TWM587509U - Artificial intelligence sensor device - Google Patents

Abstract

An artificial intelligence sensor device having a body fixed to a DUT, the body comprising: at least one sensor module; a communication module; and a control module with at least one memory unit and a computing unit, wherein the control module is coupled to the at least one sensor module and the communication module, and the computing unit receives at least one physical sensing data collected by the at least one sensor unit which comparing with a default health state model stored in the memory unit, an automatic identification function of an abnormal state is provided.

Description

Artificial intelligence sensor device

This case relates to the field of control systems, in particular to a sensor device with artificial intelligence.

With the improvement of production technology, the scale of the equipment in the factory is getting larger and larger, the relationship between the systems is getting closer, and the equipment itself is becoming more and more complicated. However, due to long-term operation, problems such as mechanical wear, changes in lubrication conditions, and misalignment will affect the operation of the equipment.

Therefore, if the equipment does not have obvious problems (such as irregular vibration) for a long time during operation, although it can still maintain normal operation in the short term, it will definitely cause equipment damage in the long run and affect its performance. Therefore, if a possible failure of the equipment can be predicted at an early stage, the equipment will be repaired and maintained immediately. Once the failure occurs, it will even cause downtime or industrial safety incidents, and the economic losses caused will be considerable.

Known technologies such as the Republic of China I380548 "motor abnormality detection and protection device and method" patent are characterized by including: a motor abnormality detection and protection device, electrically connected to a motor, including: a sensing unit, which senses the A state of the motor and generating at least a first sensing signal and a second sensing signal output; an abnormality judging unit electrically connected to the sensing unit, the abnormality judging unit having a first preset value, wherein when When the amount of variation between the received first sensing signal and the second sensing signal exceeds the first preset value, it is determined as an abnormal state, and a warning signal output is generated; and a control unit is electrically connected to the The abnormality judgment unit is configured to control the motor to stop running after receiving the warning signal and according to the warning signal.

However, the patented architecture uses general reference conditions such as preset voltage, current, or speed variations as detection protection for the judgment of transient abnormal changes in the motor, and has the following disadvantages:

First, the preset reference conditions are limited, and it is impossible to adjust the judgment based on the particularity of individual equipment or user needs.

Second, it cannot be applied to a variety of devices (such as a semiconductor photomask transfer box) and cannot be combined with a variety of sensing units (such as a sound sensor and a temperature / humidity sensor module).

In addition, the prior art attempts to meet the above requirements. It uses wireless communication to transmit data collected by sensors to a gateway or cloud platform for calculation and establishment of a model, and then executes the model as a reference condition judgment on the device side. However, it also caused a delay in the speed of transmission and a security problem with data being stolen. Therefore, there is an urgent need in the art for a novel sensor device.

One of the purposes of this case is to disclose an artificial intelligence sensor device whose computing unit receives at least one physical quantity sensing data collected by at least one sensor unit for comparison with a preset health state model stored in a memory unit. Provides automatic identification of abnormal conditions.

Another purpose of this case is to disclose an artificial intelligence sensor device, the sensor module of which has a sensor sampling rate setting unit, which can set the sampling rate of the sensor module according to the needs of the user.

Another object of this case is to disclose an artificial intelligence sensor device having an information processing device for receiving the at least one physical quantity sensing data transmitted by a communication module, and performing data marking, feature extraction, and establishing a preset health state model. Calculation, and then save the preset health state model to the memory unit.

Another purpose of this case is to disclose an artificial intelligence sensor device that can be adjusted according to user needs to adjust a preset health state model. By monitoring the health status of the object to be measured and the quality of the processed product at any time, it provides automatic identification of abnormal conditions. In order to achieve the effects of preventive maintenance, reduce the loss caused by downtime, reduce industrial safety incidents, improve yield and productivity, and achieve a better user experience.

In order to achieve the foregoing object, an artificial intelligence sensor device is proposed, which has a body that can be fixed on an object to be measured. The body includes: at least a sensor module; a wireless communication module; and a control module. Including at least a memory unit and an arithmetic unit, wherein the control module is respectively coupled to the at least one sensor module and the communication module, and the arithmetic unit receives the at least one sensor unit The collected at least one physical quantity sensing data is used for comparison with a preset health state model stored in the memory unit, in order to provide an automatic identification function of an abnormal state.

In one embodiment, the control module is a control module selected by a group consisting of a micro control unit, a field programmable logic gate array, or a micro signal processor.

In one embodiment, the sensor module is composed of a gas sensor module, a sound sensor module, a gyroscope sensor module, a temperature / humidity sensor module, a pressure sensor module, and an acceleration sensor module. , A magnetic field sensor module, a current sensor module and a physiological signal sensor module.

In one embodiment, the sensor module further includes a sensor sampling rate setting unit.

In one embodiment, the communication module is a wired communication module or a wireless communication module, the wired communication module is an RS-485 communication module, and the wireless communication module is a 3G, 4G, 5G, WiFi, Bluetooth or low-power WAN module.

In one embodiment, it further includes an information processing device to receive the at least one physical quantity sensing data transmitted by the communication module, and perform data marking, feature extraction, and operations for establishing the preset health state model.

In one embodiment, the preset health state model is stored in the memory unit through the communication module or an external memory.

In one embodiment, the body further includes a power supply unit, which is an AC power source, a DC power source, or a lithium-ion battery.

In one embodiment, the control module further includes a light emitting unit, and the light emitting unit is a light emitting diode unit.

In an embodiment, the communication module further has a function of transmitting the abnormal state of the object to be tested to a handheld device, an information device, a gateway, a programmable logic controller, or a cloud platform. .

In order to enable your review committee to better understand the structure, characteristics and purpose of this case, the drawings and detailed description of the preferred embodiments are attached as follows.

Please refer to Figs. 1a to 1b together, wherein Fig. 1a shows a schematic diagram of the artificial intelligence sensor device of a preferred embodiment of the case, and Fig. 1b shows a communication module of the artificial intelligence sensor device of a preferred embodiment of the case Schematic diagram of the use case of transmitting the abnormal state of the object under test.

As shown in FIG. 1 a, the artificial intelligence sensor device of the present case has a body 10 that can be fixed on an object to be measured 200.

The main body 10 includes at least one sensor module 100; a communication module 110; and a control module 120.

The object to be measured 200 is, for example but not limited to, a motor, a machine tool, a wearing device, or an alarm.

The sensor module 100 is, for example, but not limited to, a gas sensor module, a sound sensor module, a gyroscope sensor module, a temperature / humidity sensor module, a pressure sensor module, and an acceleration sensor module. , A magnetic field sensor module, a current sensor module and a physiological signal sensor module.

The communication module 110 is a wired communication module or a wireless communication module. The wired communication module is, for example but not limited to, an RS-485 communication module; the wireless communication module is, for example but not limited to, a 3G, 4G. , 5G, WiFi, Bluetooth Low Energy (BLE) or Low Power Wide Area Network (LPWAN) module; the low power wide area network module such as but not limited to LoRA (Long Rang), NB-IoT (Narrow Band Internet of Things) and Sigfox.

The control module 120 includes at least a memory unit 121 and an operation unit 122. The control module 120 is coupled to the at least one sensor module 110 and the communication module 110, and the operation unit 122 receives at least one physical quantity sensing data collected by the at least one sensor unit 110. For comparing with a preset health state model stored in the memory unit 121, in order to provide an automatic identification function of an abnormal state.

The control module 120 is, for example, but not limited to, a control module selected by a group consisting of a micro control unit (MCU), a field programmable logic gate array (FPGA), or a micro digital signal processor (uDSP). group.

As shown in FIG. 1b, the communication module 110 further has an abnormal state for transmitting the object 200 to be tested to a handheld device 300, an information device 400, a gateway 500, and a programmable logic controller ( Programmable Logic Controller (PLC) 600, or a Cloud Platform 700.

Please refer to FIG. 2, which is a schematic structural diagram of an artificial intelligence sensor device including an information processing device according to a preferred embodiment of the present invention.

As shown in FIG. 2, the artificial intelligence sensor device of the present case further includes an information processing device 20. The information processing device 20 is configured to receive the at least one physical quantity sensing data transmitted by the communication module 110, and perform data marking, feature extraction, and operations for establishing the preset health state model.

The preset health state model is stored in the memory unit 121 through the communication module 110 or an external memory.

The following are examples of several common applications, but not limited to them.

(I) Motor vibration detection application:

This application is to install the body of the artificial intelligence sensor device of the case on the object to be measured (a motor in this example), and use the at least one sensor module to collect at least one physical quantity sensing data (in this example, an acceleration sensor module is used And the sound sensor module collect various vibration data), and then transmit the data to the information processing device through the communication module.

Use the information processing device to perform data labeling, feature extraction, and calculation of establishing the preset health state model, and finally save the established preset health state model to the control model through the communication module or the external memory. The memory unit contained in the group.

In this way, the artificial intelligence sensor device in this case can collect the physical quantity sensing data of the object to be measured (the motor in this example) according to the at least one sensor module currently used for the preset health state stored with the memory unit. The models are compared to automatically identify different vibration states of the motor, such as whether the motor speed is normal, whether it has been impacted by external forces or foreign objects have invaded the blade. The above-mentioned vibration states can be defined according to user requirements. By monitoring the health status of the object to be measured (motor in this example) at any time, an automatic status recognition function is provided to achieve the effect of preventive maintenance and reduce the loss caused by shutdown.

(Two) vibration, pressure, temperature and humidity and current detection applications of machine tools:

This application is to install the body of the artificial intelligence sensor device of the case on the object to be measured (a machine tool in this example), and use the at least one sensor module to collect at least one physical quantity sensing data (in this example, an acceleration sensor module is used The sound sensor module collects data of various vibrations during machining, the pressure sensor module collects data of various force levels of the machine tool, and the temperature and humidity sensor module and current sensor module to collect data), and then through the The communication module transmits data to the information processing device.

Use the information processing device to perform data labeling, feature extraction, and calculation of establishing the preset health state model, and finally save the established preset health state model to the control model through the communication module or the external memory. The memory unit contained in the group.

In this way, the artificial intelligence sensor device of the present case can collect the physical quantity sensing data of the object to be measured (a machine tool in this example) according to the at least one sensor module currently used for the preset health stored with the memory unit. The state models are compared to automatically identify the different vibration states of the machine tool, the magnitude of the force and the amount of current, such as whether the machine tool is excessively vibrated during processing, whether it has been hit by an external force, whether the magnitude of the force is within a set range, and whether the amount of current is excessive. higher. All of the above will seriously affect the normal operation of the precision machine tool and produce key factors of defective products. The above-mentioned different vibration and force states can be defined according to user needs. By monitoring the health of the object to be measured (tool machine in this example) and the quality of processed products at any time, it provides automatic identification of abnormal conditions to prevent it. The effect of maintenance and the effect of improving yield and productivity.

(3) Application of vibration detection for semiconductor equipment:

This application is to install the body of the artificial intelligence sensor device in this case on the object to be measured. This example is a semiconductor photomask transfer box (Reticle SMIF Pod; RSP) or a wafer transfer box (Front Opening Unified Pod; FOUP). The at least one sensor module collects at least one kind of physical quantity sensing data (in this example, the acceleration sensor module is used to collect data of various vibrations of the semiconductor detection equipment when the photomask box or wafer transfer box is moved), and then through the communication The module sends data to the information processing device.

Use the information processing device to perform data labeling, feature extraction, and calculation of establishing the preset health state model, and finally save the established preset health state model to the control model through the communication module or the external memory. The memory unit contained in the group.

In this way, the artificial intelligence sensor device in this case can collect acceleration sensing data of the object to be measured (in this example, a semiconductor photomask transfer box or a wafer transfer box) according to the at least one sensor module currently used to communicate with the object. The preset health state model stored in the memory unit is compared, and the magnitude of different vibrations when the photomask box or wafer transfer box moves during the detection process is automatically identified, and then the situation of particles caused by the vibration is analyzed and calculated. . The above-mentioned extremely small particles are the key factors that seriously affect the normal operation of the semiconductor photomask or wafer and produce defective products. The above-mentioned different vibration states can be defined according to user needs. By monitoring the health status of the semiconductor photomask and wafer at any time during the inspection and movement, it can provide automatic identification of abnormal conditions to improve yield and productivity. effect.

(4) Environmental monitoring applications in manufacturing plants:

This application is to install the body of the artificial intelligence sensor device of the case on the object to be measured (this example is a manufacturing factory environment), and use the at least one sensor module to collect at least one physical quantity sensing data (this example is to use an acceleration sensor The module and the sound sensor module collect the vibration data of the factory during the production and manufacturing process, and the temperature / humidity sensor module and the gas sensor module collect the data of the factory environment and working conditions), and then transmit the data through the communication module To the information processing device.

Use the information processing device to perform data marking, feature extraction, and calculation of establishing the preset health state model. Finally, the established preset health state model is stored in the control model through the communication module or the external memory. The memory unit contained in the group.

In this way, the artificial intelligence sensor device of the present case can collect physical quantity sensing data of the object to be measured ((this example is the environment of a manufacturing factory) according to the at least one sensor module currently used for storing the memory with the memory unit. The preset health state model is used for comparison, and the factory automatically recognizes the vibration and changes in environmental conditions during the production and manufacturing processes, such as whether there is an industrial safety event such as an earthquake, abnormal air-conditioning operation, and leakage of unknown gas in the factory. It is a key factor that seriously affects the normal operation of the manufacturing factory and then produces a shutdown of the production line or even safety at work. The different vibration, temperature, humidity and gas states can be defined according to user needs, and the environmental conditions of the manufacturing factory are monitored at any time. Provide automatic identification of abnormal conditions to achieve the effects of preventive maintenance, improve yield and productivity, and minimize the losses caused by industrial safety incidents and downtime.

(5) Voice recognition applications of the alarm system:

This application is to install the body of the artificial intelligence sensor device of the case on the object to be measured (this example is an alarm system), and use the at least one sensor module to collect at least one physical quantity sensing data (this example is to use a sound sensor module Collect various environmental sound data), and then transmit the data to the information processing device through the communication module.

Use the information processing device to perform data labeling, feature extraction, and calculation of establishing the preset health state model, and finally save the established preset health state model to the control model through the communication module or the external memory. The memory unit contained in the group.

In this way, the artificial intelligence sensor device in this case can collect the physical quantity sensing data of the object to be measured (the alarm system in this example) according to the at least one sensor module currently used for the preset health stored with the memory unit. The state models are compared to automatically identify the state of the environment, such as whether a gunshot or window glass has been broken. The above-mentioned different voice recognitions can be defined according to user needs. By monitoring the environmental conditions at any time, an automatic recognition function of abnormal states is provided to achieve the effect of improving personal safety.

(6) Gesture and gesture recognition applications for wearable devices:

This application is to install the body of the artificial intelligence sensor device of the case on the object to be measured (this example is a wearable device, such as a smart watch or smart glasses), and use the at least one sensor module to collect at least one physical quantity sensing data ( This example uses an acceleration sensor module to collect data on various gestures and postures of a user when using a wearable device), and then transmits the data to the information processing device through the communication module.

Use the information processing device to perform data labeling, feature extraction, and calculation of establishing the preset health state model, and finally save the established preset health state model to the control model through the communication module or the external memory. The memory unit contained in the group.

In this way, the artificial intelligence sensor device in this case can collect the physical quantity sensing data of the object to be measured (the wearable device in this example) according to the at least one sensor module currently used for the preset stored with the memory unit. The health status model is compared to automatically recognize different gestures and postures of the user, such as when the user wants to watch the smart watch and make a movement of raising his wrist, walking, running, cycling, swimming, falling, etc., or The user uses the smart glasses to make a motion control menu of nodding or shaking his head. The above-mentioned different gestures and postures can be defined according to user needs, by identifying the actions and states of the wearable device user, and providing automatic recognition of abnormal states to achieve a better user experience.

Please refer to FIG. 3, which illustrates a schematic diagram of an artificial intelligence sensor device including a power supply unit according to a preferred embodiment of the present invention.

As shown in the figure, the body 10 of the artificial intelligence sensor device of the present case further includes a power supply unit 130 for supplying electric energy. The power supply unit 130 is, for example, but not limited to, an AC power source, a DC power source, or a lithium ion battery.

Please refer to FIG. 4, which is a schematic structural diagram of an artificial intelligence sensor device including a light emitting unit and a sensor sampling rate setting unit according to a preferred embodiment of the present invention.

As shown in the figure, the control module 120 of the artificial intelligence sensor device of the present case further includes a light-emitting unit 123 to provide a light-emitting warning effect of the abnormal state, such as but not limited to a light-emitting diode unit.

In addition, the sensor module 100 further has a sensor sampling rate setting unit 101, which can set the sampling rate of the sensor module 100 according to user needs.

With the design disclosed above, this case has the following advantages:
1. The artificial intelligence sensor device of the present case has a computing unit that receives at least one physical quantity sensing data collected by at least one sensor unit for comparison with a preset health state model stored in a memory unit, in order to provide anomalies Automatic identification of status.
2. The artificial intelligence sensor device in this case has a sensor module with a sensor sampling rate setting unit, which can set the sampling rate of the sensor module according to user needs.
3. The artificial intelligence sensor device of the present case has an information processing device for receiving the at least one physical quantity sensing data transmitted by the communication module, and performing operations of data labeling, feature extraction and establishment of a preset health state model, and then Save the preset health state model to the memory unit.
4. The artificial intelligence sensor device in this case can be adjusted according to the user's needs to adjust the preset health state model. By monitoring the health status of the object to be measured and the quality of the processed products at any time, it can provide automatic identification of abnormal conditions to achieve prevention. Maintenance, reduce the loss caused by downtime, reduce industrial safety incidents, improve yield and productivity, and achieve a better user experience.

What is disclosed in this case is a preferred embodiment. For example, those who have partial changes or modifications that are derived from the technical ideas of this case and are easily inferred by those skilled in the art, do not depart from the scope of patent rights in this case.

To sum up, regardless of the purpose, method and effect, this case is showing its technical characteristics that are quite different from the conventional ones, and it is first of all suitable for practical use. It is also in line with the requirements of the new type of patent. Granting patents at an early date will benefit society and feel good.

10‧‧‧ Ontology

20‧‧‧ Information Processing Device

100‧‧‧ Sensor Module

101‧‧‧Sensor sampling rate setting unit

110‧‧‧Communication Module

120‧‧‧Control Module

121‧‧‧Memory unit

122‧‧‧ Computing Unit

123‧‧‧Light-emitting unit

130‧‧‧Power supply unit

200‧‧‧ Object to be tested

300‧‧‧ handheld device

400‧‧‧ Information Device

500‧‧‧Gateway

600‧‧‧ Programmable Logic Controller

700‧‧‧ Cloud Platform

FIG. 1 a is a schematic diagram of an artificial intelligence sensor device according to a preferred embodiment of the present invention.
FIG. 1b is a schematic diagram illustrating a use case in which the communication module of the artificial intelligence sensor device transmits an abnormal state of the object to be measured in a preferred embodiment of the present invention.
FIG. 2 is a schematic structural diagram of an artificial intelligence sensor device including an information processing device according to a preferred embodiment of the present invention.
FIG. 3 is a schematic structural diagram of an artificial intelligence sensor device including a power supply unit according to a preferred embodiment of the present invention.
FIG. 4 is a schematic structural diagram of an artificial intelligence sensor device including a light emitting unit and a sensor sampling rate setting unit according to a preferred embodiment of the present invention.

Claims (10)

An artificial intelligence sensor device has a body that can be fixed on an object to be measured. The body includes:
At least one sensor module;
A communication module; and a control module including at least a memory unit and an arithmetic unit,
The control module is respectively coupled to the at least one sensor module and the communication module, and the computing unit receives at least one physical quantity sensing data collected by the at least one sensor unit for Compare with a preset health state model stored in the memory unit, in order to provide an automatic identification function of an abnormal state. According to the artificial intelligence sensor device described in the first patent application scope, the control module is a control module selected by a group consisting of a micro control unit, a field programmable logic gate array or a micro-digital signal processor. group. The artificial intelligence sensor device described in item 1 of the patent application scope, wherein the sensor module is composed of a gas sensor module, a sound sensor module, a gyroscope sensor module, a temperature / humidity sensor module, a A sensor module selected by a group consisting of a pressure sensor module, an acceleration sensor module, a magnetic field sensor module, a current sensor module, and a physiological signal sensor module. According to the artificial intelligence sensor device described in item 3 of the patent application scope, the sensor module further has a sensor sampling rate setting unit. For example, the artificial intelligence sensor device described in the first patent application scope, wherein the communication module is a wired communication module or a wireless communication module, the wired communication module is an RS-485 communication module, and the wireless communication The module is a 3G, 4G, 5G, WiFi, Bluetooth or low power wide area network module. The artificial intelligence sensor device described in item 1 of the patent application scope further includes an information processing device to receive the at least one physical quantity sensing data transmitted by the communication module, and perform data marking, feature extraction and The operation of establishing the preset health state model. The artificial intelligence sensor device according to item 6 of the scope of patent application, wherein the preset health state model is stored in the memory unit through the communication module or an external memory. The artificial intelligence sensor device according to item 1 of the patent application scope, wherein the body further includes a power supply unit, which is an AC power source, a DC power source, or a lithium ion battery. The artificial intelligence sensor device according to item 1 of the patent application scope, wherein the control module further includes a light-emitting unit, and the light-emitting unit is a light-emitting diode unit. The artificial intelligence sensor device according to item 1 of the scope of the patent application, wherein the communication module further has a method for transmitting the abnormal state of the object to be tested to a handheld device, an information device, a gateway, and a programmable logic. Controller, or a cloud platform function.