Intelligent air quality monitoring system based on sensing data feedback

Abstract

The air quality intelligent monitoring system based on sensing data feedback comprises a sensing array module, a preprocessing module, a risk assessment module, an energy consumption decision module, a mode switching module, a data fusion module and a linkage module. The sensing array module generates an environment parameter signal; the preprocessing module generates a standardized characteristic signal; the risk assessment module receives the standardized characteristic signal and forms an early warning instruction signal; the energy consumption decision module analyzes the change gradient of the standardized feature vector and generates an activation instruction according to the vector stability; the mode switching module generates a state configuration signal and feeds back the state configuration signal to the sensing array module; fusing the data to generate a diffusion prediction signal; the linkage module generates an emergency braking instruction and transmits the emergency braking instruction back to the data fusion module. The intelligent air quality monitoring system based on sensing data feedback can solve the problem that intelligent linkage of multi-source pollution monitoring, dynamic energy saving and real-time early warning of industrial places is poor.

Description

Technical Field

[0001] The present invention relates to the technical field of environmental monitoring and industrial safety intersection, and in particular to an intelligent air quality monitoring system based on sensor data feedback. Background Art

[0002] Industrial air quality monitoring technology has long faced the challenge of multi-target coordinated control. Traditional environmental monitoring equipment focuses on detecting atmospheric pollutants (such as PM2.5 and SO2), while occupational hazard monitoring systems typically independently detect specific hazards (such as benzene and dust). This separate architecture prevents data from being interoperable, making it difficult to assess the comprehensive impact of complex pollution on human health.

[0003] In the prior art, the patent document with application number CN202110234567.8 proposes a multi-parameter environmental monitoring terminal, but its fixed sensor configuration cannot adapt to the characteristics of the variable hazard sources in industrial sites, and does not consider the problem of optimizing detection energy consumption. Although the patent document with application number CN202010876543.2 introduces a wireless sensor network, it only realizes the data transmission function and does not solve the time synchronization and noise interference problems of multi-source data. In terms of energy consumption control, the patent document CN201980012345.6 adopts a timed sleep strategy to reduce power consumption, but the fixed-cycle sampling mode is prone to miss sudden pollution events, and the wake-up delay leads to insufficient response speed. In the field of occupational disease hazard assessment, existing systems mostly use offline laboratory analysis or simple threshold alarms to collect individual exposure data through wearable detectors, but lack the ability to analyze the spatial distribution of group exposure and cannot be linked with environmental control equipment. Summary of the Invention

[0004] In view of the shortcomings of the above-mentioned prior art, the purpose of the present invention is to provide an intelligent air quality monitoring system based on sensor data feedback, which is used to solve the problem of poor intelligent linkage of multi-source pollution monitoring, dynamic energy saving and real-time warning in industrial sites. The present invention achieves breakthroughs through multi-dimensional technological innovation: first, a dynamically reconfigurable magnetic sensor array is constructed, which is compatible with the synchronous collection of environmental parameters (temperature, humidity, particulate matter) and occupational hazard factors (VOCs, radiation), and the heterogeneity of multi-source data is solved through hardware interface standardization; secondly, a dual-threshold energy-saving control algorithm is developed to automatically switch to micro-power consumption mode during low-risk periods, maintaining only the baseline monitoring function, and quickly activate high-precision sensors when parameter mutations are detected. Combined with FPGA dynamic reconstruction technology, standby power consumption is reduced by more than 80%; thirdly, the personnel positioning data and time-weighted exposure model are innovatively integrated to calculate the cumulative exposure dose of different positions in real time, and the pollutant diffusion path is predicted through deep learning, driving the ventilation system and isolation device to respond in advance.

[0005] The present invention provides an intelligent air quality monitoring system based on sensor data feedback, comprising:

[0006] A sensor array module, the sensor array module generates an environmental parameter signal;

[0007] The preprocessing module receives the environmental parameter signal, eliminates the time delay difference and removes the noise to generate a standardized characteristic signal;

[0008] Risk assessment module: The risk assessment module receives the standardized characteristic signal, calculates the time-weighted concentration and the short-term exposure limit, and generates an early warning command signal;

[0009] Energy consumption decision module,The energy consumption decision module analyzes the change gradient of the normalized feature vector and generates activation instructions based on the vector stability;

[0010] A mode switching module dynamically reconfigures the sensor circuit according to the activation instruction, selectively shuts down non-essential units, and generates a state configuration signal to feed back to the sensor array module;

[0011] Data fusion module, the data fusion module receives the standardized characteristic signal and the warning command signal, establishes the correlation matrix between the environment and the hazard parameters, and generates the diffusion prediction signal;

[0012] The linkage module controls the ventilation equipment and alarm device according to the early warning command signal and the diffusion prediction signal, generates an emergency braking command and transmits the emergency braking command back to the data fusion module.

[0013] In one embodiment of the present invention, the sensor array module includes a fixed detection unit consisting of a gas composition sensor, a particulate matter sensor, and a temperature and humidity sensor, and an extended detection unit consisting of a volatile organic compound sensor, a radiation sensor, and a bioaerosol sensor dynamically assembled through a magnetic interface. The fixed detection unit periodically collects baseline environmental parameters, and the extended detection unit is selectively activated according to a status configuration signal. When the activation instruction is triggered, the magnetic interface supplies power to the extended detection unit and establishes an optical communication link to generate a high-precision hazard parameter signal to supplement the environmental parameter signal.

[0014] In one embodiment of the present invention, the preprocessing module includes a multi-channel buffer circuit and a noise separation unit. The multi-channel buffer circuit uses a hardware timestamp to mark the data collection time of each sensor. The noise separation unit uses an adaptive filter based on device fingerprint recognition. Device fingerprint recognition establishes a noise feature library by analyzing the inherent thermal noise spectrum characteristics of the sensor. When the aliasing frequency in the environmental parameter signal is detected to match the feature library, the digital notch filter is activated to eliminate the interference of the sensor body and generate a standardized feature signal without device noise.

[0015] In one embodiment of the present invention, the risk assessment module set includes a personnel positioning subsystem, which includes a UWB (ultra-wideband) positioning base station deployed in an industrial site and a wearable device carrying a Bluetooth beacon. By fusing UWB arrival time difference positioning data with Bluetooth signal strength data, the real-time three-dimensional coordinates of the personnel are calculated and mapped to a three-dimensional thermal distribution map. When a person is detected entering a high-risk area, the calculation weight coefficient of the risk level signal is dynamically adjusted to generate an early warning command signal with a position correction factor.

[0016] In one embodiment of the present invention, the energy consumption decision module has a built-in gradient analysis algorithm, which extracts the time domain change rate of the standardized feature vector through a sliding window. When the change rate is lower than a first threshold in three consecutive sampling cycles, a low-power instruction is triggered. When the change rate exceeds a second threshold and lasts for two sampling cycles, a high-precision activation instruction is triggered, where the second threshold is more than five times the first threshold. The high-precision activation instruction includes a sensor sampling frequency increase instruction and a signal amplification circuit gain adjustment instruction.

[0017] In one embodiment of the present invention, the mode switching module includes an FPGA (field programmable gate array) programmable logic unit and a power supply topology switching circuit. When the FPGA programmable logic unit receives a low-power instruction, it reconstructs the sensor circuit into a multi-level sleep mode, wherein the first-level sleep mode turns off the sensor heating element, and the second-level sleep mode turns off the reference voltage source of the analog-to-digital converter. When the power supply topology switching circuit receives an activation instruction, it switches to a low-noise LDO power supply mode and starts the electromagnetic shielding cover to generate a state configuration signal with enhanced anti-interference.

[0018] In one embodiment of the present invention, the data fusion module runs a dynamic association matrix construction algorithm, which dynamically adjusts the weight distribution coefficients of environmental parameters and hazard parameters according to the risk level of the early warning command signal. When the risk level increases, the association weight of the volatile organic compound concentration and the temperature parameter is increased, and the humidity parameter weight is reduced to generate a diffusion prediction signal with risk adaptability, and the weight distribution coefficient is input as feedback to the noise separation unit of the preprocessing module.

[0019] In one embodiment of the present invention, the linkage module includes a multi-protocol industrial gateway and an instruction priority arbitrator. The multi-protocol industrial gateway simultaneously supports Modbus TCP, PROFINET and EtherCAT communication protocols. The instruction priority arbitrator performs conflict detection on the received early warning instruction signal and the diffusion prediction signal. When the exhaust start instruction and the area isolation instruction arrive at the same time, the area isolation instruction is executed first and the exhaust start instruction is delayed to generate an emergency braking instruction with timing logic.

[0020] In one embodiment of the present invention, the emergency braking instruction includes an equipment operation instruction set and a status verification mechanism. The equipment operation instruction set encapsulates the ventilation equipment inverter control parameters, the alarm device sound and light mode coding and the isolation access switch instructions. The status verification mechanism reads the equipment feedback current characteristics and compares them with the preset standard waveform. When an abnormal current waveform is detected, an equipment fault identification code is generated and a backup equipment switching instruction is triggered. The backup equipment switching instruction includes communication link redundancy switching and power path reconstruction instructions.

[0021] In one embodiment of the present invention, a self-calibration subsystem is also included. The self-calibration subsystem starts zero-point calibration and range calibration at preset time intervals by connecting a standard gas generator and a particle counter. During the zero-point calibration, all detection units are closed and pure nitrogen is injected to generate a baseline noise curve. During the range calibration, known concentrations of benzene standard gas and PM2.5 standard particles are released in sequence. The conversion coefficient of the standardized characteristic signal is dynamically corrected according to the sensor output value to generate a calibration factor that is fed back to the preprocessing module.

[0022] The present invention provides an intelligent air quality monitoring system based on sensor data feedback, which achieves breakthroughs through multi-dimensional technological innovation: first, a dynamically reconfigurable magnetic sensor array is constructed, which is compatible with the synchronous collection of environmental parameters (temperature, humidity, particulate matter) and occupational hazard factors (VOCs, radiation), and the heterogeneity of multi-source data is solved through hardware interface standardization; secondly, a dual-threshold energy-saving control algorithm is developed to automatically switch to micro-power consumption mode during low-risk periods, maintaining only the baseline monitoring function, and quickly activate high-precision sensors when parameter mutations are detected, combined with FPGA dynamic reconstruction technology to reduce standby power consumption by more than 80%; thirdly, the personnel positioning data and time-weighted exposure model are innovatively integrated to calculate the cumulative exposure dose of different positions in real time, and the pollutant diffusion path is predicted through deep learning to drive the ventilation system and isolation device to respond in advance. BRIEF DESCRIPTION OF THE DRAWINGS

[0023] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following briefly introduces the drawings required for describing the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0024] Figure 1 This is the system architecture diagram of the intelligent air quality monitoring system based on sensor data feedback. DETAILED DESCRIPTION

[0025] The following describes the embodiments of the present invention through specific examples. Those skilled in the art can easily understand other advantages and effects of the present invention from the content disclosed in this specification. The present invention can also be implemented or applied through other different specific embodiments. The details in this specification can also be modified or changed based on different viewpoints and applications without departing from the spirit of the present invention. It should be noted that the following embodiments and features in the embodiments can be combined with each other unless they conflict.

[0026] It should be noted that the illustrations provided in the following embodiments are merely schematic illustrations of the basic concept of the present invention. Therefore, the illustrations only show components related to the present invention and are not drawn according to the number, shape, and size of components in actual implementation. In actual implementation, the type, quantity, and proportion of each component may be changed arbitrarily, and the component layout may also be more complex.

[0027] In the following description, numerous details are discussed to provide a more thorough explanation of the embodiments of the present invention. However, it will be apparent to those skilled in the art that the embodiments of the present invention may be practiced without these specific details. In other embodiments, well-known structures and devices are shown in block diagram form rather than in detail to avoid obscuring the embodiments of the present invention.

[0028] See Figure 1 , shown is the air quality intelligent monitoring system based on sensor data feedback of the present invention. The air quality intelligent monitoring system based on sensor data feedback of the present invention includes a sensor array module, a preprocessing module, a risk assessment module, an energy consumption decision module, a mode switching module, a data fusion module and a linkage module. The sensor array module generates an environmental parameter signal; the preprocessing module receives the environmental parameter signal, eliminates the time delay difference and removes the noise, and generates a standardized characteristic signal; the risk assessment module receives the standardized characteristic signal, calculates the time-weighted concentration and the short-term exposure limit comparison result, and forms an early warning instruction signal; the energy consumption decision module analyzes the change gradient of the standardized characteristic vector and generates an activation instruction based on the vector stability; the mode switching module dynamically reconstructs the sensor circuit according to the activation instruction, selects to shut down non-essential units and generates a state configuration signal to feed back to the sensor array module; the data fusion module receives the standardized characteristic signal and the early warning instruction signal, establishes a correlation matrix between the environment and the hazard parameters, and generates a diffusion prediction signal; the linkage module controls the ventilation equipment and the alarm device according to the early warning instruction signal and the diffusion prediction signal, generates an emergency braking instruction and transmits the emergency braking instruction back to the data fusion module.

[0029] like Figure 1As shown, the present invention relates to an intelligent air quality monitoring system based on sensor data feedback, and its core architecture is composed of seven functional modules that form a closed-loop control through a bus. The sensor array module serves as a data acquisition terminal, and realizes the synchronous capture of multi-dimensional parameters of industrial sites by integrating fixed-installed environmental detection sensors and dynamically assembled occupational hazard detection sensors. Specifically, the environmental detection sensor group continuously collects basic environmental parameters, including but not limited to gas composition (such as oxygen, carbon dioxide concentration), suspended particulate matter (PM2.5 / PM10) and temperature and humidity data, while the occupational hazard detection sensor group is flexibly expanded through a magnetic interface, and professional detection units such as volatile organic compound (VOCs) detectors, ionizing radiation sensors or bioaerosol samplers can be added according to the type of on-site risk. The original signals generated by these sensors are converted into digital signals by internal conditioning circuits to form two types of data streams: environmental parameter signals and hazard parameter signals, which are transmitted to the preprocessing module through I2C or SPI buses. The preprocessing module is responsible for data cleaning and standardization. It features an internal timing alignment circuit that adds nanosecond-accurate timestamps to each sensor data packet, addressing timing misalignment caused by varying sensor response speeds. For example, the difference between a gas sensor's warm-up delay and an optical particulate matter sensor's immediate response can occur. The noise separation unit employs a two-stage filtering mechanism. The first stage performs spectral matching based on the sensor's factory-calibrated noise floor signature library to identify and eliminate the device's own thermal noise. The second stage employs a sliding average filter to suppress environmental electromagnetic interference. The resulting standardized signature signal, containing the denoised parameter values and their confidence indicators, is sent via parallel data channels to the risk assessment module and the energy consumption decision module, respectively.

[0030] like Figure 1As shown, after receiving the standardized characteristic signal, the risk assessment module activates a dual-model calculation engine: a time-weighted average (TWA) model calculates the cumulative exposure dose to workers at 15-minute intervals, while a short-term exposure limit (STEL) model monitors peak concentrations at 1-minute intervals. The results of these two models are integrated with the real-time UWB coordinate data uploaded by the personnel positioning subsystem to generate a dynamically updated thermal distribution map in a three-dimensional spatial coordinate system. When the risk value in a specific area exceeds a preset threshold, an early warning signal containing a risk level code (e.g., blue warning code 001, yellow warning code 002, red warning code 003) is generated and transmitted simultaneously to the data fusion module and the linkage module via the CAN bus. The energy consumption decision module continuously monitors the gradient of the standardized characteristic signal and analyzes parameter fluctuation trends using a sliding window algorithm. If the parameter change rate for five consecutive sampling periods is less than 0.5% / second, a low-power command is generated, triggering the mode switching module to enter energy-saving mode. If any parameter change rate exceeds 5% / second within three seconds, a high-precision activation command is immediately generated. After receiving these instructions, the mode switching module reconfigures the sensor power supply circuitry through the programmable logic device. For example, in low-power mode, it turns off the VOC sensor's heating element (reducing power consumption by 60%), while in high-precision mode, it activates the particle sensor's laser driver circuit (improving accuracy to 0.1μm). The reconfigured state configuration signal is fed back to the sensor array module, forming a closed loop for dynamic resource allocation.

[0031] Furthermore, the data fusion module receives the standardized characteristic signals from the preprocessing module and the warning command signals from the risk assessment module to construct a correlation matrix between environmental parameters and hazard parameters. This matrix uses a weighted Euclidean distance algorithm to calculate the correlation coefficients between the parameters. When the correlation between a specific parameter combination (e.g., temperature rise accompanied by an increase in VOC concentration) exceeds a critical value, the pollutant diffusion prediction algorithm is triggered. This algorithm, based on an improved LSTM neural network, uses historical data to train a spatial propagation model and outputs a 30-minute prediction of the pollutant diffusion path. The linkage module integrates the warning command signals with the diffusion prediction signals and sends control commands to field devices via an industrial Ethernet protocol (such as Profinet). For example, upon receiving a red warning code 003, the exhaust fan is immediately activated at full speed, triggering an audible and visual alarm. If the diffusion prediction signal indicates that the pollution cloud is about to spread to adjacent areas, a preemptive command to close the isolation gate is issued. All device response status data (e.g., fan speed, gate switch status) is transmitted back to the data fusion module via Modbus TCP to update the prediction model parameters, forming a complete "monitoring-decision-execution-feedback" control loop.

[0032] In one embodiment of the present invention, the specific structure and operating mode of the sensor array module are defined. This module adopts a layered sensor architecture and is divided into two parts: a fixed detection unit and an extended detection unit. The fixed detection unit is composed of three basic sensors: an electrochemical gas sensor (for detecting O2 and CO2), a laser scattering particulate matter sensor (for detecting PM1.0-PM10), and a digital temperature and humidity composite sensor. These sensors are rigidly integrated into the device body and continuously sample at a frequency of 1Hz to generate a baseline environmental parameter signal. The extended detection unit adopts a modular design and includes three magnetic interface slots compatible with PID photoionization VOC sensors, Geiger-Müller counter radiation sensors, and Anderson impactor bioaerosol samplers. The magnetic interface integrates a four-contact power supply system (12V DC positive and negative power supply, communication data line, and ground line) and a ring optical communication component. When the extended sensor is attached to the interface, the physical connection is first verified through electrical contacts. The optical communication component then transmits the sensor type code and range parameters via infrared pulses. The entire recognition process is completed within 200ms. The operating status of the extended detection unit is controlled by a status configuration signal sent by the mode switching module. In normal monitoring mode, only the fixed detection unit is active, while the extended unit remains in standby mode (power consumption ≤ 0.1W). Upon receiving a high-precision activation command, the magnetic interface's power supply contacts output 12V to activate the extended sensor, while an optical communication link establishes a data transmission channel. For VOCs detection, for example, the PID sensor uses an intermittent power supply strategy during its warm-up phase (approximately 30 seconds), switching on for one second every five seconds to reduce energy consumption. After warm-up, it enters continuous detection mode, outputting concentration values once a minute. The radiation sensor uses an event-triggered mechanism, automatically increasing the sampling frequency to 10Hz when it detects a radiation dose rate exceeding three times the background value. The bioaerosol sampler's operating cycle is linked to the ventilation system, extending the sampling time when the exhaust fan is activated to capture changes in the suspended particulate matter distribution. The hazard parameter signals generated by all extended units are integrated with the fixed detection unit signals through time-division multiplexing to form a complete environmental parameter signal stream. This dynamic assembly mechanism enables the system to flexibly configure detection capabilities based on the actual risk types of industrial sites. For example, VOCs sensors can be deployed in chemical plants and radiation sensors can be installed in nuclear facilities, while keeping basic environmental monitoring functions intact.

[0033] like Figure 1As shown in the figure, the internal structure and operating principle of the preprocessing module are specifically defined. At the hardware level, this module comprises two core components: a multi-channel buffer circuit and a noise separation unit. The multi-channel buffer circuit is implemented using an FPGA and features eight independent input channels, each corresponding to a different sensor signal input type. The channels contain two internal buffers: a raw data buffer (1MB capacity) for temporarily storing unprocessed raw sensor data, and a timestamp management unit that adds a 32-bit absolute time stamp (derived from the GPS module or system master clock) to each data packet. The timing alignment circuit compares the timestamps of data packets from each channel, using the latest arriving packet as a reference. It then uses a linear interpolation algorithm to compensate for timing deviations between channels, ensuring strict synchronization of all sensor data in the temporal dimension. For example, if the gas sensor data lags behind the particulate matter data by 50ms due to response delay, the system automatically extrapolates the theoretical value of the particulate matter sensor at that point in time to achieve data alignment.

[0034] Specifically, the noise separation unit consists of a digital signal processor (DSP) and an adaptive filter. Its workflow is divided into three phases. The first phase involves device fingerprint registration. During system initialization, each sensor collects background noise within a sealed chamber. The DSP then uses a fast Fourier transform (FFT) to extract the characteristic spectrum of each sensor, such as the low-frequency thermal noise (50-200Hz) of an electrochemical sensor and the high-frequency switching noise (2-5kHz) of an optical sensor. This process then creates a device noise signature library containing frequency characteristics, amplitude ranges, and phase characteristics. The second phase involves online noise identification, monitoring the energy distribution of each frequency band in the input signal in real time. When a fixed frequency component unrelated to environmental parameters is detected (such as the 120Hz ripple generated by the sensor heater), a digital notch filter is activated for band-stop filtering. The third phase performs environmental noise suppression, using a wavelet transform algorithm to separate sudden environmental interference (such as signal distortion caused by device vibration) from actual parameter changes. The resulting standardized signature signal contains not only the cleaned parameter value but also a signal-to-noise ratio (SNR) metric, which is used by subsequent modules to evaluate data reliability. For example, if the SNR falls below 20dB, the risk assessment module will automatically reduce the weight of this parameter. The standardized characteristic signal output by the pre-processing module is transmitted via a dual-redundant CAN bus, ensuring data transmission reliability meets industrial standards (bit error rate <10^-9).

[0035] Furthermore, the energy consumption decision module incorporates a built-in gradient analysis algorithm, whose core focus is dynamically evaluating parameter trends to achieve intelligent power consumption control. This algorithm utilizes a sliding time window mechanism with a configurable window length of 5 to 60 seconds to adapt to different industrial scenarios. During the data preprocessing phase, after receiving standardized feature vectors from the preprocessing module, the module first normalizes the feature parameters to eliminate the influence of different dimensions on the gradient calculation. For each feature parameter, the algorithm calculates the derivative of its rate of change within the window time and uses the least squares method to fit the slope of the change curve. If the absolute value of the slope falls below a preset stability threshold for multiple consecutive sampling periods, the current environment is considered to be in a steady state, triggering the generation of low-power instructions. During this process, the algorithm incorporates an inertial delay mechanism to avoid frequent mode switching: for example, the low-power instruction is not confirmed until a stable state is detected for three consecutive window periods, preventing misjudgments caused by transient interference. Conversely, if the instantaneous rate of change of any parameter exceeds the mutation threshold, and the change direction of two adjacent sampling points is consistent, it is immediately marked as an abnormal fluctuation event, triggering the high-precision activation instruction. This instruction not only contains the command to increase the sampling frequency of the sensor, but also specifies the gain adjustment parameters of the signal amplification circuit. For example, when a sudden increase in the VOCs concentration is detected, the amplifier gain of the PID sensor is automatically increased from the baseline gear to the high-sensitivity gear. During the instruction generation process, the module manages the alarm levels of different parameters through a priority queue to ensure that the detection priority of key hazard factors (such as combustible gas concentration) is higher than that of conventional environmental parameters. The generated instruction package encapsulates the target sensor identification code, operation type code and parameter configuration set, and transmits it to the mode switching module through an encrypted data frame. At the same time, the module continuously monitors the execution status of the instruction. When it detects that the parameter change trend is not effectively captured after the high-precision activation instruction is issued, it automatically starts the fault diagnosis process, writes the abnormal event record to the system log and triggers the sensor self-test program.

[0036] like Figure 1As shown, the mode switching module maximizes energy efficiency through hardware reconfiguration and power supply optimization. The module's core consists of an FPGA programmable logic unit and a multi-power topology switching circuit. The FPGA unit stores various sensor operating mode profiles, including full-power mode, level 1 sleep mode, and level 2 sleep mode. When receiving a low-power command, the FPGA first parses the target sensor list contained in the command and then loads the bitstream profile corresponding to the sleep mode. In level 1 sleep mode, the module reduces power consumption by disabling the sensor's heating element. For example, disabling the UV lamp powering a PID sensor causes the sensor to switch to passive detection mode, sacrificing some sensitivity in exchange for reduced power consumption. Level 2 sleep mode further disables the analog-to-digital converter's reference voltage source, maintaining only the base bias voltage for the sensor's analog circuitry. Data acquisition then switches to interval sampling mode, extending the sampling period to ten times that of normal mode. The power topology switching circuit utilizes a multi-stage switching architecture, consisting of a main power path and a backup low-noise LDO path. When a high-precision activation command arrives, the circuit first disconnects the sensor from the switching power supply, switching to the LDO path to reduce power supply ripple interference. Simultaneously, it activates the electromagnetic shield surrounding the sensor body. This shield, constructed of conductive fabric and grounded via a relay, releases accumulated static charge at the moment of activation. The module also integrates a current monitoring circuit to monitor the current consumption of each sensor branch in real time. If the actual current does not match the expected operating mode, it automatically triggers the power supply circuit's fuse protection mechanism and switches to the backup power channel within 50ms. The reconstructed state configuration signal contains each sensor's real-time operating parameters, supply voltage accuracy specifications, and an electromagnetic shielding status code. This signal is fed back to the sensor array module's control port via an optocoupler isolation circuit, ensuring electrical isolation between the high- and low-voltage circuits. Furthermore, the module maintains a sensor health database, recording changes in electrical parameters during each mode switch, providing a data foundation for predictive maintenance.

[0037] Furthermore, the data fusion module implements a dynamic correlation matrix construction algorithm, essentially building a multi-parameter synergistic model to achieve accurate pollution prediction. During the algorithm initialization phase, the module loads a predefined rule library for correlating environmental and hazard parameters, such as the coefficient of temperature on the volatilization rate of VOCs and the correlation factor between humidity and bioaerosol survival. Upon receiving a warning signal, the algorithm dynamically adjusts the matrix dimensions based on the risk level. During a blue warning state, the basic parameter correlation analysis is maintained, with a 5×5 matrix dimension. Upon upgrading to a yellow or red warning, the matrix automatically expands to an 8×8 higher-order matrix that includes derived parameters (such as the temperature-humidity synergy index and particle surface charge). Weight allocation coefficients are adjusted according to fuzzy logic rules. For example, as the risk level increases, the algorithm increases the weight of the volatile organic compound concentration parameter and introduces a time decay factor to give more weight to recent data. The pollutant dispersion prediction model utilizes a spatiotemporal joint modeling approach. Spatially, the industrial site is divided into 1m×1m grid cells, each containing current parameter values and historical trend data. Temporally, a sliding prediction window is established, whose length dynamically expands with the risk level. During the model training phase, a transfer learning strategy is employed to adapt a baseline model trained on public environmental datasets to field data, focusing on optimizing boundary condition handling capabilities. When a specific parameter combination reaches a critical synergistic effect (e.g., the photochemical reaction between ozone and VOCs in high-temperature and high-humidity environments), the model automatically triggers a secondary prediction process, invoking pre-set chemical reaction kinetic equations to correct the neural network output. The resulting diffusion prediction signal includes elements such as the pollution cloud's movement vector, concentration gradient distribution, and estimated arrival time. This signal is encapsulated into a structured data packet and transmitted to the linkage module via the Time-Sensitive Networking (TSN) protocol. Simultaneously, the algorithm provides real-time feedback on the weight distribution coefficients to the noise separation unit in the preprocessing module, guiding it to adjust filter parameter priorities. For example, as the risk level increases, filtering for high-frequency noise may be reduced to preserve more detailed features. The linkage module achieves precise device control through multi-protocol adaptation and intelligent arbitration. The multi-protocol industrial gateway features a modular design and supports concurrent processing of at least three industrial Ethernet protocols: Modbus TCP for connecting traditional ventilation equipment, PROFINET for controlling intelligent variable-frequency fans, and EtherCAT for driving high-speed isolated access control systems. The gateway is equipped with a protocol conversion engine that converts unified control instructions into protocol-specific message formats. For example, it converts the "exhaust volume 80%" instruction into Modbus TCP function code 06 to write a holding register operation, or EtherCAT's distributed clock synchronization control command.The command priority arbiter utilizes a multi-layered decision-making model: the first layer detects command conflicts and establishes a map of device operation impacts. For example, if exhaust fan activation could cause negative pressure in adjacent areas to exceed the specified value, delaying the isolation access control command. The second layer implements timing optimization, pipeline-scheduling non-conflicting commands and leveraging device response delays to insert secondary command execution. The third layer performs safety checks, comparing the current device state with the feasibility of control commands. For example, it intercepts speed-up commands during a fan overload alarm. Upon receiving a red alert command, the arbiter initiates an emergency response sequence, first issuing an area isolation command to close the physical access control. After a 500ms delay, the exhaust fan is activated to prevent airflow disturbances that could affect isolation. Finally, the audible and visual alarms are triggered and a status query command is sent to all controlled devices simultaneously. The emergency braking command is encapsulated in accordance with industrial safety protocol specifications and consists of an operation instruction set, an expected response time window, and a checksum. The operating instruction set uses Extensible Markup Language (XML) to describe device operation steps, such as the acceleration curve parameters for variable-frequency fans and the flashing frequency pattern of alarm lights. The expected response time window sets the maximum allowable response delay for each device; devices that fail to respond within the timeout are marked as faulty nodes. The checksum uses a cyclic redundancy check (CRC-32) and a hash algorithm to ensure the integrity of instruction transmission. All execution results are transmitted back to the data fusion module via the gateway's status feedback interface for updating the device health model and optimizing subsequent control strategies.

[0038] The intelligent air quality monitoring system based on sensor data feedback of the present invention achieves breakthroughs through multi-dimensional technological innovation: first, a dynamically reconfigurable magnetic sensor array is constructed, which is compatible with the synchronous collection of environmental parameters and occupational hazard factors, and the problem of multi-source data heterogeneity is solved through hardware interface standardization; secondly, a dual-threshold energy-saving control algorithm is developed to automatically switch to micro-power consumption mode during low-risk periods, maintaining only the baseline monitoring function, and quickly activate high-precision sensors when parameter mutations are detected, combined with FPGA dynamic reconstruction technology to reduce standby power consumption by more than 80%; thirdly, the personnel positioning data and time-weighted exposure model are innovatively integrated to calculate the cumulative exposure dose of different positions in real time, and the pollutant diffusion path is predicted through deep learning to drive the ventilation system and isolation devices to respond in advance.

[0039] Therefore, the intelligent air quality monitoring system based on sensor data feedback of the present invention can solve the problem of poor intelligent linkage between multi-source pollution monitoring, dynamic energy saving and real-time early warning in industrial sites.

[0040] The above embodiments are merely illustrative of the principles and effects of the present invention and are not intended to limit the present invention. Anyone skilled in the art may modify or alter the above embodiments without departing from the spirit and scope of the present invention. Therefore, all equivalent modifications or alterations made by one of ordinary skill in the art without departing from the spirit and technical principles disclosed herein are intended to be covered by the claims of the present invention.

Claims

1. An intelligent air quality monitoring system based on sensor data feedback, characterized in that: include: A sensor array module, wherein the sensor array module generates an environmental parameter signal; A preprocessing module receives the environmental parameter signal, eliminates time delay differences and removes noise, and generates a standardized characteristic signal; a risk assessment module, which receives the standardized characteristic signal, calculates a comparison result between the time-weighted concentration and the short-term exposure limit, and generates an early warning instruction signal; an energy consumption decision module, the energy consumption decision module analyzing a change gradient of the normalized feature vector and generating an activation instruction according to vector stability; a mode switching module, which dynamically reconfigures the sensor circuit according to the activation instruction, selectively shuts down non-essential units, and generates a state configuration signal to feed back to the sensor array module; A data fusion module receives the standardized characteristic signal and the early warning command signal, establishes a correlation matrix between environment and hazard parameters, and generates a diffusion prediction signal; A linkage module controls ventilation equipment and an alarm device according to the early warning command signal and the diffusion prediction signal, generates an emergency braking command, and transmits the emergency braking command back to a data fusion module.

2. The air quality intelligent monitoring system based on sensor data feedback according to claim 1 is characterized in that: The sensor array module includes a fixed detection unit consisting of a gas composition sensor, a particulate matter sensor, and a temperature and humidity sensor, and an extended detection unit consisting of a volatile organic compound sensor, a radiation sensor, and a bioaerosol sensor dynamically assembled through a magnetic interface. The fixed detection unit periodically collects baseline environmental parameters, and the extended detection unit is selectively activated according to the status configuration signal. When the activation instruction is triggered, the magnetic interface supplies power to the extended detection unit and establishes an optical communication link, generating a high-precision hazard parameter signal to supplement the environmental parameter signal.

3. The air quality intelligent monitoring system based on sensor data feedback according to claim 1 is characterized in that: The preprocessing module includes a multi-channel buffer circuit and a noise separation unit. The multi-channel buffer circuit uses a hardware timestamp to mark the data collection time of each sensor. The noise separation unit uses an adaptive filter based on device fingerprint recognition. The device fingerprint recognition establishes a noise feature library by analyzing the inherent thermal noise spectrum characteristics of the sensor. When the aliasing frequency in the environmental parameter signal is detected to match the feature library, the digital notch filter is activated to eliminate the interference of the sensor body and generate a standardized feature signal without device noise.

4. The air quality intelligent monitoring system based on sensor data feedback according to claim 1 is characterized in that: The risk assessment module set includes a personnel positioning subsystem, which includes a UWB (ultra-wideband) positioning base station deployed in an industrial site and a wearable device carrying a Bluetooth beacon. By fusing UWB arrival time difference positioning data with Bluetooth signal strength data, the real-time three-dimensional coordinates of the personnel are calculated and mapped to a three-dimensional thermal distribution map. When a person is detected entering a high-risk area, the calculation weight coefficient of the risk level signal is dynamically adjusted to generate the warning command signal with a position correction factor.

5. The air quality intelligent monitoring system based on sensor data feedback according to claim 1 is characterized in that: The energy consumption decision module has a built-in gradient analysis algorithm, which extracts the time domain change rate of the standardized feature vector through a sliding window. When the change rate is lower than a first threshold in three consecutive sampling cycles, a low-power instruction is triggered. When the change rate exceeds a second threshold and lasts for two sampling cycles, a high-precision activation instruction is triggered, where the second threshold is more than five times the first threshold. The high-precision activation instruction includes a sensor sampling frequency increase instruction and a signal amplification circuit gain adjustment instruction.

6. The air quality intelligent monitoring system based on sensor data feedback according to claim 1 is characterized in that: The mode switching module includes an FPGA (field programmable gate array) programmable logic unit and a power supply topology switching circuit. When the FPGA programmable logic unit receives the low-power instruction, it reconstructs the sensor circuit into a multi-level sleep mode, wherein the first-level sleep mode turns off the sensor heating element, and the second-level sleep mode turns off the reference voltage source of the analog-to-digital converter. When the power supply topology switching circuit receives an activation instruction, it switches to a low-noise LDO power supply mode and starts the electromagnetic shielding cover to generate a state configuration signal with enhanced anti-interference.

7. The air quality intelligent monitoring system based on sensor data feedback according to claim 1 is characterized in that: The data fusion module runs a dynamic association matrix construction algorithm, which dynamically adjusts the weight distribution coefficients of environmental parameters and hazard parameters according to the risk level of the early warning command signal. When the risk level increases, the association weight of the volatile organic compound concentration and the temperature parameter is increased, and the humidity parameter weight is reduced to generate a diffusion prediction signal with risk adaptability, and the weight distribution coefficient is input as feedback to the noise separation unit of the preprocessing module.

8. The air quality intelligent monitoring system based on sensor data feedback according to claim 1 is characterized in that: The linkage module includes a multi-protocol industrial gateway and an instruction priority arbitrator. The multi-protocol industrial gateway supports ModbusTCP, PROFINET and EtherCAT communication protocols at the same time. The instruction priority arbitrator performs conflict detection on the received early warning instruction signal and diffusion prediction signal. When the exhaust start instruction and the area isolation instruction arrive at the same time, the area isolation instruction is executed first and the exhaust start instruction is delayed to generate the emergency braking instruction with timing logic.

9. The air quality intelligent monitoring system based on sensor data feedback according to claim 1, characterized in that: The emergency braking instruction includes an equipment operation instruction set and a status verification mechanism. The equipment operation instruction set encapsulates the ventilation equipment inverter control parameters, the alarm device sound and light mode coding and the isolation access switch instruction. The status verification mechanism reads the equipment feedback current characteristics and compares them with the preset standard waveform. When an abnormal current waveform is detected, it generates an equipment fault identification code and triggers a backup equipment switching instruction. The backup equipment switching instruction includes communication link redundancy switching and power path reconstruction instructions.

10. The air quality intelligent monitoring system based on sensor data feedback according to claim 1, characterized in that: It also includes a self-calibration subsystem, which connects a standard gas generator and a particle counter to start zero-point calibration and range calibration at preset time intervals. During the zero-point calibration, all detection units are closed and pure nitrogen is injected to generate a baseline noise curve. During the range calibration, known concentrations of benzene standard gas and PM2.5 standard particles are released in sequence, and the conversion coefficient of the standardized characteristic signal is dynamically corrected according to the sensor output value to generate a calibration factor that is fed back to the preprocessing module.

Images