Abstract:Comparison analysis is a key tool to analyze signal and to detect faults in engineering. There are three different kinds of algorithms are set up to compare signals. Based on these new algorithms, several practical approaches are built to detach gradual change and to diagnose abrupt faults. A practical example is given from a space-flight engineering mission to show that methods given in this paper are valuable.

Abstract:A fault trend prediction problem for a class of exponential fault process is studied. Under the condition that the measure variable is disturbed by a stationary noise, the fault process is detected by statistical analysis of the measurement firstly. According to the model assumption, the parameters of the fault trend process can be obtained by using STF ( strong tracing filter). After the extraction of trend component, the modeling error series becomes a stationary series, which can be used for normal ARMA time series analysis. Finally, the whole prediction can be acquired by combining trend prediction and time series analysis. Computer simulations validate the effectiveness of the proposed method.

Abstract:The problem of optimal reliable guaranteed cost control with regional poles constraints is investigated for a class of uncertain discrete - time systems subject to sensor failures. The sufficient conditions are derived for the faulty closed - loop systems meeting such performance requirements by using matrix inequalities. The existing result on the upper bound of guaranteed cost function is improved. Based on linear matrix inequalities (LMI) approach, the reliable state - feedback controller is designed to guarantee, for possible sensor failures, the closed - loop system satisfying the pre-specified regional pole index and having the optimal quadratic cost performance. Finally, the simulative example demonstrates the validity of the proposed method.

Abstract:Based on the fatigue test results of SAE 1045 medium carbon steel, the multi-axial fatigue life predication models for different ratios of torsional nominal stress amplitude to the axial one are studied respectively. Then a unified multi-axial fatigue life predication model is founded. It is found that the different ratios induce different models. In order to get the relationship between the axial model and the unified multi-axial model, it is supposed that the fatigue strength exponent and the fatigue ductility exponent are constants, while the fatigue strength coefficient times 1.5435 and the fatigue ductility coefficient times 0.6713, the maximum principle stain amplitude replaces the axial push-pull strain amplitude, then the revised axial fatigue life predication model is equivalent to the unified multi-axial model. Therefore, the revised axial fatigue life predication model can be used to predict the multi-axial fatigue life for any ratio of torsional nominal stress amplitude to the axial one.

Abstract:For the active fault-tolerant control method of non-linear system with multi-failures, the identification algorithm of the improved CMAC based on balanced learning is used to estimate faults function in this paper. The fault - tolerant control is realized by utilizing the effective control law reconfiguration strategy based on sliding model control algorithm. Finally, a simulation is done for the examples of double - fault cases. The simulation results demonstrate the effectiveness of the proposed tolerant - fault control method.

Abstract:The design problem of guaranteed cost reliable controller for a class of uncertain discrete fuzzy systems with actuator failures is considered. Based on a more general continuous model of actuator failure, the sufficient condition for the existence of state - feedback guaranteed reliable controller is derived from using fuzzy Lyapunov function and linear matrix inequality (LMI) technique. Furthermore, the optimal reliable controller design problem is formulated as a quasi - convex optimization problem.

Abstract:The robust H∞ fault estimation problem for a class of LPV plants which depend affinely on a vector of time-varying parameters has been studied in this paper. Based on the gain scheduling techniques, a new method for the design of a robust H∞ fault estimator is developed. The resulting LPV estimator has the same parameter dependence as the plants, and can be characterized via a set of linear matrix inequalities (LMIs). To demonstrate the proposed method, a numerical example is considered. Simulation results show that the designed LPV fault estimator is effective for a class of LPV plants.

Abstract:For the networked control system (NCS), the uncertainty of network - induced delays is converted to the uncertainty of the parameter matrix, and the states are transformed to the augmented states to eliminate the effect of the asynchronism. Based on Lyapunove stability theory and Linear Matrix Inequality (LMI), the sufficient conditions for closed - loop of networked control system possessing asymptotically stable against sensor or actuator failure are given, and the corresponding design method of robust fault-tolerant controller is presented via solving several linear matrix inequalities. Finally, numerical examples are given to demonstrate the effectiveness and feasibility of the proposed approach.

Abstract:The phase center of each element antenna is not necessarily to be determined accurately for beam forming of planar phased-array antenna, but it is necessary for that of non-planar phased-array antenna, or phase is not set correctly and antenna performance is decreasing. In this paper, a new method of computing the element antenna phase center is proposed, and the element antenna phase center of a certain alternating - direction communication antenna is computed, and the effect of element antenna phase center on beam forming of non-planar phased-array antenna is studied. The simulation results demonstrate the effectiveness of the new method.

Abstract:In networked control systems the information is exchanged using a network among control system components. The performance of the closed - loop system is not only determined by the characteristics of the control system, but also determined by the scheduling manner imposed by network. The designer of fault diagnosis should take the new peculiarities into account. In this paper, networked control systems are modeled based on information scheduling scheme and the fault detection problems of them are studied. The residual signal is generated using an unknown input observer. Sufficient conditions for the existence of unknown input observer are derived. Several stability results regarding the state estimation error are established. Simulation results are given to show the effectiveness of the design.

Abstract:In this paper, a broadband micro-strip patch antenna is designed and manufactured, which is comprised of many substrates with different thickness and dielectric constants. This antenna is fed, through the coupling slot, by a semi-annular micro-strip line. The simulated and tested results show that the 47% impedance band (VSWR<2) and 8.4dB gain are achieved. This antenna can be widely used in the field of broadband plane array antennas.

Abstract:Aiming at the problem of target identification in multi-sensor data fusion, a method of target identification based on the rough set is proposed. Through this method the rough set theory is used to analyze data, the classing rule is derived from the matrix and discernible matrix and discernible function of the known decision making system and thus creating the rule sets. The effectiveness is illustrated by an example.

Abstract:PID control based on BP neural networks, proceeding from the basic principle, learning rules and algorithms of BP neural networks, is introduced in this paper. Better control qualities are obtained by using the dynamic BP algorithms of three-layer networks. Simulation results indicate that the PID control based on BP neural networks displays a strong robustness in dealing with the nonlinear and uncertain system.

Abstract:Generalization performance is one of the most important problems of intelligent approaches for parameters forecasting. This paper presents an ensemble of support vector regression (ESVR) which has better generalization performance than other intelligent approaches. To increase the diversity among individuals of ensemble, the paper proposes an individual generating approach based on clustering technique. Firstly, ESVR is used to classify training samples as several subclasses that are used to train different individuals with different kernel functions. The ensemble weights of individuals are determined by the generalization errors on the validation sets. ESVR is tested on the parameter drift data of gyroscope. By comparing single SVR, single neural network, neural network ensemble and combination approach with ESVR in generalization performance, the results reveal that ESVR has better generalization performance than other intelligent approaches in most cases.

Abstract:In conventional adaptive array processing, the array processing performance will be degraded seriously because of the steering error, which must be calibrated. Based on the analysis of the effect of steering errors on adaptive beam-forming performance, a new method of calibrating the errors is presented, in which the array data and the uncertainty of the steering errors are utilized for iteratively computing the factor used in modifying the array steering vector. Adaptive beam-forming is performed with the calibrated steering vector. The theoretical analysis and simulation results show that the new method is better in robustness to the steering errors.

Abstract:This paper presents a novel technique to reduce the number of operations in Multiplierless implementations of linear DSP transforms based on shifting and adding , CSD ,and sub-expressions. The complexity of multiplier blocks can be significantly reduced by using an efficient number system. First it gives the 10 bits CSD representation and definition of Sub-expressions. Then Statistical Characterization of CSD Code is studied , it found that the five-term Sub-expressions elimination . Through the design and implementation of FIR, our method will use the less adders than ordinary schemes.

Abstract:A new adaptive bit and power allocation algorithm in single user OFDM system is proposed. At the request of object BER, the bit allocation algorithm is used in two steps to allocate appropriate bits to each subcarrier, finally the sum power is minimized. The simulation results shows that the performance of the proposed algorithm is approaching to that of the optimization method, and the complexity of the algorithm is greatly reduced.

Abstract:Two different design schemes about firewall IPS module are brought up. In the first scheme snot-inline technique and the QUEUE action of netfilter are used to achieve dropping of attack data packet. In the second one, IPS about Denial of Service attack is achieved with the benefit of combination of synccokies, the new netfilter fuzzy match, PSD match, U32 match with an improvement on the firewall kernel. After comprehensive comparison, the module is developed according to the second scheme. The experiment shows that the designed module works well in defending main DOS attack.

Abstract:The parameters affecting networking reliability of ground-to-air data link are analyzed. Based on the probability distributions of failure conditions of such parameters, the reliability analysis method of each node is induced. Finally the three-node data link model is taken for example, its reliability is analyzed in several states and the reliability index is deduced.

Abstract:The relation between the length and the minimum distance of four dimensional binary self-orthogonal codes are discussed. The nonexistence of self-orthogonal codes with parameters \[ 15m + 5, 4, 8m + 2 \] and that with parameters \[ 15m + 12, 4, 8m + 6 \] are proved, thus for each n ≥ 8, the minimum distance dso of optimal \[n, 4 \] self - orthogonal is determined. At last, for each n≥ 8 a generator matrix that generates a \[n, 4, dso (n) \] self-orthogonal code is presented, and their weight polynomials are determined.

Abstract:The problem of satisfactory fault - tolerant control system design for linear discrete uncertain systems against actuator failures with sector pole index, steady variance index and H - infinity constraints is addressed. A more practical and general model of actuator failures is adopted. Based on linear matrix inequality approach, the consistency theory on sector pole index, steady variance index and H - infinity constraints is set up, and the ranges of consistent indices are analyzed in details. Furthermore, the effective controller design method for systems with constraints of consistent indices is proved. Numerical examples are given to illustrate the validity of the design method.

Abstract:Based on the analysis of the cooperative engagement of SAM troops and air arm, the selection and optimization method of air -ground cooperative engagement is put forward. Firstly, the overall combat efficiency under different cooperative engagements is calculated and the best pattern is selected. Secondly, the adjusting plan is given according to the strength ratio between enemy and ourselves, and simultaneously the relevant mathematic models are given. This method is proved to be effective by a certain example. The method can provide reference for the decision - making of air - ground cooperative engagement.

Abstract:To deal with the lack of detailed depiction of neutral gas in PIC simulation, the paper presents a two-dimensional model of multi-element and multi-phase neutral gas to depict explicitly the fields of neutral gas in plasma region. Numerical methods of solving the proposed equations as well as various boundary conditions are put forward. The above model and method are applied to the simulation of adiabatic gas diffusion. The results obtained well accord with the theory. The model can be used in improving PIC simulation programs.

Abstract:Considering the characteristic of logistics supply, this paper proposes a new evaluation method based on the idea of multi-layer state space, and establishes a feedback network mathematical model. In this way, the problems of integrate modeling and system optimizing can be solved effectively. Practice proves that the method is reasonable and effective.

Abstract:Based on the fundamental notions of intuitionistic fuzzy sets, and in consideration of the effects of the three factors-the membership, the non-membership and the probability distribution, two extensions of intuitionistic fuzzy sets, i.e. statistical intuitionistic fuzzy sets and multi-value intuitionistic fuzzy sets, are presented. The calculations and relations of extending the intuitionistic fuzzy sets are introduced. Thus, the intuitionistic fuzzy sets are further extended and applied.