A Correction Method Suitable for Dynamical Seasonal Prediction

Based on the hindcast results of summer rainfall anomalies over China for the period 1981-2000 by the Dynamical Climate Prediction System （IAP-DCP） developed by the Institute of Atmospheric Physics, a correction method that can account for the dependence of model＇s systematic biases on SST anomalies is proposed. It is shown that this correction method can improve the hindcast skill of the IAP-DCP for summer rainfall anomalies over China, especially in western China and southeast China, which may imply its potential application to real-time seasonal prediction.

Systematic Errors of Zonal-Mean Flow in Dynamical Monthly Prediction and Its Improvement

An analysis of a large number of cases of 500 hPa height monthly prediction shows that systematic errors exist in the zonal mean components which account for a large portion of the total forecast errors, and such errors are commonly seen in other prediction models. To overcome the difficulties of the numerical model, the authors attempt a 'hybrid' approach to improving the dynamical extended-range (monthly) prediction. The monthly pentad-mean nonlinear dynamical regional prediction model of the zonal-mean geopotential height (wave number 0) based on a large amount of data is constituted by employing the reconstruction of phase-space theory and the spatio-temporal series predictive method. The dynamical prediction of the numerical model is then combined with that of the nonlinear model, i.e., the pentadmean zonal-mean height produced by the nonlinear model is transformed to its counterpart in the numerical model by nudging during the time integration. The forecast experiment results show that the above hybrid approach not only reduces the systematic error in zonal mean height by the numerical model, but also makes an improvement in the non-axisymmetric components due to the wave-flow interaction.

Preliminary Studies on Predicting the Tropical Indian Ocean Sea Surface Temperature through Combined Statistical Methods and Dynamic ENSO Prediction

The sea surface temperature （SST） in the In- dian Ocean affects the regional climate over the Asian continent mostly through a modulation of the monsoon system. It is still difficult to provide an a priori indication of the seasonal variability over the Indian Ocean. It is widely recognized that the warm and cold events of SST over the tropical Indian Ocean are strongly linked to those of the equatorial eastern Pacific. In this study, a statistical prediction model has been developed to predict the monthly SST over the tropical Indian Ocean. This model is a linear regression model based on the lag relationship between the SST over the tropical Indian Ocean and the Nino3.4 （5°S-5°N, 170°W-120°W） SST Index. The pre- dictor （i.e., Nino3.4 SST Index） has been operationally predicted by a large size ensemble E1 Nifio and the Southern Oscillation （ENSO） forecast system with cou- pled data assimilation （Leefs_CDA）, which achieves a high predictive skill of up to a 24-month lead time for the equatorial eastern Pacific SST. As a result, the prediction skill of the present statistical model over the tropical In- dian Ocean is better than that of persistence prediction for January 1982 through December 2009.

Comparative Analysis of ARIMA and LSTM Model-Based Anomaly Detection for Unannotated Structural Health Monitoring Data in an Immersed Tunnel

Structural Health Monitoring(SHM)systems have become a crucial tool for the operational management of long tunnels.For immersed tunnels exposed to both traffic loads and the effects of the marine environment,efficiently identifying abnormal conditions from the extensive unannotated SHM data presents a significant challenge.This study proposed amodel-based approach for anomaly detection and conducted validation and comparative analysis of two distinct temporal predictive models using SHM data from a real immersed tunnel.Firstly,a dynamic predictive model-based anomaly detectionmethod is proposed,which utilizes a rolling time window for modeling to achieve dynamic prediction.Leveraging the assumption of temporal data similarity,an interval prediction value deviation was employed to determine the abnormality of the data.Subsequently,dynamic predictive models were constructed based on the Autoregressive Integrated Moving Average(ARIMA)and Long Short-Term Memory(LSTM)models.The hyperparameters of these models were optimized and selected using monitoring data from the immersed tunnel,yielding viable static and dynamic predictive models.Finally,the models were applied within the same segment of SHM data,to validate the effectiveness of the anomaly detection approach based on dynamic predictive modeling.A detailed comparative analysis discusses the discrepancies in temporal anomaly detection between the ARIMA-and LSTM-based models.The results demonstrated that the dynamic predictive modelbased anomaly detection approach was effective for dealing with unannotated SHM data.In a comparison between ARIMA and LSTM,it was found that ARIMA demonstrated higher modeling efficiency,rendering it suitable for short-term predictions.In contrast,the LSTM model exhibited greater capacity to capture long-term performance trends and enhanced early warning capabilities,thereby resulting in superior overall performance.

Research on Quantitative Identification of Three-Dimensional Connectivity of Fractured-Vuggy Reservoirs

The fractured-vuggy carbonate oil resources in the western basin of China are extremely rich.The connectivity of carbonate reservoirs is complex,and there is still a lack of clear understanding of the development and topological structure of the pore space in fractured-vuggy reservoirs.Thus,effective prediction of fractured-vuggy reservoirs is difficult.In view of this,this work employs adaptive point cloud technology to reproduce the shape and capture the characteristics of a fractured-vuggy reservoir.To identify the complex connectivity among pores,fractures,and vugs,a simplified one-dimensional connectivity model is established by using the meshless connection element method(CEM).Considering that different types of connection units have different flow characteristics,a sequential coupling calculation method that can efficiently calculate reservoir pressure and saturation is developed.By automatic history matching,the dynamic production data is fitted in real-time,and the characteristic parameters of the connection unit are inverted.Simulation results show that the three-dimensional connectivity model of the fractured-vuggy reservoir built in this work is as close as 90%of the fine grid model,while the dynamic simulation efficiency is much higher with good accuracy.

Dynamic prediction of gas emission based on wavelet neural network toolbox

This paper presents a method for dynamically predicting gas emission quantity based on the wavelet neural network （WNN） toolbox. Such a method is able to predict the gas emission quantity in adjacent subsequent time intervals through training the WNN with even time-interval samples. The method builds successive new model with the width of sliding window remaining invariable so as to obtain a dynamic prediction method for gas emission quantity. Furthermore, the method performs prediction by a self-developed WNN toolbox. Experiments indicate that such a model can overcome the deficiencies of the traditional static prediction model and can fully make use of the feature extraction capability of wavelet base function to reflect the geological feature of gas emission quantity dynamically. The method is characterized by simplicity, flexibility, small data scale, fast convergence rate and high prediction precision. In addition, the method is also characterized by certainty and repeatability of the predicted results. The effectiveness of this method is confirmed by simulation results. Therefore, this method will exert practical significance on promoting the application of WNN.

Dynamic prediction of landslide displacement using singular spectrum analysis and stack long short-term memory network

An accurate landslide displacement prediction is an important part of landslide warning system. Aiming at the dynamic characteristics of landslide evolution and the shortcomings of traditional static prediction models, this paper proposes a dynamic prediction model of landslide displacement based on singular spectrum analysis(SSA) and stack long short-term memory(SLSTM) network. The SSA is used to decompose the landslide accumulated displacement time series data into trend term and periodic term displacement subsequences. A cubic polynomial function is used to predict the trend term displacement subsequence, and the SLSTM neural network is used to predict the periodic term displacement subsequence. At the same time, the Bayesian optimization algorithm is used to determine that the SLSTM network input sequence length is 12 and the number of hidden layer nodes is 18. The SLSTM network is updated by adding predicted values to the training set to achieve dynamic displacement prediction. Finally, the accumulated landslide displacement is obtained by superimposing the predicted value of each displacement subsequence. The proposed model was verified on the Xintan landslide in Hubei Province, China. The results show that when predicting the displacement of the periodic term, the SLSTM network has higher prediction accuracy than the support vector machine(SVM) and auto regressive integrated moving average(ARIMA). The mean relative error(MRE) is reduced by 4.099% and 3.548% respectively, while the root mean square error(RMSE) is reduced by 5.830 mm and 3.854 mm respectively. It is concluded that the SLSTM network model can better simulate the dynamic characteristics of landslides.

Time series prediction of mining subsidence based on a SVM

In order to study dynamic laws of surface movements over coal mines due to mining activities,a dynamic prediction model of surface movements was established,based on the theory of support vector machines(SVM) and times-series analysis.An engineering application was used to verify the correctness of the model.Measurements from observation stations were analyzed and processed to obtain equal-time interval surface movement data and subjected to tests of stationary,zero means and normality.Then the data were used to train the SVM model.A time series model was established to predict mining subsidence by rational choices of embedding dimensions and SVM parameters.MAPE and WIA were used as indicators to evaluate the accuracy of the model and for generalization performance.In the end,the model was used to predict future surface movements.Data from observation stations in Huaibei coal mining area were used as an example.The results show that the maximum absolute error of subsidence is 9 mm,the maximum relative error 1.5%,the maximum absolute error of displacement 7 mm and the maximum relative error 1.8%.The accuracy and reliability of the model meet the requirements of on-site engineering.The results of the study provide a new approach to investigate the dynamics of surface movements.

New multicast scheme based on Dynamic mobility prediction in mobile IPv6 environment

A new mobile multicast scheme called mobility prediction based mobile multicast(MPBMM) was proposed. In MPBMM, when a mobile node (MN) roams among subnets during a multicast session, MN predicts the next subnet, to which MN will attach, by the information of its position and mobility speed, consequently speeds up the handoff procedure. Simulation results show that the proposed scheme can minimize the loss of multicast packets, reduce the delay of subnet handoff, decrease the frequency of multicast tree reconfiguration, and optimize the delivery path of multicast packets. When MN moves among subnets at different speeds (from 5 to 25 ms), the maximum loss ratio of multicast packets is less than0.2%, the maximum inter-arrival time of multicast packets is 117 ms, so the proposed scheme can meet the QoS requirements of real-time services. In addition, MPBMM can support the mobility of multicast source.

Novel Hybrid Physics‑Informed Deep Neural Network for Dynamic Load Prediction of Electric Cable Shovel

Electric cable shovel(ECS)is a complex production equipment,which is widely utilized in open-pit mines.Rational valuations of load is the foundation for the development of intelligent or unmanned ECS,since it directly influences the planning of digging trajectories and energy consumption.Load prediction of ECS mainly consists of two types of methods:physics-based modeling and data-driven methods.The former approach is based on known physical laws,usually,it is necessarily approximations of reality due to incomplete knowledge of certain processes,which introduces bias.The latter captures features/patterns from data in an end-to-end manner without dwelling on domain expertise but requires a large amount of accurately labeled data to achieve generalization,which introduces variance.In addition,some parts of load are non-observable and latent,which cannot be measured from actual system sensing,so they can’t be predicted by data-driven methods.Herein,an innovative hybrid physics-informed deep neural network(HPINN)architecture,which combines physics-based models and data-driven methods to predict dynamic load of ECS,is presented.In the proposed framework,some parts of the theoretical model are incorporated,while capturing the difficult-to-model part by training a highly expressive approximator with data.Prior physics knowledge,such as Lagrangian mechanics and the conservation of energy,is considered extra constraints,and embedded in the overall loss function to enforce model training in a feasible solution space.The satisfactory performance of the proposed framework is verified through both synthetic and actual measurement dataset.

Protein Secondary Structure Prediction with Dynamic Self-Adaptation Combination Strategy Based on Entropy

The algorithm based on combination learning usually is superior to a singleclassification algorithm on the task of protein secondary structure prediction. However,the assignment of the weight of the base classifier usually lacks decision-makingevidence. In this paper, we propose a protein secondary structure prediction method withdynamic self-adaptation combination strategy based on entropy, where the weights areassigned according to the entropy of posterior probabilities outputted by base classifiers.The higher entropy value means a lower weight for the base classifier. The final structureprediction is decided by the weighted combination of posterior probabilities. Extensiveexperiments on CB513 dataset demonstrates that the proposed method outperforms theexisting methods, which can effectively improve the prediction performance.

On numerical earthquake prediction

Can earthquakes be predicted？ How should people overcome the difficulties encountered in the study of earthquake prediction？ This issue can take inspiration from the experiences of weather forecast. Although weather forecasting took a period of about half a century to advance from empirical to numerical forecast, it has achieved significant success. A consensus has been reached among the Chinese seismological community that earth- quake prediction must also develop from empirical fore- casting to physical prediction. However, it is seldom mentioned that physical prediction is characterized by quantitatively numerical predictions based on physical laws. This article discusses five key components for numerical earthquake prediction and their current status. We conclude that numerical earthquake prediction should now be put on the planning agenda and its roadmap designed, seismic stations should be deployed and observations made according to the needs of numerical prediction, and theoretical research should be carried out.

Dynamic environmentprediction of spacecraft usingdynamic substructuring

A low frequency dynamic environment prediction of spacecraft using dynamic substructu- ring is presented. The dynamic environment could be used to describe the level of the excitation on the spacecraft itself and auxiliary equipment. In addition, the dynamic environment is a criterion for the structural dynamic design as well as the ground verification test. The proposed prediction method could solve two major problems. The first is the time consumption of analyzing the whole spacecraft model due to the huge amount of degrees of freedom, and the second is multi-source for component structural dynamic models from distributive departments. To demonstrate the feasibility and efficien- cy, the proposed prediction method is applied to resolve a launching satellite case, and the results were compared with those obtained by the traditional prediction technology using the finite element method.

Improvement of System Dynamic Precision by Dynamic Prediction

According to the characteristic of the sensor inertia, the dynamic prediction to improve the system dynamic precision is presented in this paper. With the recurrence calculation of time constant of the sensor, the system dynamic precision is greatly improved. The example using this method is given.

Production Dynamic Prediction Method of Waterflooding Reservoir Based on Deep Convolution Generative Adversarial Network(DC-GAN)

The rapid production dynamic prediction of water-flooding reservoirs based on well location deployment has been the basis of production optimization of water-flooding reservoirs.Considering that the construction of geological models with traditional numerical simulation software is complicated,the computational efficiency of the simulation calculation is often low,and the numerical simulation tools need to be repeated iteratively in the process of model optimization,machine learning methods have been used for fast reservoir simulation.However,traditional artificial neural network(ANN)has large degrees of freedom,slow convergence speed,and complex network model.This paper aims to predict the production performance of water flooding reservoirs based on a deep convolutional generative adversarial network(DC-GAN)model,and establish a dynamic mapping relationship between well location deployment and output oil saturation.The network structure is based on an improved U-Net framework.Through a deep convolutional network and deconvolution network,the features of input well deployment images are extracted,and the stability of the adversarial model is strengthened.The training speed and accuracy of the proxy model are improved,and the oil saturation of water flooding reservoirs is dynamically predicted.The results show that the trained DC-GAN has significant advantages in predicting oil saturation by the well-location employment map.The cosine similarity between the oil saturation map given by the trained DC-GAN and the oil saturation map generated by the numerical simulator is compared.In above,DC-GAN is an effective method to conduct a proxy model to quickly predict the production performance of water flooding reservoirs.

Dynamic prediction of over-excavation gap due to posture adjustment of shield machine in soft soil

The probability analysis of ground deformation is becoming a trend to estimate and control the risk brought by shield tunnelling.The gap parameter is regarded as an effective tool to estimate the ground loss of tunnelling in soft soil.More specifically,x,which is a gap parameter component defined as the over(or insufficient)excavation due to the change in the posture of the shield machine,may contribute more to the uncertainty of the ground loss.However,the existing uncertainty characterization methods for x have several limitations and cannot explain the uncertain correlations between the relevant parameters.Along these lines,to better characterize the uncertainty of x,the multivariate probability distribution was developed in this work and a dynamic prediction was proposed for it.To attain this goal,1523 rings of the field data coming from the shield tunnel between Longqing Road and Baiyun Road in Kunming Metro Line 5 were utilized and 44 parameters including the construction,stratigraphic,and posture parameters were collected to form the database.According to the variance filter method,the mutual information method,and the value of the correlation coefficients,the original 44 parameters were reduced to 10 main parameters,which were unit weight,the stoke of the jacks(A,B,C,and D groups),the pressure of the pushing jacks(A,C groups),the chamber pressure,the rotation speed,and the total force.The multivariate probability distribution was constructed based on the Johnson system of distributions.Moreover,the distribution was satisfactorily verified in explaining the pairwise correlation between x and other parameters through 2 million simulation cases.At last,the distribution was used as a prior distribution to update the marginal distribution of x with any group of the relevant parameters known.The performance of the dynamic prediction was further validated by the field data of 3 shield tunnel cases.

Energy consumption dynamic prediction for HVAC systems based on feature clustering deconstruction and model training adaptation

The prediction of building energy consumption offers essential technical support for intelligent operation and maintenance of buildings,promoting energy conservation and low-carbon control.This paper focused on the energy consumption of heating,ventilation and air conditioning(HVAC)systems operating under various modes across different seasons.We constructed multi-attribute and high-dimensional clustering vectors that encompass indoor and outdoor environmental parameters,along with historical energy consumption data.To enhance the K-means algorithm,we employed statistical feature extraction and dimensional normalization(SFEDN)to facilitate data clustering and deconstruction.This method,combined with the gated recurrent unit(GRU)prediction model employing adaptive training based on the Particle Swarm Optimization algorithm,was evaluated for robustness and stability through k-fold cross-validation.Within the clustering-based modeling framework,optimal submodels were configured based on the statistical features of historical 24-hour data to achieve dynamic prediction using multiple models.The dynamic prediction models with SFEDN cluster showed a 11.9%reduction in root mean square error(RMSE)compared to static prediction,achieving a coefficient of determination(R2)of 0.890 and a mean absolute percentage error(MAPE)reduction of 19.9%.When compared to dynamic prediction based on single-attribute of HVAC systems energy consumption clustering modeling,RMSE decreased by 12.6%,R2 increased by 4.0%,and MAPE decreased by 26.3%.The dynamic prediction performance demonstrated that the SFEDN clustering method surpasses conventional clustering method,and multi-attribute clustering modeling outperforms single-attribute modeling.

Dynamic prediction of landslide life expectancy using ensemble system incorporating classical prediction models and machine learning

With the development of landslide monitoring system,many attempts have been made to predict landslide failure-time utilizing monitoring data of displacements.Classical models(e.g.,Verhulst,GM(1,1),and Saito models)that consider the characteristics of landslide displacement to determine the failuretime have been investigated extensively.In practice,monitoring is continuously implemented with monitoring data-set updated,meaning that the predicted landslide life expectancy(i.e.,the lag between the predicted failure-time and time node at each instant of conducting the prediction)should be re-evaluated with time.This manner is termed“dynamic prediction”.However,the performances of the classical models have not been discussed in the context of the dynamic prediction yet.In this study,such performances are investigated firstly,and disadvantages of the classical models are then reported,incorporating the monitoring data from four real landslides.Subsequently,a more qualified ensemble model is proposed,where the individual classical models are integrated by machine learning(ML)-based meta-model.To evaluate the quality of the models under the dynamic prediction,a novel indicator termed“discredit index(b)”is proposed,and a higher value of b indicates lower prediction quality.It is found that Verhulst and Saito models would produce predicted results with significantly higher b,while GM(1,1)model would indicate results with the highest mean absolute error.Meanwhile,the ensemble models are found to be more accurate and qualified than the classical models.Here,the performance of decision tree regression-based ensemble model is the best among the various ML-based ensemble models.

THE MONTHLY PREDICTION EXPERIMENTS USING A COUPLED ANALOGY-DYNAMICAL MODEL

Considering from point of view of the dynamics,it is convenient to regard the field to be predicted as a small disturbance superposed on the historical analogous field,and thus the statistical technique can be used in combining with the dynamics.Along this line,a coupled atmosphere-earth surface analogy-dynamical model is formulated and applied to making monthly prediction. This approach facilitated the utility of the useful information contained in both the historical data set and the initial field to improve the dynamic model based solo on the latter and show better skill in prediction.

Dynamic recrystallization behavior of burn resistant titanium alloy Ti-25V-15Cr-0.2Si

Dynamic recrystallization (DRX) behavior in β phase region for the burn resistant titanium alloy Ti?25V?15Cr?0.2Si was investigated with a compression test in the temperature range of 950?1100 °C and the strain rate of 0.001?1 s?1. The results show that deformation mechanism of this alloy in hot deformation is dominated by DRX, and new grains of DRX are evolved by bulging nucleation mechanism as a predominant mechanism. DRX occurs more easily with the decrease of strain rate and the increase of deformation temperature. Grain refinement is achieved due to DRX during the hot deformation at strain rate range of 0.01?0.1 s?1 and temperature range of 950?1050 °C. DRX grain coarsening is observed for the alloy deformed at the higher temperatures of 1100 °C and the lower strain rates of 0.001 s?1. Finally, in order to determine the recrystallized fraction and DRX grain size under different deformation conditions, the prediction models of recrystallization kinetics and recrystallized grain sizes were established.