JP-7855534-B2 - State detection device and system

JP7855534B2JP 7855534 B2JP7855534 B2JP 7855534B2JP-7855534-B2

Inventors

大島俊
山本敬亮
小野豪一

Assignees

株式会社日立製作所

Dates

Publication Date: 20260508
Application Date: 20230215
Priority Date: 20220518

Claims (16)

Multiple sensors, A calculation unit that outputs data detected by the aforementioned multiple sensors, Equipped with, The aforementioned arithmetic unit, A processing unit that performs data processing is provided, The aforementioned processing apparatus is The digitized time-series signal data from the aforementioned multiple sensors is converted into data relating to spectral intensity. Based on the aforementioned spectral intensity data, a pseudo-image is generated. The aforementioned simulated image is analyzed to output the classification result of the equipment's condition. The aforementioned processing apparatus is The transformation is performed by performing a discrete Fourier transform or fast Fourier transform on the data of the time-series signal, then nonlinearly transforming the values corresponding to the spectral intensity, and finally quantizing the expression word length of the spectral intensity. A state detection device characterized by the following features .
A state detection device according to claim 1 , The aforementioned processing apparatus is The aforementioned pseudo-image is analyzed using a convolutional neural network. A state detection device characterized by the following features.
A state detection device according to claim 1 , The aforementioned processing apparatus is In the above conversion, a process is performed to replace some of the values of the data relating to spectral intensity. A state detection device characterized by the following features.
A state detection device according to claim 1 , The aforementioned multiple observations Multiple sensor groups are configured, each containing a different type of sensor. The aforementioned group of observations is They are each placed in different inspection locations. The aforementioned processing apparatus is Multiple of the aforementioned pseudo-images are generated, In this process, the spectral intensity data for the same type of sensor is placed in the same region among the multiple pseudo-images being generated. A state detection device characterized by the following features.
A state detection device according to claim 1 , The aforementioned multiple observations Multiple sensor groups are configured, each containing a different type of sensor. The aforementioned group of observations is They are each placed in different inspection locations. The aforementioned processing apparatus is Multiple of the aforementioned pseudo-images are generated, In this process, the spectral intensity data for sensors placed at the same inspection location is placed in the same region among the multiple pseudo-images generated. A state detection device characterized by the following features.
A state detection device according to claim 2 , The aforementioned processing apparatus is Based on the output of the aforementioned convolutional neural network, the time width or period of the Discrete Fourier Transform or Fast Fourier Transform is set. A state detection device characterized by the following features.
Multiple sensors, A calculation unit that outputs data detected by the aforementioned multiple sensors, Equipped with, The aforementioned arithmetic unit, A processing unit that performs data processing is provided, The aforementioned processing apparatus is The digitized time-series signal data from the aforementioned multiple sensors is converted into data relating to spectral intensity. Based on the aforementioned spectral intensity data, a pseudo-image is generated. The aforementioned simulated image is analyzed to output the classification result of the equipment's condition. The aforementioned processing apparatus is The above transformation is performed by the Discrete Fourier Transform or the Fast Fourier Transform, The aforementioned processing apparatus is The aforementioned pseudo-image is analyzed using a convolutional neural network. The status detection device is The system further comprises a camera for acquiring camera images and an input device for the user, The aforementioned processing apparatus is The convolutional neural network processes the camera image or pseudo-image selected by the user. A state detection device characterized by the following features .
Sensors and, A calculation unit that outputs the data detected by the aforementioned sensor, Equipped with, The aforementioned arithmetic unit, A processing unit that performs data processing is provided, The aforementioned processing apparatus is The digitized time-series signal data from the aforementioned sensor is converted into data relating to spectral intensity. Based on the aforementioned spectral intensity data, a pseudo-image is generated. The aforementioned simulated image is analyzed to output the classification result of the equipment's condition. The aforementioned processing apparatus is The pseudo-image is generated by repeatedly arranging multiple rows of data relating to the spectral intensity corresponding to the aforementioned sensor. A state detection device characterized by the following features.
A state detection device according to claim 8 , The aforementioned processing apparatus is The above transformation is performed by the Discrete Fourier Transform or the Fast Fourier Transform. A state detection device characterized by the following features.
A state detection device according to claim 8 , The aforementioned processing apparatus is The aforementioned pseudo-image is analyzed using a convolutional neural network. A state detection device characterized by the following features.
A state detection device according to claim 9 , The aforementioned processing apparatus is The aforementioned pseudo-image is analyzed using a convolutional neural network. A state detection device characterized by the following features.
Multiple sensors, A calculation unit that outputs data detected by the aforementioned multiple sensors, Equipped with, The aforementioned arithmetic unit, A processing unit that performs data processing is provided, The aforementioned processing apparatus is The digitized time-series signal data from the aforementioned multiple sensors is converted into data relating to spectral intensity. Based on the aforementioned spectral intensity data, a pseudo-image is generated. The aforementioned simulated image is analyzed to output the classification result of the equipment's condition. The aforementioned processing apparatus is The transformation is performed by orthogonally transforming the data of the time-series signal, then nonlinearly transforming the values corresponding to the spectral intensity, and finally quantizing the expression word length of the spectral intensity. The orthogonal transformation is performed using at least one of the following: discrete cosine transform, discrete sine transform, discrete Fourier transform (which is slower than the Fast Fourier Transform), or Walsh-Hadamard transform. A state detection device characterized by the following features .
Multiple sensors, A calculation unit that outputs data detected by the aforementioned multiple sensors, Equipped with, The aforementioned arithmetic unit, A processing unit that performs data processing is provided, The aforementioned processing apparatus is The digitized time-series signal data from the aforementioned multiple sensors is converted into data relating to spectral intensity. Based on the aforementioned spectral intensity data, a pseudo-image is generated. The aforementioned simulated image is analyzed to output the classification result of the equipment's condition. The aforementioned processing apparatus is The aforementioned transformation is performed by combining at least two of the following: discrete cosine transform, discrete sine transform, discrete Fourier transform, fast Fourier transform, and Walsh-Hadamard transform. A state detection device characterized by the following features.
Multiple sensors, A calculation unit that outputs data detected by the aforementioned multiple sensors, Equipped with, The aforementioned arithmetic unit, A processing unit that performs data processing is provided, The aforementioned processing apparatus is The digitized time-series signal data from the aforementioned multiple sensors is converted into data relating to spectral intensity. Based on the aforementioned spectral intensity data, a pseudo-image is generated. The aforementioned simulated image is analyzed to output the classification result of the equipment's condition. The status detection device is It has an allocation table that stores the allocation of orthogonal transformations used in the aforementioned transformation, The aforementioned processing apparatus is Based on the aforementioned allocation table, the transformation is performed using at least one of the following: discrete cosine transform, discrete sine transform, discrete Fourier transform, fast Fourier transform, and Walsh-Hadamard transform. A state detection device characterized by the following features.
A state detection device according to claim 14 , The aforementioned processing apparatus is The allocation table is updated based on the information input during on-chip learning. A state detection device characterized by the following features.
A system using a state detection device comprising: multiple sensors; a calculation unit that outputs data detected by the multiple sensors; the calculation unit comprising a processing unit that performs data processing; the processing unit converting the digitized time-series signal data from the multiple sensors into spectral intensity data; generating a pseudo-image based on the spectral intensity data; and analyzing the pseudo-image to output a classification result of the equipment's state . The system includes a computer that determines the combination of orthogonal transformation and image analysis processing algorithm to be used for the transformation based on the detection accuracy and resource information of the state detection device, and configures the state detection device based on the result, or sends an instruction to configure the state detection device based on the result. A system characterized by the following features.

Description

This invention relates to a state detection device. The aging of infrastructure and plant facilities is progressing, making their maintenance and management a critical social issue. Consequently, there is a growing demand for automated monitoring technologies for these facilities. Therefore, in recent years, systems that detect external abnormalities (such as scratches and cracks) in facilities by applying AI image recognition processing to camera and satellite images are being put into practical use. However, methods using only images have limitations due to image resolution constraints and the fact that they can only detect external abnormalities. Therefore, there is a need for a technology that can detect the state of equipment with high accuracy and reliability by analyzing time-series signals from various sensors attached to or placed near the equipment. However, traditionally, analyzing time-series signals required manually developing specialized analysis algorithms tailored to the type of sensor, signal characteristics, and the type of state to be detected. As a result, building such systems quickly and at low cost was not easy. On the other hand, there are also known attempts to identify states by processing time-series signals from sensors using AI such as deep learning. However, generally speaking, deep learning capable of handling time-series signals faces challenges due to the difficulty in training neural networks and the significant time and effort required. Furthermore, Patent Document 1 discloses a technology for detecting equipment anomalies by arranging time-series signals from a sensor to convert them into a pseudo-RGB image, and then analyzing this image using image recognition AI employing a convolutional neural network and a support vector machine. The convolutional neural network extracts features from the image, and the support vector machine uses these features to determine whether or not an anomaly exists (binary determination). Patent Document 1 enables the application of an image recognition AI that is easy to train by replacing anomalies in time-series signals with anomalies in images. However, in the case of sensors such as vibration sensors, the phase of the sensor signal and the phase difference between sensor signals can have countless variations, resulting in countless variations in the converted images. Therefore, new challenges arise in ensuring the accuracy of training and inference. Furthermore, to promote widespread adoption in real-world settings, it is necessary to be able to build a state detection system based on low-cost edge devices with limited hardware resources. Japanese Patent Publication No. 2020-144619 A diagram showing an example of the configuration of the state detection device in the first embodiment.A diagram illustrating in detail an example of the pseudo-image generation process in the first embodiment.A diagram illustrating in detail an example of the pseudo-image generation process in the first embodiment.A diagram showing an example of the configuration of the state detection device in the second embodiment.A diagram showing an example of a convolutional neural network used in the second embodiment.A diagram illustrating an example of the learning process for the convolutional neural network used in the second embodiment.A diagram showing an example configuration of the state detection device in the third embodiment.A diagram illustrating an example of the effect of nonlinear quantization in a state detection device.A diagram showing an example configuration of the state detection device in the fourth embodiment.A diagram showing an example of pseudo-image generation processing using a strength-processing unit.A diagram showing an example of pseudo-image generation processing using a strength-processing unit.A diagram showing an example configuration of the state detection device in the fifth embodiment.A diagram illustrating an example of multiple pseudo-images generated by a multiple pseudo-image generation unit.A diagram illustrating an example of multiple pseudo-images generated by a multiple pseudo-image generation unit.A diagram showing an example configuration of the state detection device in the sixth embodiment.A diagram illustrating an example of the operation of the detection and control unit.A diagram showing an example configuration of the state detection device in the seventh embodiment.A diagram illustrating an example of pseudo-image generation processing in the eighth embodiment.A diagram illustrating an example of a pseudo-image.A diagram illustrating an example of a pseudo-image.A diagram showing an example configuration of the state detection device in the ninth embodiment.A diagram showing an example configuration of the state detection device in the ninth embodiment.A diagram showing an example configuration of the state detection device in the ninth embodiment.A diagram showing an example of the system configuration in the t