Real-time Prediction Problems Research Articles

Data Analytics and Reporting API – A Reliable Tool for Data Visualization and Predictive Analysis

Data analytics, the science of analyzing raw data is widely being used in a broad range of applications to makeeffective, efficient and timely decisions in the real-time prediction problems. Analytics requires the user to havean in-depth knowledge about the various steps involved in the process for arriving at the conclusions. Hence, tomake it easier for the naïve users, Data Analytics and Reporting API (DARAPI) provides Analytics as a Service inthe form of an Application Programming Interface (API) has been proposed. DARAPI is aimed to ease the processof analyzing the data by the users without requiring their expertise in the field. It accepts the input file from theuser and performs pre-processing techniques at various stages including imputation of missing data, detectionand replacement of outliers, encoding of categorical variables, and etc. Furthermore, feature engineering is performed, based on which DARAPI will execute the different classification/regression models and finally deliversthe model which provides the best accuracy for future predictions. This entire system is rendered to the user inthe form of an API that can be called from any device that is Internet enabled. DARAPI stands unique with its embedded feedback mechanism which generates constructive input for future predictions. This feature enhancesthe performance of the system in comparison with the existing tools. DARAPI proves beneficial not only to thenaïve users but also to the experts by saving their time and efforts needed in understanding the data.

A Function Approximation Approach to the Prediction of Blood Glucose Levels

The problem of real time prediction of blood glucose (BG) levels based on the readings from a continuous glucose monitoring (CGM) device is a problem of great importance in diabetes care, and therefore, has attracted a lot of research in recent years, especially based on machine learning. An accurate prediction with a 30, 60, or 90 min prediction horizon has the potential of saving millions of dollars in emergency care costs. In this paper, we treat the problem as one of function approximation, where the value of the BG level at time t+h (where h the prediction horizon) is considered to be an unknown function of d readings prior to the time t. This unknown function may be supported in particular on some unknown submanifold of the d-dimensional Euclidean space. While manifold learning is classically done in a semi-supervised setting, where the entire data has to be known in advance, we use recent ideas to achieve an accurate function approximation in a supervised setting; i.e., construct a model for the target function. We use the state-of-the-art clinically relevant PRED-EGA grid to evaluate our results, and demonstrate that for a real life dataset, our method performs better than a standard deep network, especially in hypoglycemic and hyperglycemic regimes. One noteworthy aspect of this work is that the training data and test data may come from different distributions.

Soft - sensing modeling based on ABC - MLSSVM inversion for marine low - temperature alkaline protease MP fermentation process

BackgroundAiming at the characteristics of nonlinear, multi-parameter, strong coupling and difficulty in direct on-line measurement of key biological parameters of marine low-temperature protease fermentation process, a soft-sensing modeling method based on artificial bee colony (ABC) and multiple least squares support vector machine (MLSSVM) inversion for marine protease fermentation process is proposed.MethodsFirstly, based on the material balance and the characteristics of the fermentation process, the dynamic “grey box” model of the fed-batch fermentation process of marine protease is established. The inverse model is constructed by analyzing the inverse system existence and introducing the characteristic information of the fermentation process. Then, the inverse model is identified off-line using MLSSVM. Meanwhile, in order to reduce the model error, the ABC algorithm is used to correct the inverse model. Finally, the corrected inverse model is connected in series to the marine alkaline protease MP fermentation process to form a composite pseudo-linear system, thus, real-time on-line prediction of key biological parameters in fermentation process can be realized.ResultsTaking the alkaline protease MP fermentation process as an example, the simulation results demonstrate that the soft-sensing modeling method can solve the real-time prediction problem of key biological parameters in the fermentation process on-line, and has higher accuracy and generalization ability than the traditional soft-sensing method of support vector machine.ConclusionsThe research provides a new method for soft-sensing modeling of key biological parameters in fermentation process, which can be extended to soft-sensing modeling of general nonlinear systems.

A Dynamic Analytics Method Based on Multistage Modeling for a BOF Steelmaking Process

This paper proposes a dynamic analytics method based on the least squares support vector machine with a hybrid kernel to address real-time prediction problems in the converter steelmaking process. The hybrid kernel function is used to enhance the performance of the existing kernels. To improve the model’s accuracy, the internal parameters are optimized by a differential evolution algorithm. In light of the complex mechanisms of the converter steelmaking process, a multistage modeling strategy is designed instead of the traditional single-stage modeling method. Owing to the dynamic nature of the practical production process, great effort has been made to construct a dynamic model that uses the prediction error information based on the static model. The validity of the proposed method is verified through experiments on real-world data collected from a basic oxygen furnace steelmaking process. The results indicate that the proposed method can successfully solve dynamic prediction problems and outperforms other state-of-the-art methods in terms of prediction accuracy. Note to Practitioners —With the development of cyber-physical systems, abundant real-time data have been collected from the converter steelmaking process. These data provide an opportunity to solve product quality prediction problems using data-driven models. This paper proposes a dynamic analytics method based on the least squares support vector machine with a hybrid kernel to address this challenging issue. To improve the model’s performance, a differential evolution algorithm is used to optimize its parameters. Because of the fierce physicochemical reaction in the converter furnace, it is difficult for a single-stage model to achieve accurate predictions. Thus, a multistage modeling strategy is proposed to address this difficulty, and a dynamic model based on feedback error is developed to realize real-time prediction. We verify the effectiveness of the proposed method using real data from a basic oxygen furnace (BOF) steelmaking process. The computational results reveal that the proposed method has a higher prediction accuracy than other methods, making it helpful in guaranteeing the specified product quality and in maintaining stable BOF operation.

Symbiotic Structure Learning Algorithm for Feedforward Neural-Network-Aided Grey Model and Prediction Applications

The learning ability of neural networks (NNs) enables them to solve time series prediction problems. Off-line training can be applied to design the structure and weights of NNs when sufficient training data are available. However, this may be inadequate for applications that operate in real time, possess limited memory size, or require online adaptation. Furthermore, the structural design of NNs (i.e., the number of hidden neurons and connected topology) is crucial. This paper presents a novel algorithm, called the symbiotic structure learning algorithm (SSLA), to enhance a feedforward neural-network-aided grey model (FNAGM) for real-time prediction problems. Through symbiotic evolution, the SSLA evolves neurons that cooperate well with each other, and constructs NNs from the neurons with hyperbolic tangent and linear activation functions. During construction, the hidden neurons with the linear activation function can be simplified to a few direct connections from the inputs to the output neuron, leading to a compact network topology. The NNs share the fitness value with participating neurons, which are further evolved through neuron crossover and mutation. The proposed SSLA was evaluated through three real-time prediction problems. Experimental results showed that the SSLA-derived FNAGM possesses a partially connected NN with few hidden neurons and a compact topology. The evolved FNAGM outperforms other methods in prediction accuracy and continuously adapts the NN to the dynamic changes of the time series for real-time applications.

Predictive flow-field estimation

We are considering the problem of real-time prediction of 3D turbulent velocity fields based on a small number of scalar measurements. The method of proper orthogonal decomposition (POD) allows for the decomposition of an ensemble of velocity fields into a set of spatial basis functions and a set of temporal coefficients. The computation of the temporal coefficients is by no means a trivial matter, especially when one is faced with a large number of modes. In this paper we discuss the use of radial basis function (RBF) models to capture the discrete time evolution and nonlinear dynamics of the POD coefficients. Further, we propose the use of regularized regression techniques to generate models that provide mappings between the POD coefficients and scalar measurements. As a final step towards real-time prediction, the state–space RBF models and regression measurement models are combined using unscented Kalman filters to produce optimal solutions such that a balance between the state models and measurement models is achieved. The proposed methods are tested for two specific cases. The classical Lorenz model is chosen to demonstrate the use and effectiveness of RBF models as a potential candidate for state models. Flow around a wall-mounted cube in a channel at R e = 20,000 is considered as the second case. The aim for the second case is to be able to accurately predict the POD coefficients outside the ensemble. It is shown that a large number of POD coefficients is required to approximate the velocity fields with sufficient accuracy. The RBF models are created based on only the temporal information available from the initial ensemble, and it is shown that the RBF model is able to correctly approximate the high-dimensional phase space. Combined with the unscented Kalman filter it is indeed possible to track the evolution of the POD coefficients for a long time. The robustness of the filter is demonstrated by considering the presence of noise in measurements and using measurement information at time steps greater than the evolution time step.

Predicting future behavior of transient events rapidly enough to evaluate remedial control options in real-time

Electric utilities are becoming increasingly interested in using synchronized phasor measurements from around power systems to enhance their protection and remedial action control strategies. Accordingly, the task of predicting future behavior of the power system before it actually occurs has become an important area of research. This paper presents and analyses several approaches for solving the real-time prediction problem. In order to solve power systems with detailed load models fast enough for real-time prediction, the authors present a new piecewise constant current load model approximation technique that can solve a model as complex as the New England 39 bus system with composite voltage dependent loads much faster than in real-time. If the reduced order model is too large for real-time solution, then a pattern recognition tool such as decision trees can be trained off line to associate the post-fault phasor measurements with the outcome of future behavior. In this case also, the piecewise constant current technique would be needed to perform the offline training set generation with sufficient speed and accuracy. >