scholarly journals Medium- and long-term runoff forecasting based on a random forest regression model

2020 ◽  
Vol 20 (8) ◽  
pp. 3658-3664
Author(s):  
Chen Shijun ◽  
Wei Qin ◽  
Zhu Yanmei ◽  
Ma Guangwen ◽  
Han Xiaoyan ◽  
...  
Keyword(s):  
Random Forest ◽  
Power Plants ◽  
Moving Average ◽  
Support Vector ◽  
Random Forest Regression ◽  
Autoregressive Moving Average Model ◽  
Runoff Forecasting ◽  
Svm Model ◽  
Moving Average Model

Abstract Medium- and long-term runoff forecasting is closely related to the generation capacity forecasting of cascade hydropower stations, which is of great significance to power plants when arranging production plans and assisting market decisions. In order to improve the accuracy of runoff forecasting, an attempt was made to use random forest regression (RFR) to model the medium- and long-term runoff forecasting and to further make a verification based on the actual monthly runoff data of Mupo and Chuntangba stations. By comparison with the forecast results attained through a support vector machine (SVM) and an integrated autoregressive moving average model (IARMA), the results showed that the RFR model had the lowest mean square error (MSE) among the three methods. In addition, the coefficients of determination R2 of the RFR for the two stations increased by 0.0261 and 0.0295 compared with the SVM model, and the R2 rose by 0.1134 and 0.1332 compared with the IARMA model. The comparison of the three methods showed that the RFR had higher forecasting accuracy as well as stronger reliability and practicability than the IARMA model and the SVM model, so the RFR provided a new idea and method for the study of runoff forecasting.

scholarly journals Mold Level Predict of Continuous Casting Using Hybrid EMD-SVR-GA Algorithm

Processes ◽  
2019 ◽  
Vol 7 (3) ◽  
pp. 177 ◽  
Author(s):  
Zhufeng Lei ◽  
Wenbin Su
Keyword(s):  
Continuous Casting ◽  
Production Control ◽  
Moving Average ◽  
Time Series Prediction ◽  
Casting Process ◽  
Support Vector ◽  
Intrinsic Mode Functions ◽  
Autoregressive Moving Average Model ◽  
Mode Decomposition ◽  
Moving Average Model

The prediction of mold level is a basic and key problem of continuous casting production control. Many current techniques fail to predict the mold level because of mold level is non-linear, non-stationary and does not have a normal distribution. A hybrid model, based on empirical mode decomposition (EMD) and support vector regression (SVR), is proposed to solve the mold level in this paper. Firstly, the EMD algorithm, with adaptive decomposition, is used to decompose the original mold level signal to many intrinsic mode functions (IMFs). Then, the SVR model optimized by genetic algorithm (GA) is used to predict the IMFs and residual sequences. Finally, the equalization of the predict results is reconstructed to obtain the predict result. Several hybrid predicting methods such as EMD and autoregressive moving average model (ARMA), EMD and SVR, wavelet transform (WT) and ARMA, WT and SVR are discussed and compared in this paper. These methods are applied to mold level prediction, the experimental results show that the proposed hybrid method based on EMD and SVR is a powerful tool for solving complex time series prediction. In view of the excellent generalization ability of the EMD, it is believed that the hybrid algorithm of EMD and SVR is the best model for mold level predict among the six methods, providing a new idea for guiding continuous casting process improvement.

scholarly journals Data-driven runoff forecasting for Minjiang River: a case study

2020 ◽  
Vol 20 (6) ◽  
pp. 2284-2295
Author(s):  
Yuqiang Wu ◽  
Qinhui Wang ◽  
Ge Li ◽  
Jidong Li
Keyword(s):  
Optimal Allocation ◽  
Data Driven ◽  
Function Parameter ◽  
Support Vector ◽  
Runoff Forecasting ◽  
Minjiang River ◽  
Svm Model ◽  
Runoff Prediction

Abstract Long-term runoff forecasting has the characteristics of a long forecast period, which can be widely applied in environmental protection, hydropower operation, flood prevention and waterlogging management, water transport management, and optimal allocation of water resources. Many models and methods are currently used for runoff prediction, and data-driven models for runoff prediction are now mainstream methods, but their prediction accuracy cannot meet the needs of production departments. To this end, the present research starts with this method and, based on a support vector machine (SVM), it introduces ant colony optimization (ACO) to optimize its penalty coefficient C, Kernel function parameter g, and insensitivity coefficient p, to construct a data-driven ACO-SVM model. The validity of the method is confirmed by taking the Minjiang River Basin as an example. The results show that the runoff predicted by use of ACO-SVM is more accurate than that of the default parameter SVM and the Bayesian method.

Application Research of Support Vector Machine Based on Particle Swarm Optimization in Runoff Forecasting

2012 ◽  
Vol 226-228 ◽  
pp. 2303-2307
Author(s):  
Li Xue Wang ◽  
Li Na Wang ◽  
Guo Feng Li ◽  
Ce Luan ◽  
Fei Fei Sun
Keyword(s):  
Support Vector Machine ◽  
Optimization Procedure ◽  
Support Vector ◽  
Application Research ◽  
Generalization Capability ◽  
Runoff Forecasting ◽  
Svm Model ◽  
Basic Optimization ◽  
Monthly Runoff

In view of the little sample, less data problems, mid-and-long term hydrologic forecasting is a case of which, Support Vector Machine (SVM) can solve this kind of problems perfectly. This paper introduced the basic optimization procedure and PSO-SVM modeling procedure. The PSO-SVM model has been applied in forecasting the monthly runoff of Dahuofang reservoir. The comparison between PSO-SVM and not-optimized SVM implied that the PSO-SVM has a fast convergence speed and strong generalization capability, also the related error has been decreased from 15.5% to 11.9%.

scholarly journals Multi-Channel Singular Spectrum Analysis on Geocenter Motion and Its Precise Prediction

Sensors ◽  
2021 ◽  
Vol 21 (4) ◽  
pp. 1403
Author(s):  
Xin Jin ◽  
Xin Liu ◽  
Jinyun Guo ◽  
Yi Shen
Keyword(s):  
Solid Earth ◽  
Spectrum Analysis ◽  
Mean Squared Error ◽  
Moving Average ◽  
Singular Spectrum Analysis ◽  
Singular Spectrum ◽  
Autoregressive Moving Average Model ◽  
Seasonal Characteristics ◽  
Geocenter Motion ◽  
Moving Average Model

Geocenter is the center of the mass of the Earth system including the solid Earth, ocean, and atmosphere. The time-varying characteristics of geocenter motion (GCM) reflect the redistribution of the Earth’s mass and the interaction between solid Earth and mass loading. Multi-channel singular spectrum analysis (MSSA) was introduced to analyze the GCM products determined from satellite laser ranging data released by the Center for Space Research through January 1993 to February 2017 for extracting the periods and the long-term trend of GCM. The results show that the GCM has obvious seasonal characteristics of the annual, semiannual, quasi-0.6-year, and quasi-1.5-year in the X, Y, and Z directions, the annual characteristics make great domination, and its amplitudes are 1.7, 2.8, and 4.4 mm, respectively. It also shows long-period terms of 6.09 years as well as the non-linear trends of 0.05, 0.04, and –0.10 mm/yr in the three directions, respectively. To obtain real-time GCM parameters, the MSSA method combining a linear model (LM) and autoregressive moving average model (ARMA) was applied to predict GCM for 2 years into the future. The precision of predictions made using the proposed model was evaluated by the root mean squared error (RMSE). The results show that the proposed method can effectively predict GCM parameters, and the prediction precision in the three directions is 1.53, 1.08, and 3.46 mm, respectively.

scholarly journals Analysis and prediction of polar motion using MSSA method

2021 ◽  
Vol 73 (1) ◽  
Author(s):  
Xin Jin ◽  
Xin Liu ◽  
Jinyun Guo ◽  
Yi Shen
Keyword(s):  
Polar Motion ◽  
Rotation Axis ◽  
Moving Average ◽  
Singular Spectrum Analysis ◽  
Autoregressive Moving Average ◽  
Rotational Axis ◽  
Autoregressive Moving Average Model ◽  
System Service ◽  
Moving Average Model ◽  
Polar Motion Prediction

AbstractPolar motion is the movement of the Earth's rotational axis relative to its crust, reflecting the influence of the material exchange and mass redistribution of each layer of the Earth on the Earth's rotation axis. To better analyze the temporally varying characteristics of polar motion, multi-channel singular spectrum analysis (MSSA) was used to analyze the EOP 14 C04 series released by the International Earth Rotation and Reference System Service (IERS) from 1962 to 2020, and the amplitude of the Chandler wobbles were found to fluctuate between 20 and 200 mas and decrease significantly over the last 20 years. The amplitude of annual oscillation fluctuated between 60 and 120 mas, and the long-term trend was 3.72 mas/year, moving towards N56.79 °W. To improve prediction of polar motion, the MSSA method combining linear model and autoregressive moving average model was used to predict polar motion with ahead 1 year, repeatedly. Comparing to predictions of IERS Bulletin A, the results show that the proposed method can effectively predict polar motion, and the improvement rates of polar motion prediction for 365 days into the future were approximately 50% on average.

scholarly journals Empirical methods for the estimation of Southern Ocean CO<sub>2</sub>: support vector and random forest regression

2017 ◽  
Vol 14 (23) ◽  
pp. 5551-5569 ◽  
Author(s):  
Luke Gregor ◽  
Schalk Kok ◽  
Pedro M. S. Monteiro
Keyword(s):  
Random Forest ◽  
Southern Ocean ◽  
Synthetic Data ◽  
Good Effect ◽  
Support Vector ◽  
Co2 Uptake ◽  
Empirical Methods ◽  
Random Forest Regression ◽  
Proxy Variables ◽  
The Impact

Abstract. The Southern Ocean accounts for 40 % of oceanic CO2 uptake, but the estimates are bound by large uncertainties due to a paucity in observations. Gap-filling empirical methods have been used to good effect to approximate pCO2 from satellite observable variables in other parts of the ocean, but many of these methods are not in agreement in the Southern Ocean. In this study we propose two additional methods that perform well in the Southern Ocean: support vector regression (SVR) and random forest regression (RFR). The methods are used to estimate ΔpCO2 in the Southern Ocean based on SOCAT v3, achieving similar trends to the SOM-FFN method by Landschützer et al. (2014). Results show that the SOM-FFN and RFR approaches have RMSEs of similar magnitude (14.84 and 16.45 µatm, where 1 atm  =  101 325 Pa) where the SVR method has a larger RMSE (24.40 µatm). However, the larger errors for SVR and RFR are, in part, due to an increase in coastal observations from SOCAT v2 to v3, where the SOM-FFN method used v2 data. The success of both SOM-FFN and RFR depends on the ability to adapt to different modes of variability. The SOM-FFN achieves this by having independent regression models for each cluster, while this flexibility is intrinsic to the RFR method. Analyses of the estimates shows that the SVR and RFR's respective sensitivity and robustness to outliers define the outcome significantly. Further analyses on the methods were performed by using a synthetic dataset to assess the following: which method (RFR or SVR) has the best performance? What is the effect of using time, latitude and longitude as proxy variables on ΔpCO2? What is the impact of the sampling bias in the SOCAT v3 dataset on the estimates? We find that while RFR is indeed better than SVR, the ensemble of the two methods outperforms either one, due to complementary strengths and weaknesses of the methods. Results also show that for the RFR and SVR implementations, it is better to include coordinates as proxy variables as RMSE scores are lowered and the phasing of the seasonal cycle is more accurate. Lastly, we show that there is only a weak bias due to undersampling. The synthetic data provide a useful framework to test methods in regions of sparse data coverage and show potential as a useful tool to evaluate methods in future studies.

scholarly journals A Set of State–Space Models at a High Disaggregation Level to Forecast Italian Industrial Production

J ◽  
2019 ◽  
Vol 2 (4) ◽  
pp. 508-560
Author(s):  
Riccardo Corradini
Keyword(s):  
State Space ◽  
Industrial Production ◽  
Forecast Error ◽  
Moving Average ◽  
Predictive Ability ◽  
State Space Models ◽  
Vector Autoregressive ◽  
Autoregressive Moving Average Model ◽  
New Strategy ◽  
Moving Average Model

Normally, econometric models that forecast the Italian Industrial Production Index do not exploit information already available at time t + 1 for their own main industry groupings. The new strategy proposed here uses state–space models and aggregates the estimates to obtain improved results. The performance of disaggregated models is compared at the same time with a popular benchmark model, a univariate model tailored on the whole index, with persistent not formally registered holidays, a vector autoregressive moving average model exploiting all information published on the web for main industry groupings. Tests for superior predictive ability confirm the supremacy of the aggregated forecasts over three steps horizon using absolute forecast error and quadratic forecast error as a loss function. The datasets are available online.

Sign in / Sign up

Export Citation Format

Share Document