Method, device and storage medium for constructing electromechanical system function module

Abstract

The present application belongs to the technical field of mechatronic system design, and particularly relates to a mechatronic system function module construction method, device, equipment and storage medium. The specific process of the method is as follows: step one, obtaining the requirement-function-structure mapping relationship of the mechatronic system; step two, based on the mapping relationship, performing mechatronic system function module division; the specific process is as follows: S201, based on the requirement-function-structure mapping relationship, constructing a requirement-function matrix and a function-structure mapping matrix, determining the correlation degree of the structure component on the function attribute and the correlation relationship on the structure attribute, and establishing the comprehensive correlation degree of the structure component; S202, determining the index of module division, including the cohesion degree of the function module and the coupling degree between the function modules, establishing a target function based on the criterion of high clustering degree and low coupling degree; S203, based on the target function, identifying the module grouping mode, and forming the function module division.

Description

Technical Field

[0001] This invention belongs to the field of electromechanical system design technology, and specifically relates to a method, apparatus, equipment and storage medium for constructing functional modules of an electromechanical system. Background Technology

[0002] Currently, the functional reconfiguration of multidisciplinary coupled systems mainly relies on traditional modular design and optimization methods.

[0003] Modular design: The system is decomposed into several independent modules, each responsible for a specific function, and interoperability between modules is achieved through predefined interfaces. This method can improve the flexibility and reusability of the design to a certain extent.

[0004] Optimization methods: Optimize the combination of modules using methods such as linear programming and dynamic programming, usually with certain specific performance indicators (such as cost, weight, and time) as optimization objectives.

[0005] These methods typically simplify the design process by decomposing the system into several independent modules, but they are insufficient when dealing with multidisciplinary coupling relationships. For example, some existing technologies use linear programming or dynamic programming methods for module combination and optimization, but these methods can usually only handle simple coupling relationships within a single discipline. For coupling relationships in complex systems involving multiple disciplines, especially optimization under resource constraints, there is a lack of effective solutions.

[0006] 1) Limited ability to handle complex coupling relationships: Existing technologies, when dealing with multidisciplinary coupled systems, can typically only handle relatively simple module combinations. They lack effective solutions for complex coupling relationships between disciplines, especially for optimization under resource constraints. This limitation can easily lead to conflicts or inconsistencies between functional modules in the system design, thereby affecting the overall optimization effect of the system.

[0007] 2) Lack of multi-objective optimization capability: Existing technologies typically employ single-objective optimization methods, which can only achieve optimality on a single performance metric, and cannot simultaneously consider the optimization of multiple objectives (such as cost, resource utilization, and system performance). This limitation makes it difficult for the system to achieve overall optimality when facing multi-objective constraints, affecting the overall performance of the system and the effectiveness of resource allocation.

[0008] 3) Insufficient resource utilization efficiency: Due to insufficient consideration of resource constraints in existing methods, the design scheme suffers from low resource utilization efficiency, potentially leading to resource waste or shortage. This not only affects the system's economy but may also negatively impact the overall system performance.

[0009] 4) Poor flexibility and adaptability of modular design: Existing modular designs often lack sufficient flexibility and adaptability when dealing with dynamically changing design requirements or external environments, making it difficult to respond and adjust quickly. This limitation makes it difficult for the system to maintain high efficiency and stable performance in changing operating environments. Summary of the Invention

[0010] This invention aims to address the shortcomings of existing technologies in the design of multidisciplinary coupled systems in handling complex coupling relationships and resource optimization, which often lead to low overall system performance and resource utilization efficiency. This invention proposes a multidisciplinary functional reconfiguration modeling method based on resource optimization. By optimizing the combination of functional modules in the system, it ensures that the system achieves global optimality in resources and performance while satisfying multiple constraints. This method can effectively address the challenges posed by resource constraints and multidisciplinary coupling in complex system design, thereby improving the overall system efficiency.

[0011] The technical solution for implementing the present invention is as follows:

[0012] In a first aspect, an embodiment of this application provides a method for constructing functional modules of an electromechanical system, the specific process of which is as follows:

[0013] Step 1: Obtain the requirement-function-structure mapping relationship of the electromechanical system;

[0014] Step two: Based on the mapping relationship, divide the electromechanical system into functional modules; the specific process is as follows:

[0015] S201. Based on the requirement-function-structure mapping relationship, construct the requirement-function matrix and the function-structure mapping matrix, determine the degree of correlation of structural components in functional attributes and the correlation of structural attributes, and establish the comprehensive correlation degree of structural components.

[0016] S202, The indicators for determining the module division include: the cohesion of functional modules and the coupling between functional modules. An objective function is established based on the criteria of high clustering degree and low coupling degree.

[0017] S203, Based on the objective function, identify the module grouping pattern and form a functional module division.

[0018] Furthermore, the functional attributes described in this invention are: the degree of similarity and correlation between the two structural components in terms of functional composition and functional relationship; the structural attributes are: the correlation between the two structural components in multiple disciplines.

[0019] Furthermore, the comprehensive correlation degree described in this invention is:

[0020]

[0021] Where α and β are the set weight parameters, For functional relevance, This refers to the degree of correlation in structural attributes.

[0022] Furthermore, the cohesion of this invention is calculated as follows:

[0023]

[0024] In the formula |S k |For functional module S k The size of the set, c ij For functional module S k The overall correlation between structural component i and structural component j.

[0025] Furthermore, the two functional modules S of the present invention k S k′ The formula for calculating the coupling degree between them is:

[0026]

[0027] Furthermore, step S203 of the present invention is implemented using the fuzzy C-means clustering method, and the specific process is as follows:

[0028] First, clustering methods are used to group functional components, grouping components with similar functions or structural dependencies into the same module to form a preliminary set of module partitioning schemes;

[0029] Secondly, grouping is evaluated and selected based on the objective function;

[0030] Finally, the final module partitioning scheme is determined by evaluating the functional integrity, independence, and cost factors of each module.

[0031] Furthermore, step S203 of the present invention employs a genetic algorithm to implement functional module division, specifically as follows:

[0032] The constraints include: (1) full coverage constraint: ensuring that each component is assigned to exactly one module; (2) module non-empty constraint: ensuring that each module contains at least one component.

[0033] Define the objective function:

[0034]

[0035] Where λ is the equilibrium parameter.

[0036] Based on constraints and a defined objective function, a genetic algorithm is used to divide functional modules.

[0037] Furthermore, this invention also includes a step for rapid functional reconfiguration based on the Cooperative Optimization (CO) method. Before utilizing CO for functional reconfiguration, basic assumptions are proposed regarding modular design, resource constraints, rapid reconfiguration, and cooperative optimization.

[0038] Modular design assumption: Based on the modular concept, the overall system is divided into multiple levels, including the top-level system level, subsystem level, and module level;

[0039] Source constraint assumption: It is assumed that the budget C, time T, materials M, and manufacturing process are finite and known throughout the entire refactoring process;

[0040] Rapid refactoring assumption: Rapid refactoring involves not only performance optimization within modules, but also requires ensuring coordinated operation between various functional subsystems at the system level;

[0041] Collaborative optimization: The objective function Z for system-level optimization sys (X) and subsystem-level optimization objectives Subject-level optimization objective function Coupled through the following relationship:

[0042]

[0043] Among them, w k and w i It is a weighting coefficient, representing the trade-off between system-level goals and the goals of individual disciplines, g(z kl ) represents the constraint function for the coupling relationship between subsystem k and subsystem l, g(z) ij ) represents the constraint function representing the coupling relationship between module i and module j, λ kl and λ ij It is a penalty system for coupling relationships;

[0044] Constraints for module-level optimization:

[0045]

[0046] subject tog(z ij ) = 0

[0047] Secondly, embodiments of the present invention also provide an electromechanical system functional module construction device, comprising:

[0048] The first processing module is used to obtain the requirement-function-structure mapping relationship of the electromechanical system;

[0049] The second processing module is used to divide the electromechanical system into functional modules based on the mapping relationship, specifically as follows:

[0050] Based on the requirement-function-structure mapping relationship, a requirement-function matrix and a function-structure mapping matrix are constructed to determine the degree of correlation of structural components in terms of functional attributes and the correlation of structural attributes, and to establish the comprehensive correlation degree of structural components.

[0051] The indicators for determining module partitioning include: the cohesion of functional modules and the coupling between functional modules. An objective function is established based on the criteria of high clustering degree and low coupling degree.

[0052] Based on the objective function, the module grouping pattern is identified, and the functional module division is formed.

[0053] Thirdly, embodiments of the present invention also provide an electromechanical system functional module construction device, comprising: a transceiver, a processor, a memory, and a program or instructions stored in the memory and executable on the processor; when the processor executes the program or instructions, it implements the steps in the electromechanical system functional module construction method as described above.

[0054] Fourthly, embodiments of the present invention also provide a readable storage medium having a program or instructions stored thereon, wherein the program or instructions, when executed by a processor, implement the steps in the electromechanical system functional module construction method as described above.

[0055] Beneficial effects:

[0056] The electromechanical system functional module construction method provided by this invention determines the functional modules and structural components within the electromechanical system based on the system functional requirements. Considering the task requirements of low-cost design, the optimal module division principle is high cohesion and low coupling. This design principle not only helps improve the reliability and performance of the system but also simplifies the manufacturing process and improves tooling changeover efficiency. Based on this design principle, design resources are allocated to each functional module to obtain the completed electromechanical system. This completed electromechanical system reduces the possibility of improper resource allocation or functional module conflicts in later development, lowers the overall design and development costs, improves the efficiency of system development, can quickly respond to dynamic changes in design requirements, ensures efficient operation in various environments, and enhances the adaptability and scalability of the system. Attached Figure Description

[0057] Figure 1 This is the basic process for dividing the functional modules in the embodiments of the present invention;

[0058] Figure 2 This is a flowchart illustrating the requirement-function-structure mapping relationship of an electromechanical system according to an embodiment of the present invention.

[0059] Figure 3 This is a flowchart of fuzzy C-means clustering provided in an embodiment of the present invention;

[0060] Figure 4 A basic flowchart of cross-processing provided in this embodiment of the invention;

[0061] Figure 5 A flowchart of the genetic algorithm provided in an embodiment of the present invention;

[0062] Figure 6 A flowchart illustrating the multidisciplinary functional reconstruction mathematical model technology that considers resource optimization, provided for embodiments of the present invention. Detailed Implementation

[0063] To make the technical problems, technical solutions and advantages of the present invention clearer, a detailed description will be given below in conjunction with the accompanying drawings and specific embodiments.

[0064] Basic concept definitions:

[0065] System Functions: In complex electromechanical systems, functions refer to the specific capabilities of each subsystem at different operational stages, such as target detection, tracking, navigation, motion control, and task execution. Through the parallel or serial operation of various functions within the system, the system can accurately and reliably complete the predetermined task objectives. The realization of these functions depends on the coordinated coupling between the subsystems to ensure optimal overall performance.

[0066] Requirements: In the design and application of complex electromechanical systems, requirements determine how the system should operate in a specific working environment to meet expected objectives. Requirements may change throughout the system's lifecycle, from design to use. For example, the system may need to accurately identify and track targets while performing complex operational tasks in noisy or disruptive environments.

[0067] Modular or functional modules: Modular design of complex electromechanical systems can greatly simplify the design process, enabling the system to better adapt to different operating environments and task requirements. A complex electromechanical system typically consists of multiple independent modules, such as sensing modules, drive modules, control modules, and execution modules. Each module comprises several parts and sub-components, responsible for a set of related functions within the subsystem, and connects and adapts to other modules through predefined interfaces. For complex electromechanical systems, modules can be classified according to their functional structure characteristics into general-purpose modules, dedicated functional modules, and series modules. General-purpose modules are usually standard components in the system, employing a universal standardized architecture and possessing high compatibility, such as navigation modules. Dedicated functional modules are highly customized modules designed for specific tasks or operating environments, such as identification modules and high-precision control modules for specific targets. Series modules fall between general-purpose and dedicated modules, typically consisting of a series of different models of functional modules using a universal interface, such as drive modules with different power ratings or sensing modules with different precision levels.

[0068] Basic Components: Components are the basic building blocks and sub-units of various modules in a complex electromechanical system. Taking a complex electromechanical system as an example, a navigation module may contain multiple sensors (such as infrared sensors and radar sensors), signal processors, and communication equipment. Components are the basic units that realize the functions of each functional module. They integrate with other components through internal interfaces to achieve the overall function of the system.

[0069] The design concept of this invention is as follows:

[0070] In the design of complex, multidisciplinary coupled systems, reasonable modularization is fundamental to achieving system optimization and efficient operation. To ensure that the system can achieve optimal performance under resource constraints, it is necessary to rationally divide the system into functional modules. The principle of modular design is high cohesion and low coupling, meaning that each module should have highly related functions, while the dependencies between modules should be minimized.

[0071] When designing complex electromechanical systems, considering the requirement of low-cost design, the optimal modularization principle is high cohesion and low coupling (also known as separation). This design principle not only helps improve system reliability and performance but also simplifies the manufacturing process and improves tooling changeover efficiency. Furthermore, modularization must comprehensively consider additional indicators such as the feasibility of modular manufacturing, cost-effectiveness, and standardization of interfaces between modules to ensure the overall efficiency and flexibility of the system in design, manufacturing, and maintenance.

[0072] like Figure 1 As shown in the figure, an embodiment of this application provides a method for constructing functional modules of an electromechanical system, such as... Figure 1 As shown, the specific process is as follows:

[0073] Step 1: Based on the Function-Behavior-Structure (FBS) module functional structure mapping technology, obtain the requirement-function-structure mapping relationship of the electromechanical system.

[0074] In the modular design of complex electromechanical systems, it is often necessary to consider complex operational requirements, multiple functional constraints, and the coupling relationships between various structural components. Therefore, the premise and foundation for module division is to clarify the task requirements and functional details, and to clarify the functional structural mapping relationships between the various module components of the system.

[0075] Function-Behavior-Structure (FBS) modular functional structure mapping is a commonly used method for system functional structure analysis and modeling. This method achieves modeling and decomposition between function and structure through a hierarchical mapping mechanism that maps function to behavior and behavior to structure. Specifically, the FBS method treats system function as a high-level design goal, behavior as the specific manifestation of function, and structure as the specific physical implementation required to achieve these behaviors.

[0076] In complex electromechanical systems, modular design utilizes the Functional Breakdown Structure (FBS) method. Through detailed analysis of the system's function, behavior, and structure, it provides crucial input for subsequent module functional partitioning. The specific process is as follows:

[0077] Requirements analysis: Define the overall functionality of the system and the specific functional requirements of each subsystem. These functional requirements may include operational accuracy, response speed, and environmental adaptability.

[0078] Behavior mapping: Mapping functions to specific behaviors. In this step, each function is broken down into multiple executable behaviors. For example, a navigation function might be mapped to a series of specific behaviors such as path planning, speed adjustment, and attitude control.

[0079] Structure mapping: Behaviors are further mapped to the physical structure of the system. This includes determining the physical implementation of each behavior, such as the placement of sensors, the design of actuators, and the architecture of the control system.

[0080] The Functional Structure Mapping (FBS) method described above clarifies the relationships between system requirements, behaviors, and structure, providing clear guidance for module partitioning. The goal of module partitioning is to ensure that each module has the ability to independently implement specific functions while effectively collaborating with other modules through standardized interfaces. The application of the FBS method in modular design helps optimize overall system performance and improves system flexibility and scalability. To better suit the modular design of modern complex electromechanical systems, an FBS-based module functional structure mapping technique is introduced. Its specific process is as follows: Figure 2 As shown.

[0081] During the full lifecycle requirements analysis phase: It is necessary to collect and analyze the overall requirements of complex electromechanical systems as comprehensively as possible, including operating scenarios, operational needs, technical specifications, and expected performance. The intended use, performance indicators, operating environment, and budget constraints of the system must be defined in detail. This includes specific requirements for the system's key technical specifications. Simultaneously, the requirements analysis should also consider manufacturing, maintainability, and upgrade requirements. The results of the analysis will directly guide subsequent system layer division, function identification, and module structure design, ensuring that the design can meet all operational requirements while possessing future flexibility and scalability. During the design phase, considering the design goals of modular design and scalable adjustment, requirements analysis and functional decomposition will be added, deleted, or modified during system design iterations. The future scalability requirements of modules should be considered. During the maintenance phase, each module should be easily accessible and replaceable to facilitate daily maintenance and upgrades, thereby reducing lifecycle maintenance costs. Core requirements include reusability, compatibility, and ease of transportation. During the tooling phase, core requirements should include modular assembly, easy module disassembly, and standardized interfaces to improve tooling efficiency. During the usage phase, i.e., the system's operation and execution phase, the system should be guaranteed to have a certain level of effectiveness to achieve specific task objectives. Common task requirements include operational flexibility, guaranteed control precision, and high efficiency.

[0082] In the system-level partitioning and functional decomposition phase, complex electromechanical systems are divided into several subsystems according to their functions. Taking a certain complex electromechanical system as an example, it is divided into several main subsystems according to their functions:

[0083] Structural Subsystem: The structural subsystem forms the supporting framework of the entire complex electromechanical system, responsible for maintaining the system's physical integrity and structural stability. Its main functions include providing a load-bearing and support platform, ensuring secure connections between modules, enabling electrical and signal connections, and providing necessary physical protection for the system. The structural subsystem is the foundation of all other subsystems, and its design must consider multiple aspects such as material strength, durability, and thermal protection.

[0084] Control Subsystem: The control subsystem is responsible for the operation control and monitoring of the entire system, ensuring that the system can execute tasks according to predetermined instructions during operation. Its main functions include executing operational control, real-time monitoring of system status, and managing communication between the system and external devices. The control subsystem typically integrates advanced electronic control units and software to achieve precise control and efficient management in complex operating environments.

[0085] Power Subsystem: The power subsystem provides the necessary power support for the complex electromechanical system, ensuring stable operation under various operating conditions. This subsystem is primarily responsible for energy storage and management, power output control, and adjusting power distribution during operation. The design of the power subsystem needs to consider energy efficiency, power management, and power output stability to ensure sufficient energy and power support during system operation.

[0086] Navigation Subsystem: The navigation subsystem is responsible for providing the system with accurate navigation and positioning information, supporting effective path planning and target tracking in dynamic environments. It integrates various sensing technologies, such as GPS, inertial navigation systems, radar, and infrared sensors, to ensure high-precision navigation in complex environments.

[0087] This system-level functional decomposition not only clearly identifies the function and corresponding tasks of each subsystem, but also lays a solid foundation for subsequent modular design and optimization. The functional design of each subsystem needs to be closely aligned with the overall system objectives to ensure the overall coordination and consistency of the system. Furthermore, functional decomposition also needs to consider the coupling relationships between subsystems, achieving efficient communication and collaboration between them through standardized interfaces and protocols to further improve the overall performance and reliability of the system.

[0088] Functional Structure Mapping Analysis Phase: For each subsystem, further, the Function-Behavior-Structure (FBS) mapping and solution are performed at the function-structure level. Taking the dynamic subsystem as an example:

[0089] The primary function of the power subsystem is to manage energy supply and provide operating power for the system. These two functions are implemented by different components. For energy management, the power subsystem is mainly responsible for storing, supplying, and regulating energy flow to ensure that the drive unit or other power devices receive adequate energy under various operating conditions. Key components include energy storage units, energy pumps, energy transmission pipelines, and electronic control units.

[0090] By refining the functions of the power subsystem, the role of each component in the overall system function and the interrelationships between components can be clarified. This functional structure mapping analysis provides an important basis for the modular design of the system, helping to optimize system performance, improve system reliability and maintainability.

[0091] Step 2: Basic Process of Functional Module Division for Complex Electromechanical Systems

[0092] Based on the aforementioned requirements-behavior-structure analysis and functional structure mapping analysis, we can further divide complex electromechanical systems into functional modules. Module division is a crucial step in system design, ensuring the rational distribution of system functions and the efficient operation of modules. The following is the basic process of functional module division:

[0093] S201, Based on the requirement-function-structure mapping relationship, construct the requirement-function matrix and the function-structure mapping matrix, determine the degree of correlation of structural components in functional attributes and the correlation relationship in structural attributes, and establish the comprehensive correlation degree of structural components; the specific process of this step is as follows:

[0094] Requirements-Function-Structure Mapping Decomposition: A detailed analysis of the input requirements list and functional decomposition is performed, mapping them to structural components capable of implementing these functions. This process constructs a requirements-function matrix and a function-structure matrix. Based on these, and considering the physical relationships between components, a comprehensive design structure matrix is ​​established.

[0095] To systematically evaluate the interrelationships between structural components, a comprehensive evaluation method can be used, combining functional and structural attributes for quantitative calculation. Functional attributes: the degree of similarity and correlation between two structural components in terms of functional composition and functional relationship; structural attributes: the correlation between two structural components in mechanical, electrical, and other disciplines.

[0096] The correlation between functional attributes can be quantified using the aforementioned "function-structure" correlation matrix. Functional correlation can be calculated by summing the correlation strength of each pair of component functions according to certain weights. The "function-structure" correlation matrix is ​​defined as follows, and the formula for calculating functional correlation is:

[0097]

[0098] Among them, w k Let A be the weight of the k-th discipline. ik Let A be the component weight vector for function i. ij Let be the component weight vector for function j.

[0099] correlation in structural attributes The structural correlation between disciplines such as mechanical and electrical engineering can be scored using expert scoring methods, and then weighted according to certain weights.

[0100] By combining the evaluation results of functional and structural attributes, the final inter-component correlation assessment can be obtained. This application uses a simple additive linear combination to achieve this, i.e., the comprehensive correlation calculation formula is:

[0101]

[0102] Here, α and β are the set weight parameters.

[0103] Through the quantitative analysis of the above process, the correlation matrix of the structural components can be obtained.

[0104] S202, The indicators for determining the module division include: the cohesion of functional modules and the coupling between functional modules. An objective function is established based on the criteria of high clustering degree and low coupling degree.

[0105] Establish fundamental principles for module partitioning: Based on the project's actual objectives, such as low-cost design, performance reliability, and technical feasibility, formulate the most suitable fundamental principles for functional module partitioning. These principles will guide module partitioning, ensuring that modular design can balance system performance and cost-effectiveness.

[0106] The basic principle is high cohesion and low coupling: Here, we first give two indicators for quantifying the results of module partitioning: cohesion and coupling.

[0107] Cohesion is a metric used to measure the strength of the correlation between components within a module. Higher cohesion in a functional module indicates a tighter connection between its components, meaning these components work together to achieve the module's functionality. Let's define a functional module S. k The formula for calculating cohesion is:

[0108]

[0109] In the formula |S k |For functional module S k The size of the set, c ij For functional module S k The overall correlation between structural component i and structural component j.

[0110] Coupling degree is an indicator used to measure the degree of dependency or coupling between two functional modules. The lower the coupling degree between two functional modules, the more independent they are. If a functional module has low coupling degree with other functional modules, it indicates high independence; changes to other modules have little impact on this module, meaning it has high compatibility. Define two functional modules S. k S k′ The formula for calculating the coupling degree between them is:

[0111]

[0112] On the one hand, it requires maximizing cohesion, ensuring that the functions and components within each module are closely related, and that functional modules can implement their functions as independently as possible. On the other hand, it requires minimizing coupling, reducing direct dependencies between modules, so that individual modules can be flexibly replaced or modified without affecting the functionality of other modules.

[0113] S203, Based on the objective function, identify the module grouping pattern and form a functional module division scheme.

[0114] In this embodiment, clustering can be used to determine the functional module partitioning scheme. Specifically, clustering analysis is used to group functional components, grouping components with similar functions or structural dependencies into the same module. Through clustering analysis, module grouping patterns are identified, and a preliminary set of module partitioning schemes is formed for further evaluation and selection. Specifically, the Fuzzy C-means (FCM) clustering algorithm is chosen to implement the functional module clustering of components. Fuzzy C-means (FCM) is a soft clustering method that uses membership degrees to represent the relationship between each data point, thereby determining the cluster to which each functional component belongs (i.e., the functional module). Therefore, the potential relationships and dependencies between functions can be automatically identified through the Fuzzy C-means clustering algorithm, thereby determining which functions should be grouped into one module. The specific process is as follows: Figure 3 As shown.

[0115] Module Partition Analysis and Adjustment: Finally, by evaluating factors such as the functional completeness, independence, and cost of each module, the final module partitioning scheme is determined. If necessary, the module partitioning results are adjusted to ensure the rationality and feasibility of the system design.

[0116] Furthermore, in the above embodiments, functional module division can also be implemented using a genetic algorithm.

[0117] Fuzzy clustering provides an effective method for initial functional module partitioning, but it still has limitations in many more complex real-world constraints and objectives. Genetic algorithms, a commonly used metaheuristic optimization technique, can generate more refined module partitioning solutions. Compared to directly calling clustering algorithms, genetic algorithms can consider more complex optimization objectives and user-defined constraints, outputting possible functional module partitioning schemes based on these constraints.

[0118] Modularizing structural components is equivalent to dividing a certain set (all components) into several subsets (functional modules). Therefore, the modularization problem of components can be transformed into a mathematical model of set partitioning with specific objectives and constraints. By solving this mathematical model, the final functional module partitioning result can be obtained.

[0119] Problem Definition: Given a set S = {c1, c2, ..., cn} of n components. n The task is to partition the dataset into k disjoint subsets S1, S2, ..., Sn. k Each subset represents a functional module, and each component belongs to one and only one module. The degree of association between structural component i and structural component j is known to be c. ij .

[0120] Decision variable: Define a two-dimensional binary decision variable x. ik , where x ik =1 indicates that structural component i is assigned to module set k.

[0121] Constraints:

[0122] (1) Full Coverage Constraint: Ensure that each component is assigned exactly to one module.

[0123]

[0124] (2) Module non-empty constraint: Ensure that each module contains at least one component.

[0125]

[0126] (3) The above constraints based on decision variables can also be expressed in a set-based form.

[0127]

[0128] Objective function: The basic objective is to maximize the cohesion within modules while minimizing the coupling between modules. Defined as f cohesion (S k Module S k cohesion, f coupling (S i ,S j ) represents module S i and S j The degree of coupling between them. The optimization objective of maximizing cohesion can be expressed as:

[0129]

[0130] The optimization objective of minimizing coupling can be expressed as:

[0131]

[0132] This is a multi-objective optimization problem. For convenience, a balance parameter λ∈[0,1] is introduced to adjust the importance of cohesion and coupling. This leads to the comprehensive optimization objective:

[0133]

[0134] Based on the above definitions, the basic process of functional module partitioning based on genetic algorithms is as follows: Figure 3 As shown,

[0135] First, an initial functional module partitioning scheme is generated. Then, a fitness function is constructed based on the objective function, and the fitness of the initial population composed of multiple allocation schemes is evaluated using the fitness function. Second, chromosomes in the population are processed through operations such as selection, crossover, and mutation to generate new individuals with higher fitness. Through generational inheritance and evolution, the allocation scheme with the highest fitness is obtained. The specific process is as follows: Figure 4 As shown.

[0136] (1) Coding Design

[0137] In this embodiment, a chromosome (individual) is represented by a complete functional module partitioning scheme. To intuitively represent the meaning of chromosome encoding, integer encoding is used, with each gene representing a component, and the gene value indicating the functional module group assigned to that component. Using x... i The value of the i-th gene locus in chromosome X[k], i.e., the module grouping of the i-th component, is mathematically represented as follows:

[0138] X k =(x1,…,x i ,…,x n ),x i ∈1,2,…k

[0139] Similarly, N functional module partitioning schemes can be generated as an initial population for subsequent genetic algorithm iterations. Typically, the quality and diversity of the initial population solutions significantly impact the algorithm's convergence speed and the final solution. Therefore, in the module partitioning problem, a heuristic strategy based on problem characteristics can be used to generate the initial population, such as pre-grouping components according to their inter-component relationships.

[0140] (2) Fitness function

[0141] The fitness function is a function used in optimization algorithms to measure the performance of each individual component in the problem optimization process. In the genetic algorithm for functional module partitioning in this project, the fitness function is used to evaluate the quality of a solution for a given module partitioning scheme. The independence between components can be evaluated by calculating the component pair correlation between different modules.

[0142] Based on the mathematical model in the previous section, for a given module partitioning scheme X, its fitness function is:

[0143]

[0144] (3) Selection Strategy

[0145] The purpose of selection strategies is to select a subset of highly fit individuals from the current population according to certain rules to generate the next generation. Common selection strategies include Roulette Wheel Selection, Elitism, and Tournament Selection. In Roulette Wheel Selection, the probability of each individual being selected is proportional to its fitness. Elitism directly replicates a subset of the most fit individuals from the current population to the next generation. Tournament Selection randomly selects a group of individuals and then chooses the individual with the highest fitness from among them.

[0146] (4) Cross operator

[0147] In genetic algorithms, the crossover operator is an operation used to generate new individuals. Crossover simulates the gene exchange process in biological evolution, generating new offspring by exchanging and combining chromosome segments from two parent individuals. Common crossover operators include single-point crossover, two-point crossover, and multi-point crossover. After crossover, it is necessary to check whether the offspring meet the constraints. If not, crossover must be repeated. Taking two-point crossover as an example, the basic process is as follows: Figure 5 As shown:

[0148] (5) Mutation operator

[0149] Mutation is a random perturbation in genetic algorithms that mimics the mutation of biological genes. It is mainly used to introduce new gene combinations, help the algorithm avoid getting stuck in local optima, and increase the possibility of exploring new solutions.

[0150] For an existing chromosome population, the steps for mutation operations are as follows:

[0151] 1. Select mutant individuals: based on the preset mutation probability P mut To determine which individuals need to undergo mutation.

[0152] 2. Selection of mutation sites: For a selected chromosome / individual, for each gene locus, there is a certain probability P′. mut It has mutated.

[0153] 3. Perform mutation: For the selected genes, randomly select one or more gene locations to mutate. There are various mutation methods, such as simple bit flipping (for binary encoding) and randomly assigning new values ​​(for integer or real number encoding).

[0154] 4. Constraint check: Check whether the new genome meets the constraints. If it does, proceed to step 5; otherwise, return to step 2.

[0155] 5. Apply the variation results: Update the individual's genome and reassess its fitness.

[0156] For scenarios with a large number of iterations, the mutation probability can often be dynamically adjusted based on the number of generations. This ensures a broad search within the solution space at the beginning of the iteration to maintain population diversity and prevent getting trapped in local optima. Conversely, during the convergence phase, it ensures a detailed search within the solution space to prevent the optimal solution from being compromised and to improve convergence speed. The formula for calculating the adaptive mutation probability is shown below:

[0157]

[0158] Where, p mmax p mmin These represent the maximum and minimum mutation probabilities, respectively; A is the convergence constant, typically A = 9.903438. max f is the fitness of the best individual in the population; f′ is the fitness of the individual to be crossovered; f is the fitness of the individual to be mutated. This represents the average fitness of the population.

[0159] A genetic algorithm is used for iterative solutions to optimize the cohesion within modules and the coupling between modules. Through iterative selection, crossover, and mutation operations, the genetic algorithm continuously improves the quality of the module partitioning scheme, ultimately yielding the optimal module partitioning scheme.

[0160] By employing the above method to scientifically and rationally divide complex electromechanical systems into modules, not only can the overall design and development efficiency of the system be improved, but the independence and maintainability of each module are also ensured. Each module focuses on a specific functional task, reducing dependencies between modules and thus lowering the system's complexity. This design approach allows for rapid modification and upgrades of individual modules during system design iterations or functional expansions, minimizing the impact on the entire system and ensuring its flexibility and scalability.

[0161] Furthermore, in yet another embodiment of this application, multidisciplinary functional reconfiguration with optimal resource utilization is also considered.

[0162] Functional module partitioning and optimization aim to improve the structural rationality and design efficiency of a system. However, in practical applications, the finiteness of resources (such as time, cost, and energy) is often a key factor affecting the overall system performance. Resource optimization methods can further improve the overall efficiency and economy of a system by optimizing resource allocation within a given functional module framework. Simultaneously, during design iterations and functional expansions, the system often needs to improve performance without increasing resource consumption. Resource optimization models can provide a flexible framework for the system, making it more resilient during functional refactoring and minimizing the impact on existing resources.

[0163] The technical flowchart of the multidisciplinary functional reconfiguration mathematical model considering resource optimization is as follows: Figure 6 As shown.

[0164] Within the framework of modular partitioning and optimization, complex electromechanical systems are divided into multiple functional modules, each involving the integration and optimization of multiple disciplines in its design. While this modular design significantly improves the system's flexibility and adaptability, enabling it to better cope with different task requirements and environmental changes, it also exacerbates the coupling complexity between disciplines. Introducing multidisciplinary design optimization (MDO), which utilizes distributed computing technology and concurrent engineering principles, allows for the simultaneous optimization of multiple disciplines during the design process of complex systems, thereby improving the overall design efficiency. The design benefits of the MDO method can be expressed as:

[0165] Δ Design =∑ i Δ Disciplinary +Δ MDO ,

[0166] Where, Δ Design The total benefit of the MDO method design is represented by ∑ i Δ Disciplinary Δ represents the sum of the benefits of single-discipline design. MDO This represents the increase in benefits after introducing the MDO method. This formula demonstrates that the MDO method can indeed increase the benefits of design.

[0167] Collaborative Optimization (CO) is a multi-level, multidisciplinary design optimization (MDO) method specifically designed to handle the coupling relationships between multiple disciplines in complex systems. However, in the CO method, the optimization task of the entire system is divided into two levels: system-level optimization and discipline-level optimization. But in the design of complex electromechanical systems, there are often subsystem-level optimizations below the system-level optimization; therefore, an improved CO method is introduced.

[0168] To facilitate the establishment of a multidisciplinary functional reconfiguration mathematical model that considers resource optimization, basic assumptions are first proposed regarding four aspects: modular design, resource constraints, rapid reconfiguration, and collaborative optimization.

[0169] Modular design assumption: Based on the modular concept, the overall system is divided into multiple levels, including the top-level system level, subsystem level, and module level. Each level handles different optimization tasks and coupling relationships. The top-level system level mainly handles the coupling between subsystems and global resource allocation; the subsystem level is responsible for optimizing the performance and coordination of internal functional modules; and the module level specifically optimizes the design variables of functional modules. Modular design improves the operability of the design and allows for rapid response and adaptability of the system through independent reconfiguration of modules, even with limited resources. Therefore, this assumption is reasonable and necessary. The system is divided into m subsystems, and the set of variables involved in each subsystem k (k = 1, 2, ..., m) is X = {X...} 1 ,X 2 ,…,X m Within each subsystem k, there are n... k There are three functional modules, and the design variable set for each module i is x. i Then the design variable set of the subsystem is The coupling relationship between subsystems is mediated by the interface variable Z. kl This indicates the dependency relationship between subsystem k and subsystem l.

[0170] Resource Constraint Assumptions: It is assumed that resources (such as budget C, time T, materials M, and manufacturing process) are finite and known throughout the refactoring process, and these resource constraints remain constant throughout the design process. The design and optimization of each functional module must be carried out under these strict resource constraints to ensure the feasibility and economy of the final solution. In real-world engineering projects, resource constraints are an unavoidable reality. By clearly defining resource constraints, we can effectively control costs, shorten development cycles, and ensure the practical applicability of the system in complex environments during the design process; therefore, resource constraint assumptions are also necessary. Let the total budget of the system be C. total The time limit is T. total Material restrictions are M total The constraints can be expressed as:

[0171]

[0172] Among them, C i (X k ), T i (X k M i (X k ) represent the budget, time, and material consumption of the i-th subsystem, respectively.

[0173] Rapid refactoring assumption: It is assumed that the various functional subsystems of the system can be rapidly refactored when requirements change. Rapid refactoring involves not only performance optimization within modules but also ensuring coordinated operation between functional subsystems at the system level to avoid performance degradation or functional failure due to refactoring. Let the new task requirement be d, then the system needs to find a new set of design variables X′ within a finite time, such that:

[0174]

[0175] Here, Z(X′,d) represents the comprehensive performance objective function of the system under the new task requirements. This objective function needs to be minimized or maximized by the refactored design variable X′.

[0176] Collaborative Optimization System: This system assumes that system-level optimization and discipline-level optimization can be separated and reversed, but through iteration and coordination mechanisms, it ensures that the functional refactoring of each module ultimately conforms to the overall system-level goal. System-level optimization handles the coupling relationships and resource allocation between subsystems, while discipline-level optimization optimizes the functional modules of their respective subsystems under given resource constraints. The two coordinate through information transmission and feedback mechanisms to ultimately achieve optimal overall system performance. In complex systems, relying solely on centralized optimization is insufficient to effectively handle the complex coupling relationships between multiple subsystems. The CO method, through hierarchical optimization, achieves efficient management of complex systems, especially under resource constraints, ensuring the consistency between the independent optimization of each discipline and the overall goal. The objective function Z of the system and optimization is... sys (X) and subsystem-level optimization objectives Subject-level optimization objective function Coupled through the following relationship:

[0177]

[0178] Among them, w k and w i It is a weighting coefficient, representing the trade-off between system-level goals and the goals of individual disciplines, g(z kl ) represents the constraint function for the coupling relationship between subsystem k and subsystem l, g(z) ij) represents the constraint function representing the coupling relationship between module i and module j, λ kl and λ ij This is the penalty coefficient for coupling relationships. Below are the constraints for module-level optimization:

[0179]

[0180] subject to g(z ij ) = 0

[0181] Specific application process of the functional module rapid refactoring model

[0182] In the rapid reconfiguration of complex electromechanical systems, the CO (Optimization Framework) serves as the core optimization framework, ensuring that each functional module can complete the reconfiguration quickly and efficiently, given sufficient resources. The CO method reduces the difficulty of handling complex coupling relationships and improves overall optimization efficiency through hierarchical optimization at the system, subsystem, and module levels. The following is a detailed application of the CO method in this model.

[0183] (1) Input:

[0184] Requirement d: Requirements will directly affect the design goals and optimization directions given by the system.

[0185] Resource Limitation: C total ,T total M total Resource constraints include specific limitations on budget, time, materials, and manufacturing processes.

[0186] Initial design state x0: The current design state of the system and the initial configuration of each subsystem.

[0187] (2) Internal logic:

[0188] The CO method consists of two parts: system-level and subject-level, and finally, the two are iteratively optimized.

[0189] Top-level system-level optimization

[0190] Task decomposition: Based on the new requirement d, the system-level optimization module decomposes the task into specific objectives x for each subsystem module. i This includes the performance requirements, resource requirements, and interface requirements between modules for each module.

[0191] Resource allocation: System-level optimization allocates resource quota C to each functional module based on current resource constraints. i ,T i M i These resource quotas need to be determined through optimized calculations.

[0192] Coupling Coordination: System-level optimization is responsible for handling the coupling relationships between subsystems, ensuring that each subsystem can work in coordination after functional refactoring. This process is achieved by minimizing the coupling constraint function g(z). ij To achieve this.

[0193] Subsystem-level optimization

[0194] Independent optimization: Within the framework of system-level optimization, subject-level optimization operates within its respective resource constraints C. i ,T i M i Next, independently optimize the design variable x for each module. i And calculate the subject-level objective function. The optimal value.

[0195] Feedback mechanism: The results of discipline-level optimization are returned as feedback information to system-level optimization in order to adjust the global design variable x and coupling relationship. The feedback process updates the coupling variable z. kl accomplish.

[0196] Module-level optimization: Within the framework of subsystem-level optimization, module-level optimization specifically addresses the design variables of each functional module, such as material selection and structural parameters, to ensure that the module performs optimally under the subsystem-level objectives.

[0197] Iterative optimization: Coordination between the system-level and discipline-level objectives is achieved through multiple rounds of iteration until the system-level objective function Z is reached. sys (X) converges to the optimal solution. In each iteration, the system-level optimization updates the global design variables, and the subject-level optimization updates the module design variables.

[0198] (3) Output:

[0199] Refactoring scheme X′: The final output includes the optimized configuration of each subsystem module.

[0200] System performance evaluation E(X′): The performance evaluation result E(X′) of the reconstructed system is analyzed by calculating the comprehensive performance index Z(X′,d) to ensure that the system achieves optimal performance under the new task requirements.

[0201] Resource Usage Report: A detailed resource usage report R(X′) includes resource consumption and remaining resource allocation for each subsystem and module. This report helps the design team further optimize resource management and provides a reference for future refactoring tasks.

[0202] This invention proposes a solution scheme based on a genetic algorithm for the multi-level optimization design problem of complex electromechanical systems, and elaborates on its solution steps in detail. Through hierarchical optimization strategies and collaborative optimization methods, we can effectively address the multidisciplinary coupling and resource constraints in complex electromechanical systems.

[0203] This application can also utilize linear programming or dynamic programming instead of multi-objective genetic algorithms for module combinatorial optimization. While these methods may be more efficient in handling simple systems, they are prone to getting trapped in local optima in complex, multidisciplinary systems, making it difficult to reach the global optimum.

[0204] This application can also employ a hierarchical optimization strategy, breaking down the multidisciplinary system into several subsystems for separate optimization before overall integration. While this approach is feasible in certain scenarios, its optimization effect is often inferior to the overall optimization method.

[0205] The embodiments of this application have the following effects:

[0206] Improving resource utilization efficiency: This invention introduces a resource optimization-based modeling method, enabling effective analysis and allocation of limited resources during the system design phase, maximizing resource utilization. Compared to existing technologies, this method reduces resource waste, lowers the overall system cost, and thus improves the system's economic efficiency.

[0207] Enhancing overall system performance: Employing optimization tools such as multi-objective genetic algorithms, this invention can achieve global optimization of system performance in complex multidisciplinary coupling relationships. Compared with existing technologies, this invention not only excels in single performance indicators but also achieves optimal balance among multiple performance indicators, meeting diverse design requirements and thus comprehensively improving the overall system performance.

[0208] Enhancing System Flexibility and Adaptability: This invention provides a flexible modular combination strategy, enabling the system to better cope with changing requirements under different task and environmental conditions. Compared with existing technologies, the system can quickly respond to dynamic changes in design requirements, ensuring efficient operation in various environments and improving the system's adaptability and scalability.

[0209] Reduced Design and Development Costs: By implementing global optimization during the system design phase, this invention effectively reduces potential resource allocation issues or functional module conflicts during later development, thereby lowering overall design and development costs and improving system development efficiency. Compared to existing methods, this invention not only reduces rework and adjustments during development but also shortens product time-to-market.

[0210] This invention also provides an electromechanical system functional module construction device, comprising:

[0211] The first processing module is used to obtain the requirement-function-structure mapping relationship of the electromechanical system;

[0212] The second processing module is used to divide the electromechanical system into functional modules based on the mapping relationship, specifically as follows:

[0213] Based on the requirement-function-structure mapping relationship, a requirement-function matrix and a function-structure mapping matrix are constructed to determine the degree of correlation of structural components in terms of functional attributes and the correlation of structural attributes, and to establish the comprehensive correlation degree of structural components.

[0214] The indicators for determining module partitioning include: the cohesion of functional modules and the coupling between functional modules. An objective function is established based on the criteria of high clustering degree and low coupling degree.

[0215] Based on the objective function, the module grouping pattern is identified, and the functional module division is formed.

[0216] This invention also provides an electromechanical system functional module construction device, including: a transceiver, a processor, a memory, and a program or instructions stored in the memory and executable on the processor; when the processor executes the program or instructions, it implements the steps in the electromechanical system functional module construction method as described above.

[0217] This invention also provides a readable storage medium storing a program or instructions thereon, which, when executed by a processor, implement the steps in the electromechanical system functional module construction method as described above.

[0218] The above description represents the preferred embodiments of the present invention. It should be noted that those skilled in the art can make various improvements and modifications without departing from the principles of the present invention, and these improvements and modifications should also be considered within the scope of protection of the present invention.

Claims

1. A method of building a functional module of an electromechanical system, characterized by, The specific process is: Step one, obtaining the requirement-function-structure mapping relationship of the electromechanical system; Step two, based on the mapping relationship, performing electromechanical system function module division; the specific process is: S201, based on the requirement-function-structure mapping relationship, constructing a requirement-function matrix and a function-structure mapping matrix, determining the correlation degree of the structure components in the functional attribute and the correlation relationship in the structure attribute, and establishing the comprehensive correlation degree of the structure components; S202, determining the module division indicators including the cohesion degree of the function modules and the coupling degree between the function modules, and establishing a target function based on the high clustering degree and low coupling degree criteria; S203, based on the target function, identifying the module grouping mode, and forming the function module division; It also includes a function rapid reconstruction step based on the collaborative optimization method CO. Before using CO to realize function reconstruction, basic assumptions are made in terms of modular design, resource constraints, rapid reconstruction, and collaborative optimization. Among them, The modular design assumption is to divide the overall system into multiple levels based on the modular concept, including the top-level system, the subsystem level, and the module level; Source constraints assumptions: assume that the budget , time , materials and manufacturing processes are limited and known throughout the reconstruction process; Let the total budget of the system be , the time limit be , the material limit be , and the constraint condition be represented as: wherein, budget, time and material consumption of the i-th subsystem, respectively;​ The rapid reconstruction assumption is that rapid reconstruction not only involves performance optimization within the module, but also requires ensuring the coordinated operation of various functional subsystems at the system level; Co-optimization: system-level optimization objective function and subsystem-level optimization objectives , discipline-level optimization objective function coupled through the following relationship: wherein, and are weight coefficients representing trade-off between system level objectives and discipline objectives, represents a constraint function of coupling relationship between subsystem k and subsystem l, represents a constraint function of coupling relationship between module i and module j, and are penalty coefficients of coupling relationship; The constraint conditions for module-level optimization are: 。 2. The method of claim 1, wherein the MEMS building block is a MEMS building block of claim 1. The function attribute is the similarity and correlation degree of two structure components in functional composition and functional relationship; the structure attribute is the correlation relationship of two structure components in multiple disciplines.

3. The method of claim 2, wherein the MEMS building block is a MEMS building block of claim 1. The comprehensive correlation degree is: wherein, and is a set weight parameter, is a functional correlation, is a correlation on a structural attribute The cohesion degree calculation is: ， In the formula is the size of the set of functional modules is the size of the set of functional modules is the structural component in the functional module is the comprehensive relevance of the structural component is the comprehensive relevance of the structural component​ 4. The method of claim 3, wherein the MEMS building block is a MEMS resonator. Two functional modules The coupling degree calculation formula between them is: 。 5. The method of claim 1, wherein the MEMS building block is a MEMS resonator. The step S203 is realized by using the fuzzy C-means clustering method, and the specific process is: First, use the clustering method to group the function components, and group the components with similar functions or mutual dependence in structure into the same module to form a preliminary module division scheme set; Second, evaluate and select the grouping based on the target function; Finally, the final module division scheme is determined by evaluating the function integrity, independence, and cost factors of each module.

6. The method of claim 1, wherein the MEMS building block is a MEMS resonator. The step S203 realizes function module division by using a genetic algorithm, specifically: The constraint conditions include: (1) complete coverage constraint: ensure that each component is assigned to exactly one module, (2) non-empty module constraint: ensure that each module contains at least one component; The target function is set as: ， balance parameters; Based on the constraint conditions and the set target function, the genetic algorithm is used to realize function module division.

7. An electromechanical system functional module building device for performing the method of any one of claims 1-6, wherein, It includes: A first processing module for obtaining the requirement-function-structure mapping relationship of the electromechanical system; A second processing module for performing electromechanical system function module division based on the mapping relationship, specifically: Based on the requirement-function-structure mapping relationship, a requirement-function matrix and a function-structure mapping matrix are constructed to determine the correlation degree of the structure components in the functional attribute and the correlation relationship in the structure attribute, and to establish the comprehensive correlation degree of the structure components; Determine the module division indicators including the cohesion degree of the function modules and the coupling degree between the function modules, and establish a target function based on the high clustering degree and low coupling degree criteria; Based on the target function, identify the module grouping mode, and form the function module division.

8. An electromechanical systems device, comprising: It includes: a transceiver, a processor, a memory, and a program or instructions stored on the memory and executable on the processor; the processor implements the steps in the method for constructing a functional module of an electromechanical system according to any one of claims 1-6 when executing the program or instructions.

9. A readable storage medium, characterized by, a program or instructions stored thereon, the program or instructions being executed by a processor to implement the steps in the method for constructing a functional module of an electromechanical system according to any one of claims 1-6.

Images