Parameterized Scheduling of Topological Patterns in Signal Processing Dataflow Graphs

In recent work, a graphical modeling construct called “topological patterns” has been shown to enable concise representation and direct analysis of repetitive dataflow graph sub-structures in the context of design methods and tools for digital signal processing systems [1]. In this paper, we present a formal design method for specifying topological patterns and deriving parameterized schedules from such patterns based on a novel schedule model called the scalable schedule tree. The approach represents an important class of parameterized schedule structures in a form that is intuitive for representation and efficient for code generation. We demonstrate our methods for topological pattern representation, scalable schedule tree derivation, and associated dataflow graph code generation using a case study for image processing.

Evaluation of the microscale structure and performance of asphalt mixtures under different design methods

In this paper, we propose an efficient design method for a thermal camera with continuous zoom based on the research and manufacturing experience of the thermal camera. In addition, it is divided into system design method, optical design method, mechanical design method, and electronic design method. First, we propose an effective NUC compensation method and a lens-specific sensitivity design method in terms of system. Second, we propose a zoom trajectory design method considering the temperature effect on the optical aspect. Third, it suggests the minimization of optical axis shaking between magnification conversion in terms of mechanism. Finally, we propose a lens-specific temperature compensation method and a speed conversion algorithm according to the zoom interval as an electronic aspect.

Reliability of design approaches for axially loaded offshore piles and its consequences with respect to the North Sea

In this article, the fast algorithms for the discrete Tchebichef transform (DTT) are proposed for input sequences of lengths in the range from 3 to 8. At present, DTT is widely applied in signal processing, image compression, and video coding. The review of the articles related to fast DTT algorithms has shown that such algorithms are mainly developed for input signal lengths 4 and 8. However, several problems exist for which signal and image processing with different apertures is required. To avoid this shortcoming, the structural approach and a sparse matrix factorization are applied in this paper to develop fast real DTT algorithms for short-length input signals. According to the structural approach, the rows and columns of the transform matrix are rearranged, possibly by changing the signs of some rows or columns. Next, the matched submatrix templates are extracted from the matrix structure and decomposed into a matrix product to construct the factorization of an initial matrix. A sparse matrix factorization assumes that the butterfly architecture can be extracted from the transform matrix. Combining the structural approach with a sparse matrix factorization, we obtained the matrix representation with reduced computational complexity. Based on the obtained matrix representation, the fast algorithms were developed for the real DTT via the data flow graphs. The fast algorithms for integer DTT can be easily obtained using the constructed data flow graphs. To confirm the correctness of the designed algorithms, the MATLAB R2023b software was applied. The constructed factorizations of the real DTT matrices reduce the number of multiplication operations by 78% on average compared to the direct matrix-vector product at signal lengths in the range from 3 to 8. The number of additions decreased by 5% on average within the same signal length range.

Automatic vectorization is crucial for improving the performance of computationally intensive programs. Existing compilers use conservative optimization strategies for automatic vectorization, which, in many cases, lead to the loss of vectorization opportunity. Studies have shown that the use of machine learning algorithms to build a performance prediction model is beneficial to improve the program performance. The model input is program features, and the output is the predicted optimization strategies or the program performance related to the optimization. In this paper, we focus on a computational intensive loop structure-tensor contraction, which is common in quantum chemical simulations. Most existing machine learning methods rely on control and data flow graphs as features to represent programs, but different tensor contraction kernels have the same control and data flow graphs. In addition, the existing methods often use the same kind of learning algorithm to construct a learning model, which is prone to overfitting and low-precision problems. In this paper, we propose an automatic vectorization performance enhancement method based on ensemble learning. We construct an ensemble learning model to predict the performance of tensor contraction kernels with different vectorization strategies and select the best vectorization strategy for the kernels. According to the storage access patterns of the tensor contraction kernels, we propose a static method for features representation. Based on the multi-algorithm ensemble learning strategy, we obtain better learning results than the single learning algorithm. The experimental results show that the prediction model achieves 88% and 87% prediction efficiency on two different platforms with different instruction sets, data types, and compilers. Compared with the existing methods, the prediction efficiency is greatly improved. In addition, the average peak performance is $2.96 \times $ of Intel ICC 12.0 and $2.98 \times $ of GCC 4.6 compiler, respectively.

The design of p-cycles in WDM mesh networks has been extensively studied. However, most of studies only deal with a single link failure rather than node failure. In this paper, we develop a new scalable and efficient design method for computing node-protecting p-cycles. The performance of our new proposed design method makes an obvious improvement 20% in capacity redundancy over that of the previous design method. The conventional design methods formulate the problem of pcycle design as an Integer Linear Program (ILP). To solve the ILP, the prerequisite is to a priori enumerated all possible pcycle candidates. For a large network, the resulting ILP maybe intractable as the huge number of cycles may exist. We propose a new design and solution method based on large scale optimization tools, namely Column Generation (CG), where pcycles are generated on-line when needed. The main advantage of our CG-based method is that no p-cycles are a-priori off-line enumerated, the generation of the promising set of p-cycles is embedded in the optimization process. Extensive experiments on several network instances have been conducted for comparison. Experimental results show that our new proposed methods outperforms the previous method in terms of capacity efficiency.

Asphalt treated base (ATB) materials have shown good road performance. There are some differences in the mix design methods used for ATB and Asphalt Concrete (AC) and this paper investigates four different design methods for ATB. Each design method is analyzed by using volume and mechanical indexes, and pavement performance. Based on this analysis, the proper design method of ATB is proposed. 1. ATB Design Method This paper describes the comparison of experimental data with specification criteria recommend in the SUPERPAVE method, the Bailey method, and a mechanical method. The analysis includes determining the effect of different design methods on asphalt pavement performance.

An efficient technique of designing spatial matrix filter for array signal preprocessing based on convex programming was proposed. Five methods were considered for designing the filter. In design method 1, we minimized the passband fidelity subject to the controlled overall stopband attenuation level. In design method 2, the objective function and the constraint in the design method 1 were reversed. In design method 3, the optimal matrix filter which has the general mean square error was considered. In design method 4, the left stopband and the right stopband were constrained with specific attenuation level each, and the minimized passband fidelity was received. In design method 5, the optimization objective function was the sum of the left stopband and the right stopband attenuation levels with the weighting factors 1 andγ, respectively, and the passband fidelity was the constraints. The optimal solution of the optimizations above was derived by the Lagrange multiplier theory. The relations between the optimal solutions were analyzed. The generalized singular value decomposition was introduced to simplify the optimal solution of design methods 1 and 2 and enhanced the efficiency of solving the Lagrange multipliers. By simulations, it could be found that the proposed method was effective for designing the spatial matrix filter.

Design method for relief well trajectory to avoid side-track operations and control the relative distance

Structural behaviour of cold-formed thin-walled short steel channel columns at elevated temperatures. Part 2: Design calculations and numerical analysis

This paper designs and optimizes a new structure hybrid material B‐pillar assembly. Establish a finite element analysis model for the side impact test of an electric vehicle to verify the model's accuracy. Construct a finite element model of the hybrid material B‐pillar assembly. Combined with the static performance analysis of the B‐pillar, the laminate design of the carbon fiber composite B‐pillar reinforcement plate and optimization. For the outer and inner panels of the B‐pillar, the surrogate model method and the multiobjective optimization method are combined to set 11 design variables and 13 responses. The optimal Latin hypercube design method was used to randomly sample each design variable and fit the radial basis function (RBF) approximate model. The modified second‐generation non‐dominated sorting genetic algorithm (MNSGA‐II) was used to carry out multiobjective optimization of the mixed material B‐pillar assembly, and the multicriteria decision‐making method was used to obtain the optimal solution for the Pareto solution set. A drop weight impact test was performed on the hybrid material B‐pillar assembly. The results showed that the peak collision force of the hybrid material B‐pillar was reduced by 43.7% compared with the original steel B‐pillar, the specific energy absorption was increased by 17.5%, and the side impact safety performance was better. After optimization, the weight of the mixed material B‐column was reduced by 26.44%. The improvement rates of the intrusion amount at P2 and the intrusion speed at P1 were 23.99% and 0.63%, respectively, and the lightweight effect and side impact safety were significantly improved.Highlights The finite element analysis parameters of carbon fiber composite materials were constructed through experiments, and the basis for simulation analysis under various working conditions of the B‐pillar based on performance‐driven optimization was established. A hybrid material B‐pillar design method is proposed, which consists of a variable‐thickness high‐strength steel B‐pillar outer plate, a variable‐thickness high‐strength steel B‐pillar inner plate, and a carbon fiber composite B‐pillar reinforcement plate. The super layer of the carbon fiber composite B‐pillar reinforcement plate was optimized and the laying plan of the carbon fiber composite B‐pillar reinforcement plate was determined. Performance‐driven optimization is adopted to realize the collaboration of multilevel design and optimization methods such as B‐pillar conceptual design, detailed design, and multiobjective optimization. A modified NSGA‐II algorithm was proposed and combined with the multicriteria decision‐making method and the entropy‐weighted gray relation analysis method to determine the Pareto optimal solution and the optimal design scheme for the mixed material B‐pillar.

Centrifugal pump is widely used in various fields of the national economy. Pumps, of which 70% are centrifugal pumps, consume 20.9% of the total electricity generation nationwide in China in 2009, according to the statistics. In this paper, the research advances in design methods and characteristics of internal flow fields for centrifugal pumps, carried out by the Research Center of Fluid Machinery Engineering and Technology of Jiangsu University, is introduced. First of all, the Greater Flow Design method, Non-Overload Design method and Splitter Blade Offset Design method are fully discussed through systematically studies. In addition, three methods of performance prediction for centrifugal pumps are adopted, and a CAD/CFD software for hydraulic design and analysis is initially developed. Secondly, to deepen the understanding of characteristics of internal flow in centrifugal pumps, the numerical simulation and PIV measurement investigation have been done. Then the unsteady flow induced vibration and noise are studied considering fluid-structure interaction and fluid-sound interaction. The control of the unsteady flow in centrifugal pumps is also attempted to optimize the design method for centrifugal pumps. In addition, the pressure pulsation and vibration behaviors under both cavitation and non-cavitation conditions are obtained and analyzed. Finally, some detecting methods and criterion for cavitation incipiency and development are put forward.

Design of drainage culverts considering critical storm duration

Lasers that produce femtosecond pulses require dispersion-controlled mirrors. Several researchers have reported on thin-film design methods for these mirrors that are similar to TMDs that produce a stopband. These mirror designs have increased bandwidth over quarter-wave stacks and phase control of the reflected group delay, or GD [also group-delay dispersion (GDD) or wavelength-dependent GDD]. For example, Szipocs et al. reported the use of a monotonically varying quarter-wave stack (starting design) that is optimized to produce a required GDD as a function of wavelength. This monotonically varying stack is referred to as a chirped coating and is similar to progressive series stacks (see Sec. 2.3.2). Tempea et al. reported the use of a layer-thickness modulation method for chirped mirror coatings. This method generates a starting design with improved GDD by chirping the period of the modulated layers while simultaneously increasing the modulation amplitude. In this case, less optimization is required to adjust or smooth the GD of the starting design to target values. Matuschek et al. reported on double-chirped mirrors where more sophisticated analytical design methods produce starting designs that are very close to the desired reflectance and GD. Here, double-chirped starting designs have several features: independently chirped, low and high refractive-index layers; specific groups of layers chirped for impedance matching; and an impedance-matched, broadband antireflection (AR) coating placed on top of the mirror coating to further suppress oscillations in the GD. This starting design requires minimal optimization to improve its performance. Southwell reported on the design of rugate filters using wavelet design methods to increase the bandwidth of the stopband. A wavelet is defined as a fully apodized sine wave refractive-index profile (see Fig. 1.6). This method increases the bandwidth of the resulting stopband by using several overlapping wavelets. These wavelet designs are free of stopband harmonics and reflection ripples in the passbands. Note that consideration of GD is not part of this design method. In all of the above design methods, the objective is to accomplish one or both of the following: (1) to widen the bandwidth and increase the reflectance of the first-order stopband; and (2) to achieve the desired GD (and GDD) versus wavelength. This chapter starts by examining chirped TMDs from a more general perspective, like Chapters 2 and 3, but for only the spectral properties of multiple stopbands. Next, the GD of a chirped-TMD design is briefly discussed. This chapter concludes by considering a half-modulation TMD function and its effect on stopband positions and GD performance. The limitations of both methods are also addressed.