Satellite single-machine product multi-dimensional risk quantitative analysis method and system

Abstract

The invention discloses a satellite single-machine product multi-dimensional risk quantitative analysis method and system, and relates to the field of satellite single-machine product risk analysis, and the method comprises the following steps: carrying out the supplier, maturity and reliability evaluation of a satellite single-machine product, and generating a multi-dimensional evaluation result; according to a multi-dimensional evaluation result, constructing a multi-modal fuzzy distribution function and a risk level fuzzy control rule; based on the multi-modal fuzzy distribution function and a risk level fuzzy control rule, performing defuzzification by adopting an area center algorithm, and outputting a quantized risk value; and carrying out quantitative risk value matching in combination with a preset risk level evaluation rule, and determining the risk level of the satellite single-machine product. According to the method, supplier evaluation index weights are dynamically optimized through an entropy weight method, product maturity is adaptively graded through a fuzzy C-means clustering algorithm, risks are accurately quantified through multi-modal fuzzy modeling, and objectivity, adaptability and precision of satellite single-machine product risk assessment are remarkably improved.

Description

Technical Field

[0001] The present invention relates to the field of risk analysis of stand-alone satellite products, and more particularly to a multi-dimensional risk quantitative analysis method and system for stand-alone satellite products. Background Art

[0002] As a core component of a spacecraft, the reliability of a standalone satellite directly determines the success or failure of its on-orbit mission. However, these products are characterized by high technical complexity, long development cycles, and a diverse supply chain. Traditional risk assessment methods struggle to meet the demands of high-precision, dynamic management.

[0003] Traditional supplier evaluations often rely on fixed weightings, such as those for indicators like supply quality and delivery progress, based on expert experience. These methods are unable to adapt to the dynamic nature of supplier performance data, leading to delayed evaluation results. This is particularly true in scenarios involving multiple purchases or sudden quality issues, creating a risk of misjudgment and failing to truly reflect a supplier's capabilities.

[0004] Existing maturity assessments often rely on manual experience or simple thresholds, lacking quantitative modeling of a product's technical status. Manual grading is susceptible to subjective cognitive biases and cannot dynamically respond to fluctuations in product development, leading to a disconnect between maturity levels and actual risks.

[0005] Furthermore, existing risk quantification methods often rely on single fuzzy models or static rule bases, making it difficult to accurately characterize the differentiated characteristics of suppliers, maturity, and reliability. For example, reliability assessments that rely solely on normal distribution models are unable to adapt to the failure patterns of wear-and-tear products. Furthermore, defuzzification algorithms lack robustness and are susceptible to noise, causing quantification results to deviate from the true risk level.

[0006] Therefore, how to design a multi-dimensional risk quantification analysis method for satellite stand-alone products to break through the limitations of static weights, subjective grading and single models to achieve objectivity, accuracy and timeliness of satellite stand-alone product risk assessment is an urgent problem that technical personnel in this field need to solve. Summary of the Invention

[0007] In view of this, the present invention provides a multi-dimensional risk quantification analysis method for satellite stand-alone products. By dynamically optimizing the weights of supplier evaluation indicators, adaptively grading product maturity, and accurately quantifying risks, the objectivity, adaptability, and accuracy of satellite stand-alone product risk assessment are improved, providing strong guarantees for the safe and reliable operation of satellite systems.

[0008] In order to achieve the above object, the present invention adopts the following technical solutions:

[0009] In a first aspect, the present invention provides a method for quantitatively analyzing the multi-dimensional risks of a single satellite product, comprising the following steps:

[0010] S1. Conduct supplier, maturity, and reliability assessments of satellite stand-alone products and generate multi-dimensional assessment results;

[0011] S2. Constructing a multimodal fuzzy distribution function and a risk level fuzzy control rule based on the multi-dimensional evaluation results;

[0012] S3. Based on the multimodal fuzzy distribution function and the risk level fuzzy control rule, an area center algorithm is used to perform defuzzification and output a quantitative risk value;

[0013] S4. Match the quantitative risk values in accordance with preset risk level evaluation rules to determine the risk level of the satellite stand-alone product.

[0014] Furthermore, in S1, the supplier evaluation adopts an entropy weight algorithm, including:

[0015] S111. Calculate the information entropy E of each evaluation indicator j ;

[0016]

[0017] in, represents the relative score of the i-th supplier in the j-th evaluation indicator; x ij represents the original score of the i-th supplier in the j-th evaluation indicator, and m is the number of suppliers;

[0018] S112. Using information entropy E j Dynamically modified weight ω' j ;

[0019]

[0020] Where n represents the number of evaluation indicators;

[0021] S113, based on the modified weight ω' j , generate supplier comprehensive score S and divide it into grades L S .

[0022] Furthermore, in S1, the maturity assessment includes:

[0023] S121, normalize the input parameters to generate the feature vector x j ;

[0024]

[0025] Among them, N d Indicates the number of design verifications, N d,max Indicates the maximum number of design verifications, N fIndicates the number of flight assessments, N f,max represents the maximum number of flight tests, σ represents the stability of production batches, N d Indicates the maximum stability of the production batch;

[0026] S122, based on the feature vector x j , combined with the fuzzy C-means clustering algorithm for hierarchical optimization, the output maturity level L m .

[0027] Furthermore, in S122, the objective function J of the fuzzy C-means clustering algorithm is expressed as:

[0028]

[0029] Among them, k represents the number of clusters, p represents the number of data points, and μ ij represents the membership of the jth sample to the i-th cluster center, α represents the fuzzy factor, and x j represents the data vector of the jth sample, v i represents the i-th cluster center vector.

[0030] Furthermore, in S1, the reliability assessment includes:

[0031] S131. Determine product type and failure mode based on the design characteristics and historical failure data of satellite stand-alone products;

[0032] S132. Perform reliability model matching selection based on product type and failure mode determination results;

[0033] S133. Perform integrity check on the input data of the selected model, modify the model parameters based on historical failure cases, and output a reliability value that conforms to the actual project.

[0034] Furthermore, in S2, the multimodal fuzzy distribution function includes: a multimodal input membership function and an output membership function; wherein the multimodal input membership function includes:

[0035] Adopting trapezoidal membership function s for suppliers i (x,a i ,b i ,c i ,d i ):

[0036]

[0037] Among them, x represents the input variable, a i Indicates the left boundary point, b i Indicates the starting point of the ascending segment, c i Indicates the end point of the descending segment, di Indicates the right boundary point;

[0038] The Gaussian membership function μ is used for maturity j (x):

[0039]

[0040] Among them, μ j represents the mean of Gaussian distribution, σ j represents the standard deviation of Gaussian distribution;

[0041] The Sigmoid function R is used for reliability k (x):

[0042]

[0043] Among them, h k Represents the parameter that controls the steepness of the Sigmoid function, x k Represents the center point of the Sigmoid function.

[0044] Furthermore, in S2, a risk level fuzzy control rule is constructed, including:

[0045] S21. Define the fuzzy rule base to clarify supplier level, maturity level, and reliability value;

[0046] S22. Calculate the membership values of supplier, maturity, and reliability based on the multimodal fuzzy distribution function to generate initial risk level fuzzy control rules;

[0047] S23. Use historical data to verify the validity of each initial risk level fuzzy control rule, eliminate rules that are inconsistent with actual risk records, and obtain risk level fuzzy control rules.

[0048] Furthermore, in S3, the area center algorithm is expressed as:

[0049]

[0050] Among them, μ cen represents the quantitative risk value, u represents the quantitative domain variable of the risk value, and A(u) represents the membership function.

[0051] Furthermore, the S4 includes:

[0052] When the risk level L r ≥3, triggering a multi-level warning mechanism:

[0053] Level 1 warning: If the risk reaches level 3, the interface will be marked in red;

[0054] Level 2 warning: If the risk reaches level 4, it will be sent to the responsible engineer’s mobile terminal;

[0055] Level 3 warning: When the risk reaches level 5, an emergency plan will be automatically generated, including supplier replacement, redundant design activation and on-orbit maintenance plan.

[0056] In a second aspect, the present invention provides a multi-dimensional risk quantification analysis system for a single satellite product, comprising:

[0057] Multi-dimensional evaluation module: used to evaluate the supplier, maturity and reliability of satellite stand-alone products and generate multi-dimensional evaluation results;

[0058] Fuzzy distribution function construction module: used to construct a multimodal fuzzy distribution function and risk level fuzzy control rules based on the multi-dimensional evaluation results;

[0059] Quantitative risk value calculation module: used for defuzzification based on the multimodal fuzzy distribution function and the risk level fuzzy control rule, using the area center algorithm to output the quantitative risk value;

[0060] Risk level determination module: used to match the quantitative risk value in combination with preset risk level evaluation rules to determine the risk level of the satellite stand-alone product.

[0061] It can be seen from the above technical solution that compared with the prior art, the technical solution of the present invention has the following advantages:

[0062] Beneficial effects:

[0063] 1. This solution dynamically adjusts the weights of supplier evaluation indicators through the entropy weight method, overcoming the subjectivity and static nature of traditional fixed-weight methods. By combining historical data and multi-dimensional evaluation of supplier performance, it corrects the importance of indicators in real time, significantly improving the objectivity and adaptability of supplier evaluation and avoiding evaluation bias caused by changes in data distribution.

[0064] 2. By introducing the fuzzy C-means clustering algorithm, the product maturity level is automatically divided based on parameters such as the number of design verifications, the number of flight assessments, and the stability of production batches. Compared with the traditional grading method driven by human experience, this technology quantifies clustering characteristics through mathematical modeling, reduces human intervention, and ensures that the maturity assessment results are more closely aligned with engineering practice. It is especially suitable for the dynamic grading needs of products with complex technical status.

[0065] 3. Further, fuzzy distribution functions with different modes, such as trapezoidal, Gaussian, and Sigmoid, are designed for suppliers, maturity, and reliability. The area center algorithm is used for defuzzification to accurately map the fuzzy logic reasoning results into quantitative risk values. Compared with a single fuzzy model or subjective threshold division, this method greatly improves the accuracy and robustness of risk quantification through multimodal fusion, while supporting a multi-level early warning mechanism to enhance the timeliness and operability of risk management. BRIEF DESCRIPTION OF THE DRAWINGS

[0066] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are merely embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on the provided drawings without paying any creative work.

[0067] Figure 1 A flowchart of a multi-dimensional risk quantification analysis method for a single satellite product provided by an embodiment of the present invention;

[0068] Figure 2 A schematic diagram of the process of constructing risk level fuzzy control rules provided by an embodiment of the present invention;

[0069] Figure 3 A quantitative relationship diagram of risk, supplier, and maturity provided in an embodiment of the present invention;

[0070] Figure 4 A quantitative relationship diagram of risk, supplier, and reliability provided by an embodiment of the present invention;

[0071] Figure 5 A quantitative relationship diagram of risk, maturity, and reliability provided by an embodiment of the present invention;

[0072] Figure 6 A schematic diagram of the framework of a multi-dimensional risk quantification analysis system for a single satellite product provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0073] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0074] Example 1;

[0075] like Figure 1As shown, this embodiment provides a multi-dimensional risk quantitative analysis method for a single satellite product, including the following steps:

[0076] S1. Conduct supplier, maturity, and reliability assessments of satellite stand-alone products and generate multi-dimensional assessment results;

[0077] S2. Constructing a multimodal fuzzy distribution function and a risk level fuzzy control rule based on the multi-dimensional evaluation results;

[0078] S3. Based on the multimodal fuzzy distribution function and the risk level fuzzy control rule, an area center algorithm is used to perform defuzzification and output a quantitative risk value;

[0079] S4. Match the quantitative risk values in accordance with preset risk level evaluation rules to determine the risk level of the satellite stand-alone product.

[0080] This method uses the entropy weight method to dynamically optimize the weights of supplier evaluation indicators, the fuzzy C-means clustering algorithm to adaptively classify product maturity, and multimodal fuzzy modeling to accurately quantify risks. It significantly improves the objectivity, adaptability and accuracy of risk assessment of single-unit satellite products, while supporting a multi-level early warning mechanism and enhancing the timeliness and operability of risk management.

[0081] The following further details each step in the above method;

[0082] In this embodiment, S1, a supplier, maturity, and reliability assessment of a single satellite product is performed to generate a multi-dimensional assessment result;

[0083] Among them, the supplier evaluation adopts the entropy weight algorithm, including:

[0084] S111. Calculate the information entropy E of each evaluation indicator j ;

[0085]

[0086] in, represents the relative score of the i-th supplier in the j-th evaluation indicator; x ij represents the original score of the i-th supplier in the j-th evaluation indicator, and m is the number of suppliers;

[0087] S112. Using information entropy E j Dynamically modified weight ω' j ;

[0088]

[0089] Where n represents the number of evaluation indicators;

[0090] S113, based on the modified weight ω'j , generate supplier comprehensive score S and divide it into grades L S .

[0091] In this embodiment, the evaluation indicators in the supplier evaluation include: supply quality, quality issues and impacts, delivery schedule, product price, service quality, and quality improvement. The evaluation results obtained based on these indicators are shown in Table 1 below:

[0092] Table 1

[0093]

[0094]

[0095] Supplier evaluation uses an entropy weighting method to calculate the information entropy of each evaluation indicator and dynamically adjusts the weight accordingly. This method can adjust the importance of each indicator in real time based on actual data changes, avoiding the subjective bias and static issues caused by traditional fixed weights, and improving the objectivity and adaptability of supplier evaluation.

[0096] Further, maturity assessment includes:

[0097] S121, normalize the input parameters to generate the feature vector x j ;

[0098]

[0099] Among them, N d Indicates the number of design verifications, N d,max Indicates the maximum number of design verifications, N f Indicates the number of flight assessments, N f,max represents the maximum number of flight tests, σ represents the stability of production batches, N d Indicates the maximum stability of the production batch;

[0100] S122, based on the feature vector x j , combined with the fuzzy C-means clustering algorithm for hierarchical optimization, the output maturity level L m .

[0101] Among them, the objective function J of the fuzzy C-means clustering algorithm is expressed as:

[0102]

[0103] Among them, k represents the number of clusters, p represents the number of data points, and μ ij represents the membership of the jth sample to the i-th cluster center, α represents the fuzzy factor, and x j represents the data vector of the jth sample, v i represents the i-th cluster center vector.

[0104] In this embodiment, the specific product status in the maturity assessment includes: principle prototype, engineering prototype, system test, single flight assessment, multiple flight assessments, technical status solidification, small batch trial production and mass production; the obtained assessment results are shown in Table 2 below:

[0105] Table 2

[0106]

[0107]

[0108] The maturity assessment utilizes a fuzzy C-means clustering algorithm to normalize input parameters and automatically assign maturity levels. Compared to traditional grading methods driven by human experience, this approach quantifies clustering characteristics through mathematical modeling, reduces human intervention, and makes maturity assessment results more accurate to engineering realities. This makes it particularly suitable for the dynamic grading requirements of products with complex technical statuses.

[0109] Furthermore, in S1, the reliability assessment includes:

[0110] S131. Determine product type and failure mode based on the design characteristics and historical failure data of satellite stand-alone products;

[0111] S132. According to the product type and the failure mode determination results, the reliability model matching selection is carried out; specifically, if it is a success-failure type product, the binomial distribution model is used to evaluate its success probability under a fixed number of tests; if the main failure mode of the product is accidental failure, the exponential distribution model is used to evaluate its reliability under a constant failure rate; if the main failure mode of the product is wear-out failure and it is difficult to characterize the state through a single performance parameter, the Weibull distribution model is used to evaluate its failure risk that increases with time or the number of uses; if the product can characterize the working state through a single performance parameter, the normal distribution model or the stress intensity model is used to analyze the degree of deviation of the performance parameter distribution from the design threshold; if it is a long-life product, a performance degradation model or an accelerated life model is established to predict long-term reliability through degradation data or accelerated test data.

[0112] S133. Perform integrity check on the input data of the selected model, modify the model parameters based on historical failure cases, and output a reliability value that conforms to the actual project.

[0113] By carefully determining product type and failure mode, it accurately matches the most suitable reliability model, such as binomial, exponential, and Weibull distributions, to accurately assess the reliability of individual satellite products. Furthermore, by verifying the integrity of input data and modifying model parameters based on historical failure cases, the engineering practicality and accuracy of the assessment results are ensured.

[0114] In this embodiment S2, a multimodal fuzzy distribution function and a risk level fuzzy control rule are constructed based on the multi-dimensional evaluation results; the multimodal fuzzy distribution function includes: a multimodal input membership function and an output membership function;

[0115] Multimodal input membership functions include:

[0116] Adopting trapezoidal membership function s for suppliers i (x,a i ,b i ,c i ,d i ):

[0117]

[0118] Among them, x represents the input variable, a i Indicates the left boundary point, b i Indicates the starting point of the ascending segment, c i Indicates the end point of the descending segment, d i Indicates the right boundary point, i = 1, 2, 3, 4, 5 is divided into 5 levels, and the meaning of each level is shown in Table 1;

[0119] The Gaussian membership function μ is used for maturity j (x):

[0120]

[0121] Among them, μ j represents the mean of Gaussian distribution, σ j Indicates the standard deviation of the Gaussian distribution, j = 1, 2, 3, 4, 5, 6, 7, 8 is divided into 8 levels, and the meaning of each level is shown in Table 2;

[0122] The Sigmoid function R is used for reliability k (x):

[0123]

[0124] Among them, h k Represents the parameter that controls the steepness of the Sigmoid function, x k Represents the center point of the Sigmoid function, k = 1, 2, 3, 4, 5 is divided into 5 levels, specifically corresponding to: 0-0.6, very low; 0.6-0.7 low; 0.7-0.8, medium; 0.8-0.9, high, 0.9-1 very high;

[0125] Furthermore, the output membership function adopts a triangular fuzzy distribution function to define the membership of the i-th risk level. The membership reaches a maximum value of 1 when x=i and drops to 0 when x=i-1 and x=i+1. The triangular fuzzy distribution function is expressed as:

[0126]

[0127] Wherein, a=i-1 is the left boundary point, b=i is the center point, and c=i+1 is the right boundary point. In this embodiment, the value range of i is 1≤i≤5, and there are 5 levels of risk in total.

[0128] Further, such as Figure 2 As shown, the risk level fuzzy control rules are constructed, including:

[0129] S21. Define the fuzzy rule base to clarify supplier level, maturity level, and reliability value;

[0130] S22. Calculate the membership values of supplier, maturity, and reliability based on the multimodal fuzzy distribution function to generate initial risk level fuzzy control rules;

[0131] S23. Use historical data to verify the validity of each initial risk level fuzzy control rule, eliminate rules that are inconsistent with actual risk records, and obtain risk level fuzzy control rules.

[0132] Furthermore, it also includes: assigning a weight to each rule based on the matching accuracy of the rule in historical data, and retaining high-confidence rules;

[0133] In this step, by adopting appropriate membership functions for different assessment dimensions, accurate fuzzy processing of each assessment dimension is achieved. At the same time, combined with historical data verification and rule weight allocation, the effectiveness and high confidence of the risk level fuzzy control rules are ensured, thereby improving the accuracy and reliability of risk assessment data.

[0134] In this embodiment S3, based on the multimodal fuzzy distribution function and the risk level fuzzy control rule, an area center algorithm is used to perform defuzzification and output a quantitative risk value;

[0135] Among them, the area center algorithm is expressed as:

[0136]

[0137] Among them, μ cen represents the quantitative risk value, u represents the quantitative domain variable of the risk value, and A(u) represents the membership function.

[0138] The area center algorithm here fully considers the distribution characteristics of the membership function on the quantitative domain variables, and effectively avoids the limitations of a single numerical representation by calculating the area center of the membership function.

[0139] Combined with the quantitative risk value, a quantitative relationship diagram between risk and assessment factors can be constructed. When the reliability of a single satellite product is 0.8, the quantitative relationship between the risk value and the supplier and product maturity is as follows: Figure 3 As shown in the figure, when the reliability of a single-unit product is a certain value, the higher the supplier score and the higher the product maturity, the lower the product risk. When the maturity of a satellite single-unit product is level 5.5, the quantitative relationship between the risk value and the supplier and reliability is as follows: Figure 4 As shown in the figure, when the product maturity is a fixed value, the higher the supplier score, the higher the product reliability and the lower the product risk; when the satellite stand-alone product supplier score is 70, the quantitative relationship between risk value and maturity and reliability is as follows Figure 5 As shown in the figure, it can be seen that when the supplier score is a certain value, the higher the maturity of the product, the higher the reliability of the product, and the lower the product risk; in the above risk quantification relationship diagram, the blue area indicates a lower risk, and as the risk value increases, it gradually transitions to the green area of medium risk and the orange area of higher risk.

[0140] In this embodiment S4, the quantitative risk value is matched in combination with a preset risk level evaluation rule to determine the risk level of the satellite stand-alone product;

[0141] The risk level description of the satellite stand-alone product in this embodiment is shown in Table 3 below:

[0142] Table 3

[0143]

[0144]

[0145] Furthermore, it also includes:

[0146] When the risk level L r ≥3, triggering a multi-level warning mechanism:

[0147] Level 1 warning: If the risk reaches level 3, the interface will be marked in red;

[0148] Level 2 warning: If the risk reaches level 4, it will be sent to the responsible engineer’s mobile terminal;

[0149] Level 3 warning: When the risk reaches level 5, an emergency plan will be automatically generated, including supplier replacement, redundant design activation and on-orbit maintenance plan.

[0150] This step, combined with pre-set risk level assessment rules, accurately matches quantitative risk values, determines the risk level of individual satellite products, and triggers a multi-level early warning mechanism accordingly. This not only improves the timeliness of risk response but also ensures the comprehensiveness and operability of risk management.

[0151] In this embodiment, a multi-dimensional risk quantification analysis method for a single satellite product is proposed. The entropy weight method is used to dynamically optimize supplier evaluation weights, the fuzzy C-means clustering algorithm is used to adaptively classify product maturity, and a risk rule library is constructed in combination with a multimodal fuzzy distribution function. The area center algorithm is used to output quantitative risk values. Ultimately, through preset risk level matching and a multi-level early warning mechanism, the objectivity, adaptability, and timeliness of risk assessment are significantly improved.

[0152] Example 2;

[0153] like Figure 6 As shown, this embodiment provides a multi-dimensional risk quantification analysis system for a single satellite product, including:

[0154] Multi-dimensional evaluation module: used to evaluate the supplier, maturity and reliability of satellite stand-alone products and generate multi-dimensional evaluation results;

[0155] Fuzzy distribution function construction module: used to construct a multimodal fuzzy distribution function and risk level fuzzy control rules based on the multi-dimensional evaluation results;

[0156] Quantitative risk value calculation module: used for defuzzification based on the multimodal fuzzy distribution function and the risk level fuzzy control rule, using the area center algorithm to output the quantitative risk value;

[0157] Risk level determination module: used to match the quantitative risk value in combination with preset risk level evaluation rules to determine the risk level of the satellite stand-alone product.

[0158] This embodiment provides a multi-dimensional risk quantification analysis system for stand-alone satellite products. The system integrates four major modules: multi-dimensional assessment, fuzzy distribution function construction, quantitative risk value calculation, and risk level determination. It can comprehensively and accurately assess the risks of stand-alone satellite products, and provide strong support for risk management and decision-making through quantitative risk value output and risk level determination.

[0159] The various embodiments in this specification are described in a progressive manner, with each embodiment focusing on the differences from other embodiments. References to the same or similar parts between the various embodiments are sufficient. For the systems disclosed in the embodiments, since they correspond to the methods disclosed in the embodiments, the description is relatively simple. For relevant parts, refer to the method description.

[0160] The above description of the disclosed embodiments is intended to enable one skilled in the art to implement or use the present invention. Various modifications to these embodiments will be readily apparent to one skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present invention. Therefore, the present invention is not limited to the embodiments shown herein but is intended to conform to the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A multi-dimensional risk quantitative analysis method for a single satellite product, characterized by: The following steps are involved: S1. Conduct supplier, maturity, and reliability assessments of satellite stand-alone products and generate multi-dimensional assessment results; S2. Constructing a multimodal fuzzy distribution function and a risk level fuzzy control rule based on the multi-dimensional evaluation results; S3. Based on the multimodal fuzzy distribution function and the risk level fuzzy control rule, an area center algorithm is used to perform defuzzification and output a quantitative risk value; S4. Match the quantitative risk values in accordance with preset risk level evaluation rules to determine the risk level of the satellite stand-alone product.

2. The multi-dimensional risk quantitative analysis method for a single satellite product according to claim 1 is characterized in that: In S1, the supplier evaluation adopts the entropy weight algorithm, including: S111. Calculate the information entropy E of each evaluation indicator j ; in, represents the relative score of the i-th supplier in the j-th evaluation indicator; x ij represents the original score of the i-th supplier in the j-th evaluation indicator, and m is the number of suppliers; S112. Using information entropy E j Dynamically modified weight ω' j ; Where n represents the number of evaluation indicators; S113, based on the modified weight ω' j , generate supplier comprehensive score S and divide it into grades L S .

3. The multi-dimensional risk quantitative analysis method for a single satellite product according to claim 1, characterized in that: The S1, maturity assessment, includes: S121, normalize the input parameters to generate the feature vector x j ; Among them, N d Indicates the number of design verifications, N d,max Indicates the maximum number of design verifications, N f Indicates the number of flight assessments, N f,max represents the maximum number of flight tests, σ represents the stability of production batches, N d Indicates the maximum stability of the production batch; S122, based on the feature vector x j , combined with the fuzzy C-means clustering algorithm for hierarchical optimization, the output maturity level L m .

4. The multi-dimensional risk quantitative analysis method for a single satellite product according to claim 3 is characterized in that: In S122, the objective function J of the fuzzy C-means clustering algorithm is expressed as: Among them, k represents the number of clusters, p represents the number of data points, and μ ij represents the membership of the jth sample to the i-th cluster center, α represents the fuzzy factor, and x j represents the data vector of the jth sample, v i represents the i-th cluster center vector.

5. The multi-dimensional risk quantitative analysis method for a single satellite product according to claim 1 is characterized in that: The reliability assessment in S1 includes: S131. Determine product type and failure mode based on the design characteristics and historical failure data of satellite stand-alone products; S132. Perform reliability model matching selection based on product type and failure mode determination results; S133. Perform integrity check on the input data of the selected model, modify the model parameters based on historical failure cases, and output a reliability value that conforms to the actual project.

6. The multi-dimensional risk quantitative analysis method for a single satellite product according to claim 1, characterized in that: In S2, the multimodal fuzzy distribution function includes: a multimodal input membership function and an output membership function; wherein the multimodal input membership function includes: Adopting trapezoidal membership function s for suppliers i (x,a i ,b i ,c i ,d i ): Among them, x represents the input variable, a i Indicates the left boundary point, b i Indicates the starting point of the ascending segment, c i Indicates the end point of the descending segment, d i Indicates the right boundary point; The Gaussian membership function μ is used for maturity j (x): Among them, μ j represents the mean of Gaussian distribution, σ j represents the standard deviation of Gaussian distribution; The Sigmoid function R is used for reliability k (x): Among them, h k Represents the parameter that controls the steepness of the Sigmoid function, x k Represents the center point of the Sigmoid function.

7. The multi-dimensional risk quantitative analysis method for a single satellite product according to claim 1, characterized in that: In S2, the risk level fuzzy control rules are constructed, including: S21. Define the fuzzy rule base to clarify supplier level, maturity level, and reliability value; S22. Calculate the membership values of supplier, maturity, and reliability based on the multimodal fuzzy distribution function to generate initial risk level fuzzy control rules; S23. Use historical data to verify the validity of each initial risk level fuzzy control rule, eliminate rules that are inconsistent with actual risk records, and obtain risk level fuzzy control rules.

8. The multi-dimensional risk quantitative analysis method for a single satellite product according to claim 1, characterized in that: In S3, the area center algorithm is expressed as: Among them, μ cen represents the quantitative risk value, u represents the quantitative domain variable of the risk value, and A(u) represents the membership function.

9. The multi-dimensional risk quantitative analysis method for a single satellite product according to claim 1, characterized in that: Said S4 includes: When the risk level L r ≥3, triggering a multi-level warning mechanism: Level 1 warning: If the risk reaches level 3, the interface will be marked in red; Level 2 warning: If the risk reaches level 4, it will be sent to the responsible engineer’s mobile terminal; Level 3 warning: When the risk reaches level 5, an emergency plan will be automatically generated, including supplier replacement, redundant design activation and on-orbit maintenance plan.

10. A multi-dimensional risk quantitative analysis system for satellite stand-alone products, characterized by: include: Multi-dimensional evaluation module: used to evaluate the supplier, maturity and reliability of satellite stand-alone products and generate multi-dimensional evaluation results; Fuzzy distribution function construction module: used to construct a multimodal fuzzy distribution function and risk level fuzzy control rules based on the multi-dimensional evaluation results; Quantitative risk value calculation module: used for defuzzification based on the multimodal fuzzy distribution function and the risk level fuzzy control rule, using the area center algorithm to output the quantitative risk value; Risk level determination module: used to match the quantitative risk value in combination with preset risk level evaluation rules to determine the risk level of the satellite stand-alone product.

Images