Unmanned ship multi-stage task reliability assessment method based on given confidence coefficient

Abstract

The invention provides an unmanned ship multi-stage task reliability assessment method based on a given confidence coefficient, and relates to the technical field of unmanned ship reliability analysis, and the method comprises the steps: S1, determining system composition equipment, fault distribution and a series-parallel connection relation; s2, multi-level data fusion; s3, calculating the confidence coefficient of the single-stage task according to the target confidence coefficient corresponding to the target reliability of the multi-stage task; s4, multi-stage task reliability aggregation under single-stage task confidence is set; and S5, judging whether the reliability evaluation result of the multi-stage task of the unmanned ship meets the reliability requirement under the target confidence degree or not. The invention provides an unmanned ship multi-level data fusion method, and realizes an unmanned ship task reliability verification method based on combination of a small amount of task reliability test data and performance function verification. Through a multi-stage task reliability aggregation technology, a task reliability calculation method of the unmanned surface vehicle under a set confidence coefficient is provided based on task reliability evaluation results of all stages of the unmanned surface vehicle.

Description

Technical Field

[0001] This invention relates to the field of unmanned surface vessel (USV) reliability analysis technology, and specifically to a method for assessing the reliability of a USV in a multi-stage mission based on a given confidence level. Background Technology

[0002] Task reliability refers to the ability of a product to perform its specified functions within a given task profile (i.e., the entire task execution cycle). Its evaluation includes not only the overall stability of the system, but also the role of redundancy design in ensuring task continuity.

[0003] Currently, equipment mission reliability verification mainly relies on internal and external field physical test data conducted in the later stages of engineering development. Its theoretical basis is statistical laws (law of large numbers). Therefore, to obtain an assessment result with a given confidence level, it is necessary to significantly increase the test sample size and test time. During equipment evaluation, mission reliability verification primarily involves statistically analyzing the equipment's operating time and failure frequency during the evaluation process. Based on the definition of a severe failure, the number of severe failures is determined, and the lower confidence limit of the mean severe failure interval (MFR) at a given confidence level is calculated. Then, the lower confidence limit of the equipment's mission reliability for a given mission time is calculated. However, in practical engineering applications, due to constraints in development funding and development cycles, it is impossible to conduct a large number of mission reliability tests. With insufficient sample size, the objectivity and authenticity of the mission reliability verification results are insufficient. Therefore, it is considered to combine a small amount of mission reliability test data with operating time and failure data from performance and functional verification for mission reliability verification. However, the data collected during performance and functional verification is equipment-level data. Therefore, how to fuse equipment-level data to obtain secondary system data, and further obtain primary system data (equipment-level data), is a current technical challenge in equipment mission reliability verification.

[0004] In the field of unmanned surface vessels (USVs), as newly developed equipment, the development of USVs is constrained by research and development funding and timelines, which significantly limits the conduct of mission reliability tests. Manned ship missions typically involve only propulsion and power systems, communication systems, etc., during the departure and return phases, making the navigation mission assessment less challenging and providing ample test data. However, USV navigation missions involve assessments of situational awareness, autonomous route planning, and autonomous obstacle avoidance capabilities. The limited time for actual testing and the scarcity of available test data make it difficult to verify the mission reliability of USVs.

[0005] On the other hand, the mission profile of unmanned surface vessels is usually composed of multiple phases of missions, such as autonomous departure, area monitoring, collaborative operation and autonomous return. Each phase is highly dependent on equipment (such as navigation systems and communication relay equipment), and is affected by the marine environment (wind, waves, salt spray). The distribution of equipment failures is unknown, and traditional mission reliability verification methods are difficult to apply.

[0006] The system components required for each stage of the task differ, and equipment test data can only be used to assess the reliability of equipment for a specific stage of the task. There is existing theoretical research on reliability assessment for multi-stage tasks. Many scholars, considering the characteristics of unmanned and intelligent systems—complex hierarchical structures, logical compositions, and multi-stage tasks—have addressed the state dependency problem of shared equipment between stages based on Markov processes, and have utilized methods such as Dynamic Fault Tree (DFT), Bayesian networks, Generalized Stochastic Petri Nets (GSPN), and Monte Carlo simulation to achieve multi-stage task assessment.

[0007] However, these theoretical methods have limitations: On the one hand, the basic assumption of using current methods for evaluation is that the failure distribution of the system's components is known. However, in actual engineering applications, the distribution of important components of intelligent marine equipment is usually unknown. Although the above methods are theoretically complete, they are often not applicable to the task reliability assessment of current intelligent systems. On the other hand, when calculating the reliability of multi-stage tasks, it is generally assumed that when the confidence level of the reliability of each stage task is the same as the confidence level of the multi-stage task, the reliability of each stage task is obtained by combining the reliability of each stage task. However, this calculation method ignores the impact of simply combining the task reliability on the confidence level when the distribution of each stage task is different. Therefore, current theoretical research is insufficient in terms of the scientific nature of the calculation method and the authenticity of the calculation results. Summary of the Invention

[0008] To address the shortcomings of existing technologies, the present invention aims to provide a multi-stage mission reliability assessment method for unmanned surface vessels (USVs) based on a given confidence level. This method addresses the limitations of USV mission profiles and experimental implementation, which significantly restrict the conduct of mission reliability tests. Furthermore, current theoretical research on multi-stage mission reliability suffers from deficiencies in the scientific rigor of calculation methods and the veracity of settlement results. This invention proposes a method for calculating the mission reliability of equipment at a given confidence level, based on the mission reliability assessment results of each stage of the intelligent equipment using multi-stage mission reliability aggregation technology.

[0009] Specifically, this invention provides a method for assessing the reliability of unmanned surface vessels (USVs) in a multi-stage mission based on a given confidence level, which includes the following steps: S1: Define the system components, fault distribution, and series-parallel relationships: Divide the unmanned surface vessel (USV) system into a power system, autonomous navigation system, control system, identification system, and communication system. Based on the USV's system and equipment composition, and in conjunction with the phased mission reliability model, determine the fault distribution and series-parallel relationships of the USV's equipment. S2: Multi-level data fusion: Classify the fault distribution and series-parallel relationships of unmanned surface vessel equipment, and fuse the equipment's working time and fault data to the system level based on the classification results to obtain the system's working time and fault data. S3: Determine the confidence level for a single-stage task: Based on the target confidence level and target task reliability set for the multi-stage task assessment, determine the confidence level required to assess the reliability of a single-stage task. Calculate the confidence level of the reliability of a multi-stage task : ; in: ; ; ; In the formula, is the reliability variable; M is a constant value selected based on the accuracy requirements of numerical calculation; m is the number of multi-stage tasks; This represents the i-th stage task, in Confidence value under precision operator; Increase the confidence level of the reliability of multi-stage tasks The single-stage task confidence is calculated to be equal to the set target confidence level. S4: Set the reliability aggregation of multi-stage tasks under the single-stage task confidence level: Based on the single-stage task confidence level set in S3, combined with the classification results in S2 and the system's working time and fault data, calculate the reliability confidence lower limit of each stage task; through the multi-stage task reliability aggregation method, merge the reliability evaluation results of each stage task under the single-stage task confidence level to obtain the reliability evaluation results of multi-stage tasks under the target confidence level; S5: Determine whether the evaluation results meet the reliability requirements under the target confidence level: Compare the evaluation results obtained in S4 with the set target mission reliability. If the evaluation results reach the target mission reliability threshold, the unmanned surface vessel passes the mission reliability assessment.

[0010] Furthermore: In S2, the classification of fault distribution and series-parallel relationships of unmanned surface vessel (USV) equipment includes: A data fusion method for a serial system composed of devices with different exponential distributions; A data fusion method for a serial system composed of devices with the same exponential distribution; A data fusion method for a parallel system composed of devices with the same exponential distribution; A data fusion method for a K / N system composed of devices with the same exponential distribution.

[0011] Furthermore: In S4, the method for calculating the lower limit of the reliability confidence level of a series system composed of devices with different exponential distributions includes: When a component of the system malfunctions: ; ; Sub-confidence level is The lower confidence limit for the reliability of the task is: ; When the components of the system are not malfunctioning: ; Confidence level is The lower confidence limit for the reliability of the task is: ; In the formula, Normally distributed quantiles; Indicates the number of devices in the system; Indicates the equivalent operating time of each unit device; This indicates the number of failures in each unit device: Indicates the testing time of the merged system; Indicates the number of failures in the merged system; Indicates the duration of one task; This indicates the number of equivalent tasks.

[0012] Furthermore: In S4, the method for calculating the lower confidence limit of the reliability of a series system composed of devices with the same exponential distribution includes: When a component of the system fails, the sub-confidence level is The lower confidence limit for the reliability of the task is: ; in, ; In the formula, For degrees of freedom and confidence level of Quantiles of the distribution; When no component of the system malfunctions, the sub-confidence level is: The lower confidence limit for the reliability of the task is: ; In the formula, Indicates the test time of the equipment; Indicates the testing time of the merged system; Indicates the number of failures in the merged system; Indicates confidence level; Indicates the duration of one task; Indicates the number of equivalent tasks; This represents the number of equivalent tasks after merging; Indicates the number of times the task failed.

[0013] Furthermore: In S4, the method for calculating the lower confidence limit of the reliability of a parallel system composed of devices with the same exponential distribution includes: The confidence level is The lower confidence limit for the reliability of the task is: ; in, ; In the formula, For degrees of freedom and confidence level of Quantiles of the distribution; When no component of the system malfunctions, the confidence level is: The lower confidence limit for the reliability of the task is: ; In the formula, Indicates the number of devices in the system; Indicates the total testing time of the equipment; Indicates the testing time of the merged system; Indicates the number of failures in the merged system; Indicates confidence level; Indicates the duration of one task; Indicates the number of equivalent tasks; This represents the number of equivalent tasks after merging.

[0014] Furthermore: In S4, the calculation method for the lower confidence limit of the reliability of a K / N system composed of devices with the same exponential distribution includes: When a component of the system malfunctions: The system's MTBF at a sub-confidence level of The lower confidence limit at that time: ; When the components of the system are not malfunctioning: Sub-confidence level is The lower confidence limit for the reliability of the task is: ; In the formula, X represents the number of devices that are functioning normally; Indicates the number of devices included in the system; Indicates the minimum number of working devices; Indicates the total system test time; Indicates the testing time of the merged system; Indicates the number of failures in the merged system; Indicates the duration of one task; Indicates the number of equivalent tasks; Indicates the confidence level.

[0015] Furthermore: In S4, when a multi-stage task consists of m completely identical and independent stage tasks, the i-th stage task has a confidence level of... The lower confidence limit of reliability at that time is Multi-stage tasks in terms of confidence The lower confidence limit of task reliability for: .

[0016] Furthermore: In S4, when a multi-stage task consists of multiple single-stage tasks with not entirely identical distributions and which are independent of each other, the confidence level of each single-stage task... The lower confidence limit for the reliability of the task is as follows: Multi-stage tasks in terms of confidence The lower confidence limit of task reliability for: .

[0017] Furthermore, S4 also includes: calculating the lower confidence limit of the reliability of the first-stage task at the determined confidence level, using the confidence level of the single-stage task determined in S3. ; In the formula, This is the lower confidence limit for the reliability of the first phase of the mission; This represents the total equivalent test time for the first phase of the mission. For the equivalent number of faults, For degrees of freedom is The determined sub-confidence level is time Distribution value.

[0018] Compared with the prior art, the beneficial effects of the present invention are as follows: 1. This invention addresses the complexity of unmanned surface vessel (USV) mission profiles and the limitations of experimental implementation. To resolve the problems in multi-stage mission reliability verification, it proposes a multi-level data fusion method for intelligent systems, realizing a phased mission reliability verification method for intelligent systems based on a combination of limited mission reliability test data and performance and functional verification. On the other hand, this invention proposes a method for calculating the mission reliability of equipment at a given confidence level based on the mission reliability assessment results of each stage of the unmanned surface vessel using multi-stage mission reliability aggregation technology.

[0019] 2. This invention also proposes a multi-stage mission reliability verification method for unmanned surface vessels. By using a multi-level data fusion method for system components, the working time and fault data of the equipment level are fused to obtain the working time and fault data of the secondary system. Then, the secondary system data are fused to obtain the working time and fault data of the primary system, thereby realizing the fusion of equipment-level data to system-level data and obtaining the reliability of each stage of the mission. Based on the reliability and confidence requirements of each stage of the mission, the reliability of the multi-stage mission is aggregated to realize the multi-stage mission reliability verification under a given confidence level. Attached Figure Description

[0020] Figure 1 This is a flowchart of the multi-stage mission reliability assessment method for unmanned surface vessels based on a given confidence level disclosed in this invention. Figure 2 This is a cross-sectional view of a multi-stage operation task of the unmanned surface vessel multi-stage mission reliability assessment method based on a given confidence level disclosed in this invention. Detailed Implementation

[0021] Hereinafter, embodiments of the present invention will be described with reference to the accompanying drawings.

[0022] like Figure 1 As shown, a method for multi-stage mission reliability assessment of unmanned surface vessels based on a given confidence level includes the following steps: S1: Based on the system and equipment composition of the unmanned surface vessel and combined with the phased mission reliability model, clarify the fault distribution and series-parallel relationships of the equipment; S2: Multi-level data fusion; The fault distribution and series-parallel relationships of the equipment can be categorized as follows: A data fusion method for a serial system composed of devices with different exponential distributions; A data fusion method for a serial system composed of devices with the same exponential distribution; A data fusion method for a parallel system composed of devices with the same exponential distribution; A data fusion method for a K / N system composed of devices with the same exponential distribution.

[0023] This further distinguishes between data fusion methods for when a fault occurs and when no fault occurs. Based on the actual situation in engineering applications, an appropriate method is selected to fuse the equipment's operating time and fault data to the system level, thereby obtaining the system's operating time and fault data.

[0024] (1) Data fusion method for a series system composed of devices with different exponential distributions; If the failure rates of the components in a system all follow an exponential distribution, then for a system composed of devices connected in series with different exponential distributions, the system operating time and failure data can be obtained from the following model: ①When none of the system's components are malfunctioning: The total test time and total number of failures of the merged system are calculated using the following formulas. ): ; ② When all components of the system fail: The total test time and total number of failures of the merged system are calculated using the following formulas. ): ; ③ When some components of the system malfunction while others do not: The total test time and total number of failures of the merged system are calculated using the following formulas. ): ; When using the above formulas, the operating time and number of failures of the constituent equipment need to be adjusted according to the failure status of the constituent equipment as follows: Divide the equipment data into two groups (no failure group and failure group): ,remember , ,remember .like That is, it exists and The smallest device will Corresponding Recorded as , and will Compress to The information from other devices should be substituted into the above formula for synthesis; otherwise, the above formula should be used directly for calculation.

[0025] Specifically, the multi-level data fusion methods under different conditions are as follows: The system contains the following number of devices: ; No. The total testing time for each index device is: ; No. The number of failures for each index device is: ; The system test time is: ; The number of system failures is: ; Sub-confidence level: ; (2) Data fusion method for a series system composed of devices with the same exponential distribution; If the failure rates of the components in a system all follow an exponential distribution, then for a system composed of devices connected in series with the same exponential distribution, the system operating time and failure data can be obtained from the following model: ①When a component of the system malfunctions: ; ; ②When none of the system's components malfunction: , ; The total test time and total number of failures of the merged system are calculated using the following formulas. ): ; In the formula, the number of devices included in the system is: ; The testing time for the equipment is: ; The number of equipment failures is: ; The system test time is: ; The number of system failures is: ; Confidence level: ; One task time: ; Equivalent number of tasks: ; Equivalent number of tasks after merging: ; Number of mission failures: .

[0026] (3) Data fusion method for parallel systems composed of devices with the same exponential distribution: If the failure rates of the components in a system all follow an exponential distribution, then for a system composed of parallel devices with the same exponential distribution, the system operating time and failure data can be obtained from the following model: ①When a component of the system malfunctions: ; ; In the formula: , ; ②When none of the system's components malfunction: Sub-confidence The confidence level γ is converted into a success-or-failure type task, and γ is set to 0.99.

[0027] , ; In the formula, ; The total test time and total number of failures of the merged system are calculated using the following formulas. ): ; In the formula, the number of devices included in the system is: ; The total testing time for the equipment is: ; The number of equipment failures is: ; The system test time is: ; The number of system failures is: ; Sub-confidence level: ; One task time: ; Equivalent number of tasks: ; Equivalent number of tasks after merging: ; Number of mission failures: .

[0028] (4) Data fusion method for K / N system composed of devices with the same exponential distribution; If the failure rates of the components in a system all follow an exponential distribution, then for a system composed of N devices with the same exponential distribution, the system is considered normal only when K or more devices are functioning correctly. The system's operating time and failure data can be obtained from the following model: ①When a component of the system malfunctions: ; ; In the formula: , ; ②When none of the system's components malfunction: confidence level The confidence level γ is converted into a success-or-failure type task, and γ is generally taken as 0.99.

[0029] In the formula, .

[0030] The total test time and total number of failures of the merged system are calculated using the following formulas. ): ; In the formula, the number of devices included in the system is: ; The total testing time for the equipment is: ; The number of equipment failures is: ; The system test time is: ; The number of system failures is: ; Sub-confidence level: ; One task time: ; Equivalent number of tasks: ; Equivalent number of tasks after merging: ; Number of mission failures: ; S3: Based on the confidence level requirements corresponding to the target reliability of multi-stage task evaluation, set the confidence level for single-stage task evaluation: Confidence level of reliability of multi-stage tasks It can be calculated using the following formula: ; In the formula: ; ; ; In the formula, For reliability variables; M is a constant value selected based on the accuracy requirements of numerical calculations (M can be 10). 4 10 5 10 6 (etc.); m is the number of multi-stage tasks; This represents the i-th stage task, in Confidence value under precision operator; According to M, we can obtain And then to obtain Thus obtain Finally, the confidence level is obtained. .

[0031] When evaluating the reliability of a multi-stage task, the confidence level of the task reliability is crucial. The value is given; when the given value is... At this point, calculations can be performed using the formula described above, by setting the confidence level for the single-stage task. This improves the confidence level of the multi-stage task reliability obtained by fusing the reliability of each stage of the task. equal .

[0032] The methods for calculating task reliability under different conditions are as follows: (1) Calculation method for the reliability of a series system composed of devices with different exponential distributions: set up , The task reliability is calculated as follows, taking the combined equivalent number of tasks and failure count: ①When a component of the system malfunctions: ; ; Sub-confidence level is The lower confidence limit for the reliability of the task is: ; In the formula, Normally distributed Quantiles.

[0033] ②When the components of the system are not malfunctioning: ; Confidence level is The lower confidence limit for the reliability of the task is: ; In the formula, the number of devices included in the system is: ; The equivalent operating time of each unit is: ; The number of failures in each unit device is : The testing time for the merged system is: ; The number of failures in the merged system is: ; One task time: ; The equivalent number of tasks is: ; Sub-confidence level: .

[0034] (2) Calculation method for the reliability of a series system composed of devices with the same exponential distribution: ①When a component of the system malfunctions: Using the total system test time and total number of failures data ( The system's MTBF was calculated at a confidence level of ), The lower confidence limit at that time: ; In the formula, For degrees of freedom and confidence level of Quantiles of the distribution.

[0035] The confidence level is The lower confidence limit for the reliability of the task is: ; ②When the components of the system are not malfunctioning: Confidence level is The lower confidence limit for the reliability of the task is: ; In the formula, the test time of the equipment is: ; The testing time for the merged system is: ; The number of failures in the merged system is: ; Confidence level: ; One task time: ; Equivalent number of tasks: ; Equivalent number of tasks after merging: ; Number of mission failures: .

[0036] (3) Calculation method for the reliability of parallel systems composed of devices with the same exponential distribution; ①When a component of the system malfunctions: Using the total system test time and total number of failures data ( The system's MTBF was calculated at a sub-confidence level of ), The lower confidence limit at that time: ; In the formula, For degrees of freedom and confidence level of Quantiles of the distribution.

[0037] The confidence level is The lower confidence limit for the reliability of the task is: ; ②When the components of the system are not malfunctioning: Confidence level is The lower confidence limit for the reliability of the task is: ; In the formula, the number of devices included in the system is: ; The total testing time for the equipment is: ; The testing time for the merged system is: ; The number of failures in the merged system is: ; Sub-confidence level: ; One task time: ; Equivalent number of tasks: ; Equivalent number of tasks after merging: .

[0038] (4) Calculation method for the reliability of a K / N system composed of devices with the same exponential distribution: If the failure rates of all components in a system follow an exponential distribution, then for a system composed of N devices with the same exponential distribution, the system is considered normal only when K or more devices are functioning correctly. The system reliability can be calculated as follows: ①When a component of the system malfunctions: Using the total system test time and total number of failures data ( The system's MTBF was calculated at a confidence level of ), The lower confidence limit at that time: ; ②When the components of the system are not malfunctioning: Confidence level is The lower confidence limit for the reliability of the task is: ; In the formula, the number of devices that can work normally is X; The system includes the following number of devices: ; The minimum number of working devices is: ; The total system test time is: ; The testing time for the merged system is: ; The number of failures in the merged system is: ; One task time: ; Equivalent number of tasks: ; Sub-confidence level: ; S4: Set the reliability aggregation of multi-stage tasks under the confidence level of a single-stage task: (1) When a multi-stage task consists of m single-stage tasks with identical and independent distributions, the i-th stage task has a sub-confidence of... The lower confidence limit of reliability at that time is Then the multi-stage task has a certain confidence level The lower confidence limit of task reliability for: ; (2) When a multi-stage task consists of multiple single-stage tasks with not completely identical distributions and which are independent of each other, the confidence level of each stage task is... The lower confidence limit for the reliability of the task is as follows: Multi-stage tasks in terms of confidence The lower confidence limit of task reliability for: ; At this point, the confidence level of the reliability of the multi-stage task. It can be calculated using the following formula: ; In the formula: ; ; ; In the formula, M is a constant value selected according to the accuracy requirements of numerical calculation (M can be 10). 4 10 5 10 6 (etc.), which can be obtained from M. And then to obtain Thus obtain Finally, the confidence level is obtained. .

[0039] Therefore, in order to obtain the target confidence level The reliability of a multi-stage task can be improved by adjusting the confidence level of each stage. , making Then, the given time can be obtained. At that time, target confidence level The target reliability of multi-stage tasks .

[0040] S5: Determine whether the reliability assessment results of the unmanned surface vessel's multi-stage mission meet the reliability requirements under the target confidence level. Based on the single-stage task confidence obtained in step S3 Calculate the confidence level for a single-stage task. The confidence level at that time. When a multi-stage mission consists of two phases: a first phase (i.e., the navigation mission) and a second phase (i.e., the operational mission), the lower confidence limit for the reliability of the navigation phase is determined. for: ; In the formula, This refers to the total equivalent test time during the navigation mission phase. For the equivalent number of faults, For degrees of freedom is Confidence level is The chi-square distribution value at time.

[0041] Further determined by the mission duration of a single flight mission The calculation yielded the navigation mission at a confidence level of The reliability of the task at that time is: ; Similarly, the task phase is obtained with a confidence level of The reliability of the task at that time is: ; in, This refers to the duration of one mission in the first phase of the mission, namely the navigation mission. This refers to the second stage of the task, which is one task time for the assignment.

[0042] (3) Multi-stage task reliability calculation; Based on the above calculation results, we can obtain the confidence level. The reliability of the next multi-stage task is: ; like ≥ Then the ship's equipment meets the requirements at a confidence level. The reliability requirements of the task must be met; otherwise, they will not be satisfied.

[0043] Specifically, to assess whether the multi-stage mission reliability of intelligent ships meets the target confidence level... The task reliability has reached the target reliability. The requirements can be achieved through the following steps: Collect the total equipment testing time provided by manufacturers during the equipment development process, including component manufacturers, finished product suppliers, and final assembly workshops. Number of failures .

[0044] Based on the series and parallel connections of each device, a suitable multi-level data fusion method is selected to calculate the total equipment-level test time. and number of failures The total time for the integrated ship-wide equivalent test and number of failures Collect the vessel's overall working time during mooring trials, sea trials, and naval trials. and number of failures The test data of the entire vessel ( Equivalent whole-ship data after integration with equipment () By combining these data, the total test time for the entire vessel's navigation mission is obtained. and number of failures .

[0045] Since a multi-stage mission consists of navigation and operational tasks, it is necessary to calculate the reliability of the multi-stage mission. At this time, the mission reliability during the navigation phase must be calculated first. and operational phase reliability .

[0046] ; Before calculating task reliability, the confidence level should be determined first, therefore, based on the target confidence level required when evaluating the reliability of a multi-stage task. Calculate the confidence level of mission reliability during the navigation and operational phases.

[0047] Due to the confidence level of multi-stage task reliability It can be calculated using the following formula: ; In the formula: ; ; ; In the formula, M is a constant value selected according to the accuracy requirements of numerical calculation (M can be 10). 4 10 5 10 6 According to M, it can be obtained that... And then to obtain Thus obtain Finally, the confidence level is obtained. .

[0048] Therefore, in the known The value is the target confidence level. At that time, the confidence level of the navigation mission and Operation 3333 mission can be adjusted. This makes the equation hold, thus allowing us to calculate the confidence level of the multi-stage task reliability. = Sub-confidence of mission reliability during the navigation and operational phases .

[0049] The sub-confidence level obtained from the above adjustments Calculate the sub-confidence level. Lower confidence limit of mission reliability during the navigation phase for: ; In the formula, This refers to the total equivalent test time during the navigation mission phase. For the equivalent number of faults, For degrees of freedom is Confidence level is The chi-square distribution value at time.

[0050] Further determined by the mission duration of a single flight mission The calculation yielded the navigation mission at a confidence level of The reliability of the task at that time is: ; Similarly, the task phase is obtained with a confidence level of The reliability of the task at that time is: ; in, This refers to the first phase of the mission, which is one mission duration for the navigation mission. This refers to the second stage of the task, which is one task time for the assignment.

[0051] Based on the above calculation results, we can obtain the confidence level. The reliability of the next multi-stage mission objective is: ; like Then the ship's equipment meets the requirements at a confidence level. The reliability requirements of the task must be met; otherwise, they will not be satisfied.

[0052] Example 1 like Figure 2 As shown, Figure 2 This is a cross-sectional view of a multi-stage mission of an unmanned surface vessel.

[0053] Taking a certain unmanned surface vessel as an example, we conducted verification that the reliability of a multi-stage mission is no less than 0.7 under the requirement of a confidence level of 0.8.

[0054] This multi-stage mission consists of two phases: Phase A (navigation mission) and Phase B (operational mission). According to the mission profile, when an unmanned surface vessel (USV) performs a multi-stage mission, it sails to the designated mission execution area after preparation, then performs the operation within a specified time, and finally returns to the designated area. The duration of the voyage mission includes the departure phase and the return phase.

[0055] Test data from each device was collected, and the device's operating time and fault data were fused to obtain a multi-level fused overall vessel operating time of 2481 hours and an equivalent fault count of 1. Combining the overall vessel operating time and fault data from testing and use, the overall vessel operating time was calculated to be 3291 hours, with a fault count of 1. Relevant data are shown in Table 1. Table 1. Mission Reliability Verification Data To obtain the target reliability of multi-stage tasks at a target confidence level of 0.8, the multi-stage task reliability aggregation method at a given confidence level is used to calculate the stage task confidence level of navigation task A and operation task B as 0.86.

[0056] At a stage mission confidence level of 0.86, the lower confidence limit for the reliability of mission A of this ship is calculated as follows: ; The mission duration for this ship, Task A, is 236 hours, and its reliability is: ; Similarly, the reliability of task B at a confidence level of 0.86 is: ; Therefore, the reliability of this ship in multi-stage missions at a confidence level of 0.8 is: .

[0057] Through data collection, multi-level data fusion, and multi-stage mission reliability aggregation, the lower confidence limit of the multi-stage mission reliability of the vessel was found to be 0.73 at a confidence level of 0.8, which meets the requirement of mission reliability of 0.7. Therefore, the vessel passed the mission reliability assessment.

[0058] The embodiments described above are merely preferred embodiments of the present invention and are not intended to limit the scope of the present invention. Various modifications and improvements made by those skilled in the art to the technical solutions of the present invention without departing from the spirit of the present invention should fall within the protection scope defined by the claims of the present invention.

Claims

1. A method for assessing the reliability of unmanned surface vessels (USVs) in a multi-stage mission based on a given confidence level, characterized in that: It includes the following: S1: Define the system components, fault distribution, and series-parallel relationships: Divide the unmanned surface vessel (USV) system into a power system, autonomous navigation system, control system, identification system, and communication system. Based on the USV's system and equipment composition, and in conjunction with the phased mission reliability model, determine the fault distribution and series-parallel relationships of the USV's equipment. S2: Multi-level data fusion: Classify the fault distribution and series-parallel relationships of unmanned surface vessel equipment, and fuse the equipment's working time and fault data to the system level based on the classification results to obtain the system's working time and fault data. S3: Determine the confidence level for a single-stage task: Based on the target confidence level and target task reliability set for the multi-stage task assessment, determine the confidence level required to assess the reliability of a single-stage task. Calculate the confidence level of the reliability of a multi-stage task : ； in: ； ； ； In the formula, is the reliability variable; M is a constant value selected based on the accuracy requirements of numerical calculation; m is the number of multi-stage tasks; This represents the i-th stage task, in Confidence value under precision operator; Increase the confidence level of the reliability of multi-stage tasks The single-stage task confidence is calculated to be equal to the set target confidence level. S4: Set the reliability aggregation of multi-stage tasks under the single-stage task confidence level: Based on the single-stage task confidence level set in S3, combined with the classification results in S2 and the system's working time and fault data, calculate the reliability confidence lower limit of each stage task; through the multi-stage task reliability aggregation method, merge the reliability evaluation results of each stage task under the single-stage task confidence level to obtain the reliability evaluation results of multi-stage tasks under the target confidence level; S5: Determine whether the evaluation results meet the reliability requirements under the target confidence level: Compare the evaluation results obtained in S4 with the set target mission reliability. If the evaluation results reach the target mission reliability threshold, the unmanned surface vessel passes the mission reliability assessment.

2. The method for assessing the reliability of unmanned surface vessels in a multi-stage mission based on a given confidence level as described in claim 1, characterized in that: In S2, the fault distribution and series-parallel relationships of unmanned surface vessel (USV) equipment are classified as follows: A data fusion method for a serial system composed of devices with different exponential distributions; A data fusion method for a serial system composed of devices with the same exponential distribution; A data fusion method for a parallel system composed of devices with the same exponential distribution; A data fusion method for a K / N system composed of devices with the same exponential distribution.

3. The method for assessing the reliability of unmanned surface vessels in a multi-stage mission based on a given confidence level as described in claim 2, characterized in that: In S4, the methods for calculating the lower limit of the reliability confidence of a series system composed of devices with different exponential distributions include: When a component of the system malfunctions: ； ； Sub-confidence level is The lower confidence limit for the reliability of the task is: ； When the components of the system are not malfunctioning: ； Confidence level is The lower confidence limit for the reliability of the task is: ； In the formula, Normally distributed quantiles; Indicates the number of devices in the system; Indicates the equivalent operating time of each unit device; This indicates the number of failures in each unit device: Indicates the testing time of the merged system; Indicates the number of failures in the merged system; Indicates the duration of one task; This indicates the number of equivalent tasks.

4. The method for assessing the reliability of unmanned surface vessels in a multi-stage mission based on a given confidence level as described in claim 2, characterized in that: In S4, the methods for calculating the lower confidence limit of the reliability of a series system composed of devices with the same exponential distribution include: When a component of the system fails, the sub-confidence level is The lower confidence limit for the reliability of the task is: ； in, ; In the formula, For degrees of freedom and confidence level of Quantiles of the distribution; When no component of the system malfunctions, the sub-confidence level is: The lower confidence limit for the reliability of the task is: ； In the formula, Indicates the test time of the equipment; Indicates the testing time of the merged system; Indicates the number of failures in the merged system; Indicates confidence level; Indicates the duration of one task; Indicates the number of equivalent tasks; This represents the number of equivalent tasks after merging; Indicates the number of times the task failed.

5. The method for assessing the reliability of unmanned surface vessels in a multi-stage mission based on a given confidence level as described in claim 2, characterized in that: In S4, the method for calculating the lower confidence limit of the reliability of a parallel system composed of devices with the same exponential distribution includes: The confidence level is The lower confidence limit for the reliability of the task is: ； in, ; In the formula For degrees of freedom and confidence level of Quantiles of the distribution; When no component of the system malfunctions, the confidence level is: The lower confidence limit for the reliability of the task is: ； In the formula, Indicates the number of devices in the system; Indicates the total testing time of the equipment; Indicates the testing time of the merged system; Indicates the number of failures in the merged system; Indicates confidence level; Indicates the duration of one task; Indicates the number of equivalent tasks; This represents the number of equivalent tasks after merging.

6. The method for assessing the reliability of unmanned surface vessels in a multi-stage mission based on a given confidence level as described in claim 2, characterized in that: In S4, the calculation method for the lower confidence limit of the reliability of a K / N system composed of devices with the same exponential distribution includes: When a component of the system malfunctions: The system's MTBF at a sub-confidence level of The lower confidence limit at that time: ； When the components of the system are not malfunctioning: Sub-confidence level is The lower confidence limit for the reliability of the task is: ； In the formula, X represents the number of devices that are functioning normally; Indicates the number of devices included in the system; Indicates the minimum number of working devices; Indicates the total system test time; Indicates the testing time of the merged system; Indicates the number of failures in the merged system; Indicates the duration of one task; Indicates the number of equivalent tasks; Indicates the confidence level.

7. The method for assessing the reliability of unmanned surface vessels in a multi-stage mission based on a given confidence level as described in claim 1, characterized in that: In S4, when a multi-stage task consists of m completely identical and independent stage tasks, the i-th stage task has a confidence level of... The lower confidence limit of reliability at that time is Multi-stage tasks in terms of confidence The lower confidence limit of task reliability for: 。 8. The method for assessing the reliability of unmanned surface vessels in a multi-stage mission based on a given confidence level as described in claim 1, characterized in that: In S4, when a multi-stage task consists of multiple single-stage tasks with not entirely identical distributions and which are independent of each other, the confidence level of each single-stage task is... The lower confidence limit for the reliability of the task is as follows: Multi-stage tasks in terms of confidence The lower confidence limit of task reliability for: 。 9. The method for assessing the reliability of unmanned surface vessels in a multi-stage mission based on a given confidence level as described in any one of claims 1-8, characterized in that: S4 also includes: calculating the lower confidence limit of the reliability of the first-stage task at the determined confidence level, using the confidence level of the single-stage task determined in S3. ； In the formula, This is the lower confidence limit for the reliability of the first phase of the mission; This represents the total equivalent test time for the first phase of the mission. For the equivalent number of faults, For degrees of freedom is The determined sub-confidence level is time Distribution value.

Images