Biogas is a sustainable energy source produced through anaerobic digestion (AD), which converts organic waste into methane and carbon dioxide. Accurate prediction of biogas yield is essential for stable and efficient operation. However, this task is difficult due to the nonlinear dynamics of AD systems and frequent missing values in sensor data. Here, we propose a twostage framework. First, we introduce BiogasGAN, a generative adversarial network designed to impute missing values in multivariate time series data. It reconstructs incomplete sensor records while preserving temporal and cross-variable relationships. Second, we present BiogasNET, a hybrid deep learning model that combines convolutional layers, long short-term memory (LSTM) units, and attention mechanisms to forecast biogas production from imputed data. We evaluate our framework on real-world biogas plant datasets. Experimental results show that BiogasNET achieves state-of-the-art performance, with RMSE and MAE as low as 0.029 and 0.022, respectively. Ablation studies confirm the value of each model component, and comparisons with conventional machine learning methods highlight its robustness. Overall, our approach provides an effective and practical solution for biogas yield prediction in real-world environments.

Although Transformer has been successfully applied to Vision tasks in various fields, its large computational cost and performance degradation due to divergence from the language task are issues to be addressed. In this paper, we introduce token pruning to Mask2Former, a state-of-the-art segmentation method, to reduce computational cost and improve recognition accuracy without additional training. Multi-Scale Token Pruning (MSTP) works effectively on the multi-scale feature tokens of Mask2Former and can be universally implemented with various conventional token pruning methods. Experimental results show that introducing Top-K (norm+rand) MSTP into the Mask2Former of Swin-L backbone achieves +0.28 (56.31) mIoU on the ADE20K benchmark with +5.7% speed up. With this improvement, Mask2Former+MSTP can achieve mIoU equivalent to the large and powerful BEiT-UperNet with 1/4 of the computational complexity. In addition, +0.04 (57.86) PQ for COCO panoptic and +0.19 (63.36) mIoU for Mapillary Vistas are achieved, showing particular usefulness in complex semantic tasks with a large number of categories.

Wood drying technology plays a key role in extending service lifetime of wood as the moisture content has a great influence on the wood quality. This paper presents a moisture content prediction model based on the deep long short-term memory (LSTM) autoencoders with stochastic sensitivity (DLASS) to extract a hidden representation of input data. The DLASS uses multiple LSTM encoders to learn more informative hidden representations from unseen samples, which are then decoded using multiple LSTM decoders. The DLASS is trained by minimizing the perturbed error from historical moisture content data. Furthermore, a nonlinear fully connected feedforward neural network as a regression layer is applied to predict moisture content using hidden representations learned by the DLASS. The DLASS is applied to real-world industrial data of Sugi wood processed by a drying kiln made by SECEA from August 4 to 19, 2008. Multiple test cases and comparisons with existing classical and state-of-the-art models show that the DLASS model yields more accurate moisture content prediction results and has high generalization capability. To be specific, the DLASS yields the lowest MAE (0.026), MAPE (0.280), and RMSE (0.058) for predicting moisture content during the wood drying process.

Wireless power transfer (WPT) has attracted much attention in recent years. It is highly convenient and is expected to be put to practical use in a variety of devices. Wireless power transfer systems have the problem that the power transmission efficiency and the transmitted power change depending on the changes in the position and distance of the transmission and receiving coils and the size of the load resistance. In this paper, the efficiency is controlled by controlling the DC-DC converter at the power receiving side, and the output voltage is controlled at the power transmitting side using the operator theory. The effectiveness of the system was confirmed by simulation and the experiments using the actual system.

In this paper, a photovoltaic module integrated microinverter with gradationally controlled voltage sources and a series connected buck converter or a full bridge inverter as an active filter is proposed. Recently, typical commercially used microinveter can achieve high efficiency. By using microinverter system, high Maximum Power Point Tracking (MPPT) efficiency by its small photovoltaic area can be achieved. The proposed circuit is a combination of a gradationally controlled voltage source with low switching frequency, a series connected active filter with high switching frequency and an unfolding circuit. The power to be converted in this circuit can be reduced to 1/8 as compared with a conventional full power conversion inverter. The circuit construction, the controller for the proposed circuits are described. In addition, the grid connected operations are verified by circuit simulation.

This paper gives an optimal control law to eliminate ripples in inter-sample intervals of transient state of multi-rate sampled-control systems. As for ripples in steady state, there already exist several methods to remove the ripples. But to reduce the ripples in transient-state, so far there exists only trial and error method and there does not exist a systematic method. In this paper, first a reference model is defined which generates a reference output with no ripples and at sampled points, it has the same outputs with continuous time multi-rate control systems. Then the ripples in transient state are measured by the integrals of squared output errors. And equations to calculate the integrals are given by solving of Lyapunov equations. Finally the optimal control law is derived by differentiating the equations by the control input. The obtained optimal control input has a state feedback form and can be calculated at the short sampling time.

This paper concerns a swing suppression control of a variable length link utilizing the length change of a shape memory alloy (SMA) wire. In general, SMA actuators exhibit hysteresis characteristic, which affects the control of application using the length change of SMA actuators. To suppress the sway angle of a variable length pendulum efficiently, the feedback control system to make the length of the SMA wire track to the reference input which represents the desired length of the link is designed. In this paper, a SMA actuator model which consists of thermal model and hysteresis model is represented. A SMA actuator is considered as a controlled object, an operator theoretic control method by using the SMA actuator model is proposed based on right coprime factorization approach. The designed nonlinear feedback control system is compensated the hysteresis effect and the slow response of the SMA actuator. As a stabilization control method for the variable length link, the energy-based control approach using the kinetic energy of rotation and the potential energy of the link is employed for the purpose of obtaining the reference input of the feedback system. From the simulation results, the effectiveness of the proposed method to suppress the sway angle of a variable length link by using a SMA actuator is illustrated.

For loosely coupled wireless power transfer systems, strongly magnetic coupling is becoming a viable scheme to realize power transfer over medium distances. The method using DC-DC circuit is regarded as the most promising way to realize impedance matching for such systems. However, due to the nonlinear nature of rectifying circuit, it is difficult to track the optimal load accurately and the robust stability can not be guaranteed. Based on the above considerations, one tracking operator-based optimal load control method is proposed in this paper. The proposed control system can track the optimal load with high accuracy even when the output load varies, the tracking performance and stability can be verified. Moreover, the robust stability is considered using operator theory. Finally, simulation results are presented to confirm the effectiveness of the proposed control scheme.

Fractional calculus is defined by expanded integer order integration and differentiation. Most of research results describing mathematical modeling by fractional calculus deal with physical or electrical phenomena, not heat transfer phenomenon. Also, there is no research result that describes the effectiveness of operator based nonlinear controller with fractional calculus. In this paper, the mathematical modeling of heat transfer phenomenon of process control system by fractional calculus is proposed. Also, an operator based nonlinear control system with fractional calculus is designed. Then, we utilized a fractional PID controller as the tracking controller. Finally, simulation is conducted to verify the effectiveness of the proposed fractional controller. We confirm that the robust stability against the uncertainty of fractional dynamics is guaranteed.

Proportional-integral-derivative (PID) control schemes have been widely used in most industrial control systems. However, it is difficult to determine a suitable set of PID gains because most industrial systems have nonlinearity. On the other hands, the cerebellar model articulation controller (CMAC) classified as neural networks has been proposed, and design scheme of an intelligent PID controller has been proposed. However, the CMAC-PID controller has a problem that CMACs used as PID tuners must be trained in an online manner to get their optimal weights. In order to train CMACs in an offline manner, a combination of CMAC learning and a fictitious reference iterative tuning (FRIT) scheme, which is called CMAC-FRIT scheme, has been proposed in our previous research, and an effectiveness of the method has been evaluated only by simulations. FRIT is a scheme to determine control parameters of linear controllers by using a set of experimental data. According to the CMAC-FRIT scheme, a CMAC-PID tuner can be trained in an offline manner by using a set of operating data. In this research, the proposed CMAC-PID controller is implemented and applied to a magnetic levitation device.

In this paper, operator-based robust control for nonlinear uncertain system is considered by using robust right coprime factorization approach. In details, the effects from uncertainties for nonlinear uncertain systems are analyzed through internally stable control design. For the stabilizing system, an internal model control (IMC) like operator-based control design is proposed. Based on the proposed design scheme, the effect from uncertainties is eliminated. By using operator theory-based approach, the system is robustly stable and the desired tracking performance can be realized. Finally, the effectiveness of the proposed design scheme is demonstrated by a simulation example. Note to Practitioners-Robust stability and tracking problems are two well-known topics in the field of control, which play an important role in real application and attract the researchers' attention all the time. Especially, these issues on nonlinear systems still remain challenging owing to complexity and the nonlinear characteristic property. Another challenging issue in nonlinear control is the effects from uncertainties. Motivated by these issues, robust control design by using operator theoretic-based approach has caused for concern owing to its effectiveness in dealing with nonlinearity and uncertainties. This paper provides a survey of the research in the above issues that would help practitioners in controlling nonlinear uncertain systems and in understanding the advantages of operator-based approach.