Real-time control system for the hazardous status and conditions of AIoT-based workers using a combination of biosignal sensors and multilayer signal decomposition technology.

Abstract

The challenge is to enable a small number of managers to manage a large number of workers while not only eliminating the source of exposure to safety blind spots, but also to efficiently deal with, predict, and prevent hazardous situations. [Solution] The present invention relates to an AIoT-based real-time control system for worker hazardous conditions and situations utilizing a composite biosignal sensor and multilayer signal decomposition technology. More specifically, it enables the construction of a real-time control system that allows managers to monitor individual hazardous or abnormal conditions and situations of workers that may occur in various industrial sites in real time, based on a variety of composite sensors and multilayer signal decomposition technology for each individual worker.

Description

Technical Field

[0001] The present invention relates to an AIoT-based real-time control system (AIoT BASED REAL-TIME CONTROL SYSTEM OF DANGEROUS CONDITION AND SITUATION FOR WORKERS USING THE MULTI BIO SIGNAL SENSOR AND MULTILAYER SIGNAL TECHNOLOGY) that enables the construction of a real-time control system so that managers can monitor individual dangers, abnormal states, and situations of workers that may occur at various industrial sites in real time. By doing so, not only can a small number of managers manage a large number of workers while cutting off the source of exposure of the safety blind spots of the workers, but also the dangerous situations can be efficiently addressed, predicted, and prevented.

Background Art

[0002] The disaster mortality rate at various industrial sites such as shipbuilding and heavy industries, chemical factories, and construction industries ranks first among OECD member countries (four times that of the United States and 20 times that of the United Kingdom). The continuous increase in the number of disaster victims and deaths reflects a poor working environment. In fact, the actual site is in a poor working environment where a small number of safety managers have to manage a large number of workers.

[0003] On the other hand, in the case of the manufacturing industry, when one disaster death accident occurs, it corresponds to an average loss of 63 billion won, and the impact on corporate productivity and image cannot be avoided.

[0004] Therefore, the importance of the safety and health of workers in industrial sites has increased, and due to the increasing innovative requirements for the existing manufacturing environment due to the Fourth Industrial Revolution, there is a need for technology that can prevent safety accidents in advance through the immediate prior detection of abnormal signs of workers in industrial sites.

[0005] On the other hand, Patent Document 1 relates to a safety management system that can create a safe working environment by monitoring and paying attention to the surrounding environmental conditions of an industrial work site. It discloses an industrial environmental safety management system device that receives environmental sensing signals from a sensor module, determines the hazard to workers in or entering the room, and provides information for controlling entry and exit or for repairs.

[0006] However, conventional technologies such as the one described in Patent Document 1 above rely on installing temperature and gas sensors within the worker's workspace to determine the presence or absence of danger. Not only does this fail to detect the worker themselves, i.e., any physical abnormalities of the worker, but it also lacks the functionality for managers to monitor individual dangers or abnormal conditions of workers in real time by relying solely on specific single sensors such as temperature and gas sensors. Furthermore, it remains practically impossible for a small number of managers to manage a large number of workers, and there are problems in preventing safety blind spots and dangerous situations for workers. [Prior art documents] [Patent Documents]

[0007] [Patent Document 1] Republic of Korea Published Patent Gazette No. 10-2022-0167786 (December 21, 2022) "Industrial Environmental Safety Management System Device" [Overview of the project] [Problems that the invention aims to solve]

[0008] The present invention aims to solve the above-mentioned problems by enabling the construction of a real-time control system that allows managers to monitor individual dangers or abnormal conditions and situations of workers that may occur in various industrial sites in real time, based on diverse composite sensors and multi-layer signal decomposition technology for each individual worker. This will not only enable a small number of managers to manage a large number of workers while blocking the exposure of workers in safety blind spots, but also enable efficient response, prediction, and prevention of danger situations.

[0009] The problems that this invention aims to solve are not limited to those described above. Other problems that this invention aims to solve, not mentioned herein, will be clearly understood by those with ordinary skill in the art to which this invention belongs from the description below. [Means for solving the problem]

[0010] The present invention includes: safety equipment (S1) worn by a worker; attachment part (S2) provided at one end inside the safety equipment (S1) to which a biosignal measuring unit (100) is attached; a biosignal measuring unit (100) attached to the attachment part (S2) to measure the worker's biosignals; and a signal processing unit (200) that receives the biosignals measured by the biosignal measuring unit (100), determines the worker's state based on the transmitted biosignals, and outputs the result. The biosignal measuring unit includes: a plurality of biosignal measuring means (110) for measuring the worker's biosignals; a biosignal receiving unit (120) that receives the plurality of biosignals from the biosignal measuring means (110); a biosignal processing unit (130) that supplies the plurality of biosignals received from the biosignal receiving unit (120) to corresponding biosignal storage units (140); and a biosignal processing unit (130) that receives the biosignals supplied from the biosignal processing unit (130), converts them into data, and stores them. The system includes a storage unit (140) and a transmission unit (150) that communicates with the signal processing unit (200) and provides the biological signals stored in the biological signal storage unit (140) to the signal processing unit (200). The signal processing unit (200) includes an amplification unit (210) that communicates with the transmission unit (150) and amplifies the analog biological signals received through the transmission unit (150); an ADC (220) that converts the analog biological signals amplified through the amplification unit (210) into digital biological signals; a state determination unit (230) that determines the state of the worker based on the biological signals converted into digital biological signals; and an output unit (240) that communicates with the terminals of the worker and the manager and provides the worker state information determined by the state determination unit (230) to the terminals of the worker and the manager.

[0011] Here, the above-mentioned biological signals include brain waves, heart rate, motion acceleration, and temperature; the above-mentioned biological signal measuring means (110) includes a brain wave measuring sensor (112), a temperature measuring sensor (113), a heart rate measuring sensor (114), and an acceleration measuring sensor (115); and the above-mentioned biological signal storage unit (140) includes a brain wave signal storage unit (141) for digitizing and storing brain wave signals; a heart rate signal storage unit (142) for digitizing and storing heart rate signals; a motion acceleration signal storage unit (143) for digitizing and storing motion acceleration; and a temperature signal storage unit (144) for digitizing and storing temperature.

[0012] On the other hand, the state determination unit (230) may include a learning module (231) that stores and learns biosignal information measured by the biosignal measurement unit (100) over a certain period of time, and a determination module (232) that compares the biosignal information learned by the learning module (231) with the biosignal information generated in real time by the biosignal measurement unit (100) to determine whether or not there is an abnormality in the worker's state.

[0013] The judgment module (232) may include: a biosignal data storage unit (232-1) that stores learned biosignal information and biosignal information generated in real time; a change value calculation unit (232-2) that calculates a change value by substituting the biosignal information generated in real time into the learned biosignal information; and a state derivation unit (232-3) that determines the state by substituting the change value into previously learned state classification information.

[0014] Furthermore, the above-mentioned change value calculation unit (232-2) includes a first change value calculation unit (232-2a) that calculates change values ​​based on electroencephalograms; a second change value calculation unit (232-2b) that calculates change values ​​based on heart rate; a third change value calculation unit (232-2c) that calculates change values ​​based on temperature; and a fourth fluctuation value calculation unit (232-2d) that calculates change prices based on acceleration values.

[0015] On the other hand, the electroencephalogram (EEG) sensor (112), temperature sensor (113), heart rate sensor (114), and acceleration sensor (115) are each mounted on a flexible circuit board module (116), and the flexible circuit board module (116) consists of a flexible circuit board section (116a); a power supply section (116b) mounted on the flexible circuit board section (116a); a sensor mounting section (116c) for fixing each sensor mounted on the flexible circuit board section (116a); and the flexible circuit board section (116a) The system includes a wireless communication unit (116d) mounted on top of the above-mentioned sensors, which transmits signals detected from each of the above-mentioned sensors to the above-mentioned biosignal receiving unit (120); the power supply unit (116b), the sensors installed in the sensor installation unit (116d), and the wireless communication unit (116d) are interconnected through electrical wiring; the flexible substrate unit (116a) is made of a polyimide film; the wireless communication unit (116d) uses Bluetooth, Zigbee, UWB, or Wi-Fi; and the electrical wiring uses conductive fibers. [Effects of the Invention]

[0016] This invention enables the construction of a real-time control system that allows managers to monitor individual hazards or abnormal conditions and situations of workers that may occur in various industrial sites in real time, based on diverse composite sensors and multi-layer signal decomposition technology for each individual worker. This not only allows a small number of managers to manage a large number of workers while blocking exposure to safety blind spots, but also has the effect of efficiently dealing with, predicting, and preventing the occurrence of hazards. [Brief explanation of the drawing]

[0017] [Figure 1-2] This is a block diagram showing the overall configuration of an AIoT-based worker hazard status and situation real-time control system utilizing a composite biosignal sensor and multilayer signal decomposition technology according to the present invention. [Figure 3] This is a block diagram illustrating a real-time control system for worker hazardous conditions and situations on an AIoT platform utilizing a composite biosignal sensor and multilayer signal decomposition technology according to the present invention. [Figure 4] This is a block diagram embodying the biological signal measuring means according to the present invention. [Figure 5] This is a block diagram embodying the storage unit for biological signals according to the present invention. [Figure 6] This is a block diagram embodying the state determination unit according to the present invention. [Figure 7] This is a block diagram embodying the determination module according to the present invention. [Figure 8] This is a block diagram embodying the change value calculation unit according to the present invention. [Figure 9] This is a block diagram embodying the flexible substrate module according to the present invention. [Figure 10] This is a plan view showing the mounting portion to which the silicon pad according to the present invention is applied. [Figure 11] This is a side view showing the mounting portion to which the silicon pad according to the present invention is applied.

Embodiments for Carrying Out the Invention

[0018] The advantages and features of the embodiments of the present invention, and the methods for achieving them, will become clear by referring to the embodiments described in detail hereinafter together with the accompanying drawings. However, the present invention is not limited to the embodiments disclosed below, and can be embodied in various different forms. Merely, these embodiments are provided to complete the initiation of the present invention and to fully inform those with ordinary knowledge in the technical field to which the present invention pertains of the scope of the invention. The present invention is only defined by the scope of the claims. Throughout the specification, the same reference numerals refer to the same components. When it is determined that a specific description of a notification function or configuration may unnecessarily obscure the gist of the present invention in explaining an embodiment of the present invention, the detailed description thereof will be omitted. And the terms described hereinafter are terms defined in consideration of the functions in the embodiments of the present invention, and these can be changed according to the intentions or customs of users, operators, etc. Therefore, the definitions should be made based on the content throughout this specification.

[0019] Hereinafter, a preferred embodiment of the present invention will be described with reference to the attached drawings.

[0020] FIGS. 1 and 2 are block diagrams showing the overall configuration of an AIoT infrastructure worker danger status and situation real-time control system using a composite biosignal sensor and a multi-layer signal decomposition technique according to the present invention. FIG. 3 is a block diagram that embodies the AIoT infrastructure worker danger status and situation real-time control system using a composite biosignal sensor and a multi-layer signal decomposition technique according to the present invention. FIG. 4 is a block diagram that embodies the biosignal measurement means according to the present invention. FIG. 5 is a block diagram that embodies the biosignal storage unit according to the present invention. The embodied block diagram is a block diagram based on judgment. FIG. 8 is a block diagram that embodies the change value calculation unit according to the present invention. FIG. 9 is a block diagram that embodies the flexible substrate module according to the present invention. FIG. 10 is a plan view showing the mounting part to which a silicon pad according to the present invention is applied. FIG. 11 is a side view showing the mounting part to which a silicon pad according to the present invention is applied.

[0021] Referring to FIGS. 1 to 11, the AIoT infrastructure worker danger status and situation real-time control system using a composite biosignal sensor and a multi-layer signal decomposition technique according to the present invention includes a safety equipment (S1), a mounting part (S2), a biosignal measurement unit (100), and a signal processing unit (200).

[0022] The above-mentioned safety equipment (S1) can include all the equipment worn by the worker during work. As an example, it can include work clothes, safety helmets, safety shoes, gloves, masks, etc. In addition, various equipment according to the type of work and the working environment can be applied.

[0023] The above-mentioned attachment part (S2) serves as an intermediary for attaching the biosignal measurement unit (100) to safety equipment (S1). When direct attachment of the biosignal measurement unit (100) to safety equipment (S1) is difficult, the attachment part (S2) can be used as an intermediary. For example, for work clothes and safety helmets, a band in a form corresponding to the jacket and trousers or the inside of the safety helmet can be used. In the case of safety shoes, an insole can be used, and in the case of gloves and masks, the inner lining can be used. The above-mentioned biosignal measurement unit (100) is attached to the above-mentioned attachment part (S2) and, as a means of measuring the biosignals of a worker, a biosignal measurement means (110), a biosignal receiving unit (120), and after the biosignal receiving unit (120) receives multiple biosignals measured by the biosignal measurement means (110), the multiple biosignals measured by the biosignal measurement means (110) are each received by the receiving unit (110), and then received through the biosignals. Then, the biological signals supplied from the biological signal processing unit (130) are received by the biological signal storage unit (140), converted into data, and stored. The biological signals stored in the biological signal storage unit (140) are then provided to the signal processing unit (200) via a transmitting unit (150) that communicates with the signal processing unit (200). Here, the biological signals include brain waves, heart rate, motion acceleration, and temperature, and are not necessarily limited to these signals; a variety of biological signals can be collected depending on the type of work and the work environment.

[0024] On the other hand, if the biosignal measurement means (110) is an electroencephalogram (EEG), heart rate, motion acceleration, and temperature, it may include an EEG measurement sensor (112), a temperature measurement sensor (113), a heart rate measurement sensor (114), and an acceleration measurement sensor (115).

[0025] When considering this, the above-mentioned biosignal storage unit (140) may include an electroencephalogram (EEG) signal storage unit (141) for digitizing and storing EEG signals, a heart rate signal storage unit (142) for digitizing and storing heart rate signals, a motion acceleration signal storage unit (143) for digitizing and storing motion acceleration signals, and a temperature signal storage unit (144) for digitizing and storing temperature signals.

[0026] In other words, the signals measured by each sensor can be broken down and stored for each individual item, and this can be used to determine the worker's condition, thereby enabling the construction of a multi-layer signal decomposition technology platform system.

[0027] On the other hand, the electroencephalogram (EEG) sensor (112), temperature sensor (113), heart rate sensor (114), and acceleration sensor (115) can be mounted on a flexible circuit board module (116) to improve coupling efficiency. More specifically, the flexible circuit board module (116) includes a power supply unit (116b), a sensor mounting unit (116c) for fixing each sensor, and a wireless communication unit (116d) for transmitting signals detected from each sensor to the biosignal receiving unit (120). The power supply unit (116b), the sensors installed in the sensor mounting unit (116d), and the wireless communication unit (116d) are interconnected through electrical wiring.

[0028] Here, the flexible substrate portion (116a) is fitted with a polyimide film to provide flexibility, the wireless communication portion (116d) is fitted with Bluetooth, ZigBee, UWB (Ultra Wide Band), or Wi-Fi, and the electrical wiring can be made of conductive fibers, in which case known fibers containing or coated with carbon nanotubes or metal particles can be used.

[0029] The above-mentioned signal processing unit (200) is a means for receiving the biological signal measured by the biological signal measurement unit (100) and determining the worker's condition based on the transmitted biological signal, and includes an amplification unit (210), an ADC (Analog Digital Converter, 220), a condition determination unit (220), and an output unit (240). More specifically, it receives an analog biological signal through the amplification unit (210) which communicates with the above-mentioned transmission unit (150), amplifies it, and then converts it into a bio-signal converted through the amplified signal to the ADC (210), and provides worker condition information determined by the above-mentioned condition determination unit (230) to the terminals of the worker and the manager through the output unit (240) which communicates with the terminals of the worker and the manager.

[0030] On the other hand, the state determination unit (230) determines the worker's state through the learning module (231) and the determination module (232). More specifically, it stores the biosignal information measured by the biosignal measurement unit (100) over a certain period of time and learns it through the learning module (231). Then, the determination module (232) compares the biosignal information learned by the learning module (231) with the biosignal information generated in real time by the biosignal measurement unit (100) to determine whether there is an abnormality in the worker's state.

[0031] The judgment module (232) includes a biosignal data storage unit (232-1), a change value calculation unit (232-2), and a state derivation unit (232-3). More specifically, the learned biosignal information and the biosignal information generated in real time are stored through the biosignal data storage unit (232-1), the change value calculation unit (232-2) substitutes the biosignal information generated in real time into the learned biosignal information to calculate the change value, and then the state derivation unit (232-3) determines the state by substituting the change value into the previously learned state classification information.

[0032] Here, the classification information mentioned above is pre-learned information that, depending on the change value, classifies the state as state A if the range of the change value is A. For example, if the change value is 10 or more, it means information that classifies the state as "dangerous" or "abnormal deep pulse".

[0033] Furthermore, when the biological signals are electroencephalograms, heart rate, motion acceleration, and temperature, the above-mentioned change value calculation unit (232-2) may include a first change value calculation unit (232-2a) that calculates the change value based on electroencephalograms, a second change value calculation unit (232-2b) that calculates the change value based on heart rate, and a fourth change value calculation unit (232-2d) that calculates the change value based on temperature.

[0034] On the other hand, the learning module (231) and the decision module (232) can utilize publicly available artificial intelligence (AI) modules. In other words, it is possible to construct an AIoT (Artificial Intelligence of Things) platform system through the integration of the various sensors and the AI ​​modules mentioned above.

[0035] Here, the artificial intelligence module described above can be a CNN (Convolutional Neural Networks) model that includes an Inception module in its GAP (Global Average Pooling) layer. The artificial intelligence module can calculate the desired data by learning weights of multiple inputs through deep learning. Furthermore, various artificial intelligence network models such as RNN (Recurrent Neural Network), DNN (Deep Neural Network), and DRNN (Dynamic Recurrent Neural Network) can be used for such learning.

[0036] Here, RNN is a deep learning technique that considers current and past data simultaneously, and a cyclic neural network (RNN) refers to a neural network in which the connections between units constituting the artificial neural network form a directed cycle. Furthermore, various methods can be used to construct a cyclic neural network (RNN), but representative examples include the Fully Recurrent Network, Hopfield Network, Elman Network, ESN (Echostate network), LSTM (Long Short Termemory network), Bi-directional RNN, and CTRNN (Continuous-time RNN). In addition, methods such as slope descent, Hessian Free Optimization, and Global Optimization Method can be used to train a cyclic neural network (RNN).

[0037] Furthermore, general CNNs have the problem of high processing speed and computational complexity. In this invention, iterative calculations are required through immediate additional data, so a lighter algorithm is needed. Therefore, a GAP (Global Average Pooling) layer and an Inception module can be applied to the CNN model. Generally, CNNs have many convolutional layers, resulting in a large number of filters. A large number of filters means that a large number of feature maps are accumulated. In other words, the dimensionality of the CNN is very large. To handle high dimensions, a large number of parameters capable of handling that dimension are required. However, if there are too many parameters, problems such as overfitting during training may occur. Therefore, a method is needed to reduce the dimensionality by reducing the number of parameters used in the filters. The layer that plays this role in a CNN is the pooling layer. In this case, GAP solves this problem by extracting representative values ​​(averages) from each feature and immediately transferring them to the classification group. As a result, it plays the role of directly associating the feature map with the category while maintaining the spatial information that the previous feature map had, so a confidence map (=feature map) can be obtained that shows which part of the obtained feature map played the main role in classifying it into the corresponding class. Because it does not require separate parameter optimization, it has the advantage of not only requiring less computation but also blocking the over-fitting problem. The Inception structure is a form in which a small neural network is added inside the main neural network, and the small internal neural network is called an Inception module. Inception modules can reduce feature maps through 1x1 convolution, and have effective characteristics in terms of processing speed and computational cost.

[0038] This allows the above artificial intelligence module to apply a CNN model that includes the Inception module to the GAP (Global Average Pooling) layer.

[0039] In other words, the present invention enables the construction of a real-time control system based on diverse composite sensors and multi-layer signal decomposition technology for each individual worker, allowing managers to monitor individual dangers or abnormal conditions and situations that may occur in various industrial sites in real time. This enables a small number of managers to manage a large number of workers while not only blocking exposure to safety blind spots, but also enabling efficient response, prediction, and prevention of danger situations.

[0040] On the other hand, a biosignal measuring unit (100) is attached to a mounting part (S2) provided at one end of the safety equipment (S1), and the mounting part (S2) and the biosignal measuring unit (100) are in continuous contact with the worker.

[0041] In other words, when considering the use of a woven material for the attachment part (S2) that has relatively high hardness or causes an uncomfortable wearing experience due to considerations such as rigidity, a silicone pad (P) can be provided on the contact surface between the attachment part (S2) and the worker to improve wearing comfort. Furthermore, since poor contact may occur in the multiple sensors included in the biosignal measurement unit (100) due to sweat during prolonged wear, multiple airflow channels (P1) can be formed in the silicone pad. In other words, the wearing comfort and breathability for the worker can be improved with the above configuration. On the other hand, in the case of the silicone pad (P) described above, it can be manufactured in foam form to further improve the wearing comfort by providing cushioning, etc. In the case of foam, it can be manufactured by adding 2 to 4 parts by weight of an inorganic filler, which is a mixture of silica and carbon black in a weight ratio of 1:1 to 2:1 as an inorganic filler to reinforce the physical properties of the silicone resin, 1 to 3 parts by weight of dicumyl peroxide as a crosslinking agent, and 2 to 5 parts by weight of azodicarbonamide as a foaming agent to 100 parts by weight of silicone resin. The above composition may reduce the above content and efficiency. The present invention is not limited to the above embodiments, and has a diverse range of applications. Without departing from the gist of the present invention as claimed, anyone with ordinary skill in the art to which the present invention belongs can carry out various modifications. [Explanation of symbols]

[0042] S1: Safety equipment S2: Mounting part P: Silicone pad P1: Pneumatic fluidized furnace 100: Biosignal measurement device 110: Biosignal measurement means 112: Electroencephalogram (EEG) measurement sensor 113: Temperature measurement sensor 114: Heart rate measurement sensor 115: Accelerometer 120: Biosignal receiving unit 130: Biosignal Processing Unit 140: Biosignal storage unit 141: Brainwave signal memory unit 142: Heart rate signal storage unit 143: Motion acceleration signal storage unit 144: Temperature signal storage device 150: Transmitter 200: Signal Processing Unit 210: Amplifier section 220: ADC 230: State determination unit 231: Learning Module 232: Decision Module 232-1: Biosignal data storage unit 232-2: Change Value Calculation Unit 232-2a: First change value calculation unit 232-2b: Second change value calculation unit 232-2c: Third change value calculation unit 232-2d: Fourth change value calculation unit; 232-3: State derivation section; 240: Output section

Claims

1. Safety equipment worn by workers (S1); A mounting section (S2) is provided at one end of the above-mentioned safety equipment (S1) to allow the biosignal measurement unit (100) to be attached; A biosignal measuring unit (100) is attached to the above-mentioned mounting unit (S2) to measure the biosignals of the worker; and The biological signals measured by the above-mentioned biological signal measurement unit (100) are transmitted, and the system includes a signal processing unit (200) that determines the worker's condition based on the transmitted biological signals and outputs the result. The above-mentioned biosignal measurement unit (100) is, Multiple biosignal measuring means (110) for measuring the biosignals of a worker; A biosignal receiving unit (120) that receives multiple biosignals using the biosignal measuring means (110) described above; A biosignal processing unit (130) supplies multiple biosignals received from the biosignal receiving unit (120) to the corresponding biosignal storage unit (140); A biosignal storage unit (140) receives the biosignal supplied from the biosignal processing unit (130) above, converts it into data, and stores it; and A transmitting unit (150) communicates with the above-mentioned signal processing unit (200) and provides the biological signals stored in the above-mentioned biological signal storage unit (140) to the signal processing unit (200); The above signal processing unit (200) is, An amplification unit (210) communicates with the above-mentioned transmitting unit (150) and amplifies the analog biological signal received through the transmitting unit (150); ADC(220) converts the analog biosignal amplified through the above amplification unit (210) into a digital biosignal; A state determination unit (230) that determines the worker's state based on the biosignal converted into the above digital biosignal; and Includes an output unit (240) that communicates with the terminals of the worker and the manager and provides the worker status information determined by the status determination unit (230) to the terminals of the worker and the manager. A real-time control system for worker hazardous conditions and status on an AIoT platform, characterized by the use of a composite biosignal sensor and multilayer signal decomposition technology.

2. The above biosignals include brain waves, heart rate, motion acceleration, and temperature. The above-mentioned biosignal measurement means (110) includes an electroencephalogram (EEG) sensor (112), a temperature sensor (113), a heart rate sensor (114), and an acceleration sensor (115). The above-mentioned biosignal storage unit (140) is an electroencephalogram (EEG) signal storage unit (141) that converts EEG signals into data and stores them; A heart rate signal storage unit (142) that digitizes and stores heart rate signals; a motion acceleration signal storage unit (143) that digitizes and stores motion acceleration; and Includes a temperature signal storage unit (144) that digitizes and stores the temperature as data. A real-time control system for worker hazardous conditions and status on an AIoT platform utilizing a composite biosignal sensor and multilayer signal decomposition technology as described in claim 1.

3. The above-mentioned state determination unit (230) A learning module (231) that stores and learns the biosignal information measured by the biosignal measurement unit (100) for a certain period of time; and The learning module (231) compares the biosignal information it has learned with the biosignal information generated in real time by the biosignal measurement unit (100) to determine whether or not there is an abnormality in the worker's condition; this includes a judgment module (232). A real-time control system for worker hazardous conditions and status on an AIoT platform utilizing a composite biosignal sensor and multilayer signal decomposition technology as described in claim 1.

4. The above decision module (232) is, A biosignal data storage unit (232-1) stores learned biosignal information and biosignal information generated in real time; A change value calculation unit (232-2) calculates a change value by substituting the biosignal information generated in real time into the learned biosignal information; and Includes a state derivation unit (232-3) that determines the state by substituting the above change values ​​into already learned state classification information. A real-time control system for worker hazardous conditions and status on an AIoT platform utilizing a composite biosignal sensor and multilayer signal decomposition technology as described in claim 3.

5. The above change value calculation unit (232-2) is, A first change value calculation unit (232-2a) calculates change values ​​based on electroencephalogram (EEG); a second change value calculation unit (232-2b) calculates change values ​​based on heart rate; A third change value calculation unit (232-2c) that calculates the change value based on temperature; and Includes a fourth change value calculation unit (232-2d) that calculates the change value based on the acceleration value. A real-time control system for worker hazardous conditions and status on an AIoT platform utilizing a composite biosignal sensor and multilayer signal decomposition technology as described in claim 4.

6. The electroencephalogram (EEG) sensor (112), temperature sensor (113), heart rate sensor (114), and acceleration sensor (115) are each mounted on a flexible circuit board module (116). The above flexible circuit board module (116) is, Flexible substrate section (116a); A power supply unit (116b) is mounted on the flexible circuit board portion (116a) mentioned above; Sensor mounting section (116c) for fixing each sensor mounted on the flexible substrate section (116a); and It includes a wireless communication unit (116d) mounted on the flexible substrate portion (116a) above, which transmits signals detected from each of the above sensors to the biosignal receiving unit (120); The power supply unit (116b), the sensor installation unit (116d), and the wireless communication unit (116d) installed in the above-mentioned unit are interconnected through electrical wiring. The above flexible substrate portion (116a) is made by applying a polyimide film. The above wireless communication unit (116d) applies Bluetooth, Zigbee, UWB, or Wi-Fi. The above electrical wiring uses conductive fibers. A real-time control system for worker hazardous conditions and status on an AIoT platform utilizing a composite biosignal sensor and multilayer signal decomposition technology as described in claim 2.

Images