In this paper, we proposed a new diagnosis and control method for damaged system based on the Digital Twin technology. Using the proposed method, a target system can be controlled under optimum condition even when some trouble occurred in a target system. In the proposed method, a system is control based on a physical model which parameters are tuned by measured data of a target system. We applied this method to a 3D printer to verify its ability and got good result.

In this paper, we propose an equivalent viscoelastic model for non-destructive inspection of water main. First, a static loading test is performed using the buried pipe to obtain the restoring force characteristics. Second, the step response test is conducted by the step loading. Furthermore, the damping coefficient is estimated using the transient response. As a result, when the ratio of water content was 30%, the soft spring characteristics was observed, and when the ratio of water content was 40%, the hard spring characteristics was observed. The relaxation behavior is observed in the restoring force characteristic. Moreover, the linear spring constant was obtained as the 22.31×10⁶[N/m] in case of water content 40%, the linear spring constant was obtained as the 8.49×10⁶[N/m] in case of water content 30%, respectively. In addition, the curve fitting analysis is performed using the transient response. As the result, the relaxation behavior were good agreement with the exponential relaxation model and Prandtl-Eyring model.

1D Modeling Study Group of The Japan Society of Mechanical Engineers; Dynamics, Measurement and Control Division is focusing most on the formulation of 1D models. Formulation of a 1D model by applying mechanical, fluid, thermal and electrical systems based on a five-level 1D modeling method to embody the concept of 1D-CAE, which is considered to be very useful for physics formulation.

Reciprocating compressors are used to compress hydrogen gas, but it has been reported that each component wears out due to long-term continuous operation, resulting in a decrease in production. We have focused on the fact that the vibration of the compressor changes when each component wears out, and have attempted to monitor the wear by vibration. In this paper, I focused on the wear of the crosshead pin and created a mechanical model to reproduce the impact vibration caused by the phenomenon. As a result, I succeeded in generating two characteristic impact vibrations close to the experimental values in the model simulation. Using this model, I estimated the amount of wear, and although it did not match the actual amount of wear, I confirmed an increase or decrease in the amount of wear.

For comparative experiments between the vibration signal and the current signal, an intelligent fault diagnosis method based on multiclass convolutional neural network (MCNN) has been proposed to identify faults in complex rotor system. Firstly, the both signals were recorded simultaneously under steady-state for four kinds of speed. Secondly, the signal processing technique is chosen to solve the problem of modeling noise instances as true underlying relationship for MCNN. Finally, a one-versus-one and a comprehensive MCNN have been trained with both signal at various conditions individually and collectively, respectively. And the results revealed that the accuracy of the vibration signal is better than the current signal whether it is structure faults or the external bearing faults. Moreover, the fault diagnosis performance is investigated for the wide range of parameters in MCNN.

Since bearings are important components in rotating machinery equipment, it is important to regularly monitor and diagnose the condition of bearings in critical equipment in order to prevent sudden and serious accidents. In recent years, research on intelligent and automatic diagnosis technology using AI technology has been conducted for the diagnosis of abnormalities in bearings. In particular, deep learning has been attracting attention in the field of equipment diagnosis because it is an AI technology with high feature extraction capability. In this study, we proposed a method to automatically perform feature extraction and state classification using convolutional neural network (CNN), a type of deep learning, after removing noise from vibration signals measured for bearing diagnosis using statistical filters. As a result of various verification experiments, it was found that the proposed method can achieve highly accurate diagnosis of bearing abnormalities even in noisy environments.

With full cooperation from experienced diagnosticians and engineers, we have developed a system that automates rotational machinery diagnostics using multimodal deep learning method. The multimodal approach has advantage in mimicking diagnosticians’ thinking process and works well with current deep learning frameworks. We discuss how the multimodal model is built and how it behaves on robustness testing. We also introduce originally developed wireless acceleration sensor device and cloud-based application which allows us to update model online.

It is important to predict the condition of pump shaft at sliding bearing because of the wear in long time usage. However, it is difficult and not clearly known the relationship between the vibration signal and severity of wear in pump shaft. In this study, in order to monitor the condition of the slide bearings, an experimental device for the rotor systems with vertical shaft and horizontal shaft supported with water was developed. Long-term rotation experiments were conducted for both shafts. In these experiments, the torque, bearing temperature, and displacement in the x and y directions were measured and recorded. Torque and bearing temperature signals were used to label the condition as “normal” or “abnormal” state. Then, various features were calculated from displacement signals. Support vector machine (SVM) model was trained by some specific labeled feature signals. The trained SVM model could identify the state and transition of state in other time zones.

In petrochemical plants, equipment for efficiently compressing low-molecular-weight gases is widely used. Among them, the reciprocating compressor has a long piston rod, which enables efficient compression at both ends of the piston. Since the inside of the compressor is covered by a casing, it is not possible to grasp the inside condition of the actual machine. In this study, the acceleration of the crosshead, piston rod, and piston were collected by using a small experimental machine that was manufactured based on the vibration characteristics of the actual machine so that the internal conditions could be grasped. The simulation results using a 12-degree-of-freedom dynamic model of the small experimental machine were used to identify parameters of the narrow-band time series data using the frequency considered to be the piston rod bending vibration during operation, in order to reproduce the shock fluctuation of one crankshaft revolution obtained from the experiment. The acceleration of the simulation results using the parameters obtained from the results was able to reproduce the characteristic shock fluctuations at two locations of the crankshaft single rotation.

This paper concerns the detection method of contact-type failure based on nonlinear piezoelectric impedance modulation. When the contact condition of a contact-type failure is fluctuated by the low-frequency vibration, the amplitude and phase of high-frequency vibration are modulated in synchronization with low-frequency vibration (nonlinear wave modulation). Contact-type failure detection based on nonlinear piezoelectric impedance modulation is the method that these modulations is measured as the fluctuation of electromechanical impedance. In this paper, the measurement of fluctuation of instantaneous electromechanical admittance is performed as investigation of the nonlinearity between low-frequency vibration and fluctuation of electromechanical impedance. Firstly, the overview of nonlinear piezoelectric impedance modulation and the single-degree-of-freedom model are introduced. Secondly, the measurement of frequency characteristics of instantaneous electromechanical admittance is performed. The relationship between the force frequency and admittance fluctuation is revealed.

Recently, the deteriorated buried water mains are increased, as a result, many accidents (leakage, burst and so on) are occurred. Owing to prevent accidents, efficient maintenance based on Non-Destructive Inspection (NDI) is required. However, there is no existing effective technology for evaluate buried main. Therefore, it is necessary to expose water mains in the inspection by the past technology. In this paper, we proposed the no-exposure diagnosis method through the connecting pipe. To develop an inspection equipment for estimating the deteriorated condition of water mains, the in-plane bending deformation vibration modes are induced by white noise excitation through connecting pipes. Furthermore, the finite element analysis was performed for understanding the eigenfrequencies and the modal shape of the coupled vibratory system. Moreover, the simplified model was derived based on the coupled ring model, the eigenfrequencies by an above model were good agreement with the calculation values by finite element analysis.

Recently, the landslide disaster is frequently occurred caused by frequently heavy rainfall. The before detection technology of the landslide is required to prevent the human live. In previous work, the internal soil state estimation technology based on soil-structure coupling was already proposed by an author. Above technique is composed from the elastic pile and electromagnetic exciter, in addition, the eigen mode of low frequency band under 250Hz lower was used. However, there is a problem about measurement resolution in the low frequency band. In this paper, consideration for expansion to high frequency band of proposed method was considered. First, the new vibration pile structure was proposed. In addition, the fundamental operation verification was considered in Laboratory bench set. Furthermore, operational modal analysis was performed in actual field.

In this paper, the free rocking vibration behavior of one-degree-of-freedom systems is considered for realize the before detection of furniture tipping. First, the sensing module is fabricated using the angular velocity sensor focused on the furniture rocking motion. Moreover, the free rocking vibration experiment was performed using the above sensing module. In addition, the numerical simulation was conducted by 4th Runge-Kutta method. As the result, the measurement signal using the proposed sensing module was good agreement with the numerical simulation. Furthermore, the detection algorithm of the rocking vibration is derived by the correlation coefficient between the measurement signal and the numerical simulation prediction signal of Runge-Kutta method.

In recent years, the deterioration of water main to be social problem, because of occurrence of water leakage and burst accidents. Owing to prevent the accidents, it is important to condition assessment for water main. In previous work, Non-Destructive Inspection (NDI) based on in-plane bending mode was proposed. However, the previous works were not considered to influence by buried situation. In this paper, the buried situation dependency of in-plane bending mode frequency is considered focused on the instantaneous frequency analysis. As the result, the nonlinear free oscillation behavior was observed in the imitate thinning pipe in buried situation.

Diagnosis of the condition of rolling bearings is important for maintaining the machine operating rate. In recent years, research on bearing condition diagnosis methods using machine learning has been actively conducted. In these diagnoses, the failure state data obtained from bearings that have been given initial defects such as indentations in advance are often used. However, in actual condition diagnosis of bearing, it is important to detect abnormalities at the initial stage, so it is necessary to use bearing data from the normal condition to the occurrence of damage in machine learning. In this study, an experimental system that can apply an arbitral motion with any frequency to the rotating shaft was developed. A vibration experiment of deep groove ball bearings was conducted. The 3 axes (x,y,z) acceleration, 2 axes (x,y) displacement, and acoustic emission signal were measured and analyzed. Measures of envelop spectrum and kurtosis were calculated and observed.

In recent years, owing to the effects of the global coronavirus pandemic and decline of the working-age population, production efficiency and labor productivity are being actively improved in manufacturing and production sites by the introduction of digital technologies. Given these trends, we have developed a system to determine the manufacturing statuses of machines at the early stage to aid production efficiency within a manufacturing company. To detect failures and abnormalities in hydraulic presses at the early stages, we have built and commenced operation of a failure diagnostic system using machine learning. A sensor that measures vibration acceleration is attached to the cylinder, oil pump, and pump drive motor, which are the main parts of the press machine; signals are continuously collected, and the signal for normal operation is modeled on the basis of the standard deviation, crest factor, and maximum signal values. Failures and abnormalities are also detected on the basis of the amount of temporal changes and deviations of the model. Thus, the possibility to predict the time to failure by monitoring such variations is shown.

Recently researches are being conducted to generate a proper acoustic emission (AE) wave numerically and to compare it theoretically. In this paper the generation of AE signal is theoretically analyzed for a homogeneous material applying surface load. Aluminum is selected as the homogenous material. Simulations for the validations of the developed computational algorithms are performed based on the established theories. Under similar conditions, AE wave generation and propagation experiments are also performed on the aluminum surface to validate the presented theory. The analysis for both the detected and the simulated waves are shown here. Among the AE wave parameters, the characteristics of maximum amplitude is selected for the comparison of the experimental and the simulated waves. The simulated wave is modified considering the influence of different transfer functions on the theoretical wave. The maximum amplitude of both the simulated and the experimental waves attenuates with the increase of source to receiver distance for different input loads.

Since electric motors (including generators) are used and play crucial roles in various industries, it is important to diagnose their conditions appropriately to avoid troubles in advance. A motor is a device that converts the electric energy into the rotational (mechanical) energy. Electric motors consist of multisystem and operate through the interaction between the magnetic field and electric current in windings to generate the torque applied on the motor shaft. So, it is effective to comprehensively diagnose the motor conditions by monitoring multi-measurement data such as vibration, electronic current and temperature etc. In the previous work, we formulated the model of a condenser motor with magnetic circuit network method in order to evaluate dynamic behavior including a motor shaft, and showed its effect to electric current. In response to this result, we focus on a three-phase induction motor which is widely used in industry, and verify the model by comparing with FEM results about its electromagnetic characteristics.

For plant maintenance in the electric power industry, the method is shifted from Time Based Maintenance(TBM) to Condition Based Maintenance(CBM). CBM is carried out appropriately after continuously evaluating the condition of the equipment and making a scientific and rational judgment. As a result, an extension of regular inspection cycle and reduction of maintenance costs are investigated in the industry by optimizing the maintenance. Lubricating oil diagnosis, which is one of the diagnostic techniques of CBM, is said to be highly sensitive to failures. On the other hand, it has been difficult to apply to online monitoring. Therefore, we developed a device that distinguishes particles in oil from other particles, and found a useful index for wear evaluation by monitoring the ratio of time change in each particle size range using a contamination classification.

Elevators are widely used as the main vertical transportation in the building. In order to ensure the safe operation of the elevator, it is necessary to estimate the damage condition of each component. The rope doesn’t only transmit the driving force of the motor but also supports the load of the car. Therefore, it is particularly important to estimate the damage condition of the rope. Since the rope is damaged unevenly in the longitudinal direction, an inspector needs to identify the most damaged position of the rope. The rope moves along the whole distance of the elevator shaft, so that the inspector needs a lot of time for the conventional rope checking procedure. Therefore, an alternative method is required to reduce the rope checking time. In this paper, we propose a new method to estimate the worn-out rope diameter by using the specific information from the elevator system. The rope diameter can be derived from the rotation number of the traction sheave, which is related to the travel length of the car. The method is evaluated by experiments and we can conclude that the estimated rope diameter corresponds to the actual measurements.