Research Institute on General Development and Evaluation of Equipment

Beijing, China

Research Institute on General Development and Evaluation of Equipment

Beijing, China
SEARCH FILTERS
Time filter
Source Type

Zheng N.,PLA Air Force Aviation University | Zhang L.,PLA Air Force Aviation University | Zhang B.,PLA Air Force Aviation University | Zhao Y.,Research Institute on General Development and Evaluation of Equipment | Si W.,PLA Air Force Aviation University
Proceedings of 2016 Prognostics and System Health Management Conference, PHM-Chengdu 2016 | Year: 2016

According to the characteristics of power system fault data, a new attribute reduction method of rough set is proposed, which is used to diagnose the power system fault of complex equipment. First, the attribute reduction problem is transformed to the set covering problem and the correlation coefficient matrix is built based on the correlation matrix. Then based on the selection principle of high quality feature set, the random divergence mechanism is introduced, the disturbance searching algorithm is designed and the result of attribute reduction is obtained. At last, the UCI data set and one analog circuit are used to confirm the algorithm. The simulation results indicate that the fault diagnosis time is reduced greatly and the accurate rate is more than 91%, which showed the practicability of the algorithm. © 2016 IEEE.


Wang J.,Jilin University | Wang J.,Research Institute on General Development and Evaluation of Equipment | Du W.,Jilin University | Yu C.,CAS Changchun Institute of Applied Chemistry | Liang Y.,Jilin University
Journal of Bionanoscience | Year: 2013

With the increasing number of protein sequences without known structures, it is important and urgent to realize many aspects of new proteins in experimental methods or computational methods. In this study, protein domains are considered as the basic units of protein folding, evolution, and function. Support vector machine (SVM) method is used to predict the domain boundary of protein. Protein profiles, predicted structures, predicted relative solvent accessibility and carbon atomic coordinates are applied in the prediction model. The testing and training set contains 125 unique protein sequences (no sequences folds were present in both the testing and training sets). We have achieved a sensitivity of 67% and specificity of 40.5% for multi-domain protein chains with a 10-fold cross validation by using this model. Copyright © 2013 American Scientific Publishers.


Xing C.,Changchun University | Xing C.,Jilin University | Wan L.,Research Institute on General Development and Evaluation of Equipment | Wang J.,Jilin University | Liang Y.,Jilin University
Journal of Bionanoscience | Year: 2013

In this paper, on the basis of principal component analysis, we use least squares support vector machine (LS-SVM) to predict tRNA. Appearance frequencies of single nucleotide, 2-nucleotides, (G-C)% and (A-T)% were chosen as characteristic inputs. Results from tests showed that the prediction accuracy was 90.51% on prokaryotic tRNA dataset. The results indicate that the proposed method is adoptable for prokaryotic ncRNA prediction. Copyright © 2013 American Scientific Publishers.


Zhou C.,Jilin University | Wan L.,Research Institute on General Development and Evaluation of Equipment | Liang Y.,Jilin University
ICACTE 2010 - 2010 3rd International Conference on Advanced Computer Theory and Engineering, Proceedings | Year: 2010

Classification and prediction of different cancers based on gene expression profiles are important for cancer diagnosis, cancer treatment and medication discovery. The k nearest neighbor algorithm (k-NN) is one easy and efficient machine learning method for cancer classification and the parameter k is crucial. In this paper, we integrate minimum spanning tree (MST) and k nearest neighbor algorithm (k-NN) for cancer classification. The MST is designed for the selection of parameter k and the nearest neighbors for k-NN. Firstly we build a minimum spanning tree (MST) based on Euclidean distance between each two samples for gene expression data only including one unknown class sample. Secondly for unknown class sample in the gene expression data, we find the connected samples and then apply majority vote principle. Thirdly if there are tied votes then we expend the connected samples with the nearest neighbors for unknown class sample until all the samples are expended or the class for unknown sample is obtained. This hybrid algorithm is referred to as MSTNN. The hybrid algorithm MSTNN is compared with kNN and other 3 existing classification algorithms on CNS dataset, Colon dataset and Lung dataset, 3 binary class gene expression datasets and 3 multi-class gene expression datasets: Leukemial, Leukemia2, and Leukemia3 invol.ving human cancers. The MSTNN algorithm improves 5.65% better than k-NN on average LOOCV accuracy and 13.80% better than kNN on testing datasets classification average accuracy, and achieves the best performance in all the 5 algorithms. The results demonstrate that the proposed MSTNN algorithm is feasible to classify human cancers. © 2010 IEEE.


Wang Y.,Jilin University | Wu C.,Jilin University | Wu C.,CAS Institute of Automation | Wan L.,Research Institute on General Development and Evaluation of Equipment | Liang Y.,Jilin University
2010 The 2nd International Conference on Computer and Automation Engineering, ICCAE 2010 | Year: 2010

When faults occur in power systems, it is hard to manually deal with the fault data reported by the system of supervisory control and data acquisition (SCADA) because of the huge amount of alarm information. In this paper, we study the problem of power system fault diagnosis by using support vector machine (SVM), and enhance the ability of fault diagnosis through optimizing support vectors. The results of simulation tests demonstrate the effectiveness of the proposed automatic fault diagnosis method. ©2010 IEEE.


Sun B.-C.,Research Institute on General Development and Evaluation of Equipment | Sun B.-C.,Beijing Institute of Technology | Fan J.-F.,Beijing Information Science and Technology University
Chinese Control Conference, CCC | Year: 2013

A hybrid differentiator based on linear and nonlinear parts is introduced. A deeply strapdown guidance and control (DSGC) system for low-cost munitions is presented using body-fixed seeker and miniature inertial measurement unit (MIMU). The three-dimension line-of-sight (LOS) dynamics for guidance command are described using the missile-target geometry. Due to the weakness of strapdown seeker, the aforementioned hybrid differentiator is then introduced to obtain and filter the inertial LOS rate for proportional guidance law. The analysis and trajectory simulation results verify the performance of the proposed algorithm. © 2013 TCCT, CAA.


Sun B.,Research Institute on General Development and Evaluation of Equipment | Sun B.,Beijing Institute of Technology | Fan J.,Beijing Information Science and Technology University
Hongwai yu Jiguang Gongcheng/Infrared and Laser Engineering | Year: 2014

A novel deeply strapdown guidance & control (DSGC) framework based on strapdown seeker and strapdown inertial measurement unit was presented for miniature guided munitions applied to UAV, soldier shoulder and other platforms. Due to that there was no available guidance information of DSGC, the theoretic transform equations based on missile-target geometric relationship and roll -stabilized simplified equations were introduced, then a nonlinear differentiator was proposed for stable digital angular rate calculation. A PI compensated two -loop acceleration autopilot was presented as the inner loop of guidance system. The trajectory simulation for UAV platform and soldier shoulder was analyzed, respectively. The results showed that the DSGC technology was a well solution for the precisely guided miniature munitions especially the short-range. Wherein, the multiple-constraints design for guidance & control system was the critical factor for application.


Yin S.,CAS Changchun Institute of Optics and Fine Mechanics and Physics | Ming B.,Research Institute on General Development and Evaluation of Equipment | Gao H.,CAS Changchun Institute of Optics and Fine Mechanics and Physics | Wang T.,CAS Changchun Institute of Optics and Fine Mechanics and Physics
Hongwai yu Jiguang Gongcheng/Infrared and Laser Engineering | Year: 2012

To depress the guidance message noise of air-to-surface semi-strapdown imaging seeker, the control loop model of stability tracking was established, the transform characteristic of various error and output was derived. The power spectrum response of main noise was obtained. Then, the weighted angular rate of line of sight was gained. In order to verify the validity of the weighted angular rate of line of sight, the simulation were completed in the guidance loop. The results show that the weighted method can effectively compressed the noise measurement information of seeker. The semi-strapdown TV seeker was achieved with semi-physical way, the double closed-loop was completed in TV seeker to make disturbance isolation and track the target. The methods for extracting guided information through weighted angular rate of line of sight and direct extraction methods were compared. The experimental results show that the weighted extraction method of line of sight angular rate noise variance is 0.098 (°)/s, which is lower than the direct synthesis of noise of 46%. The weighted method can improve the problem of large semi-strapdown sight angular rate noise.


Fan J.-F.,Beijing Information Science and Technology University | Sun B.-C.,Research Institute on General Development and Evaluation of Equipment | Sun B.-C.,Beijing Institute of Technology | Su Z.,Beijing Information Science and Technology University | Dong S.-Y.,Ordnance Engineering College
Chinese Control Conference, CCC | Year: 2012

The available feedback topologies for tactical missile autopilot were detailed and examined using optimal control. There were transformations that could provide equivalent closed loop properties for the different topologies. The open loop properties of autopilots would be different, however. Compared with the limitations of traditional stability margins, a novel vector margin approach was introduced to establish a standard performance level and determine the best autopilot topology from a robustness perspective. A static unstable plant as example was used throughout. The analysis and simulation results validate the presented approach. © 2012 Chinese Assoc of Automati.


Zhheng N.,PLA Air Force Aviation University | Zhang L.,PLA Air Force Aviation University | Zhang B.,PLA Air Force Aviation University | Wang W.,PLA Air Force Aviation University | And 2 more authors.
MATEC Web of Conferences | Year: 2016

At present, the fault signals of surface to air missile equipment are hard to collect and the accuracy of fault diagnosis is very low. To solve the above problems, based on the superiority of wavelet transformation on processing non-stationary signals and the advantage of SVM on pattern classification, this paper proposes a fault diagnosis model and takes the typical analog circuit diagnosis of one power distribution system as an example to verify the fault diagnosis model based on Wavelet Transformation and SVM. The simulation results show that the model is able to achieve fault diagnosis based on a small amount of training samples, which improves the accuracy of fault diagnosis. © 2016 Owned by the authors, published by EDP Sciences.

Loading Research Institute on General Development and Evaluation of Equipment collaborators
Loading Research Institute on General Development and Evaluation of Equipment collaborators