An architecture model complexity quantification analysis method for engineering complex systems

Abstract

The application belongs to the field of model-based system engineering application and practice, and relates to a method for quantitatively analyzing complexity of an architecture model of an engineering complex system. A plurality of alternative system architecture models are designed in advance, and then an architecture model to be evaluated is selected. The system architecture model is abstracted according to graph theory to obtain a graph representation of the architecture model, and the complexity of each vertex in the graph is determined (1). Then, the graph representation of the architecture model is converted into a matrix form of the architecture model, and the complexity of each connection relationship in the matrix is determined (2). Then, the topological complexity of the matrix is determined (3). Then, the complexity of the architecture model is calculated according to (1), (2) and (3), and the complexity values of all the alternative architecture models are calculated until the complexity values of different architecture models are compared to evaluate, compare and optimize the architecture model of the system or product. The selected architecture model has the lowest complexity and the lowest implementation difficulty, and the calculation method is convenient for engineering implementation.

Description

A method for quantitative analysis of the complexity of architectural models for complex engineering systems Technical Field

[0001] This application belongs to the field of model-based systems engineering applications and practices, and specifically relates to a method for quantitative analysis of the complexity of architectural models for complex engineering systems. Background Technology

[0002] As performance requirements increase and constraints on general quality characteristics (robustness, reliability, maintainability, etc.) become more stringent, engineering systems or products are becoming increasingly complex. This makes their development more challenging and often leads to delays or cost overruns. Excessive cost overruns or longer-than-expected delays can cause project failures, which can largely be attributed to the inability to quantify and manage the complexity of the designed systems or products.

[0003] Therefore, the quantification, analysis, and optimization management of complexity are becoming increasingly important in system and product design. A core issue in applying the MBSE design methodology to the design of complex systems or products is the system architecture design, which describes the dependencies and topology of components or subsystems within the system. The internal architecture of a complex system can be described as a complex network or graph, where components or subsystems are interconnected as independent nodes, and these connections represent interactions between components or subsystems. The system's functionality and efficiency are influenced by its architecture. In the design of any large-scale engineering system, adopting a simple, robust, and efficient architecture is crucial. Typically, as the functionality of the designed system increases, its architecture also becomes more complex.

[0004] Although there is some theoretical research on system complexity analysis in the MBSE field, there is no clear and applicable method for measuring the complexity of a system or product in engineering practice. Summary of the Invention

[0005] The purpose of this application is to provide a method for quantitative analysis of the complexity of architectural models for complex engineering systems, in order to solve the problem that it is difficult to effectively measure the complexity of systems or products in the prior art.

[0006] The technical solution of this application is: a method for quantitative analysis of the complexity of architectural models of complex engineering systems, including:

[0007] The system or product architecture model is designed to obtain x alternative system architecture models, and the architecture model is represented by M. x Indicated, m represents the architectural model M. x The number of connections in the middle;

[0008] Select an architecture model M that needs to be evaluated. x Transform the graph into a graph, denoted by G, and calculate the number of vertices n in graph G, where n represents the architectural model M. x The number of submodules;

[0009] Transform the graph-based architecture model into a matrix form to obtain matrix A. Count the number of non-zero elements in matrix A, denoted by m.

[0010] The computational complexity C for each vertex in graph G i Then the complexity of n vertices is C i The complexity of summing the modules is C. sum ,Right now

[0011] The complexity L of calculating each connection in matrix A j Then the complexity of m connections is L j The complexity of summing up the connections between modules is L. connect ,Right now

[0012] Calculate the matrix energy T of matrix A E Then determine the complexity T of the topology. topology ==T E / n;

[0013] Computational architecture model M x The complexity F sys The complexity F sys Including parameter one, "module complexity C" sum "Parameter 2: Complexity L of inter-module connections" connect The complexity T of the topology and parameter three. topology ", i.e. F sys =C sum +L connect *T topology ;

[0014] Through complexity F sys Determine the alternative architecture model M x The complexity calculation results;

[0015] Select the next architecture model M to be evaluated. x To obtain the architecture model M x The complexity F sys Determine whether all candidate models have been analyzed. If so, then evaluate all architecture models M that need to be evaluated. x The complexity analysis was completed, and then the architecture models M that had been evaluated were compared. x The complexity F sys Numerical values, choosing complexity Fsys The model with the lowest numerical value is selected as the final architecture model M.

[0016] M x .

[0017] Preferably, the graph G can be selected as a directed graph or an undirected graph.

[0018] Preferably, matrix A can be selected as an adjacency matrix or an affinity matrix.

[0019] Preferably, the complexity C i and complexity L j The calculation method can be determined by either "expert scoring" or "maturity level ratio calculation".

[0020] Preferably, the matrix energy T E The matrix energy T is greater than zero. E It is obtained by matrix singular value decomposition.

[0021] Preferably, at least two alternative architecture models are designed and selected.

[0022] This application presents a method for quantitatively analyzing the complexity of architectural models for complex engineering systems. This method involves pre-designing multiple alternative system architecture models, selecting one for evaluation, abstracting the architecture model using graph theory to obtain its graph representation, and determining the complexity of each vertex in the graph (1). The graph representation is then transformed into a matrix form, and the complexity of each connection in the matrix is ​​determined (2). The topological complexity of the matrix is ​​then determined (3). The complexity of the architecture model is calculated based on (1), (2), and (3), and this process is repeated until the complexity values ​​of all alternative architecture models are calculated. The complexity values ​​of different architecture models are then compared to evaluate, compare, and optimize the architecture model of the system or product. The selected architecture model has the lowest complexity and is the easiest to implement. This calculation method is easy to implement in engineering, and the result is unique. Furthermore, the complexity calculation results can provide an objective basis for the selection and design of system and product architecture models. Attached Figure Description

[0023] To more clearly illustrate the technical solutions provided in this application, the accompanying drawings will be briefly described below. Obviously, the drawings described below are merely some embodiments of this application.

[0024] Figure 1 is a schematic diagram of the overall process of this application;

[0025] Figure 2 on the left shows the alternative architecture M1 of this application, and Figure 2 on the right shows the alternative architecture M2 of this application;

[0026] Figure 3 (left) is an undirected graph of alternative architecture M1 in this application; Figure 3 (right) is an undirected graph of alternative architecture M2 in this application.

[0027] Figure 4 (left) shows the adjacency matrix of alternative architecture M1 in this application; Figure 4 (right) shows the adjacency matrix of alternative architecture M2 in this application.

[0028] Figure 5 is a schematic diagram of the expert scoring method for this application. Detailed Implementation

[0029] To make the objectives, technical solutions, and advantages of this application clearer, the technical solutions in the embodiments of this application will be described in more detail below with reference to the accompanying drawings.

[0030] A method for quantitative analysis of the architectural complexity of complex engineering systems is provided. This method can quantitatively evaluate and analyze the architectural complexity of the designed system or product and can be applied to the architectural design and analysis phases of any complex engineering project or product.

[0031] As shown in Figure 1, the steps include:

[0032] Step S100: Design the system or product architecture model to obtain x alternative system architecture models. The architecture model is represented by M. x express;

[0033] In order to select the optimal architecture model and provide at least two alternative system architecture models, i.e., x = 2, 3..., whether using object-oriented systems engineering (OOSEM) theory or harmonious systems engineering (Harmony-SE) theory, establishing an architecture model is a necessary prerequisite activity.

[0034] Step S200: Select an architecture model M to be evaluated. x Transform the graph into a graph, denoted by G, and calculate the number of vertices n in graph G, where n represents the architectural model M. x The number of submodules;

[0035] Preferably, graph G can be a directed graph or an undirected graph, where vertices represent sub-modules in the graph, as shown in Figure 2, and the lines connecting sub-modules represent the connection relationships between sub-modules.

[0036] Step S300: Convert the graph-form architecture model into a matrix form to obtain matrix A. Count the number of non-zero elements in matrix A, denoted by m.

[0037] Preferably, matrix A can be selected as an adjacency matrix or an association matrix, which can be selected according to actual needs. Non-zero elements in the statistical matrix represent connections between sub-modules.

[0038] Step S400: Calculate the complexity C for each vertex in graph G. i The value of i ranges from [1-n], and then the complexity of n vertices is C. i The complexity of summing the modules is C. sum ,Right now

[0039] C i The calculation method can adopt either "expert scoring" or "maturity level ratio calculation"; C i The suggested value range is [0-10].

[0040] Step S500: Calculate the complexity L of each connection relationship in matrix A. j The value of j ranges from [1-m]; then the complexity of m connections is L. j The complexity of summing up the connections between modules is L. connect ,Right now

[0041] L j The calculation method can adopt either "expert scoring" or "maturity level ratio calculation". j The suggested value range is [0-1].

[0042] The formula for calculating the maturity level ratio is:

[0043]

[0044] Where k is the complexity system, with a value range of [0-10], TRL max TRL represents the highest maturity level. min TRL represents the highest maturity level. c This represents the actual maturity level of submodule C.

[0045] Step S600: Calculate the matrix energy T of matrix A. E Then determine the complexity T of the topology. topology ==T E / n, where n is the number of submodules in graph G;

[0046] Preferably, the matrix energy T E It can be obtained by using matrix singular value decomposition.

[0047] Step S700, Computational Architecture Model M x The complexity F sys The complexity F sys Including parameter one, "module complexity C" sum "Parameter 2: Complexity L of inter-module connections" connectThe complexity T of the topology and parameter three. topology ", i.e. F sys =C sum +L connect *T topology Each F sys Corresponding to an M x ;

[0048] Step S800: Select the next architecture model M to be evaluated. x Repeat steps S200-S700 to obtain the architecture model M. x The complexity F sys Determine whether all candidate models have been analyzed. If so, then evaluate all architecture models M that need to be evaluated. x Analysis complete;

[0049] Step S900, with a complexity F sys Evaluate the alternative architecture model M x :F sys The smaller the value, the better the system's architecture model M. x The lower the complexity, the easier the implementation, and the lower the priority should be; F sys The larger the value, the higher the system architecture model M. x The higher the complexity, the more difficult the implementation, and therefore it should be chosen last; complexity F is selected. sys The model with the lowest numerical value is selected as the final architecture model M. x The complexity F sys At the very least, the selected architecture model M x It has the lowest implementation difficulty.

[0050] This application pre-designs multiple alternative system architecture models, then selects one architecture model to be evaluated, abstracts the system architecture model according to graph theory to obtain a graph representation of the architecture model, and determines the complexity of each vertex (representing a sub-module in the architecture model) (1); then the graph representation of the architecture model is transformed into a matrix form of the architecture model, and the complexity of each connection relationship in the matrix (representing the connection relationship between sub-modules in the architecture model) is determined (2); then the topological complexity of the matrix is ​​determined (3); then the complexity of the architecture model is calculated based on (1), (2), and (3), until the complexity values ​​of all alternative architecture models are calculated, and the complexity values ​​of different architecture models are compared to evaluate, compare, and optimize the architecture model of the system or product. The selected architecture model has the lowest complexity and the lowest implementation difficulty. This calculation method is easy to implement in engineering, and the result is unique; at the same time, the complexity calculation result can provide an objective basis for the selection and design of the architecture model of the system and product.

[0051] This method is particularly suitable for the evaluation and optimization of complex systems and product architecture models.

[0052] The following is an illustration using a specific example;

[0053] As one specific implementation method, a method for quantitative analysis of the architectural model complexity of complex engineering systems includes:

[0054] Step S100: Design two alternative architectures, M1 (Figure 2 left) and M2 (Figure 2 right);

[0055] Step S200: Convert the alternative architecture M1 of the system into an undirected graph (Figure 3 left), and count the number of vertices of M1 as 5;

[0056] Step S300: Transform the undirected graph of M1 into an adjacency matrix (Figure 4 left), and count the number of non-zero elements in M1 to be 10.

[0057] In step S400, according to the maturity level ratio calculation method, the complexities of the five sub-modules in M1 are calculated as follows: 2, 2, 2, 1, 1. The module complexity C of M1 is then calculated. sum =8;

[0058] In step S500, following the expert scoring method as shown in Figure 5, the complexities of the 10 connections in M1 are calculated as follows: 0.5, 0.5, 0.5, 1, 1, 0.5, 1, 1. Finally, the complexity L of the inter-module connections in M1 is obtained. connect =6;

[0059] Step S600: Calculate the topological complexity T of M1. topology =1.12;

[0060] Step S700: Calculate the complexity F of the alternative architecture model M1. sys =15.2;

[0061] In step S800, repeat steps S200-S800 to convert the system candidate architecture M2 into an undirected graph (Figure 3, right), and count the number of vertices in M2 as 5; convert the undirected graph of M2 into an adjacency matrix (Figure 4, right), and count the number of non-zero elements in M1 as 8; and calculate the complexity F of the candidate architecture model M2 in sequence according to steps S500-S800. sys =18.28.

[0062] Step S900: Compare the complexity values ​​F of the alternative architecture models M1 and M2. sys The complexity F of architecture model M1 sys Given the smaller size, alternative architecture model M1 should be selected as the system architecture.

[0063] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the technical scope disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

Claims

1. A method for quantitative analysis of the complexity of architectural models for complex engineering systems, characterized in that, include: The system or product architecture model is designed to obtain x alternative system architecture models. The architecture model is used... Indicate; select an architecture model that needs to be evaluated. Transform the graph into a graph, denoted by G, and calculate the number of vertices n in graph G, where n represents the architectural model M. x The number of submodules; transform the graph-based architecture model into a matrix form, obtaining matrix A, and count the number of non-zero elements in matrix A, denoted by m, where m represents the architecture model M. x The number of connections in graph G; the computational complexity of calculating the number of vertices in graph G. Then the complexity of n vertices The complexity of the module is obtained by summing. ,Right now The complexity of calculating each connection in matrix A Then the complexity of m connections The complexity of summing up the connections between modules ,Right now ; Calculate the matrix energy of matrix A Then determine the complexity of the topology. = / n; Computational architecture model complexity complexity Including the complexity of the "parameter one" module The complexity of connections between modules "parameter 2" The complexity of the topology with parameter three ",Right now Through complexity Identify alternative architecture models The complexity calculation results are used to select the next architecture model to be evaluated. To obtain the architecture model complexity Determine if all candidate models have been analyzed; if so, then evaluate all architecture models that need to be evaluated. The complexity analysis was completed, and then the architecture models that had been evaluated were compared. complexity Numerical value, choosing complexity The model with the lowest numerical value is selected as the final architecture model. The matrix A is an adjacency matrix or an incidence matrix; the matrix energy The matrix energy is greater than zero. It is obtained by matrix singular value decomposition.

2. The method for quantitative analysis of architectural model complexity of complex engineering systems as described in claim 1, characterized in that: The graph G can be a directed graph or an undirected graph.

3. The method for quantitative analysis of architectural model complexity of complex engineering systems as described in claim 1, characterized in that: The complexity and complexity It is determined by "expert scoring" or "maturity level ratio calculation".

4. The method for quantitative analysis of architectural model complexity of complex engineering systems as described in claim 1, characterized in that: Design and select at least two alternative architecture models.

Images