Wearable early warning device, method and equipment for state recognition

Abstract

The embodiment of the invention belongs to the field of electroencephalogram collection safety, and particularly relates to a wearable early warning device, method and equipment for state recognition. The device comprises an electroencephalogram acquisition module, an electroencephalogram processing module, a core embedded processing module and a multi-mode early warning module. Real-time brain waves of an operator are collected through the brain wave collection module; the electroencephalogram processing module is used for carrying out feature extraction on the electroencephalogram and preliminarily detecting an operation state; and the core embedded processing module performs accurate state identification according to the characteristic result, and sends an early warning instruction to the multi-mode early warning module when a fatigue state is determined, so as to trigger multi-mode alarms such as sound, light, vibration and the like. The problem that in the prior art, real-time and accurate monitoring and early warning cannot be carried out on the fatigue state of personnel in complex industrial environments such as oil and gas stations is effectively solved, and the real-time performance, the accuracy and the system robustness of high-risk post safety monitoring are remarkably improved.

Description

Technical Field

[0001] The embodiments in this specification relate to the field of EEG data acquisition security, and in particular to a wearable early warning device, method and apparatus for state recognition. Background Technology

[0002] Oil and gas stations, as typical high-risk industrial sites, demand extremely high levels of concentration from their personnel. Especially during nighttime patrols, holiday shifts, or continuous operations, fatigue often leads to decreased attention or even temporary sleep, posing a significant safety risk. Currently, most stations rely on traditional video surveillance or shift-based patrols for personnel monitoring, which suffers from inherent limitations such as poor real-time performance and difficulty in identifying covert behaviors. In recent years, electroencephalogram (EEG) signals, which directly reflect the state of human brain activity, have been applied to scenarios such as driver fatigue monitoring and medical status recognition. However, applying EEG behavior recognition systems to oil and gas industrial stations remains a technological gap. Existing EEG monitoring equipment is often bulky, consumes a lot of power, and is expensive. More importantly, it generally lacks specific design for the high-noise, high-electromagnetic-interference environment and stringent explosion-proof requirements of oil and gas stations, making it difficult to meet the actual needs of frontline industrial sites.

[0003] Therefore, there is an urgent need for a wearable EEG monitoring and early warning solution that can adapt to the special operating environment of oil and gas stations, has high reliability, and can be deployed in a portable manner, in order to solve the problem that existing technologies cannot achieve accurate and real-time identification and early warning of personnel status in complex industrial environments. Summary of the Invention

[0004] To address the lack of reliable, real-time monitoring methods for worker fatigue in high-risk industrial environments such as oil and gas stations, and the problems of existing EEG monitoring devices being bulky, power-consuming, lacking anti-interference capabilities, and failing to meet explosion-proof requirements, this specification provides a wearable early warning device, method, and equipment for state recognition. This device can acquire workers' EEG signals through a head-mounted structure and perform embedded signal analysis to determine their attention and fatigue status in real time. Upon detecting an anomaly, it immediately triggers a dual-channel warning system of vibration and voice. Simultaneously, utilizing a lightweight wearable design and localized processing capabilities, it meets the urgent needs of demanding duty positions in oil and gas stations, coal mines, and hydropower stations for portable deployment, industrial safety, and high robustness.

[0005] To address the aforementioned technical problems, a first aspect of this specification provides a wearable early warning device for state recognition. The device includes: an EEG acquisition module, an EEG processing module, a core embedded processing module, and a multimodal early warning module; wherein: The EEG acquisition module is used to collect the real-time brain waves of the workers. The EEG processing module, connected to the EEG acquisition module, is used to extract features from the real-time EEG waves; and The work status of the data acquisition personnel is detected based on the fatigue algorithm and the real-time EEG. The core embedded processing module is used to perform state recognition based on the feature extraction result and the work status. If the state recognition result indicates that the data collection operator is in a state of fatigue, an early warning command is sent to the multimodal early warning module. A multimodal early warning module, connected to the core embedded processing module, is used to generate early warnings based on received early warning commands.

[0006] Furthermore, the device also includes a transmission positioning module. The transmission and positioning module locates the wearable early warning device via wireless transmission and positioning, and is used for data communication between the EEG processing module, the early warning module, and the mobile device.

[0007] Furthermore, the EEG acquisition module also includes a flexible dry contact patch electrode array. The flexible dry contact patch electrode array employs a differential acquisition structure to suppress common-mode noise; The differential acquisition structure includes a signal electrode and a reference electrode, used to synchronously acquire the differential signal between the signal electrode and the reference electrode.

[0008] Furthermore, the core embedded processing module also includes, An adaptive threshold recognition model is constructed based on the initial EEG and the fatigue recognition model. The adaptive threshold recognition model and the real-time EEG are used to determine the worker's work status.

[0009] Furthermore, the adaptive threshold recognition model constructed based on historical EEG waves further includes, When the operator first wears the warning device, the operator collects the electroencephalogram (EEG) data in a conscious state to obtain the initial EEG waves. Calculate the energy ratio of each frequency band in the initial EEG: = Pα / (Pα + Pβ + Pθ), Among them, P α The baseline alpha wave power at wakefulness; P β The baseline beta wave power at wakefulness; P θ The baseline theta wave power; This is the initial energy ratio threshold.

[0010] Furthermore, determining the worker's work status through the adaptive threshold recognition model and the real-time features further includes: Furthermore, determining the worker's work status through the adaptive threshold recognition model and the real-time features further includes: Using the monitoring interval within a continuous time period as a judgment period, the characteristic changes within several recent time windows are statistically analyzed. If at least two of the following conditions are met, it is determined to be continuous fatigue: α-wave power P α Decrease > 20%, theta wave power P θ Increase > 30%, signal-to-noise ratio (SNR) < 10 dB.

[0011] Furthermore, the EEG acquisition module uses the TGAM1 EEG acquisition chip to output raw EEG signals, attention index, relaxation index, and frequency band power parameters.

[0012] Furthermore, all modules in the device are embedded within the explosion-proof safety helmet housing, which has a protection rating of Ex ib IIB T4 Gb / IP66.

[0013] Furthermore, the device also includes a vital signs detection unit, used to acquire the vital signs parameters of the worker based on the real-time electroencephalogram (EEG), and input the vital signs parameters into a fatigue recognition model to obtain the vital signs status of the worker.

[0014] Furthermore, the vital signs detection unit further includes, After acquiring the real-time EEG, the heart rate peak of the worker is extracted within a preset time window by using 0.8 to 3 Hz bandpass filtering and peak detection to estimate the worker's heart rate (HR). The respiratory cycle of the worker is extracted by filtering and peak detection at 0.05 to 0.5 Hz to estimate the worker's respiratory rate (RR). The heart rate (HR) and respiratory rate (RR) are used as vital sign parameters to distinguish between short-term inattention and persistent drowsiness or napping.

[0015] A third aspect of the embodiments of this specification provides a wearable early warning method based on state recognition, the method comprising: Real-time brainwaves of the personnel involved in the data collection operation; Feature extraction is performed on the real-time EEG; and The work status of the data acquisition personnel is detected based on the fatigue algorithm and the real-time EEG. Based on the results of feature extraction and the work status, status identification is performed. If the status identification result indicates that the data collection operator is in a state of fatigue, an early warning command is sent to the multimodal early warning module. The multimodal early warning module generates an early warning based on the received early warning command.

[0016] Furthermore, based on the fatigue algorithm and the real-time EEG detection, the work status of the data acquisition personnel is further included, An adaptive threshold recognition model is constructed based on the initial EEG and fatigue recognition model. The adaptive threshold recognition model and the real-time EEG are used to determine the worker's work status.

[0017] A third aspect of the embodiments of this specification provides a computer device including a memory, a processor, and a computer program stored in the memory, wherein the computer program, when executed by the processor, performs instructions of the method described in any of the foregoing embodiments.

[0018] A third aspect of the embodiments of this specification provides a computer storage medium having a computer program stored thereon, which, when run by a processor of a computer device, executes instructions for the methods described in any of the foregoing embodiments.

[0019] A fourth aspect of the embodiments of this specification provides a computer program product, the computer program product including a computer program, which, when run by a processor of a computer device, executes instructions of the method described in any of the foregoing embodiments.

[0020] To make the objectives, technical solutions, and advantages of the embodiments in this specification clearer, the technical solutions in the embodiments of this specification will be described clearly and completely below. The embodiments of this specification provide a wearable early warning device for state recognition. Its core lies in achieving efficient and reliable monitoring and early warning of work status through the following collaborative modules: The device actively collects real-time brainwave signals from the worker through an EEG acquisition module, providing a high-quality data foundation for subsequent analysis; the collected signals are then subjected to deep feature extraction by a connected EEG processing module, and a fatigue algorithm is used to perform preliminary detection of the work status, thereby transforming the raw signals into feature indicators with clear physiological significance; the core embedded processing module performs accurate state recognition based on the results of the above feature extraction and state detection, and once it determines that the worker is in a fatigued work state, it immediately generates and sends an early warning command; finally, the multimodal early warning module triggers the corresponding early warning signal upon receiving the command, providing timely and effective warnings to the worker through multiple sensory channels. The embodiments in this specification achieve localized, real-time identification and closed-loop early warning of personnel fatigue status through the coordinated operation and embedded processing of the above modules. At the same time, by utilizing the integrated wearable design of the device and its stable operation capability in complex industrial environments, it effectively meets the robustness, real-time and portability requirements of personnel safety monitoring for high-risk positions such as oil and gas stations. Attached Figure Description

[0021] To more clearly illustrate the technical solutions in the embodiments or prior art of this specification, the drawings used in the description of the embodiments or prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this specification. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0022] Figure 1 The diagram shown is a schematic diagram of the implementation system of the wearable early warning device for status recognition according to the embodiments of this specification; Figure 2 The diagram shown is a schematic of a wearable early warning device for status recognition provided in an embodiment of this specification; Figure 3 The diagram shown is a schematic diagram of a flexible dry contact patch electrode array according to an embodiment of this specification. Figure 4 The diagram shown is a circuit diagram of differential acquisition in an embodiment of this specification; Figure 5 The diagram shown is a structural diagram of the wearable warning device according to an embodiment of this specification; Figure 6 The diagram shown is a flowchart of a wearable early warning method for state recognition according to an embodiment of this specification; Figure 7The diagram shown is a flowchart illustrating the detection of the worker's work status based on a fatigue algorithm and real-time EEG data collection in an embodiment of this specification. Figure 8 The diagram shown is a structural diagram of a computer device according to an embodiment of this specification.

[0023] Explanation of symbols in the attached drawings: 101. Wearable early warning device; 102. Terminal; 1011. EEG acquisition module; 1012. EEG processing module; 1013. Core embedded processing module; 1014. Multimodal early warning model. Detailed Implementation

[0024] The technical solutions in the embodiments of this specification will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this specification, and not all embodiments. Based on the embodiments in this specification, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this specification.

[0025] It should be noted that the terms "first," "second," etc., in the specification, claims, and accompanying drawings are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover a non-exclusive inclusion; for example, a process, method, apparatus, product, or device that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or devices.

[0026] This specification provides the operational steps of the methods described in the embodiments or flowcharts, but based on conventional or non-inventive labor, more or fewer operational steps may be included. The order of steps listed in the embodiments is merely one possible execution order among many and does not represent the only possible execution order. In actual system or device products, the methods shown in the embodiments or drawings can be executed sequentially or in parallel.

[0027] It should be noted that the acquisition, storage, use, and processing of data in the technical solutions of the embodiments of this specification all comply with the relevant provisions of national laws and regulations.

[0028] It should be noted that in the embodiments of this specification, certain software, components, models and other existing solutions in the industry may be mentioned. These should be regarded as exemplary and are only intended to illustrate the feasibility of implementing the technical solutions of the embodiments of this specification. However, it does not mean that the applicant has used or necessarily used such solutions.

[0029] like Figure 1 The diagram shows a system configuration of a wearable early warning device for status recognition provided in an embodiment of this specification. The core of the system is the wearable early warning device 101, which is configured to independently complete the acquisition, processing, status recognition, and local early warning of electroencephalogram (EEG) signals without a central server. As an optional component, the system may also include a terminal 102 (such as a mobile device or a backend monitoring unit). This terminal 102 can establish a wireless connection with the wearable early warning device 101 to receive alarm event records, historical personnel status data, and location information uploaded by the device, thereby enabling remote traceability and management. The core workflow of the wearable early warning device 101 does not rely on communication with the terminal 102, thus ensuring the real-time performance and reliability of early warnings in high-risk work environments.

[0030] It should be noted that, Figure 1 The example shown is merely one application environment provided by this disclosure. In practical applications, other application environments may also be included, which are not limited in the embodiments of this specification.

[0031] To address the problems existing in the prior art, this specification provides a wearable early warning device, equipment, and storage medium with status recognition capabilities. Figure 2 The diagram illustrates the workflow of a wearable warning device for status recognition provided in an embodiment of this specification. The diagram depicts the process by which the wearable warning device identifies a person's status. The order of steps listed in the embodiment is merely one possible execution order among many and does not represent the only possible order. In actual system or device products, the methods shown in the embodiments or accompanying drawings can be executed sequentially or in parallel. Specifically, as shown... Figure 2 As shown, the method may include: an EEG acquisition module 1011, an EEG processing module 1012, a core embedded processing module 1013, and a multimodal early warning module 1014; wherein: The EEG acquisition module 1011 is used to acquire the real-time EEG waves of the operator. The EEG processing module 1012, connected to the EEG acquisition module 1011, is used to extract features from the real-time EEG waves; and The work status of the data acquisition personnel is detected based on the fatigue algorithm and the real-time EEG. The core embedded processing module 1013 is used to perform state recognition based on the result of feature extraction and the work status. If the result of the state recognition is that the data collection operator is in a state of fatigue, then an early warning command is sent to the multimodal early warning module. The multimodal early warning module 1014 is connected to the core embedded processing module 1013 and is used to generate an early warning based on the received early warning command.

[0032] This specification provides a wearable early warning device for status recognition. Its core lies in achieving efficient and reliable monitoring and early warning of work status through the following collaborative modules: The device actively collects real-time EEG signals from workers via an EEG acquisition module, providing a high-quality data foundation for subsequent analysis; the acquired signals are then subjected to deep feature extraction by a connected EEG processing module, and a fatigue algorithm is used for preliminary detection of work status, thereby transforming the raw signals into feature indicators with clear physiological significance; the core embedded processing module performs precise status recognition based on the results of the above feature extraction and status detection. Once it determines that the worker is in a fatigued work state, it immediately generates and sends an early warning command; finally, the multimodal early warning module triggers the corresponding early warning signal upon receiving the command, providing timely and effective warnings to the worker through multiple sensory channels. This specification, through the collaborative operation and embedded processing of the above modules, achieves localized, real-time recognition and closed-loop early warning of personnel fatigue status. Simultaneously, by utilizing the device's integrated wearable design and stable operation in complex industrial environments, it effectively meets the robustness, real-time performance, and portability requirements for personnel safety monitoring in high-risk positions such as oil and gas stations.

[0033] The EEG acquisition module, as a front-end sensing unit, is installed on the inside of the helmet using a flexible dry contact patch electrode array. Its core can be a low-cost EEG acquisition chip such as the TGAM series, which is responsible for collecting the raw real-time brain waves of the operator and can simultaneously output preliminary calculation indicators such as attention and relaxation.

[0034] The EEG processing module is directly connected to the acquisition module and is responsible for deep feature extraction of the input real-time EEG waves. This process first involves signal preprocessing and noise reduction. A second-order IIR bandpass filter built into the microcontroller performs bandpass filtering on the signal from 0.5 to 45 Hz, combined with the hardware notch filter on the acquisition chip itself, to eliminate EMG and 50 / 60 Hz power frequency noise interference. Subsequently, discrete wavelet transform is used to analyze the clean signal, and a 5-level decomposition using the db4 basis function is performed to separate key EEG bands such as α, β, and θ. The energy values ​​of each band are calculated in real-time using a fast integer algorithm in the embedded environment. Finally, by calculating the energy spectral entropy and combining the power of each band, a time-frequency feature vector for fatigue assessment is formed.

[0035] The core embedded processing module performs final state recognition based on the extracted features. This module uses an STM32 microcontroller as its hardware core to build and run an adaptive threshold recognition model. This model establishes an individualized baseline by collecting EEG data from the worker during their initial wearing of the device while they are awake, and dynamically adjusts it based on short-term averages during operation, thereby achieving real-time and accurate judgment of fatigue status. If the recognition result indicates that the worker is in a fatigued working state, the module immediately generates and sends an early warning command.

[0036] Upon receiving an early warning command, the multimodal early warning module triggers sound, light, and vibration alerts independently or in combination through its built-in voice playback chip and vibration motor, thereby effectively warning personnel in high-noise industrial environments and forming a complete local closed-loop alarm.

[0037] In addition, the system integrates a wireless communication module and a replaceable lithium battery. All modules are compactly integrated into an explosion-proof safety helmet shell that meets the Ex ib IIB T4 Gb / IP66 protection level, ensuring long-term, stable, and independent operation in complex industrial environments such as oil and gas stations, and realizing the core functions of front-end identification and local alarm.

[0038] In one embodiment of this specification, the illustrated device further includes a transmission positioning module 1015. The transmission and positioning module 1015 locates the wearable early warning device through wireless transmission and positioning, and is used for data communication between the EEG processing module, the early warning module, and the mobile device.

[0039] Specifically, the transmission and positioning module establishes a connection with the core embedded processing module, achieving two core functions through Bluetooth Low Energy or other wireless communication protocols: first, it uses the built-in positioning unit to perform real-time positioning of the wearable warning device; second, it is responsible for establishing a communication link between the core embedded processing module and external mobile devices or background monitoring systems. Specifically, this module packages the status recognition results, alarm event records, and corresponding positioning information generated by the core embedded processing module and wirelessly sends them to the mobile device, thereby achieving remote monitoring of personnel status and traceability management of work trajectories. Because the data transmission process occurs after status recognition and local alarm, the alarm command of the multimodal warning module is directly and in real-time triggered by the core embedded processing module, ensuring the immediacy of safety warnings. Its execution is completely independent of the communication status between the transmission and positioning module and external devices.

[0040] In one embodiment of this specification, the EEG acquisition module 1011 further includes a flexible dry contact patch electrode array. The flexible dry contact patch electrode array employs a differential acquisition structure to suppress common-mode noise; The differential acquisition structure includes a signal electrode and a reference electrode, used to synchronously acquire the differential signal between the signal electrode and the reference electrode.

[0041] In one embodiment of this specification, such as Figure 3 As shown, the EEG acquisition module specifically includes two parallel electrode contact strips in the left forehead region to form a first differential acquisition channel; and two parallel electrode contact strips in the right forehead region to form a second differential acquisition channel. This electrode array employs a differential acquisition structure specifically designed for industrial environments to effectively suppress common-mode noise. Specifically, this differential acquisition structure is based on an EEG-REF dual-electrode architecture, using shielded twisted-pair cables for signal transmission. For common-mode interference generated by strong interference sources such as solenoid valves and high-voltage motors in oil and gas stations, this structure can provide a common-mode rejection ratio of no less than 90 dB, significantly improving signal quality. In one specific implementation, the system uses the TGAM1 EEG acquisition module as the core acquisition unit. This module provides a standard differential acquisition interface with three contacts: EEG, REF, and GND. It continuously outputs 12-bit resolution raw EEG voltage signals at a sampling frequency of 512 Hz, while simultaneously outputting signal quality values ​​(0-200), attention index (0-100), relaxation index (0-100), and power parameters for eight frequency bands: Δ, Θ, α, β, and γ. This data is transmitted to the main control MCU in real time via the UART serial port in standard data frame format, providing a complete data foundation for subsequent signal processing and status recognition.

[0042] In another embodiment of this specification, in order to further improve the accuracy and individual adaptability of state recognition, an adaptive threshold recognition model is constructed based on the initial EEG and the fatigue recognition model; The adaptive threshold recognition model and the real-time EEG are used to determine the worker's work status.

[0043] The adaptive threshold recognition model built based on historical EEG waves further includes, When the operator first wears the warning device, the operator collects the electroencephalogram (EEG) data in a conscious state to obtain the initial EEG waves. Calculate the energy ratio of each frequency band in the initial EEG: = P α / (P α + P β + P θ ), Among them, P α The baseline alpha wave power at wakefulness; P β The baseline beta wave power at wakefulness; Pθ The baseline theta wave power; This is the initial energy ratio threshold.

[0044] During the initial wearing phase, the system recorded the operator's EEG signals during approximately 30 seconds of conscious state. Power estimation was performed on the three frequency bands: α (8-13 Hz), β (13-30 Hz), and θ (4-7 Hz), yielding P. α P β P θ And calculate the baseline energy ratio according to the above formula. As an individualized level of awareness for this worker, In one embodiment of this specification, the core embedded processing module is configured to build and run an adaptive threshold recognition model to achieve accurate judgment of the work status. The model construction begins in the baseline establishment phase. When the worker first wears the warning device, the system collects their EEG data during 30 seconds of wakefulness as initial EEG data and calculates the energy ratio of each frequency band. = Pα / (Pα + Pβ + Pθ), where Pα, Pβ, and Pθ represent the power of the α-wave, β-wave, and θ-wave, respectively, to establish an individualized fatigue assessment benchmark. During the real-time monitoring phase, the model is dynamically updated using a sliding window mechanism. The system continuously calculates the real-time energy ratio E(t) in 5-second windows and calculates the difference between it and the baseline value ΔE(t) = E(t) - When a continuous decrease in ΔE is detected and a synchronous increase in the theta wave power Pθ, it is determined to be a decreasing trend in attention. To ensure the model's adaptability, the system introduces a learning rate μ (ranging from 0.1 to 0.2), according to... ←(1-μ) The formula +μE(t) dynamically adjusts the baseline value, achieving continuous optimization of the individualized threshold. Regarding the determination of persistent fatigue state, the system employs a multi-indicator joint judgment mechanism. A persistent fatigue state is determined when any two of the following conditions are met within a consecutive 30-second time window (at least three consecutive windows): alpha wave power decreases by more than 20%; theta wave power increases by more than 30%; and the signal-to-noise ratio is below 10 dB. Once the determination is established, the system immediately activates a level-two alarm mode, broadcasting a "Please take a rest" prompt via the voice chip, while simultaneously triggering the vibration motor for a double warning. The system also features a self-recovery detection function; when the alpha wave power is detected to recover by more than 10%, the alarm state is automatically deactivated, completing a full closed loop from monitoring and early warning to recovery.

[0045] At the circuit implementation level, the acquisition module adopts an anti-common-mode noise acquisition scheme specifically designed for industrial high-interference environments.

[0046] In another embodiment of this specification, the acquisition circuit diagram is as follows: Figure 4 As shown, the front end uses a differential acquisition circuit (EEG–REF two electrodes) to suppress common-mode noise and ground potential drift, ensuring signal stability. This design has better anti-interference performance than consumer-grade devices with single-electrode acquisition. The circuit diagram is explained below. U1 is the sensor, and pins 4 and 5 are differential signals. Although the output signals of the differential signals are paired and differential, the amplitude is only in the millivolt range. Modern ADCs, such as 12 / 14 / 16-bit ADCs, typically have a full-scale range of 0–3.3 V or 0–5 V, which has limited resolution. Therefore, directly acquiring the weak skin output signal will cause the signal to be submerged in quantization noise or power supply ripple. At the same time, the output signal of U1 is easily affected by common-mode signals such as power frequency interference and power supply fluctuations in the complex electromagnetic environment of oil and gas stations. Directly connecting it to the ADC will lead to significant quantization errors. Therefore, a low-noise, high-gain, differential input amplifier circuit is required. An AD8226 instrumentation amplifier was designed. In order to achieve accurate amplification of weak signals, the AD8226 instrumentation amplifier circuit was designed with the following features.

[0047] High input impedance (>10 GΩ); low noise (8 nV / √Hz @ 1kHz); common-mode rejection ratio (CMRR) > 100dB; flexible gain setting (determined by external resistor); rail-to-rail output, suitable for single-supply systems.

[0048] The formula for designing the gain of the AD8226 amplifier circuit is:

[0049] in, The gain resistor is set by the user. Based on the TMR2901 output range (±5 mV~±50 mV), considering the ADC input range of 0–3.3V, and assuming the target full-scale signal is 3V, the amplification gain needs to be set to approximately 60 times. When an 866-ohm resistor is selected, we get:

[0050] This gain value can effectively amplify a ±50 mV signal to ±2.9 V, meeting the ADC sampling requirements while maintaining the system's linearity and low noise characteristics.

[0051] One of the key performance indicators of an instrumentation amplifier is the Common-Mode Rejection Ratio (CMRR). It measures the amplifier's ability to distinguish between "useful signals (differential)" and "interference signals (common-mode)." The AD8226 has extremely strong suppression capabilities for common-mode signals (such as power supply coupling interference and long PCB trace interference), and theoretically its CMRR is:

[0052] in: Differential gain; Common-mode gain; Common-mode rejection ratio (CMRR) is measured in decibels (dB). When the same interference (e.g., 50Hz power line noise) is introduced simultaneously on two input signal lines, this is called "common-mode interference." Ideally, an instrumentation amplifier only amplifies the difference between the two inputs, ignoring these "identical" signals. However, in reality, circuits cannot be perfectly ideal, so a small amount of common-mode signal will still be amplified. A higher CMRR indicates a more "intelligent" amplifier—it can accurately distinguish between the desired signal and the interference signal. The AD8226 has a CMRR > 100dB when the gain is > 10, effectively eliminating common-mode components introduced by interfering magnetic fields, power supply ripple, and ambient temperature changes. This is particularly crucial for the "tiny changes in residual magnetism" detected by the TMR2901, which is the key reason for choosing the AD8226 over a conventional op-amp.

[0053] It is worth emphasizing that the above-mentioned complete signal processing, feature extraction, status recognition and alarm decision-making process are all completed independently in the local embedded system without relying on any external computing resources or central server, which ensures the reliability and real-time performance of the system in special industrial environments such as oil and gas stations.

[0054] Determining the worker's work status through the adaptive threshold recognition model and the real-time features further includes... Using the monitoring interval within a continuous time period as a judgment period, the characteristic changes within several recent time windows are statistically analyzed. If at least two of the following conditions are met, it is determined to be continuous fatigue: α-wave power P α Decrease > 20%, theta wave power P θ Increase > 30%, signal-to-noise ratio (SNR) < 10 dB.

[0055] The EEG acquisition module uses the TGAM1 EEG acquisition chip to output raw EEG signals, attention index, relaxation index, and frequency band power parameters.

[0056] In one embodiment of this specification, the EEG acquisition module is configured to acquire multiple types of electroencephalographic signals. Specifically, the system acquires microvolt-level bioelectrical signals from the cerebral cortex, the data of which includes: raw EEG signals with a sampling rate of 512 Hz and a resolution of 12 bits; power characteristic signals of each band (Δ, Θ, α, β, γ) obtained after processing; attention and relaxation indices generated by a dedicated algorithm; and signal-to-noise ratio or signal quality parameters used to automatically verify electrode wearing status. These signals together constitute the complete data foundation for subsequent fatigue state analysis.

[0057] In one embodiment of this specification, such as Figure 5 As shown, all modules in the device are embedded in the explosion-proof safety helmet shell, and the shell has a protection rating of Ex ib IIB T4 Gb / IP66.

[0058] Preferably, the wearable EEG acquisition module includes four flexible dry contact electrode strips disposed on the inner side of the forehead of a safety helmet. Specifically: two parallel electrode contact strips are set in the left forehead region to form the first differential acquisition channel; two parallel electrode contact strips are set in the right forehead region to form the second differential acquisition channel. The four electrode contact strips are connected to the electrode input terminals of the TGAM EEG acquisition chip via wires, enabling dual-channel differential EEG acquisition of the left and right frontal regions.

[0059] Each electrode contact strip includes: The flexible insulating substrate layer is made of at least one of polyimide (PI), polyethylene terephthalate (PET), thermoplastic polyurethane (TPU), or silicone rubber. A conductive layer is disposed on the skin-facing side of the flexible insulating substrate. The conductive layer material is at least one of silver / silver chloride (Ag / AgCl), gold-plated copper, conductive polymer, conductive fabric, or silver-containing conductive silicone. The conductive layer surface can form several micro-bumps or comb-like tooth structures to increase the contact area and reduce the scalp-electrode contact resistance, achieving stable dry contact EEG acquisition without the need for conductive paste. By adopting a flexible conductive patch structure, the electrode contact strip can conform to the curvature of the forehead and slight head movements, significantly reducing the pressure and displacement artifacts of traditional hard metal electrodes during prolonged wear, while also exhibiting good resistance to sweat corrosion and mechanical reliability in oil and gas station environments.

[0060] In another embodiment of this specification, the device further includes a vital signs detection unit, used to acquire the vital signs parameters of the worker based on the real-time electroencephalogram (EEG), and input the vital signs parameters into a fatigue recognition model to obtain the vital signs status of the worker.

[0061] The vital signs detection unit further includes, After acquiring the real-time EEG, the heart rate peak of the worker is extracted within a preset time window by using 0.8 to 3 Hz bandpass filtering and peak detection to estimate the worker's heart rate (HR). The respiratory cycle of the worker is extracted by filtering and peak detection at 0.05 to 0.5 Hz to estimate the worker's respiratory rate (RR). The heart rate (HR) and respiratory rate (RR) are used as vital sign parameters to distinguish between short-term inattention and persistent drowsiness or napping.

[0062] For example, suppose the EEG signal output by TGAM is The sampling frequency is (For example, 512 Hz), the vital signs detection unit performs the following steps: For The preprocessed signal is obtained by performing a 50 Hz power frequency notch filter and a 0.5 Hz high-pass filter. To remove DC drift and power frequency interference; a sliding update is performed with a window size of 10 seconds and a window interval of 1 second, denoted as the [number]th [analysis window]. A time window.

[0063] Then, heart rate (HR) extraction (using ECG artifacts from prefrontal EEG) is performed within each window: right A 0.8–3 Hz bandpass filter was applied to obtain a signal with enhanced ECG components: , calculate The envelope or first derivative of the heartbeat waveform is used, and an adaptive threshold peak detection algorithm is employed to extract the position of the heartbeat peak. : The peak threshold can be taken as the root mean square value of the current window. 1.5–2.0 times; The interval between two adjacent peaks is limited to (Corresponding heart rate 40–200 bpm), remove abnormal peaks.

[0064] Count of heartbeats detected within the calculation window Window duration is The heart rate estimate for this window is: , when If the heart rate estimation window is less than a preset lower limit (e.g., 3) or the standard deviation of the peak interval is greater than 0.5 s, the heart rate estimation for that window is deemed invalid. Data marked as missing will not be used for subsequent fatigue assessment.

[0065] Respiratory rate (RR) extraction utilizes low-frequency amplitude modulation: by As input, perform low-frequency filtering to obtain the breathing modulation components: Option 1: To After rectification or envelope detection, a low-pass filter (cutoff frequency 0.5 Hz) is applied to obtain... ; Option 2: For The result was obtained by directly using a 0.05–0.5 Hz bandpass filter. .

[0066] For the obtained respiratory signals or Peak detection was performed to extract the respiratory peak. The interval between adjacent peaks is required to be within (Corresponds to 6–30 times / minute).

[0067] Let the first The number of respiratory peaks within each window is The respiratory rate is then estimated as follows: , Finally, vital sign parameters and EEG characteristics were integrated into the fatigue model: For each time window Simultaneously calculate the power of the brainwave frequency band: , And the energy ratio is obtained: , The vital signs detection unit outputs a feature vector: , The fatigue state identification model will As input, the result is determined by an adaptive threshold or classifier: Simply daydreaming: Slightly lower Slightly increased, but HR and RR differed from individual baseline by less than 10%; Drowsiness / Nap: Significant decline The rate continues to rise, while the heart rate (HR) decreases significantly and the response rate (RR) slows down or becomes irregular.

[0068] Compared with the scheme that relies solely on α / θ power changes to determine fatigue, this embodiment introduces HR and RR as auxiliary vital sign parameters, which can effectively distinguish between "short-term distraction" and "continuous drowsiness / nap" states, and significantly reduce the false alarm rate.

[0069] Based on the same concept as this specification, embodiments of this specification also provide a wearable early warning method based on state recognition, such as... Figure 6 As shown, the method includes: Step 601: Collect real-time EEG data of the workers; Step 602: Extract features from the real-time EEG; and The work status of the data acquisition personnel is detected based on the fatigue algorithm and the real-time EEG. Step 603: Based on the feature extraction results and the work status, perform status identification. If the status identification result indicates that the data collection operator is in a state of fatigue, then send an early warning command to the multimodal early warning module. Step 604: The multimodal early warning module generates an early warning based on the received early warning command.

[0070] Among them, such as Figure 7 As shown, based on the fatigue algorithm and the real-time EEG detection, the work status of the data acquisition personnel further includes, Step 701: Construct an adaptive threshold recognition model based on the initial EEG and fatigue recognition model; Step 702: Determine the worker's work status using the adaptive threshold recognition model and the real-time EEG.

[0071] In one embodiment of this specification, a computer device is also provided for implementing the methods described in any of the above embodiments, such as... Figure 8 The diagram illustrates the structure of a computer device according to an embodiment of this specification. The computer device 802 may include one or more processors 804, such as one or more central processing units (CPUs), each of which can implement one or more hardware threads. The computer device 802 may also include any memory 806 for storing information of any kind, such as code, settings, data, etc. Without limitation, for example, the memory 806 may include any type of RAM, any type of ROM, flash memory, hard disk, optical disk, etc. More generally, any memory can use any technology to store information. Furthermore, any memory can provide volatile or non-volatile retention of information. Furthermore, any memory may represent a fixed or removable component of the computer device 802. In one case, when the processor 804 executes associated instructions stored in any memory or combination of memories, the computer device 802 can perform any operation of the associated instructions. The computer device 802 also includes one or more drive mechanisms 808 for interacting with any memory, such as hard disk drive mechanisms, optical disk drive mechanisms, etc.

[0072] Computer device 802 may also include an input / output module 810 (I / O) for receiving various inputs (via input device 812) and providing various outputs (via output device 814). A specific output mechanism may include a presentation device 816 and an associated graphical user interface (GUI) 818. In other embodiments, the input / output module 810 (I / O), input device 812, and output device 814 may be omitted, and the device may function solely as a computer device within a network. Computer device 802 may also include one or more network interfaces 820 for exchanging data with other devices via one or more communication links 822. One or more communication buses 824 couple the components described above together.

[0073] Communication link 822 can be implemented in any way, such as via a local area network, a wide area network (e.g., the Internet), a point-to-point connection, or any combination thereof. Communication link 822 may include any combination of hardwired links, wireless links, routers, gateway functions, name servers, etc., governed by any protocol or combination of protocols.

[0074] Corresponding to Figures 6 to 7 In addition to the methods described above, embodiments of this specification also provide a computer-readable storage medium storing a computer program that, when executed by a processor, performs the steps of the methods described above.

[0075] This specification also provides computer-readable instructions, wherein when a processor executes the instructions, the program therein causes the processor to perform the following... Figures 6 to 7 The method shown.

[0076] It should be understood that in the various embodiments of this specification, the sequence number of each process does not imply the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of this specification.

[0077] It should also be understood that, in the embodiments of this specification, the term "and / or" is merely a description of the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent: A existing alone, A and B existing simultaneously, and B existing alone. Additionally, the character " / " in this specification generally indicates that the preceding and following related objects have an "or" relationship.

[0078] Those skilled in the art will recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed in this specification can be implemented in electronic hardware, computer software, or a combination of both. To clearly illustrate the interchangeability of hardware and software, the components and steps of each example have been generally described in terms of functionality in the foregoing description. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementations should not be considered beyond the scope of the embodiments in this specification.

[0079] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the specific working processes of the systems, devices, and units described above can be referred to the corresponding processes in the foregoing method embodiments, and will not be repeated here.

[0080] In the several embodiments provided in this specification, it should be understood that the disclosed systems, apparatuses, and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the couplings or direct couplings or communication connections shown or discussed may be indirect couplings or communication connections through some interfaces, devices, or units, or they may be electrical, mechanical, or other forms of connection.

[0081] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of the embodiments described in this specification, depending on actual needs.

[0082] Furthermore, the functional units in the various embodiments of this specification can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit.

[0083] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solutions of the embodiments of this specification, in essence, or the parts that contribute to the prior art, or all or part of the technical solutions, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, a wearable warning device, or a network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of this specification. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0084] This specification uses specific embodiments to illustrate the principles and implementation methods of this specification. The descriptions of the above embodiments are only for the purpose of helping to understand the methods and core ideas of the embodiments in this specification. At the same time, for those skilled in the art, there will be changes in the specific implementation methods and application scope based on the ideas of the embodiments in this specification. Therefore, the content of this specification should not be construed as a limitation on the embodiments in this specification.

Claims

1. A wearable early warning device with status recognition, characterized in that, The device shown includes: an EEG acquisition module, an EEG processing module, a core embedded processing module, and a multimodal early warning module; wherein: The EEG acquisition module is used to collect the real-time brain waves of the workers. The EEG processing module, connected to the EEG acquisition module, is used to extract features from the real-time EEG waves; and The work status of the data acquisition personnel is detected based on the fatigue algorithm and the real-time EEG. The core embedded processing module is used to perform state recognition based on the feature extraction result and the work status. If the state recognition result indicates that the data collection operator is in a state of fatigue, an early warning command is sent to the multimodal early warning module. A multimodal early warning module, connected to the core embedded processing module, is used to generate early warnings based on received early warning commands.

2. The wearable early warning device for status recognition according to claim 1, characterized in that, The device shown also includes a transmission and positioning module. The transmission and positioning module locates the wearable early warning device via wireless transmission and positioning, and is used for data communication between the EEG processing module, the early warning module, and the mobile device.

3. The wearable early warning device for status recognition according to claim 1, characterized in that, The EEG acquisition module also includes a flexible dry contact patch electrode array. The flexible dry contact patch electrode array employs a differential acquisition structure to suppress common-mode noise; The differential acquisition structure includes a signal electrode and a reference electrode, used to synchronously acquire the differential signal between the signal electrode and the reference electrode.

4. The wearable early warning device for status recognition according to claim 1, characterized in that, The core embedded processing module also includes, An adaptive threshold recognition model is constructed based on the initial EEG and fatigue recognition model. The adaptive threshold recognition model and the real-time EEG are used to determine the worker's work status.

5. The wearable early warning device for status recognition according to claim 4, characterized in that, The adaptive threshold recognition model, constructed based on historical EEG waves, further includes... When the operator first wears the warning device, the operator collects the electroencephalogram (EEG) data in a conscious state to obtain the initial EEG waves. Calculate the energy ratio of each frequency band in the initial EEG: = Pα / (Pα + Pβ + Pθ), Among them, P α The baseline alpha wave power at wakefulness; P β The baseline beta wave power at wakefulness; P θ The baseline theta wave power; This is the initial energy ratio threshold.

6. The wearable early warning device for status recognition according to claim 4, characterized in that, Determining the worker's work status through the adaptive threshold recognition model and the real-time features further includes... Using the monitoring interval within a continuous time period as a judgment period, the characteristic changes within several recent time windows are statistically analyzed. If at least two of the following conditions are met, it is determined to be continuous fatigue: α-wave power P α Decrease > 20%, theta wave power P θ Increase > 30%, signal-to-noise ratio (SNR) < 10 dB.

7. The wearable early warning device for status recognition according to claim 1, characterized in that, The EEG acquisition module uses the TGAM1 EEG acquisition chip to output raw EEG signals, attention index, relaxation index, and frequency band power parameters.

8. The wearable early warning device for status recognition according to claim 1, characterized in that, All modules in the device are embedded in the explosion-proof safety helmet shell, which has a protection rating of Ex ib IIB T4 Gb / IP66.

9. The wearable early warning device for status recognition according to claim 1, characterized in that, The device also includes a vital signs detection unit, used to acquire the vital signs parameters of the worker based on the real-time electroencephalogram (EEG), and input the vital signs parameters into a fatigue recognition model to obtain the vital signs status of the worker.

10. The wearable early warning device for status recognition according to claim 9, characterized in that, The vital signs detection unit further includes... After acquiring the real-time EEG, the heart rate peak of the worker is extracted within a preset time window by using 0.8 to 3 Hz bandpass filtering and peak detection to estimate the worker's heart rate (HR). The respiratory cycle of the worker is extracted by filtering and peak detection at 0.05 to 0.5 Hz to estimate the worker's respiratory rate (RR). The heart rate (HR) and respiratory rate (RR) are used as vital sign parameters to distinguish between short-term inattention and persistent drowsiness or napping.

11. A wearable early warning method based on state recognition, characterized in that, The methods shown include: Real-time brainwaves of the personnel involved in the data collection operation; Feature extraction is performed on the real-time EEG; and The work status of the data acquisition personnel is detected based on the fatigue algorithm and the real-time EEG. Based on the results of feature extraction and the work status, status identification is performed. If the status identification result indicates that the data collection operator is in a state of fatigue, an early warning command is sent to the multimodal early warning module. The multimodal early warning module generates an early warning based on the received early warning command.

12. The wearable early warning method for status recognition according to claim 11, characterized in that, The fatigue algorithm and the real-time EEG detection of the operator's work status further include, An adaptive threshold recognition model is constructed based on the initial EEG and fatigue recognition model. The adaptive threshold recognition model and the real-time EEG are used to determine the worker's work status.

13. A computer device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the computer program, it implements the method according to any one of claims 11 to 12.

Images