Auxiliary Weight Research Articles

Accurate density measurement of a small solid sample using a combination of hydrostatic weighing and pressure-of-flotation methods

Abstract A new high-precision density measurement method for small non-silicon samples was developed. Hydrostatic weighing (HW) and the pressure-of-flotation method (PFM) are high-precision techniques for determination solid sample densities. However, in HW, the uncertainty of the density measurement is quite large when the volume of the sample is much smaller than that of a 1 kg silicon sample owing to the uncertainty of the auxiliary weights during the measurement. Additionally, the PFM is not applicable for measuring the density of non-silicon materials. We developed a new method that combines HW and the PFM. We prepared a small 45 g Si3N4 sphere sample with a diameter of 30 mm. By employing two silicon discs with these masses that were nearly equivalent to the apparent masses of the sample, the same auxiliary weights were used for both the standards and the sample. The new method was validated by measuring the density of the 45 g Si3N4 sphere with a relative standard uncertainty of 4.6 × 10−6, which is significantly lower than that obtained by traditional HW methods. This approach demonstrates substantial improvements in the precision of small-sample density measurements, thus offering a robust alternative to conventional techniques.

Ground and Excited States from Ensemble Variational Principles

The extension of the Rayleigh-Ritz variational principle to ensemble states ρw≡∑kwk|Ψk⟩⟨Ψk| with fixed weights wk lies ultimately at the heart of several recent methodological developments for targeting excitation energies by variational means. Prominent examples are density and density matrix functional theory, Monte Carlo sampling, state-average complete active space self-consistent field methods and variational quantum eigensolvers. In order to provide a sound basis for all these methods and to improve their current implementations, we prove the validity of the underlying critical hypothesis: Whenever the ensemble energy is well-converged, the same holds true for the ensemble state ρw as well as the individual eigenstates |Ψk⟩ and eigenenergies Ek. To be more specific, we derive linear bounds d−ΔEw≤ΔQ≤d+ΔEw on the errors ΔQ of these sought-after quantities. A subsequent analytical analysis and numerical illustration proves the tightness of our universal inequalities. Our results and particularly the explicit form of d±≡d±(Q)(w,E) provide valuable insights into the optimal choice of the auxiliary weights wk in practical applications.

Relaxation and optimal finiteness domain for degenerate quadratic functionals. One-dimensional case

The aim of this paper is the study, in the one-dimensional case, of the relaxation of a quadratic functional admitting a very degenerate weight w, which may not satisfy both the doubling condition and the classical Poincaré inequality. The main result deals with the relaxation on the greatest ambient space L0(Ω) of measurable functions endowed with the topology of convergence in measure w dx. Here w is an auxiliary weight fitting the degenerations of the original weight w. Also the relaxation w.r.t. the L2(Ω, w˜)-convergence is studied. The crucial tool of the proof is a Poincaré type inequality, involving the weights w and w, on the greatest finiteness domain Dw of the relaxed functionals.

Product color emotional design based on 3D knowledge graph

To address the problem of fragmentation, integration difficulties in fuzzy front-end information, and ambiguity in color emotion knowledge representation and conversion within the current product color emotion design stage, this paper proposes a method based on 3D Knowledge Graph. The proposed approach aims to integrate product color emotion design into the “data knowledge + artificial intelligence” growth model, facilitating a beneficial knowledge output cycle driven by data. As for the proposed approach, it divides the design problem into three stages: in the first one, big data web crawler technology and natural language processing were employed to extract knowledge related to product color emotion design. In the second phase, the construction of a product color emotion imagery association model using the RotatE knowledge graph, thereby achieving knowledge fusion of product color emotion design, and all accumulated knowledge data is integrated into a 3D Knowledge Graph for visualization. Finally, in the third phase, the PageRank algorithm is applied to calculate primary and auxiliary color weight parameters, simulating the product color synergy mechanism and determining the color synergy effects. Then, realize the knowledge generation of product color emotional design. This approach combines the human visual experience feedback with extensive big data analysis, and accurately outputs the product color emotion design scheme that meets the user's emotional needs. Moreover, the effectiveness and applicability of the method are verified by an illustrative example involving modern machine tool.

Development and Validation of a Smart Architecture for Thyristor Valves

A high voltage thyristor converter is realized by many valve sections, whose volume is approximately occupied for only the 10% by thyristors and for the 10% by the relevant gate drivers. The remaining 80% is taken by passive auxiliary circuits, needed to protect thyristors during turn-on and turn-off commutations. This work represents a preliminary validation of an innovative architecture that aims to reduce the auxiliary circuit cost, volume, and weight of the overall valve, through the investigation of active, instead of passive solutions. The work starts from investigation of the phenomena behind the valve transient behaviors and proceeds with simulations and experimental tests, using a reduced scale circuit prototype. The match between the obtained results validates the investigated active configurations, confirming that the proposed solution can be used for improving thyristor valves.

Pattern-Reconfigurable Sparse Linear Array Synthesis Under Minimum Element Spacing Control by Alternating Sequential Quadratic Programming

A new method called alternating sequential quadratic programming (ASQP) is proposed to synthesize pattern-reconfigurable sparse linear arrays with minimum element spacing control. The method can find the common element positions and multiple sets of excitations for generating multiple reconfigurable patterns which accurately meet their given upper and lower pattern bounds. In addition, by introducing auxiliary weighting coefficients and collective excitation coefficient vectors and choosing them as optimization variables alternately, the proposed method can appropriately incorporate the minimum element spacing constraint into the pattern synthesis. Synthesized results show that the proposed method can give satisfactory reconfigurable pattern performance but save much more elements compared with some existing methods.

Diluted binary neural network

Binary neural networks (BNNs) are promising on resource-constrained devices because they reduce memory consumption and accelerate inference effectively. However, they are still potential on performance improvement. Prior studies attribute performance degradation of BNNs to limited representation ability and gradient mismatch. In this paper, we find that it also results from the mandatory representation of small full-precision auxiliary weights to large values. To tackle with this issue, we propose an approach dubbed as Diluted Binary Neural Network (DBNN). Besides avoiding mandatory representation effectively, the proposed DBNN also alleviates sign flip problem to a large extent. For activations, we jointly minimize quantization error and maximize information entropy to develop the binarization scheme. Compared with existing sparsity-binarization approaches, DBNN trains network from scratch without other procedures and achieves larger sparsity. Experiments on several datasets with various networks demonstrate the superiority of our approach.

Adversarial Auxiliary Weighted Subdomain Adaptation for Open-Set Deep Transfer Bridge Damage Diagnosis.

Deep learning models have been widely used in data-driven bridge structural damage diagnosis methods in recent years. However, these methods require training and test datasets to satisfy the same distribution, which is difficult to satisfy in practice. Domain adaptation transfer learning is an efficient method to solve this problem. Most of the current domain adaptation methods focus on close-set scenarios with the same classes in the source and target domains. However, in practical applications, new damage caused by long-term degradation often makes the target and source domains dissimilar in the class space. For such challenging open-set scenarios, existing domain adaptation methods will be powerless. To effectively solve the above problems, an adversarial auxiliary weighted subdomain adaptation algorithm is proposed for open-set scenarios. Adversarial learning is introduced to proposed an adversarial auxiliary weighting scheme to reflect the similarity of target samples with source classes. It effectively distinguishes unknown damage from known states. This paper further proposes a multi-channel multi-kernel weighted local maximum mean discrepancy metric (MCMK-WLMMD) to capture the fine-grained transferable information for conditional distribution alignment (sub-domain alignment). Extensive experiments on transfer tasks between three bridges verify the effectiveness of the algorithm in open-set scenarios.

Linear Programming Bounds for Approximate Quantum Error Correction Over Arbitrary Quantum Channels

While quantum weight enumerators establish some of the best upper bounds on the minimum distance of quantum error-correcting codes, these bounds are not optimized to quantify the performance of quantum codes under the effect of arbitrary quantum channels that describe bespoke noise models. Herein, for any Kraus decomposition of any given quantum channel, we introduce corresponding quantum weight enumerators that naturally generalize the Shor-Laflamme quantum weight enumerators. We establish an indirect linear relationship between these generalized quantum weight enumerators by introducing an auxiliary exact weight enumerator that completely quantifies the quantum code's projector, and is independent of the underlying noise process. By additionally working within the framework of approximate quantum error correction, we establish a general framework for constructing a linear program that is infeasible whenever approximate quantum error correcting codes with corresponding parameters do not exist. Our linear programming framework allows us to establish the non-existence of certain quantum codes that approximately correct amplitude damping errors, and obtain non-trivial upper bounds on the maximum dimension of a broad family of permutation-invariant quantum codes.

On 2MP-, MP2- and C2MP-inverses for rectangular matrices

This paper introduces 2MP-inverses, MP2-inverses, and C2MP-inverses, for rectangular matrices following a different approach to that used in the recent literature. These new inverses generalize some classical inverses in the literature. Instead of considering a system of matrix equations as usually, in order to define 2MP-inverses and MP2-inverses, we consider a construction from oblique projectors represented by means of outer generalized inverses. We use an adequate equivalence relation, and then we pass to the quotient set in order to get the most simple canonical representative. An interesting advantage of our extension of CMP inverses from square to rectangular matrices is that we do not need any auxiliary weight matrix, but we are using the own matrix A for doing it. In addition, some properties and representations of 2MP-, MP2-, and C2MP-inverses are given.

Deep weighted joint distribution adaption network for fault diagnosis of blast furnace ironmaking process

The blast furnace (BF) ironmaking process has the following characteristics: (1) few fault data (2) large fluctuation in data distribution. For robust fault diagnosis, some transfer learning-based methods are proposed. They mostly aim at aligning only the marginal and conditional distribution discrepancy of datasets. However, in actual application, ignoring the prior distribution will lead to degraded domain adaptation performance. Hence, we propose a new fault diagnosis method called deep weighted joint distribution adaption network (DWJDAN) to align the probability distributions of the dataset more comprehensively. The DWJDAN consists of three modules: condition recognition, class prior distribution adaptation, and joint distribution adaptation. The condition recognition is constructed by a convolutional network (CNN) to extract features and generate health labels of BF. The class prior distribution adaptation generates class-specific auxiliary weights to exploit the prior probability on the source and target domains. The joint distribution adaption facilitates the CNN to learn domain-invariant features by minimizing maximum mean discrepancy (MMD) with weighted data. Experiments using actual industrial BF data show that the proposed method can obtain good results in diagnosing the abnormal conditions of the BF ironmaking process.

Indirect adaptive control of multi-input-multi-output nonlinear singularly perturbed systems with model uncertainties

In this paper, two indirect adaptive control schemes for a class of multi-input-multi-output nonlinear singularly perturbed systems with partially unknown models and parameters are presented. Firstly, the original system dynamic equation is reformulated into a new form with identity control gain matrices, and a new multi-time-scale singularity-free neural network is employed to represent the new dynamic equation. Subsequently, an online identification scheme is proposed to update neural network weights where a set of auxiliary weight error vectors are used such that better convergence property can be achieved. Based on identification results, a singularity-free singular perturbation controller is developed to control the unknown nonlinear system. By using the singularity-free neural network and singular perturbation technique, the complexity in controller design for a singularly perturbed system is reduced, and the potential singularity problem is avoided. Moreover, a singularity-free dynamic surface control scheme is also proposed, and the “explosion of complexity” issue is relieved. Compared to conventional direct adaptive dynamic surface control schemes which use gradient-like updating laws and tracking errors to train the neural networks, the singularity-free dynamic surface controller is designed indirectly and the neural networks are updated using the identification errors. Therefore, better identification and control performance is achieved, and the potential singularity problem is also circumvented. The stability of the closed-loop system is rigorously proved via the Lyapunov approach, and the effectiveness of proposed identification and control schemes is demonstrated by simulations.

Sum/Difference Pattern Synthesis With Dynamic Range Ratio Control for Arbitrary Arrays

This article proposes a method which generates a set of weights to synthesize sum or difference pattern with precisely controlled sidelobe level (SLL), null, and dynamic range ratio (DRR) for arbitrary arrays. Our pattern synthesis approach reduces mutual coupling (MC) between the neighboring elements and complexity of the feeding network design. However, the formulated optimization problem is nonconvex due to the nonconvex objective function and fractional DRR constraint. To tackle it, we first introduce two sets of auxiliary variables: one for DRR constraint and the other for sidelobe and null constraints. By doing so, we then decompose the original optimization problem into three sets of subproblems characterized by the auxiliary variables and weight variables. To facilitate the subproblem with weight variables reaching its optimum values, we derive an appropriate range of step size. Finally, we iteratively solve these subproblems to obtain the solution to the original problem. Extensive experiments employing non-equispaced linear and rectangular arrays, concentric ring array, and cylinder array, are implemented to demonstrate that the developed approach can accurately control SLLs, null and DRR for arbitrary arrays.

Design of home intelligent robot of internet of things

An aging society is the challenge we will face and deal with. It is necessary and important to explore the problems and solutions faced by an aging society. In response to this phenomenon, we designed an IoT home smart robot based on ESP8266 and STM32 to improve the quality of life of the elderly. The robot can realize functions such as environmental monitoring, fire warning, electrical power distribution, light control, auxiliary weight lifting, weather forecast, etc. Connecting to the Baidu cloud platform can realize data visualization, allowing children to remotely view the family's data and equipment conditions in the cloud, and has the function of automatically mopping the floor to reduce the burden of housework. On the basis of the mechanical structure design, combined with the current emerging Internet of Things technology and electronic control technology, mechatronics, the research and development cost is low, the function is complete, and it has strong practicability and innovation.

Kinship Verification Based on Cross-Generation Feature Interaction Learning.

Kinship verification from facial images has been recognized as an emerging yet challenging technique in many potential computer vision applications. In this paper, we propose a novel cross-generation feature interaction learning (CFIL) framework for robust kinship verification. Particularly, an effective collaborative weighting strategy is constructed to explore the characteristics of cross-generation relations by corporately extracting features of both parents and children image pairs. Specifically, we take parents and children as a whole to extract the expressive local and non-local features. Different from the traditional works measuring similarity by distance, we interpolate the similarity calculations as the interior auxiliary weights into the deep CNN architecture to learn the whole and natural features. These similarity weights not only involve corresponding single points but also excavate the multiple relationships cross points, where local and non-local features are calculated by using these two kinds of distance measurements. Importantly, instead of separately conducting similarity computation and feature extraction, we integrate similarity learning and feature extraction into one unified learning process. The integrated representations deduced from local and non-local features can comprehensively express the informative semantics embedded in images and preserve abundant correlation knowledge from image pairs. Extensive experiments demonstrate the efficiency and superiority of the proposed model compared to some state-of-the-art kinship verification methods.

Synthesis of Multibeam Sparse Circular-Arc Antenna Arrays Employing Refined Extended Alternating Convex Optimization

A refined extended alternating convex optimization (REACO) method is presented to synthesize multibeam sparse circular-arc antenna arrays with minimum element spacing control by considering real antenna array structure characteristics. This method consists of the initial step and a few refining steps. At the initial step, an initial array with dense elements distributed on a circular-arc is considered, and its array manifold vector is described by rotating a simulated isolated element pattern (IEP) without considering element mutual coupling. The collective excitation coefficient vector (CECV) and its energy bound are introduced for each element, and consequently, the common element positions for generating desired multibeam patterns can be found by minimizing the number of active CECVs under multiple constraints. This minimization problem is further formulated as performing a sequence of alternating convex optimization (ACO) in which the CECV and an auxiliary weighting vector are alternately chosen as the optimization variables so that the minimum element spacing constraint can be easily dealt with. Once the initial optimization step is finished, a few refining steps are performed in which the element positions and excitations are successively updated in each step by renewing the array manifold vector through rotating the simulated nearby active element patterns (AEPs) of the antenna array obtained at the previous step. In such a way, the mutual coupling can be incorporated into the multibeam sparse array synthesis. An example of synthesizing a sparse circular-arc conformal array with 23 beams covering the space from −63.25° to 63.25° is conducted to validate the effectiveness and advantage of the proposed method.

Weighted and Class-Specific Maximum Mean Discrepancy for Unsupervised Domain Adaptation

Although maximum mean discrepancy (MMD) has achieved great success in unsupervised domain adaptation (UDA), most of existing UDA methods ignore the issue of class weight bias across domains, which is ubiquitous and evidently gives rise to the degradation of UDA performance. In this work, we propose two improved MMD metrics, i.e., weighted MMD (WMMD) and class-specific MMD (CMMD), to alleviate the adverse effect caused by the changes of class prior distributions between source and target domains. In WMMD, class-specific auxiliary weights are deployed to reweigh the source samples. In CMMD, we calculate the MMD for each class of source and target samples. Since the class labels of target samples are unknown for UDA problem, we present a classification expectation-maximization algorithm to estimate the pseudo-labels of target samples on the fly and update the model parameters using estimated labels. The proposed methods can be flexibly incorporated into deep convolutional neural networks to form WMMD and CMMD based domain adaptation networks, which we called WDAN and CDAN, respectively. By combining WMMD with CMMD, we present a CWMMD based domain adaptation network (CWDAN) to further improve classification performance. Experiments show that, both WMMD and CMMD benefit the classification accuracy, and our CWDAN can achieve compelling UDA performance in comparison with MMD and the state-of-the-art UDA methods.

Comparison of different interpolation methods for spatial distribution of soil organic carbon and some soil properties in the Black Sea backward region of Turkey

The purpose of this research is to compare the spatial variability of soil organic carbon (SOC) in four adjacent land uses including the cultivated area, the grassland area, the plantation area and the natural forest area in the semi - arid region of Black Sea backward region of Turkey. Some of the soil properties, including total nitrogen, SOC, soil organic matter, and bulk density were measured on a grid with a 50 m sampling distance on the top soil (0–15 cm depth). Accordingly, a total of 120 samples were taken from the four adjacent land uses. Data was analyzed using geostatistical methods. The methods used were: Block kriging (BK), co - kriging (CK) with organic matter, total nitrogen and bulk density as auxiliary variables and inverse distance weighting (IDW) methods with the power of 1, 2 and 4. The methods were compared using a performance criteria that included root mean square error (RMSE), mean absolute error (MAE) and the coefficient of correlation (r). The one - way ANOVA test showed that differences between the natural (0.6653 ± 0.2901) - plantation forest (0.7109 ± 0.2729) areas and the grassland (1.3964 ± 0.6828) - cultivated areas (1.5851 ± 0.5541) were statistically significant at 0.05 level (F = 28.462). The best model for describing spatially variation of SOC was CK with the lowest error criteria (RMSE = 0.3342, MAE = 0.2292) and the highest coefficient of correlation (r = 0.84). The spatial structure of SOC could be well described by the spherical model. The nugget effect indicated that SOC was moderately dependent on the study area. The error distributions of the model showed that the improved model was unbiased in predicting the spatial distribution of SOC. This study's results revealed that an explanatory variable linked SOC increased success of spatial interpolation methods. In subsequent studies, this case should be taken into account for reaching more accurate outputs.

Synthesis of Unequally Spaced Linear Antenna Arrays With Minimum Element Spacing Constraint by Alternating Convex Optimization

A novel method called alternating convex optimization is presented to synthesize unequally spaced linear arrays with minimum element spacing constraint. In this method, the problem of synthesizing an unequally spaced array is formulated as a sequence of alternating convex optimization problems, and the excitation vector and auxiliary weighting vector are alternately chosen as the optimization variables. The minimum spacing constraint for considering the physical element antenna size can be easily imposed in this alternating optimization process. Two examples for synthesizing unequally spaced linear arrays with focused and shaped patterns are provided to validate the effectiveness and advantages of the proposed method.

The method of similar trajectories with branching according to parametric maximum of the auxiliary weight

Abstract The weighted method of similar trajectories (MST) allows one to construct estimators of functionals on a single Markov chain simultaneously for a given range of parameters of the problem. Choosing an appropriate Markov chain, we take into account additional conditions providing the finiteness of MST variance. A modification of the weighted MST with branching of chain trajectory is constructed in the paper according to the parametric maximum of the auxiliary weight. It is proved that the computational cost of this algorithm is bounded if the basis functionals are also bounded. Numerical study of the efficiency of the modified MST in comparison with analog modelling was carried out on the example of the standard problem of transfer theory on estimation of the probability of albedo and transmission of a particle.