Search

The nonlinear representation of active devices plays an important role in microwave circuit design. Whereas, it takes a long time to extract a large amount of large signal data, and the problem of memory resource and CPU occupancy becomes significant. In order to address the problems in traditional large-signal modeling methods, in this paper an X-parameter modeling method for microwave power devices based on extreme learning machine (ELM) is proposed. To demonstrate the effectiveness of this method, a double layer back propagation (BP) neural network model is established. Then, harmonic balance simulations are used to verify the accuracy of these two models. After comparisons, it is proved that the three harmonic errors of double layer BP neural network model are 9.525dBm, 1.309dBm and 14.593dBm, respectively, and the three harmonic errors of ELM model are 0.673 dBm, 0.314 dBm, 3.09 dBm, respectively. Furthermore, the three harmonic modulus errors of double layer BP neural network model are 0.031, 0.002, 7.665e-4, respectively, and the errors of ELM model are 0.005, 0.001, 8.38e-5, respectively. Finally, in order to verify the accuracy of the predicted model in circuit design, the predicted X-parameter is used in the design of power amplifier. Moreover, the errors of the double layer BP neural network prediction model at 2.5 GHz, 5 GHz and 7.5 GHz are 1.142 dBm, 1.436 dBm and 2.294 dBm, respectively. The output power error of the ELM model at 2.5 GHz, 5 GHz and 7.5 GHz are 0.089 dBm, 0.311 dBm and 0.309 dBm, respectively. These experimental results demonstrate that the established ELM model is an efficient and valid approach for modeling GaN high electron mobility transistor types of nonlinear microwave devices.

In this paper, a traveling wave model is proposed to explore its influence on the aerodynamic drag of a Ahmed model, the experimental and numerical results of aerodynamic drag coefficient CD for the Ahmed model are in good agreement. Then by defining the aerodynamic benefit coefficient ΔCD as the evaluation index for the orthogonal experiment, range analysis is conducted to determine the influences of the amplitude A, wavelength λ and frequency ω of the wave and the vehicle speed u on ΔCD. After the analysis it can been found that λ has the least importance among these parameters, hence A, ω and u are used to construct the 105 samples for training the BP neural network to predict ΔCD, results show that ΔCD obtained from the neural network is significantly affected by the parameters of traveling wave. The prediction accuracy of the network is furtherly verified by another 15 samples which are also built on A, ω and u, and the corresponding data overlap rate of ΔCD is 96 %, so it can be concluded that the BP neural network constructed in this paper is accurate enough to predict ΔCD. [ABSTRACT FROM AUTHOR]

For the problems of slow regulation and poor tracking performance of proportional integral controllers, a control method combining BP neural network and fractional-order PIλDμ（BP-FOPID） was proposed. First, the mathematical model based on the topology of the Vienna rectifier was introduced. Then, a double closed-loop control was designed according to the mathematical model. The outer loop used a BP neural network to perform nonlinear fitting of the reference current, and the inner loop used a fractional order PIλDμ controller to track the reference current output by the outer loop. In addition, due to the problem of imbalance in the voltage of the upper and lower bridge arms on the DC side, this article adopted an improved SVPWM modulation strategy that takes into account small vectors. Finally, the corresponding simulation model was built in MATLAB/Simulink. The simulation results show the effectiveness and superiority of the proposed BP neural network control and fractional-order PIλDμ controller. [ABSTRACT FROM AUTHOR]

Copyright of Environmental Science & Technology (10036504) is the property of Editorial Board of Environmental Science & Technology and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)

Copyright of China Sciencepaper is the property of China Sciencepaper and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)

Broiler chickens are traditionally weighed by steelyard or platform scale, which is time-consuming and labor-intensive. Broiler chickens usually exhibit stress-related behavior during weighing. The 3D camera-based weighing system for broiler chickens can only weigh the broiler chicken in the monitoring area. Usually, it makes poor weight prediction due to poor segmentation especially when the broiler chicken is flapping its wings. To solve these issues, we developed one simple and low-cost weighing system with high stability and accuracy. A validity value extraction method from dynamic weighing was proposed. Then, an improved amplitude-limiting filtering algorithm and a BP neural networks model were developed to avoid accidental interference. The BP neural networks model used daily weight gain, day-age, average velocity, and the weight data after filtering algorithm as the input layer. The weighing system was tested in a commercial Beijing Fatty Chickens house with Beijing Fatty Chickens. We tested thirteen groups of Beijing Fatty Chickens of different weights, from 500 g to 1800 g in intervals of 100 g, using the three different methods: no filtering algorithm or BP neural networks, only the improved amplitude-limiting filtering algorithm and a hybrid of the improved amplitude-limiting filtering algorithm and BP neural networks. The results showed that the hybrid algorithm had a better performance in minimizing the error, lowering from the original 6% down to 3%. The accurate weight data was transmitted to the remote service platform for further decision-making, such as activity analysis, feeding management, and health alerts.

Due to fuel price fluctuations and greenhouse gas emissions caused by international shipping, the reduction of ship fuel consumption is important. In this paper, a ship energy consumption prediction model is established based on the BP neural network optimized by the Levenberg-Marquardt (LM) algorithm to predict ship energy consumption. Firstly, the ship navigation data are analyzed. Ship speed, draught and current speed are selected as the main factors affecting ship energy consumption and their weights are determined by the spearman and regression analysis. Then, the ship energy consumption model is established based on the LM-BP algorithm. Finally, ship speed, draught and current velocity are used as the input and ship energy consumption is considered as output for the model to validate the model. The predicted results is compared with the real energy comsumption and the final RMSE is approximately 259.74 kW. For ships sailing with power ranging from 6000 kW to 8500 kW, the model prediction error is 3.06 %. The results of this study could be used to optimize ship operational plans to improve the economics of waterway transport. • The effect of factors are quantified on ship energy consumption through Spearman analysis and regression analysis. • A prediction model based on LM-BP algorithm is established to predict ship energy consumption. • The model is trained and verified using the real data from 40 voyages of the ship. • The strength of the model is shown by comparing the results with the BP and LSTM models. [ABSTRACT FROM AUTHOR]

Copyright of Control Theory & Applications / Kongzhi Lilun Yu Yinyong is the property of Editorial Department of Control Theory & Applications and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)

Residual stress plays a key role in the mechanical properties and geometry stability of the printing parts during Fused Deposition Modeling (FDM). Being the representative thermoplastic in this type of manufacturing process, the process parameters of polylactic acid (PLA) in FDM have a significant impact on the residual stress of PLA. According to the multiphysics model established, the residual stress decreases with increasing layer thickness, printing speed or platform temperature. Because such traditional finite element analysis takes a long time to calculate the stress, a back propagation (BP) neural network model is established for rapid stress prediction under different processing parameters during FDM. Only 0.56 s is required with such model during each run and the prediction error is controlled in 10%. A close loop system is simulated with temperature modification to mimic an ideal real-time feedback. Meanwhile, by off-line adjusting the FDM process parameters, the residual stress along X direction can be controlled to a certain range from experiments. An intelligent additive manufacturing system can be envisaged with the possibility of stress state modification at any time and any position in the future. [ABSTRACT FROM AUTHOR]

In this paper, a new evaluation method for under-graduate education quality is proposed based on Artificial Intelligence Neural Network Back-Propagation (BP) algorithm and stress testing. Using this method, a publically available indicator pool is constructed, consisting of 19 variables in 4 dimensions such as Teaching Attitude, Teaching Content, Teaching Approach, and Basic Characteristic of Teachers, which impact under-graduates' mastery of knowledge and capacity building. After the BP neural network algorithm is used to learn the optimum parameters for this evaluation model, sensitivity test is applied to identify the indicators that have significant effects on the quality of education. Furthermore, scenario analysis is utilized to explore the influence of the quality of education under pre-specified situations, which provides theoretical and empirical support for evaluating under-graduate teaching, improving education quality, and enriching teacher resources. [ABSTRACT FROM AUTHOR]

Copyright of Engineering Mechanics / Gongcheng Lixue is the property of Engineering Mechanics Editorial Department and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)

Abstract Achieving accurate speed prediction provides the most critical support parameter for high-level energy management of plug-in hybrid electric vehicles. Nowadays, people often drive a vehicle on fixed routes in their daily travels and accurate speed predictions of these routes are possible with random prediction and machine learning, but the prediction accuracy still needs to be improved. The prediction accuracy of traditional prediction algorithms is difficult to further improve after reaching a certain accuracy; problems, such as over fitting, occur in the process of improving prediction accuracy. The combined prediction model proposed in this paper can abandon the transitional dependence on a single prediction. By combining the two prediction algorithms, the fusion of prediction performance is achieved, the limit of the single prediction performance is crossed, and the goal of improving vehicle speed prediction performance is achieved. In this paper, an extraction method suitable for fixed route vehicle speed is designed. The application of Markov and back propagation (BP) neural network in predictions is introduced. Three new combined prediction methods, all named Markov and BP Neural Network (MBNN) combined prediction algorithm, are proposed, which make full use of the advantages of Markov and BP neural network algorithms. Finally, the comparison among the prediction methods has been carried out. The results show that the three MBNN models have improved by about 19%, 28%, and 29% compared with the Markov prediction model, which has better performance in the single prediction models. Overall, the MBNN combined prediction models can improve the prediction accuracy by 25.3% on average, which provides important support for the possible optimization of plug-in hybrid electric vehicle energy consumption.

Rapid development of railway has heightened the need for the researches on hypereutectoid heavy rail steels. Artificial intelligence method has become an effective tool to realize materials composition design. In this paper, BP neural network models are constructed to determine the relationship among (Cr, P, S, V) alloying elements, mechanical performance and microstructure of hypereutectoid rail steels. Analysis based on this model reveals that Cr is the most prominent element for mechanical properties. The tensile strength, yield strength and hardness can be improved with the increasing content of Cr and V. The addition of P and S seems to decrease the strength and hardness of rail steels. Furthermore, the addition of (Cr, P, S, V) has a slight impact on the content of pearlite dual phases. The increase of (Cr, V) and decrease of (P, S) can contribute to an increase in ferrite content with the associated decrease in cementite. Experimental results agree well with the prediction based on the BP neural network model. This work provides an excellent basis for assessing the mechanical performance and microstructure of hypereutectoid heavy rail steels.

The multi-physical field full-coupling simulation of solid oxide fuel cell (SOFC) stack requires huge computational resources. Repeated iteration of highly non-linear calculation is easy to cause oscillation and lead to solution failure. At present, the simulation of SOFC stack models mainly focuses on the co-flow condition and counter-flow condition models. Most of them are simplified models that simplify the stack scale or physical field. In this paper, a SOFC decoupling model based on machine learning is established, and the full three-dimensional and multi-physical fields of the cross-flow large-scale SOFC stack are simulated. The model is divided into three parts for calculation, unit cell model, alternative mapping model, and cross-flow large-scale SOFC stack model. The alternative mapping model obtained by the BP neural network algorithm replaces the nonlinear multi-physics equations in the traditional model. Compared with the traditional method, the decoupling model can greatly reduce the computing resources and improve the stability of computing. In this paper, the experimental data of the single cell and the 30-layer stack are used to calibrate and verify the simulation results of stack. Studying the performance of the SOFC stack under different parameter conditions. Temperature, flow uniformity, gas mole fraction, and voltage distribution in the SOFC stack under different inlet flow rates and stack currents are obtained. Obtaining the output power and fuel utilization rate of the stack under different working conditions.

[Introduction] Limestone-gypsum wet desulfurization is a flue gas desulfurization technology with the most extensive application and the most mature technology in coal-fired power plants. Limestone slurry preparation is one of the high-energy -consuming processes, which has complex production process and difficult to correspond to material consumption and energy consumption. There is no reasonable and effective energy consumption prediction model. In order to establish a reliable pulping system energy consumption model that can guide the optimization of production parameters. [Method] Based on actual operating data of a 600 MW power plant, the controllable quantity in the production process was selected as input, and the time-delay relationship between each input variable was adjusted by mutual information theory. The improved BP neural network combined with genetic algorithm (GA) was used to establish the unit pulp energy consumption model of the pulping system. [Result] The experimental results show that compared with the unadjusted timing GA-BP model and the standard BP algorithm model, the calculation results of the GA-BP model with time series adjustment can more accurately approach the actual production data of the pulping system. [Conclusion] The established model can be applied to the energy optimization study of the slurry preparation process.

As train running speeds increase, the wheel–rail interactions of high‐speed trains are becoming more complicated, and predicting and monitoring wheel wear are becoming increasingly important for the safe operation of high‐speed trains. Therefore, identifying the critical factors that affect the wear of wheel–rail interactions and developing novel methods to predict wheel wear are of great importance. In this work, SIMPACK is used to establish a dynamic model of a high‐speed train and to investigate the normal and lateral contact forces of the wheel–rail interfaces and the wear of the wheels for a train passing through a specially designed route that consists of straight‐line, smooth‐curved, and circular tracks. The wheel wear is predicted by means of the Archard wear model based on the SIMPACK analysis, and the wear is validated by a backpropagation neural network (BPNN) classification based on daily measured data provided by the Beijing Railway Administration. The results from the SIMPACK dynamic simulation and the BPNN classification show that the position of a wheel in a bogie has a significant effect on the wheel wear, but the position of a carriage in a train does not have a significant effect on the wheel wear. The findings from this study are very useful for the maintenance and safe operation of high‐speed trains. [ABSTRACT FROM AUTHOR]

In this article, we propose a novel method for real-time detection of NaCl concentration based on ultrasonic impedance method. Specifically, our method is based on the fact that different concentration of NaCl leads to differences in acoustic characteristics of ultrasonic waves. We conducted an experimental study on the relationship between the concentration and the amplitude of ultrasonic echo. Our evaluations show that our approach can push the linear fitting degree ${r}^{{2}}$ to 0.9967. Then we used the BP neural network to classify and identify the collected sample data and achieved 100% classification accuracy. We also established a concentration–amplitude prediction model based on the BP neural network, and the result shows that the error between the predicted value and the ground truth is less than 8.36%. This method realizes non-contact measurement and can be applied to real-time monitoring of NaCl concentration in large, closed containers. [ABSTRACT FROM AUTHOR]

For the hub motor electric vehicle (HM-EV), the drive motor is directly integrated with the wheel. The unbalanced magnetic pull (UMP) of hub motor would be generated by magnet gap deformation under road surface roughness excitation. The longitudinal and vertical dynamic performances of the HM-EV system are therefore deteriorated. Firstly, to analyze and optimize the longitudinal and vertical dynamic performance of the HM-EV system, a new ten-degree-of-freedom mathematical quarter HM-EV system model equipped with air suspension model, permanent magnet brushless direct current (PM BLDC) hub motor model and rigid ring tire model is proposed. The UMP of PM BLDC hub motor is taken into consideration in this model. A HM-EV system model validation test bench is constructed. The accuracy of the model is verified by experiment. Secondly, based on quarter HM-EV system model, the BP neural network is adopted to calculate the longitudinal and vertical UMP. The relative error between results calculated by BP neural networks and electromagnetic formula is less than 5% and root-mean-square error (RMSE) is less than 2. With proposed BP neural networks calculation method, UMP calculation time is shortened by 70.3%. Finally, the adjustable force is introduced and model predictive control (MPC) method is used to suppress the longitudinal and vertical vibration of HMEV system. Two control methods, namely model predictive control (MPC) and constrained optimal control (COC) are proposed. The simulation results show that by applying MPC, the RMS value of evaluation indexes are decreased by 17.21%–44.10% respectively, which is better than COC (−14.42%–17.21%). With MPC, longitudinal and vertical vibration are suppressed. Comparison of two UMP calculation methods with MPC controller is conducted. The relative errors of evaluation indexes are within 3.85%. Therefore, the driving safety and riding comfort of the HM-EV are improved compared to the passive suspension and COC active suspension. [ABSTRACT FROM AUTHOR]

Tidal phenomenon is a significant dynamical phenomenon in the ocean, and the accurate prediction of tide is an important task for various maritime activities. This paper proposes analysis method considering tidal periodicity and apply it to the actual tide prediction. The results prove that this method can solve the delay problem in tide prediction, improve the accuracy of prediction. Compared with the tidal harmonic analysis method, the prediction result of this method is more accurate and requires less data for short-term tidal forecast. Although this study can only provide an accurate forecast for 3 days, it is enough to deal with risks. How to improve the accuracy of long-term prediction is one of the future research directions.

In order to detect the oleic acid content of rapeseed quickly and accurately, we propose, in this paper, an artificial BP neural networks based model for predicting oleic acid content by using rapeseed's hyperspectral information. Four types of spectral features are selected for our investigation, namely multifractal index, sensitive band, trilateral parameters, and spectral index. Both univariate variable and multiple variables are considered as our model input. The result shows that the combined feature has higher precision and better stability than when using a single parameter. An interesting finding shows that the combined feature involving multifractal parameters can significantly improve the model performance. Taking the combined feature {MF-h(0), SB-DR574, SPI-NDSI(R575, R576)} as the model input, the constructed BP (back propagation) neural networks model has the highest precision, with the coefficient of determination (R2) 0.8753, root mean square error (RMSE) 1.0301, and relative error (RE) 1.047%. This result provides some experience for the rapid detection of rapeseed's oleic acid content. [ABSTRACT FROM AUTHOR]

Abstract Background At present, the proportion of undiagnosed diabetes in Chinese adults is as high as 15.5%. People with diabetes who are not treated and controlled in time may have various complications, such as cardiovascular and cerebrovascular diseases and diabetic foot disorders, which not only seriously affect the quality of life of people with diabetes but also impose a heavy burden on families and society. Therefore, prevention and control of type 2 diabetes is of great significance. Methods We constructed a logistic regression model, a neural network model and a decision tree model to analyse the risk factors for type 2 diabetes and then compared the prediction accuracy of the different models by calculating the area under the relative operating characteristic (ROC) curve and back-inputting the data into the model. Results The prevalence of type 2 diabetes in 4177 subjects who were not diagnosed with type 2 diabetes was 9.31%. The most influential factors associated with type 2 diabetes were triglyceride (TG) ≥ 1.17 mmol/L (odds ratio (OR) =2.233), age ≥ 70 years (OR = 1.734), hypertension (OR = 1.703), alcohol consumption (OR = 1.674), and total cholesterol≥5.2 mmol/L (TC) (OR = 1.463). The prediction accuracies of the three prediction models were 90.8, 91.2, and 90.7%, respectively, and the areas under curve (AUCs) were 0.711, 0.780, and 0.698, respectively. The differences in the AUCs after back propagation (BP) of the neural network model, logistic regression model and decision tree model were statistically significant (P

Abstract Background Hand, foot and mouth disease (HFMD) is a rising public health problem and has attracted considerable attention worldwide. The purpose of this study was to develop an optimal model with meteorological factors to predict the epidemic of HFMD. Methods Two types of methods, back propagation neural networks (BP) and auto-regressive integrated moving average (ARIMA), were employed to develop forecasting models, based on the monthly HFMD incidences and meteorological factors during 2009–2016 in Jiangsu province, China. Root mean square error (RMSE) and mean absolute percentage error (MAPE) were employed to select model and evaluate the performance of the models. Results Four models were constructed. The multivariate BP model was constructed using the HFMD incidences lagged from 1 to 4 months, mean temperature, rainfall and their one order lagged terms as inputs. The other BP model was fitted just using the lagged HFMD incidences as inputs. The univariate ARIMA model was specified as ARIMA (1,0,1)(1,1,0)12 (AIC = 1132.12, BIC = 1440.43). And the multivariate ARIMAX with one order lagged temperature as external predictor was fitted based on this ARIMA model (AIC = 1132.37, BIC = 1142.76). The multivariate BP model performed the best in both model fitting stage and prospective forecasting stage, with a MAPE no more than 20%. The performance of the multivariate ARIMAX model was similar to that of the univariate ARIMA model. Both performed much worse than the two BP models, with a high MAPE near to 40%. Conclusion The multivariate BP model effectively integrated the autocorrelation of the HFMD incidence series. Meanwhile, it also comprehensively combined the climatic variables and their hysteresis effects. The introduction of the climate terms significantly improved the prediction accuracy of the BP model. This model could be an ideal method to predict the epidemic level of HFMD, which is of great importance for the public health authorities.

Abstract In this paper, a prediction method based on artificial neural network was developed to rapidly determine the residual shear resistance of reinforced concrete (RC) beams after fire. Firstly, the temperature distribution along the beam section was determined through finite element analysis using software ABAQUS. A residual shear strength calculation model was developed and validated using the test data. Using this model, 384 data entries were derived for training and testing. The input layer of neural network involved parameters of beam height, beam width, fire exposure time, cross-sectional area of stirrup, stirrup spacing, concrete strength, and concrete cover thickness. The output was the shear resistance of RC beams. It was found that use of BP neural network could precisely predict the post-fire shear resistance of RC beams. The predicted data were highly consistent with the target data. Thus, this is a novel method for computing post-fire shear resistance of RC beams. Using this new method, further investigation was also made on the effects of different parameters on the shear resistance of the beams.

In order to realize fully flexible variable valve lift and improve the intake efficiency of the engine, a new electromagnetic valve-train (EMVT) was designed to replace the traditional CAM in the engine valve-train. However, as the durability of electromagnetic linear actuator (ELA) in the EMVT was lower than that of CAM, and there might be some problems during a long running time. This paper presented an improved online fault diagnosis method combining BP Neural Networks with Grey Relation Analysis (GRA) to analyze current signals, which could realize the monitoring, diagnosis of electromagnetic linear actuator faults and early warning so as to prevent unnecessary situations such as accidental cylinder stop of the engine. The results showed that this method had a high fault diagnosis rate, high speed, reliability and practicability.

Background: Although the spatio-temporal structure of muscle activation in cutting have been studied extensively, including time-varying motor primitives and time-invariant motor modules under various conditions, the factorisation methods suitable for cutting are unclear, and the extent to which each factorisation method loses information about movement during dimensionality reduction is uncertain., Research Question: To clarify the extent to which NMF, PCA and ICA retain information about movement when downscaling, and to explore the factorisation method suitable for cutting., Methods: The kinematic data during cutting was captured with a Vicon motion capture system, from which the kinematic features of the pelvic centre of mass were calculated. NMF, PCA and ICA were used to obtain muscle synergies based on sEMG of the cutting at different angles, respectively. A back propagation neural network was constructed using temporal component of synergy as input and the kinematics data of pelvic as output. Calculation of the Hurst index using fractal analysis based on the temporal component of muscle synergy., Results: The quality of sEMG reconstruction is significantly higher with ICA ( P  < 0.01) than with NMF and PCA for the cutting. The NMF reconstruction has a high degree of preservation of movement, whereas the ICA loses movement information about the most of the swing phase, and the PCA loses information related to the change of direction. Hurst index less than 0.5 for all three angles of cutting., Significance: NMF is suitable for extracting muscle synergies in all directions of cutting. Information related to movement may be lost when using PCA and ICA to extract the synergy of cutting. The high individual variability of muscle synergy in cutting may be responsible for the loss of movement information in muscle synergy., Competing Interests: The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper., (© 2024 The Authors.)

Vehicle speed prediction plays a critical role in energy management strategy (EMS). Based on the adaptive particle swarm optimization–least squares support vector machine (APSO-LSSVM) algorithm with BP neural network (BPNN), a novel closed-loop vehicle speed prediction system is proposed. The database of a vehicle internet platform was adopted to construct a speed prediction model based on the APSO-LSSVM algorithm. Furthermore, a BPNN is established according to the local high-precision nonlinear fitting relationship between the predicted value and error so as to correct the prediction value. Then, the results are returned to the APSO-LSSVM model for calculating the minimum fitness function, thus obtaining a closed-loop prediction system. Finally, equivalent fuel consumption minimization strategy (ECMS) based EMS was performed. According to the simulation results, the RMSE performance is 0.831 km/h within 5 s, which is over 20% higher than other performances. Additionally, the training time is 15 min within 5 s, which is advantageous over BPNN. Furthermore, fuel consumption increases by 6.95% compared with the dynamic-programming algorithm and decreased by 5.6%~10.9% compared with the low accuracy of speed prediction. Overall, the proposed method is crucial for optimizing EMS as it is not only effective in improving prediction accuracy but also capable of reducing training time.

Taking advantages of the BP neural network in nonlinear mapping and generalization capabilities for multi‐parameter problems，the BP neural networks with three hidden layers were constructed to predict the compressive strength of laminates containing embedded fiber wrinkles. The compressive failure was numerically simulated based on a three‐dimensional damage model with the LaRC criterion. The numerical results were used as data samples for the networks training. An algorithm based on the golden section method was proposed to quickly determine the range of the neurons number in the hidden layer of the BP neural networks. Then the best number of the neurons was finally determined by comparing the prediction results and the assessment indicators of different cases. The results show that，the error of the strengths of the laminates with the maximum wrinkle angles of 5.6° ，9.9° and 11.4° predicted by the developed BP neural networks are 3.4%，4.6%，and -0.01%，respectively. The approach developed in the present work to predict the strength of composite materials based on the BP neural networks provides an effective way for the strength evaluation of composite materials in application. [ABSTRACT FROM AUTHOR]

The construction of a pest detection algorithm is an important step to couple "ground-space" characteristics, which is also the basis for rapid and accurate monitoring and detection of pest damage. In four experimental areas in Sanming City, Jiangle County, Sha County and Yanping District in Fujian Province, sample data on pest damage in 182 sets of Dendrolimus punctatus were collected. The data were randomly divided into a training set and testing set, and five duplicate tests and one eliminating-indicator test were done. Based on the characterization analysis of the host for D. punctatus damage, seven characteristic indicators of ground and remote sensing including leaf area index, standard error of leaf area index (SEL) of pine forest, normalized difference vegetation index (NDVI), wetness from tasseled cap transformation (WET), green band (B2), red band (B3), near-infrared band (B4) of remote sensing image are obtained to construct BP neural networks and random forest models of pest levels. The detection results of these two algorithms were comprehensively compared from the aspects of detection precision, kappa coefficient, receiver operating characteristic curve, and a paired t test. The results showed that the seven indicators all were responsive to pest damage, and NDVI was relatively weak; the average pest damage detection precision of six tests by BP neural networks was 77.29%, the kappa coefficient was 0.6869 and after the RF algorithm, the respective values were 79.30% and 0.7151, showing that the latter is more optimized, but there was no significant difference (p > 0.05); the detection precision, kappa coefficient and AUC of the RF algorithm was higher than the BP neural networks for three pest levels (no damage, moderate damage and severe damage). The detection precision and AUC of BP neural networks were a little higher for mild damage, but the difference was not significant (p > 0.05) except for the kappa coefficient for the no damage level (p < 0.05). An "over-fitting" phenomenon tends to occur in BP neural networks, while RF method is more robust, providing a detection effect that is better than the BP neural networks. Thus, the application of the random forest algorithm for pest damage and multilevel dispersed variables is thus feasible and suggests that attention to the proportionality of sample data from various categories is needed when collecting data. [ABSTRACT FROM AUTHOR]

In order to detect the oleic acid content of rapeseed quickly and accurately, we propose, in this paper, an artificial BP neural networks based model for predicting oleic acid content by using rapeseed’s hyperspectral information. Four types of spectral features are selected for our investigation, namely multifractal index, sensitive band, trilateral parameters, and spectral index. Both univariate variable and multiple variables are considered as our model input. The result shows that the combined feature has higher precision and better stability than when using a single parameter. An interesting finding shows that the combined feature involving multifractal parameters can significantly improve the model performance. Taking the combined feature {MF-h(0), SB-DR574, SPI-NDSI(R575, R576)} as the model input, the constructed BP (back propagation) neural networks model has the highest precision, with the coefficient of determination (R2) 0.8753, root mean square error (RMSE) 1.0301, and relative error (RE) 1.047%. This result provides some experience for the rapid detection of rapeseed’s oleic acid content.

Intrusion detection is a hot topic in network security. This paper proposes an intrusion detection method based on improved artificial bee colony algorithm with elite-guided search equations (IABC elite) and Backprogation (BP) neural networks. The IABC elite algorithm is based on the depth first search framework and the elite-guided search equations, which enhance the exploitation ability of artificial bee colony algorithm and accelerate the convergence. The IABC elite algorithm is used to optimize the initial weight and threshold value of the BP neural networks, avoiding the BP neural networks falling into a local optimum during the training process and improving the training speed. In this paper, the BP neural networks optimized by IABC elite algorithm is applied to intrusion detection. The simulation on the NSL-KDD dataset shows that the intrusion detection system based on the IABC elite algorithm and the BP neural networks has good classification and high intrusion detection ability. [ABSTRACT FROM AUTHOR]

Abstract Estimating the parameters of an uncooperative space target is vital important in the on-orbit servicing missions (OOS). Considering the measurement failures caused by the complexed space environment during the measuring procedure, this paper proposes a novel dual vector quaternions based mixed artificial neural network estimating algorithm (DVQ-MANN) to estimate the parameters of the uncooperative space target. Firstly, the dual vector quaternions (DVQ) are utilized to set up the relative kinematics and dynamics model. When the measurements are available, an Extended Kalman Filter (EKF) of the DVQ-MANN will operate to accomplish parameter estimations. When the measurements failures occur, the artificial neural networks(ANN) of the DVQ-MANN will work. In the designed DVQ-MANN, the first ANN is a three-layer feedforward neural network to estimate the states of the uncooperative space target. Besides, a novel deep convolution neural network is designed to estimate the covariance matrix of the states for its advantages in processing high dimensional inputs. By training well off-board and updating onboard, the proposed ANNs of the proposed DVQ-MANN can make reliable estimations under measurements failures. Finally, the proposed DVQ-MANN is validated by mathematical simulations to show its robust performances. [ABSTRACT FROM AUTHOR]

Climate change is the main factor affecting the country's vulnerability, meanwhile, it is also a complicated and nonlinear dynamic system. In order to solve this complex problem, this paper first uses the analytic hierarchy process (AHP) and natural breakpoint method (NBM) to implement an AHP-NBM comprehensive evaluation model to assess the national vulnerability. By using ArcGIS, national vulnerability scores are classified and the country's vulnerability is divided into three levels: fragile, vulnerable, and stable. Then, a BP neural network prediction model which is based on multivariate linear regression is used to predict the critical point of vulnerability. The function of the critical point of vulnerability and time is established through multiple linear regression analysis to obtain the regression equation. And the proportion of each factor in the equation is established by using the partial least-squares regression to select the main factors affecting the country's vulnerability, and using the neural network algorithm to perform the fitting. Lastly, the BP neural network prediction model is optimized by genetic algorithm to get the chaotic time series BP neural network prediction model. In order to verify the practicability of the model, Cambodia is selected to be an example to analyze the critical point of the national vulnerability index. [ABSTRACT FROM AUTHOR]

Copyright of Journal of Shenyang Agricultural University is the property of Journal of Shenyang Agricultural University Editorial Department and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)

Global economic growth and increasing carbon emissions pose significant challenges to managing and optimizing industry chains. Against the backdrop of carbon peak and carbon neutrality goals, manufacturing enterprises forgo economic efficiency to achieve low-carbon goals. Therefore, balancing the economic efficiency and carbon emissions of the entire industrial chain has become a vital issue for manufacturing enterprises. Based on the existing research, "a dual carbon industrial chain" is proposed in this study to reveal the characteristics of the industrial chain under the dual carbon goals. A collaborative optimization model under the "dual carbon industrial chain" was constructed by combining complex networks, back-propagation (BP) neural networks, and clonal selection algorithm to optimize the economic efficiency and carbon emissions of the entire industrial chain. Using a chip industry chain as an example, an optimization solution from the perspective of the overall industry chain is proposed. The results show that the proposed method can optimize the economic benefits and carbon emissions of the entire industrial chain, which has specific theoretical and practical significance for developing modern industrial chains with a dual carbon background. • Take the perspective of the whole industry chain as an entry point. • Analyze the industrial chain from the perspective of complex network. • Combine BP neural networks, complex networks, and clonal selection algorithm to propose an optimization model. • Take a real chip industry chain as an example to demonstrate the model's performance. [ABSTRACT FROM AUTHOR]

In response to the current problem of significant differences between old and new equipment in testing electrical parameter data, as well as the lack of a unified parameter measurement standard in the IC testing industry, we propose an equivalent processing method for electrical parameter data using BP neural networks. Firstly, the method involves mapping the electrical parameter data measured by both the old and new equipment and subsequently trimming the test data. Then, the BP neural network is applied to the equivalent processing of the dataset, aiming to unify the test limits and electrical parameter data. This approach facilitates the extrapolation of new data from old equipment data, enabling a seamless data transfer and the harmonization of industry standards. This method effectively reduces the testing difficulty and misjudgment rate, thereby improving the testing efficiency for semiconductor manufacturers. The experimental results demonstrate that the proposed algorithm can accurately infer the new equipment's test data from the old equipment's data, ensuring data smooth transition. Furthermore, the modification of electrical parameter data does not affect the yield. • BP neural network was used to trim and process the test parameters. • The equivalent treatment method is used to unify the standards between the old and new equipment. • Through deep learning-based electrical parameter data modeling to reduce the error rate during chip testing. [ABSTRACT FROM AUTHOR]