This paper studies the lightweight method of the front rail of domestic mid-size cars. The average crushing force of the thin-walled beam composed of multiple right angles is derived based on the theory of the crushing of rectangular thin-walled beam. Meanwhile, the substructure dissipated energy and force are obtained by target cascading of the acceleration pulse in 65 km/h frontal crash. Based on the force and the lightweight target of substructures, the new cross-section of the rails is designed with the theory of crushing of multiple right angles thin-walled beam and bending of rectangular thin-walled beam. Then the wall thickness of the new rails is modified by finite element analysis. The weight of the new front rails designed by lightweight method is reduced by 20% and the frontal crashworthiness is remained reserved.

A co-simulation model of whole-vehicle dynamics was established based on MATLAB/ Simulink and AMESim. Also a precise dynamics model of torque converter, locking clutch and torsional coupling power transmission was developed. The Proportional-Integral-Derivative (PID) controller of the lock-up clutch was designed using torque converter pump wheel and turbine speed as slip control parameters. The MATLAB Optimization Toolbox was used to optimize the slip speed based on the evaluation of the level of torque fluctuations and transmission efficiency. Simulation results show that, when the vehicle is started with small throttle, the lock-up and slip control should not be used due to unstable condition of the engine;when the vehicle is started with medium throttle, large throttle and full throttle opening, the engine works in its best working condition because of the involvement of the lock-up clutch that actively intervenes the engine operating point. The slip and lock-up control of the lock-up clutch achieves a better allocation of power transmission and balances the relationship between the energy loss and NVH.

The internal flow field and the far-field aerodynamic noise of a regenerative are calculated using a two-dimensional Computational Fluid Dynamic (CFD) model, and the simulation results are analyzed. It is shown that, in the far-field noise spectrum of the blower, the frequency, which is 42 times of the blower rotation frequency, is related to the number of blades of the inlet/outlet and the radial clearance between the impeller and the partition of the casing. The far-field noise peak frequencies of the blower are different when the clearance is larger or smaller than 0.5 mm. The effects of impeller parameters, such as the blade angle, blade number, and the clearance between the impeller and the partition of the casing, on the far-field noise of the blower are analyzed using the range analysis. The influence order of these parameters is determined, and the radial clearance has maximum influence on blower noise.

The influence of abnormal states on road traffic is analyzed. Then an estimation method of travel time feasible to different abnormal states is proposed. Some key technologies in this method are studied, including determination of sampling size based on GPS data, estimation of travel time with sufficient or insufficient samples based on GPS data, and BP neural network based travel time fusion method. Finally, the proposed method is verified by simulations under two abnormal states, one is serious traffic accident and the other one is dense fog. Simulation results show the proposed method can more accurately estimate the travel time under different abnormal states.

A cost function with combined trip under park and ride was put forward. Then Stochastic User Equilibrium (SUE) model was proposed under the condition of multi-modal traffic super-network based on the Logit model. The proposed model was solved with the Method of Successive Weight Average (MSWA). A case study was conducted to verify the feasibility of the model and the algorithm. The results show that the MSWA is superior to the Method of Successive Average (MSA) when solving Logit-SUE model. The model and the algorithm can be used to determine the reasonable parking position and riding site. They can also be used to estimate the reasonable range of parking fee in car park and the rental cost when transferring to bicycles.

According to historical statistics of the traffic volume, considering the differences of traffic flow directions, an intersection traffic control period division method is proposed based on the processing of average hourly traffic volume. In this control method, by calculation of the cycle length of traffic signal at intersection, a control period threshold value is set. Using this method the 24 h of a day is divided into several traffic control periods at the intersection of Liangqing road and Lixi road, Wuxi City. This provides a scientific basis for the establishment of signal timing programs with multiple periods a day. Compared with traditional period division method, the calculation results by the proposed method are in good agreement with the actual traffic flow.

Turning ratios at intersections were used as the prior information to derive the linear equations of the traffic flow conservation system. Based on that, the network sensors location problem (NSLP) was discussed. A model trying to infer the flow vector over the road network was proposed. The model is proven to have no unique solutions. Considering the roads importance on the traffic management and sensors cost on the installation and maintenance, a weighted NSLP model was proposed. The weighted NSLP model uses the weight parameters to represent the road importance and sensors cost. The optimal object of the weighted model is to maximize the total weights over the network. Using the theory of matroid, a greedy algorithm was designed to solve the model. A proof for the existence of feasible solutions was given finally.

To predict the real-time collision risk for rear-end crashes at freeway recurrent bottlenecks, in this study, the features of trajectories of individual vehicles when encountering kinematic waves were investigated, the collision conditions were analyzed, aggregate analysis of the trajectories of multiple vehicles was conducted. Then a model to predict the collision risks was developed using aggregated traffic flow data. Results show that the rear-end collision probability during propagation of kinematic waves was significantly related to the collision risk index and standard deviation of upstream occupancy. The model was used to predict the collisions in the bottleneck section of a real-world freeway;the predictions were consistent with observations. The highest collision risk occurs when traffic changes from free-flow to congested states. The propagation of kinematic waves in congested traffic also increases the rear-end collision risk.

A trammel pendulum model to simulate liquid sloshing in tanks with circular and oval cross-sections is developed. Motion equations of the trammel pendulum in inertial coordinate system, non-inertial coordinate system, and in a coordinate system that moves in a complex way are derived. In the inertial coordinate system, the motion equation of trammel pendulum is numerically solved. The angular position, angular velocity and hunting frequency of the pendulum are analyzed. It is found that a linear trammel pendulum model can be used to simulate liquid sloshing in partially-filled tank vehicles. This research can provide theoretical basis and actual assistance to investigate the performance, driving stability and auxiliary driving technology of tank vehicles.

The variable temperature penetration tests were conducted to derive the compaction temperature from hot-mix asphalt to war-mix asphalt using penetration index. It is based on the principle of analog of hot mixture equivalent shear strength. The variable temperature penetration testing method and process were investigated, and the reasonable compaction temperature ranges of four types of warm-mix asphalt were derived. Compared with variable-temperature control compaction test, the penetration test can fully reflect the impact of the change of asphalt viscosity on the workability of the mixture. The penetration test is also sensitive to temperature with the entire operating temperature range. The operation of the penetration test is easy and it takes shorter time.

Under considerations of the capacity reduction factor and fire load factor, the limit state equations of the component reliability analysis were established. The reliability of the reinforced concrete member after fire was analyzed. The mechanical properties, e.g. yield strength, and sectional properties, e.g. cross-sectional area, etc, of reinforced bar are considered to be random variables. A number of parameters that affect the structural reliability, including the fire load, ratio of floor area to the total area of the fire compartment, dead load, and live load are also taken as random variables. The change of the load effect of the limit state equations after fire was considered. Then the reliability of reinforced concrete member was analyzed. Results show that fire load effect has high impact on the reliability of the structure after fire. So the fire load can not be ignored under some conditions.

Numerical and experimental studies were carried out on a 2L HPCR diesel engine to explore the effects of fuel injection pressure on the processes of fuel spray, air-fuel mixing and the formation of harmful emission. Results show that, as the injection pressure increases, the spray penetration increases, the Sauter Mean Diameter (SMD) of fuel drips significantly decreases, and the atomization quality is improved. Furthermore, with the increase in injection pressure, the diffusion capability of fuel is improved, the distribution volume increases and the oxidation of soot is enhanced. On the whole, with the increase in fuel injection pressures, the soot emission is reduced and the NO emission is deteriorated. However, the increase in injection pressure can result in higher degree of fuel richness in early period of combustion and further deterioration of air-fuel mixing. When the injection pressure reaches a certain level, it can weak the sensibility of its influence on soot emission.

In order to quantitatively evaluate the rationality of matching scheme between engine and torque converter, a series of dimensionless matching evaluation indexes were developed according to the ideal matching principle, and a standardized scoring method for every index was put forward. Considering the differences of importance of the evaluation indexes in different applications, relative weights of each index was determined using analytic hierarchy process. A matching evaluation system consisting of matching evaluation vector and matching evaluation function was established based on an improved radar chart method. Taking the powered matching demand as an example, three groups of hydrodynamic torque converters for a total of sixteen matching schemes were quantitatively evaluated. Evaluation results show that, using the proposed method, the optimal matching scheme can be determined accurately.

Due to the limitations of existing three-axis force sensors, a three-dimensional (3D) force measurement platform is designed and the corresponding calibration device is developed. This platform is based on the parallel technique of four 3D force sensors, and is used for parameter measurements of high-speed bogie. To solve the dimension coupling interference of the 3D force measurement platform, the output characteristics of three orientations under uniaxial loading are analyzed, and the static calibration matrix is obtained. Testing results show that the maximal relative error of the designed platform is 0.42 and the measurement accuracy is ±0.25% under eccentric load. Under simultaneous tri-axial loads, with the increases in X-and Y-axial loads, the measurement accuracy declines to ±1.5%. This 3D force measurement platform has good performance of anti-deviated load, and the dimensional coupling interference is solved in a better way.

The crucial issues in Failure Mode and Effect Analysis (FMEA) for machine tools are the great uncertainty or unreasonable weights of severity, occurrence and detection effect factors which are determined by the experience of experts. To overcome such disadvantages, a modified FMEA was presented based on Fuzzy Data Envelopment Analysis (Fuzzy-DEA). A method to determine the weights of the effect factors was proposed based on the advantages of DEA that can assess the effectiveness of each unit. According to the relationship between the weights and the effect factors, a model of the risk priorities of failure modes was developed when a certain level of "h" was given. The rank of failure mode criticalities was determined according to the risk priorities. A case study was provided to demonstrate the use of the proposed FMEA in the design of CNC lathe and show its potential application and benefits. This study verified the feasibility of the proposed method.

Flexible rolling for surface of sheet metal is a novel method, which combines the traditional rolling technology with the multi-point forming. The curvature of the rigid rolls is adjusted by several control points in the rolling process to meet the requirements of forming forces of different surface work pieces. The three-dimensional (3D) plastic deformation of the surface sheet and the sheet feeding can be achieved by the rolling reductions and the rolls rotation. This article discusses the basic principles, forms and forming characteristics of the surface flexible rolling technology. The formula of work rolls bending radius of the circle is derived. The correctness of the method is verified by simulation of explicit dynamic finite element analysis. The effect of the maximum rolling reduction on rolling horizontal and vertical curvature is given. A surface flexible rolling device is developed for the first time and several 3D surfaces, such as spherical pieces and saddle-shaped pieces, are obtained using this device. The work pieces have good surface rolling effect and the results of the test and simulation prove the feasibility and practicality of the surface flexible rolling method.

Multi-point Thermoforming (MPTF) is a new flexible technique for forming three-dimensional polymer sheet parts. The parameters of polycarbonate according to G'Sell's constitutive equation were determined and a Finite Element Model(FEM) of MPTF process of polymer sheet was established. A series of numerical simulations were carried out to analyze the influence factors on dimpling in MPTF. The results show that the dimples on the formed part diminish with the decrease of punch element size. The large punch element radius can improve the surface quality and the shape accuracy of the formed part. Experiments were performed on a MPTF experimental setup and the results show that proper selection of the dimension of multi-point die can significantly improve the forming quality.

During the process of splitting connecting rod, the area defect of fracture surface often occurs due to particle shedding. The area defect dimension should be limited to ensure the quality of products. In this paper, the data information of connecting rod fracture surface was obtained and the three dimensional fracture surface was reconstructed based on reverse engineering idea. Then the morphology of the fracture surface was displayed using TIN method. Meanwhile, the defect area of fracture surface of connecting rod was quantitatively determined using triangular patch stitching and summation principles. Simulation results show that the area of con-rod joint surface obtained by splitting is 13.1% larger than that obtained by machining. Compared to empirical value, the particle shedding limit sizes of fracture surface has been substantially improved without reducing the bearing capacity and the use performance. This is quite significant to con-rod fracture splitting production.

The stretch bend forming process of stainless steel roof beam of metro vehicle is studied by simulation. The causes and controlling methods of the forming defects such as wrinkles, cross-section distortions and low precision of profile are investigated. Results show that the effects of pre- and post-stretching elongation on wrinkling defect are not evident, while, increasing the wrap stretching elongation can reduce wrinkling defects effectively. Compensation to the cross-section shrinkage of the profile can eliminate the cross-section distortion of the eversion of small edge. Adopting reasonable process parameters of stretch bending and precisely compensated bending mold can reduce springback obviously;therefore, the profile precision is improved and the stretch bending parts with high quality are obtained.

Based on static weight loss method and metallographic method, this paper investigated the effect of soaking time on the oxidation behavior of typical cold rolled hot stamping steel 22MnB5. With the results of the thickness of oxidation and decarburization scales, the regression equations of weight loss and reduction of oxide scales were built. The characteristic parameter named equivalent interfacial-heat-transfer-coefficient (IHTC) was got by inverse numerical calculation with ABAQUS, which was taken to describe the effects of thickness of oxidation and decarburization scales and loads based on related heat transfer dynamics performance in process of hot stamping. The results indicate that the weight loss per unit area and thickness reduction of B1500HS steel with 1.6mm thickness can reach 0.471 mg/mm² and 99.3 μm respectively, the depth of decarburization can reach 87.0 μm at 900 ℃. The oxidation of the mentioned steels obeys the parabolic law, which presents high oxidation rate at initial stage of soaking, then slows down obviously after 4~5 min. The multiple correlation coefficients between regression analysis and weight loss and reduction of oxide scales are over 0.95. The equivalent IHTC has negative relationship with the thickness of oxidation scale and it also increases with the interface load. The results show that oxidation behavior has a significant influence on heat transfer dynamics performance in hot stamping process.

The nanoindentation of single-layer graphene sheets is simulated using molecular dynamics simulation. The nano-mechanical properties and failure mechanisms of single-layer graphene sheets are investigated at atomic scale. A typical load-displacement curve is obtained and the influences of the radii of indenter and graphene, temperature and different boundary conditions on the simulation results are analyzed. Observation of the deformation process shows that, when the indenter displacement reaches a critical depth, h_c, the single-layer graphene sheet undergoes plastic ripping damage due to the atomic bond breaking beneath the indenter. Further analysis of the stress distribution on the graphene around the destruction area reveals that, when a gap is generated on the graphene, the maximum stress decreases rapidly and a homogenization distribution occurs. In addition, repeated loading-unloading processes are performed on the graphene sheet. When the penetration depth is less than the critical value, h_c, the sheet undergoes an entire elastic deformation. However, when penetration depth is larger than h_c, the deformed hexagon beneath the indenter can not be fully restored to the original state. Easier deformation of graphene may initiate from these bonds, which in turn, results in significant decrease in strength and deformation of the graphene. In addition, the elastic modulus and deformation mechanism of the graphene sheet are strongly dependent on the temperature. The rise of temperature leads to the decrease in the elastic modulus and the failure limit to some extent. Different radii of the single-layer graphene sheets affect the critical penetration depth on nanoindentation. This is due to the homogenization distribution of deformation in a wide range on graphene sheet. But it has a little effect on the elastic modulus and failure stress.

The hardness of rock, strong grind and break of drill bit, and hard/soft interlayer formation are all factors to slow down the drilling speed, shorten service life of the drill bit, resulting in long drilling cycle. Soil animals may provide inspirations to solve such problem as they can dig hole on the ground effectively. The coupling bionic theory, the non-pressure immersion method, hot pressing and welding methods are used in the design and manufacturing of ternary coupling bionic impregnated diamond oil bit (called bionic oil bits). Three parameters, including bionic non-smooth structure, material and the whole bit structure, were considered in the design and manufacturing. Drilling text of the bionic oil bit was conducted in Puguang D-103 Wells. Test results show that the penetration rate and service life of the bionic oil bit are higher than tricone bit, and its performance is stable.

Based on the input observer, an estimator of individual cylinder Air-Fuel Ratio (AFR) in a SI engine is proposed. The fuel path model, which is constituted by fuel wall-wetting model, exhaust gas mix model, exhaust gas transfer model and universal exhaust gas oxygen (UEGO) sensor model, is established. Then, under the same working condition and with the same input, the developed model is validated using the precise engine model of enDYNA. Furthermore, an input observer based on regional pole placement is proposed to design an AFR estimator. Finally, the rationality and feasibility of the proposed observer and AFR estimator are validated by real-time and offline simulations on a simulation platform for automotive engine control.

A kind of rigid-flexible hybrid parallel close looped linkage mechanism was developed for obstacle-crossing robot. The plane model of the rigid-flexible hybrid system was set up. The dynamic model was established using Lagrange dynamical equations and modal equations. Simulations of the proposed model were conducted and comparison was made with the rigid system. Results show that the trend of energy variation of the new system is similar to that of the rigid system;so, it can be used under the same application condition. The curve of kinetic energy of the rigid-flexible system is gentler than the rigid system with the action of external force and possesses more significant buffering characteristics. Therefore the new system is more suitable for obstacle-crossing under severe impacting situation.

To meet the requirement of Electromagnetic Interference (EMI) peak suppression of inverter, inverse m sequence Random Phase Pulse Width Modulation (RPPWM) is proposed. The generation of pseudo-random inverse m sequence and RPPWM principle are presented. The spectral distribution expression of inverse m sequence and the main waveforms of RPPWM are deduced. By adopting the equivalent spectrum analysis mode, which avoids direct analysis of the nonlinear comparator and reduces the difficulty of analysis, the analytical expression of RPPWM waveform spectrum is obtained. By comparison with waveform spectrum expression of square wave RPPWM, it is proved that inverse m sequence RPPWM has superior Electromagnetic Compatibility (EMC) characteristic from theory aspect. Experimental results indicate that inverse m sequence RPPWM has better EMI peak suppression effect in the vicinity of the odd switching frequencies. The combination of inverse m sequence RPPPWM and other Random Frequency Pulse Width Modulation (RFPWM) has an excellent application prospect.

Vector control schemes for induction motor drive systems have been widely adopted in high performance applications. Vector control transforms the control of an induction motor to that of separately excited dc motor by creating independent channels for flux and torque control. Crucial to the success of a vector control scheme is the knowledge of the instantaneous position of the rotor flux. The position of the rotor flux is estimated in the direct vector control scheme, which requires a priori knowledge of the machine parameters. Identification of machine parameters is a basic step in designing a motor controller. In traditional method, voltage required for off-line identification of induction motor parameters is commonly obtained by filter circuit and data acquisition equipment, and the equivalent circuit of the induction motor is usually modified. The reason that traditional off-line identification result of induction motor parameters is not suitable to online model is analyzed. The internal resistance of the power devices is equivalised to the stator resistance of the motor first. Then the voltage distortion caused by dead-time is eliminated through offline voltage correction. Finally, a genetic algorithm is used to identify the equivalent electrical parameters of induction motor. Experiment results show that the parameters obtained by this method can accurately describe the steady-state and transient response of the induction motor under specific inverter power supplying conditions.

The inductive power transfer(IPT)technology was applied in the field of household appliances,then a wireless power for household appliances can be achived. Based on the stroboscopic mapping method and fixed point theory, the operating frequency model and output power model were proposed first. Then, the influence law of the resonant network parameters on the frequency stability was analyzed. Finally a design flow for resonant network parameters to make the operating frequency naturally reach to a relatively stable status was presented. Simulation and experimental results verify that IPT systems developed with the proposed design flow have good frequency stability as the adding, removing or power disturbance of household appliances in the pick-up side.

Yangjiagou Quarry slope situated in Changchun was selected as subject investigated. The layout and selection rules of control points, polar coordinate measurement with reflectorless Total Station Instrument(TSI) and the position precision assessment based on error theory were studied in control surveying. Also, the 3 D coordinates and precision indices of slope control points were obtained. Furthermore, the control point results were applied to 3 D slope modeling and position precision of the models was assessed. For orientation validation measurement, a set of points not on the same line of slope discontinuities were measured with reflectorless TSI. Using spatial geometry and azimuth concepts, an orientation calculation model based on spatial coordinates was established and discontinuity orientations were obtained. Orientation measurements of four-point method with reflectorless TSI were compared with that of three-point method with reflectorless TSI and compass measurements, the results shows that the four-point method is effective and reliable. Thus, orientations obtained with four-point method were served as the standard to compare with ones interpreted from 3 D slope model with digital close range photogrammetry(DCRP) for the steep slope and the orientation measurement precision of discontinuities based on DCRP were evaluated. This study demonstrates the feasibility of rock mass slope control surveying with reflectorless TSI, the accuracy of 3 D slope modeling and the reliability of orientation information acquisition of discontinuities with DCRP.

The separate adaptive branching heuristics can not give full play to the advantages of adaptive constraint solving's expertise.To overcome this shortcoming,a new algorithm,AdaptBranch^surv,is proposed, which combines the Survivors-first learning value ordering heuristics with adaptive branching scheme.The experimental results of some typical Benchmark problems show that the proposed algorithm AdaptBranch^surv can significantly improve the efficiency of constraint solving.

In order to overcome the shortcomings of conventional gesture recognizers, improve the accuracy of recognition and take human-computer interaction into consideration, a new template based recognition algorithm called Grid was proposed. Using this algorithm, the user input gesture given by point set is recognized by rotation, translation, scaling normalization and resampling. In particular, this algorithm uses a gird based method to resample the most representative points of the gesture. Experiment results show that the proposed Grid has higher accuracy with a tolerable larger time cost. This algorithm provides a reference for touch-screen based gesture interaction technique.

A dimension reduction method is proposed for data streams based on Johnson-Lindenstrauss transform to meet the requirement of real time performance, which can also be used in asynchronous cases. This method uses a Johnson-Lindenstrauss transform to project the stream data from high dimensional space to low dimensional space, and the pairwise distance is preserved after the projection. The transform matrix is constructed by random hash functions, and the random hash functions are generated by the seeds, therefore the dimension reduction process is completed at the same time of the updates. This method reduces the computational complexity, and an algorithmic sublinear-time dimension reduction of data stream is gained. The experimental results show that this method can improve the efficiency and get a good accuracy.

In this article, the content style of Service Level Agreement (SLA) is expanded and an improved performance management mechanism of SLA is proposed, which are based on SLA content monitoring and analysis of user evaluation. Accordingly, a new pricing management framework is established and a resource dynamic pricing strategy is proposed based on utility. Experiment results show that the proposed pricing strategy maximally meets the demand for the SLA negotiation, which saves users' cost and maximizes users' satisfaction.

A novel vehicle active scheduling model at intersection is proposed based on Vehicle-to-Vehicle (V2V) and Vehicle-to-Infrastructure (V2I) communication and autonomous system, including vehicle motion, traffic light control and other aspects. Vehicles in the same lane and direction form a team via V2V communication. The headmost vehicle communicates proactively with traffic light when approaching to intersection, and then transmits the required information for scheduling. After summarizing all information of vehicles from different directions, the traffic light makes a scheduling decision and distributes a time-slice to vehicle applicants. Simulations show that the proposed model is better than the traditional model in reducing fuel consumption and waiting time when the traffic flow is not overloaded.

A human abnormal behavior detecting approach was proposed based on optical flow features in the motion area. An improved dual background model, which includes an adaptive running average background model and a HSV background model, was developed to indicate the variation of background pixels in order to increase the robustness against illustration change and environmental disturbance. The model can reliably extract the motion area. Foreground was obtained from video sequences by background subtraction. The motion area was labeled as several regions of interest, and the optical features in each labeled region were obtained using Lucas-Kanade algorithm. Amplitude based weighted unit energy derived from the optical flow features was defined to measure the anomaly of human activity. Experiments were conducted on various videos indoor and outdoor, and the results were presented to verify the effectiveness of the proposed scheme.

In order to solve the problems of multi-source fusion and quantitative awareness in the network security situation awareness, the particle swarm optimization is applied to search the exponent weight for different reliability data source for the D-S evidence fusion, and the multi-source fusion was obtained. Based on the fusion, the threaten gene acquisition method with adaptability was studied through the discretization to normal distribution, and the situation factor extraction was accomplished. According to the fusion and factor extraction, a formal network security situation awareness method was proposed. This method could generate service, host and network system level situation evolution curves. Simulation experiments show that the proposed fusion method can increase the detection rate and decrease the false detection rate. The formal quantitative method is able to be aware of the dynamic change of the threat and meets the goal of monitoring the network in an effective manner.

Current visualization techniques failed to effectively display the high dimension multi-objective optimization problems. To overcome this disadvantage, a new sub-diagram visualization technology based on single objective fitting is proposed. The new visualization technology displays the Pareto solution set in sub-diagram form whose number is the same as objectives. Additionally, the Pareto approximate set is drawn by the fitting location in the sub-diagram. The proposed method displays effectively the convergence and distribution of the Pareto approximate set;meanwhile, the relative merits of the performance on a single solution in each dimension objective and the comparison of the performance of different solutions in the same objective are displayed effectively. Numerical experiments show that the new visualization technique plays a key role in helping decision-makers carry on analysis and decision for multi-objective optimization problems.

An improved trust model based on recommendation evidence for P2P network is proposed. The aim of the model is to resolve the problem of identifying dishonesty recommendation peers in traditional trust model for unstructured P2P network. The new model updates the trust records of the transaction node and recommendation nodes simultaneously according to the transaction results. It adopts the decay function, which meets interpersonal characteristics, to describe the decay process of the trust records. The D-S evidence theory is introduced to the computation of the trust model, which integrates direct evidence and recommendation evidence into trust evidence. The model searches trust evidence through probability based Gossip algorithm. The experiment results show that the proposed model can increase the successful transaction rate and improve the network environment.

The existing algorithms for Prolate Spherical Wave Functions (PSWFs) demand large amount of computation and high complexity in hardware implementation. To overcome the above weaknesses, a new algorithm with high accuracy and low complexity in both time and space domains was proposed in terms of computing the state equation of differential system. Then a simulation model of the system is constructed to generate PSWFs. This simulation model is easy for hardware implementation. The proposed algorithm was compared with the one proposed by Parr B. Results show that, when both algorithms obtain identical waveform and spectrum, the one proposed in this paper has much lower complexity. Simulation results of the system model show that the output waveforms are the same as the theoretical solutions, which verify the correctness and feasibility of the constructed system model.

In order to solve the problem of imaging system instability caused by high work frequency of the image composing circuit on the satellite, the camera structure was innovated. A composing and compression method of time delay integrate charge coupled device (TDICCD) based on field programmable gate array (FPGA) platform for satellite remote sensing data was proposed. First, the signal flow was improved. Then, based on TDICCD linear array push-broom imaging mode characteristics, an improved predictive code method based on least mean square (LMS) was used for composing and compression. Finally, experiments were conducted on a platform to validate the proposed method. Experimental results indicate that the system clock is reduced to 80 MHz and 60 MHz, respectively, from the original 110MHz in the lossless and 2 times compression. The peak signal-to-noise ratio (PSNR) of the reconstructed image is bigger than 84 dB;the average processing time of one line image is 57.3 μs, which is smaller than camera's smallest line period. The proposed algorithm is stable and it can achieve the composing and compression of multi-channel TDICCD remote image within the camera.

In order to detect signals in low frequency channels with non-Gaussian noise, a novel adaptive receiver is developed. A Markov Chain Monte Carlo (MCMC) algorithm is proposed to estimate the fading coefficients and the noise model parameters simultaneously using the training sequence;then the signal is estimated by the optimal decision rule. A commonly used model, which mixes the symmetric α distribution and the Gaussian distribution, is used to model the noise. Simulation results show that the performance of proposed adaptive receiver can approach that of the optimal receiver with known channel model parameters.

The uniform wideband digital channelized receiver can not dynamically adapt to the changes of the instantaneous signal. To resolve this problem, a structural design of dynamically channelized receiver is proposed based on Cosine-Modulated filter banks. The synthetic processing matrix is used to complete near perfect reconstruction of the signal. The theory and design methods of filter prototype and the near perfect reconstruction of cosine modulated filter bands are analyzed. According to the discussion of a channel design method using uniform cosine modulated filter, a non-uniform channelized structure is designed based on the factor that different sub-band signal extraction requires different multiples. The fast algorithm of the fourth discrete cosine transform of the non-uniform channelized structure is discussed. The new signal can be reconstructed by changing the synthetic matrix based on the subchannel energy detection link when the signal dynamically changes. Simulation results verify the effectiveness of the proposed dynamic channelized receiver structure.

Traditional multi-dimensional assignment data association algorithm for multi-passive-sensor system ignores the random errors introduced by least square estimation. To overcome such problem, a new data association algorithm based on the Kullback-Leibler divergence is proposed. The Kullback-Leibler divergence between the probability density function of pseudo measurements and the most posterior probability density function works as the association cost. Meanwhile, an analytical form for kullback-Leibler divergence is calculated to alleviate the computation. The correct data association ratios of the traditional algorithm and the proposed one are compared by simulation experiments. The results show that the Kullback-Leibler divergence reflects the association probability more accurately and the proposed algorithm can achieve better performance.

A high priority backoff algorithm is proposed for congestion avoidance, which introduces the concept of rejected frames. If the network is under extreme congestion, it rejects the sender request to alleviate the network pressure. According to the random variation of data attribute of the sending nodes, the algorithm influences the CW parameters to deal with high priority service. Experiment is carried out in NS-2 simulation platform to compare various priority data flow throughputs and packet retransmission frequency. Results show that Prioritized Binary Exponential Backoff(PBEB) is an effective method to improve the successful access probability of high priority service.

Due to the fact that the existing domestic technologies applied for practically monitoring the cold-chain are fairly limited, the loss of the cold foods during the logistics is much higher than that of international standards. To solve this problem, a new method for monitoring cold chain logistics was proposed and developed based on RFID, GPS and GPRS technologies,which can be used to monitor the condition of fresh foods in real time and publish the condition and quality of the cold foods on-line, including the parameters of the temperature, picture and location information of goods in transit. The warning message will be automatically sent to the drivers and centrally control office when the temperature exceeds a given threshold. The laboratory experiment results have shown that this system can be used to ensure that the goods in transit is always maintained in a safe condition, the loss of the foods during logistics can be effectively reduced.

To investigate the technology of extracting carotenoid from freeze drying yolk by microwave extraction, the effects of four factors, including solid-liquid ratio, microwave power, microwave temperature and microwave extraction time, on the yield of carotenoid were analyzed and compared. The optimal technological conditions were obtained using orthogonally rotational combination design. The results show that the order of factors affecting the yield of carotenoid is as follows: solid-liquid ratio, microwave time, microwave power, microwave temperature;the optimal technological conditions are: solid-liquid ratio of 23 mL/g, microwave power of 600 W, microwave temperature of 43 ℃, microwave time of 18 min;under such technological conditions the yield of carotenoid is 23.99 mg/(100 g).

Fermented Oviductus Rana beverage was developed with Oviductus Rana Enzymatic hydrolysate by Kefir starter culture. Based on the results of single factor tests, the fermentation technology, including the fermentation time, inoculum amount, and sucrose addition, were optimized using Box-Behnken center composite design combined with response surface technology. The results obtained from the study show that the decreasing order of the factors that effect the sensory evaluation under the fermentation temperature of 30℃ is: inoculum amount, sucrose addition, fermentation time. The optimized technology conditions are: fermentation time of 9.76 h, sucrose concentration of 8%, inoculum addition of 4.53%. Under the above condition, the product has good flavor and taste, and there is no significant difference between the experiment results and the model prediction.