An Integrated Analysis of Yield Prediction Models:A Comprehensive Review of Advancements and Challenges

The growing global requirement for food and the need for sustainable farming in an era of a changing climate and scarce resources have inspired substantial crop yield prediction research.Deep learning(DL)and machine learning(ML)models effectively deal with such challenges.This research paper comprehensively analyses recent advancements in crop yield prediction from January 2016 to March 2024.In addition,it analyses the effectiveness of various input parameters considered in crop yield prediction models.We conducted an in-depth search and gathered studies that employed crop modeling and AI-based methods to predict crop yield.The total number of articles reviewed for crop yield prediction using ML,meta-modeling(Crop models coupled with ML/DL),and DL-based prediction models and input parameter selection is 125.We conduct the research by setting up five objectives for this research and discussing them after analyzing the selected research papers.Each study is assessed based on the crop type,input parameters employed for prediction,the modeling techniques adopted,and the evaluation metrics used for estimatingmodel performance.We also discuss the ethical and social impacts of AI on agriculture.However,various approaches presented in the scientific literature have delivered impressive predictions,they are complicateddue to intricate,multifactorial influences oncropgrowthand theneed for accuratedata-driven models.Therefore,thorough research is required to deal with challenges in predicting agricultural output.

A modified stochastic model for LS+AR hybrid method and its application in polar motion short-term prediction

Short-term(up to 30 days)predictions of Earth Rotation Parameters(ERPs)such as Polar Motion(PM:PMX and PMY)play an essential role in real-time applications related to high-precision reference frame conversion.Currently,least squares(LS)+auto-regressive(AR)hybrid method is one of the main techniques of PM prediction.Besides,the weighted LS+AR hybrid method performs well for PM short-term prediction.However,the corresponding covariance information of LS fitting residuals deserves further exploration in the AR model.In this study,we have derived a modified stochastic model for the LS+AR hybrid method,namely the weighted LS+weighted AR hybrid method.By using the PM data products of IERS EOP 14 C04,the numerical results indicate that for PM short-term forecasting,the proposed weighted LS+weighted AR hybrid method shows an advantage over both the LS+AR hybrid method and the weighted LS+AR hybrid method.Compared to the mean absolute errors(MAEs)of PMX/PMY sho rt-term prediction of the LS+AR hybrid method and the weighted LS+AR hybrid method,the weighted LS+weighted AR hybrid method shows average improvements of 6.61%/12.08%and 0.24%/11.65%,respectively.Besides,for the slopes of the linear regression lines fitted to the errors of each method,the growth of the prediction error of the proposed method is slower than that of the other two methods.

A Time Series Short-Term Prediction Method Based on Multi-Granularity Event Matching and Alignment

Accurate forecasting of time series is crucial across various domains.Many prediction tasks rely on effectively segmenting,matching,and time series data alignment.For instance,regardless of time series with the same granularity,segmenting them into different granularity events can effectively mitigate the impact of varying time scales on prediction accuracy.However,these events of varying granularity frequently intersect with each other,which may possess unequal durations.Even minor differences can result in significant errors when matching time series with future trends.Besides,directly using matched events but unaligned events as state vectors in machine learning-based prediction models can lead to insufficient prediction accuracy.Therefore,this paper proposes a short-term forecasting method for time series based on a multi-granularity event,MGE-SP(multi-granularity event-based short-termprediction).First,amethodological framework for MGE-SP established guides the implementation steps.The framework consists of three key steps,including multi-granularity event matching based on the LTF(latest time first)strategy,multi-granularity event alignment using a piecewise aggregate approximation based on the compression ratio,and a short-term prediction model based on XGBoost.The data from a nationwide online car-hailing service in China ensures the method’s reliability.The average RMSE(root mean square error)and MAE(mean absolute error)of the proposed method are 3.204 and 2.360,lower than the respective values of 4.056 and 3.101 obtained using theARIMA(autoregressive integratedmoving average)method,as well as the values of 4.278 and 2.994 obtained using k-means-SVR(support vector regression)method.The other experiment is conducted on stock data froma public data set.The proposed method achieved an average RMSE and MAE of 0.836 and 0.696,lower than the respective values of 1.019 and 0.844 obtained using the ARIMA method,as well as the values of 1.350 and 1.172 obtained using the k-means-SVR method.

An Enhanced Ensemble-Based Long Short-Term Memory Approach for Traffic Volume Prediction

With the advancement of artificial intelligence,traffic forecasting is gaining more and more interest in optimizing route planning and enhancing service quality.Traffic volume is an influential parameter for planning and operating traffic structures.This study proposed an improved ensemble-based deep learning method to solve traffic volume prediction problems.A set of optimal hyperparameters is also applied for the suggested approach to improve the performance of the learning process.The fusion of these methodologies aims to harness ensemble empirical mode decomposition’s capacity to discern complex traffic patterns and long short-term memory’s proficiency in learning temporal relationships.Firstly,a dataset for automatic vehicle identification is obtained and utilized in the preprocessing stage of the ensemble empirical mode decomposition model.The second aspect involves predicting traffic volume using the long short-term memory algorithm.Next,the study employs a trial-and-error approach to select a set of optimal hyperparameters,including the lookback window,the number of neurons in the hidden layers,and the gradient descent optimization.Finally,the fusion of the obtained results leads to a final traffic volume prediction.The experimental results show that the proposed method outperforms other benchmarks regarding various evaluation measures,including mean absolute error,root mean squared error,mean absolute percentage error,and R-squared.The achieved R-squared value reaches an impressive 98%,while the other evaluation indices surpass the competing.These findings highlight the accuracy of traffic pattern prediction.Consequently,this offers promising prospects for enhancing transportation management systems and urban infrastructure planning.

Research on the IL-Bagging-DHKELM Short-Term Wind Power Prediction Algorithm Based on Error AP Clustering Analysis

With the continuous advancement of China’s“peak carbon dioxide emissions and Carbon Neutrality”process,the proportion of wind power is increasing.In the current research,aiming at the problem that the forecasting model is outdated due to the continuous updating of wind power data,a short-term wind power forecasting algorithm based on Incremental Learning-Bagging Deep Hybrid Kernel Extreme Learning Machine(IL-Bagging-DHKELM)error affinity propagation cluster analysis is proposed.The algorithm effectively combines deep hybrid kernel extreme learning machine(DHKELM)with incremental learning(IL).Firstly,an initial wind power prediction model is trained using the Bagging-DHKELM model.Secondly,Euclidean morphological distance affinity propagation AP clustering algorithm is used to cluster and analyze the prediction error of wind power obtained from the initial training model.Finally,the correlation between wind power prediction errors and Numerical Weather Prediction(NWP)data is introduced as incremental updates to the initial wind power prediction model.During the incremental learning process,multiple error performance indicators are used to measure the overall model performance,thereby enabling incremental updates of wind power models.Practical examples show the method proposed in this article reduces the root mean square error of the initial model by 1.9 percentage points,indicating that this method can be better adapted to the current scenario of the continuous increase in wind power penetration rate.The accuracy and precision of wind power generation prediction are effectively improved through the method.

Development and validation of a circulating tumor DNA-based optimization-prediction model for short-term postoperative recurrence of endometrial cancer

BACKGROUND Endometrial cancer(EC)is a common gynecological malignancy that typically requires prompt surgical intervention;however,the advantage of surgical management is limited by the high postoperative recurrence rates and adverse outcomes.Previous studies have highlighted the prognostic potential of circulating tumor DNA(ctDNA)monitoring for minimal residual disease in patients with EC.AIM To develop and validate an optimized ctDNA-based model for predicting shortterm postoperative EC recurrence.METHODS We retrospectively analyzed 294 EC patients treated surgically from 2015-2019 to devise a short-term recurrence prediction model,which was validated on 143 EC patients operated between 2020 and 2021.Prognostic factors were identified using univariate Cox,Lasso,and multivariate Cox regressions.A nomogram was created to predict the 1,1.5,and 2-year recurrence-free survival(RFS).Model performance was assessed via receiver operating characteristic(ROC),calibration,and decision curve analyses(DCA),leading to a recurrence risk stratification system.RESULTS Based on the regression analysis and the nomogram created,patients with postoperative ctDNA-negativity,postoperative carcinoembryonic antigen 125(CA125)levels of<19 U/mL,and grade G1 tumors had improved RFS after surgery.The nomogram’s efficacy for recurrence prediction was confirmed through ROC analysis,calibration curves,and DCA methods,highlighting its high accuracy and clinical utility.Furthermore,using the nomogram,the patients were successfully classified into three risk subgroups.CONCLUSION The nomogram accurately predicted RFS after EC surgery at 1,1.5,and 2 years.This model will help clinicians personalize treatments,stratify risks,and enhance clinical outcomes for patients with EC.

Short-term train arrival delay prediction:a data-driven approach

Purpose-To optimize train operations,dispatchers currently rely on experience for quick adjustments when delays occur.However,delay predictions often involve imprecise shifts based on known delay times.Real-time and accurate train delay predictions,facilitated by data-driven neural network models,can significantly reduce dispatcher stress and improve adjustment plans.Leveraging current train operation data,these models enable swift and precise predictions,addressing challenges posed by train delays in high-speed rail networks during unforeseen events.Design/methodology/approach-This paper proposes CBLA-net,a neural network architecture for predicting late arrival times.It combines CNN,Bi-LSTM,and attention mechanisms to extract features,handle time series data,and enhance information utilization.Trained on operational data from the Beijing-Tianjin line,it predicts the late arrival time of a target train at the next station using multidimensional input data from the target and preceding trains.Findings-This study evaluates our model’s predictive performance using two data approaches:one considering full data and another focusing only on late arrivals.Results show precise and rapid predictions.Training with full data achieves aMAEof approximately 0.54 minutes and a RMSEof 0.65 minutes,surpassing the model trained solely on delay data(MAE:is about 1.02 min,RMSE:is about 1.52 min).Despite superior overall performance with full data,the model excels at predicting delays exceeding 15 minutes when trained exclusively on late arrivals.For enhanced adaptability to real-world train operations,training with full data is recommended.Originality/value-This paper introduces a novel neural network model,CBLA-net,for predicting train delay times.It innovatively compares and analyzes the model’s performance using both full data and delay data formats.Additionally,the evaluation of the network’s predictive capabilities considers different scenarios,providing a comprehensive demonstration of the model’s predictive performance.

Slope stability prediction based on a long short-term memory neural network:comparisons with convolutional neural networks,support vector machines and random forest models

The numerical simulation and slope stability prediction are the focus of slope disaster research.Recently,machine learning models are commonly used in the slope stability prediction.However,these machine learning models have some problems,such as poor nonlinear performance,local optimum and incomplete factors feature extraction.These issues can affect the accuracy of slope stability prediction.Therefore,a deep learning algorithm called Long short-term memory(LSTM)has been innovatively proposed to predict slope stability.Taking the Ganzhou City in China as the study area,the landslide inventory and their characteristics of geotechnical parameters,slope height and slope angle are analyzed.Based on these characteristics,typical soil slopes are constructed using the Geo-Studio software.Five control factors affecting slope stability,including slope height,slope angle,internal friction angle,cohesion and volumetric weight,are selected to form different slope and construct model input variables.Then,the limit equilibrium method is used to calculate the stability coefficients of these typical soil slopes under different control factors.Each slope stability coefficient and its corresponding control factors is a slope sample.As a result,a total of 2160 training samples and 450 testing samples are constructed.These sample sets are imported into LSTM for modelling and compared with the support vector machine(SVM),random forest(RF)and convo-lutional neural network(CNN).The results show that the LSTM overcomes the problem that the commonly used machine learning models have difficulty extracting global features.Furthermore,LSTM has a better prediction performance for slope stability compared to SVM,RF and CNN models.

Deep Learning for Financial Time Series Prediction:A State-of-the-Art Review of Standalone and HybridModels

Financial time series prediction,whether for classification or regression,has been a heated research topic over the last decade.While traditional machine learning algorithms have experienced mediocre results,deep learning has largely contributed to the elevation of the prediction performance.Currently,the most up-to-date review of advanced machine learning techniques for financial time series prediction is still lacking,making it challenging for finance domain experts and relevant practitioners to determine which model potentially performs better,what techniques and components are involved,and how themodel can be designed and implemented.This review article provides an overview of techniques,components and frameworks for financial time series prediction,with an emphasis on state-of-the-art deep learning models in the literature from2015 to 2023,including standalonemodels like convolutional neural networks(CNN)that are capable of extracting spatial dependencies within data,and long short-term memory(LSTM)that is designed for handling temporal dependencies;and hybrid models integrating CNN,LSTM,attention mechanism(AM)and other techniques.For illustration and comparison purposes,models proposed in recent studies are mapped to relevant elements of a generalized framework comprised of input,output,feature extraction,prediction,and related processes.Among the state-of-the-artmodels,hybrid models like CNNLSTMand CNN-LSTM-AM in general have been reported superior in performance to stand-alone models like the CNN-only model.Some remaining challenges have been discussed,including non-friendliness for finance domain experts,delayed prediction,domain knowledge negligence,lack of standards,and inability of real-time and highfrequency predictions.The principal contributions of this paper are to provide a one-stop guide for both academia and industry to review,compare and summarize technologies and recent advances in this area,to facilitate smooth and informed implementation,and to highlight future research directions.

Acute Complication Prediction and Diagnosis Model CLSTM-BPR:A Fusion Method of Time Series Deep Learning and Bayesian Personalized Ranking

Acute complication prediction model is of great importance for the overall reduction of premature death in chronic diseases.The CLSTM-BPR proposed in this paper aims to improve the accuracy,interpretability,and generalizability of the existing disease prediction models.Firstly,through its complex neural network structure,CLSTM-BPR considers both disease commonality and patient characteristics in the prediction process.Secondly,by splicing the time series prediction algorithm and classifier,the judgment basis is given along with the prediction results.Finally,this model introduces the pairwise algorithm Bayesian Personalized Ranking(BPR)into the medical field for the first time,and achieves a good result in the diagnosis of six acute complications.Experiments on the Medical Information Mart for Intensive Care IV(MIMIC-IV)dataset show that the average Mean Absolute Error(MAE)of biomarker value prediction of the CLSTM-BPR model is 0.26,and the average accuracy(ACC)of the CLSTM-BPR model for acute complication diagnosis is 92.5%.Comparison experiments and ablation experiments further demonstrate the reliability of CLSTM-BPR in the prediction of acute complication,which is an advancement of current disease prediction tools.

Prediction of discharge in a tidal river using the LSTM-based sequence-to-sequence models

The complexity of river-tide interaction poses a significant challenge in predicting discharge in tidal rivers.Long short-term memory(LSTM)networks excel in processing and predicting crucial events with extended intervals and time delays in time series data.Additionally,the sequence-to-sequence(Seq2Seq)model,known for handling temporal relationships,adapting to variable-length sequences,effectively capturing historical information,and accommodating various influencing factors,emerges as a robust and flexible tool in discharge forecasting.In this study,we introduce the application of LSTM-based Seq2Seq models for the first time in forecasting the discharge of a tidal reach of the Changjiang River(Yangtze River)Estuary.This study focuses on discharge forecasting using three key input characteristics:flow velocity,water level,and discharge,which means the structure of multiple input and single output is adopted.The experiment used the discharge data of the whole year of 2020,of which the first 80%is used as the training set,and the last 20%is used as the test set.This means that the data covers different tidal cycles,which helps to test the forecasting effect of different models in different tidal cycles and different runoff.The experimental results indicate that the proposed models demonstrate advantages in long-term,mid-term,and short-term discharge forecasting.The Seq2Seq models improved by 6%-60%and 5%-20%of the relative standard deviation compared to the harmonic analysis models and improved back propagation neural network models in discharge prediction,respectively.In addition,the relative accuracy of the Seq2Seq model is 1%to 3%higher than that of the LSTM model.Analytical assessment of the prediction errors shows that the Seq2Seq models are insensitive to the forecast lead time and they can capture characteristic values such as maximum flood tide flow and maximum ebb tide flow in the tidal cycle well.This indicates the significance of the Seq2Seq models.

The application of neural networks to comprehensive prediction by seismology prediction method

BP neural networks is used to mid-term earthquake prediction in this paper. Some usual prediction parameters of seismology are used as the import units of neural networks. And the export units of neural networks is called as the character parameter W_0 describing enhancement of seismicity. We applied this method to space scanning of North China. The result shows that the mid-term anomalous zone of W_0-value usually appeared obviously around the future epicenter 1~3 years before earthquake. It is effective to mid-term prediction.

Comprehensive Evaluation and Prediction of Enhancement of Boiling Heat Transfer with Additives

A model of evaluation and prediction of enhancement of boiling heat transfer with additives has been proposed according to fuzzy fundamentals. Correlative appraisement of boiling heat transfer augmentation was done with the model based on 39 additives which were tested by the authors and other researchers. The results show that the evaluation of 35 additives is consistent with experiments, which means that the accuracy of the model is 89.7 percent. In addition, the prediction of the ability of boiling heat transfer enhancement with sodium oleate, polyethylene glycol and Tween-40 is also in good agreement with correspondent experiments.

Comprehensive Approaches to Earthquake Prediction

There may be various anomalies from fields of multi-discipline before large earthquakes, including seismological anomalies and the anomalous changes in crustal deformation, underground fluids, geochemistry, geomagnetic fields, ground resistivity, etc. Precursors of a single discipline only reflect the process of earthquake preparation from one aspect. Only when various precursors from multi-disciplined fields are put into a unified process of earthquake preparation can the process be understood and identified comprehensively. The comprehensive methods used in earthquake prediction are outlined in this paper. The relationship that the forthcoming main shock bears with the precursory features such as the time-space distribution, the order of the precursors, etc. is analyzed. Research shows that a few of the main shocks can be predicted with the comprehensive methods to some extent.

A Comprehensive Price Prediction System Based on Inverse Multiquadrics Radial Basis Function for Portfolio Selection

Price prediction plays a crucial role in portfolio selection (PS). However, most price prediction strategies only make a single prediction and do not have efficient mechanisms to make a comprehensive price prediction. Here, we propose a comprehensive price prediction (CPP) system based on inverse multiquadrics (IMQ) radial basis function. First, the novel radial basis function (RBF) system based on IMQ function rather than traditional Gaussian (GA) function is proposed and centers on multiple price prediction strategies, aiming at improving the efficiency and robustness of price prediction. Under the novel RBF system, we then create a portfolio update strategy based on kernel and trace operator. To assess the system performance, extensive experiments are performed based on 4 data sets from different real-world financial markets. Interestingly, the experimental results reveal that the novel RBF system effectively realizes the integration of different strategies and CPP system outperforms other systems in investing performance and risk control, even considering a certain degree of transaction costs. Besides, CPP can calculate quickly, making it applicable for large-scale and time-limited financial market.

Landslide displacement prediction based on the ICEEMDAN,ApEn and the CNN-LSTM models

Landslide deformation is affected by its geological conditions and many environmental factors.So it has the characteristics of dynamic,nonlinear and unstable,which makes the prediction of landslide displacement difficult.In view of the above problems,this paper proposes a dynamic prediction model of landslide displacement based on the improvement of complete ensemble empirical mode decomposition with adaptive noise(ICEEMDAN),approximate entropy(ApEn)and convolution long short-term memory(CNN-LSTM)neural network.Firstly,ICEEMDAN and Ap En are used to decompose the cumulative displacements into trend,periodic and random displacements.Then,the least square quintic polynomial function is used to fit the displacement of trend term,and the CNN-LSTM is used to predict the displacement of periodic term and random term.Finally,the displacement prediction results of trend term,periodic term and random term are superimposed to obtain the cumulative displacement prediction value.The proposed model has been verified in Bazimen landslide in the Three Gorges Reservoir area of China.The experimental results show that the model proposed in this paper can more effectively predict the displacement changes of landslides.As compared with long short-term memory(LSTM)neural network,gated recurrent unit(GRU)network model and back propagation(BP)neural network,CNN-LSTM neural network had higher prediction accuracy in predicting the periodic displacement,with the mean absolute percentage error(MAPE)reduced by 3.621%,6.893% and 15.886% respectively,and the root mean square error(RMSE)reduced by 3.834 mm,3.945 mm and 7.422mm respectively.Conclusively,this model not only has high prediction accuracy but also is more stable,which can provide a new insight for practical landslide prevention and control engineering.

Traffic prediction enabled dynamic access points switching for energy saving in dense networks

To meet the ever-increasing traffic demand and enhance the coverage of cellular networks,network densification is one of the crucial paradigms of 5G and beyond mobile networks,which can improve system capacity by deploying a large number of Access Points(APs)in the service area.However,since the energy consumption of APs generally accounts for a substantial part of the communication system,how to deal with the consequent energy issue is a challenging task for a mobile network with densely deployed APs.In this paper,we propose an intelligent AP switching on/off scheme to reduce the system energy consumption with the prerequisite of guaranteeing the quality of service,where the signaling overhead is also taken into consideration to ensure the stability of the network.First,based on historical traffic data,a long short-term memory method is introduced to predict the future traffic distribution,by which we can roughly determine when the AP switching operation should be triggered;second,we present an efficient three-step AP selection strategy to determine which of the APs would be switched on or off;third,an AP switching scheme with a threshold is proposed to adjust the switching frequency so as to improve the stability of the system.Experiment results indicate that our proposed traffic forecasting method performs well in practical scenarios,where the normalized root mean square error is within 10%.Furthermore,the achieved energy-saving is more than 28% on average with a reasonable outage probability and switching frequency for an area served by 40 APs in a commercial mobile network.

A Short-Term Climate Prediction Model Based on a Modular Fuzzy Neural Network

In terms of the modular fuzzy neural network (MFNN) combining fuzzy c-mean (FCM) cluster and single-layer neural network, a short-term climate prediction model is developed. It is found from modeling results that the MFNN model for short-term climate prediction has advantages of simple structure, no hidden layer and stable network parameters because of the assembling of sound functions of the self-adaptive learning, association and fuzzy information processing of fuzzy mathematics and neural network methods. The case computational results of Guangxi flood season (JJA) rainfall show that the mean absolute error (MAE) and mean relative error (MRE) of the prediction during 1998-2002 are 68.8 mm and 9.78%, and in comparison with the regression method, under the conditions of the same predictors and period they are 97.8 mm and 12.28% respectively. Furthermore, it is also found from the stability analysis of the modular model that the change of the prediction results of independent samples with training times in the stably convergent interval of the model is less than 1.3 mm. The obvious oscillation phenomenon of prediction results with training times, such as in the common back-propagation neural network (BPNN) model, does not occur, indicating a better practical application potential of the MFNN model.

An Approach for Improving Short-Term Prediction of Summer Rainfall over North China by Decomposing Interannual and Decadal Variability

A statistical downscaling approach was developed to improve seasonal-to-interannual prediction of summer rainfall over North China by considering the effect of decadal variability based on observational datasets and dynamical model outputs.Both predictands and predictors were first decomposed into interannual and decadal components.Two predictive equations were then built separately for the two distinct timescales by using multivariate linear regressions based on independent sample validation.For the interannual timescale,850-hPa meridional wind and 500-hPa geopotential heights from multiple dynamical models＇ hindcasts and SSTs from observational datasets were used to construct predictors.For the decadal timescale,two well-known basin-scale SST decadal oscillation （the Atlantic Multidecadal Oscillation and the Pacific Decadal Oscillation） indices were used as predictors.Then,the downscaled predictands were combined to represent the predicted/hindcasted total rainfall.The prediction was compared with the models＇ raw hindcasts and those from a similar approach but without timescale decomposition.In comparison to hindcasts from individual models or their multi-model ensemble mean,the skill of the present scheme was found to be significantly higher,with anomaly correlation coefficients increasing from nearly neutral to over 0.4 and with RMSE decreasing by up to 0.6 mm d-1.The improvements were also seen in the station-based temporal correlation of the predictions with observed rainfall,with the coefficients ranging from-0.1 to 0.87,obviously higher than the models＇ raw hindcasted rainfall results.Thus,the present approach exhibits a great advantage and may be appropriate for use in operational predictions.

A real-time prediction method for tunnel boring machine cutter-head torque using bidirectional long short-term memory networks optimized by multi-algorithm

Based on data from the Jilin Water Diversion Tunnels from the Songhua River(China),an improved and real-time prediction method optimized by multi-algorithm for tunnel boring machine(TBM)cutter-head torque is presented.Firstly,a function excluding invalid and abnormal data is established to distinguish TBM operating state,and a feature selection method based on the SelectKBest algorithm is proposed.Accordingly,ten features that are most closely related to the cutter-head torque are selected as input variables,which,in descending order of influence,include the sum of motor torque,cutter-head power,sum of motor power,sum of motor current,advance rate,cutter-head pressure,total thrust force,penetration rate,cutter-head rotational velocity,and field penetration index.Secondly,a real-time cutterhead torque prediction model’s structure is developed,based on the bidirectional long short-term memory(BLSTM)network integrating the dropout algorithm to prevent overfitting.Then,an algorithm to optimize hyperparameters of model based on Bayesian and cross-validation is proposed.Early stopping and checkpoint algorithms are integrated to optimize the training process.Finally,a BLSTMbased real-time cutter-head torque prediction model is developed,which fully utilizes the previous time-series tunneling information.The mean absolute percentage error(MAPE)of the model in the verification section is 7.3%,implying that the presented model is suitable for real-time cutter-head torque prediction.Furthermore,an incremental learning method based on the above base model is introduced to improve the adaptability of the model during the TBM tunneling.Comparison of the prediction performance between the base and incremental learning models in the same tunneling section shows that:(1)the MAPE of the predicted results of the BLSTM-based real-time cutter-head torque prediction model remains below 10%,and both the coefficient of determination(R^(2))and correlation coefficient(r)between measured and predicted values exceed 0.95;and(2)the incremental learning method is suitable for realtime cutter-head torque prediction and can effectively improve the prediction accuracy and generalization capacity of the model during the excavation process.