Method and system for generating capability boundary of architecture utility based on causal topology

Abstract

This invention relates to a method and system for generating utility capability boundaries for system architectures based on causal topology, comprising: S1. Receiving and parsing the system architecture; S2. Constructing a causal topology graph; S3. Generating a total utility function describing the overall capability of the system architecture; S4. Selecting several variables from the total utility function and setting the variation range of each variable, performing regularized variation deduction on the selected variables, thereby generating utility capability boundaries; S5. Repeating steps S1 to S4 for different system architecture schemes to generate utility capability boundaries corresponding to different system architecture schemes, performing visualization overlay, and then performing attribution analysis. Advantages: By formally modeling different causal topological structures in the system architecture and introducing a utility capability generation and boundary construction mechanism, the system characterizes the reachability boundary characteristics of system capabilities under different structural paradigms, revealing the influence mechanism of structural patterns on the formation of capability boundaries.

Description

Technical Field

[0001] This invention relates to the field of complex system architecture design and evaluation, and specifically to a method and system for generating system architecture utility capability boundaries based on causal topology. Background Technology

[0002] With the development of complex engineering systems and system-wide tasks, system architecture analysis has gradually shifted from evaluating the performance of single equipment or systems to analyzing system capabilities across multiple factors, levels, and interactions. Existing research typically understands system capability as the overall function or task effect that a system can achieve under given resource configuration and operational conditions, and its analytical methods mainly focus on the following two categories: (1) System architecture evaluation method based on capability index aggregation This type of method typically includes the following steps: ① Constructing a set of system capability indicators, which reflect the system's performance in terms of functionality, performance, efficiency, or robustness; ② Calculating the values ​​of each indicator for different architectural schemes under given operating conditions; ③ Aggregating the multidimensional indicators using methods such as weighted summation, utility functions, or analytic hierarchy process (AHP) to obtain a comprehensive evaluation result of the system capability; ④ Ranking or selecting different architectural schemes based on the comprehensive evaluation result. This type of method focuses on the capability level of a given architectural scheme under given conditions.

[0003] (2) System architecture capability analysis method based on simulation or optimization This type of method typically includes: ① establishing a system architecture model to describe the components and their connections within the system; ② simulating or performing optimization calculations on the system under set operating scenarios or task conditions; ③ obtaining the system's performance capabilities under specific parameters or resource configurations based on the simulation or optimization results; and ④ comparing and analyzing different architecture schemes through multiple simulations or optimizations. This type of method can reflect the architecture's performance capabilities under specific conditions, but it usually depends on specific scenarios or parameter settings.

[0004] The above analytical method has the following shortcomings: 1) Existing methods typically evaluate system capabilities under fixed parameters or scenarios, treating capability as a specific numerical value or score result, without systematically analyzing the upper and lower limits or range of change of capability under varying key variables. This makes it impossible to answer capability boundary questions such as "under the current system architecture, what is the highest level that capability can reach, and to what extent will it degrade at the lowest level?", and the potential space of system capability lacks clear characterization.

[0005] 2) Existing methods often treat the system architecture as a fixed input, focusing on the impact of resource allocation or operational strategies on capability outcomes, without explicitly modeling the causal constraints introduced by different structural paradigms (such as serial, collaborative, and redundant). Therefore, although it is possible to compare the capabilities of different architectural solutions in a certain scenario, it is difficult to reveal how the structural pattern itself limits or expands the reachable boundary of system capabilities, and the mechanism by which structural design affects capability formation is opaque.

[0006] 3) There is a lack of unified measurement and expression forms among capability indicators, simulation results, and optimization outputs, making it difficult to systematically compare the capability results of different architecture schemes on the same scale. Therefore, even if capability differences between different architecture schemes are observed, it is difficult to clearly attribute changes in capability boundaries to specific causal topological differences, and can only remain at the level of empirical or qualitative analysis.

[0007] Based on this, the present invention is proposed. Summary of the Invention

[0008] The purpose of this invention is to provide a method and system for generating the utility capability boundary of a system architecture based on causal topology. By formally modeling different causal topological structures in the system architecture and introducing a utility capability generation and boundary construction mechanism, the system characterizes the reachability boundary characteristics of the system capability under different structural paradigms and reveals the influence mechanism of structural patterns on the formation of capability boundaries.

[0009] To achieve the above objectives, the technical solution of the present invention is as follows: A method for generating the utility capability boundary of a system architecture based on causal topology includes the following steps: S1. Receive and parse the input system architecture description information, and transform it into an internally processable object model; S2. Construct a causal graph based on the object model, identify and mark the subgraph structures in the causal graph that conform to the predefined topological paradigm according to the predefined judgment rules, and form a causal topological graph represented by a formal language; S3. Generate a total utility function describing the overall capabilities of the system architecture based on the causal topology graph; S4. Select several variables from the total utility function and set the variation range of each variable. Under the premise of keeping the causal topology unchanged, perform regularized change deduction on the selected variables to obtain the upper and lower bounds of the system architecture utility capability under the corresponding causal topology, thereby generating the utility capability boundary. S5. Repeat steps S1 to S4 for different system architecture schemes to generate the utility capability boundaries corresponding to each system architecture scheme. Visualize and overlay the utility capability boundaries of different system architecture schemes and perform attribution analysis. Use the attribution analysis results as the basis for system architecture design adjustment, causal topology optimization, or resource allocation improvement.

[0010] Furthermore, the system architecture description information can be generated using structured text or through graphical modeling tools.

[0011] Furthermore, in step S1, the system architecture description information includes at least the system components, the connection relationships between the components, and the task objectives and capability indicators; Each unit in the system must be associated with its type identifier and performance parameters; The inter-unit connection relationships are used to describe data flow, control flow, or resource dependencies; The task objectives and capability indicators are used to clarify the tasks that the system architecture needs to complete and to list the capability indicators to be evaluated.

[0012] Furthermore, step S2 specifically includes the following processes: S21. Abstract the performance parameters, task objectives, and capability indicators of each system component into causal variable nodes; S22. Based on the connection relationship between the units, define directed edges between causal variable nodes to form a causal graph; S23. Analyze the formed causal graph, and based on the necessity relationship, sufficiency relationship and resource constraint characteristics between causal variable nodes, identify the subgraph structure that conforms to the predefined paradigm according to the predefined judgment rules, and label the identified subgraph structure with a topological paradigm label; The predefined determination rules include at least the following: The unique necessity rule states that if, in a subgraph structure of a causal graph, causal variable node A is the only direct predecessor of causal variable node B, and there are no other alternative paths that can make causal variable node B valid, and causal variable node B is the only direct predecessor of causal variable node C, then the subgraph structure is determined to satisfy the unique necessity relation and is identified as a serial topology paradigm. At the same time, it is necessary to determine the rule that if the establishment of causal variable node A requires the simultaneous satisfaction of inputs from two or more predecessor causal variable nodes, and the absence of any predecessor causal variable node will cause causal variable node A to fail to be established, then the subgraph structure is determined to satisfy the simultaneous necessity relationship and is identified as a cooperative topology paradigm. Any sufficient determination rule, if the establishment of causal variable node A can be independently triggered by any one of the multiple predecessor causal variable nodes, and each predecessor causal variable node is functionally equivalent, then the subgraph structure is determined to satisfy any sufficient relation and is identified as an alternative redundant topology paradigm. S24. Represent the causal topology graph with the topological paradigm label in a formal language and store it in the paradigm library.

[0013] Furthermore, step S3 includes the following specific processes: S31. Define a standardized utility function for each causal variable node, mapping the physical parameters to utility values ​​in the interval [0, 1]. S32. Based on the causal topology graph formed in step S2, select the corresponding utility combination operator according to the paradigm type, and perform bottom-up utility aggregation calculation; For the serial topology paradigm, a bottleneck-type combinatorial operator is adopted, that is, the system utility within the paradigm is defined as the minimum value or equivalent product of the utility of each causal variable node; For the cooperative topology paradigm, a joint contribution combinatorial operator is adopted, that is, multiplication, weighted summation or other fusion operations are performed on the utility of parallel branches; For alternative redundant topology paradigms, a selective combination operator is used, that is, the path with the greatest utility contribution is selected from the set of available paths, or the expected utility is calculated when considering the availability probability. S33. When there are multiple levels or nested causal topology graphs in the system architecture, the local utility calculation and aggregation are completed layer by layer in a bottom-up recursive manner until the total utility function describing the overall capability of the system architecture is generated.

[0014] Furthermore, step S4 includes the following specific processes: S41. Select several key variables from the independent variables of the total utility function and set a range of variation for each key variable; Key variables should meet the following rules: conduct sensitivity analysis on each causal variable in the total utility function, and select variables whose weights on the total utility value of the system exceed a preset threshold; The range of variation is determined based on the performance parameters in step S1 or user-defined input. S42. While keeping the causal topology unchanged, perform regularized variation deductions on key variables to analyze the achievable range of the system's utility capacity under different parameter combinations, including: Upper bound calculation involves finding a set of parameter values ​​within the range of variation of the key variables that maximizes the value of the total utility function. Lower bound calculation involves finding a set of parameter values ​​within the range of variation of the key variables that minimizes the value of the total utility function. Boundary generation: For a single key variable, fix other variables, traverse the entire interval of the key variable, record the upper and lower bound values ​​corresponding to each point, and form two curves. The area between the two curves is the utility capability boundary. For multiple key variables, generate a multi-dimensional boundary hypersurface or display a two-dimensional cross-section through projection. S43. Combine the calculated upper and lower bounds of utility capacity with the corresponding key variable parameter values ​​and output them in a data file or graphical format.

[0015] Furthermore, in step S5, the attribution analysis includes: Qualitative analysis is used to identify the system architecture scheme that covers the widest range of utility capability boundaries. Quantitative attribution involves selecting parameter points, comparing the upper and lower bounds of the utility capacity of different schemes at those parameter points, analyzing the reasons for the differences, and directly linking these reasons to the differences in topological paradigms.

[0016] A causal topology-based system for generating architecture utility capability boundaries for the method, characterized in that it comprises: The data parsing module is used to receive and parse the input architecture description information and transform it into an internally processable object model; The causal topology paradigm construction module is used to construct a causal graph based on the object model, identify and mark the subgraph structures in the causal graph that conform to the predefined topology paradigm according to predefined judgment rules, and form a causal topology graph represented in a formal language. The utility capacity generation module is used to generate a total utility function describing the overall capabilities of the system architecture based on the causal topology graph. The utility capability boundary construction module is used to select several variables from the total utility function and set the variation range of each variable. Under the premise of keeping the causal topology unchanged, the selected variables are subjected to regularized change deduction to obtain the upper and lower bounds of the system's utility capability under the corresponding causal topology, thereby generating the utility capability boundary. The topology impact attribution analysis module is used to obtain several utility capability boundaries generated by several system architecture schemes, visualize and overlay the utility capability boundaries of different system architecture schemes and perform attribution analysis, and use the attribution analysis results as the basis for system architecture design adjustment, causal topology diagram optimization or resource allocation improvement.

[0017] The advantages of this invention are: 1. Existing technologies typically perform single-point simulations only for specific operating conditions or nominal parameters, reflecting only the instantaneous capability of the system under ideal or specific conditions, and failing to reveal the full performance picture under parameter fluctuations. This invention expands the evaluation dimension from a single numerical point to a capability envelope region in a multi-dimensional parameter space by constructing an upper and lower bound for the achievable utility capability. This mechanism not only systematically analyzes the potential space of the system's capability but also intuitively demonstrates the robustness and limiting performance of the system under different perturbations, enabling designers to fully understand "what the system can do" and "the range in which it operates stably," effectively avoiding blind optimism or conservative estimations caused by single-point evaluations.

[0018] 2. Traditional evaluation methods often neglect internal connectivity logic and employ simple weighted summations, thus masking the impact of structural differences on performance. This invention innovatively introduces causal topological paradigms such as series, synergy, and redundancy into the analysis process, establishing a mathematical mapping relationship between structural form and capability boundaries. By identifying and labeling these topological paradigms, this invention can accurately calculate the nonlinear effects of different paradigms on the total utility function, making it analyzable and comparable how structural design constrains or extends capability boundaries.

[0019] 3. Existing technologies, when comparing different solutions, often only conclude that "solution A is superior to solution B," but fail to explain the root cause of the differences. This invention, by superimposing and comparing utility capability boundaries generated under different causal topological paradigms and conducting attribution analysis, can directly link the expansion or contraction of capability boundaries to specific structural pattern changes. This provides a clear direction for iterative optimization of the system architecture, thereby significantly improving the scientific nature and decision-making efficiency of architecture design and shortening the R&D iteration cycle. Attached Figure Description

[0020] Figure 1 This is a schematic diagram of the architecture of the system for generating the utility capability boundary based on causal topology in the embodiment. Figure 2 This is a flowchart illustrating the method for generating the system architecture utility capability boundary based on causal topology in this embodiment. Figure 3 This is a schematic diagram comparing the capability boundary ranges of two system architecture schemes in an example application case; Figure 4 This is a schematic diagram of the capability boundary curves of two system architecture schemes in the example application case; Figure 5 This is a schematic diagram illustrating the sensitivity analysis of environmental parameters for two system architecture schemes in the example application case; Figure 6 This is a schematic diagram comparing the capability improvements of two system architecture solutions in an example application case. Detailed Implementation

[0021] The embodiments of the present invention will be further described below. It should be understood that the described embodiments are merely some, not all, of the embodiments of this application. All other embodiments obtained by those skilled in the art based on the embodiments of this application without inventive effort are within the scope of protection of this application. The terminology used in the embodiments of this application is for the purpose of describing specific embodiments only and is not intended to limit the application. The singular forms “a,” “the,” and “the” used in the embodiments of this application and the appended claims are also intended to include the plural forms, unless the context clearly indicates otherwise.

[0022] This embodiment proposes a method for generating the system architecture utility capability boundary based on causal topology, such as... Figure 2 As shown, it includes the following steps: S1. Input and Parsing Architecture: Receives and parses the architecture description information input by the user, transforms it into an internally processable object model, and prepares for subsequent abstract modeling; S2. Constructing a causal topological paradigm: Based on the object model, construct a causal graph. According to predefined judgment rules, identify and mark the subgraph structures in the causal graph that conform to the predefined topological paradigm, forming a causal topological graph represented by a formal language. S3. Generate system utility capability: Generate a total utility function describing the overall capability of the system architecture based on the causal topology graph; S4. Constructing the utility capability boundary: Select several variables from the total utility function and set the variation range of each variable. Under the premise of keeping the causal topology unchanged, perform regularized change deduction on the selected variables to obtain the upper and lower bounds of the system architecture utility capability under the corresponding causal topology, thereby generating the utility capability boundary. S5. Comparative Analysis and Structural Attribution: Repeat steps S1 to S4 for different system architecture schemes to generate the utility capability boundaries corresponding to each system architecture scheme. Visualize and overlay the utility capability boundaries of different system architecture schemes (i.e., draw the utility capability boundaries of different system architecture schemes in the same coordinate system) and perform attribution analysis. Use the results of the attribution analysis as the basis for system architecture design adjustment, causal topology optimization, or resource allocation improvement.

[0023] In step S1 of this embodiment, the system architecture description information can be generated using structured text (such as XML) or through graphical modeling tools, and its content includes at least the system components, the connection relationships between the components, the task objectives and capability indicators.

[0024] The system's constituent units, such as sensors, communication nodes, decision centers, and execution units, each need to be associated with its type identifier and performance parameters (such as detection probability, processing latency, reliability, etc.).

[0025] The inter-unit connectivity is used to describe data flow, control flow, or resource dependencies, such as the output of sensor A serving as the input of decision center B.

[0026] The mission objectives and capability indicators are used to clarify the tasks that the system architecture needs to complete (such as area surveillance and target strike), and to list the capability indicators to be evaluated (such as situation update frequency, target processing throughput, and mission completion success rate).

[0027] In this embodiment, step S2 includes the following process: S21. Abstract the performance parameters, mission objectives, and capability indicators of each system component into causal variable nodes. For example, abstract the radar's "detection probability" into node V_detect, and the communication link's "transmission bandwidth" into node V_bw.

[0028] S22. Based on the inter-unit connection relationships, define directed edges between causal variable nodes to represent influence relationships, forming a causal graph. For example, if V_detect (detection probability) and V_bw (transmission bandwidth) jointly affect V_info_quality (information quality), then draw two directed edges pointing to V_info_quality respectively.

[0029] S23. Analyze the formed causal graph. Based on the necessity relationship, sufficiency relationship and resource constraint characteristics between causal variable nodes, identify the subgraph structure that conforms to the predefined paradigm according to the predefined judgment rules, and label the identified subgraph structure with a topological paradigm. The subgraph structure in this step refers to the local connection pattern with specific topological characteristics extracted from the entire complex system causal graph (large graph), which is composed of some nodes and some edges.

[0030] The predefined determination rules include at least the following: The unique necessity rule states that if, in a subgraph structure of a causal graph, causal variable node A is the only direct predecessor of causal variable node B, and there are no other alternative paths that can make causal variable node B valid, and causal variable node B is the only direct predecessor of causal variable node C, then the subgraph structure is determined to satisfy the unique necessity relation and is identified as a serial topology paradigm. At the same time, it is necessary to determine the rule that if the establishment of causal variable node A requires the simultaneous satisfaction of inputs from two or more predecessor causal variable nodes, and the absence of any predecessor causal variable node will cause causal variable node A to fail to be established, then the subgraph structure is determined to satisfy the simultaneous necessity relationship and is identified as a cooperative topology paradigm. Any sufficient determination rule, if the establishment of causal variable node A can be independently triggered by any one of the multiple predecessor causal variable nodes, and each predecessor causal variable node is functionally equivalent, then the subgraph structure is determined to satisfy any sufficient relation and is identified as an alternative redundant topology paradigm. S24. Represent the causal topology graph with the topological paradigm label in a formal language (such as a set of equations or logical statements) and store it in the paradigm library.

[0031] In this embodiment, step S3 includes the following process: S31. Define a standardized utility function for each causal variable node, mapping the physical parameters to utility values ​​in the interval [0, 1]. The specific form of the utility function can be selected based on task characteristics, engineering experience, or historical data, but all must satisfy monotonicity and normalization constraints to ensure the comparability of utilities for different causal variables. An example is shown below: Example 1: For a time delay T, define the utility function U(T) = exp(-αT), where α is the attenuation coefficient. The smaller T is, the higher the utility.

[0032] Example 2: For “precision P”, define the utility function U(P) = P (assuming P itself is in the interval [0,1].

[0033] S32. Based on the causal topology graph formed in step S2, select the corresponding utility combination operator according to the paradigm type, and perform bottom-up utility aggregation calculation: For the serial topology paradigm, a bottleneck-type combinatorial operator is adopted, that is, the system utility within the paradigm is defined as the minimum value or equivalent product of the utility of each causal variable node, in order to characterize the "weakest link" effect. For the cooperative topology paradigm, a joint contribution combinatorial operator is adopted, that is, multiplying, weighting, or performing other fusion operations on the utility of parallel branches to reflect the cooperative enhancement or mutual constraint relationship. For alternative redundant topology paradigms, a selective combination operator is adopted, that is, the path with the greatest utility contribution is selected from the set of available paths, or the expected utility is calculated when considering the availability probability, so as to reflect the improvement effect of the redundant structure on the stability of capability.

[0034] S33. When there are multiple levels or nested causal topology graphs in the system architecture, the local utility calculation and aggregation are completed layer by layer in a bottom-up recursive manner until the total utility function describing the overall capability of the system architecture is generated.

[0035] In this embodiment, step S4 includes the following process: S41. Select several key variables from the independent variables of the total utility function and set a range of variation for each key variable.

[0036] Key variables should meet the following rules: conduct sensitivity analysis on each causal variable in the total utility function, and select variables whose weights on the total utility value of the system exceed a preset threshold.

[0037] The range of variation is determined based on the performance parameters or user-defined input in step S1.

[0038] S42. While keeping the causal topology unchanged, perform regularized change deduction on key variables to analyze the achievable range of the system's utility capacity under different parameter combinations; The regularized change deduction refers to generating multiple sets of variable value combinations within a set variable change range based on a preset sampling strategy (such as equal-step discretization sampling, endpoint extreme value sampling, or random sampling based on probability distribution); inputting each set of variable value combinations into the total utility function, and performing step-by-step calculations using the utility combination operator defined in the causal topology graph to obtain the distribution set of the total utility value of the system; based on this distribution set, extracting the maximum value of the utility capability as the upper bound and the minimum value as the lower bound, and plotting the continuous boundary surface of the utility capability as a function of key variables, as detailed below: Upper Realm Calculate and find a set of parameter values ​​within the range of variation of the key variables that maximizes the total utility function F: ; The Lower World Calculate and find a set of parameter values ​​within the range of variation of the key variables that minimizes the total utility function F: ; Boundary generation involves fixing other variables for a single key variable, traversing the entire interval of the key variable, recording the upper and lower bound values ​​for each point, forming two curves, and the area between them is the utility capability boundary; for multiple key variables, a multidimensional boundary hypersurface is generated or a two-dimensional cross-section is displayed through projection.

[0039] S43. Combine the calculated upper and lower bounds of utility capacity with the corresponding key variable parameter values ​​and output them in a data file or graphical format.

[0040] In this embodiment, the attribution analysis in step S5 includes: Qualitative analysis identifies the system architecture scheme that covers the widest range of utility capability boundaries, i.e., observe which scheme has a higher and wider overall boundary area. A higher overall boundary area indicates a better capability baseline, while a wider overall boundary area (larger gap between upper and lower boundaries) indicates that the capability is more sensitive to parameter changes. Quantitative attribution involves selecting parameter points, comparing the upper and lower bounds of the utility capacity of different schemes at those parameter points, analyzing the reasons for the differences, and directly linking these reasons to the differences in topological paradigms.

[0041] like Figure 1 As shown, this embodiment also proposes a causal topology-based system for generating architecture utility capability boundaries for the above method, comprising: The data parsing module is used to receive and parse the input architecture description information and transform it into an internally processable object model; The causal topology paradigm construction module is used to construct a causal graph based on the object model, identify and mark the subgraph structures in the causal graph that conform to the predefined topology paradigm according to predefined judgment rules, and form a causal topology graph represented in a formal language. The utility capacity generation module is used to generate a total utility function describing the overall capabilities of the system architecture based on the causal topology graph. The utility capability boundary construction module is used to select several variables from the total utility function and set the variation range of each variable. Under the premise of keeping the causal topology unchanged, the selected variables are subjected to regularized change deduction to obtain the upper and lower bounds of the system's utility capability under the corresponding causal topology, thereby generating the utility capability boundary. The topology impact attribution analysis module is used to obtain several utility capability boundaries generated by several system architecture schemes, visualize and overlay the utility capability boundaries of different system architecture schemes and perform attribution analysis, and use the attribution analysis results as the basis for system architecture design adjustment, causal topology diagram optimization or resource allocation improvement.

[0042] To clearly illustrate the core of this embodiment, a simplified area surveillance task is selected as an application case to compare the two infrastructure solutions: Option 1 (Centralized Serial): 1 radar → 1 dedicated communication link → 1 command center → 1 strike unit; Option 2 (Distributed Redundancy): 2 radars (with identical functions) → 2 communication links (one primary and one backup) → 2 command nodes (hot backup) → 2 strike units (any of which can respond).

[0043] Step S1: Input and parse the system architecture, that is, input the system architecture description information and parse the architecture.

[0044] System components: Radar R: Detection probability P_d (affected by weather factor x1, x1∈[0.5, 1], the smaller the value, the worse the weather); Communication link C: Transmission success rate P_c (affected by interference factor x2, x2∈[0.5, 1]); Command node H: Decision accuracy P_h (assuming constant = 0.95); Strike unit W: Damage probability P_w (assuming constant = 0.9).

[0045] Inter-unit connection relationship: As shown by the arrow above, data / control flow is transmitted sequentially.

[0046] Task objectives and capability indicators: Overall task success rate P_task.

[0047] Step S2: Construct causal topological paradigms, that is, construct and identify causal topologies.

[0048] Analyze the above description information, construct a cause-effect graph, and identify the paradigm.

[0049] Cause-effect graph for Option 1: V_R(P_d) → V_C(P_c) → V_H(P_h) → V_W(P_w) → V_task Paradigm matching result: The entire link is identified as a serial topology paradigm because each node is the unique predecessor of the next node and its function is irreplaceable.

[0050] Scheme 2 Cause-effect graph (taking the perception layer as an example, the communication layer and decision layer are similar): V_R1(P_d) → V_task and V_R2(P_d) → V_task (both point directly to the same task node, indicating that either success can contribute).

[0051] Paradigm matching results: Radars R1 and R2 were identified as a candidate redundant topology paradigm.

[0052] Step S3: Generate the system's utility capability, that is, generate the utility expression.

[0053] The system automatically maps utility combination operators based on the paradigm.

[0054] Serial paradigm operator: The overall utility is the product of the utilities of each node (characterizing the effect of the weakest link).

[0055] Therefore, the total utility expression for Option 1 is: P_task_A = P_d(x1) * P_c(x2) * P_h * P_w.

[0056] Alternative redundancy paradigm operator: The overall utility is the maximum of the utilities of each branch, or the reliability formula for parallel systems considering probability. For simplicity, 1 - (1 - branch utility) is used. n Model. The perception layer utility of Scheme 2 (taking dual radar redundancy as an example) is: P_d_redundant =1 - [1 - P_d(x1)] 2 .

[0057] Therefore, the total utility expression for Scheme 2 (assuming full-link dual redundancy) is: P_task_B=[1-(1-P_d(x1)) 2 ]*[1-(1-P_c(x2)) 2 ]*P_h*[1-(1-P_w) 2 ].

[0058] Step S4: Construct the utility capability boundary, i.e., calculate the utility capability boundary.

[0059] Key variables and range of variation: The key variables affected by the environment were identified as x1 (weather factors) and x2 (disturbance factors), both with a range of [0.5, 1]. Among them, x1=1 indicates the best weather conditions, and x1=0.5 indicates the worst weather conditions; x2=1 indicates no disturbance, and x2=0.5 indicates the strongest disturbance.

[0060] Boundary calculation: Upper bound (optimal case): When x1=1 and x2=1, the calculated upper bound of the task success rate for Scheme 1 is P_task_A_max=0.8550, and the upper bound of the task success rate for Scheme 2 is P_task_B_max=0.9405. Figure 3 As shown, under optimal environmental conditions, the redundant architecture only improves upon the tandem architecture by 8.6%.

[0061] Lower bound (worst-case scenario): When x1=0.5 and x2=0.5, the calculated lower bound of the task success rate for Scheme 1 is P_task_A_min=0.2137, and the lower bound of the task success rate for Scheme 2 is P_task_B_min=0.5290. Figure 3 As shown, under worst-case conditions, the task success rate of the redundant architecture is 2.47 times that of the serial architecture.

[0062] Step S5: Comparative analysis and structural attribution.

[0063] The comparative analysis is as follows: Performance fluctuation range: such as Figure 3As shown, the task success rate of Scheme 1 fluctuated by 0.641 (from 0.8550 to 0.2137), while the fluctuation range of Scheme 2 was 0.411 (from 0.9405 to 0.5290), and the stability of the redundant architecture was improved by 35.8%.

[0064] Capability boundary analysis: Using x1 as the variable and fixing x2 = 0.75 (medium disturbance level), we iterate through the range of x1 from 0.5 to 1, calculating the P_task value for both schemes, as shown below. Figure 4 The two capability boundary curves are shown. Sensitivity analysis revealed that Scheme 1 has a sensitivity coefficient of 0.641 for environmental parameters, while Scheme 2's sensitivity coefficient decreases to 0.432, indicating that the redundancy architecture's sensitivity to environmental changes is reduced by 32.6%. Figure 5 As shown.

[0065] The quantitative attribution is as follows: like Figure 6 As shown, under typical severe operating conditions (x1=0.5, x2=0.75), the calculated task success rate of Scheme 2 (0.6613) is 106.2% higher than that of Scheme 1 (0.3206), achieving a doubling of performance. This advantage can be attributed to the alternative redundancy topology paradigm widely adopted in Scheme 2: Quantification of structural effects: The redundancy paradigm transforms the risk of single-point failure in a serial structure into enhanced reliability of a parallel system through parallel path design. Sensitivity Reduction Mechanism: The redundant architecture reduces the system's sensitivity to changes in environmental parameters from 0.641 to 0.432, a reduction of 32.6%. This reflects the buffering effect of the redundant structure on environmental disturbances; when the performance of one unit deteriorates due to environmental degradation, the backup unit can still maintain the basic functions of the system. Robustness enhancement verification: The redundant architecture maintains a task success rate above 0.5290 in the worst case, while the serial architecture only achieves 0.2137 in the worst case. This quantitative result demonstrates that the redundant topology paradigm can improve the lower bound of the system's capability by more than 100%, providing a solid guarantee for the reliable operation of the system in highly uncertain environments.

[0066] Structure-Capability Mapping: The calculation results in this case verify the core claim of the method in this embodiment, namely, that there is a direct mapping relationship between specific causal topological paradigms and specific capability characteristics. The cascade paradigm corresponds to sensitive and vulnerable performance characteristics, while the redundancy paradigm corresponds to robust and reliable performance characteristics. This mapping relationship provides a quantifiable decision-making basis for the optimized design of the system architecture, enabling architects to scientifically select appropriate combinations of topological paradigms based on the degree of uncertainty in the task environment.

[0067] The above embodiments are only used to explain the concept of the present invention, and are not intended to limit the protection of the present invention. Any non-substantial modifications made to the present invention using this concept should fall within the protection scope of the present invention.

Claims

1. A method for generating the utility capability boundary of a system architecture based on causal topology, characterized in that, Includes the following steps: S1. Receive and parse the input system architecture description information, and transform it into an internally processable object model; S2. Construct a causal graph based on the object model, identify and mark the subgraph structures in the causal graph that conform to the predefined topological paradigm according to the predefined judgment rules, and form a causal topological graph represented by a formal language; S3. Generate a total utility function describing the overall capabilities of the system architecture based on the causal topology graph; S4. Select several variables from the total utility function and set the variation range of each variable. Under the premise of keeping the causal topology unchanged, perform regularized change deduction on the selected variables to obtain the upper and lower bounds of the system architecture utility capability under the corresponding causal topology, thereby generating the utility capability boundary. S5. Repeat steps S1 to S4 for different system architecture schemes to generate the utility capability boundaries corresponding to each system architecture scheme. Visualize and overlay the utility capability boundaries of different system architecture schemes and perform attribution analysis. Use the attribution analysis results as the basis for system architecture design adjustment, causal topology optimization, or resource allocation improvement.

2. The method for generating the system architecture utility capability boundary based on causal topology as described in claim 1, characterized in that, The system architecture description information can be generated using structured text or through graphical modeling tools.

3. The method for generating the system architecture utility capability boundary based on causal topology as described in claim 1, characterized in that, In step S1, the system architecture description information includes at least the system components, the connection relationships between the components, the task objectives, and the capability indicators; Each unit in the system must be associated with its type identifier and performance parameters; The inter-unit connection relationships are used to describe data flow, control flow, or resource dependencies; The task objectives and capability indicators are used to clarify the tasks that the system architecture needs to complete and to list the capability indicators to be evaluated.

4. The method for generating the system architecture utility capability boundary based on causal topology as described in claim 3, characterized in that, In step S2, the specific details are as follows: Includes the following processes: S21. Abstract the performance parameters, task objectives, and capability indicators of each system component into causal variable nodes; S22. Based on the connection relationship between the units, define directed edges between causal variable nodes to form a causal graph; S23. Analyze the formed causal graph, and based on the necessity relationship, sufficiency relationship and resource constraint characteristics between causal variable nodes, identify the subgraph structure that conforms to the predefined paradigm according to the predefined judgment rules, and label the identified subgraph structure with a topological paradigm label; The predefined determination rules include at least the following: The unique necessity rule states that if, in a subgraph structure of a causal graph, causal variable node A is the only direct predecessor of causal variable node B, and there are no other alternative paths that can make causal variable node B valid, and causal variable node B is the only direct predecessor of causal variable node C, then the subgraph structure is determined to satisfy the unique necessity relation and is identified as a serial topology paradigm. At the same time, it is necessary to determine the rule that if the establishment of causal variable node A requires the simultaneous satisfaction of inputs from two or more predecessor causal variable nodes, and the absence of any predecessor causal variable node will cause causal variable node A to fail to be established, then the subgraph structure is determined to satisfy the simultaneous necessity relationship and is identified as a cooperative topology paradigm. Any sufficient determination rule, if the establishment of causal variable node A can be independently triggered by any one of the multiple predecessor causal variable nodes, and each predecessor causal variable node is functionally equivalent, then the subgraph structure is determined to satisfy any sufficient relation and is identified as an alternative redundant topology paradigm. S24. Represent the causal topology graph with the topological paradigm label in a formal language and store it in the paradigm library.

5. The method for generating the system architecture utility capability boundary based on causal topology as described in claim 3, characterized in that, Step S3 includes the following specific processes: S31. Define a standardized utility function for each causal variable node, mapping the physical parameters to utility values ​​in the interval [0, 1]. S32. Based on the causal topology graph formed in step S2, select the corresponding utility combination operator according to the paradigm type, and perform bottom-up utility aggregation calculation; For the serial topology paradigm, a bottleneck-type combinatorial operator is adopted, that is, the system utility within the paradigm is defined as the minimum value or equivalent product of the utility of each causal variable node; For the cooperative topology paradigm, a joint contribution combinatorial operator is adopted, that is, multiplication, weighted summation or other fusion operations are performed on the utility of parallel branches; For alternative redundant topology paradigms, a selective combination operator is used, that is, the path with the greatest utility contribution is selected from the set of available paths, or the expected utility is calculated when considering the availability probability. S33. When there are multiple levels or nested causal topology graphs in the system architecture, the local utility calculation and aggregation are completed layer by layer in a bottom-up recursive manner until the total utility function describing the overall capability of the system architecture is generated.

6. The method for generating the system architecture utility capability boundary based on causal topology as described in claim 3, characterized in that, Step S4 includes the following specific processes: S41. Select several key variables from the independent variables of the total utility function and set a range of variation for each key variable; Key variables should meet the following rules: conduct sensitivity analysis on each causal variable in the total utility function, and select variables whose weights on the total utility value of the system exceed a preset threshold; The range of variation is determined based on the performance parameters in step S1 or user-defined input. S42. While keeping the causal topology unchanged, perform regularized variation deductions on key variables to analyze the achievable range of the system's utility capacity under different parameter combinations, including: Upper bound calculation involves finding a set of parameter values ​​within the range of variation of the key variables that maximizes the value of the total utility function. Lower bound calculation involves finding a set of parameter values ​​within the range of variation of the key variables that minimizes the value of the total utility function. Boundary generation: For a single key variable, fix other variables, traverse the entire interval of the key variable, record the upper and lower bound values ​​corresponding to each point, and form two curves. The area between the two curves is the utility capability boundary. For multiple key variables, generate a multi-dimensional boundary hypersurface or display a two-dimensional cross-section through projection. S43. Combine the calculated upper and lower bounds of utility capacity with the corresponding key variable parameter values ​​and output them in a data file or graphical format.

7. The method for generating the system architecture utility capability boundary based on causal topology as described in claim 6, characterized in that, In step S5, the attribution analysis includes: Qualitative analysis is used to identify the system architecture scheme that covers the widest range of utility capability boundaries. Quantitative attribution involves selecting parameter points, comparing the upper and lower bounds of the utility capacity of different schemes at those parameter points, analyzing the reasons for the differences, and directly linking these reasons to the differences in topological paradigms.

8. A system for generating the utility capability boundary of a system architecture based on causal topology, used to perform the method described in any one of claims 1 to 7, characterized in that, include: The data parsing module is used to receive and parse the input architecture description information and transform it into an internally processable object model; The causal topology paradigm construction module is used to construct a causal graph based on the object model, identify and mark the subgraph structures in the causal graph that conform to the predefined topology paradigm according to predefined judgment rules, and form a causal topology graph represented in a formal language. The utility capacity generation module is used to generate a total utility function describing the overall capabilities of the system architecture based on the causal topology graph. The utility capability boundary construction module is used to select several variables from the total utility function and set the variation range of each variable. Under the premise of keeping the causal topology unchanged, the selected variables are subjected to regularized change deduction to obtain the upper and lower bounds of the system's utility capability under the corresponding causal topology, thereby generating the utility capability boundary. The topology impact attribution analysis module is used to obtain several utility capability boundaries generated by several system architecture schemes, visualize and overlay the utility capability boundaries of different system architecture schemes and perform attribution analysis, and use the attribution analysis results as the basis for system architecture design adjustment, causal topology diagram optimization or resource allocation improvement.

Images