumption node i through mutual communication, and update them within each node i using the gradient projection algorithm and gradient descent method:

[0059]

[0060]

[0061]

[0062] in:

[0063]

[0064]

[0065] Furthermore, for each constrained node c in the entire industrial production process, the global energy consumption optimization device can, after obtaining information about its neighboring energy consumption nodes, summarize the local variables of the neighboring energy consumption nodes of constrained node c. To determine the local decision variable x of the constraint node. c Afterwards, the global energy optimization device can also calculate the rated working constraint function g of the constraint node c. c (x c )-W c The value of the constraint function is returned to the neighboring constraint nodes of the constraint node c, indicating whether they satisfy the constraint restrictions, along with the gradient of the rated working constraint function.

[0066] Furthermore, after completing this iteration, the global energy consumption optimization device can also optimize the local decision variables x after distributed optimization. i Calculate the total energy consumption of the entire industrial production process, and then determine whether the distributed optimization has converged based on the trend of the total energy consumption.

[0067] Please refer to Figure 5 , Figure 5 A schematic diagram of the iterative results of the full-process energy consumption minimization solution provided according to some embodiments of the present invention is shown.

[0068] like Figure 5 As shown, before completing 500 iterations, the calculated total energy consumption of the entire industrial production process has not reached a stable value. The global energy consumption optimization device can therefore determine that the distributed optimization has not converged and proceed to the next iteration. Afterwards, once 500 iterations have been completed, the calculated total energy consumption of the entire industrial production process tends to stabilize. The global energy consumption optimization device can then determine that the distributed optimization has converged, and thus, based on the local decision variable x of the current round... i The value of is used to determine the optimal operating conditions for the industrial production process.

[0069] In summary, by transmitting information between each energy consumption node i and the constraint node c, and by combining the Lagrange duality method and the parametric projection algorithm for optimization under constraints, this invention can continuously adjust the raw material configuration and process operation conditions of each relevant unit node. Under the premise of meeting the production constraint requirements, it can achieve the optimization goal of reducing the energy consumption of the entire process, thereby solving the optimization problem under multiple constraints.

[0070] Please refer to further information. Figures 6A to 6D And Table 1. Figures 6A to 6D The diagram illustrates the convergence results of multiple constraints in an ethylene industrial production process according to some embodiments of the present invention. Table 1 shows a comparison between the distributed optimization results provided by some embodiments of the present invention and the optimization results obtained using a sequential model.

[0071] Table 1

[0072]

[0073]

[0074] like Figures 6A to 6D As shown in Table 1, compared with the solution results in typical simulation software, the distributed optimization method proposed in this invention can not only find the optimal solution under various constraints such as compressor constraints and total feed flow, but also has a faster solution speed and better solution results, and can further reduce the total energy consumption by 17.7%.

[0075] Furthermore, the distributed optimization algorithm provided by this invention only needs to establish a mathematical model for local production unit i to obtain the global optimization result. The comparison results shown in Table 1 demonstrate that this local mathematical model can more accurately reflect the actual energy consumption changes of the corresponding production unit i.

[0076] Furthermore, by setting constraint node c and establishing a corresponding constraint model, the distributed optimization algorithm provided by this invention can well adapt to the characteristics of complex coupling and correlation between production units in continuous industrial processes and the large number of optimization variables, and can meet the actual constraint requirements of each production unit, thereby quickly obtaining the global optimal solution and showing good optimization effect.

[0077] Although the methods described above are illustrated and depicted as a series of actions for the sake of simplicity, it should be understood and appreciated that these methods are not limited by the order of the actions, as some actions may occur in a different order and / or concurrently with other actions from the illustrations and descriptions herein or not illustrated and described herein but which may be understood by those skilled in the art, according to one or more embodiments.

[0078] The prior description of this disclosure is provided to enable any person skilled in the art to make or use this disclosure. Various modifications to this disclosure will be apparent to those skilled in the art, and the general principles defined herein may be applied to other variations without departing from the spirit or scope of this disclosure. Therefore, this disclosure is not intended to be limited to the examples and designs described herein, but should be accorded the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A method for global energy consumption optimization in an industrial production process, characterized in that, Includes the following steps: Identify multiple production units involved in energy consumption during industrial production; Each of the aforementioned production units is identified as an energy consumption node, and an energy consumption model for each of the aforementioned production units is determined. Production units subject to constraints are identified as constraint nodes, and constraint models for these production units are determined. These constraints include separable coupling equality constraints between production units. ,in, It is a production unit i Coupling equality constraint functions between neighboring nodes; Based on the local decision variables of each production unit Determine the energy consumption objective function for each production unit. ; According to the energy consumption objective function of each production unit Determine the distributed optimization function for total energy consumption. ; Summarize the local variables of the energy consumption nodes of the neighboring nodes of the constrained node. To determine the local decision variables of the constraint node. and calculate production unit c The value of the rated working constraint function ; Return to the neighboring constraint nodes of the constraint node whether they satisfy the constraint restrictions, and the gradient of the nominal working constraint function. ; as well as Through communication between the nodes, distributed optimization is performed on each production unit under the constraints based on the distributed optimization function to reduce the total energy consumption of the industrial production process and achieve global optimization of the total energy consumption of the industrial production process.

2. The global energy consumption optimization method as described in claim 1, characterized in that, Before performing distributed optimization for each of the production units, the global energy consumption optimization method further includes the following steps: Sensitivity analysis was performed on each of the production units to determine the local decision variables for each production unit. .

3. The global energy consumption optimization method as described in claim 2, characterized in that, The local decision variables This includes the process operation parameters and / or raw material flow rate of the production unit.

4. The global energy consumption optimization method as described in claim 1, characterized in that, The constraints also include local variable range constraints and / or rated operating constraints for production units, wherein, The local variable range constraint includes ,in, Local decision variables The minimum value, Local decision variables The maximum value, The rated operating constraints include ,in, These are the local decision variables of the nominal working constraint function. .

5. The global energy consumption optimization method as described in claim 4, characterized in that, The step of reducing the total energy consumption of the industrial production process by performing distributed optimization on each of the production units under the constraints based on the distributed optimization function through communication between the nodes includes: Before the start of each iteration, indivisible constraint information is obtained through communication between each energy-consuming node and its neighboring constraint nodes. And through communication between each energy-consuming node and its neighboring energy-consuming nodes, the node information of the neighboring energy-consuming nodes is obtained. To determine the equality constraint information for each of the energy consumption nodes. ; The dual variables of each energy-consuming node are updated using the dual ascent method. ;as well as Through communication between the energy-consuming nodes, the dual variables of the neighboring nodes of each node are obtained, and the dual variables are updated within each node using gradient projection and gradient descent methods. ， in, It is a primitive variable. It is an auxiliary variable. It is a dual variable.

6. The global energy consumption optimization method as described in claim 5, characterized in that, After performing distributed optimization on each production unit under the constraints based on the distributed optimization function via communication between the nodes, the global energy consumption optimization method further includes the following steps: Based on the local decision variables after distributed optimization Calculate the total energy consumption of the industrial production process; Based on the trend of total energy consumption, determine whether the distributed optimization has converged; In response to the determination that the distributed optimization has not converged, proceed to the next iteration; and In response to the conclusion of the distributed optimization convergence, based on the local decision variables The value of is used to determine the optimal operating conditions for the industrial production process.

7. The global energy consumption optimization method as described in claim 1, characterized in that, The communication between the nodes includes directed graphs and undirected graphs.

8. A global energy consumption optimization device for an industrial production process, characterized in that, include: Memory; as well as A processor, connected to the memory, and configured to implement the global energy consumption optimization method for an industrial production process as described in any one of claims 1 to 7.

9. A computer-readable storage medium storing computer instructions thereon, characterized in that, When the computer instructions are executed by the processor, the global energy consumption optimization method for the industrial production process as described in any one of claims 1 to 7 is implemented.

Images