Mixed Input Research Articles

SMT 2.0: A Surrogate Modeling Toolbox with a focus on hierarchical and mixed variables Gaussian processes

The Surrogate Modeling Toolbox (SMT) is an open-source Python package that offers a collection of surrogate modeling methods, sampling techniques, and a set of sample problems. This paper presents SMT 2.0, a major new release of SMT that introduces significant upgrades and new features to the toolbox. This release adds the capability to handle mixed-variable surrogate models and hierarchical variables. These types of variables are becoming increasingly important in several surrogate modeling applications. SMT 2.0 also improves SMT by extending sampling methods, adding new surrogate models, and computing variance and kernel derivatives for Kriging. This release also includes new functions to handle noisy and use multi-fidelity data. To the best of our knowledge, SMT 2.0 is the first open-source surrogate library to propose surrogate models for hierarchical and mixed inputs. This open-source software is distributed under the New BSD license.22https://github.com/SMTorg/SMT.

A multi-cycle pump for efficient separation of CO2 or other trace gas from a mixture of gases

An idealized gas separation pump is described for a membrane filtration system that conserves compressional energy by transferring it from one component to another operating at a different phase of the compression-decompression cycle. An example of the basic principle, we consider a pump that resembles an internal combustion engine with pistons in cylinders and three valves, one for mixed gas input, another for select gas output through a membrane, and a third for non-select gas output to the environment. With the transfer of compressional energy from the decompressing cylinder to the compressing cylinder, the energy required is the sum of frictional losses and the amount needed to compress only the trace gas that is to be separated. Compressional energy for the non-select is recovered. The mechanical energy required is the energy to compress the cylinder, NRT(lnC−1+1∕C)∼3NRT in the isothermal case for number of moles of select gas, N, gas constant R, temperature T, and compression factor C ∼ 50, multiplied by 1 + 2ϵ∕f ∼ 2 for frictional loss fraction ϵ ∼ 6% comparable to that in internal combustion engines and an input select gas molar fraction f ∼ 10% as for CO2 in flue gas. With T = 300 K and N = 2.3 × 104 per ton for CO2, this energy becomes ∼ 6NRT ∼ 94 kWh/tCO2. The advantages of a system like this are that the compression ratio through the membrane can be large and the timescale for compression can be adjusted to maximize the throughput for a given selectivity, overcoming the usual selectivity-permeance trade-off. A disadvantage is that the throughput is generally low for volumetric pumps at practical scales.

A 10-Gb/s low-power inverter-based optical receiver front-end in 0.13-[formula omitted] CMOS process

In the design of CMOS optical receivers, it is challenging to compromise the bandwidth, noise, and gain of the transimpedance amplifier (TIA). The inverter-based cascaded structure is often used in TIA designs. However, the traditional 3-stage TIA needs bandwidth expansion techniques to get good performance. In the designed front-end, we propose a 5-stage cascaded TIA that achieves performance improvement without using bandwidth expansion technique. Besides, the stability of the main amplifier (MA) is improved by adopting a feed-forward compensation technique. Designed in a 0.13-μm CMOS process, the proposed front-end exhibits a bandwidth of 8.9 GHz, a transimpedance gain of 123.5 dB Ω and an average input-referred noise of 12.2 pA/Hz. The whole circuit consumed 19.1 mW from mixed input of 1 V and 1.5 V supply voltages.

A multi-resolution physics-informed recurrent neural network: formulation and application to musculoskeletal systems.

This work presents a multi-resolution physics-informed recurrent neural network (MR PI-RNN), for simultaneous prediction of musculoskeletal (MSK) motion and parameter identification of the MSK systems. The MSK application was selected as the model problem due to its challenging nature in mapping the high-frequency surface electromyography (sEMG) signals to the low-frequency body joint motion controlled by the MSK and muscle contraction dynamics. The proposed method utilizes the fast wavelet transform to decompose the mixed frequency input sEMG and output joint motion signals into nested multi-resolution signals. The prediction model is subsequently trained on coarser-scale input-output signals using a gated recurrent unit (GRU), and then the trained parameters are transferred to the next level of training with finer-scale signals. These training processes are repeated recursively under a transfer-learning fashion until the full-scale training (i.e., with unfiltered signals) is achieved, while satisfying the underlying dynamic equilibrium. Numerical examples on recorded subject data demonstrate the effectiveness of the proposed framework in generating a physics-informed forward-dynamics surrogate, which yields higher accuracy in motion predictions of elbow flexion-extension of an MSK system compared to the case with single-scale training. The framework is also capable of identifying muscle parameters that are physiologically consistent with the subject's kinematics data.

Organic geochemical characteristics and sedimentary environment of source rocks in Northern and Southern Linhe Depression, Hetao Basin

The source rocks in the Linhe Depression of the Hetao Basin are mainly developed within the Paleogene Linhe Formation (E3l) and the Lower Cretaceous Guyang Formation (K1g). The organic geochemical characteristics of the area are still need to be clarified, which restricts the subsequent exploration and development. Therefore, this study will systematically analyze the generation potential, sedimentary environment, and source material of the two sets of source rocks. The experimental results show that: (1) The hydrocarbon generation potential of source rocks in the northern area is more significant than that in the southern area, and the hydrocarbon generation potential of the Linhe Formation is greater than that of the Guyang Formation. (2) The source material in the northern region is mainly the input of higher plants, and the source material in the southern region is the mixed source input of algae and higher plants. (3) The northern and southern source rocks of the Linhe Formation were deposited in a strongly reducing environment of salt water and fresh water, respectively. The source rocks in the northern and southern Guyang Formation were deposited in a freshwater weak reducing environment and freshwater weak oxidizing environment, respectively.

Spatial distribution, source apportionment, and assessment of marine water quality parameters in the Bohai Sea, China

A two–year (2020−2021) survey dataset of six water quality parameters (pH, dissolved oxygen (DO), chemical oxygen demand (COD), dissolved inorganic nitrogen (DIN), soluble reactive phosphate (SRP), and petroleum pollutants) was used to investigate their spatial distribution in the Bohai Sea and quantify their potential sources. There were significant differences in spatial distribution of the parameters. High concentrations of COD, DIN and SRP were found in three bays, with terrestrial input being the main pollution source. Phosphorus–limiting conditions were present in the Bohai Sea. High petroleum pollutant concentrations were identified in port areas, offshore oilfields, and adjacent areas. The pH was above the global oceanic average and there were no signs of acidification. The contribution of the mixed terrestrial inputs, maritime transportation, and offshore oil exploitation sources, oceanic and associated biotic sources, and seawater–atmosphere exchange and atmospheric deposition sources to water quality were 63.4 %, 8.0 %, and 28.6 %, respectively.

Combined Excitation and System Parameter Identification of Dynamic Systems by an Inverse Meta-Model

In the inverse problem, it is common that either the corresponding component parameters or the corresponding input signals are obtained for a given output or response. Most model-based solutions to the inverse problem involve optimization using the so-called forward model. The forward model typically comprises the mechanistic model in some form. Most commonly, inverse problems are formulated in a static setting where a wealth of theoretical results and numerical methods are available. However, there are many important dynamic applications wherein time-dependent information needs to be discerned from time-dependent data. Recently, data-based approaches, or model-free methods, have been invoked whereby feature extraction methods such as Support vector machines (SVM) and artificial neural networks (ANN) are used. Herein we develop an inverse solution for dynamic systems through easy-to-understand least-squares meta-model mathematics. The input and output training data are interchanged, so that a mixed input comprising both component parameters and discrete-time excitations can be found for a given discrete-time output. Single-value decomposition (SVD) makes any matrix inversion tractable. The inverse meta-model is compared to the optimization method and ANN using mechanistic models for fidelity, and is shown to have better accuracy and much increased speed.

Spatial price discrimination in a mixed duopoly input market

We uniquely examine an upstream mixed duopoly engaging in spatial price discrimination across a continuum of downstream markets. The monopoly firms in those markets face elastic final demand creating double marginalization. The upstream public firm faces a cost disadvantage relative to its private rival that declines as it is partially privatized. We show that the fully public firm improves social welfare relative to a private duopoly when it is not overly inefficient and when differentiation is sufficiently large. We also show that whenever this is the case, there exists an optimal degree of partial privatization that better aligns the trade-off between production costs and pricing distortions.

Codesign of Quantized Dynamic Output Feedback MPC for the Takagi–Sugeno Model

We present a co-design of measurement quantized dynamic output feedback model predictive control (DOFMPC) for Takagi-Sugeno model with bounded disturbance. The system output is quantized by a dynamic quantizer before it is transmitted to the DOFMPC controller. Hence, we utilize the dynamic output feedback control law with quantized output signal and consider the mixed input and quantized output constraint for the controller design. By optimizing the quantizer and controller paramters on-line, the control performance is enhanced. Moreover, we formulate a 2-leveled optimizations, with the upper level optimizing the performance index and the lower level optimizing the soft constraint in a lexicographic order, for the co-design of the DOFMPC controller and dynamic quantizer. Thus, there are more degrees of freedom for tighting the soft constraints. The recursive feasibility and stability of the proposed approaches are guaranteed. The applicability of the proposed approach is illustrated by a simulation example.

Adaptive attitude angle constrained fault-tolerant control of hypersonic vehicle with unknown centroid shift

This work takes a further step into the adaptive fault-tolerant control (FTC) scheme to cope with the challenges deriving from the hypersonic vehicle (HSV) in the presence of unknown centroid shift, input saturation, and actuator fault. The influence of unknown centroid shift is mainly manifested in the following aspects: 1) uncertainties of the system input matrix, 2) mixed system uncertainties, and 3)eccentric moments. A novel attitude state constraint control strategy is technically proposed by incorporating the extended barrier Lyapunov function (Ex-BLF) and the Nussbaum-type function into the backstepping design to keep the safe flight of HSV. A unified controller is designed to handle the cases with/without state constraints based on the Ex-BLF with its auxiliary system. For unknown mixed system input matrix due to the coupling of uncertainties of the inertial matrix, actuator fault, and system input saturation, a specific Nussbaum-type function assisted by a state-depend auxiliary system is designed to compensate for the influence of those time-varying nonlinear terms. As a consequence, combined with the Lyapunov stability theory, it is confirmed that the proposed FTC scheme ensures that all the closed-loop signals are bounded. Simulation results are carried out to illustrate the effectiveness and advantage of the proposed control scheme.

Distribution, Sources, and Risk Assessment of Organochlorine Pesticides in Water from Beiluo River, Loess Plateau, China.

The Loess Plateau has been a focus of public discussion and environmental concerns over the past three decades. In this study, in order to investigate the effect of OCP pollution in water of the Beiluo River, concentrations of 25 OCPs at 17 locations in the water were examined. The results showed that the concentration of ∑OCPs in the water ranged from 1.76 to 32.57 ng L-1, with an average concentration of 7.23 ng L-1. Compared with other basins in China and abroad, the OCP content in the Beiluo River was at a medium level. Hexachlorocyclohexane (HCH) pollution in the Beiluo River was mainly from the mixed input of lindane and technical HCHs. Dichlorodiphenyltrichloroethane (DDT) pollution was mainly from the mixed input of technical DDTs and dicofol. Most of the OCP pollution came from historical residues. The risk assessment results showed that hexachlorobenzene (HCB) and endosulfan had high ecological risks in the middle and lower reaches of the Beiluo River. Most residual OCPs were not sufficient to pose carcinogenic and non-carcinogenic health risks to humans. The results of this study can provide a reference for OCP prevention and control and watershed environmental management.

Mixed-Language Input and Infant Volubility: Friend or Foe?

Language mixing is a common feature of many bilingually-raised children's input. Yet how it is related to their language development remains an open question. The current study investigated mixed-language input indexed by observed (30-second segment) counts and proportions in day-long recordings as well as parent-reported scores, in relation to infant vocal activeness (i.e., volubility) when infants were 10 and 18 months old. Results suggested infants who received a higher score or proportion of mixed input in one-on-one social contexts were less voluble. However, within contexts involving language mixing, infants who heard more words were also the ones who produced more vocalizations. These divergent associations between mixed input and infant vocal development point for a need to better understand the causal factors that drive these associations.

Economic, environmental, and social assessment of the replacement of coal-fired thermoelectric plants for solar and wind energy in Chile

In 2019, a plan for the early shutdown of coal-fired power plants in Chile was announced, which will contribute to cleaner production in the energy sector and electricity-intensive industries. For the above, this research simulates the phase-out of coal-fired thermoelectric generation by solar and wind energy with an intersectoral model. Previous literature using a traditional input‒output model can evaluate only the effects of increased renewable energy investment. In contrast, this study is novel because it uses a mixed input‒output model that also allows the production of the thermoelectric, solar, and wind sectors to be exogenously modified. The results show that net national production rises by USD 977 million since the drop in thermoelectric energy is more than compensated by the investment and production in renewable energies. However, employment is reduced by 24 thousand jobs. In addition, replacing coal-fired thermoelectric plants would reduce CO2 emissions at the national level by 37%. The magnitude of the emissions reduction under this climate policy is much higher than the reduction yielded by carbon taxes, environmental taxes on multiple pollutants, and renewable energy subsidies previously evaluated in Chile by other studies. Thus, it is concluded that this policy is more compatible with carbon neutrality. Finally, in the cities where coal-fired thermoelectric plants are located, the social health benefits from lower premature mortality are valued at USD 27 million.

Search for mono-Higgs signals in bb¯ final states using deep neural networks

We study mono-Higgs signatures emerging in an illustrative new physics scenario involving Standard Model Higgs boson decays to bottom quark pairs using hybrid deep neural networks. We use a multilayer perceptron to analyze the kinematic observables and optimize the signal-to-background discrimination. The global color flow structure of hard jets emerging from the decay of heavy particles with different color charges is crucial to single out the mono-Higgs signature. Upon embedding the different color flow structures for signal and backgrounds into constructed images, we use a convolution neural network to analyze the latter. Specifically, the approach takes initially mono-type data as input, frittering away invaluable multisource and multiscale information. We then discuss a general architecture of hybrid deep neural networks that supports instead mixed input data. In comparison with single input deep neural networks, like multilayer perceptron or convolution neural networks, the hybrid deep neural networks provide higher capacity in feature extraction and thus in signal versus background classification performance. We provide reference results for the case of the high-luminosity Large Hadron Collider.

Organic matter pores in the chang 7 lacustrine shales from the Ordos Basin and its effect on reflectance measurement

To quantify the pore characteristics of various macerals in Chang 7 lacustrine shales, macerals were effectively identified according to their optical and morphological characteristics, and the nanoscale pore structure of macerals was observed by scanning electron microscope. Meanwhile, the reflectances of different positions in the same pieces of vitrinite or solid bitumen with heterogeneous pores development were measured. The results showed that the average contents of sapropelinite, liptinite, vitrinite, inertinite and solid bitumen are 42.7%, 8.7%, 13.6%, 13.8% and 21.2%, respectively, which suggests that the source of the organic matter of the Chang 7 shales is a mixed source input. The organic pores of Chang 7 shales are enriched, and the pore shapes are mostly round or elliptical. The pore size of organic pores has a wide distribution, mainly concentrate in the range of 100–400 nm, and the average plane porosity of organic pores is 10.13%. The size order of the organic pores in various macerals is: solid bitumen < bituminite < alginite < vitrinite < fusinite < liptinite. The abundance order of organic matter pores of each maceral is as follows: alginite > fusinite > bituminite > solid bitumen > vitrinite > liptinite. OM pores are mainly contributed by bituminite, solid bitumen and fusinite. The plane porosity of bituminite increases with maturity. In the process of thermal evolution, the plane porosity of fusinite is distributed in the two ranges of 20%–28% and 1%–7%. The former is mainly the primary pores of the fusinite itself, and the latter is the secondary pores formed in the thermal evolution. As for the organic pores of other macerals, no obvious thermal evolution law was found. Meanwhile, the surface imperfections of vitrinite or solid bitumen is enhanced by the enrichment of organic pores (an increase in pore size or pore number), which may result in the underestimation of their reflectances.

The Impact of Formal Agricultural Credit on Farm Productivity and Its Utilization in Khyber Pakhtunkhwa, Pakistan

Agricultural mechanization and the use of mixed inputs ensure higher productivity. In this regard, the Government of Pakistan has executed an approach for the provision of agricultural credit to smallholders for improving agricultural production. However, many issues are involved in the extent, utilization and impacts of credit. This study aims to explore the impact of credit, its utilization, misuse and factors determining the extent of agricultural credit in Pakistan. A semi-structured questionnaire was developed to collect data from 316 farmers in Mardan District. The study was analyzed through statistical tools such as a paired t-test, ANOVA and multiple regressions. Results showed that agricultural credit enhanced crops’ production. However, according to percentage use, misuse of the credit was more common than its proper utilization. Farmers utilized credit for land preparation, fertilizers, seeds, pesticides and daily labor. They misused agricultural credit for healthcare, education of children, domestic needs and business. Results of the regression model showed that farmers’ age, experience, farm size, farm income, farm labor and land ownership were determinants of the extent of agricultural credit. Policy measures should be taken to stop the misuse of agricultural credit to achieve the target set for agricultural productivity.

Geochemical Responses to Natural and Anthropogenic Settings in Salt Lakes Sediments from North-Eastern Romanian Plain

Chemical analysis was performed on sediment core samples collected from three salt lakes, Amara Lake, Caineni Lake, and Movila Miresii Lake, located in the northeast of the Romanian Plain. The concentration of 10 main elements, 6 heavy metals (HMs), 8 rare earth elements (REEs), and 10 trace elements (TEs)-determined using neutron activation analysis (NAA)-showed variability dependent on the depth sections, lake genesis and geochemical characteristics (oxbow, fluvial harbor/liman and loess saucer type). The assessment of pollution indices (contamination factor, pollution load index, geoaccumulation index, and enrichment factor) highlighted low and moderate degrees of contamination for most of the investigated elements. Principal component analysis (PCA) extracted three principal components, explaining 70.33% (Amara Lake), 79.92% (Caineni Lake), and 71.42% (Movila Miresii Lake) of the observed variability. The principal components extracted were assigned to pedological contribution (37.42%-Amara Lake, 55.88%-Caineni Lake, and 15.31%-Movila Miresii Lake), salts depositions (due to the lack of a constant supply of freshwater and through evaporation during dry periods), atmospheric deposition (19.19%-Amara Lake, 13.80%-Caineni Lake, and 10.80%-Movila Miresii Lake), leaching from soil surface/denudation, rock weathering, and mixed anthropogenic input (e.g., agricultural runoff, wastewater discharges) (13.72%-Amara Lake, 10.24%-Caineni Lake, and 45.31%-Movila Miresii Lake).

Saturated Attitude Control of Multi-Spacecraft Systems on SO(3) Subject to Mixed Attitude Constraints With Arbitrary Initial Attitude

In this paper, for multi-spacecraft systems (MSSs) with a directed complete communication topology and a time-varying virtual leader, an adaptive saturated attitude controller is proposed to achieve attitude consensus and attitude tracking under arbitrary initial attitude, mixed attitude constraints, input saturation and external disturbances. Firstly, considering the time-varying desired attitude provided by the virtual leader in a directed complete topology, an MSS attitude error function and an MSS attitude error dynamics based on SO(3) are developed. Next, an effective mixed potential function for the MSS on SO(3) is proposed for the static attitude-forbidden zones, the relative dynamic attitude-forbidden zones and the attitude-mandatory zones. In particular, different from the existing potential functions, the proposed mixed potential function is suitable for arbitrary initial attitude of the spacecraft in MSS, relaxing the restriction on the initial attitude associated with each static and dynamic attitude constraint zones. Then, an adaptive saturated attitude controller is designed to realize attitude consensus and tracking for the MSSs on SO(3) under arbitrary initial attitude, mixed attitude constraints, saturation constraints and external disturbances. Finally, simulation results of an MSS with a time-varying virtual leader are demonstrated to illustrate the efficiency of the proposed attitude controller.

Method for economic loss estimation in traditional coastal salt farming towards tidal inundation

Tidal flooding is a significant threat to salt farming regions, but this issue has not received much attention in the academic literature. Furthermore, the limited availability of data hinders further analysis of the economic impact of these recurrent flooding events. To address these challenges, this study proposes a practical method for assessing financial loss in salt farming by leveraging a combination of mixed hydrodynamic inputs, damage analysis, and GIS techniques. Specifically, the method involves developing a synthetic vulnerability curve and cost/revenue (C/R) data based on local community knowledge, which can provide a granular understanding of the impact of tidal flooding on salt farming operations. The effectiveness of this approach was demonstrated in one of the significant salt production regions in Indonesia, where the method was successfully applied. Moreover, the approach has the potential to be replicated in other coastal regions with limited data available, making it a valuable tool for policymakers and local communities. Finally, the method provides individual loss analysis at the parcel level, allowing for a more granular understanding of the financial impact of tidal flooding on salt farming operations.in Indonesia the potential to be replicated in the other coastal region with limited data available.•This method allows quick and detailed flood loss estimation using established vulnerability curves in different production stages.•It is a cost-efficient method to evaluate the impact of tidal flooding based on one-time recorded events.•It delivers individual loss analysis at the parcel level.

Develop Then Rival: A Human Vision-Inspired Framework for Superimposed Image Decomposition

A single superimposed image containing two image views causes visual confusion for both human vision and computer vision. Human vision needs a "develop-then-rival" process to decompose the superimposed image into two individual images, which effectively suppresses visual confusion. However, separating individual image views from a single superimposed image has been an important but challenging task in computer vision area for a long time. In this paper, we propose a human vision-inspired framework for single superimposed image decomposition. We first propose a network to simulate the development stage, which tries to understand and distinguish the semantic information of the two layers of a single superimposed image. To further simulate the rivalry activation/suppression process in human brains, we carefully design a rivalry stage, which incorporates the original mixed input (superimposed image), the activated visual information (outputs of the development stage) together, and then rivals to get images without ambiguity. Experimental results show that our novel framework effectively separates the superimposed images and significantly improves the performance with better output quality compared with state-of-the-art methods. The proposed method also achieves state-of-the-art results on related applications including single image reflection removal, single image rain removal, single image shadow removal, and illumination correction, etc., which validates the generalization of the framework.