A method for predicting quality status of inertial navigation system based on mission profile

Abstract

The present application discloses a method for predicting the quality situation of an inertial navigation system based on a mission profile, which relates to the field of quality situation prediction. The method includes: extracting the variance of monitoring indicators from the monitoring data corresponding to each monitoring indicator when the inertial navigation system performs a mission under the framework of a confidence rule base to calculate the weight of the optimization premise attribute, extracting the information entropy eigenvalue and calculating the optimization rule weight based on the information entropy eigenvalue, extracting the kurtosis eigenvalue and calculating the optimization individual matching degree based on the kurtosis eigenvalue, and then establishing an inertial navigation system quality situation prediction model; collecting the unprocessed monitoring data of the inertial navigation system and inputting it into the inertial navigation system quality situation prediction model to obtain the corresponding quality situation classification level. The present application can effectively and accurately predict the quality situation of the inertial navigation system.

Description

Technical Field

[0001] The present application relates to the field of quality situation prediction, and in particular to a quality situation prediction method of an inertial navigation system based on a mission profile. Background Art

[0002] With the rapid development of aerospace, aviation, navigation, and intelligent transportation, the inertial navigation system (INS) is a key navigation and control technology, and its quality stability and reliability are of vital importance. However, in actual applications, due to environmental factors, device aging, error accumulation and other reasons, the performance of the inertial navigation system will gradually decline, affecting navigation accuracy and reliability. Therefore, accurately predicting the quality status of the inertial navigation system and timely discovering and handling potential problems have become important means to ensure the safe and stable operation of the system. Summary of the invention

[0003] The purpose of this application is to provide a method for predicting the quality status of an inertial navigation system based on a mission profile, which can effectively and accurately predict the quality status of an inertial navigation system.

[0004] To achieve the above objectives, the present application provides a method for predicting the quality status of an inertial navigation system based on a mission profile, comprising:

[0005] Acquire monitoring data corresponding to multiple monitoring indicators when the inertial navigation system performs tasks.

[0006] A confidence rule base framework is constructed, and under the confidence rule base framework, the monitoring indicator variance is extracted from the monitoring data corresponding to each of the monitoring indicators to calculate the optimization premise attribute weight; the optimization premise attribute weight is the premise attribute considering the inertial navigation system sensor error.

[0007] An information entropy characteristic value is extracted from the monitoring data corresponding to each of the monitoring indicators, and an optimization rule weight is calculated based on the information entropy characteristic value under the framework of the confidence rule base; the optimization rule weight is a rule weight that takes into account the working intensity of the inertial navigation system.

[0008] A kurtosis characteristic value is extracted from the monitoring data corresponding to each of the monitoring indicators, and an optimized individual matching degree is calculated based on the kurtosis characteristic value under the framework of the confidence rule base; the optimized individual matching degree is an individual matching degree that takes into account the environmental interference of the inertial navigation system.

[0009] Based on the optimization premise attribute weights, the optimization rule weights, the optimization individual matching degrees and the confidence rule base framework, an inertial navigation system quality situation prediction model is established.

[0010] The monitoring data to be processed of the inertial navigation system is collected and input into the quality situation prediction model of the inertial navigation system to obtain the corresponding quality situation classification level.

[0011] According to the specific embodiments provided by the present application, the present application discloses the following technical effects: the present application provides a method for predicting the quality situation of an inertial navigation system based on a task profile. Under the framework of a confidence rule base, the variance of the monitoring indicator, the characteristic value of information entropy, and the characteristic value of kurtosis are extracted from the monitoring data corresponding to each monitoring indicator when the inertial navigation system performs a task, so as to quantify the influence of the corresponding working time on the quality situation, the influence of the working intensity on the quality situation, and the influence of the environmental interference on the quality situation respectively. At the same time, a comprehensive consideration is made in the form of a confidence rule base framework to calculate the corresponding optimization premise attribute weight, the optimization rule weight, and the optimization individual matching degree, so as to construct an inertial navigation system quality situation prediction model that takes into account the working time, working intensity, and environmental interference. And because the inertial navigation system quality situation prediction model comprehensively considers the data when performing the above three tasks, it can output a more practical, more effective and accurate quality situation classification level when processing the monitoring data to be processed. And after the construction of the inertial navigation system quality situation prediction model corresponding to an inertial navigation system is completed, it can be directly applied in the future, which is more efficient. BRIEF DESCRIPTION OF THE DRAWINGS

[0012] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings required for use in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0013] Figure 1 A flowchart of a method for predicting quality status of an inertial navigation system based on a mission profile is provided in one embodiment of the present application.

[0014] Figure 2 A schematic flow chart of a method for predicting quality status of an inertial navigation system based on a mission profile provided in another embodiment of the present application.

[0015] Figure 3 A schematic diagram of gyroscope and accelerometer drift data provided by an embodiment of the present application.

[0016] Figure 4 A schematic diagram of the quality situation prediction results of an inertial navigation system provided in one embodiment of the present application. DETAILED DESCRIPTION

[0017] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.

[0018] The inertial navigation system contains sensor devices such as gyroscopes and accelerometers, which have a complex structure and are difficult to establish an accurate mathematical analytical model, which brings difficulties to the quality situation prediction of the inertial navigation system. Studies have found that when the inertial navigation system works under a certain mission profile, the quality situation prediction results will be affected by factors such as the duration of the mission profile, work intensity, and external environmental interference. As a type of expert system, the confidence rule base has the greatest advantage of being able to make full use of system knowledge and monitoring information to carry out complex system modeling. Therefore, as a knowledge and data hybrid driven method, the confidence rule base can be widely used in fault diagnosis, safety assessment, quality situation prediction and other fields. Based on this, this application, under the framework of the confidence rule base, considers the mission profile of the inertial navigation system during actual work, and proposes a quality situation prediction method for the inertial navigation system based on the mission profile.

[0019] In order to make the above-mentioned objects, features and advantages of the present application more obvious and easy to understand, the present application is further described in detail below in conjunction with the accompanying drawings and specific implementation methods.

[0020] In an exemplary embodiment, Figure 1 and Figure 2 As shown, a method for predicting the quality status of an inertial navigation system based on a mission profile is provided. The method is executed by a computer device, and can be specifically executed by a computer device such as a terminal or a server alone, or by a terminal and a server together. In an embodiment of the present application, the method includes the following steps 100 to 600.

[0021] Step 100, obtaining monitoring data corresponding to a plurality of monitoring indicators when the inertial navigation system performs a task.

[0022] Step 200, constructing a confidence rule base framework, and extracting monitoring indicator variances from monitoring data corresponding to each of the monitoring indicators under the confidence rule base framework to calculate optimization premise attribute weights; the optimization premise attribute weights are premise attributes that take into account the inertial navigation system sensor errors.

[0023] The confidence rule base framework includes multiple premise attributes, attribute weights corresponding to each premise attribute, multiple confidence rules, and rule weights corresponding to each confidence rule. iIf we use a monitoring indicator to represent the quality status of the inertial navigation system, then the framework of the confidence rule base can be established as follows:

[0024] .

[0025] in, R k For the k confidence rules; x 1( t ) is the first monitoring indicator t The monitoring data corresponding to each moment, x 2( t ) is the monitoring data corresponding to the second monitoring indicator at time t; is the i-th monitoring indicator t The monitoring data corresponding to each moment, i =1,2,…,M; For the k The confidence rule i 1 evaluation result of the premise attribute, For the k The confidence rule i 2 evaluation results of the premise attributes, For the k The confidence rule i N evaluation results of premise attributes, N is the number of evaluation results; For the k The first evaluation result of the confidence rule H Confidence level of 1; For the k The Nth evaluation result of confidence rules H N The confidence level of the evaluation result characterizes the quality status of the inertial navigation system and divides it into grades; For and relationship; For the k The rule weight of each confidence rule; is the initial attribute weight of the first prerequisite attribute, is the initial attribute weight of the second prerequisite attribute, For the i The initial attribute weight of a premise attribute; the monitoring indicator corresponds to the premise attribute in the confidence rule base framework.

[0026] Research has found that after long-term operation, the performance of the sensors of the inertial navigation system begins to degrade, resulting in a decrease in the accuracy of the measurement results and errors. Variance is a concept in statistics that is used to measure the degree of deviation between a set of data and its mean. When a sensor measures the same physical quantity multiple times, its output value will fluctuate around the true value or a certain reference value. Variance can measure the fluctuation of a set of data. Specifically, the larger the variance, the greater the discreteness of the sensor output value, that is, the greater the error caused by the degradation of sensor performance; conversely, the smaller the variance, the smaller the discreteness of the sensor output value, that is, the more stable the sensor performance and the smaller the error. Therefore, in response to the problem of sensor performance degradation caused by long-term operation of the inertial navigation system, this application provides a sensor error calculation method based on the variance of monitoring data. Specifically, under the framework of the confidence rule base, the variance of the monitoring indicator is extracted from the monitoring data corresponding to each of the monitoring indicators to calculate the optimization premise attribute weight, including the following steps (21)-(23).

[0027] (21) According to the monitoring data corresponding to each of the monitoring indicators, the corresponding monitoring indicator variance is calculated using the following formula:

[0028] .

[0029] (22) According to the monitoring data and monitoring indicator variance corresponding to each of the monitoring indicators, the inertial navigation system sensor error is calculated, and the calculation formula is:

[0030] .

[0031] in, is the monitoring indicator variance of the ith monitoring indicator, T is the length of the monitoring data corresponding to the monitoring indicator; x i ( t ) is the i-th monitoring indicator t The monitoring data corresponding to each moment, t =1,2,…,T, M is the number of monitoring indicators; For the i The average value of the monitoring data corresponding to each monitoring indicator; e i is the inertial navigation system sensor error, s i ( x i ( t ), ) is the i Monitoring indicators t The monitoring data corresponding to the time The distance can be expressed by the formula:

[0032] .

[0033] (23) The inertial navigation system sensor error is combined with the attribute weight corresponding to the premise attribute to obtain the optimized premise attribute weight, and the calculation formula is:

[0034] .

[0035] .

[0036] in, For the i The optimized premise attribute weights corresponding to the premise attributes are: After initial optimization i The attribute weights of the prerequisite attributes, , , For the i The initial attribute weights of the prerequisite attributes; Characterizes the degree of ignorance of some premise attributes about the activated rules. Because in the confidence rule base framework, it cannot be guaranteed that all rules can be activated for reasoning every time. The unactivated rules just reflect the ignorance of the premise attributes about these rules.

[0037] Thus, the present application realizes the combination of the inertial navigation system sensor error and the confidence rule base, and introduces the error caused by the performance degradation of the sensor due to long-term use into the confidence rule base.

[0038] Step 300, extracting information entropy characteristic values ​​from the monitoring data corresponding to each of the monitoring indicators, and calculating optimization rule weights based on the information entropy characteristic values ​​under the framework of the confidence rule base; the optimization rule weights are rule weights that take into account the working intensity of the inertial navigation system.

[0039] Specifically, the working intensity of the inertial navigation system refers to the amount of information processed by the inertial navigation system at a certain moment. The larger the amount of information processed, the higher the working intensity, and the smaller the amount of information processed, the smaller the working intensity of the system. Information entropy is a basic concept in information theory, which is used to measure the uncertainty or amount of information of a random variable or data set. It represents the degree of chaos or disorder of the system state, and can also be understood as the amount of information required to eliminate this uncertainty. The larger the information entropy, the higher the uncertainty of the data set, the greater the amount of information contained, and the higher the working intensity of the system when processing this information; conversely, the smaller the information entropy, the lower the uncertainty of the data set, the relatively small amount of information, and the lower the working intensity of the system when processing this information.

[0040] Corresponding to this application, the calculation formula of the information entropy characteristic value is:

[0041] .

[0042] in, Indicates i The information entropy characteristic value of the monitoring indicator is T is the length of the monitoring data corresponding to the monitoring indicator; It is i The monitoring index is The probability of satisfying ; b is the base of the logarithm.

[0043] Based on the information entropy characteristic value, the work intensity is calculated using the following formula:

[0044] .

[0045] in, I i is the work intensity corresponding to the i-th monitoring indicator.

[0046] Based on the work intensity and the rule weight corresponding to the confidence rule, the optimization rule weight is calculated using the following formula:

[0047] .

[0048] .

[0049] in, To optimize the rule weights; is the rule weight after preliminary optimization, , , For the k The initial rule weights of the confidence rules, Indicates the degree of ignorance of some rules about the activation rules. So far, the present application has realized the combination of the working intensity of the inertial navigation system and the confidence rule base, and quantified the working intensity of the inertial navigation system based on information entropy and introduced it into the confidence rule base.

[0050] Step 400, extracting kurtosis eigenvalues ​​from the monitoring data corresponding to each of the monitoring indicators, and calculating optimized individual matching degrees based on the kurtosis eigenvalues ​​under the framework of the confidence rule base; the optimized individual matching degrees are individual matching degrees that take into account environmental interference of the inertial navigation system.

[0051] Specifically, during long-term, high-intensity work, the inertial navigation system will inevitably be disturbed by complex external environments, such as vibration and other factors, which will directly affect the system performance and reliability, resulting in inaccurate quality situation prediction. When the inertial navigation system is disturbed by external vibration, the monitoring data will show a more obvious impact component. Kurtosis is an indicator that is sensitive to impact signals. In vibration signal analysis, kurtosis is particularly suitable for evaluating the impact characteristics of signals. When the kurtosis value is close to 3, it means that the distribution of the vibration signal is close to the normal distribution, which usually means that the system is in a normal state or is less disturbed; when the kurtosis value is greater than 3, it means that the distribution of the vibration signal is sharper than the normal distribution and the tail is thicker, which usually means that there is a large impact component in the signal and the interference from the external environment (such as impact, vibration intensification, etc.) is relatively large; when the kurtosis value is less than 3, it means that the distribution of the vibration signal is flatter than the normal distribution and the tail is sparser. At this time, the system is less disturbed by the outside world and the system response is relatively smooth.

[0052] Corresponding to this application, the calculation formula of the kurtosis eigenvalue is:

[0053] .

[0054] in, T is the length of the monitoring data corresponding to the monitoring indicator; It is i The mean of the monitoring data corresponding to the monitoring indicators, x i ( t ) is the i Monitoring indicators t Monitoring data corresponding to each moment; It is the fourth-order central moment, which represents the average value of the fourth power of the difference between the monitoring data corresponding to the monitoring indicator and its mean; is the square of the second central moment, K i For the i The kurtosis eigenvalue of the monitoring indicator.

[0055] Based on the kurtosis eigenvalue, the corresponding environmental interference is calculated using the following formula: n i :

[0056] .

[0057] After the monitoring data is input into the confidence rule base framework, it is first converted into a unified measurement framework through the following formula:

[0058] .

[0059] When the monitoring data is interfered by the external environment, there is some uncertainty in the obtained matching degree. In this case, the following formula can be used to calculate and optimize the individual matching degree:

[0060] .

[0061] After considering the interference of indicators, the calculation formula of the remaining matching degree of monitoring data is as follows:

[0062] .

[0063] in, To optimize individual matching, characterize the i The monitoring data corresponding to each monitoring indicator is in k The individual matching degree after considering environmental interference on the reference level in the confidence rule; and They are the first i The reference values ​​of the premise attributes are given by expert knowledge; L is the number of confidence rules in the confidence rule base framework, ; For the i The remaining matching degree of each monitoring indicator after considering the uncertainty of the monitoring data indicates the degree of uncertainty information contained in the monitoring data disturbed by the environment.

[0064] Thus, the present application realizes the combination of the environmental interference of the inertial navigation system and the confidence rule base, and introduces the environmental interference of the inertial navigation system into the confidence rule base based on the kurtosis eigenvalue, thereby providing a new matching degree calculation method.

[0065] Step 500: Establishing an inertial navigation system quality situation prediction model based on the optimization premise attribute weights, the optimization rule weights, the optimization individual matching degrees and the confidence rule base framework.

[0066] Specifically, in the inertial navigation system quality situation prediction model, the following formula is used to calculate the input matching degree of the model input monitoring data relative to the confidence rule:

[0067] .

[0068] in, Indicates k The input matching degree of the premise attributes, For the k The confidence rule i Optimized individual matching of each monitoring indicator; Indicates i Optimize the prerequisite attribute weights of the prerequisite attributes.

[0069] After the monitoring data is input into the inertial navigation system quality situation prediction model, some confidence rules will be activated accordingly, and the activation weight of each confidence rule is different. The following formula can be used to calculate the rule activation weight:

[0070] .

[0071] in, For the k The rule activation weights of the confidence rules, For the k The optimized rule weights of the confidence rules.

[0072] In addition, the activation weights should satisfy two constraints: and .when , it means that the confidence rule is not activated. When a confidence rule is activated, it will produce corresponding output. k When a confidence rule is activated, the evidence reasoning algorithm is used for fusion, and the following formula is used to calculate the k Rule output for confidence rules:

[0073] .

[0074] .

[0075] in, Indicates the confidence of the model's output relative to the evaluation results; To calculate the intermediate variables, is the confidence of the jth evaluation result of the kth confidence rule.

[0076] The rule outputs of all confidence rules are integrated to obtain the final output result of the inertial navigation system quality situation prediction model. for:

[0077] ; .

[0078] in, For the i Monitoring indicators t The monitoring data corresponding to each moment, is a nonlinear model, For the j The evaluation results, for relative to The output confidence level.

[0079] So far, this application has established an inertial navigation system quality status prediction model based on the mission profile, taking into account the working time, working intensity, and external environmental interference of the inertial navigation system, and thus realizing the quality status prediction of the inertial navigation system under the mission profile.

[0080] Step 600 , collecting the monitoring data to be processed of the inertial navigation system and inputting it into the quality situation prediction model of the inertial navigation system to obtain the corresponding quality situation classification level.

[0081] In an exemplary embodiment, the present application uses the gyroscope and accelerometer drift data of the inertial navigation system as experimental support, and obtains a total of 800 sets of test data. The gyroscope and accelerometer drift data are as follows: Figure 3 As shown. The two sets of data of gyroscope and accelerometer drift are used as the input evidence sources of the inertial navigation system quality situation prediction model, and the reference levels are set to include: "low", "lower", "higher", and "high", which are represented by "L", "S", "V", and "H" respectively. The reference values ​​and reference levels are set as shown in Table 1. The output quality situation classification level includes four levels, such as: "good", "fatigue", "fault", and "uncertainty". The parameter values ​​corresponding to each level are shown in Table 2.

[0082] Table 1

[0083]

[0084] Table 2

[0085]

[0086] On this basis, an inertial navigation system quality situation prediction model is established. Through this model, different characteristic information of the inertial navigation system when performing tasks is integrated, and the working time, working intensity, and environmental interference of the inertial navigation system under the mission profile are considered at the same time, so as to achieve a more accurate prediction of the quality situation of the inertial navigation system. The confidence rule weight setting considering the mission profile is obtained by calculating the working intensity of the inertial navigation system. The calculated rule weights are shown in the second column of Table 3. There are 16 rules in total, and the initial confidence setting of each rule is shown in the last column of Table 3.

[0087] Table 3

[0088]

[0089] like Figure 4 As shown in the figure, the dotted line is the true value of the inertial navigation system quality situation, and the scattered points are the predicted result values ​​of the inertial navigation system model based on the mission profile. Figure 4 It can be seen that the method of the present application can effectively and accurately predict the quality status of the inertial navigation system.

[0090] In summary, under the framework of the trust rule base, this application uses the monitoring data of the inertial navigation system to extract variance features and inputs them into the attribute weights to characterize the errors caused by the performance degradation of the sensor due to long-term use; the amount of information of the monitoring data is calculated based on the information entropy index and input into the calculation of the rule weights to characterize the working task intensity of the inertial navigation system; the kurtosis eigenvalue of the inertial navigation system is extracted and a new matching degree calculation method is proposed based on the kurtosis eigenvalue to characterize the interference of the external environment on the inertial navigation system. Finally, an inertial navigation system quality situation prediction model based on the mission profile is established, taking into account the working time, working intensity, and environmental interference of the inertial navigation system, realizing the quality situation prediction of the inertial navigation system under the mission profile, and improving the prediction accuracy of the model. Furthermore, the method of this application can be used to make intelligent decisions for the inertial navigation system, which has good engineering application value.

[0091] The technical features of the above embodiments may be combined arbitrarily. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0092] This article uses specific examples to illustrate the principles and implementation methods of this application. The description of the above embodiments is only used to help understand the method and core ideas of this application. At the same time, for those skilled in the art, according to the ideas of this application, there will be changes in the specific implementation methods and application scope. In summary, the content of this specification should not be understood as limiting this application.

Claims

1. A method for predicting the quality status of an inertial navigation system based on a mission profile, characterized in that: The mission profile-based inertial navigation system quality situation prediction method includes: Obtain monitoring data corresponding to multiple monitoring indicators when the inertial navigation system performs tasks; Constructing a confidence rule base framework, and extracting monitoring indicator variances from monitoring data corresponding to each monitoring indicator under the confidence rule base framework to calculate optimization premise attribute weights; the optimization premise attribute weights are premise attributes that take into account inertial navigation system sensor errors; Extracting information entropy characteristic values ​​from the monitoring data corresponding to each of the monitoring indicators, and calculating optimization rule weights based on the information entropy characteristic values ​​under the framework of the confidence rule base; the optimization rule weights are rule weights that take into account the working intensity of the inertial navigation system; Extracting kurtosis characteristic values ​​from the monitoring data corresponding to each of the monitoring indicators, and calculating the optimized individual matching degree based on the kurtosis characteristic values ​​under the framework of the confidence rule base; the optimized individual matching degree is the individual matching degree considering the environmental interference of the inertial navigation system; Based on the optimization premise attribute weights, the optimization rule weights, the optimization individual matching degrees and the confidence rule base framework, a quality situation prediction model of the inertial navigation system is established; Collecting the monitoring data to be processed of the inertial navigation system and inputting it into the quality situation prediction model of the inertial navigation system to obtain the corresponding quality situation classification level; The calculation formula of the kurtosis eigenvalue is: ； in, T is the length of the monitoring data corresponding to the monitoring indicator; It is i The average value of the monitoring data corresponding to the monitoring indicators is x i ( t ) is the i Monitoring indicators t Monitoring data corresponding to each moment; It is the fourth-order central moment, which represents the average value of the fourth power of the difference between the monitoring data corresponding to the monitoring indicator and its mean; is the square of the second central moment, K i For the i The kurtosis eigenvalue of the monitoring indicator, i= 1,2,… ,M , M The number of indicators to be monitored; t =1,2,…, T ; Based on the kurtosis eigenvalue, the corresponding environmental interference is calculated using the following formula: n i : 。 2. The method for predicting the quality status of an inertial navigation system based on a mission profile according to claim 1 is characterized in that: The confidence rule base framework includes a plurality of premise attributes and an attribute weight corresponding to each premise attribute; Under the framework of the confidence rule base, the variance of the monitoring indicator is extracted from the monitoring data corresponding to each of the monitoring indicators to calculate the optimization premise attribute weight, including: Calculate the corresponding monitoring indicator variance according to the monitoring data corresponding to each of the monitoring indicators; Calculating the inertial navigation system sensor error according to the monitoring data and monitoring indicator variance corresponding to each of the monitoring indicators; The inertial navigation system sensor error is combined with the attribute weight corresponding to the premise attribute to obtain an optimized premise attribute weight.

3. The method for predicting the quality status of an inertial navigation system based on a mission profile according to claim 2 is characterized in that: The calculation formula of the monitoring index variance is: ； The calculation formula of the inertial navigation system sensor error is: ； in, For the i The monitoring indicator variance of the monitoring indicator is T is the length of the monitoring data corresponding to the monitoring indicator; x i ( t ) is the i Monitoring indicators t The monitoring data corresponding to each moment, For the i The average value of the monitoring data corresponding to each monitoring indicator; e i is the inertial navigation system sensor error, s i ( x i ( t ), ) is the i Monitoring indicators t The monitoring data corresponding to the time distance; The calculation formula of the optimization premise attribute weight is: ； ； in, is the optimized premise attribute weight corresponding to the i-th premise attribute, After initial optimization i The attribute weights of the prerequisite attributes, For the i The initial attribute weights of the prerequisite attributes; Characterizes the degree of ignorance of some premise attributes for the activation rules; the monitoring indicators correspond to the premise attributes in the framework of the confidence rule base.

4. The method for predicting the quality status of an inertial navigation system based on a mission profile according to claim 1, characterized in that: The framework of the trust rule base is: ； in, R k For the k confidence rules; x 1( t ) is the first monitoring indicator t The monitoring data corresponding to each moment, x 2( t ) is the monitoring data corresponding to the second monitoring indicator at time t, For the i Monitoring indicators t Monitoring data corresponding to each moment; For the k The confidence rule i 1 evaluation result of the premise attribute, For the k The confidence rule i 2 evaluation results of the premise attributes, For the k The confidence rule i N evaluation results of premise attributes, N is the number of evaluation results; For the k The first evaluation result of the confidence rule H A confidence level of 1, For the k The Nth evaluation result of the confidence rule H N The confidence level of the evaluation result characterizes the quality status of the inertial navigation system and divides it into grades; For and relationship; For the k The rule weight of each confidence rule; is the initial attribute weight of the first prerequisite attribute, is the initial attribute weight of the second prerequisite attribute, For the i The initial attribute weight of the premise attribute; the meaning of the confidence rule base framework is: k Confidence Rules R k Expressed as: If the prerequisite is met: x 1( t )yes ,and x 2( t )yes ,……,and yes , then the output of executing the fuzzy rules is: The first evaluation result H 1 and the corresponding confidence , No. N Evaluation results H N and the corresponding confidence , the rule weight in the fuzzy rule is , the initial attribute weight is , ,……, .

5. The method for predicting the quality status of an inertial navigation system based on a mission profile according to claim 1, characterized in that: The confidence rule base framework includes a plurality of confidence rules and a rule weight corresponding to each confidence rule; The calculation formula of the information entropy eigenvalue is: ； in, En i Indicates i The information entropy characteristic value of the monitoring indicator is T is the length of the monitoring data corresponding to the monitoring indicator, It is i The monitoring index is The probability of b is the base of the logarithm; Based on the information entropy characteristic value, the work intensity is calculated using the following formula: ； in, I i is the work intensity corresponding to the i-th monitoring indicator; Based on the work intensity and the rule weight corresponding to the confidence rule, the optimization rule weight is calculated using the following formula: ； ； in, To optimize the rule weights; is the rule weight after preliminary optimization, For the k The initial rule weights of the confidence rules, represents the degree of ignorance of some rules about the activation rules, M is the number of monitoring indicators, L is the number of confidence rules in the confidence rule base framework.

6. The method for predicting the quality status of an inertial navigation system based on a mission profile according to claim 1, characterized in that: The calculation formula for optimizing individual matching is: ； ； in, ; To optimize individual matching, characterize the i The monitoring data corresponding to each monitoring indicator is in k The individual matching degree after considering environmental interference on the reference level in the confidence rule; and They are the first i The reference value of the premise attribute; L is the number of confidence rules in the confidence rule base framework, ; For the i The remaining matching degree of each monitoring indicator after considering the uncertainty of monitoring data; Characterization i Monitoring indicators t Monitoring data corresponding to the time The value of is in the first two adjacent confidence rules under the confidence rule base framework. i The range is between the reference values ​​of the premise attributes.

7. The method for predicting the quality status of an inertial navigation system based on a mission profile according to claim 1, characterized in that: In the inertial navigation system quality situation prediction model, the following formula is used to calculate the input matching degree of the model input monitoring data relative to the confidence rule: ； in, Indicates k The input matching degree of the premise attributes, For the k The confidence rule i Optimized individual matching of each monitoring indicator; Indicates i Optimized prerequisite attribute weights of prerequisite attributes; M The number of indicators to be monitored; The rule activation weight is calculated using the following formula: ； in, For the k The rule activation weights of the confidence rules, For the k The optimized rule weights of the confidence rules; L is the number of confidence rules in the confidence rule base framework; When k When a confidence rule is activated, the evidence reasoning algorithm is used for fusion, and the following formula is used to calculate the k Rule output for confidence rules: ； ； in, Indicates the confidence of the model's output relative to the evaluation results; To calculate intermediate variables; N is the number of evaluation results, For the k Confidence rule j The confidence level of the evaluation results; The rule outputs of all confidence rules are integrated to obtain the final output result of the inertial navigation system quality situation prediction model. for: ； ； in, For the i Monitoring indicators t The monitoring data corresponding to each moment, is a nonlinear model, For the j The evaluation results, for relative to The output confidence level.

Images