Multi-variable Echo State Network Optimized by Bayesian Regulation for Daily Peak Load Forecasting

Niu, Dongxiao; Ji, Ling; Xing, Mian; Wang, Jianjun

doi:10.4304/jnw.7.11.1790-1795

Cited by 6 publications

(6 citation statements)

References 16 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Ideally, the structure of the network will be simplified while the accuracy remains. Thus, the generalization ability can be improved [21]. We only penalize the output weights, since the input weights and cycle connection weights have already been pruned when the highly constrained reservoir is designed.…”

Section: B Readout Regularizationmentioning

confidence: 99%

Learning the deterministically constructed Echo State Networks

Tang

Tiňo

Chen

2014

2014 International Joint Conference on Neural Networks (IJCNN)

View full text Add to dashboard Cite

Echo State Networks (ESNs) have shown great promise in the applications of non-linear time series processing because of their powerful computational ability and efficient training strategy. However, the nature of randomization in the structure of the reservoir causes it be poorly understood and leaves room for further improvements for specific problems. A deterministically constructed reservoir model, Cycle Reservoir with Jumps (CRJ), shows superior generalization performance to standard ESN. However, the weights that govern the structure of the reservoir (reservoir weights) in CRJ model are obtained through exhaustive grid search which is very computational intensive. In this paper, we propose to learn the reservoir weights together with the linear readout weights using a hybrid optimization strategy. The reservoir weights are trained through nonlinear optimization techniques while the linear readout weights are obtained through linear algorithms. The experimental results demonstrate that the proposed strategy of training the CRJ network tremendously improves the computational efficiency without jeopardizing the generalization performance, sometimes even with better generalization performance.

show abstract

Section: B Readout Regularizationmentioning

confidence: 99%

Learning the deterministically constructed Echo State Networks

Tang

Tiňo

Chen

2014

2014 International Joint Conference on Neural Networks (IJCNN)

View full text Add to dashboard Cite

show abstract

“…Similarly, the inputs and outputs of the Y-direction model correspond to the Y-direction data, while those of the X-direction model correspond to data in all three directions. In addition, in order to train the neural network more effectively and improve its generalization ability [19], compared with Levenberg-Marquardt algorithm, which converges fast, in this paper, Bayesian regularization algorithm with penalty function is chosen. The number of layers in this network is set to three, in which the activation function of the hidden layer is the sigmoid function, the activation function of the output layer is the linear function, and the number of training iterations is 200.…”

Section: Model Constructing and Datamentioning

confidence: 99%

Prediction Model of the Mechanical Behavior of a Fuel Cell Stack under Strengthened Road Vibrating Conditions

Hou

et al. 2021

Modelling and Simulation in Engineering

View full text Add to dashboard Cite

In this paper, a data-oriented model has been presented by nonlinear autoregressive exogenous model (NARX) neural network, which aims at predicting the mechanical behavior of a fuel cell stack for vehicle under the real-life operational conditions. A 300-hour vibration test with reproduction of SVP road spectrum was completed on a Multi-Axial Simulation Table. At the same time, data acquisition of drive displacement and acceleration response on stack was carried out in every 50 hours. All data collected were used to train and evaluate the model based on NARX. Result shows that the prediction model built is of good precision and consistent with the actual situation.

show abstract

“…In the training of the ESN only a specific hour of the day is taken into account. In [25] the ESN is used for a multi-variate TS prediction and the reservoir is trained with a Bayesian regularization technique. In order to avoid overfitting in the regression step, neurons and redundant connections are pruned.…”

Section: Related Workmentioning

confidence: 99%

“…The smooth component, instead, is processed with an ESN and the final prediction is obtained integrating all the components. An interesting forecasting model using ESN on a multivariate TS can be found in [25], where the prediction is performed using a special ESN with a different reservoir for each variable: this allows to better catch the dynamic of each single variable. Because of the presence of multiple reservoirs, the number of output connections from the internal matrices to the output is huge and could lead to overfitting during the training; for this reason authors propose a method for pruning these connections.…”

Section: Related Workmentioning

confidence: 99%

“…A preemptive analysis of the TS can help us to understand the nature of the seasonality and the variance in the data. In this preprocessing phase it is also important to identify the presence of multiple seasonalities, hidden daily and/or hourly patterns, which could be better treated using multiple forecasting systems [9], [25] or with models designed to deal with multiple seasonality [27], [28], [36]. Fourier frequency analysis and the auto-correlation function (ACF), evaluated up to sufficient high number of lags, are the most commonly used tools for seasonality study.…”

Section: The Analyzed Time Seriesmentioning

confidence: 99%

See 1 more Smart Citation

Short-Term Electric Load Forecasting Using Echo State Networks and PCA Decomposition

et al. 2015

View full text Add to dashboard Cite

In this paper, we approach the problem of forecasting a time series (TS) of an electrical load measured on the Azienda Comunale Energia e Ambiente (ACEA) power grid, the company managing the electricity distribution in Rome, Italy, with an echo state network (ESN) considering two different leading times of 10 min and 1 day. We use a standard approach for predicting the load in the next 10 min, while, for a forecast horizon of one day, we represent the data with a high-dimensional multi-variate TS, where the number of variables is equivalent to the quantity of measurements registered in a day. Through the orthogonal transformation returned by PCA decomposition, we reduce the dimensionality of the TS to a lower number k of distinct variables; this allows us to cast the original prediction problem in k different one-step ahead predictions. The overall forecast can be effectively managed by k distinct prediction models, whose outputs are combined together to obtain the final result. We employ a genetic algorithm for tuning the parameters of the ESN and compare its prediction accuracy with a standard autoregressive integrated moving average model.

show abstract

Multi-variable Echo State Network Optimized by Bayesian Regulation for Daily Peak Load Forecasting

Cited by 6 publications

References 16 publications

Learning the deterministically constructed Echo State Networks

Learning the deterministically constructed Echo State Networks

Prediction Model of the Mechanical Behavior of a Fuel Cell Stack under Strengthened Road Vibrating Conditions

Short-Term Electric Load Forecasting Using Echo State Networks and PCA Decomposition

Contact Info

Product

Resources

About