Neural Network Prediction Model Research Articles

Recovery of iron from copper slag using jatropha oil as a green reductant: Investigation of pyrolysis analysis, reaction kinetic and artificial neural network model prediction

Jatropha oil (JO) is a non-edible vegetable oil and a potential renewable energy source. In this paper, JO was proposed as a reductant to recover iron from copper slag (CS), and the effects of JO dosage, reduction temperature and reduction time on the recovery of iron by deep reduction of CS were investigated. Based on Coats-Redfern method, a thermogravimetric analyzer was used to explore the reactivity, mass loss behavior, and kinetic mechanism of CS reduction by JO, and to compare the reduction characteristics of anthracite and JO char when used as a reductant. Additionally, artificial neural network (ANN) modeling was used to predict the effects of reductant type, temperature, and heating rate on mass loss during CS reduction. The results showed that the optimum condition for CS reduction when JO was used as the reductant was 3 wt% JO at 1250 °C for 45 min, which resulted in 84.97 % Fe recovery and 94.58 % Fe grade in CS; the reduction mechanism of CS followed a one-dimensional diffusion model (D1); and ANN5 (15*1) was the optimum ANN model to predict the reduction of CS. This research introduces innovative strategies for the environmentally friendly and efficient recovery of iron from iron-containing furnace slag.

Prediction of Buildings’ Settlements Induced by Deep Foundation Pit Construction Based on LSTM-RA-ANN

In view of the shortcomings of existing methods for predicting the settlement of surrounding buildings caused by deep foundation pit construction, this study uses the monitoring data of a foundation pit project in Shanghai and divides the construction process of the pit into three working conditions, that is, enclosure construction, earthwork excavation, and basement support construction. The attention mechanism and residual update are integrated into the artificial neural network (ANN) model, and the root-mean-square error, average absolute error, and determination coefficient are used as the evaluation indices of the model. The artificial neural network prediction model LSTM-RA-ANN for building settlements in deep foundation pit construction was then established. The prediction performance of the model was also analysed under different working conditions, and the influences of the main factors (including the soil parameter, monitoring point location, activation function, hyperparameter, and input number) on the evaluation index was further explored. The results indicate that the performances of the established LSTM-RA-ANN model are closely related to the construction conditions, the predicted settlements agree well with the monitored ones in three working conditions with the greatest errors occurring at a later time of the working conditions, and the prediction accuracy of the great–small order corresponds to basement support, enclosure construction, and earthwork excavation respectively. The farther the monitoring point is from the edge of the pit, the better the model performance is. The activation function, initial learning rate, and maximum iteration batch have a great influence on the evaluation indices of the model, while the number of input points has little effect on the evaluation indices. These results may serve as a reference for the safe construction and normal operation of foundation pit engineering.

Identifying the risk of depression in a large sample of adolescents: An artificial neural network based on random forest.

This study aims to develop an artificial neural network (ANN) prediction model incorporating random forest (RF) screening ability for predicting the risk of depression in adolescents and identifies key risk factors to provide a new approach for primary care screening of depression among adolescents. The data were from a large cross-sectional study conducted in China from July to September 2021, enrolling 8635 adolescents aged 10-17 with their parents. We used the Patient health questionnaire (PHQ-9) to rate adolescent depression symptoms, using scales and single-item questions to collect demographic information and other variables. Initial model variables screening used the RF importance assessment, followed by building prediction model using the screened variables through the ANN. The rate of depression symptoms in adolescents was 24.6%, and the depression risk prediction model was built based on 70% of the training set and 30% of the test set. Ten variables were included in the final prediction model with a model accuracy of 85.03%, AUC of 0.892, specificity of 89.79%, and sensitivity of 70.81%. The top 10 significant factors of depression risk were adolescent rumination, adolescent self-esteem, adolescent mobile phone addiction, peer victimization, care in parenting styles, overprotection in parenting styles, academic pressure, conflict in parent-child relationship, parental rumination, and relationship between parents. The ANN model based on the RF effectively identifies depression risk in adolescents and provides a methodological reference for large-scale primary screening. Cross-sectional studies and single-item scales limit further improvements in model accuracy.

Determining the optimum layer combination for cross-laminated timber panels according to timber strength classes using Artificial Neural Networks

The primary aim of this work was to determine the effects of production parameters, such as wood species and timber strength classes, on some mechanical properties of cross-laminated timber (CLT) panels using artificial neural network (ANN) prediction models. Subsequently, using the models obtained from the analyses, the goal was to identify the optimum layer combinations of timber strength classes used in the middle and outer layers that would provide the highest mechanical properties for CLT panels. CLT panels made from spruce and alder timbers, as well as hybrid panels created from combinations of these two wood species, were produced. The strength classes of the timbers were determined non-destructively according to the TS EN 338 (2016) standard using an acoustic testing device. The bending strength and modulus of elasticity values of the CLT panels were determined destructively according to the TS EN 408 (2019) standard. According to ANN results, the optimum timber strength classes and layer combinations were determined for bending strength as C24-C27-C24 for spruce CLT, D18-D24-D18 for alder CLT, C30-D40-C30 and D18-C30-D18 for hybrid panels; and for modulus of elasticity, C22-C27-C22 for spruce, D35-D30-D35 for alder, C16-D24-C16, and D24-C24-D24 for hybrid panels.

A method to assist designers in optimizing the exterior styling of vehicles based on key features

Aiming to assist the design and management personnel of automotive products to make correct judgments and give optimization solutions for the design of exterior styling, and to solve the defects of the traditional method which relies too much on experience and personal ability, a neural network-based key feature analysis and optimization method of styling is proposed. Firstly, the user attributes are analyzed and quantified, and the weighted optimization of user evaluation data is completed based on the multidimensional Gaussian distribution function to construct a more objective car styling evaluation dataset. Secondly, a particle swarm algorithm is proposed to optimize the prediction model of radial basis function neural network (PSO-RBFNN) which deeply analyzes the mapping relationship between different car perspectives and users’ perceptual preferences. The identification of styling components is completed with the annotation of feature points, and a parametric expression method of styling feature data with multi-viewpoint weight optimization is proposed. Third, the model interpretation is based on the embedded feature selection method to discover the key feature points affecting user perception, and the key common features of styling are obtained through the coupling relationship analysis of feature points. Finally, the optimal distribution range of key features is given based on statistical analysis, and the application process is shown by an example of optimization to verify the feasibility of the theory. Compared with the traditional methods, PSO-RBFNN has better prediction accuracy. Meanwhile, the modeling optimization method based on key features has the great advantages of objective accuracy and high efficiency compared with the traditional theory.

Interplay of roughness and wettability in microchannel fluid flows—Elucidating hydrodynamic details assisted by deep learning

Under microconfinement, the complex interaction between surface roughness and fluid slippage yields unexpected variations in friction factor and drag reduction. These variations arise from the combined effects of roughness and hydrophobic interactions of the surface with the hydrodynamic field. Our study investigates alterations in frictional characteristics within long microchannels, considering fluid slippage, hydraulic diameter, and roughness. This exploration holds promise for precise drag reduction control applications for lab-on-a-chip and small-scale devices. To address computational limitations in analyzing diverse hydrodynamic conditions, we employ an artificial neural network prediction model, validated with experimental and numerical results. Contrary to the macroscopic conclusions obtained from the Moody chart, our findings indicate that fluid slippage, apart from surface roughness, significantly influences the friction factor. The interdependencies of friction factor on the flow and fluid parameters are thoroughly studied toward the proposition of a new slip-modified constricted flow friction factor formula, predicting friction in microchannels with combined roughness and hydrophobicity effects. This combined numerical and machine-learning approach presents a noteworthy counterpart to the moody chart at microscales offering the potential for a unified continuum-based description to include interfacial effects.

Machine intelligence accelerated design of conductive MXene aerogels with programmable properties

Designing ultralight conductive aerogels with tailored electrical and mechanical properties is critical for various applications. Conventional approaches rely on iterative, time-consuming experiments across a vast parameter space. Herein, an integrated workflow is developed to combine collaborative robotics with machine learning to accelerate the design of conductive aerogels with programmable properties. An automated pipetting robot is operated to prepare 264 mixtures of Ti3C2Tx MXene, cellulose, gelatin, and glutaraldehyde at different ratios/loadings. After freeze-drying, the aerogels’ structural integrity is evaluated to train a support vector machine classifier. Through 8 active learning cycles with data augmentation, 162 unique conductive aerogels are fabricated/characterized via robotics-automated platforms, enabling the construction of an artificial neural network prediction model. The prediction model conducts two-way design tasks: (1) predicting the aerogels’ physicochemical properties from fabrication parameters and (2) automating the inverse design of aerogels for specific property requirements. The combined use of model interpretation and finite element simulations validates a pronounced correlation between aerogel density and compressive strength. The model-suggested aerogels with high conductivity, customized strength, and pressure insensitivity allow for compression-stable Joule heating for wearable thermal management.

Multi-task aquatic toxicity prediction model based on multi-level features fusion

IntroductionWith the escalating menace of organic compounds in environmental pollution imperiling the survival of aquatic organisms, the investigation of organic compound toxicity across diverse aquatic species assumes paramount significance for environmental protection. Understanding how different species respond to these compounds helps assess the potential ecological impact of pollution on aquatic ecosystems as a whole. Compared with traditional experimental methods, deep learning methods have higher accuracy in predicting aquatic toxicity, faster data processing speed and better generalization ability. ObjectivesThis article presents ATFPGT-multi, an advanced multi-task deep neural network prediction model for organic toxicity. MethodsThe model integrates molecular fingerprints and molecule graphs to characterize molecules, enabling the simultaneous prediction of acute toxicity for the same organic compound across four distinct fish species. Furthermore, to validate the advantages of multi-task learning, we independently construct prediction models, named ATFPGT-single, for each fish species. We employ cross-validation in our experiments to assess the performance and generalization ability of ATFPGT-multi. ResultsThe experimental results indicate, first, that ATFPGT-multi outperforms ATFPGT-single on four fish datasets with AUC improvements of 9.8%, 4%, 4.8%, and 8.2%, respectively, demonstrating the superiority of multi-task learning over single-task learning. Furthermore, in comparison with previous algorithms, ATFPGT-multi outperforms comparative methods, emphasizing that our approach exhibits higher accuracy and reliability in predicting aquatic toxicity. Moreover, ATFPGT-multi utilizes attention scores to identify molecular fragments associated with fish toxicity in organic molecules, as demonstrated by two organic molecule examples in the main text, demonstrating the interpretability of ATFPGT-multi. ConclusionIn summary, ATFPGT-multi provides important support and reference for the further development of aquatic toxicity assessment. All of codes and datasets are freely available online at https://github.com/zhaoqi106/ATFPGT-multi.

Performance prediction of IS centrifugal pump based on improved BP neural network

Abstract The improved BP neural network was applied to establish prediction models for rated flow and rated head separately, the neural network structures were 8-11-1 and 5-6-1 respectively. 123 sets of pump experimental data were organized as training samples, and the other 15 sets of data were used as test samples. The results show that the training of the two networks converges quickly, the predictive values of the two performance parameters have good consistency with the experimental values, and the average testing errors of rated flow and rated head are 4.2% and 4.7%, respectively. The neural network prediction models established are accurate and effective, meeting the requirements of engineering applications.

Assessment of salmon sashimi processing conditions for Listeria monocytogenes cross-contamination and effectiveness of CLPSO-BP neural network model constructed for predicting microbial transfer

During preparation stages of ready-to-eat (RTE) foods, they can be cross-contaminated with Listeria monocytogenes which could result to outbreaks of foodborne listeriosis. In this study, the effects of gloves, cutting boards and cleaning methods on the transmission of L. monocytogenes during the processing of salmon sashimi were investigated. Subsequently, an improved particle swarm algorithm optimized BP neural network prediction model (CLPSO-BP) was developed based on Lévy flight and chaotic mapping. After processing contaminated salmons with gloves, we found that the surfaces of the gloves contained a significant amount of L. monocytogenes (5.24 log CFU/g), and 10.69% of the bacteria were subsequently transferred to uninfected salmon after being processed with the same gloves. Contamination of the processing tools with L. monocytogenes was reduced by up to 3 log CFU/g and the transfer rate was reduced by up to 50% after cleaning with room temperature water and detergent. In contrast, no bacteria were found after wiping the surfaces of the tools 70 °C hot water and detergent. The mean squared error (MSE) for the CLPSO-BP modeling tests was 0.1003, indicating accurate predictions. Overall, choosing an appropriate cleaning method and changing gloves regularly can help to avoid cross-contamination when processing RTE foods.

A new IPCA-CPSO-BP Model for Predicting Gas Emission in Underground Coal Mines

Gas has become the primary factor restricting the safe and efficient production of coal mines. Gas emission prediction model is very important for mine gas emission prediction and gas disaster prevention. The improved principal component analysis (IPCA) was used to reduce the dimension of 13 influencing factors of gas emission, and the CPSO-BP neural network prediction model was built with MATLAB software to predict the absolute gas emission of Zhongtai Mining.The results show that the cumulative contribution rate of principal components is not less than 95%, and the number of principal components after improvement is reduced from the original 6 to 3, which effectively solves the problem of excessive principal components and data redundancy caused by the correlation difference between various influencing factors and gas emission amount. At the same time, the optimized particle swarm optimization algorithm alleviates the trouble that the particle swarm optimization algorithm is prone to fall into the local optimal solution. By coupling the improved particle swarm optimization algorithm with BP neural network, the problem of BP neural network's over-dependence on the initial value of weights and thresholds is solved, and the optimal initial weights and thresholds are provided for BP neural network, improving the accuracy of model prediction results. The average relative error of the original BP neural network is 1.6638%, the normalized mean square error is 0.3155, and the regression correlation index is 0.8157. The IPCA-CPSO-BP prediction model decreased to 0.5749%, 0.1143 and 0.9758, respectively. IPCA improves the data dimensionality reduction ability of principal component analysis, IPCA-CPSO-BP enhances the stability of particle swarm optimization algorithm, and significantly improves the prediction accuracy of BP model. The prediction trend is highly consistent with the actual value, which verifies the reliability of the model and provides strong data support for mine gas prevention and control.

Tension identification of bridge suspension rods based on multiple impedance parameters from self-inductive coils

ABSTRACT This study introduces a method for identifying tension in suspension rods based on impedance from self-inductive coils, grounded in electromagnetic induction and magnetic elasticity principles. The theoretical feasibility of this method was analysed through a high-frequency inductive coil model. Steel strand tensile tests, utilising self-inductive coil sensors at various frequencies, explored the relationship between resistance, inductance, and capacitance with tension. According to the experimental results, the sensitivity coefficient is used to determine the optimal prediction frequency of each impedance parameter. Within the optimal frequency range, a single impedance parameter regression prediction method based on the impedance difference ratio index was proposed. To improve the accuracy and stability of the predictions, a GA-BP neural network prediction model based on the fusion of multiple impedance parameters was proposed. The results indicate that the GA-BP neural network prediction method, which integrates multiple impedance parameters, achieves higher accuracy in identifying cable forces than the single impedance parameter regression prediction method. Specifically, at a frequency of 5 kHz, the relative error of the GA-BP neural network prediction method, which integrates multiple impedance parameters, was only 3.74%. This research offers a novel approach for future in-service tension identification in bridge suspension rods.

Interpretable deep learning insights: Unveiling the role of 1 Gy volume on lymphopenia after radiotherapy in breast cancer

BackgroundLymphopenia is known for its significance on poor survivals in breast cancer patients. Considering full dosimetric data, this study aimed to develop and validate predictive models for lymphopenia after radiotherapy (RT) in breast cancer. Material and methodsPatients with breast cancer treated with adjuvant RT were eligible in this multicenter study. The study endpoint was lympopenia, defined as the reduction in absolute lymphocytes and graded lymphopenia after RT. The dose-volume histogram (DVH) data of related critical structures and clinical factors were taken into account for the development of dense neural network (DNN) predictive models. The developed DNN models were validated using external patient cohorts. ResultsA total of 918 consecutive patients with invasive breast cancer enrolled. The training, testing, and external validating datasets consisted of 589, 203, and 126 patients, respectively. Treatment volumes at nearly all dose levels of the DVH were significant predictors for lymphopenia following RT, including volumes at very low-dose 1 Gy (V1) of organs at risk (OARs) including lung, heart and body, especially ipsilateral-lung V1. A final DNN model, combining full DVH dosimetric parameters of OARs and three key clinical factors, achieved a predictive accuracy of 75 % or higher. ConclusionThis study demonstrated and externally validated the significance of full dosimetric data, particularly the volume of low dose at as low as 1 Gy of critical structures on lymphopenia after radiation in patients with breast cancer. The significance of V1 deserves special attention, as modern VMAT RT technology often has a relatively high value of this parameter. Further study is warranted for RT plan optimization.

Prediction and Optimization Analysis of the Performance of an Office Building in an Extremely Hot and Cold Region

The White Paper on Peak Carbon and Carbon Neutral Action 2022 states that China is to achieve peak carbon by 2030 and carbon neutrality by 2060. Based on the “3060 dual-carbon” goal, how to improve the efficiency of energy performance is an important prerequisite for building a low-carbon, energy-saving, green, and beautiful China. The office performance building studied in this paper is located in the urban area of Turpan, where the climate is characterized by an extremely hot summer environment and a cold winter environment. At the same time, the building is oriented east–west, with the main façade facing west, and the main façade consists of a large area of single-layer glass curtain wall, which is affected by western sunlight. As a result, there are serious problems with the building’s energy consumption, which in turn leads to excessive carbon emissions and high life cycle costs for the building. To address the above problems, this paper analyzes and optimizes the following four dimensions. First, the article creates a Convolutional Neural Network (CNN) prediction model with Total Energy Use in Buildings (TEUI), Global Warming Potential (GWP), and Life Cycle Costs (LCC) as the performance objectives. After optimization, the R2 of the three are 0.9908, 0.9869, and 0.9969, respectively, thus solving the problem of low accuracy of traditional prediction models. Next, the NSGA-II algorithm is used to optimize the three performance objectives, which are reduced by 41.94%, 40.61%, and 31.29%, respectively. Then, in the program decision stage, this paper uses two empowered Topsis methods to optimize this building performance problem. Finally, the article analyzes the variables using two sensitivity analysis methods. Through the above research, this paper provides a framework of optimization ideas for office buildings in extremely hot and cold regions while focusing on the four major aspects of machine learning, multi-objective optimization, decision analysis, and sensitivity analysis systematically and completely. For the development of office buildings in the region, whether in the early program design or in the later stages, energy-saving measures to optimize the design have laid the foundation of important guidelines.

Research on Game-Theoretic Supply Chain Optimisation Strategies Based on Multiple Linear Regression and BP-PSO Combined Time Series Analysis

Vegetable commodities have a short cycle, and supermarkets generally replenish goods based on historical sales and demand for each commodity. Vegetable commodities are different from other commodities with relatively stable supply chain planning and market environment, so the automatic pricing and replenishment decisions of fresh food superstores on commodities are of great significance. This paper focuses on the problem of commodity pricing and replenishment strategy. Firstly, through multiple linear analysis regression analysis, the relationship between total quantity and cost plus pricing, fit a one-dimensional linear fitting equation, and visualise the correlation between sales price and total quantity, secondly, observe the trend of various types of vegetables through matlab time-series analysis, establish BP neural network prediction model, predict the total quantity of sales and total quantity of replenishment in the coming week, and formulate the commodity pricing and replenishment strategy using Linear programming to establish the optimization model, verify the feasibility of the model, the parameter normality test and residual test and visual inspection error, analyse the parameter significance, use BP neural network to test the fitting effect. The results of the study, based on the game theory of supply chain replenishment and pricing strategy to establish a mathematical model.

Development of prediction model for structural behavior and gradation of porous asphalt pavement

This study is aimed at developing prediction model for structural behavior of Porous Asphalt Pavement (PAP) using ABAQUS software and also developing ANN (Artificial Neural Network) model to predict void percentages in Porous Asphalt mix gradations. Data from the past literatures were used to analyze various mixing parameters affecting the properties of Porous Asphalt gradation mixes. The study analyzed PAPs using KENPAVE software. The findings indicated that fewer allowable repetitions to fatigue and rutting failures were observed for thinner Porous Asphalt Concrete (PAC) layer thicknesses. This suggests that thicker PAC layers may offer better resistance to fatigue and rutting failures in pavement systems. The study found that the nature of subgrade material significantly influenced the rutting performance of PAP systems. Specifically, clayey soils exhibited a 77.74% reduction in the design life of the pavement compared to gravelly soil subgrades. This highlights the importance of considering subgrade characteristics in pavement design and construction to optimize pavement performance and longevity. Higher contact pressures resulted in higher tensile stresses at the bottom of PAC layer which in turn reduced the fatigue life of PAP. The findings from ABAQUS analysis indicated that an increase in the void percentage in Porous Asphalt Concrete (PAC) mixes led to a significant increase in horizontal tensile strain at the bottom of the PAC layer, with a 12.3% increase observed. Additionally, there was a noticeable increase in tensile strain for a PAC mix with 28% voids compared to a mix with 16% voids. This suggests that higher void percentages in the PAC mixes can potentially enhance the flexibility and deformation resistance of the pavement structure, which may contribute to improved performance under various loading conditions, hence leading to 92.8% reduction in allowable load repetitions to fatigue failure.The study further revealed that an increased void percentage in Porous Asphalt Concrete (PAC) mixes resulted in a decrease in vertical stress and deflection in the PAP system. However, no significant effect on the allowable repetitions to rutting failure was observed. This suggests that higher void percentages may lead to better load distribution and reduced vertical stresses within the pavement structure, contributing to improved performance in terms of stress and deflection.Moreover, Asphalt Pavement mixes were compiled to develop an Artificial Neural Network (ANN) prediction model. The results demonstrated good conformity between predicted and actual data, with a mean square error of 0.109 and a coefficient of correlation of 0.994. This indicates that the ANN model accurately predicts pavement performance based on the compiled asphalt pavement mixes, providing a valuable tool for pavement design and analysis.

Neural network survivability approach of a wave energy converter considering uncertainties in the prediction of future knowledge

To tune the wave energy converter (WEC) controller parameters such as damping to reduce the line force during extreme wave conditions, future knowledge of the line force is required. To achieve this, the incoming wave and system state should be predicted for a few seconds in the future. It is rather an arduous task to predict the future knowledge of waves and the system’s dynamic when dealing with breaking and steep waves, and the system is subject to various nonlinear forces. The classical model-based control strategies often rely on linear assumptions to estimate the WEC dynamics for the sake of simplicity. Unlike the model-based, the data-driven approaches are free from modeling errors and the algorithms are trained over the true and noisy data to predict non-linear system behaviors. Using data-driven approaches, we are able to model nonlinear dynamics. However, new questions emerge on the accuracy of the future wave and system state predictions, and how this uncertainty propagates into the final prediction of the line force. As incorrect damping may lead to excessive line force and detrimental damage to the system, these are the knowledge gaps that need to be addressed. The main purpose of this paper is to answer these questions through a survivability strategy for wave energy converters by providing a realistic perspective on the implementation of the neural network approaches by accounting for the errors in the input data. For this purpose, a series of neural networks is designed that first predicts the surface elevation for 0.36 s ahead, i.e. corresponding to 2 s in the full-scale WEC. This future knowledge of the wave elevation is then used to predict the system state (i.e. power take-off (PTO) translator position) for the same prediction horizon based on the PTO damping. This information is then fed to a convolutional neural network (CNN) that predicts the peak line force 0.36 s ahead. This paper also evaluates the predictive capabilities of neural network (NN) models, comparing them to classical autoregressive (AR) and Kalman filter methods. While the AR model exhibits slightly higher accuracy in fixed PTO system configurations, it falters in providing real-time predictions when system parameters undergo variations. In contrast, the NN prediction model excels by establishing a robust relationship between system configuration and output signals. Especially noteworthy is its ability to outperform classical methods with minimal training on a selective set of system configurations. Then, the sensitivity of the peak line force prediction to the uncertainties in the input data from NN models and the prediction horizon is analyzed. The neural network models are trained over the experimental data subjected to extreme sea states for a point absorber wave energy converter. The results suggest that the accuracy of the surface elevation prediction has an insignificant direct effect on the peak force prediction model. However, these uncertainties are reflected in the PTO translator position prediction, and the model is considerably sensitive to the accuracy of this prediction. This sensitivity nonetheless is less notable for higher PTO damping values.

Impact of Real-Time Blind Slat Angle Control on Reducing Total Energy Consumption of a Library Building Using an Optimized Artificial Neural Network Model: A Case Study of the Six Moroccan Thermal Zones

Abstract The prevalence of extensive glazed areas in contemporary buildings contributes significantly to solar radiation infiltration, elevating energy demands and causing discomfort for occupants. Window shading devices play a pivotal role in addressing this challenge. This paper presents the development and optimization of an artificial neural network (ANN) predictive model, designed to enable real-time control of slat angles by predicting total energy loads, specifically during summer (for cooling and lighting purposes). The refined model demonstrates high precision, achieving a normalized root mean square error (nRMSE) of approximately 1.72% and a correlation coefficient (R) of around 0.999, despite utilizing limited meteorological data. Key inputs for the model include solar radiation, solar altitude, and external temperature, with a particular focus on slat reflectivity. The study assesses the efficiency of three slat types based on their reflectivity: high (80%), medium (50%), and low (20%). Additionally, the research explores the impact of window-to-wall ratio (WWR) values on the control system's efficacy, revealing a positive correlation between higher WWR values and improved energy savings through ANN slat angle control. Furthermore, the study extends the applicability of the ANN model to the six thermal zones in Morocco, affirming its generalization across diverse environmental conditions.

Establishment of urban green corridor network based on neural network and landscape ecological security

The planning and construction of urban ecological corridors play a role in restoring and improving the ecological environment of cities, promoting the movement of other species and biological factors living in urban environments and wider regions. It also profoundly influences and enriches the spiritual and cultural experiences of people living in cities and nearby spaces. The reason for the emphasis on green corridors is that many cities and towns are constantly expanding, and the urban green space system has not yet formed an effective network. At the same time, factors such as the loss of the urban natural environment, biodiversity reduction, and environmental degradation have led to the need to build urban green corridors to deal with risks. By improving the neural network model, this paper predicted the construction land scale of the urban green corridor network, which was used to adjust the land use structure of the green corridor and optimize the land use layout. This paper aims to use the upgraded neural network method to predict the scale of urban green corridor network building land, which helps to evaluate the ecological security status. It can solve the dynamic solution problem of multi-indicator variable weight problems, overcoming the influence of subjective factors in the weight-setting process. The experiment adopted the improved neural network model for prediction. The results showed that its accuracy was much higher than the gray prediction model, which has improved by about 14.81%. This paper fully proved that the improved neural network model had a high degree of fit and feasibility for predicting the land scale of urban green corridor networks. It is directly related to the rationality and practicability of the urban green corridor network planning scheme, which plays a role in guaranteeing the ecological security of the landscape.

Data-Physics Fusion-Driven Defect Predictions for Titanium Alloy Casing Using Neural Network.

The quality of Ti alloy casing is crucial for the safe and stable operation of aero engines. However, the fluctuation of key process parameters during the investment casting process of titanium alloy casings has a significant influence on the volume and number of porosity defects, and this influence cannot be effectively suppressed at present. Therefore, this paper proposes a strategy to control the influence of process parameters on shrinkage volume and number. This study constructed multiple regression prediction models and neural network prediction models of porosity volume and number for a ZTC4 casing by simulating the gravity investment casting process. The results show that the multiple regression prediction model and neural network prediction model of shrinkage cavity total volume have an accuracy of over 99%. The accuracy of the neural network prediction model is higher than that of the multiple regression model, and the neural network model realizes the accurate prediction of shrinkage defect volume and defect number through pouring temperature, pouring time, and mold shell temperature. The sensitivity degree of casing defects to key process parameters, from high to low, is as follows: pouring temperature, pouring time, and mold temperature. Further optimizing the key process parameter window reduces the influence of process parameter fluctuation on the volume and number of porosity defects in casing castings. This study provides a reference for actual production control process parameters to reduce shrinkage cavity and loose defects.