A Model Predictive Control-Based Production Line Change Propagation Control Method and System

Abstract

A production line change propagation control method based on model predictive control is characterized by the following steps: Based on multilayer network theory, key parameters of production on the production line are abstracted into a network, and a multi-dimensional network model is constructed based on the relationships between these key parameters; based on the multi-dimensional network model, the changed parameter nodes and other key nodes affected by them are identified as the first node, and a predictive model is constructed using the first node; a control cycle and data thresholds are set; the first moment of the control cycle is recorded, and predictive data for each moment within the remaining control cycle is obtained through the predictive model. This invention aims to ensure that the implementation process of production line design changes is controllable and efficient, significantly reduce the chain reactions generated during the change process, and avoid unnecessary large-scale redesign.

Description

Technical Field

[0001] This invention relates to the field of production regulation technology, and in particular to a production line change propagation control method and system based on model predictive control. Background Technology

[0002] In modern manufacturing, production line design changes are made to improve production efficiency, reduce costs, adapt to changing market demands, or accommodate technological advancements. As production line designs become increasingly complex, the parameters involved—including equipment, processes, procedures, and materials—are also multiplying. Therefore, the dissemination and updating of relevant parameters become crucial whenever design changes occur. If the changed parameters are not promptly and accurately transmitted to all stages of the production line, or if this transmission triggers a chain reaction, it can lead to the collapse of the design scheme, or even force a large-scale redesign of the entire production line. This significantly extends the design cycle, increases design costs, and impacts production efficiency and product quality.

[0003] In particular, during the production line design process, parameter changes often propagate in a "snowball effect," meaning that a small design change can trigger a chain reaction of adjustments to multiple stages and parameters, rendering the system design ineffective and even requiring a reassessment of the entire production line layout and processes. This not only increases design time and costs but can also lead to production interruptions or quality problems, severely impacting a company's market competitiveness.

[0004] Therefore, how to effectively control the propagation of parameters during production line design changes, avoid a cascading effect of changes, and ensure the controllability, timeliness, and rationality of changes has become a key technical problem that urgently needs to be solved in modern manufacturing. Summary of the Invention

[0005] To address the aforementioned shortcomings, the present invention aims to propose a production line change propagation control method and system based on model predictive control, thereby avoiding the problem of avalanche-like chain propagation caused by a change in a single parameter of the production line.

[0006] To achieve this objective, the present invention adopts the following technical solution: a production line change propagation control method based on model predictive control, comprising the following steps: Step S1: Based on multi-layer network theory, the key parameters of production on the production line are abstracted into a network, and a multi-dimensional network model is constructed based on the relationship between the key parameters.

[0007] Based on the multi-dimensional network model, identify the changed parameter nodes and other key nodes affected by them, and use them as the first node to build a prediction model.

[0008] Step S2: Set the control cycle and data threshold;

[0009] Step S3: Record the first moment of the control cycle, and obtain the prediction data for each moment in the remaining control cycle through the prediction model. Determine if all the prediction data are less than the data threshold. If all the prediction data are less than the data threshold, then proceed to step S4.

[0010] If all predicted data are not all less than the data threshold, then the control variables in the prediction model are adjusted by the objective function to obtain the updated prediction model, and step S4 is executed.

[0011] Step S4: Delete the first time point and update the first time point with the next time point. Repeat step S3 until the first time point is the last time point in the control period, and output the control variables in the prediction model.

[0012] Preferably, the key parameters include the configuration, dynamics, control, and optimization of the production line, and each key parameter contains at least one parameter node;

[0013] The criteria for constructing the multi-dimensional network model in step S1 are as follows:

[0014] Each layer of the network consists of parameter nodes with the same properties. The relationships between different parameter nodes in the production line are abstracted as edges. The parameter nodes and edges together form a multi-dimensional network model of the production line.

[0015] Preferably, the prediction model is as follows:

[0016] L(t+1)=AL(t)+Bβ(t)L(t) α +Du(t);

[0017] Where A represents the inherent coupling relationship coefficient between parameter nodes, B represents the weight coefficient of the allocation strategy, D represents the path blocking influence coefficient, α is the capacity parameter, and β(t) and u(t) are both control variables, where β(t) represents the load allocation coefficient at time t, and u(t) is the propagation path switch of the 0 / 1 vector;

[0018] L(t) = [L1(t), L1(t), ... L N (t)] T ;

[0019] Where L N (t) represents the load of the Nth parameter node at time t, where N is the total number of parameter nodes;

[0020] There is a nonlinear relationship between the load of the parameter node and the load of the parameter node at the next time step. The specific nonlinear relationship is as follows:

[0021] L i (t+1)=ωL i (t)+Ση[Lj (t)] θ ,

[0022] ω, θ, and η are all capacity parameters.

[0023] Preferably, the objective function is as follows:

[0024]

[0025] in

[0026]

[0027] Where H p To control the total length of the cycle, Q i L represents the load tracking weight of the i-th parameter node. ref,i R represents the tracking load balancing target of the i-th parameter node. i ΔC represents the threshold adjustment weight of the i-th parameter node. i (t) is also a control variable, ΔC i (t) represents the capacity threshold of the i-th parameter node at time t, S i This represents the path blocking weight of the i-th parameter node.

[0028] Preferably, the objective function also has the following constraints:

[0029]

[0030] ∑ΔC i (t+k)≤ΔC max ;

[0031] Where β max C represents the maximum load distribution factor. max This represents the capacity threshold of the node with the largest parameter.

[0032] A production line change propagation control system based on model predictive control, using the aforementioned production line change propagation control method based on model predictive control, includes: a construction module, a setting module, a monitoring module, and an output module;

[0033] The building module is used to abstract the key parameters of production on the production line into a network based on the multi-layer network theory, and to construct a multi-dimensional network model based on the relationship between the key parameters.

[0034] Based on the multi-dimensional network model, identify the changed parameter nodes and other key nodes affected by them, and use them as the first node to build a prediction model.

[0035] The settings module is used to set the control cycle and data thresholds;

[0036] The monitoring module is used to record the first moment of the control cycle and obtain the prediction data for each moment in the remaining control cycle through the prediction model. It determines that all the prediction data are less than the data threshold. If all the prediction data are less than the data threshold, the output module is called.

[0037] If all predicted data are not all less than the data threshold, the control variables in the prediction model are adjusted through the objective function to obtain the updated prediction model, and the output module is called.

[0038] The output module is used to delete the first moment, update the next moment as the first moment, and re-call the monitoring module until the first moment is the last moment in the control cycle, and output the control variables in the prediction model.

[0039] One of the above technical solutions has the following advantages or beneficial effects: the model predictive control production line change propagation control method, combined with real-time collaboration and feedback mechanism, aims to ensure that the implementation process of production line design changes is controllable and efficient, and significantly reduce the chain reaction generated during the change process, avoiding unnecessary large-scale redesign. Attached Figure Description

[0040] Figure 1 This is a flowchart of one embodiment of the present invention.

[0041] Figure 2 This is a schematic diagram of the structure of one embodiment of the system of the present invention.

[0042] Figure 3 This is a schematic diagram of a multi-dimensional network model according to an embodiment of the present invention. Detailed Implementation

[0043] Embodiments of the present invention are described in detail below. Examples of these embodiments are shown in the accompanying drawings, wherein the same or similar reference numerals denote the same or similar elements or elements having the same or similar functions throughout. The embodiments described below with reference to the accompanying drawings are exemplary and are only used to explain the present invention, and should not be construed as limiting the present invention.

[0044] In the description of embodiments of the present invention, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of indicated technical features. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more of the stated features. In the description of embodiments of the present invention, "a plurality of" means two or more, unless otherwise explicitly specified.

[0045] Furthermore, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of indicated technical features. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more of that feature. In the description of this invention, unless otherwise stated, "a plurality of" means two or more. Those skilled in the art will understand the specific meaning of the above terms in this invention based on the specific circumstances.

[0046] like Figures 1-3 As shown, a production line change propagation control method based on model predictive control includes the following steps: Step S1: Based on multi-layer network theory, the key parameters of production on the production line are abstracted into a network, and a multi-dimensional network model is constructed based on the relationship between the key parameters.

[0047] Based on the multi-dimensional network model, identify the changed parameter nodes and other key nodes affected by them, and use them as the first node to build a prediction model.

[0048] Step S2: Set the control cycle and data threshold;

[0049] Step S3: Record the first moment of the control cycle, and obtain the prediction data for each moment in the remaining control cycle through the prediction model. Determine if all the prediction data are less than the data threshold. If all the prediction data are less than the data threshold, then proceed to step S4.

[0050] If all predicted data are not all less than the data threshold, then the control variables in the prediction model are adjusted by the objective function to obtain the updated prediction model, and step S4 is executed.

[0051] Step S4: Delete the first time point and update the first time point with the next time point. Repeat step S3 until the first time point is the last time point in the control period, and output the control variables in the prediction model.

[0052] To address the problem of avalanche propagation caused by parameter changes in existing technologies, which renders designs unusable or forces large-scale production line modifications, this invention proposes a model predictive control-based production line change propagation control method. Combined with real-time collaboration and feedback mechanisms, this invention aims to ensure the controllable and efficient implementation of production line design changes, significantly reduce the chain reactions generated during changes, and avoid unnecessary large-scale redesigns.

[0053] The following example illustrates this: In a parameter network of an automotive assembly line, the load (control variable) of the core welding parameter node (node ​​1) surges due to design changes. It is necessary to dynamically adjust the load distribution through MPC to avoid cascading failures.

[0054] At this point, it is necessary to abstract the key parameters in the automotive assembly line parameter network into a network, and construct a multi-dimensional network model based on the relationships between the key parameters. This multi-dimensional network allows us to understand the other parameter nodes that need to be changed due to a change in node 1. Then, we collect these other parameter nodes and use them as the first node to construct a corresponding prediction model. This prediction model can predict the overall load of the production line at the next moment. It determines whether the load exceeds a certain threshold. If it does, it indicates that the modification of node 1 has an adverse effect on the production line, requiring modification of the corresponding control variables (such as load distribution coefficients, propagation path switches, or the load of node 1). Modifying the control variables can be achieved through a corresponding objective function. Within a control cycle, the control variables are continuously modified using the prediction model and objective function to ensure the production line meets production requirements without causing adverse effects.

[0055] It is worth mentioning that the periodic regulation is carried out through the "execution of the first step control" mechanism in this invention.

[0056] The following example illustrates this. Assume the control cycle is set to 5, and the first time step t = 1. The current prediction model is used to predict values ​​at t = 1, 2, 3, 4, and 5 respectively. If the load is exceeded, the prediction model is updated, and the first time step is changed to t = 2. The new prediction model is then used to predict values ​​at t = 2, 3, 4, and 5 respectively, and the judgment of whether the load exceeds the threshold is repeated. That is, in each control cycle, only the optimal control action at the first time step (t) is implemented, rather than executing all control commands for the entire control cycle. This mechanism achieves adaptive adjustment of complex dynamic systems by dynamically updating the system state and re-optimizing.

[0057] Preferably, the key parameters include the configuration, dynamics, control, and optimization of the production line, and each key parameter contains at least one parameter node;

[0058] The criteria for constructing the multi-dimensional network model in step S1 are as follows:

[0059] Each layer of the network consists of parameter nodes with the same properties. The relationships between different parameter nodes in the production line are abstracted as edges. The parameter nodes and edges together form a multi-dimensional network model of the production line.

[0060] The production line is divided into four dimensions: configuration, movement, control, and optimization.

[0061] Configuration: This mainly focuses on the physical planning and configuration of the production line. It involves determining the overall structure of the production line based on product production needs, process requirements, and resource allocation. Configuration design includes not only equipment selection and layout but also connection methods between equipment, transmission paths, and workstation configurations. Configuration parameters include: dimensions, length, width, and area.

[0062] Motion type: Focuses on the coordination and planning of movements of various work units and equipment in the production line. Specifically, it refers to how to ensure the smooth and efficient production process by rationally arranging various tasks (such as equipment start-up, operation, transfer, and shutdown). In designing the behavior, factors such as the working sequence of each piece of equipment, task scheduling, and action timing need to be considered to avoid unnecessary waiting and conflicts. Motion type parameters include: speed, acceleration, etc.

[0063] Control type: This aims to achieve information exchange and coordinated control among various devices and workstations in a production line. The core of control design lies in how to achieve precise interaction between information systems and physical equipment. This includes the construction of control networks, data acquisition and analysis, decision-making, and command issuance. Effective control design enables automatic monitoring, real-time scheduling, and feedback adjustments of the production line, thereby improving the flexibility and responsiveness of the production process. Seamless connection between information and the physical world makes the entire production process more intelligent and adaptive. Control type parameters include: control accuracy, response time, etc.

[0064] Optimization: The core of the production line optimization process, it achieves optimal production operation by optimizing the entire line's drive engine. The focus is on precisely controlling each link of the production line through multi-dimensional data analysis and model optimization to achieve the best balance between production efficiency, quality, and cost. It involves optimization in multiple aspects, from production scheduling to equipment maintenance, from energy management to quality control. Optimization parameters include: capacity, yield, etc. The resulting multi-line network model is as follows... Figure 3 As shown.

[0065] Preferably, the prediction model is as follows:

[0066] L(t+1)=AL(t)+Bβ(t)L(t) α +Du(t);

[0067] Where A represents the inherent coupling relationship coefficient between parameter nodes, B represents the weight coefficient of the allocation strategy, D represents the path blocking influence coefficient, α is the capacity parameter, and β(t) and u(t) are both control variables, where β(t) represents the load allocation coefficient at time t, and u(t) is the propagation path switch of the 0 / 1 vector;

[0068] L(t) = [L1(t), L1(t), ... LN (t)] T ;

[0069] Where L N (t) represents the load of the Nth parameter node at time t, where N is the total number of parameter nodes;

[0070] There is a nonlinear relationship between the load of the parameter node and the load of the parameter node at the next time step. The specific nonlinear relationship is as follows:

[0071] L i (t+1)=ωL i (t)+∑η[L j (t)] θ ,

[0072] ω, θ, and η are all capacity parameters.

[0073] The prediction model is used to make predictions during the control period, which requires predictions for all moments within the period. At this time, the data for the next moment can be obtained through nonlinear relationships, and the load data of the parameter nodes at any moment within the control period can be obtained by repeatedly using nonlinear relationships.

[0074] Preferably, the objective function is as follows:

[0075]

[0076] in

[0077]

[0078] Where H p To control the total length of the cycle, Q i L represents the load tracking weight of the i-th parameter node. ref,i R represents the tracking load balancing target of the i-th parameter node. i ΔC represents the threshold adjustment weight of the i-th parameter node. i (t) is also a control variable, ΔC i (t) represents the capacity threshold of the i-th parameter node at time t, S i This represents the path blocking weight of the i-th parameter node.

[0079] The first item Tracking load balancing target L ref Load tracking deviation cost: penalizes parameter nodes for load deviation from the ideal reference value. (Second item) The penalty threshold is frequently adjusted; the penalty parameter is the node capacity threshold ΔC. i The adjustment range of (t). The third item is... The penalty threshold is adjusted frequently; the number of decisions made to block the transmission path or the criticality of the penalty are also considered.

[0080] It's worth noting that Q is the load tracking weight, and a larger Q is assigned to highly sensitive nodes (those prone to triggering avalanches). i Strictly control its load deviation. S is the path blocking weight; for core process paths (such as welding → painting), set an extremely high S. i This can help avoid disruptions as much as possible.

[0081] Preferably, the objective function also has the following constraints:

[0082] To limit node capacity;

[0083] To constrain the range of control variables;

[0084] This is path blocking logic;

[0085] ∑ΔC i (t+k)≤ΔC max Constrained by total resource availability;

[0086] Where β max C represents the maximum load distribution factor. max This represents the capacity threshold of the node with the largest parameter.

[0087] A production line change propagation control system based on model predictive control, using the aforementioned production line change propagation control method based on model predictive control, includes: a construction module, a setting module, a monitoring module, and an output module;

[0088] The building module is used to abstract the key parameters of production on the production line into a network based on the multi-layer network theory, and to construct a multi-dimensional network model based on the relationship between the key parameters.

[0089] Based on the multi-dimensional network model, identify the changed parameter nodes and other key nodes affected by them, and use them as the first node to build a prediction model.

[0090] The settings module is used to set the control cycle and data thresholds;

[0091] The monitoring module is used to record the first moment of the control cycle and obtain the prediction data for each moment in the remaining control cycle through the prediction model. It determines that all the prediction data are less than the data threshold. If all the prediction data are less than the data threshold, the output module is called.

[0092] If all predicted data are not all less than the data threshold, the control variables in the prediction model are adjusted through the objective function to obtain the updated prediction model, and the output module is called.

[0093] The output module is used to delete the first moment, update the next moment as the first moment, and re-call the monitoring module until the first moment is the last moment in the control cycle, and output the control variables in the prediction model.

[0094] In the description of this specification, the references to terms such as "one embodiment," "some embodiments," "illustrative embodiment," "example," "specific example," or "some examples," etc., indicate that a specific feature, structure, material, or characteristic described in connection with that embodiment or example is included in at least one embodiment or example of the invention. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples.

[0095] Although embodiments of the invention have been shown and described, those skilled in the art will understand that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the claims and their equivalents.

Claims

1. A production line change propagation control method based on model predictive control, characterized in that, Includes the following steps: Step S1: Based on multi-layer network theory, the key parameters of production on the production line are abstracted into a network, and a multi-dimensional network model is constructed based on the relationship between the key parameters. Based on the multi-dimensional network model, the changed parameter nodes and other key nodes affected by them are identified as the first node, and a prediction model is constructed using the first node. Step S2: Set the control cycle and data threshold; Step S3: Record the first moment of the control cycle, and obtain the prediction data for each moment in the remaining control cycle through the prediction model. Determine if all the prediction data are less than the data threshold. If all the prediction data are less than the data threshold, then proceed to step S4. If all predicted data are not all less than the data threshold, then the control variables in the prediction model are adjusted by the objective function to obtain the updated prediction model, and step S4 is executed. Step S4: Delete the first time point and update the first time point with the next time point. Repeat step S3 until the first time point is the last time point in the control period. Output the control variables in the prediction model. The prediction model is as follows: ； Where A represents the inherent coupling coefficient between parameter nodes, B represents the weight coefficient of the allocation strategy, and D represents the path blocking impact coefficient. For capacity parameters, and All are control variables, among which This represents the load distribution coefficient at time t. A propagation path switch for 0 / 1 vectors; ； in This represents the load of the Nth parameter node at time t, where N is the total number of parameter nodes; There is a nonlinear relationship between the load of the parameter node and the load of the parameter node at the next time step. The specific nonlinear relationship is as follows: ， All are capacity parameters; The specific objective function is as follows: ； in ; ； ； in To control the total length of the cycle, This represents the load tracking weight of the i-th parameter node. This represents the tracking load balancing target of the i-th parameter node. This represents the threshold adjustment weight for the i-th parameter node. Also for controlling variables, This represents the capacity threshold of the i-th parameter node at time t. This represents the path blocking weight of the i-th parameter node.

2. The production line change propagation control method based on model predictive control according to claim 1, characterized in that, The key parameters include the configuration, dynamics, control, and optimization of the production line, and each key parameter contains at least one parameter node; The criteria for constructing the multi-dimensional network model in step S1 are as follows: Each layer of the network consists of parameter nodes with the same properties. The relationships between different parameter nodes in the production line are abstracted as edges. The parameter nodes and edges together form a multi-dimensional network model of the production line.

3. The production line change propagation control method based on model predictive control according to claim 2, characterized in that, The objective function is also subject to the following constraints: ； ； ； ； in Indicates the maximum load distribution factor. This represents the capacity threshold of the node with the largest parameter.

4. A production line change propagation control system based on model predictive control, characterized in that, The production line change propagation control method based on model predictive control according to any one of claims 1 to 3 includes: a construction module, a setting module, a monitoring module, and an output module; The building module is used to abstract the key parameters of production on the production line into a network based on the multi-layer network theory, and to construct a multi-dimensional network model based on the relationship between the key parameters. Based on the multi-dimensional network model, identify the changed parameter nodes and other key nodes affected by them, and use them as the first node to build a prediction model. The settings module is used to set the control cycle and data thresholds; The monitoring module is used to record the first moment of the control cycle and obtain the prediction data for each moment in the remaining control cycle through the prediction model. It determines that all the prediction data are less than the data threshold. If all the prediction data are less than the data threshold, the output module is called. If all predicted data are not all less than the data threshold, the control variables in the prediction model are adjusted through the objective function to obtain the updated prediction model, and the output module is called. The output module is used to delete the first moment, update the first moment with the next moment, call the monitoring module again, until the first moment is the last moment in the control cycle, and output the control variables in the prediction model. The prediction model is as follows: ； Where A represents the inherent coupling coefficient between parameter nodes, B represents the weight coefficient of the allocation strategy, and D represents the path blocking impact coefficient. For capacity parameters, and All are control variables, among which This represents the load distribution coefficient at time t. A propagation path switch for 0 / 1 vectors; ； in This represents the load of the Nth parameter node at time t, where N is the total number of parameter nodes; There is a nonlinear relationship between the load of the parameter node and the load of the parameter node at the next time step. The specific nonlinear relationship is as follows: ， All are capacity parameters; The specific objective function is as follows: ； in ; ； ； in To control the total length of the cycle, This represents the load tracking weight of the i-th parameter node. This represents the tracking load balancing target of the i-th parameter node. This represents the threshold adjustment weight for the i-th parameter node. Also for controlling variables, This represents the capacity threshold of the i-th parameter node at time t. This represents the path blocking weight of the i-th parameter node.

Images