Batch satellite integrated technology maturity evaluation method and system

Abstract

The application discloses a batch satellite comprehensive technology maturity evaluation method and system, and relates to the technical field of spacecraft batch production quality evaluation. A three-layer quality consistency evaluation framework covering manufacturing processes, product characteristics and on-orbit operation is constructed, and comprehensive evaluation of different levels of quality states is realized through various statistical analysis methods. On this basis, test, process, product, technology, fault and on-orbit data are integrated to form a multi-dimensional comprehensive technology maturity index, and the weight is adjusted in stages according to the accumulation of on-orbit data. According to the comprehensive evaluation result, the test strategy is managed in stages, a large amount of cutting can be implemented in the high maturity stage, moderate adjustment is suitable in the medium stage, and full inspection is maintained in the low maturity stage. At the same time, the on-orbit and ground linkage mechanism is introduced, so that the on-orbit anomaly can be fed back in real time and drive the ground test strategy to be dynamically tightened.

Description

Technical Field

[0001] This invention relates to the field of quality evaluation technology for mass production of spacecraft, specifically to a method and system for evaluating the comprehensive technological maturity of mass-produced satellites. Background Technology

[0002] In recent years, with the rapid growth in demand for large-scale deployment of low-Earth orbit satellite constellations, constellation projects such as SpaceX's Starlink system and OneWeb have driven a significant shift in spacecraft manufacturing models, evolving from traditional single-satellite customized development to a high-paced, large-scale mass production model. In this context, spacecraft production not only needs to meet performance and reliability requirements but also needs to balance cost control and capacity expansion. However, the traditional piece-by-piece, full-coverage testing system faces severe challenges under mass production conditions: on the one hand, maintaining a 100% screening rate will lead to a significant increase in testing resource requirements; for example, the demand for thermal vacuum testing equipment may reach several times the steady-state requirement. On the other hand, for constellations with thousands of satellites, the cumulative cost of acceptance testing alone could reach hundreds of millions of dollars. Therefore, in the context of mass production, reducing unnecessary tests and implementing reasonable test tailoring has become an inevitable trend. The prerequisite for achieving this goal is the ability to conduct scientific and reliable assessments of product quality consistency.

[0003] However, existing technologies still have significant shortcomings in product quality consistency evaluation: First, the evaluation dimensions are relatively singular. Most methods only analyze the stability of the manufacturing process based on statistical process control (SPC), failing to simultaneously cover important dimensions such as the distribution of key product characteristics and on-orbit performance, resulting in one-sided evaluation results. Second, there is a lack of an effective on-orbit data closed-loop feedback mechanism. Ground maturity evaluation cannot be dynamically corrected based on on-orbit fault information, causing a disconnect between ground assessment and actual on-orbit performance. When new on-orbit fault modes occur, it is difficult to adjust ground test strategies in a timely manner. Third, there is a lack of a unified comprehensive quantitative index system. Existing methods cannot integrate and analyze multi-source information such as manufacturing process data, product performance data, test records, fault closed-loop status, and on-orbit reliability, thus failing to provide clear and quantitative basis for test tailoring decisions.

[0004] The information disclosed in the background section is only intended to enhance the understanding of the background of this disclosure, and therefore may include information that does not constitute prior art known to those skilled in the art. Summary of the Invention

[0005] The purpose of this invention is to provide a method and system for evaluating the overall technological maturity of mass-produced satellites, in order to solve the problems mentioned in the background art.

[0006] To achieve the above objectives, the present invention provides the following technical solution: a method for evaluating the integrated technology maturity of mass-produced satellites, comprising the following steps: A quality consistency evaluation framework is established, which divides the quality consistency of mass-produced satellites into three levels: process consistency, product consistency, and reliability consistency. Process consistency is evaluated based on statistical process control to monitor manufacturing process parameters, product consistency is evaluated based on the statistical distribution of key product characteristics, and reliability consistency is evaluated based on on-orbit operation data. Statistical analysis is performed based on manufacturing process parameters to obtain process capability levels and monitor process fluctuation trends to comprehensively determine process consistency indicators. Based on the key characteristics of the products, they are classified and cross-batch statistical distribution consistency tests are carried out. The product consistency indicators are determined by combining the results of multivariate statistical analysis. By combining the satellite data from completed tests and experiments, statistical inferences are made to obtain dynamically updated test effectiveness indicators; Based on the manufacturing process execution records and fault handling records, statistical analysis was conducted to obtain process stability index and fault closure rate index, respectively. Based on on-orbit operation data, the on-orbit working time, fault conditions, fault recurrence, fault occurrence stage, and performance degradation trend are analyzed to obtain on-orbit reliability indicators. The comprehensive technology maturity index is obtained by weighting and integrating process consistency index, product consistency index, test effectiveness index, process stability index, fault closure rate index and on-orbit reliability index, and the weights are adjusted in stages as on-orbit data accumulates. The test strategy is adjusted based on the comprehensive technology maturity index, and the comprehensive technology maturity index is adjusted in conjunction with the on-orbit reliability index to tighten the test strategy when the on-orbit reliability index declines.

[0007] Preferably, when monitoring the trend of manufacturing process parameters, an exponentially weighted moving average method is used, with the smoothing coefficient λ ranging from 0.1 to 0.3 and the control limit width coefficient ranging from 2.7 to 3.0. In the process consistency evaluation, when the minimum value of the process capability index of all manufacturing process parameters is lower than 1.0, the process consistency index is directly determined to be zero to characterize the state of insufficient manufacturing process capability.

[0008] Preferably, when verifying the consistency of key product characteristics, a two-sample distribution test method based on the difference of empirical distribution function is adopted, and the judgment is made at a significance level of 0.05. At the same time, multiple key characteristic parameters are constructed into multidimensional vectors for joint distribution analysis. By calculating multivariate statistics and converting them into judgment statistics under the corresponding distribution, the overall distribution consistency between different batches is judged.

[0009] Preferably, during the on-orbit operation data analysis process, when the statistically obtained recurring failure rate exceeds 0.3, a comprehensive re-inspection is performed on subsequent batches of products; when the early failure rate exceeds 0.4, the environmental stress screening intensity is increased; when the performance degradation rate exceeds a preset threshold, the qualification criteria for ground tests are adjusted to strengthen the screening standards.

[0010] Preferably, the weights of each evaluation indicator in the calculation of the comprehensive technology maturity index are adjusted in stages according to the accumulation of on-orbit data. In the stage where there is no on-orbit data, the weights related to on-orbit reliability are set to zero and the weights of test effectiveness are increased. In the initial accumulation stage, the weight of on-orbit reliability is introduced and the weight of ground evaluation is appropriately reduced. In the full accumulation stage, the weight of on-orbit reliability is further increased and the weights of other indicators are reduced, so that the weight allocation changes dynamically with the maturity of data.

[0011] Preferably, a change impact assessment mechanism is introduced in the product consistency evaluation process. When design changes, process changes, or supply chain changes occur, key characteristic data of multiple samples before and after the change are collected, and distribution consistency test or mean difference test is performed. When the test results show that there is a significant difference, the relevant test items are resumed to ensure the consistency of the product after the change.

[0012] Preferably, during the comprehensive technology maturity assessment and test tailoring process, a dynamic linkage rule is established between on-orbit reliability indicators and test strategies. When the on-orbit reliability indicators decline, the comprehensive technology maturity indicators are reduced synchronously according to a preset ratio, and the interval levels of environmental stress screening tests are gradually rolled back. Specifically, whenever the on-orbit reliability indicators decline by a certain amount, the corresponding level is rolled back by one level.

[0013] Preferably, during the evaluation of test effectiveness, while updating the test effectiveness indicators batch by batch, a lower limit of the confidence interval is introduced as a judgment criterion. When the lower limit of the confidence interval reaches a preset threshold, it is determined that the test stability of the current batch of products meets the requirements; otherwise, the original test intensity remains unchanged.

[0014] Preferably, in the joint evaluation of process consistency and product consistency, a correlation is established between key manufacturing process parameters and corresponding key product characteristics. When abnormal fluctuations in process parameters are detected and the distribution of corresponding product characteristics is deviated, targeted reviews are performed on the relevant processes, and the corresponding batch of products is marked.

[0015] A comprehensive technology maturity assessment system for mass-produced satellites includes a process consistency assessment module, a product consistency assessment module, a test effectiveness assessment module, a process and fault management module, an on-orbit data linkage module, and a comprehensive maturity calculation and decision-making module. The process consistency evaluation module connects to the manufacturing execution system to collect manufacturing process parameter data, performs statistical analysis and calculates process capability indicators and trend statistics, determines the process status based on control limits, outputs process consistency evaluation results and generates early warnings when anomalies occur. The product consistency evaluation module extracts key characteristic data of batch products from the testing database, performs classification processing, executes distribution consistency test and multivariate analysis, outputs product consistency evaluation results and identifies abnormal batches; The test effectiveness evaluation module obtains test result data from the test integration management system, performs statistical modeling, and updates the test effectiveness indicators and their confidence intervals based on Bayesian analysis. The process and fault management module obtains process records and fault data from the quality management system, performs statistical analysis, and obtains evaluation results on process stability and fault closure rate. The on-orbit data linkage module receives on-orbit data through the telemetry and control interface, performs structured processing, calculates on-orbit reliability indicators, and generates alarms when anomalies occur. The comprehensive maturity calculation and decision-making module receives the evaluation results and performs weighted calculations according to the stage weights to obtain the comprehensive technology maturity index. It then determines the level based on the grading thresholds and outputs the trimming level.

[0016] The technical effects and advantages provided by the present invention in the above technical solution are as follows: This invention establishes a three-tiered quality consistency evaluation framework encompassing process consistency, product consistency, and reliability consistency. It systematically characterizes the quality of mass-produced satellites from three dimensions: manufacturing process parameter control, distribution of key product characteristics, and on-orbit performance. Compared to traditional methods that rely solely on statistical process control for manufacturing stability assessment, this framework extends from the process dimension to the entire process-product-application chain. This allows the evaluation results to not only reflect the production status but also reveal the actual performance and reliability levels of the product. This effectively overcomes the problems of single evaluation dimensions and one-sided results in existing technologies, significantly improving the comprehensiveness and accuracy of quality evaluation.

[0017] This invention introduces an on-orbit reliability index, directly incorporating on-orbit operational data into the technology maturity assessment system, thus realizing a dynamic closed-loop linkage mechanism between ground evaluation results and actual on-orbit performance. When the on-orbit failure rate increases or the performance degradation trend intensifies, the on-orbit reliability index decreases, which in turn leads to a synchronous decrease in the comprehensive technology maturity index, automatically triggering a tightening and adjustment of ground test strategies. This mechanism transforms quality control from traditional one-way evaluation to real-time feedback regulation, effectively compensating for the disconnect between ground evaluation and on-orbit operation in existing technologies, and improving the system's responsiveness to abnormal trends and risk prevention capabilities.

[0018] This invention constructs a six-dimensional comprehensive technology maturity index system, encompassing test effectiveness, process consistency, product consistency, process stability, fault closure rate, and on-orbit reliability. It achieves multi-source data fusion through weight allocation. Furthermore, based on the accumulation of on-orbit data, the weights are dynamically adjusted in three stages (A, B, and C), enabling the evaluation system to automatically adapt to different quality control priorities throughout the product lifecycle. In the early stages, ground data is emphasized, while the weight of on-orbit data is gradually increased in the mid-to-late stages. This provides a scientific, continuous, and quantifiable basis for experimental tailoring decisions, avoiding the uncertainty caused by empirical judgment.

[0019] In this invention, all evaluation indicators are based on clearly defined data sources and, combined with rigorously defined calculation methods and judgment rules, achieve standardization and repeatability of the quality evaluation process. Through a unified data structure, indicator model, and calculation process, complex quality information can be transformed into quantifiable and comparable evaluation results, supporting continuous iterative optimization. Simultaneously, this method possesses good engineering feasibility and can be directly embedded into production line information systems to form automated evaluation tools, enabling real-time monitoring, early warning analysis, and decision support, significantly improving the quality management level and intelligence of mass-produced satellite manufacturing processes. Attached Figure Description

[0020] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in this invention. For those skilled in the art, other drawings can be obtained based on these drawings.

[0021] Figure 1 This is a flowchart of a method for evaluating the overall technological maturity of mass-produced satellites according to the present invention.

[0022] Figure 2 This is a schematic diagram of the modules of a mass-produced satellite integrated technology maturity evaluation system according to the present invention. Detailed Implementation

[0023] Exemplary embodiments will now be described more fully with reference to the accompanying drawings. However, these exemplary embodiments can be implemented in many forms and should not be construed as limited to the examples set forth herein; rather, they are provided so that the description of this disclosure will be more complete and fully convey the concept of the exemplary embodiments to those skilled in the art.

[0024] This invention provides, for example Figure 1 The method for evaluating the overall technological maturity of mass-produced satellites, as shown, includes the following steps: In advanced production models, the quality consistency requirements for mass-produced satellites shift from individual unit reliability to batch stability control, necessitating the construction of a comprehensive evaluation system covering the manufacturing process, product characteristics, and on-orbit operational status. Quality consistency is divided into three levels: the manufacturing process level reflects the stability of the production process; the product characteristics level characterizes the performance distribution differences between batches; and the on-orbit operational level reflects the reliability performance during actual service. These three levels form a closed-loop connection from ground manufacturing to on-orbit application, ensuring that the evaluation results not only reflect the current production status but also guide subsequent production and testing strategy adjustments.

[0025] At the manufacturing process level, stability monitoring is achieved through continuous data collection and analysis of key process parameters. These parameters include assembly torque, peak welding temperature and welding time, propellant charge, solar array deployment torque, and RF link performance indicators. Each parameter has upper and lower limits to define the allowable fluctuation range. Statistical analysis is used to obtain the process mean and standard deviation, and based on this, the process capability index is calculated. in It is the process capability index, a core indicator used to measure whether a manufacturing process is stable and meets specifications. Indicates the upper limit of the specification. Indicates the lower limit of the specification. This represents the process mean. The standard deviation is the standard deviation. This index measures the position and dispersion of the process output distribution relative to the specification range. When the mean deviates from the center or fluctuates excessively, the process capability index decreases, reflecting insufficient process stability.

[0026] To improve sensitivity to trend changes, an exponentially weighted moving average statistical method is introduced, which is expressed as follows: in, This is the EWMA statistic, which is the current statistical value in the Exponentially Weighted Moving Average (EWMA) method. Indicates the current sampled value. This represents the statistical value at the previous moment. The smoothing coefficient controls the weighting of historical and current data. This method can highlight the latest changes while preserving historical trend information and has a strong ability to detect slow drift.

[0027] At the product characteristic level, consistency evaluation is achieved through statistical distribution analysis of key performance parameters. Key characteristics cover indicators such as overall satellite mass, power consumption, thermal margin, link budget margin, and attitude control accuracy, and are categorized into mandatory consistency and permissible distribution categories based on their impact on safety and functionality. For permissible distribution characteristics, a distribution consistency test is used to determine whether significant differences exist between different batches. The Kolmogorov-Smirnov statistic is employed. in, and These are the empirical distribution functions for two batches of samples, respectively. This represents the maximum deviation between two distributions. This statistic measures the degree of difference between the overall distribution patterns of the two data sets. In scenarios involving joint analysis of multiple characteristics, Hotelling is introduced. A comprehensive evaluation of statistical measures: in, and For sample size, The difference between the means vectors, To merge the covariance matrices, Its inverse matrix, This is the transpose vector. This statistic reflects the overall changes in multidimensional characteristics, avoiding biases caused by single-indicator evaluations. Reliability is evaluated through long-term operational data at the on-orbit operation level. Data dimensions include cumulative operating time, failure modes and frequencies, recurring failures, failure occurrence stages, and performance degradation trends. The mean time between failures (MTBF) is constructed based on cumulative operating time and the total number of failures. in, Mean time between failures (MTBF) refers to the average time between failures during actual on-orbit operation. This indicates the total operating time of all satellites. This represents the total number of failures. This metric is used to characterize the system's reliability level in a real-world operating environment.

[0028] Further comparison with design specifications yields on-orbit reliability evaluation metrics: in, It is an on-orbit reliability metric, used to measure whether actual on-orbit performance meets design objectives. This represents the design-required mean time between failures (MTBF), with a maximum value of 1 to limit the evaluation results. This metric reflects whether the actual reliability meets the design expectations.

[0029] Through collaborative analysis at these three levels, a full-chain evaluation is achieved, encompassing the manufacturing process, product characteristics, and on-orbit operational status. The process level provides the foundation for production stability, the product level reflects batch consistency, and the on-orbit level verifies actual reliability performance. The fusion of these three types of data transforms quality evaluation from static analysis to dynamic closed-loop monitoring, facilitating timely identification of potential risks and guiding the optimization of production and testing strategies.

[0030] To determine whether a statistic is under control, control limits need to be set, which are expressed as follows: in, Indicates the upper control limit. Indicates the lower control limit; and These are the process mean and standard deviation, respectively. For smoothing coefficients; Indicates the current sample number; The control limit width coefficient, typically between 2.7 and 3.0, is used to determine the width range of the control limits. The expression within the square root is used to correct for the fluctuation characteristics of the statistic when the initial sample size is small, so that the control limits gradually converge with the sample size. This control limit constitutes the dynamic threshold when the statistic... If the value exceeds this range, it indicates an abnormal deviation in the process, requiring intervention.

[0031] In practical applications, by establishing the aforementioned statistical model for each manufacturing process parameter, its stability can be determined in real time. When the process falls within the control limits, it is considered to be under control; when... When the upper or lower control limit is exceeded, it indicates that the parameter has abnormal fluctuations or trend deviations, and further analysis of the cause is required, such as changes in equipment status, fluctuations in process conditions, or operational deviations.

[0032] After completing the single-parameter evaluation, a comprehensive analysis of the multi-parameter results is required. Let there be a total of... For each of the manufacturing process parameters, a process capability index is calculated, and its control state is identified. First, the minimum process capability index among all parameters is determined, denoted as... This value represents the weakest link in the overall manufacturing process. Based on this, and combined with the results of control chart analysis, a process consistency index is formed.

[0033] When the process capability index of all manufacturing process parameters is not lower than 1.33, and the statistics of all parameters do not exceed the limits, the overall stability of the manufacturing process is considered to be high, and the corresponding evaluation result is: in, This represents the process consistency index, with a value of 1.0 indicating that the stability requirements are fully met. When the statistics of some parameters exceed the limits, the evaluation results need to be corrected. The calculation method is as follows: in, It is the minimum process capability index among all parameters; Indicates the number of parameters that went out of bounds; Indicates the total number of parameters; This value is used to reflect the impact of the proportion of out-of-bounds parameters on overall stability. The more out-of-bounds parameters there are, the smaller this value becomes, thereby reducing the overall evaluation result. Used to reflect the degree of process capability deficiency, the evaluation result is proportionally reduced when the minimum process capability index is below the ideal threshold of 1.33. When the minimum process capability index is below 1.0, it indicates that at least one critical process parameter deviates significantly from the specification requirements. In this case, the manufacturing process is considered to lack basic stability, and the evaluation result is directly taken as: This situation indicates that there are significant risks in the production process, and the prerequisites for subsequent quality control or trial cutting are not met.

[0034] By combining single-parameter statistical analysis with multi-parameter comprehensive evaluation using the above method, a systematic quantitative description of manufacturing process consistency is achieved. The process capability index is used to evaluate static distribution characteristics, the exponentially weighted moving average method is used to capture dynamic trends, control limits are used to identify abnormal states, and the comprehensive index is used to reflect the overall stability level. This method can identify potential risk points in a timely manner during mass production and provide data support for subsequent process optimization, quality control, and production decisions.

[0035] To ensure quality consistency control at the mass production satellite product level, a systematic assessment of the consistency level between different production batches is needed, based on the statistical distribution of key product characteristics. Since random fluctuations within a certain range are unavoidable during mass production, not all characteristics require absolute consistency. Instead, they should be categorized and managed according to their impact on safety, interface compatibility, and lifespan. For key characteristics that are allowed to fluctuate within a certain range, statistical testing methods are used to determine whether the distribution between different batches remains consistent, thus reflecting whether production stability extends to the product level.

[0036] In practical processing, for each allowed classification key characteristic, the measured data from the two most recent production batches must be selected as the analysis object, with each batch containing at least 10 satellite samples. Let the first batch sample be... The second batch of samples is ,in and These represent the sample sizes of the two batches. By performing statistical analysis on these two sets of data, their respective empirical cumulative distribution functions can be obtained, which are used to describe the distribution characteristics of the data within different value intervals.

[0037] First, the Kolmogorov-Smirnov two-sample test is used to assess the difference in distribution between the two batches. The core of this method lies in calculating the maximum deviation between two empirical distribution functions, expressed as follows: in, This represents the maximum difference between two sample distributions; This represents the empirical cumulative distribution function of the first batch of samples; Represents the empirical cumulative distribution function of the second batch of samples; symbol Indicates all possible values Find the maximum value within the range. This statistic reflects the degree of difference between the two samples in the overall distribution.

[0038] At the significance level Under the condition, the statistic Compare with the critical value. The critical value is determined by the function. It is determined that its value is 1.36 at a significance level of 0.05. Two batches are considered statistically significant when the following relationship is met: in, The coefficient is related to the significance level. and These represent the two sample sizes; the square root term is used to correct for the impact of different sample sizes on the statistical results. When this inequality holds, it indicates that the data distributions of the two batches are inconsistent, which may indicate process fluctuations or systematic biases.

[0039] Building upon univariate testing, to further improve evaluation accuracy, multivariate statistical analysis is introduced to conduct a joint consistency assessment of multiple key characteristics. (Assuming similarities are considered...) Each satellite can be represented as a key characteristic parameter. Dimensional vector: in, Indicates the first Characteristic vectors of each satellite; Indicates the first The satellite in the Values ​​on a single characteristic; symbol This indicates the transpose operation. By calculating the mean vectors for the two batches of samples separately, the mean difference vector can be obtained. This is expressed as the difference between the two batch mean vectors. The combined covariance matrix is ​​also calculated simultaneously. It is used to reflect the correlation between various characteristics and the overall fluctuation.

[0040] Based on the above statistics, construct Statistics, expressed in the following form: in, Represents multivariate statistics; and The sample size for both batches; The difference between the means vectors; Its transpose; To merge the covariance matrices; This is the inverse of the covariance matrix. This statistic comprehensively considers the correlation between multiple variables and can reflect the overall distribution differences.

[0041] To facilitate judgment, The statistic is converted to the F-statistic, which is expressed in the following form: in, This is the transformed statistic; This indicates the number of characteristics involved in the analysis; the combined terms in the denominator and numerator are used to adjust for the influence of sample size and variable dimensionality on the statistical results. At the significance level... Under the conditions, the calculated Values ​​and corresponding degrees of freedom Compare the distribution critical values, when When the value exceeds the critical value, it is considered that the two batches have significant differences in the multivariate joint distribution.

[0042] After completing the above statistical tests, a summary analysis is performed on the key characteristics of all permissible distribution classes. Let the total number of permissible distribution class characteristics be... The number of characteristics that failed the Kolmogorov-Smirnov test is The product consistency index is defined as follows: in, This represents a product consistency index, with a value ranging from 0 to 1. Indicates the number of features that failed the consistency test; This indicates the total number of characteristics involved in the evaluation. This indicator reflects the proportion of characteristics that maintain consistency. When all characteristics pass the test, the value is 1, indicating complete consistency; when a large number of characteristics fail the test, the value decreases.

[0043] When multiple variables If the test results also fail, it indicates a joint distribution shift among multiple key characteristics. In this case, the product consistency index is further revised by introducing a penalty coefficient, the expression of which is as follows: Here, 0.8 is the penalty coefficient, used to reflect the additional risk brought about by the inconsistency of multivariate distributions. Through this correction, the consistency evaluation value can be further reduced based on the univariate test results, thereby more accurately reflecting the overall status of the product.

[0044] For critical characteristics that must be specified, the requirements are much higher than those for characteristics that can be distributed. Any deviation from the specifications indicates an unacceptable deviation in the product's critical performance or safety. In this case, the product consistency metric is directly set as follows: This value indicates that the product consistency does not meet the requirements at all, and there is no possibility of further optimization or trimming.

[0045] By employing the aforementioned multivariate statistical analysis methods, combining univariate distribution tests with multivariate joint tests, and integrating them with a classification management strategy, a comprehensive evaluation of product consistency can be achieved. The Kolmogorov-Smirnov test is used to capture distributional differences in single characteristics. Tests are used to identify synergistic variations among multiple characteristics, while the construction of consistency indices enables the quantitative expression of complex statistical results. The holistic approach not only identifies significant differences between batches but also quantifies the degree of these differences, providing a reliable basis for subsequent quality control and production adjustments.

[0046] In the development of mass-produced satellites, testing and experimentation play a crucial role in identifying potential defects, verifying product performance, and ensuring delivery quality. Due to the continuously increasing sample size in mass production environments, relying solely on the test results of a single satellite is insufficient to comprehensively reflect the overall quality level. Therefore, it is necessary to introduce statistical inference methods to comprehensively analyze historical test data, thereby forming a test effectiveness evaluation index that can be dynamically updated as production progresses. This index not only reflects the current testing system's defect detection capability but also assesses the product's stability during continuous production.

[0047] In the statistical modeling process, Bayesian estimation is used to model the effectiveness of the test. This method can continuously update the understanding based on existing data, allowing the evaluation results to gradually converge as the sample size increases, thus balancing the situations of insufficient information in the early stages and sufficient data in the later stages. The core of the Bayesian method lies in obtaining the posterior distribution by combining the prior distribution with the observed data, thereby achieving a probabilistic estimate of the unknown parameters.

[0048] In the initial stage, the Beta distribution is selected as the prior distribution, and its parameters are set as follows: 、 The corresponding form is Beta(1,1). This distribution is uniform, meaning that no biased assumptions are made about the probability of passing the test without any prior information. Parameters and These represent the prior weights for the initial number of successes and failures, respectively. Under these values, the two weights are equal, reflecting the unbiased prior assumption.

[0049] As production progresses, the first When evaluating a satellite, it is necessary to statistically analyze the previous data. The performance of each satellite in all tests and experiments. Assume... Indicates as of the date The cumulative number of satellites that have passed all tests and trials without defects before the current satellite is calculated. This variable reflects the cumulative number of successful samples in the historical production process. Here, "defect-free pass" indicates a state where no abnormalities or failures were found under the established testing system. Based on the above statistics, the number of satellites that have passed all tests and trials without defects can be calculated. The posterior mean of the test validity index corresponding to each satellite time point is expressed as follows: in, Indicates the first The effectiveness indicators of testing each satellite at any given time; These are the parameters for prior success. For prior failure parameters; For the front The cumulative number of defect-free passes among the satellites; in the denominator This represents the number of satellites that have completed testing. Essentially, this expression is a Bayesian estimate of the probability of a test passing, and its result can be understood as the expected value of the probability of future product tests passing based on existing data.

[0050] As the sample size increases, the denominator gradually increases, the influence of prior parameters gradually weakens, and the posterior mean is gradually dominated by actual observation data. When a large number of satellites pass without defects... A value close to 1 indicates that the product performance is stable under the testing system; if defects occur frequently, this indicator will decrease, reflecting a high proportion of problems exposed during the testing phase.

[0051] Relying solely on the posterior mean is insufficient to fully reflect uncertainty; therefore, confidence interval analysis is needed to obtain more conservative evaluation results. The lower bound of the posterior distribution at a given confidence level can be calculated using the quantile function of the Beta distribution, expressed as follows: in, This represents the lower bound of the posterior distribution at a 95% confidence level. The quantile function represents the Beta distribution; 0.025 corresponds to the lower quantile of the two-sided 95% confidence interval; parameters Indicates the number of successes after the update; parameter This represents the updated number of failures. This lower bound reflects a conservative estimate of the test's validity, taking into account statistical uncertainty.

[0052] When the sample size is small, due to significant statistical uncertainty, this lower bound is usually lower than the posterior mean. As the sample size increases, the distribution gradually converges, and the lower bound gradually approaches the mean, indicating that the reliability of the evaluation results increases. By introducing this lower bound, overestimation due to insufficient samples or random fluctuations can be avoided, thereby improving the robustness of the evaluation.

[0053] In engineering applications, when When a preset threshold (e.g., 0.85) is reached, the product's performance in the testing system can be considered highly stable at the current confidence level. This criterion considers not only the test pass rate but also the sample size and statistical fluctuations, providing a basis for judging whether the testing system is representative.

[0054] The above method, through a star-by-star update approach, allows the test validity indicators to dynamically change with the production progress. In the initial stage, due to the limited sample size, the indicators fluctuate significantly; as production continues and data gradually accumulates, the indicators tend to stabilize, providing a more accurate reflection of the product quality status. Simultaneously, this method considers both historical data and the current state, ensuring that the evaluation results are both continuous and timely reflect changing trends.

[0055] From an overall structural perspective, the construction of the test effectiveness index represents a shift from single-test results to statistical inference across the entire batch. By introducing a Bayesian update mechanism, discrete test results can be transformed into continuous probabilistic evaluation indicators, making them comparable across different production stages. Furthermore, the introduction of confidence intervals effectively controls risk and avoids misjudgments due to insufficient samples.

[0056] In summary, this evaluation method is highly adaptable to mass production environments, providing high statistical reliability while maintaining computational simplicity. By continuously accumulating data and dynamically updating indicators, it can provide a stable basis for quality control decisions and offer quantitative support for subsequent process optimization and experimental strategy adjustments.

[0057] In the production and quality control of mass-produced satellites, in addition to statistical analysis of manufacturing process parameters and product performance, in-depth evaluation is also needed from two aspects: process execution stability and the degree of fault handling closure. This further reflects the maturity of the production system and its problem-solving capabilities. Process stability describes the consistency and standardization of each process step in the production process, while the fault closure rate measures whether quality problems have been effectively identified, analyzed, and thoroughly resolved. These two types of indicators complement each other and together constitute the key evaluation dimensions of production quality management capabilities.

[0058] In evaluating the process stability, it is necessary to review each and every production step of each satellite. Assume that the production process of a single satellite involves... These processes encompass multiple stages, including assembly, testing, debugging, and inspection. During actual execution, deviations may occur in some processes, including but not limited to parameters exceeding specifications, rework, or acceptance with concessions under conditions deviating from standards. These deviations are collectively referred to as process deviations, and the corresponding number of processes is denoted as . .

[0059] Based on the above statistical results, a process stability index can be constructed, which is expressed as follows: in, This represents a process stability index, with a value ranging from 0 to 1. Indicates the number of processes where process deviations occurred; This represents the total number of processes. This indicator reflects the percentage of processes that maintain standardized execution across all processes; when all processes are without deviation, this indicates a satisfactory outcome. ,at this time This indicates that the process is executed completely stably; when the number of deviation processes increases, the index value decreases, reflecting the existence of unstable factors in the production process.

[0060] Because the evaluation of a single satellite may be affected by random factors, a comprehensive analysis of multiple samples from continuous production is necessary to improve the reliability of the evaluation results. Typically, the five most recently produced satellites are selected, their respective process stability indices are calculated, and the average value is taken as the process stability evaluation result for the current stage. This approach smooths out short-term fluctuations and makes the indicators more representative. Simultaneously, thresholds are set to identify anomalies. When the process stability index of any satellite falls below 0.90, it indicates that there may be significant process fluctuations in that batch of production, requiring attention and root cause analysis, such as unreasonable process parameter settings, inadequate adherence to operating procedures, or unstable equipment status.

[0061] Based on the completed process stability analysis, a systematic evaluation of fault handling is also required. Various faults are inevitable during production and testing, and the quality of fault handling directly affects product reliability and the stability of subsequent batches. Therefore, a quantitative analysis of fault closure is necessary. Let the total number of faults recorded up to the current stage be... The number of faults that have completed the full processing flow is Here, completing the closed loop means that the fault has been located and analyzed, the mechanism has been confirmed, and the corrective measures have been verified, and the root cause of the problem has been effectively eliminated.

[0062] At the same time, it is also necessary to count the number of duplicate faults, denoted as . Repeated failures refer to the occurrence of the same failure mode multiple times on different satellites. Their occurrence indicates that existing corrective measures have failed to completely eliminate the root cause. The presence of repeated failures can significantly impact overall quality stability, therefore, they require penalty treatment during the evaluation process.

[0063] Taking into account the above factors, a fault closure rate index is constructed, which is expressed as follows: in, This represents the fault closure rate metric. This indicates the number of faults that have been processed in a closed loop. Indicates the total number of faults; This indicates the number of repeating faults. The formula consists of two parts: the first part... This reflects the degree of completion of fault handling, specifically the percentage of all faults that have achieved closed-loop management; Part Two This item is used to reflect the impact of recurring failures on overall quality. When the proportion of recurring failures is high, this value decreases, thereby reducing the overall evaluation result.

[0064] This indicator has a clear physical meaning. It is valid when all faults have been closed in a loop and there are no recurring faults. 0, at this time A positive result indicates a sound fault management system and thorough problem handling. A decrease in the indicator value when there are unclosed or recurring faults reflects deficiencies in the fault handling process. In particular, when the proportion of recurring faults is high, even with a large number of closed loops, the overall evaluation will be significantly affected, suggesting a need to focus on the effectiveness of root cause analysis.

[0065] Using the methods described above, production quality management capabilities can be comprehensively evaluated from two dimensions. The process stability index focuses on the standardization and consistency of production execution, reflecting the stability of the process system; the failure closure rate index focuses on problem-solving capabilities, reflecting whether the quality management system possesses continuous improvement capabilities. Combining the two provides a comprehensive characterization of the production system's preventative and corrective capabilities.

[0066] In practical applications, these two types of indicators can be used in conjunction with other quality evaluation indicators. For example, when process stability is high but the failure closure rate is low, it may indicate that although the current production process is stable, historical problems have not been completely resolved; conversely, when the closure rate is high but process stability is low, it indicates that problems can be repaired in a timely manner, but the production process itself still has significant fluctuations. By jointly analyzing these two types of indicators, the source of quality problems can be more accurately located, guiding subsequent optimization directions.

[0067] Overall, by quantitatively modeling process deviations and fault handling, the quality management process that originally relied on experience-based judgment can be transformed into a data-driven objective evaluation system, thereby improving the quality control level and the scientific nature of decision-making in the mass production of satellites.

[0068] In the full life-cycle quality evaluation system for mass-produced satellites, data from the on-orbit operation phase is crucial, as it accurately reflects the product's performance and reliability in the actual space environment. By systematically collecting and structuring on-orbit operational data, reliability evaluation indicators oriented towards actual service conditions can be established, providing feedback for ground manufacturing and testing strategies. On-orbit data processing revolves around several key dimensions, including operational time statistics, fault information analysis, recurring fault identification, fault occurrence stage classification, and performance degradation trend monitoring. These dimensions complement each other, collectively forming a complete on-orbit reliability analysis system.

[0069] Regarding operational time statistics, the cumulative on-orbit operational time of each satellite since its launch is continuously recorded via a telemetry system. Let the cumulative on-orbit operational time of the i-th satellite be . By summing up all the satellites in orbit, the total working time can be obtained: in, It represents the cumulative operating time of all satellites in orbit and is an important indicator for measuring the overall operational scale of the system; This represents the runtime of a single satellite. This time parameter serves as the basis for calculating the mean time between failures (MTBF) in subsequent reliability analysis, reflecting the time scale of the system in actual operation.

[0070] In terms of fault information processing, each on-orbit fault needs to be recorded in detail and archived in a structured manner. The record includes the fault number, satellite number, fault occurrence time (expressed as cumulative time since orbit insertion), fault phenomenon description, fault mode classification, fault location level, and fault impact level. Fault mode classification is usually based on a pre-established failure mode library to ensure the comparability of data between different batches and different satellites. By summarizing the fault records of all on-orbit satellites, the cumulative number of faults can be obtained, denoted as [missing information]. This parameter directly reflects the failure frequency of the system in the actual operating environment and is the core data source for reliability evaluation.

[0071] In the analysis of recurring faults, all faults are classified and statistically analyzed according to their fault modes to identify the recurrence of the same fault mode across different satellites. Let the number of recurring faults be... The recurrence rate is defined as the cumulative number of failure modes occurring two or more times. This statistic can be used to identify systemic problems; when a failure mode repeatedly occurs, it usually means that the related design or process issues have not been fundamentally resolved. The failure recurrence rate is further defined as: in, This is used to characterize the proportion of recurring faults among all faults. A threshold is set. When the proportion of repeated failures exceeds this threshold, it indicates that the system has a high structural risk and requires adjustments to production and testing strategies, such as resuming full testing or strengthening screening efforts.

[0072] In the analysis of failure occurrence stages, failures are divided into different stages according to their occurrence time to identify different types of effective characteristics. The early failure stage is defined as 0 to 2160 hours after orbit insertion, the accidental failure stage as 2160 to 43200 hours, and the attrition failure stage as after 43200 hours. The number of failures in each of the three stages is counted and denoted as follows: , and By analyzing the failure ratio at each stage, it is possible to determine whether the system has early defects or lifespan-related issues. When the following conditions are met... If the early failure rate is high, it usually indicates a deficiency in the production or screening process, requiring efforts to reduce the probability of early failures by strengthening environmental stress screening or optimizing process parameters.

[0073] In monitoring performance degradation trends, key performance parameters are tracked over a long period, such as the output power of the solar array, the gain of the communication link, and the performance of the propulsion system. By analyzing the changes of these parameters over time, their degradation rate can be calculated, denoted as [missing information]. The degradation rate is defined as the absolute comparison of performance changes per unit time. It is typically normalized to an annual value and used to measure the long-term performance stability of a system. A degradation rate threshold is set. This threshold corresponds to 60% of the allowable degradation during the design life, converted to an annual average rate. When the following conditions are met... If this occurs, it indicates that the performance degradation rate is exceeding expectations, requiring adjustments to the relevant design or ground testing to improve the product's reliability margin.

[0074] After completing the above multidimensional data analysis, the mean time between failures (MTBF) can be calculated based on the cumulative running time and the total number of failures, as shown in the following expression: in, It represents the actual mean time between failures, reflecting the system's reliability level in a real operating environment; Cumulative running time; This represents the cumulative number of failures. The higher this indicator is, the lower the probability of a failure occurring per unit of time, and the higher the system reliability.

[0075] To further evaluate the degree of matching between actual reliability and design objectives, an on-orbit reliability index is introduced, the expression of which is as follows: in, This represents the on-orbit reliability evaluation index; The value represents the target mean time between failures (MTBF) specified during the design phase; the fractional part represents the ratio between actual performance and the design target; the minimum function is used to limit the upper limit of the indicator to 1.0, thereby preventing the evaluation value from losing its distinguishing meaning when it exceeds the design target. When the actual reliability meets or exceeds the design requirements, the indicator value is 1; when the actual reliability is lower than the design target, the indicator decreases proportionally, reflecting the performance gap.

[0076] It is necessary to consider the situation in the early stages of production, when on-orbit data has not yet been fully accumulated. When the number of satellites in orbit is small or the cumulative operating time is less than 2160 hours, the statistical results may have significant uncertainties. In this case, this indicator can be temporarily excluded from the comprehensive evaluation system, and its weight can be adjusted to avoid unstable impacts on the overall evaluation.

[0077] The above methods enable a systematic analysis of on-orbit operational status, transforming fragmented telemetry data, fault records, and performance change information into unified quantitative indicators. These indicators not only reflect the current reliability level of the system but also reveal potential risk trends and provide closed-loop feedback for ground manufacturing, testing, and quality control. The overall approach represents a shift from single-fault statistics to multi-dimensional comprehensive analysis, fully leveraging the role of on-orbit data in the quality evaluation system.

[0078] In the mass production satellite quality evaluation system, to achieve a unified quantitative expression of multi-dimensional quality information, it is necessary to integrate multiple evaluation indicators from different sources to form a comprehensive indicator that can fully reflect the system's technological maturity. This comprehensive indicator is based on six dimensions: test effectiveness, process consistency, product consistency, process stability, fault closure rate, and on-orbit reliability. It is integrated through a weighted method, which can reflect both the contribution of each dimension to the overall quality and the dynamic changes in the focus of quality control at different stages.

[0079] The calculation format for the Integrated Technology Maturity Index is as follows: in, The Comprehensive Technology Maturity Index (CTMI) is the core quantitative result ultimately used to evaluate the overall quality level of the system. Indicators representing test effectiveness reflect the testing system's ability to detect defects and the product's pass rate. Indicators representing process consistency are used to describe the stability of the manufacturing process; This indicates a product consistency index, used to reflect the consistency of the distribution of key characteristics across different batches of products. This represents a process stability index, used to measure the consistency of production process execution; The fault closure rate is an indicator used to evaluate the completeness and effectiveness of fault handling. This indicates on-orbit reliability, which reflects the reliability performance of a product in a real-world operating environment.

[0080] parameter to These represent the weight coefficients corresponding to the six indicators mentioned above, used to describe the importance of each indicator in the comprehensive evaluation. The weights satisfy the following constraints: And each weight satisfies the non-negativity condition, that is, each This constraint ensures that the comprehensive index is a weighted average of the individual sub-indicators, giving the result a clear physical meaning and making the contribution ratio between different indicators controllable.

[0081] Because the quality control priorities of mass-produced satellites differ at different development stages, the weight allocation needs to be dynamically adjusted based on the accumulation of on-orbit data to achieve stage adaptability of the evaluation system. In the initial stage, when there is no on-orbit data or insufficient on-orbit data, the evaluation mainly relies on ground data, including test results, manufacturing processes, and product characteristics. At this time, on-orbit reliability indicators cannot provide effective support. Therefore, in this stage, the weight of on-orbit reliability is set to zero, and the remaining weights focus on ground quality indicators. The specific weight configuration is as follows: This is used to emphasize the importance of test effectiveness in the initial stages; This is used to reflect the stability of the manufacturing process; This is used to reflect the product's characteristics—consistency; This is used to measure the stability of process execution; This is used to reflect fault handling capabilities; This indicates that on-orbit reliability is not currently included in the evaluation.

[0082] As the number of satellites in orbit gradually increases and the cumulative on-orbit time exceeds a certain period, the on-orbit data begins to have statistical significance. At this point, the evaluation system needs to introduce on-orbit reliability indicators while appropriately reducing its dependence on ground data. In this stage, the weights are adjusted as follows: The weight of test validity has decreased slightly, but it still occupies an important position; The process consistency weight should be appropriately reduced; Product consistency weights are adjusted accordingly. The weight of process stability is further reduced; The fault closure rate remains unchanged; On-orbit reliability has begun to be included in the comprehensive evaluation.

[0083] When the number of satellites in orbit reaches a significant scale and the cumulative on-orbit time is long, the on-orbit data can fully reflect the long-term operational status of the product. At this point, the evaluation system should further increase the weight of on-orbit reliability, making it a core evaluation criterion. In this stage, the weight allocation is adjusted as follows: The weight of test validity continues to decline; The process consistency weight is further reduced; The weight of product consistency is reduced accordingly; The weight of process stability remains at a low level; The fault closure rate remained stable; The weight of on-orbit reliability has been significantly increased.

[0084] The aforementioned weighting adjustment strategy reflects a gradual transition from relying on ground verification to relying on on-orbit performance, enabling the evaluation system to adapt to the quality control needs of different development stages. In the initial stage, due to the lack of on-orbit data, testing and manufacturing data are mainly relied upon; in the mid-stage, on-orbit data is introduced to assist in judgment; in the later stage, on-orbit data becomes the dominant factor, thereby achieving the dynamic evolution of the evaluation system.

[0085] In engineering practice, while the above weighting configuration has general applicability, it can still be appropriately adjusted according to the specific characteristics of the mission. For example, different types of satellites differ in mission cycles, reliability requirements, and production schedules, so the weights can be fine-tuned according to actual needs. However, to ensure the stability and consistency of the evaluation system, the adjustment range of a single weight is usually no more than ±0.05 to avoid drastic fluctuations in the evaluation results.

[0086] Based on the above comprehensive indicators, a tiered judgment mechanism can be constructed to classify technology maturity into different levels, and experimental tailoring strategies can be formulated accordingly. A high overall technology maturity indicates stable production processes, good product consistency, an effective testing system, and on-orbit performance meeting expectations. In this case, repetitive screening tests can be appropriately reduced, thereby lowering costs and improving production efficiency. When the following conditions are met: This indicates that the overall quality level is relatively high, at which point environmental stress screening tests can be drastically reduced. This reduction typically involves extending the test interval or increasing the interval level, allowing some products to skip certain repetitive testing stages. This strategy is suitable for the stable production phase, helping to improve production line cycle time while reducing resource consumption within a controllable risk range. When the overall technology maturity is in the middle range, it indicates that the overall system quality is within an acceptable range, but some uncertainty still exists. At this point, it is not advisable to completely relax testing requirements; instead, a limited reduction strategy should be adopted. When the following conditions are met: This indicates a moderate system quality level. At this stage, screening tests can be adjusted appropriately, such as reducing the frequency of test execution or extending the interval for the number of consecutive passes required for upgrades, thereby gradually optimizing test resource allocation while controlling risks. This stage typically corresponds to a phase where production stability is gradually improving but not yet fully validated, requiring a balance between efficiency and reliability. When the overall technology maturity index is below the preset threshold, it indicates significant instability in the system, potentially stemming from manufacturing process fluctuations, product performance differences, or inadequate fault handling. In this case, a strict testing strategy should be maintained to ensure product quality remains unaffected. When the following conditions are met: This indicates that the current quality level is insufficient to support experimental tailoring, and the full inspection mode should continue to be implemented, with a complete testing and screening process for each satellite to minimize potential risks.

[0087] Building upon the aforementioned tiered decision-making mechanism, a dynamic linkage mechanism needs to be introduced to allow the on-orbit operational status to influence ground testing strategies in real time. Since on-orbit reliability indicators directly reflect the product's performance in a real environment, a decline in this indicator indicates a deviation between actual operational status and expectations, necessitating timely adjustments to production and testing strategies. In the calculation of comprehensive indicators, the on-orbit reliability index... As an important component, its changes will directly affect the overall evaluation results.

[0088] When the on-orbit failure rate increases or performance degradation accelerates... This will decrease, leading to The decline should be automatic. When the decline reaches a certain level and touches the grading threshold, the testing strategy tightening mechanism should be triggered automatically. For example, when a production mode that was originally in a cut-off state enters a lower grade range due to a decline in indicators, some or all screening tests should be resumed to prevent further expansion of quality risks.

[0089] To quantify this linkage, an adjustment rule based on the magnitude of changes in on-orbit reliability can be established. When on-orbit reliability indicators decline, their impact on the testing strategy is proportionally transmitted. For example, when... For every 0.1 decrease, the interval level for all environmental stress screening tests is regressed by one level. This rule reflects the direct feedback effect of on-orbit data on ground quality control, thus forming a closed loop in the quality control system.

[0090] In practice, if on-orbit reliability continues to decline, multi-level rollback may be triggered. For example, when the cumulative decline reaches 0.2 or 0.3, a rollback of level 2 or 3 may be triggered, thereby gradually restoring a more stringent test coverage. This mechanism can effectively prevent quality control issues caused by over-pruning, enabling the system to respond quickly and adjust its strategy when abnormal trends occur.

[0091] Through the aforementioned hierarchical decision-making and dynamic linkage mechanism, the experimental strategy can be transformed from static configuration to dynamic optimization. The comprehensive technology maturity index provides a unified quantitative basis, enabling data from different sources to be transformed into executable decision rules; the hierarchical judgment mechanism provides clear operational guidance for complex evaluation results; and the dynamic linkage mechanism ensures that the on-orbit operational status can be fed back to the ground production system in a timely manner, thereby forming a closed-loop control.

[0092] Overall, this method achieves close coupling between quality evaluation and experimental decision-making, enabling mass-produced satellites to improve production efficiency while ensuring reliability, and continuously optimizes quality control strategies through a data-driven approach.

[0093] To illustrate the specific application of this evaluation method, a low-Earth orbit communication constellation is selected as an example. This constellation is planned to deploy 1000 satellites, each weighing approximately 300 kg, with a designed lifespan of 5 years (43,800 hours) and a production rate of 10 satellites per month.

[0094] First, a three-tiered evaluation framework for quality consistency is established to comprehensively evaluate batch-produced satellites from three levels: process consistency, product consistency, and reliability consistency.

[0095] In the process consistency evaluation, the key manufacturing process parameters are selected as shown in the table below: Table 1: Process Consistency Monitoring Parameters In product consistency evaluation, the key characteristics and their classifications are as follows: Table 2: Key Product Characteristics In process consistency calculations, taking assembly torque as an example, the statistical results for the first 30 satellites are as follows: mean. Standard deviation Given a maximum specification limit (USL) of 12.0 and a minimum specification limit (LSL) of 8.0, the process capability index is calculated as follows: This result indicates that the process capability of this parameter is below 1.33, suggesting room for improvement. The remaining parameters... The values ​​are 1.05, 1.42, 1.56, 1.38, and 1.51 respectively, therefore the minimum value is: In exponentially weighted moving average analysis, the smoothing coefficient is taken. Assuming all parameters remain within their limits, the process consistency index is: In the product consistency evaluation, taking the total power consumption of the satellite as an example, the average power consumption of the first batch (#1, #15) was 498.2W, with a standard deviation of 18.5W; the average power consumption of the second batch (#16, #30) was 502.6W, with a standard deviation of 16.8W. The KS test was used to calculate: The critical value is: Since 0.133 < 0.496, this property passes the test. Assuming that one of the three permissible distribution properties fails, then: In the test validity calculation, 28 out of the first 30 satellites passed without defects, therefore: In the process stability calculation, the process deviations of the most recent 5 satellites are 2, 1, 3, 0, and 1 respectively, with a total of 120 processes. Therefore, the average process stability is: In the calculation of the fault closure rate, the total number of faults is 8, the closure rate is 7, and the repeated fault occurs once. Therefore: In the on-orbit reliability calculation, there are currently 20 satellites in orbit, with a cumulative operating time of 72,000 hours and 2 failures. Therefore: In the comprehensive evaluation, stage B weighting is adopted: The overall technology maturity level is: Based on the evaluation results, It falls within the medium maturity range, allowing for a limited pruning strategy. Based on the results analysis, the main limiting factors are: Insufficient torque capacity during assembly; The thermal control margin is inconsistent.

[0096] Therefore, it is crucial to optimize the consistency between assembly process control and thermal control design. Once these indicators are improved, the overall index will further increase to above 0.90, thus entering a high maturity state and enabling more extensive experimental tailoring.

[0097] This invention establishes a three-tiered quality consistency evaluation framework encompassing process consistency, product consistency, and reliability consistency. It systematically characterizes the quality of mass-produced satellites from three dimensions: manufacturing process parameter control, distribution of key product characteristics, and on-orbit performance. Compared to traditional methods that rely solely on statistical process control for manufacturing stability assessment, this framework extends from the process dimension to the entire process-product-application chain. This allows the evaluation results to not only reflect the production status but also reveal the actual performance and reliability levels of the product. This effectively overcomes the problems of single evaluation dimensions and one-sided results in existing technologies, significantly improving the comprehensiveness and accuracy of quality evaluation.

[0098] This invention introduces an on-orbit reliability index, directly incorporating on-orbit operational data into the technology maturity assessment system, thus realizing a dynamic closed-loop linkage mechanism between ground evaluation results and actual on-orbit performance. When the on-orbit failure rate increases or the performance degradation trend intensifies, the on-orbit reliability index decreases, which in turn leads to a synchronous decrease in the comprehensive technology maturity index, automatically triggering a tightening and adjustment of ground test strategies. This mechanism transforms quality control from traditional one-way evaluation to real-time feedback regulation, effectively compensating for the disconnect between ground evaluation and on-orbit operation in existing technologies, and improving the system's responsiveness to abnormal trends and risk prevention capabilities.

[0099] This invention constructs a six-dimensional comprehensive technology maturity index system, encompassing test effectiveness, process consistency, product consistency, process stability, fault closure rate, and on-orbit reliability. It achieves multi-source data fusion through weight allocation. Furthermore, based on the accumulation of on-orbit data, the weights are dynamically adjusted in three stages (A, B, and C), enabling the evaluation system to automatically adapt to different quality control priorities throughout the product lifecycle. In the early stages, ground data is emphasized, while the weight of on-orbit data is gradually increased in the mid-to-late stages. This provides a scientific, continuous, and quantifiable basis for experimental tailoring decisions, avoiding the uncertainty caused by empirical judgment.

[0100] In this invention, all evaluation indicators are based on clearly defined data sources and, combined with rigorously defined calculation methods and judgment rules, achieve standardization and repeatability of the quality evaluation process. Through a unified data structure, indicator model, and calculation process, complex quality information can be transformed into quantifiable and comparable evaluation results, supporting continuous iterative optimization. Simultaneously, this method possesses good engineering feasibility and can be directly embedded into production line information systems to form automated evaluation tools, enabling real-time monitoring, early warning analysis, and decision support, significantly improving the quality management level and intelligence of mass-produced satellite manufacturing processes.

[0101] This invention provides, for example Figure 2The illustrated integrated technology maturity assessment system for mass-produced satellites includes a process consistency assessment module, a product consistency assessment module, a test effectiveness assessment module, a process and fault management module, an on-orbit data linkage module, and an integrated maturity calculation and decision-making module. The process consistency evaluation module is used to interact with the manufacturing execution system, automatically collect measured data of manufacturing process parameters at each production station, perform statistical analysis on the manufacturing process parameters, calculate the corresponding process capability indicators and process trend statistics, and determine the process status based on preset control limits, and output the process consistency evaluation results; when the capability level of any manufacturing process parameter is lower than the set threshold or the process trend statistics exceed the limit, a process anomaly warning message is generated. The product consistency evaluation module is used to extract key characteristic data of different batches of satellites from the testing database, classify the key characteristics, perform cross-batch distribution consistency tests and multivariate joint distribution analysis for permissible distribution characteristics, and output product consistency evaluation results; when there are cases where the distribution test does not meet the consistency requirements, the corresponding batch is marked as an anomaly and recorded. The test effectiveness evaluation module is used to obtain test and experiment result data of each satellite from the test integration management system, statistically model the historical test pass status, and update the test effectiveness indicators batch by batch or satellite by satellite based on the Bayesian update mechanism. At the same time, it outputs the corresponding confidence interval range to reflect the stability and reliability of the test system. The process and fault management module is used to obtain process execution records and fault handling records from the quality management system, perform statistical analysis on process deviations to obtain process stability evaluation results, and perform comprehensive analysis on fault handling closure and recurring faults to output fault closure rate evaluation results. The on-orbit data linkage module is used to receive on-orbit satellite telemetry data through the telemetry and control interface, perform structured processing on-orbit operation data, including operation time statistics, fault information classification, recurring fault identification, fault occurrence stage division and performance degradation trend analysis, and calculate on-orbit reliability evaluation indicators based on the processing results; when the proportion of recurring faults or the performance degradation rate exceeds the preset threshold, on-orbit abnormal alarm information is generated and output. The comprehensive technology maturity calculation and decision-making module receives the process consistency evaluation results, product consistency evaluation results, test effectiveness evaluation results, process stability evaluation results, fault closure rate evaluation results, and on-orbit reliability evaluation results output by each evaluation module. Based on the current on-orbit data accumulation stage, it automatically matches the corresponding weight parameters, performs weighted fusion calculation on each evaluation result to obtain the comprehensive technology maturity index, and judges the comprehensive technology maturity based on the preset grading threshold. It outputs test tailoring level suggestions and the current quality risk status. At the same time, when the on-orbit reliability evaluation result declines, it adjusts the comprehensive technology maturity index in a coordinated manner to trigger the tightening of test strategies.

[0102] The present invention provides a method for evaluating the overall technological maturity of mass-produced satellites, which is implemented through the aforementioned evaluation system for the overall technological maturity of mass-produced satellites. For details of the specific method and process of the evaluation system for the overall technological maturity of mass-produced satellites, please refer to the embodiment of the above-mentioned method for evaluating the overall technological maturity of mass-produced satellites, which will not be repeated here.

[0103] The foregoing has only described certain exemplary embodiments of the present invention by way of illustration. Undoubtedly, those skilled in the art can modify the described embodiments in various ways without departing from the spirit and scope of the present invention. Therefore, the foregoing drawings and descriptions are illustrative in nature and should not be construed as limiting the scope of protection of the claims of the present invention.

Claims

1. A method for evaluating the overall technological maturity of mass-produced satellites, characterized in that, Includes the following steps: Establish a quality consistency evaluation framework to divide the quality consistency of batch-produced satellites into three levels: process consistency, product consistency, and reliability consistency. Statistical analysis is performed based on manufacturing process parameters to obtain process capability levels and monitor process fluctuation trends to comprehensively determine process consistency indicators. Based on the key characteristics of the products, they are classified and cross-batch statistical distribution consistency tests are carried out. The product consistency indicators are determined by combining the results of multivariate statistical analysis. By combining the satellite data from completed tests and experiments, statistical inferences are made to obtain dynamically updated test effectiveness indicators; Based on the manufacturing process execution records and fault handling records, statistical analysis was conducted to obtain process stability index and fault closure rate index, respectively. Based on on-orbit operation data, the on-orbit working time, fault conditions, fault recurrence, fault occurrence stage, and performance degradation trend are analyzed to obtain on-orbit reliability indicators. The comprehensive technology maturity index is obtained by weighting and integrating process consistency index, product consistency index, test effectiveness index, process stability index, fault closure rate index and on-orbit reliability index, and the weights are adjusted in stages as on-orbit data accumulates. The test strategy is adjusted based on the comprehensive technology maturity index, and the comprehensive technology maturity index is adjusted in conjunction with the on-orbit reliability index to tighten the test strategy when the on-orbit reliability index declines.

2. The method for evaluating the integrated technology maturity of mass-produced satellites according to claim 1, characterized in that, When monitoring the trend of manufacturing process parameters, the exponential weighted moving average method is used, with the smoothing coefficient λ ranging from 0.1 to 0.3 and the control limit width coefficient ranging from 2.7 to 3.

0. In the process consistency evaluation, when the minimum value of the process capability index of all manufacturing process parameters is lower than 1.0, the process consistency index is directly judged as zero to indicate the insufficient manufacturing process capability.

3. The method for evaluating the integrated technology maturity of mass-produced satellites according to claim 1, characterized in that, When verifying the consistency of key product characteristics, a two-sample distribution test method based on the difference of empirical distribution function is adopted, and the judgment is made at a significance level of 0.

05. At the same time, multiple key characteristic parameters are constructed into multidimensional vectors for joint distribution analysis. By calculating multivariate statistics and converting them into judgment statistics under the corresponding distribution, the overall distribution consistency between different batches is judged.

4. The method for evaluating the integrated technology maturity of mass-produced satellites according to claim 1, characterized in that, During the on-orbit operation data analysis, when the statistically obtained recurring failure rate exceeds 0.3, a comprehensive re-inspection will be carried out on subsequent batches of products. When the early failure rate exceeds 0.4, the environmental stress screening intensity is increased; when the performance degradation rate exceeds the preset threshold, the pass criteria for ground tests are adjusted to strengthen the screening standards.

5. The method for evaluating the integrated technology maturity of mass-produced satellites according to claim 1, characterized in that, In the calculation of the comprehensive technology maturity index, the weights of each evaluation index are adjusted in stages according to the accumulation of on-orbit data. In the stage where there is no on-orbit data, the weights related to on-orbit reliability are set to zero and the weight of test effectiveness is increased. In the initial accumulation stage, the weight of on-orbit reliability is introduced and the weight of ground evaluation is appropriately reduced. In the full accumulation stage, the weight of on-orbit reliability is further increased and the weights of other indicators are reduced, so that the weight allocation changes dynamically with the maturity of data.

6. The method for evaluating the integrated technology maturity of mass-produced satellites according to claim 1, characterized in that, In the product consistency evaluation process, a change impact assessment mechanism is introduced. When design changes, process changes, or supply chain changes occur, key characteristic data of multiple samples before and after the change are collected, and distribution consistency tests or mean difference tests are performed. When the test results show that there are significant differences, the relevant test items are resumed to ensure the consistency of the product after the change.

7. The method for evaluating the integrated technology maturity of mass-produced satellites according to claim 1, characterized in that, In the process of comprehensive technology maturity assessment and test tailoring, a dynamic linkage rule is established between on-orbit reliability indicators and test strategies. When the on-orbit reliability indicators decline, the comprehensive technology maturity indicators are reduced synchronously according to a preset ratio, and the interval levels of environmental stress screening tests are gradually rolled back. Specifically, whenever the on-orbit reliability indicators decline by a certain amount, the corresponding level is rolled back by one level.

8. The method for evaluating the integrated technology maturity of mass-produced satellites according to claim 1, characterized in that, During the evaluation of test effectiveness, the test effectiveness indicators are updated batch by batch, and the lower limit of the confidence interval is introduced as the judgment criterion. When the lower limit of the confidence interval reaches the preset threshold, it is determined that the test stability of the current batch of products meets the requirements; otherwise, the original test intensity remains unchanged.

9. The method for evaluating the integrated technology maturity of mass-produced satellites according to claim 1, characterized in that, In the joint evaluation of process consistency and product consistency, a correlation is established between key manufacturing process parameters and corresponding key product characteristics. When abnormal fluctuations in process parameters are detected and the distribution of corresponding product characteristics is deviated, targeted reviews are performed on the relevant processes and the corresponding batches of products are marked.

10. A mass-production satellite integrated technology maturity assessment system, used to implement the mass-production satellite integrated technology maturity assessment method according to any one of claims 1-9, characterized in that, It includes a process consistency evaluation module, a product consistency evaluation module, a test effectiveness evaluation module, a process and fault management module, an on-orbit data linkage module, and a comprehensive maturity calculation and decision-making module; The process consistency evaluation module connects to the manufacturing execution system to collect manufacturing process parameter data, performs statistical analysis and calculates process capability indicators and trend statistics, determines the process status based on control limits, outputs process consistency evaluation results and generates early warnings when anomalies occur. The product consistency evaluation module extracts key characteristic data of batch products from the testing database, performs classification processing, executes distribution consistency test and multivariate analysis, outputs product consistency evaluation results and identifies abnormal batches; The test effectiveness evaluation module obtains test result data from the test integration management system, performs statistical modeling, and updates the test effectiveness indicators and their confidence intervals based on Bayesian analysis. The process and fault management module obtains process records and fault data from the quality management system, performs statistical analysis, and obtains evaluation results on process stability and fault closure rate. The on-orbit data linkage module receives on-orbit data through the telemetry and control interface, performs structured processing, calculates on-orbit reliability indicators, and generates alarms when anomalies occur. The comprehensive maturity calculation and decision-making module receives the evaluation results and performs weighted calculations according to the stage weights to obtain the comprehensive technology maturity index. It then determines the level based on the grading thresholds and outputs the trimming level.

Images