Particle Swarm Optimization Back Propagation Research Articles

Research on the Uplift Pressure Prediction of Concrete Dams Based on the CNN-GRU Model

Dam safety is considerably affected by seepage, and uplift pressure is a key indicator of dam seepage. Thus, making accurate predictions of uplift pressure trends can improve dam hazard forecasting. In this study, a convolutional neural network, (CNN)-gated recurrent neural network, (GRU)-based uplift pressure prediction model was developed, which included the CNN model’s feature extractability and the GRU model’s learnability for time series correlation data. Then, the model performance was verified using a dam as an example. The results showed that the mean absolute errors (MAEs) of the CNN-GRU model were 0.1554, 0.0398, 0.2306, and 0.1827, and the root mean square errors (RMSEs) were 0.1903, 0.0548, 0.2916, and 0.2127. The prediction performance was better than that of the particle swarm optimization–back propagation (PSO-BP), artificial bee colony optimization–support vector machines (ABC-SVM), GRU, long short-term memory network (LSTM), and CNN-LSTM models. The method improves the utilization rate of dam safety monitoring results and has engineering utility for safe dam operations.

Compensation of environmental parameters for optical detection of chemical oxygen demand

Chemical oxygen demand (COD) reflects the degree of water damage by organic pollutants, and is an important indicator for water environment protection and marine hydrological monitoring, so it is very important to accurately measure COD. However, changes in ambient temperature and atmospheric noise cause huge deviations in the precise measurement of COD by optical-based water quality detectors. The purpose of this research is to realize the accurate measurement of COD of the optical water quality detector by compensating the environmental parameters of the water quality detector. The compensation model established in this paper is a particle swarm optimization (PSO) back-propagation neural network (BPNN), which can compensate for temperature and filter out atmospheric noise, named back-propagation neural network of particle swarm optimization of filter out atmospheric noise (FAN-PSO-BPNN). FAN-PSO-BPNN reduced the maximum relative error by 92.51%, root mean square error by 91.64%, coefficient of variation by 91.74%, and the distance between the maximum and minimum prediction errors by 92.94% compared with BPNN in filtering out atmospheric noise interference and temperature compensation. The optimization scheme proposed in this paper for BPNN is crucial and effective, and the FAN-PSO-BPNN compensation model improves the anti-interference ability of the water quality detector, guarantees the accurate measurement of COD in seawater, and helps the water quality detector to monitor in seawater for a long time.

The application of PSO-BP combined model and GA-BP combined model in Chinese and V4’s economic growth model

Abstract This paper adopts different optimization algorithms such as Genetic Algorithm (GA) and Particle Swarm Optimization Algorithm (PSO-Algorithm) to train Back-Propagation (BP) neural networks, fits the Chinese, the Czech, Slovak, Hungarian, and Polish gross domestic product (GDP) growth model (from 1995 to 2020) and makes short-term simulation predictions. We use the PSO-Algorithm and GA with strong global search ability to optimize the weights and thresholds of the network, combine them with the BP neural network, and apply the resulting Particle Swarm Optimization Back-Propagation (PSO-BP) combined model or Genetic-Algorithm Back-Propagation (GA-BP) combined model to allow the network to achieve fast convergence. Besides, we also compare the above two hybrid models with standard multivariate regression model and BP neural network with different initialization methods like normal uniform and Xavier for fitting and short-term simulation predictions. Finally, we obtain the excellent results that all the above models have achieved a good fitting effect and PSO-BP combined model on the whole has a smaller error than others in predicting GDP values. Through the technology of PSO-BP and GA-BP, we have a clearer understanding of the five countries gross domestic product growth trends, which is conducive to the government to make reasonable decisions on the economic development.

A static load position identification method for optical fiber-composite structures based on particle swarm optimization- back Propagation neural network algorithm

Aiming at researching on health monitoring of composite materials, a static load position identification method for optical-fiber composite structures based on the Particle Swarm Optimization-Back Propagation (PSO-BP) neural network algorithm is proposed. Based on the 2 × 2 optical fibers-composite structures, the PSO-BP algorithm is used to establish a nonlinear mapping between the fiber output intensity and the position. At first, a three-layer BP neural network is established. The number of the hidden layer is 30. And then the PSO algorithm is used to optimize the initial weights and thresholds of the BP neural network. Finally, a BP neural network is built using optimized initial weights and thresholds. A total of 515 sets of data samples are collected by the experimental system, of which 500 sets are used for training and 15 sets are used for the final model prediction. Simulation results show that the Mean Square Error (MSE) of the static load position prediction based on the PSO-BP algorithm is 0.0485. Compared with the position prediction model established by the BP neural network, Radial Basis Function (RBF) neural network and Support Vector Regression Machine (SVRM), the PSO-BP neural network model has a higher accuracy. The proposed method has an important application value for the research of health self-diagnosis of composite structures.

Research on improving the basis weight measurement accuracy of tissue paper based on PSO-BP neural network

The near-infrared (NIR) sensor can be used for measuring the basis weight and moisture of tissue paper, but the measurement accuracy is not ideal for this paper grade. The weight range of the tissue is 10~30 g/m2, indicating that it is a low gram weight paper. The temperature and humidity of the production environment significantly impact an NIR sensor. This paper focuses on improving the measurement accuracy of tissue paper basis weight. In order to reduce the influences of temperature and humidity, a mathematical model based on a particle swarm optimization back propagation (PSO-BP) neural network is proposed. In comparison with multiple linear regression measurement models, the basis weight measurement error with the PSO-BP model is within ± 0.5 g/m2. This model can effectively improve the measurement accuracy and has a good effect on overcoming the basis weight nonlinear effect caused by the changes in ambient temperature and humidity.

Dynamic recrystallization of Ti-6Al-4V titanium alloy based on cellular automata

Ti-6Al-4V (TC4) alloy is a (α + β) two-phase titanium alloy and has good process plasticity, superplasticity, and corrosion resistance. It is widely used in the aerospace industry. During the hot working process, the microstructure and deformation behavior affect its mechanical properties. However, most of the current research focuses on the use of metallographic methods and physical modeling methods to characterize and estimate the evolution of microstructures and lacks accuracy. Therefore, in this study, the thermal deformation behavior of TC4 titanium alloy is predicted and the dynamic recrystallization process is visualized through thermal simulation test, metallographic test, particle swarm optimization–backpropagation (PSO–BP) algorithm, and cellular automata (CA) method. The conclusions are as follows: the grain growth of TC4 titanium alloy is relatively uneven during hot compression. The increase in temperature and the decrease in strain rate are favorable to dynamic recrystallization. After hot compression, the grain texture appeared in the microstructure of TC4 titanium alloy, and the type and strength of texture changed with the change in temperature. Based on the PSO–BP algorithm, the rheological behavior of TC4 titanium alloy during hot compression is predicted. Compared with the experimental value, the correlation of the PSO–BP model is 0.9972, and the error is within 10%. Finally, using the PSO–BP algorithm as the input, combined with the CA method and the related theory of dynamic recrystallization, the dynamic recrystallization behavior of TC4 titanium alloy during hot compression is simulated. The electron backscatter diffraction test is used for comparison, and the simulated microstructure is similar to the experimental structure after removing the influence of some defects. The average grain size error is within 10%. It shows that this model can well predict the dynamic recrystallization behavior of TC4 titanium alloy during hot compression.

Predicting bedrock depth under asphalt pavement through a data-driven method based on particle swarm optimization-back propagation neural network

Performance evaluations during and after asphalt pavement construction are essential to ensure long-term and safe operations. These evaluations are performed by obtaining each layer’s modulus of the pavement structure by back-calculating the deflection response. However, bedrock under the pavement can seriously affect this analysis. To limit this effect, this paper proposes predicting the bedrock depth by utilizing a data-driven machine learning algorithm based on particle swarm optimization-back propagation neural network (PSO-BPNN) aided by falling weight deflectometer (FWD) tests. Additionally, to provide sufficient data for the PSO-BPNN, a spectral element method with fixed-end boundary conditions (B-SEM) is utilized to calculate the theoretical deflections of multiple pavements at different bedrock depths. After performing a sensitivity analysis, the peak deflections at different measurement points and three time points corresponding to the peak deflection and the two subsequent fluctuation peaks at the load center measurement point are extracted as the input of the PSO-BPNN prediction model. Due to the pervasiveness of measurement errors in field tests, the raw input data are processed by adding random and systematic errors, and the corresponding bedrock depths are taken as the output dataset. The results demonstrate that the PSO-BPNN model is effective and feasible in predicting bedrock depth. Furthermore, in contrast to existing estimation models, the PSO-BPNN prediction model is more accurate and reliable with broad application prospects in road engineering.

Regression prediction of hydrogen enriched compressed natural gas (HCNG) engine performance based on improved particle swarm optimization back propagation neural network method (IMPSO-BPNN)

Artificial neural network (ANN) methods have been rapidly developed and applied in solving nonlinear small sample problems. In this paper, an improved particle swarm algorithm optimized back propagation neural network (IMPSO-BPNN) method was proposed and used for the regression analysis and prediction of 20% (volume fraction) hydrogen enriched compressed natural gas (HCNG) engine performance. Meanwhile, various ANN and support vector machine (SVM) methods were also utilized for a comparative study. The experimental results show that the HCNG engine has the highest combustion efficiency, the maximum output torque and the minimum brake specific fuel consumption (BSFC) when operating at the maximum brake torque (MBT) timing, and the brake specific NOx (BSNOx) is also at a relatively low level. Through the comparison of multiple methods, the prediction accuracy and generalization ability of the IMPSO-BPNN model are the best overall. For example, the mean absolute percentage error (MAPE) of the optimal IMPSO-BPNN model (0.771%) is 5.85%, 12.62%, 17.96%, 7.57% and 7.88% less than that of the PSO-BPNN, GA-BPNN, BPNN, PSO-SVM and GA-SVM models (0.819%, 0.882%, 0.940%, 0.834% and 0.837%), respectively; and the correlation coefficient (R) is also higher than that of other models (0.99986). Secondly, the method shows visible superiority in the temporal dimension compared with the methods optimized by genetic algorithm and SVM method. For example, the CPU running time of the optimal IMPSO-BPNN model is reduced by 1773.32%, 149.74% and 1231.34% compared to that of the optimal GA-BPNN, PSO-SVM and GA-SVM models, respectively. Since the IMPSO-BPNN method is a flexible and general method, it is a new idea for the study of engine electronic control calibration tools.

Design of feed rate monitoring system and estimation method for yield distribution information on combine harvester

Real-time monitoring of harvester feed rate is of great significance for guiding harvesting work and improving harvesting efficiency. In this study, a feed rate monitoring system is designed and developed. The system consists of a torque sensor for measuring the header power shaft torque, an angel sensor for measuring header height, a humidity sensor for measuring grains moisture content, a Global Navigation Satellite System receiver, and a vehicle industrial computer. Data, such as the moisture content of grains and the working position of the harvester, are collected by the system during harvest. The height of the header, the moisture content of grains, and the torque of the header power shaft are selected as the input quantity, and feed rate is selected as the output quantity. A calculation model for feed rate is established using a particle swarm optimization–back propagation neural network, and then real-time monitoring of feed rate is realized. On the basis of the monitoring of feed rate, an estimation method for yield distribution information is proposed. Through analyzing the rice samples in the experimental area, the method establishes a binary linear model between the header height, the grain moisture content, and the stem-to-grain ratio of the rice fed to the harvester. The model is combined with the monitoring results of feed rate to obtain the rice yield data fed to the harvester. Then, the rice yield data are matched with the positional information at the time of harvesting to obtain a yield distribution map of the harvested area. Field experiments are conducted in a rice-growing area in Northeast China, and the average relative error of the feed rate monitoring system is 7.69%. The field sample verification accuracy of the stem-to-grain ratio model is higher than 90%, and the correlation coefficient R2 is 0.93. The estimation accuracy of the yield distribution is higher than 90%, which provides a scientific basis for agricultural decision making.

An Apple Fungal Infection Detection Model Based on BPNN Optimized by Sparrow Search Algorithm.

To rapidly detect whether apples are infected by fungi, a portable electronic nose was used in this study to collect the gas information from apples, and the collected information was processed by smoothing filtering, data dimensionality reduction, and outlier removal. Following this, we utilized K-nearest neighbors (KNN), random forest (RF), support vector machine (SVM), a convolutional neural network (CNN), a back-propagation neural network (BPNN), a particle swarm optimization–back-propagation neural network (PSO-BPNN), a gray wolf optimization–backward propagation neural network (GWO-BPNN), and a sparrow search algorithm–backward propagation neural network (SSA-BPNN) model to discriminate apple samples, and adopted the 10-fold cross-validation method to evaluate the performance of each model. The results show that SSA can effectively optimize the performance of the BPNN, such that the recognition accuracy of the optimized SSA-BPNN model reaches 98.40%. This study provides an important reference value for the application of an electronic nose in the non-destructive and rapid detection of fungal infection in apples.

Membrane Fouling Prediction Based on Tent-SSA-BP.

In view of the difficulty in obtaining the membrane bioreactor (MBR) membrane flux in real time, considering the disadvantage of the back propagation (BP) network in predicting MBR membrane flux, such as the local minimum value and poor generalization ability of the model, this article introduces tent chaotic mapping in the standard sparrow search algorithm (SSA), which improves the uniformity of population distribution and the searching ability of the algorithm (used to optimize the key parameters of the BP network). The tent sparrow search algorithm back propagation network (Tent-SSA-BP) membrane fouling prediction model is established to achieve accurate prediction of membrane flux; compared to the BP, genetic algorithm back propagation network (GA-BP), particle swarm optimization back propagation network (PSO-BP), sparrow search algorithm extreme learning machine(SSA-ELM), sparrow search algorithm back propagation network (SSA-BP), and Tent particle swarm optimization back propagation network (Tent–PSO-BP) models, it has unique advantages. Compared with the BP model before improvement, the improved soft sensing model reduces MAPE by 96.76%, RMSE by 99.78% and MAE by 95.61%. The prediction accuracy of the algorithm proposed in this article reaches 97.4%, which is much higher than the 48.52% of BP. It is also higher than other prediction models, and the prediction accuracy has been greatly improved, which has some engineering reference value.

Prediction Model of Sound Signal in High-Speed Milling of Wood-Plastic Composites.

The accuracy of the acoustic signal prediction model for wood–plastic composites milling has an important influence on the condition monitoring of the cutting process and the improvement of the machining environment. To establish a high-precision prediction model of sound signal in the high-speed milling of wood–plastic composites, high-speed milling experiments on self-developed wood–plastic composites were carried out with cemented carbide tools. A mathematical model of the relationship of the four milling parameters, including axial cutting depth, radial cutting depth, feed rate and cutting speed, and the sound signal of wood–plastic composites milling, was established by using the full-factor test method. The experimental data obtained by the orthogonal test method were used as the test samples in the mathematical model. Test results show that the prediction accuracy of the mathematical model of the sound signal in the milling of wood–plastic composites exceeds 95.4%. To further improve the prediction accuracy of the sound signal in the milling of wood–plastic composites, a prediction model was established using back propagation (BP) neural network. Then, the particle swarm optimization (PSO) algorithm was used to optimize the BP neural network, obtaining the PSO–BP neural network prediction model. The test results show that the prediction accuracy of the PSO–BP prediction model for the sound signal in the high-speed milling of wood–plastic composites exceeds 97.5%. The PSO–BP model has a better global approximation ability and higher prediction accuracy than the BP model. The research results can provide a reference basis for sound signal prediction in the high-speed milling of wood–plastic composites.

Global 1 km × 1 km gridded revised real gross domestic product and electricity consumption during 1992–2019 based on calibrated nighttime light data

As fundamental data, gross domestic product (GDP) and electricity consumption can be used to effectively evaluate economic status and living standards of residents. Some scholars have estimated gridded GDP and electricity consumption. However, such gridded data have shortcomings, including overestimating real GDP growth, ignoring the heterogeneity of the spatiotemporal dynamics of the grid, and limited time-span. Simultaneously, the Defense Meteorological Satellite Program’s Operational Linescan System (DMSP/OLS) and National Polar-orbiting Partnership’s Visible Infrared Imaging Radiometer (NPP/VIIRS) nighttime light data, adopted in these studies as a proxy tool, still facing shortcomings, such as imperfect matching results, discontinuity in temporal and spatial changes. In this study, we employed a series of methods, such as a particle swarm optimization-back propagation (PSO-BP) algorithm, to unify the scales of DMSP/OLS and NPP/VIIRS images and obtain continuous 1 km × 1 km gridded nighttime light data during 1992–2019. Subsequently, from a revised real growth perspective, we employed a top-down method to calculate global 1 km × 1 km gridded revised real GDP and electricity consumption during 1992–2019 based on our calibrated nighttime light data.

State of Health estimation and Remaining Useful Life prediction for lithium-ion batteries by Improved Particle Swarm Optimization-Back Propagation Neural Network

Accurate State of Health (SOH) estimation and Remaining Useful Life (RUL) prediction play important roles in ensuring the safe operation of the batteries and minimizing maintenance costs. It is difficult to directly measure the SOH and RUL of batteries in actual application. This paper estimates SOH and predicts RUL based on Improved Particle Swarm Optimization-Back Propagation Neural Network (IPSO-BPNN) with Health Indicators (HIs) as input. The HIs are extracted from the lithium-ion batteries charging process because the charging process is stable and easy to measure. Aiming at the nonlinear problem of batteries, Back Propagation Neural Network (BPNN) with strong generalization ability is used to estimate SOH and predict RUL. In order to solve the problem of BPNN parameter initialization, Particle Swarm Optimization (PSO) combined with variation factor is adopted in this paper to optimize the initial weights and thresholds of the neural network. In addition, the exponential decaying learning rate is adopted to improve the stability and learning efficiency of the network. Two datasets of batteries are used to verify the proposed IPSO-BPNN method. The results show that compared with the standard BPNN method, the maximum root mean square error and mean absolute error of SOH estimation results by the proposed IPSO-BPNN method are reduced to 0.78% and 1.01% respectively, which proves IPSO-BPNN method has higher accuracy and validity than standard BPNN method.

Constructing a Large-Scale Urban Land Subsidence Prediction Method Based on Neural Network Algorithm from the Perspective of Multiple Factors

The existing neural network model in urban land-subsidence prediction is over-reliant on historical subsidence data. It cannot accurately capture or predict the fluctuation in the sequence deformation, while the improper selection of training samples directly affects its final prediction accuracy for large-scale urban land subsidence. In response to the shortcomings of previous urban land-subsidence predictions, a subsidence prediction method based on a neural network algorithm was constructed in this study, from a multi-factorial perspective. Furthermore, the scientific selection of a large range of training samples was controlled using a K-shape clustering algorithm in order to produce this high-precision urban land subsidence prediction method. Specifically, the main urban area of Kunming city was taken as the research object, LiCSBAS technology was adopted to obtain the information on the land-subsidence deformation in the main urban area of Kunming city from 2018–2021, and the relationship between the land subsidence and its influencing factors was revealed through a grey correlation analysis. Hydrogeology, geological structure, fault, groundwater, high-speed railways, and high-rise buildings were selected as the influencing factors. Reliable subsidence training samples were obtained by using the time-series clustering K-shape algorithm. Particle swarm optimization–back propagation (PSO-BP) was constructed from a multi-factorial perspective. Additionally, after the neural network algorithm was employed to predict the urban land subsidence, the fluctuation in the urban land-subsidence sequence deformation was predicted with the LSTM neural network from a multi-factorial perspective. Finally, the large-scale urban land-subsidence prediction was performed. The results demonstrate that the maximum subsidence rate in the main urban area of Kunming reached −30.591 mm⋅a−1 between 2018 and 2021. Moreover, there were four main significant subsidence areas in the whole region, with uneven distribution characteristics along Dianchi: within the range of 200–600 m from large commercial areas and high-rise buildings, within the range of 400–1200 m from the under-construction subway, and within the annual average. The land subsidence tended to occur within the range of 109–117 mm of annual average rainfall. Furthermore, the development of faults destroys the stability of the soil structure and further aggravates the land subsidence. Hydrogeology, geological structure, and groundwater also influence the land subsidence in the main urban area of Kunming. The reliability of the training sample selection can be improved by clustering the subsidence data with the K-shape algorithm, and the constructed multi-factorial PSO-BP method can effectively predict the subsidence rate with a mean squared error (MSE) of 4.820 mm. The prediction accuracy was slightly improved compared to the non-clustered prediction. We used the constructed multi-factorial long short-term memory (LSTM) model to predict the next ten periods of any time-series subsidence data in the three types of cluster data (Cluster 1, Cluster 2, and Cluster 3). The root mean square errors (RMSE) were 0.445, 1.475, and 1.468 mm; the absolute error ranges were 0.007–1.030, 0–3.001, and 0.401–3.679 mm; the errors (mean absolute error, MAE) were 0.319, 1.214, and 1.167 mm, respectively. Their prediction accuracy was significantly improved, and the predictions met the measurement specifications. Overall, the prediction method proposed from the multi-factorial perspective improves large-scale, high-accuracy urban land-subsidence prediction.

City- and county-level spatio-temporal energy consumption and efficiency datasets for China from 1997 to 2017

Understanding the evolution of energy consumption and efficiency in China would contribute to assessing the effectiveness of the government’s energy policies and the feasibility of meeting its international commitments. However, sub-national energy consumption and efficiency data have not been published for China, hindering the identification of drivers of differences in energy consumption and efficiency, and implementation of differentiated energy policies between cities and counties. This study estimated the energy consumption of 336 cities and 2,735 counties in China by combining Defense Meteorological Satellite Program/Operational Line-scan System (DMSP/OLS) and Suomi National Polar-Orbiting Partnership/Visible Infrared Imaging Radiometer Suite (NPP/VIIRS) satellite nighttime light data using particle swarm optimization-back propagation (PSO-BP). The energy efficiency of these cities and counties was measured using energy consumption per unit GDP and data envelopment analysis (DEA). These data can facilitate further research on energy consumption and efficiency issues at the city and county levels in China. The developed estimation methods can also be used in other developing countries and regions where official energy statistics are limited.

Integrating DMSP-OLS and NPP-VIIRS Nighttime Light Data to Evaluate Poverty in Southwestern China

Poverty alleviation is one of the most important tasks facing human social development. It is necessary to make accurate monitoring and evaluations for areas with poverty to improve capability of implementing poverty alleviation policies. Here, this study introduced nighttime light (NTL) data to estimate county-level poverty in southwest China. First, this study used particle swarm optimization-back propagation hybrid algorithm to explore the potential relationship between two NTL data (the Defense Meteorological Satellite Program’s Operational Line Scan System data and the Suomi National Polar-orbiting Partnership Visible Infrared Imaging Radiometer Suite data). Then, we integrated two NTL data at the pixel level to establish a consistent time-series of NTL dataset from 2000 to 2019. Next, an actual comprehensive poverty index (ACPI) was employed as an indicator of multidimensional poverty at county level based on 11 socioeconomic and natural variables, and which could be the reference to explore the poverty evaluation using NTL data. Based on the correlation between the ACPI and NTL characteristic variables, a poverty evaluation model was developed to evaluate the poverty situation. The result showed the great matching relationship between DMSP-OLS and NPP-VIIRS data (R2 = 0.84). After calibration, the continuity and comparability of DMSP-OLS data were significantly improved. The integrated NTL data also reflected great consistency with socioeconomic development (r = 0.99). The RMSE between ACPI and the estimated comprehensive poverty index (ECPI) based on the integrated NTL data is approximately 0.19 (R2 = 0.96), which revealed the poverty evaluation model was feasible and reliable. According to the ECPI, we found that the magnitude of poverty eradication increased in southwest China until 2011, but slowed down from 2011 to 2019. Regarding the spatial scale, geographic barriers are a key factor for poverty, with high altitude and mountainous areas typically having a high incidence of poverty. Our approach offers an effective model for evaluation poverty based on the NTL data, which can contribute a more reliable and efficient monitoring of poverty dynamic and a better understanding of socioeconomic development.

Colorimetric characterization of color imaging systems using a multi‐inputPSO‐BPneural network

AbstractMost commonly used camera characterization methods do not use a deep learning‐based artificial neural network approach at present. This article proposes a colorimetric characterization method for color imaging systems based on the multi‐input particle swarm optimization backpropagation neural network. Combined with a particle swarm optimization algorithm for global search and a 19‐input vector, this method not only overcomes the effects of local extrema on the multi‐input backpropagation neural network, but also improves the accuracy of the common input backpropagation neural network. Images of a ColorChecker SG chart were collected using a Canon EOS 1000D camera for experimental verification, and the color differences were used to evaluate the characterization results. The results show that the color differences of the multi‐input particle swarm optimization backpropagation neural network (structure: 19‐7‐3) model are substantially better than those of the multi‐input backpropagation neural network (structure: 19‐7‐3) and common input backpropagation neural network (structure: 3‐4‐3) models. Its performance is close to that of the weighted nonlinear regression model. The multi‐input particle swarm optimization backpropagation neural network is hence an effective method for colorimetric characterization with good prediction accuracy.

Improved Particle Swarm Optimization-Based BP Neural Networks for Aero-Optical Imaging Deviation Prediction

In this article, an improved particle swarm optimization (IPSO) algorithm based on similarity and random mutation is raised. The diversity of particles in the population is decided by the size of the aggregation. When the aggregation degree of particles in the population surpass a certain threshold, the concept of similarity is used to measure the similarity between particles and global extremum, and the particles with higher similarity are discretized by mutation strategy. By increasing the particle swarm’s diversity, the population’s local and global search ability tend to balance. The weight and threshold of the back propagation (BP) neural networks are optimized by the IPSO algorithm. Then, the model of the improved particle swarm optimization back propagation neural network (IPSO-BP) is applied to the aero-optical imaging deviation prediction. The results show that the prediction accuracy of the IPSO-BP model is superior to the PSO-BP model, the extreme learning machine (ELM) model, and the least square support vector machine (LSSVM) model, and its convergence speed is faster than that of the PSO-BP neural network model. Finally, the application of deep learning in aero-optical imaging deviation prediction is analyzed compared with the IPSO-BP neural network model.

Method for identifying the impact load condition of thin-walled structure damage based on PSO-BP neural network.

Thin-walled structures (TWS) were widely used in engineering equipment, and may be subjected to impact loads to produce different degrees of structural damage during application. However, it is a difficult problem to determine the impact load conditions for these structural damages. In this study, we developed a novel method of identifying the impact load condition of the thin-walled structure damage, which is based on particle swarm optimization-backpropagation (PSO-BP) neural network. First, the known impact position and velocity are applied to the finite element model (FEM) of the TWS to produce permanent plastic deformation, and to fit the characteristic shape of the deformation is needed by invoking the multivariate polynomial function. Then, the method is devoted to build a basic data set. With impact position and velocity as input and function coefficients as output, a model of extended PSO-BP neural network is established. Besides, the basic sample set is expanded to solve the lack of samples. Ultimately, utilizing the expanded total sample set as training data, function coefficients, impact position and velocity will be outputted. On the basis of the known functional coefficients of deformed surfaces, a model of predictive PSO-BP neural network is established and predicted. Furthermore, we predicted the collision position and velocity using a conventional BP neural network in the same way. Finally, the predicted impact position and velocity is compared with the analysis results of the FEM, which verifies that the PSO-BP neural network algorithm has high accuracy.