Optimal Model Structure Research Articles

The global nutrition can be greatly improved with diet optimization

Malnutrition and nutritional imbalances resulting from a poor diet are common worldwide. Here, we present a diet structure optimization model aimed at improving the nutritional levels, while keeping the optimized diet affordable and acceptable. We optimized the current diet of 181 countries by incorporating ten major food groups. A simple diet adjustment could increase the nutritional levels by up to 11.9 % and significantly address the existing nutrient imbalances. The optimized diet is environmentally friendly, reducing the total land use and greenhouse gas emissions by 11.8 % and 8.3 %, respectively. Furthermore, compromising on some nutritional improvements could yield greater environmental benefits, which could result in 17.1 %, and 12.7 % reductions in total land use and greenhouse gas emissions, respectively. Moreover, a 28.4 % diet cost increase can lead to a maximum of 9.7 % greater nutritional improvement than the optimal increase of constant cost. We provide a global diet optimization strategy that improves both human nutrition and sustainability.

Development of an assessment-based planting structure optimization model for mitigating agricultural greenhouse gas emissions

Optimization of crop structure is an efficient way to reduce greenhouse gas (GHGs) from agriculture production. However, carbon footprint have rarely been incorporated into previous planting structure optimization models due to the challenges of assessing the spatial and temporal distribution of agricultural carbon footprint for multiple crops in irrigated districts. In addition, previous planting structure models suffered from strong subjectivity in objective function determination, and the obtained non-dominated solution set offered difficulties to decision-makers in selecting specific implementation options. To fill such gaps, an integrated accounting-assessment-optimization-decision making (AAODM) approach was proposed, which remedies the shortcomings of previous crop planting structure optimization models in carbon footprint mitigation, and overcomes the subjectivity of objective function determination and the difficulty in selecting specific implementation options. Firstly, life cycle assessment (LCA) method was used to account for the multi-year agricultural carbon footprints of multiple crops in the irrigation district. The optimization objective functions of planting structure optimization models can then be determined based on the assessment method of carbon footprint influencing factors. Next, the Non-dominated Sorting Genetic Algorithm-II (NSGA-II) was used to generate a non-dominated solution set of the optimization model. The optimal planting structure can be finally obtained based on decision making methods by determining the maximum harmonic mean (HM) and knee points (KPs) of the non-dominated solution set. The developed AAODM approach was then applied to a case study of agricultural crop management in Bayan Nur City, China. The results showed that the level of economic development was a key factor influencing the increase in carbon footprint in Bayan Nur City over the past 20 years. The regulation of the level of economic development would significantly influence the agricultural carbon footprint in Bayan Nur City. Moreover, two optimal crop cultivation patterns were provided for decision-makers by selecting solutions from the Pareto front with decision making methods. The comparison results with other methods showed that the solutions obtained by NSGA-II were superior to MOPSO in terms of carbon reduction. The developed AAODM approach for agricultural GHG mitigation could help agricultural production systems in achieving low carbon emissions and high efficiency.

Research on mechanical property prediction of hot rolled steel based on lightweight multi-branch convolutional neural network

Various process parameters and chemical compositions affect the crystal phase structure of hot-rolled steel. The real-time prediction of mechanical properties is a key problem in this field. Implementing accurate, effective, real-time, and lightweight prediction according to its technological process is of great significance. Therefore, to solve the problem of weak characterization ability of the original data, the Gramian Angular Field method is proposed to transform the original data into a two-dimensional matrix to enrich the correlation features between the data. At the same time, to solve the high real-time requirements of compact tasks and improve the accuracy of the model, a combination module of the multi-branch enhanced convolutional kernel and adaptive bottleneck layer is proposed, which can improve the performance of the model and balance the relationship between lightweight. Experiments in this paper show that compared with similar models proposed in other literature, the optimal hyperparameter structure model proposed in this paper has higher prediction accuracy and smaller model scale, and can achieve higher prediction performance in a wide range of applications.

A multi-objective optimization model of urban passenger transportation structure under low-carbon orientation considering participating subjects.

The optimization of urban passenger transport structure can effectively save energy and reduce the carbon emission of transport. However, the carbon emission and energy consumption generated by the construction of transport projects are worthy of attention. In this paper, a multi-objective optimization model of urban passenger transport structure oriented to low carbon is proposed considering passenger, operator, and construction. The optimal solution of the model is obtained based on the NSGA-II algorithm, and the validity of the model is verified with a case of Qingdao. The optimal ratio of Qingdao passenger traffic structure considering only the passenger perspective (PS), considering the passenger and operator (POS), and considering the three parties together (POCS) is obtained, respectively. The results show that the optimal structures obtained by the PS, POS, and POCS models increase the public transport passenger share by 12.74%, 20.74%, and 23.70%, compared with the actual values. From both the supply and demand sides, there has been an increase in the passenger share of public transport. The POS model is more suitable for solving structural optimization problems that do not involve construction carbon emissions in the short term. The POCS model is more suitable for long-term comprehensive structural optimization problems. The results of the study provide a reference basis for optimizing the urban passenger transport structure.

Selection of the Optimal Timber Harvest Based on Optimizing Stand Spatial Structure of Broadleaf Mixed Forests

There is increasing interest in optimizing stand structure through forest management. The forest structure influences growth and maintains the structure, promoting sustainability. Structure-based forest management (SBFM), which is based on the spatial relationships between a reference tree and its four nearest neighbors, considers the best spatial structure for the stand and promotes the development towards a healthy and stable state by selectively thinning specific trees. This management method is a scientific approach for sustainable forest management, and appropriate harvesting is the core principle of uneven-aged forest management. However, the application of this approach in the management of uneven-aged mixed stands is a challenge because their dynamics are more difficult to elucidate than those of planted or pure stands. This study presented a stand spatial structure optimization model with a transition matrix growth model for selecting suitable timber harvest during uneven-aged mixed-forest management optimization. The model was developed using three neighborhood-based structural indices (species mingling, diametric differentiation, and horizontal spatial pattern) and diameter diversity indices. The approach was applied to four broadleaf stands in the Maoershan Forest Farm of the Heilongjiang Province. The results demonstrate that optimizing the stand spatial structure with a transition matrix growth model improved the objective function values (F-index) by 23.8%, 12.8%, 14.6%, and 28.3%, and the optimal removal of trees from the stands ranged from 24.3% to 25.5%. The stand structure in the next cycle (after 5 years) was closer to the uneven-mixed state. The main conclusion of this study is that optimizing the stand spatial structure with a transition matrix growth model can improve the speed and accuracy of tree selection for harvesting in unevenly mixed forests, thus helping regulate stable and diverse forest growth.

Effect of strain on the electronic structure and optical properties of Cr-doped monolayer MoS2.

In this paper, the electronic and optical properties of Cr-doped monolayer MoS2 under uniaxial tensile strain are investigated by first-principle calculations. It is shown that uniaxial tensile strain can significantly change the electronic and optical properties of Cr-doped monolayer MoS2, and the bandgap value of the intrinsic MoS2 system gradually decreases with the increase of tensile strain, while the bandgap value of the Cr-doped MoS2 system is relatively stable. However, when the stretching reaches a certain degree, both the intrinsic and doped systems become metallic. From the analysis of the density of states, it is found that new electronic states and energy levels appear in the intrinsic MoS2 system and all Cr-doped monolayer MoS2 systems with the increase of the tensile strain, but the changes in the density of states diagrams of the Cr-doped monolayer MoS2 system are relatively small, which is mainly attributed to the effect of the Cr-doped atoms. The analysis of optical properties displays that the stretched doped system differs from the intrinsic MoS2 system in terms of dielectric function, absorption and reflection, energy loss function, and refractive index. Our results suggest that uniaxial tensile strain can be used as an effective means to modulate the electronic structure and optical properties of Cr-doped monolayer MoS2. These findings provide a theoretical basis for understanding the optoelectronic properties of MoS2 and its doped systems as well as their applications in optoelectronic devices. Based on the first principle of density functional theory framework and the CASTEP module in Materials Studio software (Perdew et al. in Phys Rev Lett 77(18):3865-3868, 1996). The structure of Cr atom-doped MoS2 systems and MoS2 systems were optimized using the generalized gradient approximation plane-wave pseudopotential method (GGA) and Perdew-Burke-Ernzerhof (PBE) generalized functions under 3%, 6%, and 9% tensile deformation, and the corresponding formation energy, bond length, electronic structure, and optical properties of the models were analyzed. The Grimme (J Comput Chem 27(15):1787-1799, 2006) vdW correction with 400 eV cutoff was used in Perdew-Burke-Ernzerhof (PBE) functional to optimize the geometry until the forces and energy converged to 0.02 eV/Å and 1.0e-5eV/atom, respectively. For each model structure optimization, the K-point grid was assumed to be 4×4×1, using the Monkhorst-Pack special K-point sampling method. After the MoS2 supercell convergence test, the plane-wave truncation energy was chosen to be 400 eV. Following geometric optimization, the iterative accuracy converged to no less than 1.0×10-5 eV/atom for total atomic energy and less than 0.02 eV/Å for all atomic forces. We created a vacuum layer of 18 Å along the Z-axis to prevent the impact of periodic boundary conditions and weak van der Waals forces between layers on the monolayer MoS2. In this paper, a total of 27 atoms were used for the 3×3×1 supercell MoS2 system, which consists of 18 S atoms and 9 Mo atoms.

A Study on Improving Musical Expression through Optimization of Artificial Neural Network Model Structure

A Study on Improving Musical Expression through Optimization of Artificial Neural Network Model Structure

The mediating roles of coping styles and academic burnout in the relationship between stressors and depressive symptoms among Chinese postgraduates.

Since few studies have incorporated factors like stressors, coping styles, and academic burnout into the same model to analyze their impacts on depressive symptoms, this research attempts to establish an optimal structural model to explore the direct and indirect effects of these factors on depressive symptoms. A total of 266 postgraduates completed questionnaires regarding coping styles, academic burnout, stressors, and depressive symptoms. The path analysis was applied for investigating the roles of coping styles and academic burnout in mediating the relationship between stressors and depressive symptoms. The total and direct effects of stressors on depressive symptoms were 0.53 and 0.31, respectively. The proportion of the direct effect of stressors on depressive symptoms to its total effect amounted to 58.50%. The indirect effects of academic burnout, positive coping style, and negative coping style on the association between stressors and depressive symptoms were 0.11, 0.04, and 0.03, taking up 20.75%, 7.55%, and 5.66% of the total effect, respectively. The serial indirect effect of positive coping style and academic burnout was 0.02, accounting for 3.77% of the total effect, while that of negative coping style and academic burnout was 0.02, taking up 3.77% of the total effect. Coping styles and academic burnout chain jointly mediate the relationship between stressors and depressive symptoms among postgraduates. Thus, encouraging postgraduates to tackle stress positively may reduce the likelihood of the development of academic burnout and further reduce depressive symptoms.

Multi-objective optimization decision-making of an underwater vehicle rotary docking skirt structure

This study determines the design parameters, analysis methods, and mesh size of a finite-element analysis model in the structural design of a rotary docking skirt (RDS) structure for an underwater vehicle. The impact of the structural design variables on the response is then studied. The sample points determined using the optimal Latin hypercube method are analyzed and fitted with the Kriging model, response surface model, and radial basis function. After comparing the fitting accuracies of the three approximation models, an approximation model of the RDS structure is determined. Next, several typical multi-objective optimization algorithms are tested and the improved nondominated sorting genetic algorithm II (INSGA-II) is selected for establishing an optimization model of the RDS structure. INSGA-II obtains the Pareto solution set of multi-objective optimization. The weights are obtained by combining the subjective and objective weights determined through the subjective-weighting analytic hierarchy process and the objective-weighting entropy weight method, respectively. Based on the combination weight, the technique for order preference by similarity to the ideal solution method is selected for multi-objective optimization decision-making. The final decision-making scheme of the rotary docking skirt structure improved the optimization effect compared to the initial scheme.

An Intelligent Approach for Predicting Mechanical Properties of High-Volume Fly Ash (HVFA) Concrete

Plastic waste (PW) is a major soild waste, which its generation continues to increase globally year in and year out. Proper management of the PW is still a challenge due to its non-biodegradable nature. One of the most convenient ways of managing plastic waste is by using it in concrete as a partial substitute for natural aggregate. However, the main shortcomings of adding plastic waste to concrete are a reduction in strength and durability. Hence, to reduce the undesirable impact of the PW in concrete, highly reactive additives are normally added. In this research, 240 experimental datasets were used to train an artificial neural network (ANN) model using Levenberg Marquadt algorithms for the prediction of the mechanical properties and durability of high-volume fly ash (HVFA) concrete containing fly ash and PW as partial substitutes for cement and coarse aggregate, respectively, and graphene nanoplatlets (GNP) as additives to cementitious materials. The optimized model structure has five input parameters, 17 hidden neurons, and one output layer for each of the physical parameters. The results were analyzed graphically and statistically. The obtained results revealed that the generated network model can forecast with deviations less than 0.48%. The efficiency of the ANN model in predicting concrete properties was compared with that of the SVR (support vector regression) and SWLR (stepwise regression) models. The ANN outperformed SVR and SWLR for all the models by up to 6% and 74% for SVR and SWLR, respectively, in the confirmation stage. The graphical analysis of the results further demonstrates the higher prediction ability of the ANN. Doi: 10.28991/CEJ-2023-09-09-04 Full Text: PDF

An improved equilibrium optimizer for numerical optimization: A case study on engineering design of the shell and tube heat exchanger

The accurate design of the shell-and-tube heat exchanger (STHE) model is crucially important for industries such as process industries, oil refining, thermic devices, and power plants. However, determining its parameters is a challenging task due to the multimodality and nonlinearity of the cost function. Due to such nature, performing algorithms face the challenges of sluggish convergence speed and the proneness to local optima, making it difficult to reach satisfactory solutions. In line with the above challenges, this paper aims to contribute by developing an improved optimization algorithm that can effectively deal with such optimization issues. The proposed algorithm, named Levy-Opposition- Equilibrium Optimizer (LOEO), combines the Equilibrium Optimizer (EO) and the strategies of opposition learning (OL) and Levy flight learning (LFL) to overcome the dilemma of early convergence and getting trapped in local optima. It does this by employing the OL and LFL strategies after performing the EO phase to explore the search space in opposite directions and exploit the search space by means of the Levy distribution pattern, respectively. The proposed LOEO approach aims to enhance the algorithm's ability (provides enhanced exploration capability, effective search space exploitation, and enhanced convergence speed) and reach higher quality solutions to optimization problems. To assess the performance of the suggested LOEO, three stages of analysis are carried out. In the first stage, the performance of the proposed algorithm is investigated on some benchmark problems, including CEC 2005 and CEC 2020, using statistical verifications, convergence behavior, and the boxplot paradigm. The second stage involves the non-parametric Friedman test to assess the significance of results, where the results indicate that the LOEO outperforms the best existing one (equilibrium whale optimization algorithm (EWOA)) by an average rank of the Friedman test greater than 28% for CEC 2005 benchmark functions while outperforming enhanced EO (EEO) by 24% for CEC 2020. Finally, the LOEO's applicability is realized through determining the optimal design structure of the STHE model. The reported results indicate that the LOEO algorithm offers accurate performance compared to competing algorithms as it saves the cost of the device by 39.81% compared to the best existing results, and thus, it is commended to adopt for new applications.

Multi-step-ahead and interval carbon price forecasting using transformer-based hybrid model.

Accurate and stable carbon price forecasts serve as a reference for assessing the stability of the carbon market and play a vital role in enhancing investment and operational decisions. However, realizing this goal is still a significant challenge, and researchers usually ignore multi-step-ahead and interval forecasting due to the non-linear and non-stationary characteristics of carbon price series and its complex fluctuation features. In this study, a novel hybrid model for accurately predicting carbon prices is proposed. The proposed model combines multi-step-ahead and interval carbon price forecasting based on the Hampel identifier (HI), time-varying filtering-based empirical mode decomposition (TVFEMD), and transformer model. First, HI identifies and corrects outliers in carbon price. Second, TVFEMD decomposes carbon price into several intrinsic mode functions (imfs) to reduce the non-linear and non-stationarity of carbon price to obtain more regular features in series. Next, these imfs are reconstructed by sample entropy (SE). Subsequently, the orthogonal array tuning method is used to optimize the transformer model's hyperparameters to obtain the optimal model structure. Finally, after hyperparameter optimization and quantile loss function, the transformer is used to perform multi-step-ahead and interval forecasting on each part of the reconstruction, and the final prediction result is obtained by summing them up. Five pilot carbon trading markets in China were selected as experimental objects to verify the proposed model's prediction performance. Various benchmark models and evaluation indicators were selected for comparison and analysis. Experimental results show that the proposed HI-TVFEMD-transformer hybrid model achieves an average MAE of 0.6546, 1.3992, 1.6287, and 2.2601 for one-step, three-step, five-step, and ten-step-ahead forecasting, respectively, which significantly outperforms other models. Furthermore, interval forecasts almost always have a PICI above 0.95 at a confidence interval of 0.1, thereby indicating the effectiveness of the hybrid model in describing the uncertainty in the forecasts. Therefore, the proposed hybrid model is a reliable carbon price forecasting tool that can provide a dependable reference for policymakers and investors.

Experimental analysis on the structural seismic behavior of steel frame-precast steel reinforced concrete (SRC) infill wall with lateral force resisting

The structural seismic performance of steel frame-precast steel reinforced concrete (SRC) infill wall with lateral force resisting is analyzed, and the structural strength of steel frame-precast SRC infill wall with lateral force resisting is improved. The structural seismic performance optimization model of SRC lateral force resisting wall based on buckling restrained brace is proposed. Through the finite element simulation software, the seismic performance and response results of ordinary steel frames, buckling restrained braced steel frames and a relatively new type of sacrificial-energy dissipation braced steel frames under earthquake are compared and analyzed to demonstrate the applicability and performance advantages of sacrificial-energy dissipation braced steel frames in the steel frame braced structure system. Under the action of horizontal earthquake, the supporting members experience reciprocating axial tension and compression cycles, which dissipate a large amount of seismic energy input into the structure. Therefore, the buckling restraint support method can be used in the structure to improve the support strength. Under horizontal reciprocating load action of earthquake, the ability to consume seismic energy through self-hysteresis of the brace is poor. Experimental research shows that, the unbalanced force formed in the beam of the frame beam under seismic action will form a plastic hinge at the beam end at both ends of the frame beam. Especially when the brace is buckling unstable and the stiffness of the frame beam is small, the plastic hinge effect at the beam end is significant. This phenomenon may cause damage to the frame beam or even local floor subsidence. The buckling restraint support has a full hysteresis area under axial tension and compression, and its mechanical performance is excellent. It is obviously superior to ordinary steel bracing in energy dissipation capacity and seismic performance. It can accurately predict the bearing capacity of reinforced concrete under strong earthquake, and the energy dissipation distribution is more in line with the requirements of “energy seismic design method”.

Stock Prediction using LSTM model

With the development of the times, investors are increasingly in demand for stock price forecasting. However, stock price fluctuations are full of uncertainty, making traditional machine learning algorithms more erroneous in long-term forecasting. Based on the LSTM model, this paper uses Tushare to obtain the historical price of stocks, and the optimal structure and best training parameters of the LSTM model in stock price prediction are determined experimentally. The prediction accuracy of the LSTM model was evaluated by MAE, and the best result was 69.15, which achieved accurate prediction of stock prices. Compared with the traditional SVR model and the ARMA model, the prediction results of LSTM are more in line with the actual value, and the prediction accuracy of the algorithm is higher.

Research on Optimization Model of Heat Exchange Fin Structure in Energy Storage System

Aiming at the thermodynamic problem of heat transfer fin in heat storage system, the solidification and melting model and thermodynamic conventional model are established by using simulation software, thermodynamic image processing and other methods. MATLABAB, COMSOL and other software are used to program the optimal rectangular fin spacing Angle and the optimal heat exchange fin structure. Firstly, based on the reason that the simulation software can simulate the heat transfer process of phase change material (PCM) in the actual process to the greatest extent, the plane modeling and thermodynamic equation method are used to analyze the distance Angle of the rectangular fin, and the melting solidification model is established, and the optimal solution is obtained. Finally, combined with the analytic hierarchy model and TOPSIS method, the triangular fin structure and rectangular fin structure were compared from three levels, namely, the heat sink area, temperature rise time and PCM maintenance time of different shapes. And create a heat transfer rate of return to more favorably analyze the data and visually present the comparative results.

Fuzzy Model Parameter and Structure Optimization Using Analytic, Numerical and Heuristic Approaches

Fuzzy systems are widely used in most fields of science and engineering, mainly because the models they produce are robust, accurate, easy to evaluate and capture real-world uncertainty better than do the classical alternatives. We propose a new methodology for structure and parameter tuning of Takagi–Sugeno–Kang fuzzy models using several optimization techniques. The output parameters are determined analytically, by finding the minimum of the root-mean-square error (RMSE) for a properly defined error function. The membership functions are simplified by considering symmetry and equispacing, to reduce the optimization problem of finding their parameters, and allow it to be carried out using the numerical method of gradient descent. Both algorithms are fast enough to finally implement a strategy based on the hill climbing approach to finding the optimal structure (number and type of membership functions) of the fuzzy system. The effectiveness of the proposed strategy is shown by comparing its performance, using four case studies found in current relevant works, to the popular adaptive network-based fuzzy inference system (ANFIS), and to other recently published strategies based on evolutionary fuzzy models. In terms of the RMSE, performance was at least 28% better in all case studies.

DEVELOPMENT OF OPTIMAL CROP PRODUCTION MODEL CONSIDERING EXISTING NATURAL-CLIMATIC RISKS INCREASING CROP YIELDS

Contemporary agriculture is a fertile ground for the effective use of economic and mathematical models, which can be evaluated to unwind several problems with characteristic optimization features: multiple solution opportunities and freedom of choice, limited production resources, and efficiency valuation. The presented study aims to develop a model of optimal crop production structure under the existing weather risks in the agricultural management system. The article reviews the basic theoretical concepts in optimizing the production structure of agricultural enterprises, examines the specific features of crop production, proves the influence of weather and climate conditions on forming the production structure, demonstrates the use of correlation and regression analysis for trend modeling and forecasting of crop yields, and offers suggestions for determining the optimal production structure. The study concludes that multivariate forecasting helps optimize management in economic organizations and ensure their development under the variability of natural conditions. Natural and climatic conditions significantly impact the development of production structures in agricultural enterprises, along with nature acting as an innate participant in the game.

Numerical Simulation of a Simplified Reaction Model for the Growth of Graphene via Chemical Vapor Deposition in Vertical Rotating Disk Reactor

The process of graphene growth by CVD involves a series of complex gas-phase surface chemical reactions, which generally go through three processes, including gas phase decomposition, surface chemical reaction, and gas phase diffusion. The complexity of the CVD process for growing graphene is that it involves not only chemical reactions but also mass, momentum, and energy transfer. To solve these problems, the method of numerical simulation combined with the reactor structure optimization model provides a good tool for industrial production and theoretical research to explore the influencing factors of the CVD growth of graphene. The objective of this study was to establish a simplified reaction model for the growth of graphene by chemical vapor deposition(CVD) in a vertical rotating disk reactor (VRD). From a macroscopic modeling perspective, computational fluid dynamics (CFD) was used to investigate the conditions for the growth of graphene by chemical vapor deposition in a high-speed rotating vertical disk reactor on a copper substrate surface at atmospheric pressure (101,325 Pa). The effects of gas temperature, air inlet velocity, base rotation speed, and material ratio on the surface deposition rate of graphene in a VRD reactor were studied, and the technological conditions for the preparation of graphene via the CVD method in a VRD reactor based on a special structure were explored. Compared with existing models, the numerical results showed the following: the ideal growth conditions of graphene prepared using a CVD method in a VRD reactor involve a growth temperature of 1310 K, an intake speed of 470 mL/min, a base speed of 300 rpm, and an H2 flow rate of 75 sccm; thus, more uniform graphene with a better surface density and higher quality can be obtained. The effect of the carbon surface deposition rate on the growth behavior of graphene was studied using molecular dynamics (MD) from a microscopic perspective. The simulation showed that the graphene surface deposition rate could control the nucleation density of graphene. The combination of macro- and microsimulation methods was used to provide a theoretical reference for the production of graphene.

Do political connections cause over-indebtedness? Empirical evidence from China’s listed construction firms

PurposeThis study aims to investigate the relationship between political connections and the over-indebtedness of firms in the construction industry. Furthermore, this study explores the moderating effect of corporate governance mechanisms with monitoring intent on this relationship.Design/methodology/approachThis study uses the data from China’s listed construction firms for the years 2010–2019 to run the fixed-effect regression. This study constructs the optimal capital structure mathematical model by following the trade-off approach.FindingsThe research results show that most of China’s listed construction firms are surprisingly over-indebted in the long run. This study affirms that political connections positively impact the over-indebtedness of China’s listed construction firms. However, corporate governance can alleviate the impact of political connections on the over-indebtedness of China’s listed construction firms.Originality/valueThere were limited studies to discuss the relationship between political connections and the over-indebtedness of construction firms, and no particular attention has been given to the moderating effect of corporate governance mechanisms on the relationship between political connections and over-indebtedness. Moreover, in calculating the over-indebtedness of China’s listed construction firms, this study considers the financial characteristics of China’s construction firms when building the mathematical model of optimal capital structure, which makes the calculation results of over-indebtedness closer to reality.

A method for constructing non-elementary Cobb – Douglas production functions

The actual scientific problem in machine learning is the development of new interpretable mathematical models, as well as methods and programs for their construction. Many well-known regression models have good interpretative properties, for example, linear and power models (Cobb – Douglas production functions). Previously, the author developed non-elementary linear regressions, the problem of constructing which was reduced to the mixed integer 0–1 linear programming problem.Based on non-elementary linear regressions, in this paper, for the first time, non-elementary Cobb-Douglas production functions are proposed, which include not only explanatory variables in degrees, but also all their possible pair combinations, transformed using binary operations min and max. The proposed models are linearized, which makes it possible to apply for their construction the mixed integer 0–1 linear programming problem formulated in the same way as for non-elementary linear regressions. As a result of its solution, the optimal structure model is automatically determined. The advantage of such a formulation is that the solution of the problem can be obtained faster than using enumeration procedures, and also that the signs of the estimates of the constructed model are guaranteed to be consistent with the meaningful meaning of the factors. At the same time, it is possible to control the requirements for the structure of the model using linear constraints on binary variables. In particular, the problem can be used to select the optimal structures for traditional elementary Cobb – Douglas production functions.The problem of modeling the gross regional product of the Tomsk region is solved. The following variables were chosen as explanatory variables: average per capita cash income of the population, investments in fixed capital, costs of innovative activities of organizations, average annual number of employees, cost of fixed assets, internal costs for research and development. The LPSolve package was chosen as the solver of the mixed integer 0–1 linear programming problem. As a result of solving this problem, the optimal structure of the non-elementary Cobb – Douglas production function was chosen, which contains all six explanatory variables in three regressors. The coefficient of determination for the constructed model turned out to be 0.997. All regression coefficients turned out to be significant according to Student's t-test, and their signs satisfy the meaningful meaning of the factors. An interpretation for the constructed model is given.