BP Neural Network Prediction Model Research Articles

An Optimized Neural Network Prediction Model for Reservoir Porosity Based on Improved Shuffled Frog Leaping Algorithm

Efficient and accurate porosity prediction is essential for the fine description of reservoirs, for which an optimized BP neural network (BPNN) prediction model is proposed. Aiming at the problem that the BPNN is sensitive to initialization and converges to local optimum easily, an improved shuffled frog leaping algorithm (ISFLA) is proposed based on roulette and genetic coding. Firstly, a roulette mechanism is introduced to improve the selection probability of elite individuals, thus enhancing the global optimization ability. Secondly, a genetic coding method is carried out by making full use of effective information such as the global and local optimal solutions and the boundary values of subgroups. Subsequently, the ISFLA algorithm is verified on 12 benchmark functions and compared with four intelligent optimization algorithms, and experimental results show its good optimization performance. Finally, the ISFLA algorithm is applied to the optimization of initial weights and thresholds of the BPNN, and a new model named ISFLA_BP is proposed to study the porosity prediction problem. The logging data is preprocessed by grey correlation analysis and deviation normalization, and then the effective prediction of porosity is achieved by natural gamma, density and other relevant parameters. The performance of ISFLA_BP model is compared with the standard three-layer BPNN and four BPNN parameter optimization methods based on swarm intelligence algorithms. Experimental results show that the proposed model has higher training accuracy, stability and faster convergence speed, with a mean square error of 0.02, and its prediction accuracy for porosity is higher than that of the other five methods.

Experimental research on weathercock interference characteristics of weathercock airspeed tube

Pitot tube is one of the most important airborne devices on the aircraft. The airspeed tube can not only provide the pilot with the airspeed of the aircraft, but also the rate of climb of the aircraft. In this paper, the wind tunnel test method is used to test the weathercock type airspeed tube used on a domestic supersonic tailless layout aircraft and it obtains the airspeed tube's angle of attack and sideslip angle weathercock interference characteristics and total static pressure aerodynamic characteristics. In this process, we find that the sideslip angle begins to offset with the angle of attack increasing and the offset is about -1° when the airspeed exceeds 1.5 Ma. Subsequently, CFD calculations are used to compare and supplement the wind vane characteristics obtained from the wind tunnel test. It is found that the pressure data collected by the airspeed tube has a certain error due to the interference characteristics of the wind vane. Finally, the pressure error is corrected by BP neural network prediction model. It provides a reference to calibrating air-data-sensors for researchers.

An Intelligent Optimization for Building Design Based on BP Neural Network and SPEA-II Multiobjective Algorithm.

With the continuous development of the field of building optimization, more and more optimization methods have sprung up, among which there are many kinds of intelligent optimization algorithms. This kind of intelligent optimization algorithm usually relies on the traditional building performance simulation method to obtain the building performance index for optimization. However, intelligent optimization algorithms generally require large-scale calculations. At the same time, the time required for building performance simulation is often limited by the complexity of building models and the configuration of computers, which leads to a long time for performance optimization, which cannot give efficient and accurate feedback to designers in engineering. Building performance optimization methods based on intelligent optimization algorithms are mainly used in scientific research and are difficult to put into practical projects. Therefore, this paper builds an accurate and efficient platform for building performance prediction and optimization to help designers make decisions combined with BP neural network and the SPEA-II multiobjective optimization algorithm. Besides, the optimization results of the case are quantitatively and qualitatively analyzed and presented in visual form based on the BP neural network prediction model. Quantitative analysis includes the evolution process of solution set, convergence process, and comprehensive quality evaluation of solution set. Qualitative analysis includes Pareto frontier and optimal architectural scheme analysis. Finally, the conclusion shows that the platform prediction and optimization can give accurate and reliable optimal solution, and the optimal building scheme is reasonable and has high engineering application value.

Fuzzy Random Prediction Model of Frost Heave Characteristics of Horizontal Frozen Metro Contact Channel in Coastal Area

The frost heave characteristics of artificial frozen soil are very important information for underground freezing engineering. It is found that both the frost heaving force and rate of the same soil layer increase with the decrease of freezing temperature. In addition, due to the comprehensive influence of freezing temperature, natural water content, dry density, and tidal flow peak value, the frost heave characteristics of different soil samples are evidently uncertain. With the aim of improving the deficiency of traditional BP neural network algorithms in solving fuzzy random engineering problems, random factor and mean square error between layers are used to modify the evaluation function of the network model. On this basis, taking tidal flow peak value, freezing temperature, natural water content, and dry density as inputs, the frost heaving force and rate of frozen soil as output values, and setting the number of hidden layer elements as 4, an improved fuzzy random BP network prediction model for frost heave characteristics was established. The new network prediction model has a smaller weight and bias, and the response tends to be smoother than the traditional one, which greatly reduces the overfitting phenomenon. The engineering example shows that the improved BP neural network prediction model can make the predicted value of frost heaving force and rate basically coincide with the measured value after effective training, and the error is controlled within 8%. Therefore, the prediction model can be used as an effective tool to predict frost heaving characteristics in Nantong metro freezing construction, and the corresponding model and method can also be extended to similar engineering cases.

Optimization of BP neural network model by chaotic krill herd algorithm

Taking kidney bean as the research object, row spacing, fertilizer application and planting density were selected as experimental factors, production for the response indicators, the chaos theory, krill herd algorithm is introduced into the BP neural network, the minimum error in training as a target, the model of weight and threshold as variables to optimize the BP neural network and chaotic krill herd algorithm BP neural network prediction model was set up (C-KHA-BP). The RMSE of C-KHA-BP model is 191.93 kg/hm2、MAE is 153.18 kg/hm2, and MAPE is 12.67%, the correlation coefficient R2 is 0.95.By solving the global optimal solution of C-KHA-BP model, the optimal row spacing of kidney bean was 72.63 cm, the fertilizer application rate was 103.91 kg/hm2, and the planting density was 30 × 104 plants /hm2. The next year, the validation test was conducted in the same test area, and the yield of kidney bean under the test scheme was 2843.2 kg /hm2, the relative error between the test result and the simulation optimization result (2949.5 kg /hm2) was only −3.65%, indicating that the fitting function of C-KHA-BP prediction model was precision and the optimization result was accurate. The results of this study can provide a new approach to the prediction and optimization of similar models in the field of grain production.

Study on intelligent prediction method of rock drillability based on Bayesian lithology classification and optimized BP neural network

With the rapid development of shale gas in northeastern Sichuan province, problems such as low drilling rate, long drilling cycle and serious bit wear are becoming more and more prominent in shale gas drilling process. This paper proposes new prediction theory with two steps based on logging data sensitive to lithology by introducing Bayesian classifier and Self Adaptive Double Chain Quantum Genetic Algorithm (S-DC-QGA) which could optimize BP neural network prediction model. In the first step, the Bayesian network classifier achieved the classification and recognition of lithology, improved the correlation of model sample data, and enhanced the accuracy and speed of lithology classification. In the second step, S-DC-QGA characterized as the global optimization is used to optimize the number of hidden layer nodes, weights and thresholds of the BP neural network. Results showed that Bayesian lithologies classification enhanced the correlation of samples, moreover optimized BP neural network solved overfitting, random initial weights of BP itself and oscillation of fitting and generalization ability with slight change of network parameters, improving the prediction accuracy of rock drillability parameters and the generalization ability of the model. The findings of this study can help for better drilling design and improvement of mechanical drilling speed.

Data Privacy Protection in News Crowdfunding in the Era of Artificial Intelligence

This paper aims to study the protection of data privacy in news crowdfunding in the era of artificial intelligence. This paper respectively quotes the encryption algorithm of artificial intelligence data protection and the BP neural network prediction model to analyze the data privacy protection in news crowdfunding in the artificial intelligence era. Finally, this paper also combines the questionnaire survey method to understand the public’s awareness of privacy. The results of this paper show that artificial intelligence can promote personal data awareness and privacy, improve personal data and privacy measures and methods, and improve the effectiveness and level of privacy and privacy. In the analysis, the survey found that male college students only have 81.1% of the cognition of personal trait information, only 78.5% of network trace information, and only 78.3% of female college students’ cognition of personal credit.

Penetration Depth Prediction Based on Data Fusion

This paper presents a penetration depth prediction model based on data fusion. The parameters of the penetration analysis are divided into different evaluation spaces, and then empirical algorithms are evaluated and the better algorithm is selected in each evaluation space. A large number of simulation data is generated to solve the problem of lack of experimental data. Two BP neural network prediction models are built based on experiment data and simulation data, respectively, and the genetic algorithm is used for parameter optimization. Finally, the attention mechanism is used to fuse the two models to generate the final dimensionless penetration depth. The experiment results show that the data fusion model has good prediction accuracy both in the whole parameter space and each evaluation space.

Research on Grey Neural Network Optimal Combination in Wind Power Generation Forecasting Technique

In this paper, combining BP neural network prediction model with grey prediction model effectively as an optimal combination forecasting technique is proposed. This model not only considers the influence factors such as wind power, wind direction and temperature, but also comprehensively considers the historical wind power generation data, through the effective combination of BP neural network prediction model and grey prediction model, the prediction accuracy is improved. The results show that the error of the new grey neural network optimal combination prediction is lower than that of the single grey prediction and BP neural network prediction, which has a certain research value.

Grid investment capability prediction based on path analysis and BP neural network

With the more complex investment environment of China’s power grid, the accurate prediction of the investment ability of power grid enterprises has become an important prerequisite for managers to make precise investment decisions. This paper first selects the factors affecting the investment capacity of the power grid from the internal and external environment, and establishes the index system of the factors affecting the investment capacity. Secondly, the path analysis is used to deeply explore the interaction relationship and influence degree of each index and investment capacity. Finally, the maximum investment capacity of the power network can be predicted based on the BP neural network prediction model. The results show that the BP neural network prediction model can achieve higher prediction accuracy when predicting the power grid investment capability.

Exploration of process conditions for coupling C4 olefins based on neural network prediction model

In today's rapidly developing economy, industrial development is increasingly being updated. C4 olefins are widely used as important chemical raw materials in the production of chemicals and pharmaceuticals, and ethanol is the main raw material for the production of C4 olefins. Investigating the process conditions for the catalytic coupling of ethanol to C4 olefins has great environmental and economic value. Therefore, in this paper, a BP neural network prediction model is used to fit and predict the experimental data based on existing chemical experiments to obtain the optimal process conditions for the highest possible yield of C4 olefins.

Prediction of PM2.5 concentration based on BP neural network optimized by bee colony algorithm

According to the ambient air pollutants data and meteorological conditions data of Mianyang City in 2017, the BP neural network model based on MATLAB is established to predict the daily average PM2.5 concentration of Mianyang City in the next two days. However, the traditional BP network has the disadvantages of slow convergence speed and easy to fall into local optimum. In order to improve the prediction accuracy of the model, an optimization algorithm is added to the prediction model to avoid the model falling into local minimum. In this paper, the bee colony algorithm is added to the prediction model to improve the accuracy of BP neural network prediction model. The data from January to November are used for training, and the data from December are used as the verification results. The results show that the optimization model can accurately predict the daily average PM2.5 concentration of Mianyang City in the next two days, which provides a new idea for the prediction of PM2.5 concentration of the city, provides a theoretical basis for the early warning and decision-making of air pollution, and also provides more reliable prediction services for people’s daily travel.

Application of Training Load Prediction Model based on Improved BP Neural Network in Sports Training of Athletes

With the enhancement of data mining technology, competitive sports informatization has become an inevitable development trend. It has become a common phenomenon to use data mining technology to help athletes train scientifically, assist coaches in rational decision-making, and improve team competitiveness. In competitive sports, cyclists' adaptation to training has a complex relationship with their physical performance. In order to explore the correlation between data and provide better training data for athletes, this study proposes a load prediction model based on BP neural network (Back propagation, BP). Considering the local convergence and random assignment of traditional BP model, an adaptive genetic algorithm with improved selection operator is used to determine the initial weights and thresholds of BP neural network to improve the accuracy of the prediction model. The experimental results show that the improved adaptive genetic algorithm improves the overall optimization ability of the BP neural network, the improved BP neural network model has good stability in the convergence process, and the algorithm can search for better weight thresholds. Compared with the basic BP neural network prediction model, the accuracy of the optimized prediction model is increased by 11.86%, and the average error value is reduced by 26.21%, which is a guide to improve the training effect of the cycling team's competitive sports.

Particle Swarm Optimization Based Fire Risk Valuation Model: Shopping-Mall

In view of the shortcomings of the existing small-scale shopping mall fire prediction models, the effectiveness and scalability of the prediction results, a BP neural network prediction model is constructed to improve the prediction accuracy by considering a variety of fire risk factors. On this basis, the convergence speed of the BP neural network is accelerated with the help of the particle swarm optimization (PSO) algorithm. Then, a mixed multi-factor shopping mall fire risk grade prediction model based on a PSO based back-propagation (PSO-BP) neural network model is proposed. The constructed prediction model can simultaneously consider climate factors (daily maximum temperature, daily average temperature, 24-h precipitation, continuous drought days, sunshine hours, daily average relative humidity, and daily average wind speed), landform factors (altitude, slope, slope direction, soil water content), combustible factors (vegetation type, combustible water content, ground cover load), and human factors (density of population, distance from human activity area). Based on the actual data and field measurement data collected by the sensor network of the shopping mall (Lahore, Pakistan), the validity of the proposed model was verified by a group of experiments. The results show that the model based on the training data set and the test samples can effectively predict the fire risk level; the computational complexity of the model is significantly lower than that of the BP neural network model alone.

Development and performance test of a vehicle-mounted total nitrogen content prediction system based on the fusion of near-infrared spectroscopy and image information

To measure the total nitrogen (TN) content of large-scale farmland soil accurately in real time and guide field fertilization, a vehicle-mounted soil TN content prediction system based on the fusion of near-infrared spectroscopy and image information was developed. A combined algorithm of uniform variable illumination (UVE) and adaptive weighted sampling (CARS) were used to select the characteristic wavelength, and the selected characteristic wavelength was used as the soil spectral information. Multiple linear regression model, partial least squares regression model, BP neural network prediction model and Catboost prediction model were chosen to predict soil TN content. After comparison, Catboost was finally selected as the final prediction model. The prediction system was composed of a mechanical part, an optical part, and a control part. The mechanical part provided space support for the prediction system, the optical part obtained the spectrum and image information of the soil, mainly including seven single-band near-infrared (NIR) filters: 945, 1045, 1200, 1300, 1450, 1535, and 1615 nm; 1450 nm is the sensitive wavelength which is used to eliminate the effect of water. The control part realized spectral data collection, image feature processing, data fusion, and the prediction of soil TN content. Soil samples were used to test the accuracy of the spectrum collection of the system. The test results showed that the highest correlation coefficient between the reflectance of the prediction system at seven sensitive wavelengths and the reflectance of the NIRQuest 512 infrared spectrometer is above 0.943. The field test results show that the R2 between the predicted soil TN content by the prediction system and the value of laboratory standard detection method is greater than 0.80 and the relative errors are less than 10%. The results also show that data fusion of spectral data and image features highlighted the common contribution of the near infrared spectrum and visible light images in the detection of soil TN content, and it can improve the prediction accuracy of the instrument. Moreover, the system can provide guidance for farmland scientific variable fertilization.

Deicing Property of Asphalt Mixture Containing Steel Wool Fiber by Electromagnetic Induction Heating

Snow and ice is one of the main problems affecting road safety in winter. In order to effectively remove the snow and ice of covering the pavement, the deicing property of asphalt mixture pavement containing steel wool fiber was introduced and investigated by electromagnetic induction heating. Based on the deicing mechanism of Faraday’s law of electromagnetic induction and the Joule’s law, the influences factors affecting deicing efficiency, including length and content of steel wool fiber, ice thickness, output current and ambient temperature were analyzed. Meanwhile, the grey correlation entropy analysis and t-test between the average deicing rate and various influencing factors were explored. BP neural network prediction models of predicting change laws of average deicing rate under different influencing factors were established. The results indicate that the average deicing rate of asphalt mixture adding steel wool fiber increases with the increase of length and content of steel wool fiber. The influence degree of each factor for the average deicing rate is in order as follows: steel wool fiber content, steel wool fiber length, output current, ambient temperature and ice thickness. BP neural network has high accuracy in predicting average deicing rate under various influencing factors and the better simulation results. It is of significance to apply the technology of “electromagnetic induction heating & steel wool fiber” to the efficient deicing of asphalt pavement.

Characteristics of Mercury Pollution and Ecological Risk Assessment in Different Degraded Grasslands of the Songnen Plains, Northeastern China

Mercury (Hg) is a global and widely distributed heavy metal pollutant. Mercury can affect human health as well as the health of ecosystems and poses ecological risks. The subjects of this study are three types of grassland in the Beidianzi region, Songnen Plains, Northeastern China, characterized by different degrees of degradation. The mercury content levels in the atmosphere, soil, and forage grass on the different grasslands were determined. In addition, the relationships between the mercury pollution levels in the atmosphere and soil, and the mercury distribution correlations between the soil and plants, were examined in detail. The potential risk index (RI), single factor index (PI), and ground accumulation index (Igeo) were used to evaluate the ecological risks. The results showed that the mercury content in the soils of three types of grassland exceeded the China national standard (GB36600-2018), and the soil mercury content in the moderately degraded grassland was the highest. The single factor index method and land accumulation index method showed that the three types of grassland were slightly polluted, while the potential risk index showed that the three types of grassland were severely polluted, and the potential risk index of the moderately degraded grassland was the highest. The potential risk index decreased with the increase of soil depth. The variation trend of atmospheric mercury content was lower in the morning and evening and higher in the afternoon. The potential risk index of atmospheric mercury indicated that all types of grassland were at severe risk. There was a significant positive correlation between atmospheric mercury and soil mercury. The mercury content in herbage increased with the increase of degradation. The BP neural network prediction model constructed had good accuracy and had certain reference value.

Technical Composition and Creation of Interactive Installation Art Works under the Background of Artificial Intelligence

Interactive installation art is a kind of art that uses specific software and computer hardware as a platform, a platform for interaction between humans and machines or different people through computer hardware. It is an interactive art that uses material installations in nature as a medium. Traditional interactive installation art is not safe and convenient, in order to solve the shortcomings of traditional interactive installation art. This article introduces artificial intelligence technology by studying the overview, development, and application of artificial intelligence. The encryption algorithm for artificial intelligence data protection and the BP neural network prediction model under artificial intelligence are also introduced to ensure the safety of interactive installation art works. The part also introduces the creation tools and creation process of interactive installation art works. Finally, in the analysis part, a questionnaire analysis of the World Expo is carried out. The results of this article show that the art of connecting inserts is the most complete and open design era. Advances in science and technology, the development of digital art, and the needs of human life have led to the development of interconnected input technologies. In addition, in the survey of people’s satisfaction with artificial intelligence, we can conclude that 89% of people think that the security of artificial intelligence technology is very high. Yes, 92% of people think that artificial intelligence technology has a fast computing speed, 86% of people think that artificial intelligence technology is low in cost.

Artificial neural network-based adaptive fluid temperature regulation of extravehicular activity suit

The extravehicular activity (EVA) suit is an important component of the life-support system for astronauts working outside of a space craft. In view of the problem that the EVA suit's thermal control system cannot be reliably operated by the astronaut during extravehicular activity, scientists are working to develop a thermal comfort prediction model based on existing physiological monitoring indicators. This model could be used as the basis for an automated thermal control system for EVA suits. For this experiment, we collected physiological data (i.e. heart rate, respiration frequency, oxygen consumption and CO2 expiratory volume) from 10 male subjects (ages 24–40) as they performed various working tasks. Subjects performed tasks wearing a simulated liquid cooling suit under 12 working conditions. We developed and applied a BP neural network prediction model to examine the relationship between thermal comfort, metabolic rate and cooling based on the human body's heat balance equation. The correlation between heart rate (HR) and metabolic rate was 0.71 ​± ​0.53, while that of respiration frequency and metabolic rate was 0.90 ​± ​0.05, that of partial CO2 pressure and metabolic rate was 0.51 ​± ​0.25, and that of skin temperature and metabolic rate was 0.30 ​± ​0.27. The BP neural network model was 91% accurate in predicting the metabolic rate, and the astronaut thermal comfort model was 89% accurate. The thermal comfort prediction model, developed based on the human body heat balance equation, provides a rapid and simple quantitative judgement of the thermal comfort status of the EVA suit microenvironment. In addition, the BP neural network model, constructed based on heart rate, respiration frequency and CO2 expiratory volume, allows for dynamic predictions of the body's metabolic rate.

Chaos-GA-BP Neural Network Power Load Forecasting Based on Rough Set Theory

In order to accurately grasp the trend of power load changes and provide guidance for stable and efficient operation of the power system, this paper proposes a Chaos-GA-BP power load forecasting model based on rough sets. First, screen the main factors that affect power load changes, remove redundant data, and optimize the input of the prediction model; Second, rely on the BP neural network prediction model to adjust its topology, optimize weights and thresholds, and build a rough set-based Chaos-GA-BP power load forecasting model; Finally, considering the impact of severe power load fluctuations on the forecast results, rough set theory is introduced to perform forecast compensation when the slope on both sides of the peak is large. Applying the above method to a one-month power load forecast in a certain area in Northeast China, the change trend of the forecast curve closely follows the actual load curve. Compared with the traditional method, the average absolute error percentage is reduced by 2.67%. The experiment shows that this method has certain practicability.