Memory-based Modeling Research Articles

Long short-term memory-based modeling and controlling method for wastewater treatment process

Considering the characteristics of large time variation, strong coupling, large time delay, and serious interference of wastewater treatment system, this article proposes an offline modeling and online controlling method based on long short-term memory network to improve multivariable control and prediction accuracy under disturbance. First, a prediction model of dissolved oxygen and nitrate nitrogen concentrations is established, and the stability of this long short-term memory-based modeling method is proven via the limitation of learning rate. Second, based on the prediction model, a multivariable long short-term memory-based controller is designed to improve the control accuracy by gradient descent algorithm, and the stability of the long short-term memory-based controller is proved by Lyapunov principle. Finally, based on the Benchmark Simulation Model No.1, the simulation experiments of long short-term memory-based modeling and controlling method, the default proportion–integration control method (proposed in the Benchmark Simulation Model No.1), and model predictive control method are conducted and compared, respectively. The results show the long short-term memory-based method owns both better approximating and controlling performance, and compared with the default proportion–integration and model predictive control, the long short-term memory-based controller reduces the integral squared error of dissolved oxygen and nitrate nitrogen concentrations by more than 94% and 80%, respectively.

Design of an Internal Model Control System based on the Database-Driven Modeling for a Hydraulic Excavator

In recent years, the Digital Transformation (DX) has been promoted by the Japanese government. New value is expected to be created by utilizing new digital technologies and data through DX. In particular, the Ministry of Land, Infrastructure, Transport and Tourism (MLIT) is developing a data integration platform that allows easy access to data and models in the cloud. On the other hand, Internal Model Control (IMC) based on the memory-based modeling (MBM) using data and model has been studied. It can be able to adjust control and internal model parameters adaptively using real-time data of the control system. However, the control performance might be deteriorated by this modeling method, because it can not calculate appropriate parameters when new system input and output data is not involved in the database. This paper presents a new design manner of an IMC system based on the database-driven modeling. The effectiveness of the proposed method is verified by the simulation that is assumed work of the hydraulic excavators.

Memory-based meso-scale modeling of Covid-19

The COVID-19 pandemic has led to an unprecedented world-wide effort to gather data, model, and understand the viral spread. Entire societies and economies are desperate to recover and get back to normality. However, to this end accurate models are of essence that capture both the viral spread and the courses of disease in space and time at reasonable resolution. Here, we combine a spatially resolved county-level infection model for Germany with a memory-based integro-differential approach capable of directly including medical data on the course of disease, which is not possible when using traditional SIR-type models. We calibrate our model with data on cumulative detected infections and deaths from the Robert-Koch Institute and demonstrate how the model can be used to obtain county- or even city-level estimates on the number of new infections, hospitality rates and demands on intensive care units. We believe that the present work may help guide decision makers to locally fine-tune their expedient response to potential new outbreaks in the near future.

Alternative approaches to the prediction of electricity prices

PurposeThis paper aims to present the results of the comparison between different approaches to the prediction of electricity prices. It is well-known that the properties of the data generation process may prefer some modeling methods over the others. The data having an origin in social or market processes are characterized by unexpectedly wide realization space resulting in the existence of the long tails in the probabilistic density function. These data may not be easy in time series prediction using standard approaches based on the normal distribution assumptions. The electricity prices on the deregulated market fall into this category.Design/methodology/approachThe paper presents alternative approaches, i.e. memory-based prediction and fractal approach compared with established nonlinear method of neural networks. The appropriate interpretation of results is supported with the statistical data analysis and data conditioning. These algorithms have been applied to the problem of the energy price prediction on the deregulated electricity market with data from Polish and Austrian energy stock exchanges.FindingsThe first outcome of the analysis is that there are several situations in the task of time series prediction, when standard modeling approach based on the assumption that each change is independent of the last following random Gaussian bell pattern may not be a true. In this paper, such a case was considered: price data from energy markets. Electricity prices data are biased by the human nature. It is shown that more relevant for data properties was Cauchy probabilistic distribution. Results have shown that alternative approaches may be used and prediction for both data memory-based approach resulted in the best performance.Research limitations/implications“Personalization” of the model is crucial aspect in the whole methodology. All available knowledge should be used on the forecasted phenomenon and incorporate it into the model. In case of the memory-based modeling, it is a specific design of the history searching routine that uses the understanding of the process features. Importance should shift toward methodology structure design and algorithm customization and then to parameter estimation. Such modeling approach may be more descriptive for the user enabling understanding of the process and further iterative improvement in a continuous striving for perfection.Practical implicationsMemory-based modeling can be practically applied. These models have large potential that is worth to be exploited. One disadvantage of this modeling approach is large calculation effort connected with a need of constant evaluation of large data sets. It was shown that a graphics processing unit (GPU) approach through parallel calculation on the graphical cards can improve it dramatically.Social implicationsThe modeling of the electricity prices has big impact of the daily operation of the electricity traders and distributors. From one side, appropriate modeling can improve performance mitigating risks associated with the process. Thus, the end users should receive higher quality of services ultimately with lower prices and minimized risk of the energy loss incidents.Originality/valueThe use of the alternative approaches, such as memory-based reasoning or fractals, is very rare in the field of the electricity price forecasting. Thus, it gives a new impact for further research enabling development of better solutions incorporating all available process knowledge and customized hybrid algorithms.

Memory-based cognitive modeling for robust object extraction and tracking

Inspired by the way humans perceive the environment, in this paper, we propose a memory-based cognitive model for visual information processing which can imitate some cognitive functions of the hum...

A Parallel Framework for In-Memory Construction of Term-Partitioned Inverted Indexes

With the advances in cloud computing and huge RAMs provided by 64-bit architectures, it is possible to tackle large problems using memory-based solutions. Construction of term-based, partitioned, parallel inverted indexes is a communication intensive task and suitable for memory-based modeling. In this paper, we provide an efficient parallel framework for in-memory construction of term-based partitioned, inverted indexes. We show that, by utilizing an efficient bucketing scheme, we can eliminate the need for the generation of a global vocabulary. We propose and investigate assignment schemes that can reduce the communication overheads while minimizing the storage and final query processing imbalance. We also present a study on how communication among processors should be carried out with limited communication memory in order to reduce the total inversion time. We present several different communication-memory organizations and discuss their advantages and shortcomings. The conducted experiments indicate promising results.

Memory-based Gaussian Mixture Background Modeling

Gaussian mixture model (GMM) is one of the best models for modeling a background scene with gradual changes and repetitive motions. However,it fails when the scene changes suddenly,e.g.,a door is opened or closed. To handle such problems,we propose a memory-based Gaussian mixture model (MGMM) inspired by the way human perceives the environment. The human memory mechanism is introduced to model the background. Each pixel of every frame is processed and transferred through three spaces:ultra-short time memory space,short time memory space,and long time memory space. The proposed memory-based model can remember what the scene has ever been,which helps the model adapt to the variation of the scene more quickly.

Recursive Memory-based Hysteresis Modeling for Solid-state Smart Actuators

This article presents a new modeling approach for the memory-dependent hysteresis phenomenon in a broad class of smart structures and systems. We propose a recursive formulation to relate the minor hysteresis trajectories to their surrounding loops. More specifically, each internal (minor) trajectory targets its previous turning point and converges to its neighboring loop with a tunable exponential rate. By applying the ‘curve alignment’ and the ‘wiping out’ properties at the turning points, we present a new strategy within the context of a memory-based hysteresis modeling framework. A Galfenol-driven micropositioning actuator and a piezoelectrically driven nanopositioning stage are used to experimentally validate the model. Galfenol exhibits large butterfly-type nonlinearity with a small hysteresis effect, while the piezoelectric actuator exhibits wide hysteresis loops. The model is able to precisely predict the major and minor hysteresis loops in both the Galfenol and piezoelectric actuators, and is expected to be effectively and conveniently applicable to general systems exhibiting memory-dependent hysteresis.

Design of a memory‐based prefilter supplementing a robust PID control system

AbstractFor systems with uncertainties, lots of PID parameter tuning methods have been proposed from the view point of the robust stability theory. However, the control performance becomes conservative using robust PID controllers. In this paper, a new two‐degree‐of‐freedom (2DOF) controller, which can improve the tracking properties, is proposed for nonlinear systems. According to the proposed method, the prefilter is designed as the PD compensator whose control parameters are tuned by the idea of a memory‐based modeling (MBM) method. Since the MBM method is a type of local modeling methods for nonlinear systems, PD parameters can be tuned adequately in an online manner corresponding to nonlinear properties. Finally, the effectiveness of the newly proposed control scheme is numerically evaluated on a simulation example. Copyright © 2008 John Wiley and Sons Asia Pte Ltd and Chinese Automatic Control Society

Design of a self-tuning memory-based GPC for nonlinear systems

In this paper, a new system identification scheme is proposed based on a memory-based modeling (MBM) method. According to the MBM method, some local models are automatically generated based on input/output data pairs of the controlled object stored in the data-base. Especially, it is well known that the MBM method works suitably on nonlinear systems. Therefore, even if the nonlinearities are contained in the controlled object, high identification accuracy can be obtained by the proposed method. Moreover, since the parameter estimates are easily applied to many existing controllers, the good control performance can be obtained for nonlinear systems by the same way to linear control schemes. In this paper, the generalized predictive control (GPC) is used as one of existing controllers, because the GPC is designed based on multi-step prediction, and is effective for systems with large time-delays, ambiguous time-delays and/or time-variant time-delays. Finally, the effectiveness of the newly proposed control scheme is numerically evaluated on a simulation example.