Back Propagation Genetic Algorithm Research Articles

A tightly coupled GNSS RTK/IMU integration with GA-BP neural network for challenging urban navigation

Abstract Intelligent transportation system is increasing the importance of real-time acquisition of positioning, navigation, and timing information from high-accuracy global navigation satellite systems (GNSS) based on carrier phase observations. The complexity of urban environments, however, means that GNSS signals are prone to reflection, diffraction and blockage by tall buildings, causing a degraded positioning accuracy. To address this issue, we have proposed a tightly coupled single-frequency multi-system single-epoch real-time kinematic (RTK) GNSS/inertial measurement unit (IMU) integration algorithm with the assistance of genetic algorithm back propagation based on low-cost IMU equipment for challenging urban navigation. Unlike the existing methods, which only use IMU corrections predicted by machine learning as a direct replacement of filtering corrections during GNSS outages, this algorithm introduces a more accurate and efficient IMU corrections prediction model, and it is underpinned by a dual-check GNSS assessment where the weights of GNSS measurements and neural network predictions are adaptively adjusted based on duration of the integrated system GNSS failure, assisting RTK/IMU integration in GNSS outages or malfunction conditions. Field tests demonstrate that the proposed prediction model results in a 68.69% and 69.03% improvement in the root mean square error in the 2D and 3D component when the training and testing data are collected under 150 s GNSS signal-blocked conditions. This corresponds to 52.43% and 51.27% for GNSS signals discontinuously blocked with 500 s.

Performance prediction and operating parameters optimization for proton exchange membrane fuel cell based on data-driven surrogate model and particle swarm optimization

The operating parameters of proton exchange membrane fuel cell (PEMFC) are critical to its performance and working life. This study presents a data-driven modeling approach combining a surrogate model with the particle swarm optimization algorithm to optimize operating parameters for PEMFC and obtain the maximum power density. The results show that the operating parameters significantly influence power density under high current densities, with inlet temperature having the most significant effect. Lower inlet temperature, relative humidity in the cathode and anode, along with higher operating pressure yield improved output performance. The Genetic Algorithm-Backpropagation Neural Network based surrogate model exhibits excellent predictive performance with correlation coefficients of 0.99896 and 0.99815 for the training and test sets, respectively. Optimized conditions achieve a 3.3% increase in power density compared to initial settings, with only a 0.15% error in simulation calculations. This data-driven approach provides valuable insights for maximizing PEMFC efficiency and performance.

Optimization of a microchannel heat sink with surrogate model and genetic algorithm

Flow maldistribution significantly impacts the performance and operation of microchannel heat sinks, which are extensively utilized across various industries. This study develops a surrogate model to address the flow maldistribution in microchannel heat sinks, employing a combination of genetic algorithm backpropagation neural networks and numerical simulations. Genetic algorithms optimize the microchannel heat sink manifold, involving five variables. The optimized manifold reduces the flow maldistribution factor by 38 % at a coolant inlet velocity of 0.1 m/s compared to that of the conventional structure. The mechanisms of this improvement are explored through comparative analyses of the manifold shape and pressure distribution. However, addressing the flow maldistribution alone proves insufficient in eliminating local hot zones within the microchannel heat sink. Consequently, this study introduces the concept of a corrected maldistribution factor and establishes a new structure guided by this factor as the optimization index. The results indicate a 47 % reduction in the maximum-to-minimum channel flow velocity ratio compared to that of the conventional structure at a coolant inlet velocity of 2 m/s. Furthermore, the maximum temperature drops by 20 K, significantly enhancing the flow and temperature distribution within the microchannel heat sink.

Seafloor topography of Huangyan seamount chain based on GA-BP algorithm from ship survey and gravity anomaly data

Seabed topography is important for marine geophysics and geodesy. However, conventional seabed topography inversion methods based on gravity data are constrained to a linear fitting, neglecting the impact of nonlinear terms. In this paper, we propose an innovative method for seabed topography inversion by establishing fresh mathematical relationships through machine learning techniques based on the Smith and Sandwell (SAS) method. Utilizing global sea depth data from the National Geophysical Data Center (NGDC) and gravity anomaly data from satellite altimetry, our study employs the SAS method for seabed topography inversion. The improved SAS method, enhanced by the Genetic Algorithm-Backpropagation (GA-BP) algorithm, is specifically applied to invert the Huangyan Seamount Chain topography in the South China Sea. The accuracy is evaluated by seabed topography models, including EOTPO1, GEBCO-2023, and S & S V19.1. The results show the GA-BP method significantly reduces residuals and improves the accuracy estimated by the SAS inversion method. The Root Mean Square Error (RMSE) and Mean Absolute Error (MAE) are decreased by 14.98% and 5.07%, respectively. The proposed method has valuable reference significance for future marine exploration endeavours around the world.

Enhancing urban blue-green landscape quality assessment through hybrid genetic algorithm-back propagation (GA-BP) neural network approach: a case study in Fucheng, China

This study employs an artificial neural network optimization algorithm, enhanced with a Genetic Algorithm-Back Propagation (GA-BP) network, to assess the service quality of urban water bodies and green spaces, aiming to promote healthy urban environments. From an initial set of 95 variables, 29 key variables were selected, including 17 input variables, such as water and green space area, population size, and urbanization rate, six hidden layer neurons, such as patch number, patch density, and average patch size, and one output variable for the comprehensive value of blue-green landscape quality. The results indicate that the GA-BP network achieves an average relative error of 0.94772%, which is superior to the 1.5988% of the traditional BP network. Moreover, it boasts a prediction accuracy of 90% for the comprehensive value of landscape quality from 2015 to 2022, significantly outperforming the BP network’s approximate 70% accuracy. This method enhances the accuracy of landscape quality assessment but also aids in identifying crucial factors influencing quality. It provides scientific and objective guidance for future urban landscape structure and layout, contributing to high-quality urban development and the creation of exemplary living areas.

Synthesis of HNTs-Ca/Zn catalyst for biodiesel production from acidulated palm oil: Optimized by GA-BP

The development of heterogeneous base catalyst with acid resistance has become a major topic in the current research on the biodiesel production process. Halloysite nanotubes (HNTs) is supported with bimetallic of calcium and zinc for modification and labeled as HNTs-Ca/Zn to show both the acid and base characteristics. Further, this supported catalyst is used to produce biodiesel from the acidulated palm oil through catalyzing transesterification in one-step in this study. The HNTs-Ca/Zn catalyst, synthesized by the impregnation method, is characterized by TG-DTG, XRD, XPS, TEM, SEM-EDX, nitrogen adsorption-desorption, CO2/NH3-TPD, FTIR and Raman. The capability of the HNTs-Ca/Zn catalyst in catalyzing transesterification is investigated, where the maximum yield of 95.22% is achieved with the molar ratio of Zn to Ca of 1.5. The advantage of HNTs-Ca/Zn1.5 catalyst is prominent when the transesterification source oil is acidulated, where the biodiesel yield of 92.66% is obtained even at the acid value of 26.02 mg KOH g−1. Besides, reusability of the catalyst is acceptable, where the yield of 77.91% is maintained after the fourth reused cycle. Further, the Genetic Algorithm-Back Propagation (GA-BP) neural network is used to train and predict the transesterification parameters of the reaction temperature, catalyst amount and methanol to oil molar ratio. The prediction of the biodiesel yield is reasonable. At the predicted reaction parameters of 158 °C, catalyst amount of 7.1 wt.% and methanol to oil molar ratio of 15.5, the predicted biodiesel yield is 98.71%, where the experimental verification value is 98.60%. Therefore, the HNTs-Ca/Zn catalyst is considered to demonstrate exceptional catalytic performance and recyclability, offering the potential for biodiesel production.

Study on the quality of Corydalis Rhizoma in Zhejiang based on multidimensional evaluation method

Ethnopharmacological relevanceThe quality requirements of Corydalis Rhizoma (CR) in different producing areas are uniform, resulting in uneven efficacy. As a genuine producing area, the effective quality control of CR in Zhejiang Province (ZJ) could provide a theoretical basis for the rational application of medicinal materials. Aim of the studyThe purpose of this study was to effectively distinguish the CR inside and outside ZJ, and provided a theoretical basis for the quality control and material basis research of ZJ CR. Materials and methodsThe core components of ZJ CR could be identified by HPLC combined with chemometrics screening, and the quality of CR from different producing areas was evaluated by a genetic algorithm-back propagation (GA-BP) neural network. Chromaticity and near-infrared (NIR) spectroscopy were used to identify CR inside and outside ZJ, and rapid content prediction was realized. The analgesic effect of CR in different regions was compared by a zebrafish analgesic experiment. Analgesic experiments in rats and analysis of the research status of quality components were used to screen the quality control components of ZJ CR. ResultsThe contents of palmatine hydrochloride (YSBMT), dehydrocorydaline (TQZJJ), tetrahydropalmatine (YHSYS), tetrahydroberberine (SQXBJ), corydaline (YHSJS), stylopine (SQHLJ), and isoimperatorin (YOQHS) in ZJ CR were higher than those in CR from outside ZJ, but the content of protopine (YAPJ) and berberine hydrochloride (YSXBJ) was lower than that in CR from outside ZJ. YHSJS and SQHLJ could be used as the core components to identify ZJ CR. The GA-BP neural network showed that the relative importance of ZJ CR was the strongest. Chroma-content correlation analysis and the NIR qualitative model could effectively distinguish CR from inside and outside of ZJ, and the NIR quantitative model could quickly predict the content of CR from inside and outside of ZJ. Zebrafish experiments showed that ZJ, Shaanxi (SX), Henan (HN), and Sichuan (SC) CR had significant analgesic effects, while Hebei (HB) CR had no significant analgesic effect. Overall comparison, the analgesic effect of ZJ CR was better than that of CR outside ZJ. The comprehensive score of the grey correlation degree between YAPJ, YSBMT, YSXBJ, TQZJJ, YHSYS, YHSJS, SQXBJ, and SQHLJ were higher than 0.9, and the research frequency were extremely high. ConclusionsThe relative importance of the content and origin of most components of ZJ CR was higher than that of CR outside ZJ. The holistic analgesic effect of ZJ CR was better than that of CR outside ZJ, but slightly lower than that of SX CR. YHSJS and SQHLJ could be used as the core components to identify ZJ CR. YAPJ, YSBMT, YSXBJ, TQZJJ, YHSYS, SQXBJ, YHSJS, and SQHLJ could be used as the quality control components of ZJ CR. The multidimensional evaluation method used in this study provided a reference for the quality control and material basis research of ZJ CR.

Forecasting and alert of atmospheric bioaerosol concentration profile based on adaptive genetic algorithm back propagation neural network, atmospheric parameter and fluorescence lidar

Bioaerosols are biologically originated particles in the atmosphere, which is mainly composed of bacteria, fungi, viruses, pollen, spores, and the fragmentation and disintegration of plants and animals. Bioaerosols are easy to be spread in the lower atmosphere and cause various epidemic diseases, which is harmful to human health. The forecasting and alert of bioaerosols have important scientific significance and reality needs. In this paper, a method is proposed for estimating and predicting the concentration profile of atmospheric bioaerosols using fluorescence lidar observational data. Using the powerful nonlinear prediction ability of artificial neural networks and through repeated training, a mathematical model can be established for the relationship among atmospheric environment, meteorological parameters, and bioaerosol concentration profiles. The input parameters are temperature and humidity, aerosol extinction coefficient, backscatter coefficient, PM2.5, PM10, SO2, NO2, CO, O3, and wind speed, and outputs the concentration profile of bioaerosols. The prediction results with the measurement relative deviation of genetic algorithm back propagation (GA-BP) neural network and adaptive genetic algorithm back propagation (AGA-BP) neural network were analyzed. The results indicate that the AGA-BP neural network can effectively predict the concentration distribution of bioaerosols, and the predicted concentrations of bioaerosols are 1793 particles × m−3, 3088 particles × m−3, 5261 particles × m−3, 7410 particles × m−3 and 9133 particles × m−3 for air quality with superior, fine, mild contamination, middle level pollution and heavy pollution at an altitude of 0.315 km, respectively. We found that the predicted concentration of pollution weather is much higher than that of good weather. Furthermore, the AGA-BP neural network was used to predict the concentration profiles of atmospheric bioaerosols under different weather conditions, which provided a new research method for forecasting and alert of atmospheric bioaerosols.

Dynamic modeling and vibration control optimization of a rotating hollow beam with ESACLD treatment

As an active and passive hybrid control technology, enhanced segmented active constrained layer damping (ESACLD) is an intelligent damping structure for suppressing adverse vibrations in flexible structures. In this study, the three-dimensional spatial dynamic model of a rotating hollow beam with improved ESACLD treatment is established based on the finite element method in a floating frame of reference. The damping effects of the ESACLD with both edge elements and incisions are considered in the dynamic model for the first time. Simulation results show the damping performance of the ESACLD is better than traditional active constrained layer damping with or without incisions (SACLD or ACLD). The Genetic Algorithm-Backpropagation (GA-BP) neural network is used to optimize the physical parameters of ESACLD, and the optimized model can effectively suppress the transverse and flapping bending vibration. It is also found that flapping and transverse bending vibrations exhibit differential dependence on nutation and precession angular velocities; higher nutation and lower precession velocities suppress flapping bending. The existing commercial finite element software has some problems, such as low efficiency and inaccuracy. The new finite element proposed in this paper uses a new form function to ensure accuracy in the simulation of rotation and vibration and has obvious advantages in dealing with rotational structures and complex boundary conditions. Research in this work can provide a design framework for vibration prediction and control of flexible beam-like structures.

Prediction modeling of cigarette ventilation rate based on genetic algorithm backpropagation (GABP) neural network

AbstractThe ventilation rate of cigarettes is an important indicator that affects the internal quality of cigarettes. When producing cigarettes, the unit may experience unstable ventilation rates, which can lead to a decrease in cigarette quality and pose certain risks to smokers. By establishing the ventilation rate prediction model, guide the design of unit parameters in advance, to achieve the goal of stabilizing unit ventilation rate, improve the stability of cigarette ventilation rate, and enhance the quality of cigarettes. This paper used multiple linear regression networks (MLR), backpropagation neural networks (BPNN), and genetic algorithm-optimized backpropagation (GABP) to construct a model for the prediction of cigarette ventilation rate. The model results indicated that the total ventilation rate was significantly positively correlated with weight (P < 0.01), circumference, hardness, filter air permeability, and open resistance. The results showed that the MLR models' (RMSE = 0.652, R2 = 0.841) and the BPNN models’ (RMSE = 0.640, R2 = 0.847) prediction ability were limited. Optimization by genetic algorithm, GABP models were generated and exhibited a little better prediction performance (RMSE = 0.606, R2 = 0.873). The results indicated that the GABP model has the highest accuracy in the prediction of predicting ventilation rate and can accurately predict cigarette ventilation rate. This method can provide theoretical guidance and technical support for the stability study of the ventilation rate of the unit, improve the design and manufacturing capabilities and product quality of short cigarette products, and help to improve the quality of cigarettes.

Optimal Operation Model of Ecological Flow of Hydropower Station Based on Genetic Algorithm and Neural Network

In view of the contradiction between the ecological flow regulation of the reservoir River and the power generation flow regulation of the reservoir with the largest power generation, this paper takes the maximum power generation and the maximum average ecological protection degree as the objective function, and uses the genetic algorithm back propagation neural networks (GA-BP) algorithm to establish the reservoir flow regulation model under the constraints of water balance, reservoir capacity, unit output and unit overflow, The optimal solution of the dispatching scheme is carried out, and the Shulehe hydropower station is taken as an example to verify the optimal scheme. The results show that the optimized dispatching model can ensure the demand of ecological flow, increase 32680 kW·h power generation and 23.85% power generation benefit, so as to provide scientific reference for the rational planning, dispatching and utilization of reservoir water resources.

Comparison of extraction processes, characterization and intestinal protection activity of Bletilla striata polysaccharides

We optimized the extraction process of Bletilla striata polysaccharides using orthogonal design, Box-Behnken design (BBD), and genetic algorithm-back propagation (GA-BP), then compared and evaluated them to confirm that the combination of BBD and GA-BP neural networks was capable of increasing polysaccharide yields and antioxidant activity. The optimal extraction parameters were as follows: liquid-to-solid ratio of 15 mL/g, extraction power of 450 W, and extraction time of 34 min. Under these conditions, the polysaccharide yield and antioxidant activity were 8.29 ± 0.50 % and 26.20 ± 0.28 (mM FE/mg). Subsequently, the polysaccharide was purified to obtain purified Bletilla striata polysaccharides 1 (pBSP1) with a Mw of 255.172 kDa. Scanning electron microscope (SEM), ultraviolet-visible detector (UV), fourier transform infrared spectrometer (FTIR), high performance liquid chromatography (HPLC), X-ray diffraction (XRD), nuclear magnetic resonance (NMR) and periodate oxidation were used to analyze the structure of pBSP1. The results showed pBSP1 had a smooth surface and a rough interior, with a composition of α-D conformation glucose (18.23 %) and β-D conformation mannose (53.77 %), and an amorphous crystal structure. According to the results of thermogravimetric and rheological tests, pBSP1 exhibits good thermal stability and viscoelastic behavior. Furthermore, pBSP1 protected lipopolysaccharide (LPS)-induced GES − 1 and Caco2 cells, the results showed pBSP1(400 μg/mL) lowered TEER synthesis in Caco2 cells as well as apoptosis and reactive oxygen species (ROS) production in both cells, indicating that pBSP1 may have an intestine protective effect.

Bed density prediction of gas–solid separation fluidized bed based on genetic algorithm-back propagation neural network

The fluctuation of bed density in the process of mineral separation in gas–solid fluidized bed is particularly critical, and accurate prediction of bed density will play a crucial role in industrial production. In this paper, the axial and radial distribution of bed density in gas–solid fluidized bed under different ratios of dense medium and gas velocities was investigated. Based on the data of this study and other scholars, the corresponding data set was constructed. A 13–14-1 bed density prediction model was established by using BP neural network algorithm. The overall correlation of the model is 0.9771. The model is more accurate to predict the bed density of single dense medium. The mean absolute percentage error is 4.17 % and the root mean square error is 0.0298. The optimized GA-BP (Genetic Algorithm-Back Propagation) neural network by genetic algorithm results are closer to the actual experimental values. The overall correlation of the model is increased to 0.9793, and the prediction accuracy can reach more than 90 %. The GA-BP neural network prediction model is more accurate in predicting fluidized beds with density fluctuation range of 1.6–2.2 g/cm3, which can meet the needs of industrial applications.

GA-BP Optimization Using Hybrid Machine Learning Algorithm for Thermopile Temperature Compensation

Thermoelectric pile, which uses non-contact infrared temperature measurement principle, is widely used in various precision temperature measuring instruments. This paper analyzes environmental temperature's influence on thermoelectric piles' measurement accuracy and proposes a environment temperature compensation based on GA-BP (Genetic Algorithm-Back Propagation) neural network. The GA algorithm makes up for the slow iterative speed and easy to fall into local optimization of BP algorithm. The experimental simulation results show that environment temperature compensation based on GA-BP can accurately correct the measurement error caused by environmental temperature and other factors.

Model for Technology Risk Assessment in Commercial Banks

As the complexity of banking technology systems increases, the prevention of technological risk becomes an endless battle. Currently, most banks rely on the experience and subjective judgement of experts and employees to allocate resources for technological risk management, which does not effectively reduce the frequency of technology-related incidents. Through an analysis of mainstream risk management models, this study proposes a technology-based risk assessment system based on machine learning. It first identifies risk factors in bank IT, preprocesses the sample data, and uses different regression prediction models to train the processed data to build an intelligent assessment model. The experimental results indicated that the Genetic Algorithm–Backpropagation Neural Network model achieved the best performance. Based on assessment indicators, indicator weight values, and risk levels, commercial banks can develop targeted prevention and control measures by applying limited resources to the most critical corrective actions, thereby effectively reducing the frequency of technology-related incidents.

Short-term power grid load forecasting based on optimized VMD and GA-BP

Abstract The present article proposes an enhanced hybrid neural network model that combines variational mode decomposition (VMD) and genetic algorithm-backpropagation (GA-BP) to tackle the accurate prediction task of nonstationary and nonlinear power demand data. In comparison to existing methods, this study employs the North Gallic Hawk Optimization (NGO) algorithm to preliminarily ensure the optimization of the number of VMD modes, K and the penalty factor, α. Furthermore, it utilizes the envelope entropy criterion to determine the minimum optimal VMD components for predicting low-order spectral features vectors. Building upon this, the GA-BP optimization method establishes prediction models for the NGO-VMD model’s individual components. When compared with traditional methods, the GA-BP model exhibits significant advantages in terms of rapid convergence, fewer iterations and high prediction accuracy. Empirical results demonstrate that this approach holds promising prospects for practical application in power demand prediction. In comparison to traditional model predictive performance, the root mean square error is reduced by 60.36% and the mean absolute percentage error by 44.34%, providing robust support to enhance prediction accuracy and promote sustainable development of power systems.

Research on dynamic splitting tensile damage characteristics of steel fiber reinforced concrete under freeze-thaw cycle environment based on AE

This paper aims to investigate the dynamic splitting tensile damage characteristics of steel fiber reinforced concrete (SFRC) and achieve early warning of the dangerous stage under freeze-thaw cycle environment through acoustic emission (AE) technology, Mel-frequency cepstral coefficients (MFCCs) and genetic algorithm-back propagation (GA-BP) neural network. The effects of fiber content and loading rate on mechanical properties, wavelet energy spectrum coefficient and wavelet time-frequency-energy maximum, as well as MFCCs of AE signals were studied herein. Then a GA-BP neural network was established to realize the classification of the damage degree of SFRC under freeze-thaw cycling with MFCCs as the signal characteristic. The results indicate that the increasing loading rate leads to more serious brittle failure, but the appropriate steel fiber content can make the cracks generated develop fully during the dynamic splitting tensile failure process by analyzing the wavelet energy spectrum coefficient and wavelet time-frequency-energy maximum. It is also found that the damage degree of SFRC can be identified by analyzing the MFCCs. Furthermore, the GA-BP neural network performs well in the classification of the damage degree of SFRC with the recognition over 90%, which is more suitable for concrete conditions with a proper loading rate and fiber content.

Preparation of multifunctional hydrogels based on co-pigment-polysaccharide complexes and establishment of a machine learning monitoring platform

Economic loss due to fish spoilage exceeds 25 billion euros every year. Accurate and real-time monitoring of the freshness of fish can effectively cut down economic loss and food wastage. In this study, a dual-functional hydrogel based on sodium alginate-co-pigment complex with volatile antibacterial and intelligent indication was prepared and characterized. The characterization results indicated that the sodium alginate-co-pigment complex successfully improved the stability and color development ability of blueberry anthocyanins and bilberry anthocyanins at different temperatures and pH. The double cross-linking network inside the hydrogel conferred it with excellent mechanical properties. During rainbow trout storage, the hydrogel indicated a color difference of 73.55 on the last day and successfully extended the shelf-life of rainbow trout by 2 days (4 °C). Additionally, four dual-channel monitoring models were constructed using machine learning. The validation error of the genetic algorithm back propagation model (GA–BP) was only 5.6e-3, indicating that GA-BP can accurately monitor the freshness of rainbow trout. The rainbow trout real-time monitoring platform built based on GA-BP model can monitor the freshness of rainbow trout in real time through the images uploaded by users. The results of this study have broad applicability in the food industry, environmental conservation, and economic sustainability.

Preschool Teachers Teaching Quality Evaluation Based on Neural Network Algorithms

Based on the current teaching situation of preschool teachers, in order to comprehensively evaluate the effectiveness of early childhood teaching, research have constructed a teaching quality evaluation model using fuzzy synthesis method and expert method. This model can handle the fuzzy relationship between evaluation indicators and achieve the evaluation of teaching quality. Considering that the teaching evaluation is influenced by many factors, the Genetic Algorithm back propagation (GA-BP) neural network algorithm is chosen for the solution model construction of the preschool teacher teaching quality. The entropy method chosen for the data calculation is to complete the preschool teaching quality evaluation. In the mean square error test for solving the model, the improved GA-BP model converged after 40 iterations with the model convergence speed increased by 34.65%. In the evaluation and prediction of preschool teacher indicators, the improved GA-BP model accurately evaluated the teacher classroom teaching indicators. In sample 3, sample 6, and sample 9, the improved GA-BP model scored 91, 89, and 88 points, respectively, close to the true scoring results. The improved model’s accuracy was high as 90% in teacher skills, personality charm, and academic research evaluation. The improved model also effected better in the application of the preschool teaching quality evaluation. The application effect of this model in teaching effectiveness evaluation and teacher quality evaluation is good, providing valuable reference for the establishment of early childhood education evaluation system.

Quantitative research on stress failure risk assessment for girth welds with unequal wall thickness of the X80 pipeline under lateral load

The cumulative length of the X80 steel pipeline extends to 17,000 km, with a predominant portion traversing mountainous terrains characterized by intricate geological formations. Recurrent geological hazards significantly impede the safe operation of oil and gas pipelines. Landslides can induce substantial axial forces on steel pipelines, significantly increasing the risk of pipeline failure. Moreover, as a result of sudden changes in mechanical structure, the pipeline's unequal wall thickness leads to local stress concentration, making it more vulnerable to external loads than the steel pipeline itself. Therefore, to analyze the mechanical response of steel pipelines under different working conditions, we establish a smooth particle hydrodynamics, finite element transverse landslide pipeline model for girth welds with unequal wall thickness and investigate a tearing accident caused by a lateral landslide in a section of the large-diameter natural gas pipeline. The landslide volume is found to be the main controlling factor for the failure of pipeline circumferential welds. According to the genetic algorithm backpropagation neural network, the von Mises stress of the most dangerous section of the pipeline is calculated based on pipeline and landslide parameters, and the error is 11.67 %. According to the API RP 581–2016 Risk-Based Inspection Methodology, a failure risk assessment matrix is established using the pipeline statute as an evaluation indicator to evaluate the landslide risk to pipelines in different regions. A variable-wall-thickness girth weld on the ZG pipeline was selected for application. This method has good applicability for evaluating the failure risk of circumferential welds with wall thicknesses of 12.8/15.3 mm under the action of small landslides.