Nonlinear supply chain system change effect propagation and data visualization method and system

Abstract

The invention discloses a nonlinear supply chain system change effect propagation and data visualization method and system, and relates to the field of supply chain cascade systems, and the method comprises the steps: constructing a mathematical model of a nonlinear supply chain cascade system; optimizing the mathematical model of the nonlinear supply chain cascade system to obtain an optimized mathematical model of the nonlinear supply chain cascade system; determining a global controller; determining a compensation controller; constructing a radial basis function neural network; optimizing the compensation controller by adopting the radial basis function neural network to obtain an optimized compensation controller; and controlling the optimized mathematical model of the nonlinear supply chain cascade system based on the global controller and the optimized compensation controller. According to the method, the problem that the propagation effect is caused by the fact that the system parameters and the structure cannot be controlled under the condition that the system parameters and the structure are unknown due to the change of the supply chain system caused by emergencies can be solved.

Description

Technical Field

[0001] The present invention relates to the technical field of supply chain cascade systems, and particularly relates to a method and system for the propagation and data visualization of change effects in a non-linear supply chain system. Background Art

[0002] With the acceleration of the process of economic globalization, supply chains have broad application prospects in fields such as manufacturing, finance, and e-commerce. However, due to emergencies such as natural disasters and trade wars, problems such as supply chain system interruptions and information blockages occur frequently, bringing significant economic losses to enterprise groups and thus threatening the safety of people's lives and property.

[0003] A supply chain system is composed of different nodes such as suppliers, producers, distributors, retailers, and customers; around the core enterprise, through the control and management of logistics, information flow, and capital flow, a complex and complete functional network is realized to deliver the final product made from raw materials to the hands of customers.

[0004] In recent years, with the continuous development of economic globalization, due to the occurrence of sudden natural events such as the epidemic and natural disasters, the structure of the supply chain system has changed. A key issue in the supply chain system is the stability of the system structure or parameters, and the system structure or parameter device is easily changed due to emergencies such as cyberattacks and trade wars. Therefore, it is essential to design a suitable compensation controller for the supply chain system to cope with the changes. Otherwise, the system may be damaged, suffocated, or collapsed. Unfortunately, there are few research results on the change control of dynamic supply chain systems, especially non-linear systems. Regarding the change design of production and supply chain systems, it can be traced back to 1997 when Wringt first reviewed the research on "engineering change" and defined engineering change as the frequent modification of product components. According to the research of Jarratt et al., product change is defined as the change of product structure, drawings, or software during the product development process. Obviously, the interruption or blockage of the supply chain usually leads to corresponding changes in product design. On the other hand, during the synchronous evolution process, the change in product design will inevitably lead to changes in the supply chain system. The synchronous evolution process can be defined as the process in which supply chain decisions are continuously modified or re-established after the behavior of certain links in the supply chain changes. Among them, the change in supply chain behavior is called the introduction of change events. Therefore, it is particularly important to adopt appropriate change control strategies for the supply chain system to eliminate the impact of these destructive factors. Summary of the Invention

[0005] The purpose of the present invention is to provide a method and system for the propagation and data visualization of change effects in a non-linear supply chain system, so as to solve the problem that the propagation effect cannot be controlled when the system parameters and structure are unknown due to changes caused by emergencies in the supply chain system.

[0006] To achieve the above object, the present invention provides the following solutions:

[0007] A method for nonlinear supply chain system change effect propagation and data visualization, comprising:

[0008] Constructing a mathematical model of a nonlinear supply chain cascade system;

[0009] Optimizing the mathematical model of the nonlinear supply chain cascade system to obtain a mathematical model of the nonlinear supply chain cascade system after change due to an unexpected event;

[0010] Determining a global controller;

[0011] Determining a compensation controller;

[0012] Constructing a radial basis neural network;

[0013] Optimizing the compensation controller by using the radial basis neural network to obtain an optimized compensation controller;

[0014] Controlling the mathematical model of the optimized nonlinear supply chain cascade system based on the global controller and the optimized compensation controller;

[0015] Constructing a multi-agent simulation system to display the inventory changes in different controller situations and switching states.

[0016] The expression of the mathematical model of the nonlinear supply chain cascade system after change due to an unexpected event of the system is as follows:

[0017]

[0018] Where x i ∈R, 1≤i≤n, respectively represent the inventory level and change rate of the i-th device or link of the system at time t, 1≤k≤i, represent the change rate of the inventory level after system change at time t, f i (1≤i≤n) is a nonlinear function related to unknown customer demand and current inventory states x1, x2... x i ; g i , 1≤i≤n, is a nonlinear function related to the production level of downstream devices and is regarded as the transition gain of the upstream production inventory level x i+1 ; 1≤k≤i, respectively represent the changed f i g i , ρ i , 1≤i≤n, represent the loss rates of each node warehouse, u(t)∈R represents the production rate at time t, u icDenote the change compensation control input of the $i$-th device or link. Denote the output $x$ after the system change. i+1 Denote the upstream production library level.

[0019] The expression of the global controller is as follows:

[0020]

[0021] Among them, $V$ n-1 is the constructed formula of the $(n - 1)$-th level Lyapunov function in the design, $\varphi$ n , $\varphi$ n-1 respectively denote the scalar functions constructed for the $n$-th and $(n - 1)$-th level nodes, $k$ n is the control gain coefficient of the $n$-th level.

[0022] Optionally, the expression of the optimized compensation controller is as follows:

[0023]

[0024] Among them, $u$ ic denotes the optimized compensation controller, $V$ i-1 is the constructed formula of the $(i - 1)$-th level Lyapunov function in the design, $e$ i-1 $=x$ i-1 $-\varphi$ i-1 is the error variable, $e$ i $=x$ i $-\varphi$ i is the error variable, denotes the error of the gain $f$ i before and after the change, denotes the optimal weight coefficient of the $i$-th level node, $h$ i $(x)$ denotes the radial basis function of the $i$-th level node, denotes the derivative of $\varphi$ i-1 , $\varphi$ i-1 denotes the scalar function constructed for the $(i - 1)$-th level node, $k$ i denotes the control gain coefficient of the $i$-th level, $z$ i $=e$ i $-\varphi$ i denotes the error between $e$ i and $\varphi$ i .

[0025] Based on the above method in the present invention, the present invention further provides a non - linear supply chain system change effect propagation and data visualization system, including:

[0026] A mathematical model construction module of the non - linear supply chain cascade system, used to construct the mathematical model of the non - linear supply chain cascade system;

[0027] The first optimization module is used to optimize the mathematical model of the non-linear supply chain cascade system to obtain the optimized mathematical model of the non-linear supply chain cascade system;

[0028] The global controller determination module is used to determine the global controller;

[0029] The compensation controller determination module is used to determine the compensation controller;

[0030] The network construction module is used to construct a radial basis neural network;

[0031] The second optimization module is used to optimize the compensation controller by using the radial basis neural network to obtain the optimized compensation controller;

[0032] The control module is used to control the optimized mathematical model of the non-linear supply chain cascade system based on the global controller and the optimized compensation controller.

[0033] Optionally, the control device further includes:

[0034] The multi-agent simulation system construction module is used to construct a multi-agent simulation system to display the changes in inventory in different controller situations and switching states.

[0035] Optionally, the expression of the first optimization module is as follows:

[0036]

[0037] where x i ∈R, 1 ≤ i ≤ n respectively represent the inventory level and change rate of the i-th device or link of the system at time t, 1 ≤ k ≤ i represents the change rate of the inventory level after the system change at time t, f i (1 ≤ i ≤ n) is a non-linear function related to the unknown customer demand and the current inventory states x1, x2... x i ; g i , 1 ≤ i ≤ n, is a non-linear function related to the production level of the downstream device and is regarded as the transition gain of the upstream production inventory level x i+1 ; 1 ≤ k ≤ i respectively represent the changed f i g i , ρ i , 1 ≤ i ≤ n represent the loss rates of each node warehouse, u(t) ∈ R represents the production rate at time t, u ic represents the change compensation control input of the i-th device or link, represents the output after the system change, x i+1Indicates the upstream production library level.

[0038] The present invention also provides an electronic device, including a memory and a processor. The memory is used to store a computer program, and the processor runs the computer program to enable the electronic device to execute the above-mentioned non-linear supply chain system change effect propagation and data visualization method.

[0039] The present invention also provides a computer-readable storage medium, which stores a computer program. When the computer program is executed by a processor, it implements the above-mentioned non-linear supply chain system change effect propagation and data visualization method.

[0040] According to the specific embodiments provided by the present invention, the following technical effects are disclosed:

[0041] By designing the control rate of the cascaded non-linear supply chain inventory system, the present invention makes the supply chain system affected by downstream information stable. For a supply chain system with unknown change conditions (structural changes), a neural network method is used to identify the structure after the change, so as to design a compensation controller or a global controller to stabilize the system. The present invention uses a neural network algorithm for data processing to approximately approach any structural parameters after the change, thereby designing a controller to realize the rationality of the entire algorithm and solve the situation of unknown structure caused by changes. BRIEF DESCRIPTION OF THE DRAWINGS

[0042] In order to more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the following will briefly introduce the drawings required in the embodiments. Obviously, the drawings in the following description are only some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other drawings can be obtained based on these drawings.

[0043] Figure 1 It is a flowchart of the non-linear supply chain system change effect propagation and data visualization method provided by the present invention;

[0044] Figure 2 It is a schematic diagram of the simplified structure of the supply chain system provided by the present invention;

[0045] Figure 3 It is a schematic diagram of the supply chain system change provided by the present invention;

[0046] Figure 4 It is a schematic diagram of the radial basis neural network provided by the present invention;

[0047] Figure 5 It is a diagram of the inventory state before and after the system change provided by the present invention;

[0048] Figure 6 It is a diagram of the uncertain demand before the system change provided by the present invention;

[0049] Figure 7 The constant requirement diagram for uncertain requirement tracking after system change provided by the present invention;

[0050] Figure 8 The change signal diagram before and after system change provided by the present invention;

[0051] Figure 9 The system performance index diagram before and after system change provided by the present invention. Detailed implementation manners

[0052] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.

[0053] The purpose of the present invention is to provide a method and system for nonlinear supply chain system change effect propagation and data visualization, so as to solve the problem that the supply chain system cannot be controlled due to unknown system parameters and structures caused by changes due to emergencies, resulting in propagation effects.

[0054] To make the above objects, features and advantages of the present invention more obvious and understandable, the present invention will be further described in detail below in conjunction with the accompanying drawings and specific implementation manners.

[0055] As Figure 1 shown, Figure 1 The flowchart of the method for nonlinear supply chain system change effect propagation and data visualization provided by the present invention, the method includes:

[0056] Step 101: Construct a mathematical model of the nonlinear supply chain cascade system.

[0057] The simplified structure of the supply chain system is as Figure 2 shown. The present invention will model the supply chain system as a nonlinear cascade system. We start with a production inventory system centered on the product manufacturing process, which has n devices and a complete supply and demand part. It can be seen from Figure 2 that from the upstream device to the downstream device, the production equipment of each device processes the materials, that is, the materials flow along the supply chain from production to supply, while the information flows in the opposite direction of the supply chain. Figure 2The solid arrows represent the logistics of a certain link in the supply chain, and the dashed arrows represent the information flow. Among them, the logistics flows from the upstream to the downstream. For example, a certain product reaches the customer's hands after undergoing multiple processes. The information flow is in the reverse direction, that is, the information such as inventory from the downstream is fed back to the upstream. Among them, from left to right, u(t) represents the shipment volume of the current factory (supplier). x n represents the nth link of this supply chain system, and so on until x1. x1 represents the very end of the supply chain system. Its meaning can be understood as the next link is to deliver to the customer, that is, the output of the system. Therefore, y(t) = x1. At the same time, d(t) represents the demand of the current customer, which can be a constant or a quantity that changes periodically. d(t) represents the target value of x1 in this system, that is to say, the normal system should achieve the output volume of d(t).

[0058] Let x i (t) ∈ R (1 ≤ i ≤ n) represent the inventory level of the ith device at time t, d(t) ∈ R represent the user demand at time t, and u(t) ∈ R represent the production rate at time t. From Figure 2 it can be seen that the inventory level of the finished product x1 in the warehouse is closely related to the user demand.

[0059] Taking into account some non-linear factors, such as transfer relationships, parameter uncertainties, etc., the production-inventory relationship of the ith device is simulated as a continuous-time non-linear model, as shown in the formula.

[0060]

[0061] Among them, represents the time derivative of the current equation state variable (i.e., the node inventory level of a certain link in the supply chain), ρ i represents the spoilage rate of the current inventory products. If the current warehouse stores products such as fruits and foods, then this parameter has an effect. The value range of this parameter is [0, 1). f i is a non-linear function, usually related to the unknown customer demand and the current inventory status x1, x2... x i ; g i is a non-linear function, related to the production level of the downstream device, and is usually regarded as the transition gain of the upstream production-inventory level x i+1 . Regarding the production-inventory level x i+1 , it is usually regarded as the virtual control of a cascaded non-linear system with a triangular structure.

[0062] Similarly, the inventory dynamic equation of the nth device can be obtained

[0063]

[0064] Among them, u represents as Figure 2The production input of the single - supply - chain inventory system shown, i.e., the shipment volume of the production - plant supplier.

[0065] Finally, the non - linear equation of the entire system is

[0066]

[0067] Equation (1.3) is the mathematical model of the non - linear supply - chain cascade system.

[0068] Step 102: Optimize the mathematical model of the non - linear supply - chain cascade system to obtain the mathematical model of the non - linear supply - chain cascade system after the system changes due to unexpected events.

[0069] When the supply - chain system is interrupted or blocked due to sudden factors such as tsunamis, flash floods, and earthquakes, in this model (1.3), it is manifested as a change in a certain structure or a certain parameter. At this time, the original controller u(t) cannot control the new system with changed structure and parameters. If no other control means are added at this time, this supply - chain system may become unstable or even diverge, resulting in serious consequences. This change can be shown by Figure 3 displayed.

[0070] In Figure 3 it can be seen that there are a total of two supply - chain systems, upper and lower. The upper one is a simplified supply - chain model operating under normal circumstances. When a change occurs at a certain moment, causing an interruption in a certain link of the supply - chain system, it will change the structure of the supply - chain system, which is shown in the schematic diagram as: when t k at the moment, the i - th link is interrupted, then this facility changes from x i to This ensures the complete operation of the supply - chain transfer. However, the previous control input u cannot ensure the stability of the supply - chain system after the change. Therefore, a compensation controller, that is, Figure 3 the u in c , is needed to compensate for the i - th facility, so that under the combined action of the global controller u and the compensation controller u c , the supply - chain system after the change is stable.

[0071] For the supply - chain system model after the change, it is different from Equation (1.3). The specific derivation process is as follows:

[0072] Figure 2 Shown is the operation of the supply - chain system. Assume that a change occurs at a certain moment, that is, facilities such as the i - th device change due to unexpected events. To make the system operate stably, a compensation scheme will naturally be adopted when the change occurs, such as Figure 3As shown. Assume that the i-th device changes. Considering the propagation effect of the supply chain system, that is, in the upstream direction of the i-th device, due to the transmission of supply chain information flow, namely the supply chain propagation effect, the system structures from the n-th to the i-th devices are all affected and changed to varying degrees. And due to the randomness of the changes, the interference terms are unknown. According to this change theory, the state equation of the current i-th changed device changes from formula (1.1) to

[0073]

[0074] In formula 1.4, the definitions of each symbol are as follows: represents the time derivative of the inventory after the current change, with a tilde on it, indicating that this system is the changed system, which is a different system from the system before the change. Similarly also represents the structure function of the current i-th device after the change, which also means after the change. However, note that the x i+1 term inventory has no tilde because the defined change model shows that due to the propagation direction of the information flow, currently it only propagates upstream to the changed device, so the downstream is not affected. According to this principle, combining the original supply chain model (1.3) with formula (1.4) and the change propagation model principle, the mathematical model of the changed supply chain system can be obtained as

[0075]

[0076] where, represents the transfer magnification of the changed system, represents the new output of the changed system.

[0077] So far, the establishment processes of the supply chain cascade model under normal conditions and the changed supply chain cascade model have both been completed. Next, how to design the global controller of the supply chain system under normal operation and the design of the changed compensation controller will be described in detail respectively.

[0078] Step 103: Determine the global controller.

[0079] The backstepping method is a classic recursive controller design method based on the Lyapunov stability theorem. The basic idea of the backstepping method lies in using the analytical expression of the time derivative of the control law designed in the previous step. Its algorithm is shown in Table 1.

[0080] Table 1 Basic algorithm of the backstepping method

[0081]

[0082] Through the algorithm in Table 1, the form of the global controller can be deduced recursively as shown in the following formula:

[0083]

[0084]

[0085] So far, through the controller (1.9), the supply chain system (1.3) before the change can be made stable, and the first goal of the present invention has been achieved, that is, to design a global controller to make the unchanged supply chain system stable.

[0086] Step 104: Determine the compensation controller.

[0087] After deriving the global controller from the above scheme principle, the following details how the compensation controller is designed. First, by comparing the system and the system, it can be found that under the influence of the propagation effect, only the system structure from the first device to the i-th device has changed, while the system from the (i + 1)-th device to the n-th device is not affected.

[0088] Define where e i represents the error between the inventory status quantity of the i-th device after the change and the inventory status quantity of the i-th device before the change. However, after the change, due to the transmission of the propagation effect, it is impossible that there is only an error at the current i-th device level. Therefore, an error system needs to be established, and the error form of the i-th device can be expressed by the formula:

[0089]

[0090] In formula (1.6), represents the time derivative of the inventory error before and after the change of the current i-th device, and e i dot represents the time derivative formula of the error. Formula (1.6) is obtained by subtracting formula (1.1) from formula (1.4). After sorting, the form of formula 1.6 is obtained. For the convenience of expression, the variable symbols are integrated.

[0091] By defining

[0092]

[0093]

[0094] Formula (1.6) is written as

[0095] in the form of, where F i and G i both represent a composite structure function, just a symbol.

[0096] According to the above definition method, the error equation of the changed system and the original system is shown in (1.7).

[0097]

[0098] Among them, F i and G i are marking symbols, representing composite functions.

[0099] And from the previous error definition, it is known that then the error system can be rewritten in the following cascaded form

[0100]

[0101] So far, we have obtained the overall model of this error system. The following task is to design the compensation controller u ic , so that the error system (1.8) before and after this change converges to 0, which means that through the combined action of the global controller and the compensation controller, even if the system changes, the system can restore to the previous level.

[0102] However, there is an unavoidable problem when designing the compensation controller u ic , that is, in the error system (1.8), due to F1+G1x2, F2+G2x3 ······ and F i +G i ·x i+1 terms, due to changes and mathematical variables, these terms are unclear and have complex composite relationships, so they cannot be directly used to design the controller. Therefore, a new method needs to be introduced to handle this, which is also the core point of the present invention. First, the tools used to handle this situation will be briefly introduced below.

[0103] Step 105: Construct a radial basis neural network.

[0104] The radial basis neural network is a three-layer feedforward neural network with a single hidden layer. Using an RBF neural network can speed up the learning speed and is suitable for real-time control scenarios. In addition, existing research has shown that an RBF neural network can approximate any continuous function with arbitrary precision, that is, the universal approximation theorem. The structure diagram of the radial basis neural network is as Figure 4 shown. In the present invention, this radial basis neural network is used to identify the uncertain parameters and structures in the above terms and solve the problem that the controller cannot be designed due to the coupling relationship. The network consists of three layers: an input layer, an intermediate layer, and an output layer. Assume that according to Figure 3As shown, if the supply chain change occurs at the i-th device, the input variables of the neural network are 2i. This is because the original system needs to input i variables, and the new system after the change also needs to input i variables. This principle is reflected in formula (1.6). The middle layer is a radial basis function, and the output can be represented through a simple summation relationship.

[0105] During training, by capturing the data after the change, that is, the relationship between the input variables and output variables of the changed nodes, and training through the gradient descent method, a successfully trained model is obtained.

[0106] The output of the neural network is Then the corresponding previous uncertain terms can be replaced with this expression, and the entire error system (1.8) can be written in the following form.

[0107]

[0108] At this time, in system (1.10), This term is independent of the system variable e, so this term can be regarded as a constant in system design.

[0109] So far, the above-mentioned variable coupling problem and uncertainty problem have been solved, and the design of the compensation controller will be introduced below.

[0110] Step 106: Optimize the compensation controller using the radial basis neural network to obtain an optimized compensation controller.

[0111] For the design of the compensation controller, the design goal is to make the new system after the change follow the original system before the change to ensure the consistency of the system before and after the change. Similarly, through simple derivation using the backstepping algorithm described before, the form of the compensation controller can be obtained by combining the neural network and the backstepping method as shown in the formula.

[0112]

[0113] So far, the second core objective of this application has been completed, that is, in the supply chain system after the change, through the original global controller and to offset the propagation effect and uncertainty impact brought by the control, the supply chain system after the change is stabilized.

[0114] Step 107: Control the mathematical model of the optimized non-linear supply chain cascade system based on the global controller and the optimized compensation controller.

[0115] Step 108: Build a multi-agent simulation system to show the situation of different controllers and the changes in inventory when switching states.

[0116] Build a multi-agent simulation system, introduce simulation examples to verify the effectiveness of the designed algorithms of the proposed solutions, and apply a data visualization system to achieve auxiliary decision-making. The visualization system realizes the display of the situations of different controllers and the changes in inventory when switching states.

[0117] Currently, the main functions of this system are: intuitively display various situations of the simulation system, and intuitively display indicators such as user satisfaction and cost according to the control strategy adopted at that time.

[0118] The advantages of the system are simple, fast, efficient and convenient. The model can be simply switched through this platform. Immediately, the supply chain system under the corresponding changed global controller, the system without the compensation controller's function after the change of the supply chain system, the simulation of the inventory situation in the supply chain system before and after the change, and the operation of the global controller and the compensation controller will appear. The advantages and disadvantages of the adopted model can be very intuitively seen.

[0119] According to the corresponding experience and knowledge provided by system development practitioners, a data visualization system was replicated. There are no excessive technical requirements for operators. Just click the mouse, and the background data will appear. When the mouse hovers over the graph, the real-time data situation can be seen. The effect diagram of each part is an animated picture, which can move, zoom in, zoom out, etc. with the movement of the mouse. Single pictures can be viewed according to the sidebar, or four pictures can be combined and viewed for comparison. The system has obtained preliminary operation results, as shown in Figures 5 - 9 shown.

[0120] Based on the above method in the present invention, the present invention further provides a non-linear supply chain system change effect propagation and data visualization system, including:

[0121] A mathematical model construction module for a non-linear supply chain cascade system, used to construct a mathematical model of the non-linear supply chain cascade system;

[0122] A first optimization module, used to optimize the mathematical model of the non-linear supply chain cascade system to obtain an optimized mathematical model of the non-linear supply chain cascade system;

[0123] A global controller determination module, used to determine the global controller;

[0124] A compensation controller determination module, used to determine the compensation controller;

[0125] A network construction module, used to construct a radial basis neural network;

[0126] A second optimization module, used to optimize the compensation controller by using the radial basis neural network to obtain an optimized compensation controller;

[0127] A control module, configured to control a mathematical model of an optimized non-linear supply chain cascade system based on the global controller and the optimized compensation controller.

[0128] The present invention further provides an electronic device, including a memory and a processor. The memory is used to store a computer program, and the processor runs the computer program to enable the electronic device to execute the above-mentioned non-linear supply chain cascade system control method.

[0129] The present invention further provides a computer-readable storage medium, which stores a computer program. When the computer program is executed by a processor, the above-mentioned non-linear supply chain cascade system control method is implemented.

[0130] In the present specification, the various embodiments are described in a progressive manner. Each embodiment focuses on the differences from other embodiments. The same or similar parts among the various embodiments can be referred to each other. For the system disclosed in the embodiments, since it corresponds to the method disclosed in the embodiments, the description is relatively simple. For the relevant parts, reference can be made to the description in the method part.

[0131] In this article, specific examples are used to elaborate on the principle and implementation manner of the present invention. The description of the above embodiments is only used to help understand the method and its core idea of the present invention. At the same time, for those of ordinary skill in the art, according to the idea of the present invention, there will be changes in the specific implementation manner and application scope. In summary, the content of this specification should not be construed as a limitation to the present invention.

Claims

1. A method for the propagation of change effects and data visualization in a non-linear supply chain system, characterized mainly by Including: Constructing a mathematical model of a non-linear supply chain cascade system; Optimizing the mathematical model of the non-linear supply chain cascade system to obtain a mathematical model of the non-linear supply chain cascade system after change due to unexpected events; Determining a global controller; Determining a compensation controller; Constructing a radial basis neural network; Optimizing the compensation controller using the radial basis neural network to obtain an optimized compensation controller; Controlling the optimized mathematical model of the non-linear supply chain cascade system based on the global controller and the optimized compensation controller; Constructing a multi-agent simulation system to show the situations of different controllers and the changes in inventory during switching states.

2. According to the method for propagating the change effect and visualizing data of the non-linear supply chain system as claimed in claim 1, the expression of the mathematical model of the non-linear supply chain cascade system after change due to unexpected events is as follows: where x i ∈R, 1 ≤ i ≤ n respectively represent the inventory level and change rate of the i-th device or link of the system at time t, 1 ≤ k ≤ i represents the change rate of the inventory level after the system change at time t, f i (1 ≤ i ≤ n) is a non-linear function related to the unknown customer demand and the current inventory status x1, x2... x i is related; g i , 1 ≤ i ≤ n, is a non-linear function related to the production level of downstream devices and is regarded as the transition gain of the upstream production inventory level x i+1 ; 1 ≤ k ≤ i respectively represent the changed f i g i , ρ i , 1 ≤ i ≤ n represents the loss rate of each node warehouse, u(t) ∈ R represents the production rate at time t, u ic represents the change compensation control input of the i-th device or link, represents the output after the system change, x i+1 represents the upstream production inventory level.

3. The method for nonlinear supply chain system change effect propagation and data visualization according to claim 2, wherein The expression of the global controller is as follows: Among them, V n-1 is the construction formula of the (n - 1)-th level Lyapunov function in the design, φ n , φ n-1 respectively represent scalar functions constructed for the n-th and (n - 1)-th level nodes, and k n is the control gain coefficient of the n-th level.

4. The method for non-linear supply chain system change effect propagation and data visualization according to claim 3, characterized in that The expression of the optimized compensation controller is as follows: Among them, u ic represents the optimized compensation controller, V i-1 is the construction formula of the (i - 1)-th Lyapunov function designed, e i-1 = x i-1 - φ i-1 is the error variable, e i = x i - φ i is the error variable, represents the error of the gain f i before and after the change, represents the optimal weight coefficient of the i-th node, h i (x) represents the radial basis function of the i-th node, represents the derivative of φ i-1 φ i-1 represents the scalar function constructed by the (i - 1)-th node, k i represents the i-th control gain coefficient, z i = e i - φ i represents the error between e i and φ i .

5. A non-linear supply chain system change effect propagation and data visualization system, characterized in that, Including: A mathematical model construction module of the non-linear supply chain cascade system, configured to construct a mathematical model of the non-linear supply chain cascade system; A first optimization module, configured to optimize the mathematical model of the non-linear supply chain cascade system to obtain a mathematical model of the non-linear supply chain cascade system after change due to unexpected events; A global controller determination module, configured to determine a global controller; A compensation controller determination module, configured to determine a compensation controller; A network construction module, configured to construct a radial basis neural network; A second optimization module, configured to optimize the compensation controller using the radial basis neural network to obtain an optimized compensation controller; A control module, configured to control the optimized mathematical model of the non-linear supply chain cascade system based on the global controller and the optimized compensation controller.

6. The non-linear supply chain system change effect propagation and data visualization system according to claim 5, characterized in that The control device further includes: A multi-agent simulation system construction module, configured to construct a multi-agent simulation system to show the situations of different controllers and the changes in inventory during switching states.

7. The non-linear supply chain system change effect propagation and data visualization system according to claim 5, characterized in that, The expression of the first optimization module is as follows: where x i ∈R, 1 ≤ i ≤ n, respectively represent the inventory level and change rate of the i-th device or link of the system at time t, 1 ≤ k ≤ i, represent the change rate of the inventory level after the system change at time t, f i (1 ≤ i ≤ n) is a non-linear function, related to the unknown customer demand and the current inventory status x1, x2... x i g i , 1 ≤ i ≤ n, is a non-linear function, related to the production level of downstream equipment, and is regarded as the transition gain of the upstream production inventory level x i+1 ; 1 ≤ k ≤ i, respectively represent the changed f i g i , ρ i , 1 ≤ i ≤ n, represent the loss rate of each node warehouse, u(t) ∈ R represents the production rate at time t, u ic represents the change compensation control input of the i-th device or link, represents the output after the system change, x i+1 represents the upstream production inventory level.

8. An electronic device, characterized in that, Including a memory and a processor, the memory is used to store a computer program, and the processor runs the computer program to enable the electronic device to execute the method for propagating the change effect and visualizing data of the non-linear supply chain system as claimed in any one of claims 1-5.

9. A computer-readable storage medium, characterized in that, It stores a computer program, and when the computer program is executed by the processor, it implements the method for propagating the change effect and visualizing data of the non-linear supply chain system as claimed in any one of claims 1-5.

Images