Optimal Model Structure Research Articles

A novel discrete evidence fusion approach by considering the consistency of belief structures

Fusion of discrete belief structures gets much attention owing to the widespread existence of discrete information in decision analysis, expert system and other fields. Recently, one method that combines discrete evidence was proposed by establishing an optimization model. However, the existing optimization model of discrete belief structure has a problem of high computational complexity since it needs to calculate all the cases of deterministic evidence combination to obtain maximum and minimum values of each focal element in combination result. In this paper, a novel method of discrete evidence fusion is proposed to reduce the computational complexity by considering the consistency of belief structures of evidence groups and finding the most consistent and most inconsistent cases. Example and application are given to illustrate the effectiveness and rationality of the proposed method.

Characteristics of a new multi-channel sensing device based on C-type photonic crystal fibers

A new C-type photonic crystal fiber array was first proposed for multi-channel sensing. This design utilized the high birefringence characteristic of C-type structure and the excellent of photonic crystal array. Surface Plasmon Resonance (SPR) was chosen as the sensing principle to get high sensitivity. Through a detailed optimization process, the planar array got an optimal structure model, whose inner layer pores and defect are the main sensing channels. If some analytes with different refractive indices were entered into these channels respectively, resonance -matching conditions would change, which would superimpose on the original birefringence effect and output multiple characteristic signals. Finally, in a multi-phase infiltration condition, the device showed 5 characteristic output peaks that corresponded to different analytes. A gradual changing sensitivity of 13,091–45,322 nm/RIU was achieved in a wide refractive index range (1.33–1.44). Such a new sensing device could be regarded as a multi-channel micro-laboratory, which had a great potential for simultaneous sensing of multiple component parameters.

Research on landscape Spatial Structure Optimization of Garden sculpture based on improved Particle Swarm Optimization algorithm

Traditional methods in landscape sculpture landscape spatial structure optimization, did not consider the harmony between sculpture landscape and garden, there is a problem of long optimization processing time. Therefore, this paper designs a new optimization method of landscape sculpture spatial structure by improving particle swarm optimization algorithm. This paper analyzes the influence of garden sculpture on the whole space of the garden, and establishes the landscape space structure optimization model of garden sculpture. The spatial optimization index of garden sculpture landscape was established through analysis, and the spatial optimization constraints were set. The improved particle swarm optimization algorithm was used to solve the optimization model, and the optimization of the sculpture landscape spatial structure was completed. Compared with the optimization method based on genetic algorithm, the experiment shows that the optimization method based on improved particle algorithm has short processing time, and the optimized vegetation area, harmony between sculpture and garden are significantly improved, with better application effect.

Особливості застосування вартісної форми моделі міжпродуктового балансу до визначення обсягових і цінових показників розвитку енергетичного сектора та інших галузей економіки країни

The deformation of price system due to the inflated prices for energy resources, raw materials and supplies has led to an unreasonable rise of prices for products, a decrease in the competitiveness of country's economy, a reduction of the demand for products and, consequently, their output. These factors have led to an increase in the share of intermediate consumption in the structure of output and are one of the main reasons for low profitability, crisis of sales and increase in debt between domestic enterprises. In this context, the aim of this article is to study the possibility of applying macroeconomic indicators of the development of economy to solve the problem of determining output in kind, as well as equilibrium prices under conditions of limited information on the forecasts of final consumption, development of an optimization model of interproduct balance, intended for the analysis of change in the volumes and prices of economic system only in terms of cost indicators of its development, in particular, the structure and volumes of gross value added. In order to use the macroeconomic indicators of economy development at constructing a forecasting fuel and energy balance in the absence of information on final consumption, we proposed a system of conditions of optimization model, which uses the gross value added of branches to form restrictions on the volumes of output in physical units of their products. These conditions use the representation of subsystems of the distribution of outputs and inputs of the model of interproduct balance as a bilinear form in terms of the volumes of output in kind and equilibrium prices. Based on such forms, we constructed a method of determining outputs and prices to solve problems of forecasting and analyzing changes in country's economy according to macroeconomic indicators of its development, in particular, gross value added, in the absence of information on final demand for products. We proposed the structure of optimization model of interproduct balance based on the system of balance equations of the distribution of expenses in branches of economy in the cost form, which uses the indicators of gross value added for the solution of problems of forecasting the output of products in natural form and their volumes at substantial changes in the structure of gross value added. Keywords: fuel and energy balance, interproduct balance, model, optimization, forecasting, gross value added

Optimization-Based Data-Enabled Modeling Technique for HVAC Systems Components

Most of the energy consumed by the residential and commercial buildings in the U.S. is dedicated to space cooling and heating systems, according to the U.S. Energy Information Administration. Therefore, the need for better operation mechanisms of those existing systems become more crucial. The most vital factor for that is the need for accurate models that can accurately predict the system component performance. Therefore, this paper’s primary goal is to develop a new accurate data-driven modeling and optimization technique that can accurately predict the performance of the selected system components. Several data-enabled modeling techniques such as artificial neural networks (ANN), support vector machine (SVM), and aggregated bootstrapping (BSA) are investigated, and model improvements through model structure optimization proposed. The optimization algorithm will determine the optimal model structures and automate the process of the parametric study. The optimization problem is solved using a genetic algorithm (GA) to reduce the error between the simulated and actual data for the testing period. The models predicted the performance of the chilled water variable air volume (VAV) system’s main components of cooling coil and fan power as a function of multiple inputs. Additionally, the packaged DX system compressor modeled, and the compressor power was predicted. The testing results held a low coefficient of variation (CV%) values of 1.22% for the cooling coil, and for the fan model, it was found to be 9.04%. The testing results showed that the proposed modeling and optimization technique could accurately predict the system components’ performance.

Automatic Model Structure Identification for Conceptual Hydrologic Models

AbstractChoosing (an) adequate model structure(s) for a given purpose, catchment, and data situation is a critical task in the modeling chain. However, despite model intercomparison studies, hypothesis testing approaches with modular modeling frameworks, and continuous efforts in model development and improvement, there are still no clear guidelines for identifying a preferred model structure. By introducing a framework for Automatic Model Structure Identification (AMSI), we support the process of identifying (a) suitable model structure(s) for a given task. The proposed AMSI framework employs a combination of the modular hydrological model RAVEN and the heuristic global optimization algorithm dynamically dimensioned search (DDS). It is the first demonstration of a mixed‐integer optimization algorithm applied to simultaneously optimize model structure choices (integer decision variables) and parameter values (continuous decision variables) in hydrological modeling. The AMSI framework is thus able to sift through a vast number of model structure and parameter choices for identifying the most adequate model structure(s) for representing the rainfall‐runoff behavior of a catchment. We demonstrate the feasibility of the approach by reidentifying given model structures that produced a specific hydrograph and show the limits of the current setup via a real‐world application of AMSI on 12 MOPEX catchments. Results show that the AMSI framework is capable of inferring feasible model structure(s) reproducing the rainfall‐runoff behavior of a given catchment. However, it is a complex optimization problem to identify model structure and parameters simultaneously. The variance in the identified structures is high due to near equivalent diagnostic measures for multiple model structures, reflecting substantial model equifinality. Future work with AMSI should consider the use of hydrologic signatures, case studies with multiple types of observation data, and the use of mixed‐integer multiobjective optimization algorithms.

Structural optimization design of lightweight rectangular reflective mirror

In order to solve the problem that the mass and the surface figure accuracy of the space reflective mirror are often contradictory in the lightweight design process, a structural optimization design of a lightweight rectangular reflective mirror of an off-axis three-reflection optical system is performed. In this study, a lightweight structure based on the center support of SiC materials is proposed. At the same time, a multi-objective optimization method is introduced. With the RMS value and Mass as the optimization targets at the same time, a mirror optimal structure model is obtained with a mass of 2.32 kg. Compared with the solid mirror, the lightweight ratio is 73.8%. Then the mirror subassembly is designed and the integrated performance of it is simulated. It shows that the RMS value of the mirror reaches respectively 2.5 nm, 2.2 nm and 7.3 nm when gravity load is applied in the directions of X, Y and Z axes. Furthermore, the RMS value is 3.2 nm when the mirror subassembly is under the load condition of uniform temperature rise of 4 ℃, which is far less than the requirement of RMS≤λ/50(λ=632.8 nm). Therefore the data meets the design requirements.

Lattice structure design optimization coupling anisotropy and constraints of additive manufacturing

Replacing solid structures with lattice structure is a way enabled by additive manufacturing (AM) to realize part lightweight design. Conventional design optimization method based on homogeneous periodic lattice structure cannot achieve the optimal structure without taking the stress magnitude and orientation into account. A design optimization method of heterogeneous conformal lattice structure coupling constraints and anisotropy of AM based on principal stress lines (PSL) is proposed. (1) The PSL is calculated based on finite element analysis to visualize the path of load transfer. (2) The load-adapted lattice structure is generated guided by the PSL. (3) In order to further optimize lattice structure, the lattice structure optimization model is established by coupling anisotropy and constraints of AM based on the fused deposition modeling (FDM) experiment. Taking a cantilever beam as a case, the maximum force-to-weight ratio and the stiffness-to-weight ratio of the optimized heterogeneous conformal lattice structure are increased by 11.8% and 41.8% respectively compared with homogeneous conformal lattice structure, which verified the feasibility of the proposed method.

Ground-Wall Insulation Structure Optimization Method Research Based on Composite Material for Stator Bars

A ground-wall insulation structure optimization method based on a composite material for a stator bar is proposed in this paper. The trend regularity of the nonuniform coefficient of the electric field for a composite dielectric fillet electrode structure is established with the finite element method, and a composite ground-wall insulation structure optimization model is established in combination with the Nelder-Mead optimization algorithm. A composite ground-wall insulation optimization structure is acquired based on this model and material parameter, reducing the maximum electric field inside the ground-wall insulation and increasing the overall electrical performance of the stator bars with composite ground-wall insulation. The measured results are consistent with the predicted trend, which verifies the validity of the proposed method.

Study on optimization model of overseas oil and gas resources structure

Due to the unreasonable structure of overseas oil and gas resources and the lack of resource optimization strategy will seriously affect the stable development of oil companies in the global field, it is an urgent problem for oil companies to provide scientific and rapid optimization solutions to meet the resource allocation objectives for managers. In view of the lack of multi-stage dynamic planning model for overseas oil and gas resource structure optimization allocation, based on the operational research theory, according to the resource characteristics of overseas oil and gas projects, a dynamic programming mathematical model is established in this paper, which aims at the allocation of resource structure. It is solved by genetic algorithm, and realized by programming with MATLAB language. This study provides a new way and method for optimizing the allocation of overseas oil and gas resources.

РАЗРАБОТКА ЦИФРОВОЙ ОПТИМИЗАЦИОННОЙ БИЗНЕС-МОДЕЛИ СЕЛЬСКОХОЗЯЙСТВЕННОГО ПРЕДПРИЯТИЯ РЕГИОНА

The purpose of the study is to digitalize the business processes of a regional enterprise basing on the elaborated production structure optimization model and taking into account the influence of the enterprise development factors. Statistical and economic, monographic, abstract-logical, computational-constructive and modeling methods of economic research were used in preparation of the article. The current state of the region’s development shows the need and readiness for digitalization of the agricultural enterprises. When using economic and mathematical methods to develop a digital business model in research, the problem of optimizing the production structure to maximize the financial result under given production conditions was solved. The main variables of the model are the area under crops and the number of farm animals. The constraints express the conditions for the use of resources; they are imposed on most variables and are written as inequalities. The solution to the problem was carried out to maximize the financial result from the products sale. The resulting optimal combination of production sectors of the organization will improve the production and financial performance of the enterprise as a whole. As the object of the study the agricultural leader of the region and an exemplary diversified enterprise was chosen - JSC “PZ Meliorator”, Marksovsky district, Saratov region. The digital optimization business model, taking into account the influence of organizational development factors, is not static and can be applied to any economy in the region.

Multistep-Ahead Prediction of Urban Traffic Flow Using GaTS Model

The mathematical models for traffic flow have been widely investigated for a lot of application, like planning transportation and easing traffic pressure by using statistics and machine learning methods. However, there remains a lot of challenging problems for various reasons. In this research, we mainly focused on three issues: (a) the data of traffic flow are nonnegative, and hereby, finding a proper probability distribution is essential; (b) the complex stochastic property of the traffic flow leads to the nonstationary variance, i.e., heteroscedasticity; and (c) the multistep-ahead prediction of the traffic flow is often of poor performance. To this end, we developed a Gamma distribution-based time series (GaTS) model. First, we transformed the original traffic flow observations into nonnegative real-valued data by using the Box-Cox transformation. Then, by specifying the generalized linear model with the Gamma distribution, the mean and variance of the distribution are regressed by the past data and homochronous terms, respectively. A Bayesian information criterion is used to select the proper Box-Cox transformation coefficients and the optimal model structures. Finally, the proposed model is applied to the urban traffic flow data achieved from Dalian city in China. The results show that the proposed GaTS has an excellent prediction performance and can represent the nonstationary stochastic property well.

The Leland\u2013Toft optimal capital structure model under Poisson observations

This paper revisits the optimal capital structure model with endogenous bankruptcy, first studied by Leland (J. Finance 49:1213–1252, 1994) and Leland and Toft (J. Finance 51:987–1019, 1996). Unlike in the standard case where shareholders continuously observe the asset value and bankruptcy is executed instantaneously without delay, the information of the asset value is assumed to be updated periodically at the jump times of an independent Poisson process. Under a spectrally negative Lévy model, we obtain the optimal bankruptcy strategy and the corresponding capital structure. A series of numerical studies provide an analysis of the sensitivity, with respect to the observation frequency, of the optimal strategies, optimal leverage and credit spreads.

Hybrid genetic and shuffled frog‐leaping algorithm for neural network structure optimization and learning model to predict free spectrum in cognitive radio

SummaryCognitive radio, which is called an intelligent radio, will dynamically access the available spectrum. This mechanism will bring revolt in wireless communication, which lightens the spectrum utilization problem. Machine learning plays a vital role in every technology. Here, in cognitive radio, it helps train the model to predict the free spectrum. The cognitive clients utilize range‐detecting procedures to detect the groups previously transmitting on them to stay away from impact with the authorized clients that prompts deferral and moderates more vitality. To decrease postponement and vitality utilization and to foresee the future use of channels, range expectation procedures are utilized. Spectrum prediction is used to anticipate the future channel status in light of gathered chronicled information; here, to take care of the issue, the neural system‐based spectrum prediction utilizes a backpropagation training model that has been proposed. To enhance the structure of the neural system and to decrease the forceful weight auxiliary pattern, genetic algorithm (GA) along with the hybrid combination of shuffled frog‐leaping algorithm (SFLA) is proposed. Here, GA has been utilized to abstain from catching nearby ideal solutions. The selection, crossover, and mutation functions were performed to build the haphazardness, which stretches out the populace unite to the set that contains the global ideal solution. SFLA has been proposed for structure improvement; the paired structure has been recommended to demonstrate the memes with the motivation behind building up a subaccumulation with lesser measurements than that of the original group where recognizing affectability and precision would be versatile with that of the primary status. Simulation results show the GA‐SFLA‐based hybrid algorithm that is used has increased the results of getting the best weights by improving the system; additionally, the proposed conspire results show high forecast accuracy.

Quantification of water resource utilization efficiency as the main driver of plant diversity in the water-limited ecosystems

Identifying the drivers of species richness is a crucial for biodiversity conservation and community ecology. In this paper, the maximum utilization efficiency of water resources available to community is determined as the assembly law in an ecologically stable community with limited water resources. A plant community diversity model that incorporates water potential impedance is established, and the optimization model of community biodiversity structure was obtained based on water niche is also derived. It is shown that this optimization model can theoretically derive a species-area function similar to the traditional island biogeography curve, and it can also verify the predicted values of the species-area relationship experimentally and validated by long-term observations of plant diversity survey data. According to the model analyses, the total number of individuals (N) and the number of species (S) are determined by the combination of the water resources factors (the total quantity of available water resources in the growth season), the individual biological factors (the water consumption per unit mass of species in unit time), the complementary factors of inter- and intra-species water niche (the difference in water potential depending on the water use efficiency of the community), and plant community function factors (the total productivity of plant community). And as the area of plant communities increases, the heterogeneity of the water environment has increased. By adaptation and coevolution, more species are able to go through environmental and biological filtration, becoming community species that can utilize more available water resources at different spatial scales. Changes in water resources resulted in the community maximizing the utilization efficiency of water resources by adjusting the number and abundance of species. The main driving force for the increasing number of species is the full utilization of available water resources.

Probabilistic SVM classifier ensemble selection based on GMDH-type neural network

Support vector machine (SVM) provides a good classification and regression ability, especially, for small sample learning. However, in practice, the learning ability of implemented SVM is occasionally far from the expected level. Group method of data handling neural network (GMDH-NN) has been applied in various fields for pattern recognition and data mining. It makes it possible to automatically find interrelations in data, to select an optimal structure of network or model and to improve the accuracy of existing algorithms. In this work we propose to take the advantages of GMDH-NN for further increasing the classification performance of SVM. One weakness of the symmetric regularity criterion of GMDH-NN is that if one of the input attributes has a relatively big range, then it may overcome the other attributes. Thus, we first define a standardized symmetric regularity criterion (SSRC) to evaluate and select the candidate models, and optimize a classifier ensemble selection approach. Secondly, we define a novel structure of initial model of GMDH-NN which is from the posterior probability outputs of SVMs. These probabilistic outputs are generated from the improved Platt’s probabilistic outputs. Thirdly, in real classification tasks, different classifiers usually have different classification advantages. So we use probabilistic SVM as base learner and integrate the probabilistic SVMs with GMDH-NN, and then propose a special classifier ensemble selection approach for probabilistic SVM classifiers based on GMDH-NN called GMDH-PSVM. Moreover, we use the Borda sorting and Random weighted Borda sorting to discuss the results of our experiments. Experiments on standard UCI datasets demonstrate the effectiveness of our method.

Optimization algorithm of the surrogate model structure based on pigeon swarm search strategy

The hypersonic vehicle has problems with multiple systems coupling and multivariate design. Hence, it is difficult to determine the surrogate model structure of the hypersonic vehicle under larger samples. To solve this problem, this paper proposes a surrogate model structure optimization method based on the idea of PIO algorithm. The method can independently search for a polynomial surrogate model structure that satisfies the accuracy and prediction performance requirements. Through comparative analysis, utilizing this method outperforms Davidson’s genetic programming algorithm. Furthermore, it was applied to the global aerodynamic data fitting and control design of a hypersonic vehicle, and the simulation results reveal that the error of the prediction data obtained by this method was less than 5%, which has a good fitting effect and can meet the engineering application requirements.

Potential for reducing carbon emissions from urban traffic based on the carbon emission satisfaction: Case study in Shanghai

The transport sector has attracted much attention as one of the main sources of carbon emissions. In this study, we constructed an optimal urban traffic structure model based on the concept of carbon emission satisfaction in order to estimate the absolute carbon emission reductions that can be achieved, and to objectively analyze the relative difficulty of achieving the emissions reduction goal. By considering Shanghai as an example, we found that rail transit is the dominant mode of transportation and that the proportion of travel in private cars can be greatly reduced, but buses should be maintained at the current level, whereas the proportion of taxis may be reduced slightly. In the existing traffic environment in Shanghai, after optimizing the urban traffic structure, we found that 47.62% of the carbon emissions reduction target can be achieved. However, given the excessive attention paid to the satisfaction of individuals and the government but the lack of consideration of the ecological environment, the carbon emission satisfaction with respect to urban traffic is low at present in Shanghai. Improving the carbon emission satisfaction by reducing the satisfaction of other targets is difficult, and the potential for reducing carbon emissions is limited for transportation. Therefore, Shanghai can only achieve its carbon emission reduction targets by implementing resource allocation, transportation technology, and urban planning measures in order to improve the existing traffic conditions, thereby achieving the goal of reducing carbon emissions but without affecting the satisfaction of other targets.

Algorithms and software for optimal management of raw materials, fuel and energy resources in blast furnace production

The structure of optimization model of optimal management of raw materials, fuel and energy resources in the blast-furnace shop of iron and steel works is represented. The following blocks are taken as system basis: 1) calculation of the set of parameters that characterize the thermal, gas-dynamic, slag and blasting modes for every blast furnaces of the shop during the base period; 2) calculation of linearized model coefficients (constants of transferring via different exposure pathways) individually for every blast furnace as well as properties of iron ore raw materials, fluxing additions, blasting parameters, parameters of fuel-enriched blast influencing the technical-and-economic indices of separate furnaces performance, their thermal, gas-dynamic and slag operation modes in the course of blast-furnace melting according to UrFU-MMK blast-furnace production model within the base period; 3) solution of tasks that consider the optimal allocation of raw materials, fuel and energy resources for the project period of blast furnaces operation; 4) analysis of obtained results and providing of recommendations on the optimization of blast furnaces parameters. The developed functional model of optimal distribution of raw materials, fuel and energy resources for the engineering and technology personnel of blast-furnace shop is illustrated; the main functions and interconnections between the separate functional blocks are defined. The functions of created “Optimal management of raw materials, fuel and energy resources in the blast-furnace production” software that is realized in the Microsoft Visual Studio 2017 (C# programming language) programming environment in the form of web application are pointed out. The program product provides the engineering and technology personnel of blast furnace shop of iron and steel works with the opportunity to solve the tasks of optimal distribution of fuel and energy resources (natural gas and oxygen consumption) within the group of blast furnaces in the different technological situations.

Shallow Landslide Susceptibility Models Based on Artificial Neural Networks Considering the Factor Selection Method and Various Non-Linear Activation Functions

Landslide susceptibility mapping is well recognized as an essential element in supporting decision-making activities for preventing and mitigating landslide hazards as it provides information regarding locations where landslides are most likely to occur. The main purpose of this study is to produce a landslide susceptibility map of Mt. Umyeon in Korea using an artificial neural network (ANN) involving the factor selection method and various non-linear activation functions. A total of 151 historical landslide events and 20 predisposing factors consisting of Geographic Information System (GIS)-based morphological, hydrological, geological, and land cover datasets were constructed with a resolution of 5 x 5 m. The collected datasets were applied to information gain ratio analysis to confirm the predictive power and multicollinearity diagnosis to ensure the correlation of independence among the landslide predisposing factors. The best 11 predisposing factors that were selected in this study were randomly divided into a 70:30 ratio for training and validation datasets, which were used to produce ANN-based landslide susceptibility models. The ANN model used in this study had a multi-layer perceptron (MLP) structure consisting of an input layer, one hidden layer, and an output layer. In the output layer, the logistic sigmoid function was used to represent the result value within the range of 0 to 1, and six non-linear activation functions were used for the hidden layer. The performance of the landslide susceptibility models was evaluated using the receiver operating characteristic curve, Kappa index, and five statistical indices (sensitivity, specificity, accuracy, positive predictive value (PPV), negative predictive value (NPV)) with the training dataset. In addition, the landslide susceptibility models were validated using the aforementioned measures with the validation dataset and were compared using the Friedman test to check the significant differences among the six developed models. The optimal number of neurons was determined based on the aforementioned performance evaluation and validation results. Overall, the model with the best performance was the MLP model with the logistic sigmoid activation function in the output layer and the hyperbolic tangent sigmoid activation function with five neurons in the hidden layer. The validation results of the best model showed a sensitivity of 82.61%, specificity of 78.26%, accuracy of 80.43%, PPV of 79.17%, NPV of 81.82%, a Kappa index of 0.609, and AUC of 0.879. The results of this study highlight the effectiveness of selecting an optimal MLP model structure for shallow landslide susceptibility mapping using an appropriate predisposing factor section method.