Proposed Prediction Strategy Research Articles

A dynamic multi-objective evolutionary algorithm using center and multi-direction prediction strategies.

Dynamic multi-objective optimization problems have been popular because of its extensive application. The difficulty of solving the problem focuses on the moving PS as well as PF dynamically. A large number of efficient strategies have been put forward to deal with such problems by speeding up convergence and keeping diversity. Prediction strategy is a common method which is widely used in dynamic optimization environment. However, how to increase the efficiency of prediction is always a key but difficult issue. In this paper, a new prediction model is designed by using the rank sums of individuals, and the position difference of individuals in the previous two adjacent environments is defined to identify the present change type. The proposed prediction strategy depends on environment change types. In order to show the effectiveness of the proposed algorithm, the comparison is carried out with five state-of-the-art approaches on 20 benchmark instances of dynamic multi-objective problems. The experimental results indicate the proposed algorithm can get good convergence and distribution in dynamic environments.

Characterization and prediction of InSAR-derived ground motion with ICA-assisted LSTM model

Interferometric Synthetic Aperture Radar (InSAR) is a highly effective and widely used approach for monitoring large-scale ground deformation. The precise and timely prediction of deformation holds significant importance in mitigating and preventing geological hazards, particularly considering the long revisit cycle of satellites and the considerable time required for data processing. In this study, we propose a strategy that predicts spatiotemporal InSAR time series based on Independent Component Analysis (ICA) and the Long Short-Term Memory (LSTM) machine learning model. Unlike traditional methods that rely on physical or statistical models, the proposed strategy leverages the power of ICA and LSTM to achieve accurate predictions without such dependencies. ICA is employed to decompose and capture the InSAR displacement signals of interest caused by various natural or anthropogenic processes and to characterize each individual signal. The spatiotemporal unsupervised K-mean cluster method is then applied to partition large-scale deformation fields into homogeneous subregions, considering the spatial variations and temporal nonlinearities of time series. This process facilitates the refinement of the model, thereby enhancing the accuracy of large-scale predictions. The neural network models are then individually constructed for each cluster, and the optimal parameters are determined through a grid search strategy. Subsequently, the proposed framework is implemented and assessed using two datasets featuring distinct deformation patterns: Case I involves land subsidence in Willcox Basin, USA, while Case II focuses on post-seismic deformation following the 12 November 2017 Mw 7.3 Sarpol-e Zahab earthquake. The results demonstrate that our proposed ICA-assisted LSTM outperforms the original LSTM model on large-scale deformation prediction, with the average prediction accuracy for one-step prediction (12 days in our case) being improved by 34% and 17% for cases I and II, respectively. Furthermore, we perform iterative predictions on the spatiotemporal InSAR measurements with varying temporal characteristics for the subsequent five steps using Sentinel-1 data and evaluate its performance and limitations. The successful prediction of land subsidence and post-seismic deformation provides further evidence that the proposed prediction strategy can be effectively employed in monitoring other large-scale geohazards characterized by prolonged and gradual deformation. This capability enables expedited decision-making and timely implementation of risk mitigation measures.

A comprehensive wind speed prediction system based on intelligent optimized deep neural network and error analysis

At present, wind power is one of the most promising clean energy sources. Due to its high variability, accurate wind speed forecasting is an essential part of the wind power industry. This paper proposes a comprehensive forecasting system consisting of three parts: Firstly, the novel SC-EWT wind speed signal reconstruction system proposed in this article can effectively extract information from wind speed signals. Secondly, the QBGA-GRU model is used for signal reconstruction prediction, where QBGA is an improved genetic algorithm proposed in this article that can improve prediction accuracy. Finally, the IPSO-BLS algorithm is used to quickly analyze and predict prediction errors and further correct the predicted results. The system can provide users with adaptive signal decomposition, high-precision deep learning prediction and error analysis functions to achieve efficient and fast short-term wind speed forecasting, which can be used in the fields of wind power integration deployment and wind turbine early warning. The experiments show that the proposed prediction strategy has good predictive ability for wind speed samples in two regions, and its error is smaller than that of commonly used intelligent algorithms. Compared with other models, the proposed comprehensive prediction strategy has better predictive performance and smaller prediction errors than other combination models.

Prediction model for multiblock data with or without intermediate blocks. Estimation by PLS Path Modeling

An important issue in modern science remains the integration of multiple data sources, considered as datasets divided into blocks of variables measured on the same set of observations. When analyzing these blocks, the path modeling approach aims to take into account a specific pattern of directed relations between them. This structure, usually set up from prior knowledge, leads to a path diagram in which the block’s components are linked to one another. Until path modeling has gained popularity in recent decades, the issue of prediction has been surprisingly neglected in the literature. To fill this gap, we propose a prediction framework dedicated to the path modeling approach. Indeed, prediction for path modeling methods is of paramount interest for: (i) selecting the optimal model with cross-validation (e.g., optimal model specification and dimension), (ii) comparing the model with other relevant ones, (iii) or predicting new observations. Here, we propose an explicit formulation of a prediction model in the context of a path modeling structure. This generic prediction model is estimated using the PLS Path Modeling algorithm, denoted PMM-PLSPM (=Prediction Model for Multiblock data estimated by PLS Path Modeling). Nevertheless, its versatility makes it possible to adapt the approach to any other composite-based methods. Two estimation strategies (PLSPM-like and PLSR-like) are proposed for prediction, both being viewed as the structural model at the variable level. At the same time, the components of multiple blocks – obtained by a relevant deflation strategy – are handled to take advantage of the full multidimensional potential of blocks. The proposed prediction strategies are applied and evaluated on real data sets.

Hybrid bidirectional long short term memory with black widow optimization for crop yield prediction using data mining

SummaryCrop yield prediction is highly significant in the agricultural sector. It helps to understand the growth rate of major food crops and identify measures to improve the overall yield. The article proposes a hybrid strategy called bidirectional long short term memory with black widow optimization (Bi‐LSTM‐BWO) for predicting the annual yield produced with improved accuracy. Initially, data augmentation is performed for the collected dataset to increase the size of the dataset and to reduce the data scarcity problem. Then, the dataset is preprocessed to improve the data's quality and remove the noise and irrelevant information. The data is cleaned, transformed, and discretized in the preprocessing stage using various techniques. Then, the preprocessed data is clustered using an enhanced K‐means clustering technique. To enhance the clustering technique, the proposed technique utilized the rain optimization algorithm that automatically computes the initial centroids to improve the clustering outcome. Finally, the prediction process is performed using the proposed Bi‐LSTM‐BWO prediction scheme. The proposed prediction strategy efficiently predicts the annual yield with a high accuracy rate and minimizes loss. The proposed technique achieves a 99.18%, 99.81% and 99.01% accuracy rates for the summer, autumn and winter yield prediction, respectively.

A novel prediction and control method for solar energy dispatch based on the battery energy storage system using an experimental dataset

The high power generation growth by photovoltaic systems needs to forecast the power generation profile during a day. It is also required to evolve the high-efficient and optimal on-grid/off-grid photovoltaic power generation units. Furthermore, some advantages can be achieved by integrating photovoltaic systems with storage devices such as battery energy storage systems. Thus, optimizing the hybrid systems comprising photovoltaic and battery energy storage systems is needed to evaluate the best capacity. In the present work, a novel control and sizing scheme is proposed for the battery energy storage system in a photovoltaic power generation plant in one-hour ahead and one-day ahead during the dispatching phase. Then, the proposed prediction strategy is recommended for solar irradiation and power utilization. The control approach comprises a predictive control method concerning a Radial Basis Function network optimized by Levenberg-Marquardt back-propagation learning algorithm. Using the RBF network for simulation leads to a WAPE% =1.68 %.

A novel approach of full state tendency measurement for complex systems based on information causality and PageRank: A case study of a hydropower generation system

The hydropower generation system is a typical complex nonlinear system with hybrid state responses. The interaction between the state responses of the system is affected closely by the coupling of hydraulic, mechanical, and electromagnetic factors and the frequent changes in working conditions during operation. However, the precise roles of the interaction relationship are unknown. Here, we show that this interaction depends on the causal coupling between subsystems, and use this relationship to propose a time series data mining and data prediction strategy based on the information causality and the PageRank algorithm. A nonlinear model is used to prove that the proposed prediction strategy can effectively reduce the dimension of auxiliary variables. Finally, the strategy is validated with a 250 MW hydropower unit. Our results show that the information causal coupling between variables is cross-scale with definite Markov orders of time series, and the prediction accuracy can be improved by considering the information transfer sequences between the prediction object variable and the causal variable in the absence of state auxiliary variables. Furthermore, the proposed method can be also applied to data mining of other complex systems and variable selection of prediction models and builds a bridge between mechanism- and data-driven approaches, which has a high engineering application value.

Predicting Box-Office Markets with Machine Learning Methods.

The accurate prediction of gross box-office markets is of great benefit for investment and management in the movie industry. In this work, we propose a machine learning-based method for predicting the movie box-office revenue of a country based on the empirical comparisons of eight methods with diverse combinations of economic factors. Specifically, we achieved a prediction performance of the relative root mean squared error of 0.056 in the US and of 0.183 in China for the two case studies of movie markets in time-series forecasting experiments from 2013 to 2016. We concluded that the support-vector-machine-based method using gross domestic product reached the best prediction performance and satisfies the easily available information of economic factors. The computational experiments and comparison studies provided evidence for the effectiveness and advantages of our proposed prediction strategy. In the validation process of the predicted total box-office markets in 2017, the error rates were 0.044 in the US and 0.066 in China. In the consecutive predictions of nationwide box-office markets in 2018 and 2019, the mean relative absolute percentage errors achieved were 0.041 and 0.035 in the US and China, respectively. The precise predictions, both in the training and validation data, demonstrate the efficiency and versatility of our proposed method.

Modeling of a simplified hybrid algorithm for short-term load forecasting in a power system network

PurposeThe purpose of this paper is to develop a hybrid algorithm, which is a blend of auto-regressive integral moving average (ARIMA) and multilayer perceptron (MLP) for addressing the non-linearity of the load time series.Design/methodology/approachShort-term load forecasting is a complex process as the nature of the load-time series data is highly nonlinear. So, only ARIMA-based load forecasting will not provide accurate results. Hence, ARIMA is combined with MLP, a deep learning approach that models the resultant data from ARIMA and processes them further for Modelling the non-linearity.FindingsThe proposed hybrid approach detects the residuals of the ARIMA, a linear statistical technique and models these residuals with MLP neural network. As the non-linearity of the load time series is approximated in this error modeling process, the proposed approach produces accurate forecasting results of the hourly loads.Originality/valueThe effectiveness of the proposed approach is tested in the laboratory with the real load data of a metropolitan city from South India. The performance of the proposed hybrid approach is compared with the conventional methods based on the metrics such as mean absolute percentage error and root mean square error. The comparative results show that the proposed prediction strategy outperforms the other hybrid methods in terms of accuracy.

A Model-Based Online Reference Prediction Strategy for Model Predictive Motion Cueing Algorithms

Interactive driving simulation has become a key technology to support the development and optimization process of modern vehicle components and driver assistance systems both in academic research and in the automotive industry. However, the validity of the results obtained within the virtual environment depends essentially on the adequate reproduction of the simulated vehicle movements and the corresponding immersion of the driver. For that reason, specific motion platform control strategies, so-called Motion Cueing Algorithms (MCA), are used to replicate the simulated accelerations and angular velocities within the physical limitations of the driving simulator best possible. In this paper, we present a novel model-based approach to predict oncoming vehicle motion at runtime. For that purpose, a virtual driver model as well as a simplified vehicle dynamics model are introduced to estimate the future driver inputs and the resulting vehicle trajectories according to the current driving situation. This additional system knowledge enables control algorithms designed on the idea of Model Predictive Control (MPC) to exploit their potential more efficiently. The performance of the proposed prediction strategy is evaluated on the basis of measurement data from a real test run in comparison to an ideal prediction and a constant reference, using a hybrid kinematics motion system as an application example.

Establishing functional brain networks using a nonlinear partial directed coherence method to predict epileptic seizures

BackgroundEpilepsy is a neurological disorder characterized by unpredictable seizures that can lead to severe health problems. EEG techniques have shown to be advantageous for studying and predicting epileptic seizures, thanks to their cost-effectiveness, non-invasiveness, portability and the capability for long-term monitoring. Linear and non-linear EEG analysis methods have been developed for the effective prediction of seizure onset, however both methods remain blind to underlying alterations of the structural and functional brain networks associated with epileptic seizures. Such information is employed in this study to develop novel method for epileptic seizure prediction. New MethodsIn this study, nonlinear partial directed coherence (NPDC) was employed as measure of functional brain networks (FBNs) and analyzed to reveal the directional flow of epilepsy-linked brain activity. A novel prediction strategy was then developed for the prediction of epileptic seizures by introducing extracted network features to an extreme learning machine (ELM). ResultsResults show that the proposed method achieved favorable performance across all subjects and in all EEG frequency bands, with best accuracy of 89.2% in beta band and an optimal prediction time of 1356.4 s in delta bands, which outperforms currently available approaches. Comparison with Existing MethodsOur NPDC based on FBNs methods approach surpasses the accuracy of pure graph theory and pure non-linear methods with a significantly increased prediction time. ConclusionsThe findings of this study demonstrate that the proposed prediction strategy is suitable for the prediction of seizure onset.

A Novel Strategy for Short-Term Prediction of Fading Channel

In this paper, a novel prediction strategy based on firefly algorithm, echo state network and Savitzky-Golay filter for short-term prediction of fading channel is proposed and introduced in detail. This hybrid strategy includes two parts, that is, optimization part and repairing part. In the former, parameters of echo state network are optimized by firefly algorithm so that an optimal prediction echo state network is obtained, then in order to improve performances in prediction strategy, prediction sample of echo state network is repaired by repairing strategy based on Savitzky-Golay filter. Finally, the classic Rayleigh channel is tested and performances are analysed and compared with classic AR, which verified the validity and correctness of our proposed prediction strategy.

Ensemble prediction-based dynamic robust multi-objective optimization methods

Abstract Many real-world multi-objective optimization problems are subject to environmental changes over time, resulting in changing Pareto-optima. Wide studies on solving dynamic multi-objective optimization problems have so far concentrated on tracking moving Pareto-optima as soon as possible. In practice, however, a new solution every time the environmental change may be different from the previous optima, causing the expensive switching-cost. To this end, dynamic robust multi-objective optimization method is developed to find robust Pareto-optima over time whose performance is acceptable for the current and subsequently changed environments. With the purpose of measuring the robustness of a candidate, its fitness values in the subsequent environments are estimated by ensemble prediction methods constructed by moving average(MA), autoregressive(AR), and single exponential smoothing(SES). MA-, SES- and AR-based sub-prediction models are synthesized by the weight sum. The weights can be the pre-set constant or the binary/real number adjusted in terms of the prediction error. To examine the performance of the developed algorithm, the proposed prediction strategies are compared with three single prediction methods for 11 dynamic benchmark functions. The experimental results indicate that ensemble prediction methods have the better robustness than the single prediction models and can effectively tackle dynamic robust multi-objective optimization problems.

Multidirectional Prediction Approach for Dynamic Multiobjective Optimization Problems.

Various real-world multiobjective optimization problems are dynamic, requiring evolutionary algorithms (EAs) to be able to rapidly track the moving Pareto front of an optimization problem once an environmental change occurs. To this end, several methods have been developed to predict the new location of the moving Pareto set (PS) so that the population can be reinitialized around the predicted location. In this paper, we present a multidirectional prediction strategy to enhance the performance of EAs in solving a dynamic multiobjective optimization problem (DMOP). To more accurately predict the moving location of the PS, the population is clustered into a number of representative groups by a proposed classification strategy, where the number of clusters is adapted according to the intensity of the environmental change. To examine the performance of the developed algorithm, the proposed prediction strategy is compared with four state-of-the-art prediction methods under the framework of particle swarm optimization as well as five popular EAs for dynamic multiobjective optimization. Our experimental results demonstrate that the proposed algorithm can effectively tackle DMOPs.

Automatic CT simulation optimization for radiation therapy: A general strategy

In radiation therapy, x-ray computed tomography (CT) simulation protocol specifications should be driven by the treatment planning requirements in lieu of duplicating diagnostic CT screening protocols. The purpose of this study was to develop a general strategy that allows for automatically, prospectively, and objectively determining the optimal patient-specific CT simulation protocols based on radiation-therapy goals, namely, maintenance of contouring quality and integrity while minimizing patient CT simulation dose. The authors proposed a general prediction strategy that provides automatic optimal CT simulation protocol selection as a function of patient size and treatment planning task. The optimal protocol is the one that delivers the minimum dose required to provide a CT simulation scan that yields accurate contours. Accurate treatment plans depend on accurate contours in order to conform the dose to actual tumor and normal organ positions. An image quality index, defined to characterize how simulation scan quality affects contour delineation, was developed and used to benchmark the contouring accuracy and treatment plan quality within the predication strategy. A clinical workflow was developed to select the optimal CT simulation protocols incorporating patient size, target delineation, and radiation dose efficiency. An experimental study using an anthropomorphic pelvis phantom with added-bolus layers was used to demonstrate how the proposed prediction strategy could be implemented and how the optimal CT simulation protocols could be selected for prostate cancer patients based on patient size and treatment planning task. Clinical IMRT prostate treatment plans for seven CT scans with varied image quality indices were separately optimized and compared to verify the trace of target and organ dosimetry coverage. Based on the phantom study, the optimal image quality index for accurate manual prostate contouring was 4.4. The optimal tube potentials for patient sizes of 38, 43, 48, 53, and 58 cm were 120, 140, 140, 140, and 140 kVp, respectively, and the corresponding minimum CTDIvol for achieving the optimal image quality index 4.4 were 9.8, 32.2, 100.9, 241.4, and 274.1 mGy, respectively. For patients with lateral sizes of 43-58 cm, 120-kVp scan protocols yielded up to 165% greater radiation dose relative to 140-kVp protocols, and 140-kVp protocols always yielded a greater image quality index compared to the same dose-level 120-kVp protocols. The trace of target and organ dosimetry coverage and the γ passing rates of seven IMRT dose distribution pairs indicated the feasibility of the proposed image quality index for the predication strategy. A general strategy to predict the optimal CT simulation protocols in a flexible and quantitative way was developed that takes into account patient size, treatment planning task, and radiation dose. The experimental study indicated that the optimal CT simulation protocol and the corresponding radiation dose varied significantly for different patient sizes, contouring accuracy, and radiation treatment planning tasks.

Short-term wind power forecasting using ridgelet neural network

Rapid growth of wind power generation in many countries around the world in recent years has highlighted the importance of wind power prediction. However, wind power is a complex signal for modeling and forecasting. Despite the performed research works in the area, more efficient wind power forecast methods are still demanded. In this paper, a new prediction strategy is proposed for this purpose. The forecast engine of the proposed strategy is a ridgelet neural network (RNN) owning ridge functions as the activation functions of its hidden nodes. Moreover, a new differential evolution algorithm with novel crossover operator and selection mechanism is presented to train the RNN. The efficiency of the proposed prediction strategy is shown for forecasting of both wind power output of wind farms and aggregated wind generation of power systems.

Short-Term Load Forecast of Microgrids by a New Bilevel Prediction Strategy

Microgrids are a rapidly growing sector of smart grids, which will be an essential component in the trend toward distributed electricity generation. In the operation of a microgrid, forecasting the short-term load is an important task. With a more accurate short-term loaf forecast (STLF), the microgrid can enhance the management of its renewable and conventional resources and improve the economics of energy trade with electricity markets. However, STLF for microgrids is a complex forecast process, mainly because of the highly nonsmooth and nonlinear behavior of the load time series. In this paper, characteristics of the load time series of a typical microgrid are discussed and the differences with the load time series of traditional power systems are described. In addition, a new bilevel prediction strategy is proposed for STLF of microgrids. The proposed strategy is composed of a feature selection technique and a forecast engine (including neural network and evolutionary algorithm) in the lower level as the forecaster and an enhanced differential evolution algorithm in the upper level for optimizing the performance of the forecaster. The effectiveness of the proposed prediction strategy is evaluated by the real-life data of a university campus in Canada.

Predicting travel time to limit congestion at a highway bottleneck

A new method is proposed to predict the travel time on a highway route with a bottleneck caused by an on-ramp. The method takes advantage of the slow variation of the bottleneck throughput when congestion exists. The predicted travel time for a vehicle leaving the origin is given by the current number of vehicles on the route divided by the estimated throughput. The latter is an average of N / T recorded as each vehicle reaches the destination where N is the number of vehicles at the start of the trip and T is the time to complete the trip. Drivers divert to an off-ramp when the predicted travel time exceeds a target value. The target could be historical average travel times of alternative routes or chosen to limit the amount of congestion. Simulations employing three-phase traffic theory show that the travel time converges to the target value and remains close to or below it with the proposed prediction strategy. Strong oscillations in travel time obtained when other strategies are used for diversion do not develop with the new method because the inherent delay is effectively removed.

A novel prediction-based strategy for object tracking in sensor networks by mining seamless temporal movement patterns

Energy saving in sensor networks has received a great deal of research attention in recent years due to its wide applications. One important research issue is energy efficient object tracking in sensor networks (OTSNs). Past studies on energy saving in OTSNs can be divided into two main directions: (1) improvement in hardware design; and (2) improvement in software approaches. Many research papers save energy in hardware design, but few discuss software approaches. The intuitive way to conserve the energy of sensor nodes is to reduce the operation time of high-powered components. Utilizing the movement patterns of objects to save energy is one software approach. However, it did not take temporal information into consideration nor did it define a suitable segmenting time unit of time interval in advance. Due to the time interval between movements is a real number, an improper segmenting time unit may not discover the useful patterns, directly resulting in the inefficient object tracking. In this paper, we propose a seamless data mining algorithm named STMP-Mine to efficiently discover the temporal movement patterns of objects in sensor networks without predefining the segmenting time unit. Moreover, we propose novel location prediction strategies that employ the discovered temporal movement patterns to reduce prediction errors to save energy. With empirical evaluation on simulated data, STMP-Mine and the proposed prediction strategies are shown to deliver excellent performance in terms of scalability and energy efficiency.

Energy efficient strategies for object tracking in sensor networks: A data mining approach

In recent years, a number of studies have been done on object tracking sensor networks (OTSNs) due to the wide applications. One important research issue in OTSNs is the energy saving strategy in considering the limited power of sensor nodes. The past studies on energy saving in OTSNs considered the object’s movement behavior as randomness. In some real applications, however, the object movement behavior is often based on certain underlying events instead of randomness completely. In this paper, we propose a novel data mining algorithm named TMP-Mine with a special data structure named TMP-Tree for efficiently discovering the temporal movement patterns of objects in sensor networks. To our best knowledge, this is the first work on mining the movement patterns associated with time intervals in OTSNs. Moreover, we propose novel location prediction strategies that utilize the discovered temporal movement patterns so as to reduce the prediction errors for energy savings. Through empirical evaluation on various simulation conditions and real dataset, TMP-Mine and the proposed prediction strategies are shown to deliver excellent performance in terms of scalability, accuracy and energy efficiency.