CN109779791A - An intelligent diagnosis method for abnormal data in solid rocket motor - Google Patents

An intelligent diagnosis method for abnormal data in solid rocket motor Download PDF

Info

Publication number
CN109779791A
CN109779791A CN201910224959.6A CN201910224959A CN109779791A CN 109779791 A CN109779791 A CN 109779791A CN 201910224959 A CN201910224959 A CN 201910224959A CN 109779791 A CN109779791 A CN 109779791A
Authority
CN
China
Prior art keywords
abnormal
neural network
parameters
data
test
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Granted
Application number
CN201910224959.6A
Other languages
Chinese (zh)
Other versions
CN109779791B (en
Inventor
卫莹
钟华
张敏
李瑛�
李强
胡博
王忠颐
曹莎
陈涛
王哲
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Observation And Control Technology Research Institute Of Xi'an Space Dynamic
Original Assignee
Observation And Control Technology Research Institute Of Xi'an Space Dynamic
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Observation And Control Technology Research Institute Of Xi'an Space Dynamic filed Critical Observation And Control Technology Research Institute Of Xi'an Space Dynamic
Priority to CN201910224959.6A priority Critical patent/CN109779791B/en
Publication of CN109779791A publication Critical patent/CN109779791A/en
Application granted granted Critical
Publication of CN109779791B publication Critical patent/CN109779791B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Landscapes

  • Testing Of Engines (AREA)

Abstract

The present invention proposes abnormal data intelligent diagnosing method in a kind of solid propellant rocket, abnormal point value is checked using t first, secondly according to BP-ML algorithm, the preparation method of weight in traditional artificial neural network is improved, is changed to that each parameter value { x of engine will be inputtediIt is assumed that a Gaussian process, meets AR (P) model, acquires weighted value by Maximum Likelihood Estimation Method, verify anomaly parameter, and carry out parameter value reconstruct.The present invention solves the problems, such as method deficiency under the conditions of the prior art, reduces the error of artificial micro-judgment, establishes abnormal data method for diagnosing faults.

Description

一种固体火箭发动机中异常数据智能诊断方法An intelligent diagnosis method for abnormal data in solid rocket motor

技术领域technical field

本发明属于固体火箭发动机试验测控技术领域,主要涉及固体火箭发动机试验中异常数据的智能自动化诊断,减少人为经验判断的误差,建立异常数据智能诊断方法。The invention belongs to the technical field of solid rocket motor test measurement and control, and mainly relates to intelligent automatic diagnosis of abnormal data in solid rocket motor tests, reduces errors in judgment by human experience, and establishes an intelligent diagnosis method for abnormal data.

背景技术Background technique

固体火箭发动机地面试验是一项高风险、高投入、高能耗、不可逆的试验项目。随着试验技术的发展,试验测控系统流程可靠性不断提高,能够做到主要参数获得完整无失误。但在试验过程中工况较为复杂,不可避免地受发动机设计、工装、传感器损坏、线缆高温烧蚀、测量块掉落等因素影响,导致试验数据异常。异常情况分为异常点值和异常参数。异常点值的判断方法有t检验,拉依达判据和肖维勒判据等,其中,t检验是统计学经典的判断方法,属于正态总体均值与方差的假设检验中的一种方法,其前提是数据总体符合正态分布,可以有效严格地检验出异常点值。The solid rocket motor ground test is a high-risk, high-investment, high-energy-consumption and irreversible test project. With the development of test technology, the process reliability of the test measurement and control system has been continuously improved, and the main parameters can be obtained without errors. However, during the test, the working conditions are relatively complex, and it is inevitably affected by factors such as engine design, tooling, sensor damage, high temperature ablation of cables, and falling measuring blocks, resulting in abnormal test data. The abnormal situation is divided into abnormal point value and abnormal parameter. Judgment methods for outlier values include t test, Laida criterion and Chauville criterion, etc. Among them, t test is a classic statistical judgment method, which belongs to a method in the hypothesis test of normal population mean and variance. , the premise is that the data as a whole conforms to a normal distribution, which can effectively and strictly detect outliers.

在试验过程中固体火箭发动机易受到冲击、温度、结构变形等因素影响,试验测得的参数信号具有非线性、非平稳特征。目前国内固体火箭发动机领域进行异常参数判断时,仍然采用传统的数据处理方法,比较简单,当试验数据出现异常时,无法判断故障原因属于测控系统故障还是发动机本身故障,需专家进行经验判读,有一定的误差,因此,在此领域应用当前较为前沿的数据挖掘方法有一定的必要性。During the test, the solid rocket motor is easily affected by factors such as impact, temperature, and structural deformation. The parameter signals measured by the test have nonlinear and non-stationary characteristics. At present, when judging abnormal parameters in the domestic solid rocket motor field, the traditional data processing method is still used, which is relatively simple. When the test data is abnormal, it is impossible to judge whether the cause of the failure belongs to the failure of the measurement and control system or the failure of the engine itself. Experts are required to conduct experience interpretation. Therefore, it is necessary to apply the current more advanced data mining methods in this field.

在试验过程中的固体火箭发动机易受到冲击、温度、结构变形等因素影响,试验测得的参数信号具有非线性、非平稳特征,对于定型阶段的发动机,性能参数处于比较稳定的状态,试验环境、统计特征不随时间变化而变化。因此,在工程实际应用中,将固体发动机地面试验过程看作平稳随机过程,其采集到的各参数信号符合平稳随机信号特征,此类信号可以应用多种数据处理方法,如人工神经网络模型、Logistic回归模型、时间序列等目前应用较为广泛的机器学习、人工智能、统计学方法。During the test, the solid rocket motor is easily affected by factors such as impact, temperature, and structural deformation. The parameter signals measured in the test have nonlinear and non-stationary characteristics. For the engine in the finalization stage, the performance parameters are in a relatively stable state, and the test environment , statistical characteristics do not change with time. Therefore, in practical engineering applications, the solid motor ground test process is regarded as a stationary random process, and the collected parameter signals conform to the characteristics of stationary random signals. Logistic regression models, time series and other widely used machine learning, artificial intelligence, and statistical methods.

固体火箭发动机试验各参数之间的关系为非平稳、非线性的,受具体试验现场因素的不同影响,所得数据有差异,各参数之间具体的函数关系形式较难获得,而人工神经网络模型具有非线性映射与泛化能力,一个3层BP神经网络能够实现对任意非线性函数进行逼近(根据Kolrnogorov定理),如下图1所示。在试验数据分析处理中,可利用它表现参数间的相互关系。人工神经网络由输入层、隐层和输出层构成。学习过程由信号的正向传播和误差的反向传播两个过程组成,通过多次调整权值,直至网络输出的误差减小到可以接受的程度,或进行到事先设定的学习次数。本发明设计了一种新的权值估计算法,学习得到因变量和自变量之间的一个非线性关系。当存在某些异常参数时,可以根据此非线性关系将其检测出来。同时,也可对异常参数进行重构和恢复。The relationship between the parameters of the solid rocket motor test is non-stationary and nonlinear, and the data obtained are different due to the different factors in the specific test site. It is difficult to obtain the specific functional relationship between the parameters, and the artificial neural network model With nonlinear mapping and generalization capabilities, a 3-layer BP neural network can approximate any nonlinear function (according to the Kolrnogorov theorem), as shown in Figure 1 below. In the analysis and processing of experimental data, it can be used to express the relationship between parameters. Artificial neural network consists of input layer, hidden layer and output layer. The learning process consists of two processes, the forward propagation of the signal and the back propagation of the error. By adjusting the weights many times, the error of the network output is reduced to an acceptable level, or the number of learning times set in advance is carried out. The invention designs a new weight estimation algorithm to learn a nonlinear relationship between the dependent variable and the independent variable. When there are some abnormal parameters, they can be detected according to this nonlinear relationship. At the same time, the abnormal parameters can also be reconstructed and restored.

发明内容SUMMARY OF THE INVENTION

针对现有的固体火箭发动机试验数据的后期分析和处理过程中没有涉及异常数据智能化诊断,基本上仍然采用传统的数据处理方法,比较简单,当试验数据出现异常时,无法判断故障原因属于测控系统故障还是发动机故障,需专家进行经验判读,有一定的误差的问题。本发明提出一种固体火箭发动机中异常数据智能诊断方法,通过经典的t检验方法,判断发动机试验中的异常点值;并设计了一种基于人工神经网络模型检验发动机异常参数的新方法,改进了权值估计算法,使用该方法从发动机试验历史数据中发现隐含的故障规律,可以有效的检验出异常参数,减少人为经验判断误差,实现异常数据智能化诊断。The post-analysis and processing of the existing solid rocket motor test data does not involve intelligent diagnosis of abnormal data. Basically, the traditional data processing method is still used, which is relatively simple. When the test data is abnormal, it is impossible to determine the cause of the failure. System failure or engine failure requires expert interpretation, and there is a certain error problem. The invention proposes an intelligent diagnosis method for abnormal data in a solid rocket motor, which judges the abnormal point value in the engine test through the classical t test method; and designs a new method for testing abnormal parameters of the engine based on an artificial neural network model, which improves the The weight estimation algorithm is used, and the implicit fault rule is found from the historical data of the engine test, which can effectively detect the abnormal parameters, reduce the judgment error of human experience, and realize the intelligent diagnosis of abnormal data.

本发明的技术方案为:The technical scheme of the present invention is:

所述一种固体火箭发动机中异常数据智能诊断方法,其特征在于:包括以下步骤:The method for intelligently diagnosing abnormal data in a solid rocket motor is characterized by comprising the following steps:

步骤1:采集固体火箭发动机试验数据,检验所选取的试验数据是否满足正态分布,若满足,则进行步骤2,否则选取其他试验数据重新进行检验;Step 1: Collect the test data of the solid rocket motor, and check whether the selected test data satisfies the normal distribution, if so, proceed to Step 2, otherwise select other test data for re-testing;

步骤2:采用t检验方法,判别出所选取的试验数据中的异常点并剔除;Step 2: Use the t test method to identify the abnormal points in the selected test data and eliminate them;

步骤3:采用BP-ML算法,利用神经网络对经过步骤2中t检验处理后的试验数据进行异常参数检验,若存在异常参数,则对异常参数进行参数值重构。Step 3: Using the BP-ML algorithm and using the neural network to test the abnormal parameters of the test data processed by the t test in step 2, if there are abnormal parameters, reconstruct the parameter values of the abnormal parameters.

进一步的优选方案,所述一种固体火箭发动机中异常数据智能诊断方法,其特征在于:步骤1中通过计算试验数据的偏度和峰度判断试验数据是否满足正态分布。In a further preferred solution, the method for intelligently diagnosing abnormal data in a solid rocket motor is characterized in that: in step 1, it is judged whether the test data satisfies the normal distribution by calculating the skewness and kurtosis of the test data.

进一步的优选方案,所述一种固体火箭发动机中异常数据智能诊断方法,其特征在于:步骤3中BP-ML算法是通过将人工神经网络BP算法中的权向量估计方法采用最大似然估计方法实现而得到。A further preferred solution, the method for intelligently diagnosing abnormal data in a solid rocket motor, is characterized in that: in step 3, the BP-ML algorithm adopts the maximum likelihood estimation method for the weight vector estimation method in the artificial neural network BP algorithm. achieved by obtaining.

进一步的优选方案,所述一种固体火箭发动机中异常数据智能诊断方法,其特征在于:步骤3中进行异常参数检验的具体过程为:A further preferred solution, the method for intelligently diagnosing abnormal data in a solid rocket motor, is characterized in that: the specific process of performing abnormal parameter inspection in step 3 is:

步骤3.1:采用无异常参数的数据作为样本数据训练神经网络,得到训练好的神经网络;Step 3.1: Use the data without abnormal parameters as sample data to train the neural network to obtain a trained neural network;

步骤3.2:将实际试验数据输入神经网络,进行神经网络修复,若修复后的输出值与目标值之间相关性差,则推断神经网络输入值中存在异常参数;若神经网络输入值中存在异常参数,则对神经网络输入值中的参数值分别进行判断,以得到具体的异常参数;Step 3.2: Input the actual test data into the neural network to repair the neural network. If the correlation between the repaired output value and the target value is poor, it is inferred that there are abnormal parameters in the input value of the neural network; if there are abnormal parameters in the input value of the neural network , then the parameter values in the input value of the neural network are judged separately to obtain specific abnormal parameters;

步骤3.3:以步骤3.2得到的异常参数作为神经网络输出参数,在神经网络输入参数中去掉异常参数,然后重构神经网络,得到修复后的异常参数。Step 3.3: Use the abnormal parameters obtained in step 3.2 as the output parameters of the neural network, remove the abnormal parameters from the input parameters of the neural network, and then reconstruct the neural network to obtain the repaired abnormal parameters.

有益效果beneficial effect

与现有技术相比,本发明具有的特点是:(1)符合固体火箭发动机试验数据特性;(2)应用广泛;(3)构造完整的异常数据检验方法;(4)完成异常数据重构过程。Compared with the prior art, the present invention has the following features: (1) conforming to the characteristics of solid rocket motor test data; (2) widely used; (3) constructing a complete abnormal data inspection method; (4) completing abnormal data reconstruction process.

本发明的附加方面和优点将在下面的描述中部分给出,部分将从下面的描述中变得明显,或通过本发明的实践了解到。Additional aspects and advantages of the present invention will be set forth, in part, from the following description, and in part will be apparent from the following description, or may be learned by practice of the invention.

附图说明Description of drawings

本发明的上述和/或附加的方面和优点从结合下面附图对实施例的描述中将变得明显和容易理解,其中:The above and/or additional aspects and advantages of the present invention will become apparent and readily understood from the following description of embodiments taken in conjunction with the accompanying drawings, wherein:

图1:三层神经网络示意图;Figure 1: Schematic diagram of a three-layer neural network;

图2:目标值与修复值示意图;Figure 2: Schematic diagram of target value and repair value;

图3:目标值与修复值相关性示意图;Figure 3: Schematic diagram of the correlation between the target value and the repair value;

图4:CCDD相关系数示意图;Figure 4: Schematic diagram of CCDD correlation coefficient;

图5:目标值与修复值相关性示意图;Figure 5: Schematic diagram of the correlation between the target value and the repair value;

图6:目标值与修复值示意图;Figure 6: Schematic diagram of target value and repair value;

图7:目标值与修复值相关性示意图;Figure 7: Schematic diagram of the correlation between the target value and the repair value;

图8:CCDD相关系数示意图;Figure 8: Schematic diagram of CCDD correlation coefficient;

图9:神经网络的权值w;Figure 9: The weight w of the neural network;

图10:异常参数wP9修复后的值与目标值对比图;Figure 10: Comparison between the repaired value of the abnormal parameter wP9 and the target value;

图11:异常参数wP9。Figure 11: Abnormal parameter wP9.

具体实施方式Detailed ways

本发明涉及的异常数据检验方法使用t检验,其前提是建立在测定值遵循正态分布的理论基础之上的。尽管从理论上理解,固体火箭发动机地面试验稳态阶段中,测量性能参数特征不随时间变化而变化,符合正态分布,但是在试验过程中,会受到不同情况的干扰,其分布将会有不同程度的偏差,因此,使用t检验前,须对试验数据进行正态分布的检验。The abnormal data test method involved in the present invention uses t test, the premise of which is established on the theoretical basis that the measured value follows a normal distribution. Although it is theoretically understood that in the steady state stage of the solid rocket motor ground test, the characteristics of the measured performance parameters do not change with time and conform to a normal distribution, but during the test, it will be disturbed by different situations, and its distribution will be different. Therefore, before using the t test, the test data must be tested for normal distribution.

根据偏度与峰度公式,当随机变量{xn}服从正态分布时,其偏度k1=0,峰度k2=3,According to the formula of skewness and kurtosis, when the random variable {x n } obeys the normal distribution, its skewness k 1 =0, kurtosis k 2 =3,

计算公式如下:Calculated as follows:

t检验中,当方差未知时,单个正态总体均值的检验方法为:In the t-test, when the variance is unknown, the test method for a single normal population mean is:

假定总体服从正态分布N(μ,σ2),μ,σ2均为未知参数,(X1,...,XN)是总体容量为n的样本,欲检验假设:通常以样本修正方差代替总体方差构造统计量。Assuming that the population follows a normal distribution N(μ,σ 2 ), μ,σ 2 are unknown parameters, (X 1 ,...,X N ) is a sample with a population size of n. To test the hypothesis: The statistic is usually constructed with the sample-corrected variance instead of the population variance.

其中,当假设H0成立时,T服从自由度为n-1的t分布,当|T|的值大时,假设不大可能成立,应否定H0,所以,对给定0<α<1,由t分布表即可得检验的临界值tα/2(n-1)使in, When the hypothesis H 0 holds, T obeys the t distribution with n-1 degrees of freedom. When the value of |T| is large, the hypothesis is unlikely to hold, and H 0 should be rejected. Therefore, for a given 0<α<1, The critical value t α/2 (n-1) of the test can be obtained from the t distribution table, so that

p{|T|≥tα/2(n-1)}=α (4)p{|T|≥t α/2 (n-1)}=α (4)

故检验So check

若|T|≥tα/2(n-1),拒绝假设H0,即认为总体均值与μ0有显著差异;若|T|<tα/2(n-1),则接受H0,即认为总体均值与μ0无显著差异。这种利用服从t分布的统计量作为检验统计量的检验方法称为t检验法。If |T|≥t α/2 (n-1), reject the hypothesis H 0 , that is, consider that the population mean is significantly different from μ 0 ; if |T|<t α/2 (n-1), accept H 0 , that is, the overall mean is considered to be not significantly different from μ 0 . This test method that uses a statistic that obeys the t-distribution as a test statistic is called the t-test method.

选取固体火箭发动机试验中测量最主要参数之一压强值p的稳态数据作为检验对象。选择某发定型试验10.5s~17.5s中的数据点值,采样率为5000Sa/s,时间间隔为0.0002s,根据公式(1)、(2)分别计算测量数据的偏度k1=0.2279,峰度k2=2.7308,正态分布特性良好,可以用t检验。The steady-state data of the pressure value p, one of the most important parameters measured in the solid rocket motor test, is selected as the test object. Select the data point value in 10.5s ~ 17.5s of a hair styling test, the sampling rate is 5000Sa/s, the time interval is 0.0002s, and the skewness k 1 =0.2279 of the measured data is calculated according to formulas (1) and (2) respectively, The kurtosis k 2 =2.7308, the normal distribution characteristics are good, and the t test can be used.

神经网络模型具有强大的泛化能力,可以非线性映射多种参数之间的关系,因此本文选择对发动机设计影响较大的推力数据作为输出神经元节点,输入层神经元节点选择与输出层相关的压强、位移、应变、温度这些参数。设此系统方程具有m维输入向量和n维输出向量,{xi}为各输入参数点值,{fi}为输出参数点值,x=(x1,x2,…,xm)T,f=(f1,f2,…,fn)T,网络对应m个输入节点、n个输出节点,zi为系统识别函数,也称为激活函数,可以选择线性函数,斜面函数,s形函数等。系统方程如下所示:The neural network model has a strong generalization ability and can non-linearly map the relationship between various parameters. Therefore, this paper selects the thrust data that has a greater impact on the engine design as the output neuron node, and the selection of the input layer neuron node is related to the output layer. parameters of pressure, displacement, strain, and temperature. Suppose this system equation has an m-dimensional input vector and an n-dimensional output vector, {x i } is the value of each input parameter point, {f i } is the value of the output parameter point, x=(x 1 ,x 2 ,...,x m ) T ,f=(f 1 ,f 2 ,...,f n ) T , the network corresponds to m input nodes and n output nodes, zi is the system identification function, also called activation function, you can choose linear function, slope function , s-shaped function, etc. The system equations are as follows:

设神经网络有M层,第l层的节点数为nl,fi l为第l层节点i的输出,则Suppose the neural network has M layers, the number of nodes in the lth layer is n l , and f i l is the output of the lth layer node i, then

其中,为第l层节点i的状态,系数行向量,fl第l-1层输出列向量。in, is the state of node i at layer l, Coefficient row vector, f l layer l-1 output column vector.

此时,根据现有的人工神经网络BP算法,系统方程中的权重系数使用有导师学习方法,给定目标函数为g(w)minAt this time, according to the existing artificial neural network BP algorithm, the weight coefficient in the system equation Using the tutored learning method, the given objective function is g(w) min :

即在给定输入参数{X,F}后,使得目标函数最小That is, after the input parameters {X, F} are given, the objective function is minimized

为输出值,根据最优化中的方法-最速梯度下降法可以求出权向量 is the output value, According to the method in the optimization - the fastest gradient descent method, the weight vector can be obtained

为了提高系统方程输出值的精确度与效率,本发明改进人工神经网络BP算法,将权向量的估计方法改为最大似然估计(ML),新算法命名为BP-ML算法,In order to improve the accuracy and efficiency of the output value of the system equation, the present invention improves the artificial neural network BP algorithm, changes the estimation method of the weight vector to maximum likelihood estimation (ML), and the new algorithm is named as the BP-ML algorithm,

根据本文公式(1)、(2)检验得,输入发动机各参数值{xi}符合正态分布,且满足独立同分布,因为各参数值{xi}均是随时间变化而变化的过程量,假定{xi}是一个Gaussian过程,满足AR(P)模型,即用此时的公式(11)代替BP算法中的公式(9),According to the test of formulas (1) and (2) in this paper, the input engine parameter values {x i } conform to the normal distribution and satisfy the independent and identical distribution, because each parameter value {x i } is a process that changes with time. It is assumed that {x i } is a Gaussian process and satisfies the AR(P) model, that is, the formula (11) at this time is used to replace the formula (9) in the BP algorithm,

Xk+p+1=a1Xk+1+a2Xk+2+...+apXk+pk+p+1. (11)X k+p+1 =a 1 X k+1 +a 2 X k+2 +...+a p X k+pk+p+1 . (11)

其中假设白噪声εm~i.i.d.N(0,σ2),(X1,X2,...,XM)满足均值为μ,协方差为Σ的正态分布。设(xp+1,xp+2,…xp+m)T为AR(P)序列的一个输入样本。则:Among them, it is assumed that white noise ε m ~ iidN(0,σ 2 ), (X 1 , X 2 ,...,X M ) satisfies a normal distribution with mean μ and covariance Σ. Let (x p+1 , x p+2 ,…x p+m ) T be an input sample of the AR(P) sequence. but:

(xp+1,xp+2,…xp+m)~N(μ,Σ)(x p+1 ,x p+2 ,…x p+m )~N(μ,Σ)

其中Σ为协方差矩阵:where Σ is the covariance matrix:

我们的目标为基于观测值xp+1,xp+2,…xp+m,估计未知参数σ2,a1,a2,...,ap的值,记θ=(σ2,a1,a2,...,ap),令x1,x2,...,xp为未知参数。公式(11)可变为:Our goal is to estimate the value of the unknown parameters σ 2 , a 1 , a 2 ,..., a p based on the observed values x p+1 , x p+2 ,…x p+m , denoted θ=(σ 2 ,a 1 ,a 2 ,...,a p ), let x 1 ,x 2 ,...,x p be the unknown parameters. Equation (11) can be changed to:

所以基于观测值xp+1,xp+2,…xp+m,我们可得So based on the observations x p+1 ,x p+2 ,…x p+m , we can get

则(εp+1p+2,…,εp+m)的联合概率密度为Then the joint probability density of (ε p+1p+2 ,…,ε p+m ) is

则似然函数为Then the likelihood function is

进而似然方程为 Then the likelihood equation is

其中1≤i≤p,1≤l≤p。where 1≤i≤p, 1≤l≤p.

首先由于我们假定模型为p阶自回归模型,所以a1,a2,...,ap≠0,根据1≤i≤p可得εp+1=εp+2=ε2p=0。进而能够解得σ2的极大似然估计为:First of all, since we assume that the model is a p-order autoregressive model, a 1 , a 2 ,..., a p ≠ 0, according to 1≤i≤p can obtain ε p+1p+22p =0. Then the maximum likelihood estimate of σ 2 can be solved as:

其次根据1≤l≤p可得Secondly, according to 1≤l≤p can be obtained

当l=1,2,...,p时When l=1,2,...,p

其中εj=xj-a1xj-p-a2xj-p+1-...-apxj-1,将此式代入(19)得where ε j =x j -a 1 x jp -a 2 x j-p+1 -...-a p x j - 1 , substitute this formula into (19) to get

由(20)可解出参数a1,a2,...,ap的值,进而可以得到参数μ,Σ的值。From (20), the values of parameters a 1 , a 2 ,..., a p can be solved, and then the values of parameters μ and Σ can be obtained.

即目标函数g(w)min中,公式(9)的方程已通过建立AR(P)模型,并通过极大似然估计方法求解出未知参数,此时ai值即为wi值,代入方程(10)中,即可求出g(w)min最小时,输出样本值达到信号重构目的。此BP-ML方法估计参数较多,比较复杂,但是提高了BP算法的精度与误差。That is, in the objective function g(w) min , the equation of formula (9) has been established by the AR(P) model, and the unknown parameters have been solved by the maximum likelihood estimation method. At this time, the value of a i is the value of wi , which is substituted into In equation (10), the output sample value can be obtained when g(w) min is the smallest achieve the purpose of signal reconstruction. This BP-ML method estimates more parameters and is more complicated, but it improves the accuracy and error of the BP algorithm.

下面详细描述本发明的实施例,所述实施例是示例性的,旨在用于解释本发明,而不能理解为对本发明的限制。The embodiments of the present invention are described in detail below, and the embodiments are exemplary and intended to explain the present invention, but should not be construed as a limitation of the present invention.

训练神经网络时选取某批抽试验中的12发数据作为训练样本,能典型的代表固体火箭发动机试验状况。在具体应用中,因为上述发明的BP-ML算法估计权值过程较为复杂,在精度要求较高的情况下适用,因此,下文实例中还是应用BP方法实现。When training the neural network, 12 rounds of data from a certain batch of sampling tests are selected as training samples, which can typically represent the test conditions of solid rocket motors. In a specific application, because the BP-ML algorithm of the above invention is relatively complex in estimating the weight value process, it is applicable in the case of high precision requirements. Therefore, the BP method is still implemented in the following example.

判定标准judgement standard

恢复值的误差判断由信号均方根作为评判标准,The error judgment of the restored value is based on the root mean square of the signal as the judgment standard,

信号均方根(SMSE)公式如下:The signal root mean square (SMSE) formula is as follows:

根据神经网络得到恢复值的效果采用不同时延的相关系数(DifferentDelayCorrelation Coefficients,DDCC)来评定,可以判断恢复值和目标值的相关程度,DDCC值越接近于1,表明相关程度越高,恢复值得到的精度越高,建立的神经网络越准确。给定相关系数阈值为5%,即|1-C|≤0.05,则认为相关性高。According to the effect of the recovery value obtained by the neural network, the correlation coefficients of different delays (Different Delay Correlation Coefficients, DDCC) can be used to evaluate, and the degree of correlation between the recovery value and the target value can be judged. The higher the accuracy obtained, the more accurate the neural network established. Given that the correlation coefficient threshold is 5%, that is, |1-C|≤0.05, the correlation is considered to be high.

其中:sT为目标输出,为恢复值,τ为时延值。Where: s T is the target output, is the recovery value, and τ is the delay value.

下面对神经网络异常参数检验进行具体描述:The following is a detailed description of the abnormal parameter test of the neural network:

步骤一:建立神经网络,训练得出网络权重值;应用三层BP人工神经网络,选取十二发某批抽型号试验数据的最主要参数推力F与压强P作为研究对象,工作时间0s~18s,时间间隔选取0.2s,即每路点值个数均为91个,输入层神经元节点选取影响输出层节点相关参数,分别选择8发数据的推力F值与压强P值,输出层选择反映发动机工作状况的关键参数压强值P1作为目标输出,P1为第一发数据的压强值,即输入参数为P1,P2,P4,P5,P6,P7,P9,P10,F1,F2,F4,F5,F6,F7,F9,F10,8路压强,8路推力,共16路参数,输出目标值为P1,隐层神经元节点数选择需综合考虑训练效率与训练效果,并不是越大越好,若节点数过多,会增加运算量,造成训练过慢,甚至会过度训练,使得输出值发散;节点数过少,则会使得训练结果较差,输出不理想,因此,经过数据分析后,节点数选择为50,训练效果最理想,误差值最小;激活函数选择双极s形函数,因为其值域在[-1,1]之间,而最大压强值达到105kPa,因此将所有输入值与目标值均缩小10-5即可。Step 1: Establish a neural network and train to obtain the network weight value; apply a three-layer BP artificial neural network, select the thrust F and pressure P, the main parameters of the twelve batches of sampled test data, as the research objects, and the working time is 0s ~ 18s , the time interval is selected as 0.2s, that is, the number of each waypoint value is 91, the input layer neuron node selects the relevant parameters affecting the output layer node, respectively selects the thrust F value and the pressure P value of the 8 data, and the output layer selects the reflection The key parameter of the engine working condition is the pressure value P 1 as the target output, and P 1 is the pressure value of the first data, that is, the input parameters are P 1 , P 2 , P 4 , P 5 , P 6 , P 7 , P 9 , P 10 , F 1 , F 2 , F 4 , F 5 , F 6 , F 7 , F 9 , F 10 , 8 channels of pressure, 8 channels of thrust, a total of 16 channels of parameters, the output target value is P 1 , hidden layer neural The selection of the number of meta nodes needs to comprehensively consider the training efficiency and training effect. The bigger the better, the larger the number of nodes. If the number of nodes is too large, the amount of computation will increase, resulting in slow training, or even over-training, resulting in divergent output values; if the number of nodes is too small, It will make the training results poor and the output unsatisfactory. Therefore, after data analysis, the number of nodes is selected as 50, the training effect is the most ideal, and the error value is the smallest; the activation function selects a bipolar s-shaped function, because its value range is in [- 1,1], and the maximum pressure value reaches 10 5 kPa, so all input values and target values can be reduced by 10-5 .

初始权值矩阵赋随机数,开始训练网络,当误差达到预定要求后,将此时的权值保存,即可输入实际的发动机试验数据作为测试样本,代人权值计算,进行参数检测。The initial weight matrix is assigned random numbers to start training the network. When the error reaches the predetermined requirement, the weight at this time is saved, and the actual engine test data can be input as a test sample, which is calculated on behalf of the human value for parameter detection.

图3表示了目标值与修复值相关性,可以看出源信号与分离后信号相关程度较高,不管是趋势还是值域,均趋于一致;并且在不同的时延下,相关系数有不同程度的提高,在τ取7时,CCDD的值最大,表明相关程度最高。SMSE=2.8687。C=1.0008,即目标值与修复制的误差精度在0.08%。Figure 3 shows the correlation between the target value and the repair value. It can be seen that the source signal and the separated signal have a high degree of correlation, regardless of the trend or the value range, they are all consistent; and under different delays, the correlation coefficients are different. When τ is 7, the value of CCDD is the largest, indicating the highest degree of correlation. SMSE=2.8687. C=1.0008, that is, the error accuracy between the target value and the repair system is 0.08%.

为了验证此恢复值的可靠性,选取另外一发数据的压强P2作为输出,在τ取21时,得到相关系数为C=1.0005,即目标值与修复制的误差精度在0.05%,相关程度最高。In order to verify the reliability of this recovery value, the pressure P 2 of another data is selected as the output. When τ is 21, the correlation coefficient is C=1.0005, that is, the error accuracy between the target value and the repair system is 0.05%, and the degree of correlation is 0.05%. Highest.

步骤二:异常参数判别;为避免将正常数据误判为异常,检测时误差阈值应大于训练误差阈值。将此时的权值保存,即可输入实际的发动机试验数据作为测试样本,代人权值计算,进行参数检测。因为所收集到的某批抽型号试验数据的压强、推力数据作为主参数,均获得率为100%,无异常情况出现,因此我们将P9中的40个点值人为扩大十倍,构造异常的参数wP9,输入值P1,P2,P4,P5,P6,P7,wP9,P10,F1,F2,F4,F5,F6,F7,F9,F10,8路压强,8路推力,共16路参数,输出目标值为P1,进行神经网络修复,得到修复后的P1,无论时延值τ取任意值时,|1-C|≥0.05,即修复后的P1与目标值P1相关性较差,由此可反推出输入值中有异常参数。同理,在我们构建的神经网络模型的前提下,想判定某一发的推力F、压强P是否有异常,则将输入值中的推力压强值替换为需要判定的参数,输出目标值不变,任为P1,得到修复后的值,并进行比较,观察相关系数是否在规定的阈值内,即|1-C|≤0.05,若满足,则无异常参数,若不满足,则有异常参数,进一步依次替换检测值,判定具体的异常参数。Step 2: Discrimination of abnormal parameters; in order to avoid misjudging normal data as abnormal, the error threshold during detection should be greater than the training error threshold. By saving the weight at this time, the actual engine test data can be input as a test sample, which can be calculated on behalf of the human value for parameter detection. Because the pressure and thrust data of a certain batch of sampling test data collected are used as the main parameters, the acquisition rate is 100%, and there is no abnormal situation. Therefore, we artificially expand the 40 point values in P9 by ten times to construct abnormality. For parameters wP9 , input values P1, P2, P4, P5 , P6 , P7 , wP9 , P10 , F1 , F2 , F4 , F5 , F6 , F7 , F 9 , F 10 , 8 channels of pressure, 8 channels of thrust, a total of 16 parameters, the output target value is P 1 , and the neural network is repaired to obtain the repaired P 1 , no matter when the delay value τ takes any value, |1- C|≥0.05, that is, the correlation between the repaired P1 and the target value P1 is poor, so it can be deduced that there are abnormal parameters in the input value. In the same way, under the premise of the neural network model we constructed, if we want to determine whether the thrust F and pressure P of a certain engine are abnormal, then replace the thrust and pressure value in the input value with the parameter to be determined, and the output target value remains unchanged. , let it be P 1 , get the repaired value and compare it to observe whether the correlation coefficient is within the specified threshold, that is |1-C|≤0.05, if it is satisfied, there is no abnormal parameter, if not, there is abnormal parameters, and further replace the detected values in turn to determine specific abnormal parameters.

步骤三:异常参数修复;根据我们训练的神经网络模型,判定出wP9为异常参数后,需要进行异常数据修复,则将图1中的输出层神经元节点设为要恢复的目标参数wP9,输入层神经元节点中去掉目标参数,分别建立各关键参数的数据,重构神经网络,训练方式同前。Step 3: Repair of abnormal parameters; according to the neural network model we trained, after it is determined that wP 9 is an abnormal parameter, abnormal data needs to be repaired, then set the output layer neuron node in Figure 1 as the target parameter to be restored wP 9 , remove the target parameter from the input layer neuron node, establish the data of each key parameter separately, reconstruct the neural network, and the training method is the same as before.

本发明应用步骤:Application steps of the present invention:

第一步:判断异常点值;选取某次固体火箭发动机试验中的最主要参数-压强值P的稳态数据。首先检验数据的正态性,根据上文(1)、(2)分别计算测量数据的偏度、峰度,偏度越接近0,峰度越接近3时,表明正态分布特性良好,可以应用t检验;其次根据公式(3)、(4)、(5),完成t检验过程,判别出异常点并剔除。The first step: determine the abnormal point value; select the steady-state data of the pressure value P, the most important parameter in a certain solid rocket motor test. First check the normality of the data, and calculate the skewness and kurtosis of the measured data according to (1) and (2) above. Apply t-test; secondly, according to formulas (3), (4), (5), complete the t-test process, identify abnormal points and eliminate them.

第二步:训练神经网络,检验出异常参数,并进行参数值重构。Step 2: Train the neural network, detect abnormal parameters, and reconstruct the parameter values.

尽管上面已经示出和描述了本发明的实施例,可以理解的是,上述实施例是示例性的,不能理解为对本发明的限制,本领域的普通技术人员在不脱离本发明的原理和宗旨的情况下在本发明的范围内可以对上述实施例进行变化、修改、替换和变型。Although the embodiments of the present invention have been shown and described above, it should be understood that the above embodiments are exemplary and should not be construed as limiting the present invention. Variations, modifications, substitutions, and alterations to the above-described embodiments are possible within the scope of the present invention without departing from the scope of the present invention.

Claims (4)

1.一种固体火箭发动机中异常数据智能诊断方法,其特征在于:包括以下步骤:1. an abnormal data intelligent diagnosis method in a solid rocket motor, is characterized in that: comprise the following steps: 步骤1:采集固体火箭发动机试验数据,检验所选取的试验数据是否满足正态分布,若满足,则进行步骤2,否则选取其他试验数据重新进行检验;Step 1: Collect the test data of the solid rocket motor, and check whether the selected test data satisfies the normal distribution, if so, proceed to Step 2, otherwise select other test data for re-testing; 步骤2:采用t检验方法,判别出所选取的试验数据中的异常点并剔除;Step 2: Use the t test method to identify the abnormal points in the selected test data and eliminate them; 步骤3:采用BP-ML算法,利用神经网络对经过步骤2中t检验处理后的试验数据进行异常参数检验,若存在异常参数,则对异常参数进行参数值重构。Step 3: Using the BP-ML algorithm and using the neural network to test the abnormal parameters of the test data processed by the t test in step 2, if there are abnormal parameters, reconstruct the parameter values of the abnormal parameters. 2.根据权利要求1所述一种固体火箭发动机中异常数据智能诊断方法,其特征在于:步骤1中通过计算试验数据的偏度和峰度判断试验数据是否满足正态分布。2. The method for intelligently diagnosing abnormal data in a solid rocket motor according to claim 1, wherein in step 1, it is judged whether the test data satisfies normal distribution by calculating the skewness and kurtosis of the test data. 3.根据权利要求1所述一种固体火箭发动机中异常数据智能诊断方法,其特征在于:步骤3中BP-ML算法是通过将人工神经网络BP算法中的权向量估计方法采用最大似然估计方法实现而得到。3. the abnormal data intelligent diagnosis method in a kind of solid rocket motor according to claim 1, is characterized in that: in step 3, BP-ML algorithm is to adopt maximum likelihood estimation by the weight vector estimation method in artificial neural network BP algorithm method is achieved. 4.根据权利要求3所述一种固体火箭发动机中异常数据智能诊断方法,其特征在于:步骤3中进行异常参数检验的具体过程为:4. the abnormal data intelligent diagnosis method in a kind of solid rocket motor according to claim 3, is characterized in that: the concrete process that carries out abnormal parameter inspection in step 3 is: 步骤3.1:采用无异常参数的数据作为样本数据训练神经网络,得到训练好的神经网络;Step 3.1: Use the data without abnormal parameters as sample data to train the neural network to obtain a trained neural network; 步骤3.2:将实际试验数据输入神经网络,进行神经网络修复,若修复后的输出值与目标值之间相关性差,则推断神经网络输入值中存在异常参数;若神经网络输入值中存在异常参数,则对神经网络输入值中的参数值分别进行判断,以得到具体的异常参数;Step 3.2: Input the actual test data into the neural network to repair the neural network. If the correlation between the repaired output value and the target value is poor, it is inferred that there are abnormal parameters in the input value of the neural network; if there are abnormal parameters in the input value of the neural network , then the parameter values in the input value of the neural network are judged separately to obtain specific abnormal parameters; 步骤3.3:以步骤3.2得到的异常参数作为神经网络输出参数,在神经网络输入参数中去掉异常参数,然后重构神经网络,得到修复后的异常参数。Step 3.3: Use the abnormal parameters obtained in step 3.2 as the output parameters of the neural network, remove the abnormal parameters from the input parameters of the neural network, and then reconstruct the neural network to obtain the repaired abnormal parameters.
CN201910224959.6A 2019-03-24 2019-03-24 Intelligent diagnosis method for abnormal data in solid rocket engine Active CN109779791B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201910224959.6A CN109779791B (en) 2019-03-24 2019-03-24 Intelligent diagnosis method for abnormal data in solid rocket engine

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201910224959.6A CN109779791B (en) 2019-03-24 2019-03-24 Intelligent diagnosis method for abnormal data in solid rocket engine

Publications (2)

Publication Number Publication Date
CN109779791A true CN109779791A (en) 2019-05-21
CN109779791B CN109779791B (en) 2021-01-01

Family

ID=66490920

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201910224959.6A Active CN109779791B (en) 2019-03-24 2019-03-24 Intelligent diagnosis method for abnormal data in solid rocket engine

Country Status (1)

Country Link
CN (1) CN109779791B (en)

Cited By (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN110239745A (en) * 2019-06-13 2019-09-17 北京深蓝航天科技有限公司 The multiple-motor parallel connection rocket control device and control method for having power redundant ability
CN110414063A (en) * 2019-06-29 2019-11-05 万翼科技有限公司 Model restorative procedure and Related product
CN110531622A (en) * 2019-09-05 2019-12-03 沈阳航空航天大学 A kind of Thrust of Solid Rocket Motor control method based on radial base neural net
CN111259927A (en) * 2020-01-08 2020-06-09 西北工业大学 Rocket engine fault diagnosis method based on neural network and evidence theory

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101907088A (en) * 2010-05-27 2010-12-08 中国人民解放军国防科学技术大学 Fault Diagnosis Method Based on Single Class Support Vector Machine
CN104239698A (en) * 2014-09-01 2014-12-24 中国航天科技集团公司第四研究院四〇一所 Time sequence correcting method of solid rocket engine vibration distortion signal
CN106460727A (en) * 2014-04-03 2017-02-22 赛峰飞机发动机公司 Method and device for monitoring a parameter of a rocket engine
CN108520301A (en) * 2018-04-17 2018-09-11 重庆大学 A Deep Belief Network Based Intermittent Fault Diagnosis Method for Circuits
CN108803554A (en) * 2017-05-04 2018-11-13 上海电气集团股份有限公司 A kind of intelligent diagnosing method of generator signal

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101907088A (en) * 2010-05-27 2010-12-08 中国人民解放军国防科学技术大学 Fault Diagnosis Method Based on Single Class Support Vector Machine
CN106460727A (en) * 2014-04-03 2017-02-22 赛峰飞机发动机公司 Method and device for monitoring a parameter of a rocket engine
CN104239698A (en) * 2014-09-01 2014-12-24 中国航天科技集团公司第四研究院四〇一所 Time sequence correcting method of solid rocket engine vibration distortion signal
CN108803554A (en) * 2017-05-04 2018-11-13 上海电气集团股份有限公司 A kind of intelligent diagnosing method of generator signal
CN108520301A (en) * 2018-04-17 2018-09-11 重庆大学 A Deep Belief Network Based Intermittent Fault Diagnosis Method for Circuits

Cited By (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN110239745A (en) * 2019-06-13 2019-09-17 北京深蓝航天科技有限公司 The multiple-motor parallel connection rocket control device and control method for having power redundant ability
CN110414063A (en) * 2019-06-29 2019-11-05 万翼科技有限公司 Model restorative procedure and Related product
CN110531622A (en) * 2019-09-05 2019-12-03 沈阳航空航天大学 A kind of Thrust of Solid Rocket Motor control method based on radial base neural net
CN110531622B (en) * 2019-09-05 2022-04-05 沈阳航空航天大学 Thrust control method of solid rocket engine based on radial basis function neural network
CN111259927A (en) * 2020-01-08 2020-06-09 西北工业大学 Rocket engine fault diagnosis method based on neural network and evidence theory
CN111259927B (en) * 2020-01-08 2022-08-05 西北工业大学 Rocket engine fault diagnosis method based on neural network and evidence theory

Also Published As

Publication number Publication date
CN109779791B (en) 2021-01-01

Similar Documents

Publication Publication Date Title
CN112731159B (en) A method for predicting and locating battery faults in battery compartments of energy storage power stations
CN109779791A (en) An intelligent diagnosis method for abnormal data in solid rocket motor
CN110807257A (en) Method for predicting residual life of aircraft engine
CN109472110A (en) A prediction method for remaining service life of aero-engine based on LSTM network and ARIMA model
CN113569338A (en) Gas compressor rotating stall early warning method based on time expansion convolutional network
CN111914883A (en) A method and device for evaluating the state of a spindle bearing based on a deep fusion network
CN112327191B (en) A battery diving probability evaluation method based on geometric feature fusion decision
CN108664690A (en) Long-life electron device reliability lifetime estimation method under more stress based on depth belief network
CN104751229A (en) Bearing fault diagnosis method capable of recovering missing data of back propagation neural network estimation values
CN105116872A (en) Fault diagnosis method based on metric learning and time sequence during industrial process
CN110533294B (en) Nuclear power plant operation fault alarm method based on artificial intelligence technology
CN115950609B (en) A Bridge Deflection Anomaly Detection Method Combining Correlation Analysis and Neural Network
CN117521512A (en) Bearing residual service life prediction method based on multi-scale Bayesian convolution transducer model
CN113988210B (en) Method, device and storage medium for repairing distorted data of structural monitoring sensor network
CN111273126B (en) Power distribution network topology rapid sensing method
CN104035431A (en) Obtaining method and system for kernel function parameters applied to nonlinear process monitoring
CN118504627B (en) Method, device, equipment and medium for generating and predicting battery performance prediction model
CN118133431B (en) Aircraft wing structure load identification method based on multi-source data fusion
CN115128469A (en) Capacity degradation analysis and online estimation method of power lithium battery based on digital twin
CN111046331A (en) Health monitoring method based on signal abnormity
CN114511025A (en) Fan fault diagnosis method and device based on weighted multi-sensor fusion filtering
CN119535227A (en) A lithium battery health state estimation method based on multi-coefficient fusion
CN119226973A (en) A method and device for detecting abnormal noise of DC transmission converter valve based on deep learning
CN117709106A (en) A mechanical structure excitation identification method based on MAP estimator
CN104200269A (en) Real-time fault diagnosis method based on online learning minimum embedding dimension network

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination
GR01 Patent grant
GR01 Patent grant