The static output feedback stabilization problem has a basic and simple structure. However, this problem is NP-hard and difficult to solve. In this study, we apply the method of deep unfolding to design a static output feedback control system. We then investigate how the performance of algorithms based on deep unfolding for static output feedback stabilization depends on the hyperparameters of the optimizer. In particular, we provide a policy for choosing appropriate hyperparameters by evaluating the algorithms' stabilization success rates and learning times for various optimizers, learning rates, loss functions, and discretization periods.

After the huge earthquake many steel structures were constructed using frame welded joints of welded construction and welded base. Steel structures are considered earthquake resistance structures however, many steel structures are constructed using frame-welded joints of fillet welded construction and welded column bases. These weld joints could result to have low capacity to absorb energy during earthquakes. Therefore, the application of structural health monitoring to the maintenance of infrastructures is extensively expected for new technology. However, in order to achieve practical utilization of new sensors, both accumulation of data and evaluation method to derive information about structural conditions from the measured data is highly required. To solve such a problem, we would like to measure the structure by health monitoring using a piezoelectric composite sensor. In this paper, measurements output voltage because using piezoelectric composite sensors, we recorded the sensor characteristics during measuring robot measurement because of changes in the thickness and shape of the base plate of the piezoelectric composite sensor. Structural FEM for sensor analysis is also introduced to evaluate the mechanism and influence of various environmental factors on the response of the structures.

In this paper, we study the problem of finding a control input plan that maximizes a profit function based on a statistical model. We employ covariate shift adaptation so that the statistical model has a high prediction ability at important covariate values. The degree of importance is determined by the profit function. To determine a control input plan based on the statistical model, we develop an iterative method that updates the statistical model and the control input plan alternately. We illustrate the effectiveness of the iterative alternate optimization method by a charge/discharge plan problem of a battery.

In a factory production line, if abnormal signals can be automatically detected from time-series fluctuations in power consumption, the defect rate of products can be reduced, and productivity can be improved. Individual period estimation is often needed to detect the abnormals because deviations from the mean period provide valuable information about the failures and the abnormals. This study proposes an online system that estimates individual cycles and detects anomalies from quasi-periodic time series data represented by energy consumption data. The proposed system extracts local patterns by convolution and identifies individual cycles and abnormal signals based on similarity vectors calculated by an attention mechanism. Experimental results show that the proposed system outperforms existing methods on several simulated data sets.