In previous researches on a model-based diagnostic system, the components are assumed mutually independent. Howerver , the assumption is not always the case because the information about whether a component is faulty ...In previous researches on a model-based diagnostic system, the components are assumed mutually independent. Howerver , the assumption is not always the case because the information about whether a component is faulty or not usually influences our knowledge about other components. Some experts may draw such a conclusion that 'if component m 1 is faulty, then component m 2 may be faulty too'. How can we use this experts' knowledge to aid the diagnosis? Based on Kohlas's probabilistic assumption-based reasoning method, we use Bayes networks to solve this problem. We calculate the posterior fault probability of the components in the observation state. The result is reasonable and reflects the effectiveness of the experts' knowledge.展开更多
This paper proposes a novel locally linear backpropagation based contribution(LLBBC) for nonlinear process fault diagnosis. As a method based on the deep learning model of auto-encoder(AE), LLBBC can deal with the fau...This paper proposes a novel locally linear backpropagation based contribution(LLBBC) for nonlinear process fault diagnosis. As a method based on the deep learning model of auto-encoder(AE), LLBBC can deal with the fault diagnosis problem through extracting nonlinear features. When the on-line fault diagnosis task is in progress, a locally linear model is firstly built at the current fault sample. According to the basic idea of reconstruction based contribution(RBC), the propagation of fault information is described by using back-propagation(BP) algorithm. Then, a contribution index is established to measure the correlation between the variable and the fault, and the final diagnosis result is obtained by searching variables with large contributions. The smearing effect, which is an important factor affecting the performance of fault diagnosis, can be suppressed as well,and the theoretical analysis reveals that the correct diagnosis can be guaranteed by LLBBC. Finally, the feasibility and effectiveness of the proposed method are verified through a nonlinear numerical example and the Tennessee Eastman benchmark process.展开更多
Applying calculation method in alloy design should be an important tendency due to its characters of inexpensive cost, high efficiency and prediction. DOS calculations of AuSn, AsSn and SbSn Sn- based alloys have ...Applying calculation method in alloy design should be an important tendency due to its characters of inexpensive cost, high efficiency and prediction. DOS calculations of AuSn, AsSn and SbSn Sn- based alloys have been investigated by employing DV - Xa method, in which different cluster models were adopted to calculate electron structure.It is proved that some regulations must be taken into ac- count in order to carry out alloy design calculation successfully,which are described in this paper in detail.展开更多
Monitoring high-dimensional multistage processes becomes crucial to ensure the quality of the final product in modern industry environments. Few statistical process monitoring(SPC) approaches for monitoring and contro...Monitoring high-dimensional multistage processes becomes crucial to ensure the quality of the final product in modern industry environments. Few statistical process monitoring(SPC) approaches for monitoring and controlling quality in highdimensional multistage processes are studied. We propose a deviance residual-based multivariate exponentially weighted moving average(MEWMA) control chart with a variable selection procedure. We demonstrate that it outperforms the existing multivariate SPC charts in terms of out-of-control average run length(ARL) for the detection of process mean shift.展开更多
Model-based diagnosis(MBD)with multiple observations shows its significance in identifying fault location.The existing approaches for MBD with multiple observations use observations which is inconsistent with the pred...Model-based diagnosis(MBD)with multiple observations shows its significance in identifying fault location.The existing approaches for MBD with multiple observations use observations which is inconsistent with the prediction of the system.In this paper,we proposed a novel diagnosis approach,namely,the Diagnosis with Different Observations(DiagDO),to exploit the diagnosis when given a set of pseudo normal observations and a set of abnormal observations.Three ideas are proposed in this paper.First,for each pseudo normal observation,we propagate the value of system inputs and gain fanin-free edges to shrink the size of possible faulty components.Second,for each abnormal observation,we utilize filtered nodes to seek surely normal components.Finally,we encode all the surely normal components and parts of dominated components into hard clauses and compute diagnosis using the MaxSAT solver and MCS algorithm.Extensive tests on the ISCAS'85 and ITC'99 benchmarks show that our approach performs better than the state-of-the-art algorithms.展开更多
This paper develops a new fault diagnosis and tolerant control framework of sensor failure(SFDTC)for complex system such as rockets and missiles.The new framework aims to solve two problems:The lack of data and the mu...This paper develops a new fault diagnosis and tolerant control framework of sensor failure(SFDTC)for complex system such as rockets and missiles.The new framework aims to solve two problems:The lack of data and the multiple uncertainty of knowledge.In the SFDTC framework,two parts exist:The fault diagnosis model and the output reconstruction model.These two parts of the new framework are constructed based on the new developed belief rule base with power set(BRB-PS).The multiple uncertainty of knowledge can be addressed by the local ignorance and global ignorance in the new developed BRB-PS model.Then,the stability of the developed framework is proved by the output error of the BRB-PS model.For complex system,the sensor state is determined by many factors and experts cannot provide accurate knowledge.The multiple uncertain knowledge will reduce the performance of the initial SDFTC framework.Therefore,in the SFDTC framework,to handle the influence of the uncertainty of expert knowledge and improve the framework performance,a new optimization model with two optimization goals is developed to ensure the smallest output uncertainty and the highest accuracy simultaneously.A case study is conducted to illustrate the effectiveness of the developed framework.展开更多
For the Purpose of obtaining the best measurements quickly in 'model based diagnosis', we constructed binary structure models, which contain only the components of candidates. The relative causality between t...For the Purpose of obtaining the best measurements quickly in 'model based diagnosis', we constructed binary structure models, which contain only the components of candidates. The relative causality between the components of the models is the same as the faulty device. Candidates are classified by structure information. The models can be directly applied to either a discrete or an analog device without any additional processing. Then a half split method is set forward, which gives an optimal measurement within O(N 2)′s computations. The algorithm used in this paper can be regarded as the extremal structure characteristics of de Kleer′s expected entropy algorithm.展开更多
Extracting, transportation and the using from fossil fuels can damage to the hydrosphere, the biosphere and the Earth＇s atmosphere. But humans always need to this valuable substance. The production of oil derivatives...Extracting, transportation and the using from fossil fuels can damage to the hydrosphere, the biosphere and the Earth＇s atmosphere. But humans always need to this valuable substance. The production of oil derivatives by means of forest waste and coal through the Fischer-Tropsch process is an appropriate solution for the cleanliness of all parts of the environment. For the production of favorite products by the synthesis of Fischer-Tropsch, the performance of the catalyst under different operating conditions should be predictable. For this reason, in this paper, eight mathematical models were determined for the selectivity of five products of methane, light hydrocarbons, gasoline, diesel and wax based on three factors of reduction temperature, time on stream, and He/CO ratio inlet gas on iron-based catalyst. The results showed that the reduction temperature factor had the most effective on the selectivity of hydrocarbon products, exception diesel, so that the increase of the reduction temperature led to increase of the selectivity of methane, light hydrocarbons, gasoline and reduce of the degree of selectivity of the wax and vice versa. For the diesel selectivity, factor of the He/CO ratio inlet gas was the most effective than other factors.展开更多
Objective To investgate the value of various parameters obtained from monoexponential,biexponential,and stretched exponential diffusion-weighted imaging models in the differential diagnosis of breast lesions.Methods A...Objective To investgate the value of various parameters obtained from monoexponential,biexponential,and stretched exponential diffusion-weighted imaging models in the differential diagnosis of breast lesions.Methods A retrospective study was performed in 54 patients with pathologically confirmed malignant tumors(n=30),benign lesions(n=34)and normal fibroglandular展开更多
基于变量预测模型的分类识别(Variable predictive model-based class discriminate,VPMCD)方法是一种新的分类识别方法,但模型类型的选择存在主观性。为了解决VPMCD方法应用于机械故障诊断过程中的模型选择问题,结合遗传算法的全局优...基于变量预测模型的分类识别(Variable predictive model-based class discriminate,VPMCD)方法是一种新的分类识别方法,但模型类型的选择存在主观性。为了解决VPMCD方法应用于机械故障诊断过程中的模型选择问题,结合遗传算法的全局优化能力,提出了基于GA-VPMCD(Genetic algorithm and variable predictive model based class discriminate)智能诊断方法。首先通过样本训练建立多个弱VPM(Variable predictive model),然后采用遗传算法优化各个弱VPM的权值,得到最优权值矩阵,最后用最优权值矩阵加权融合测试样本的弱VPM特征变量预测值,得到最佳特征变量预测值,并以误差平方和最小为辨别函数分类识别故障类型。通过GA-VPMCD方法在滚动轴承故障智能诊断中的应用实验验证了基于GA-VPMCD的故障智能诊断方法能有效地提高诊断精度和诊断系统的鲁棒性。展开更多
多变量预测模型模式识别(variable predictive model based class discriminate,简称VPMCD)利用样本特征值内在的相关性来建立特征学习模型,但是当训练样本较少时会导致模型预测不准确,因此提出了基于递归定量分析(recurrence quantific...多变量预测模型模式识别(variable predictive model based class discriminate,简称VPMCD)利用样本特征值内在的相关性来建立特征学习模型,但是当训练样本较少时会导致模型预测不准确,因此提出了基于递归定量分析(recurrence quantification analysis,简称RQA)和投票法多变量预测模型模式识别(voted variable predictive model based class discriminate,简称V-VPMCD)的故障识别方法。该方法利用了递归定量分析对非线性、非平稳信号分析的鲁棒性和样本质量不高时处理的优势,以VPMCD作为分类方法,并用投票法优化了VPMCD方法,提升了算法的稳定性和识别率。对滚动轴承不同程度、不同类型故障的模式识别实验表明,该优化算法具有较高的识别准确率和稳定性。展开更多
文摘In previous researches on a model-based diagnostic system, the components are assumed mutually independent. Howerver , the assumption is not always the case because the information about whether a component is faulty or not usually influences our knowledge about other components. Some experts may draw such a conclusion that 'if component m 1 is faulty, then component m 2 may be faulty too'. How can we use this experts' knowledge to aid the diagnosis? Based on Kohlas's probabilistic assumption-based reasoning method, we use Bayes networks to solve this problem. We calculate the posterior fault probability of the components in the observation state. The result is reasonable and reflects the effectiveness of the experts' knowledge.
基金supported by the Key Project of National Natural Science Foundation of China(61933013)Ningbo 13th Five-year Marine Economic Innovation and Development Demonstration Project(NBH Y-2017-Z1)。
文摘This paper proposes a novel locally linear backpropagation based contribution(LLBBC) for nonlinear process fault diagnosis. As a method based on the deep learning model of auto-encoder(AE), LLBBC can deal with the fault diagnosis problem through extracting nonlinear features. When the on-line fault diagnosis task is in progress, a locally linear model is firstly built at the current fault sample. According to the basic idea of reconstruction based contribution(RBC), the propagation of fault information is described by using back-propagation(BP) algorithm. Then, a contribution index is established to measure the correlation between the variable and the fault, and the final diagnosis result is obtained by searching variables with large contributions. The smearing effect, which is an important factor affecting the performance of fault diagnosis, can be suppressed as well,and the theoretical analysis reveals that the correct diagnosis can be guaranteed by LLBBC. Finally, the feasibility and effectiveness of the proposed method are verified through a nonlinear numerical example and the Tennessee Eastman benchmark process.
文摘Applying calculation method in alloy design should be an important tendency due to its characters of inexpensive cost, high efficiency and prediction. DOS calculations of AuSn, AsSn and SbSn Sn- based alloys have been investigated by employing DV - Xa method, in which different cluster models were adopted to calculate electron structure.It is proved that some regulations must be taken into ac- count in order to carry out alloy design calculation successfully,which are described in this paper in detail.
基金supported by the Qatar National Research Fund(NPRP5-364-2-142NPRP7-1040-2-293)
文摘Monitoring high-dimensional multistage processes becomes crucial to ensure the quality of the final product in modern industry environments. Few statistical process monitoring(SPC) approaches for monitoring and controlling quality in highdimensional multistage processes are studied. We propose a deviance residual-based multivariate exponentially weighted moving average(MEWMA) control chart with a variable selection procedure. We demonstrate that it outperforms the existing multivariate SPC charts in terms of out-of-control average run length(ARL) for the detection of process mean shift.
基金supported by the National Natural Science Foundation of China(Grant Nos.62076108,61972360,and 61872159).
文摘Model-based diagnosis(MBD)with multiple observations shows its significance in identifying fault location.The existing approaches for MBD with multiple observations use observations which is inconsistent with the prediction of the system.In this paper,we proposed a novel diagnosis approach,namely,the Diagnosis with Different Observations(DiagDO),to exploit the diagnosis when given a set of pseudo normal observations and a set of abnormal observations.Three ideas are proposed in this paper.First,for each pseudo normal observation,we propagate the value of system inputs and gain fanin-free edges to shrink the size of possible faulty components.Second,for each abnormal observation,we utilize filtered nodes to seek surely normal components.Finally,we encode all the surely normal components and parts of dominated components into hard clauses and compute diagnosis using the MaxSAT solver and MCS algorithm.Extensive tests on the ISCAS'85 and ITC'99 benchmarks show that our approach performs better than the state-of-the-art algorithms.
基金supported in part by the Natural Science Foundation of China under Grant Nos. 61370031,61374138, 61973046, 61833013, 61773389 and 71601168the Fundamental Research Funds for the Central Universities under Grant No. D5000210690+1 种基金the Shaanxi Outstanding Youth Science Foundation under Grant No.2020JC-34the Natural Science Foundation of Shaanxi Province under Grant Nos. 2020JM-357, 2022JQ-580,2021KJXX-22 and 2020JQ-298
文摘This paper develops a new fault diagnosis and tolerant control framework of sensor failure(SFDTC)for complex system such as rockets and missiles.The new framework aims to solve two problems:The lack of data and the multiple uncertainty of knowledge.In the SFDTC framework,two parts exist:The fault diagnosis model and the output reconstruction model.These two parts of the new framework are constructed based on the new developed belief rule base with power set(BRB-PS).The multiple uncertainty of knowledge can be addressed by the local ignorance and global ignorance in the new developed BRB-PS model.Then,the stability of the developed framework is proved by the output error of the BRB-PS model.For complex system,the sensor state is determined by many factors and experts cannot provide accurate knowledge.The multiple uncertain knowledge will reduce the performance of the initial SDFTC framework.Therefore,in the SFDTC framework,to handle the influence of the uncertainty of expert knowledge and improve the framework performance,a new optimization model with two optimization goals is developed to ensure the smallest output uncertainty and the highest accuracy simultaneously.A case study is conducted to illustrate the effectiveness of the developed framework.
文摘For the Purpose of obtaining the best measurements quickly in 'model based diagnosis', we constructed binary structure models, which contain only the components of candidates. The relative causality between the components of the models is the same as the faulty device. Candidates are classified by structure information. The models can be directly applied to either a discrete or an analog device without any additional processing. Then a half split method is set forward, which gives an optimal measurement within O(N 2)′s computations. The algorithm used in this paper can be regarded as the extremal structure characteristics of de Kleer′s expected entropy algorithm.
文摘Extracting, transportation and the using from fossil fuels can damage to the hydrosphere, the biosphere and the Earth＇s atmosphere. But humans always need to this valuable substance. The production of oil derivatives by means of forest waste and coal through the Fischer-Tropsch process is an appropriate solution for the cleanliness of all parts of the environment. For the production of favorite products by the synthesis of Fischer-Tropsch, the performance of the catalyst under different operating conditions should be predictable. For this reason, in this paper, eight mathematical models were determined for the selectivity of five products of methane, light hydrocarbons, gasoline, diesel and wax based on three factors of reduction temperature, time on stream, and He/CO ratio inlet gas on iron-based catalyst. The results showed that the reduction temperature factor had the most effective on the selectivity of hydrocarbon products, exception diesel, so that the increase of the reduction temperature led to increase of the selectivity of methane, light hydrocarbons, gasoline and reduce of the degree of selectivity of the wax and vice versa. For the diesel selectivity, factor of the He/CO ratio inlet gas was the most effective than other factors.
文摘Objective To investgate the value of various parameters obtained from monoexponential,biexponential,and stretched exponential diffusion-weighted imaging models in the differential diagnosis of breast lesions.Methods A retrospective study was performed in 54 patients with pathologically confirmed malignant tumors(n=30),benign lesions(n=34)and normal fibroglandular
文摘基于变量预测模型的分类识别(Variable predictive model-based class discriminate,VPMCD)方法是一种新的分类识别方法,但模型类型的选择存在主观性。为了解决VPMCD方法应用于机械故障诊断过程中的模型选择问题,结合遗传算法的全局优化能力,提出了基于GA-VPMCD(Genetic algorithm and variable predictive model based class discriminate)智能诊断方法。首先通过样本训练建立多个弱VPM(Variable predictive model),然后采用遗传算法优化各个弱VPM的权值,得到最优权值矩阵,最后用最优权值矩阵加权融合测试样本的弱VPM特征变量预测值,得到最佳特征变量预测值,并以误差平方和最小为辨别函数分类识别故障类型。通过GA-VPMCD方法在滚动轴承故障智能诊断中的应用实验验证了基于GA-VPMCD的故障智能诊断方法能有效地提高诊断精度和诊断系统的鲁棒性。
文摘多变量预测模型模式识别(variable predictive model based class discriminate,简称VPMCD)利用样本特征值内在的相关性来建立特征学习模型,但是当训练样本较少时会导致模型预测不准确,因此提出了基于递归定量分析(recurrence quantification analysis,简称RQA)和投票法多变量预测模型模式识别(voted variable predictive model based class discriminate,简称V-VPMCD)的故障识别方法。该方法利用了递归定量分析对非线性、非平稳信号分析的鲁棒性和样本质量不高时处理的优势,以VPMCD作为分类方法,并用投票法优化了VPMCD方法,提升了算法的稳定性和识别率。对滚动轴承不同程度、不同类型故障的模式识别实验表明,该优化算法具有较高的识别准确率和稳定性。
