Distributed networking MEMS intelligent multi-parameter gas sensing terminal cooperative monitoring method and system

Abstract

The application relates to the technical field of environmental monitoring, and provides a MEMS intelligent multi-parameter gas sensing terminal cooperative monitoring method and system based on distributed networking. The method comprises the following steps: collecting original response signals and environmental physical quantities, and solving an initial concentration vector and a local confidence interval; receiving neighbor node data, constructing a spatial gradient field based on the initial concentration vector and the local confidence interval, and generating a spatial correction factor; calculating a sampling frequency and a heating power by using the spatial correction factor and the local confidence interval, and correcting to obtain a final concentration vector; reconstructing hardware parameters according to the sampling frequency and the heating power, and sending a cooperative broadcast frame containing the final concentration vector. The method solves the problems of response lag and energy waste caused by a fixed sampling strategy in the prior art, and realizes the consideration of monitoring real-time performance and low-power operation.

Description

Technical Field

[0001] This application relates to the field of environmental monitoring technology, and in particular to a distributed network MEMS intelligent multi-parameter gas sensing terminal collaborative monitoring method and system. Background Technology

[0002] With the popularization of the Industrial Internet of Things (IIoT), gas sensors based on microelectromechanical systems (MEMS) are widely used to build distributed monitoring networks to sense changes in ambient gas concentration in real time due to their advantages such as small size and low power consumption. However, existing technologies often employ fixed sampling strategies or static data fusion methods in dynamic airflow environments, which cannot dynamically adjust terminal operating parameters based on real-time spatial concentration gradients. This results in delayed response during sudden gas leaks or unnecessary energy waste in stable conditions, making it difficult to balance the requirements of real-time monitoring and low-power operation. Summary of the Invention

[0003] This application provides a distributed networked MEMS intelligent multi-parameter gas sensing terminal collaborative monitoring method and system to solve the problems mentioned in the background art.

[0004] In a first aspect, this application provides a collaborative monitoring method for distributed networked MEMS intelligent multi-parameter gas sensing terminals, including: The system collects raw response signals and environmental physical quantities, calculates the initial concentration vector and local confidence interval, receives neighbor node data, constructs a spatial gradient field based on the initial concentration vector and the local confidence interval, and generates a spatial correction factor. It then uses the spatial correction factor and the local confidence interval to calculate the sampling frequency and heating power, and corrects them to obtain the final concentration vector. Finally, it reconstructs hardware parameters based on the sampling frequency and the heating power, and sends a cooperative broadcast frame containing the final concentration vector.

[0005] Secondly, this application provides a distributed networked MEMS intelligent multi-parameter gas sensing terminal collaborative monitoring system, comprising: The acquisition module is used to acquire raw response signals and environmental physical quantities, and calculate the initial concentration vector and local confidence interval; the receiving module is used to receive data from neighboring nodes, construct a spatial gradient field based on the initial concentration vector and the local confidence interval, and generate a spatial correction factor; the calculation module is used to calculate the sampling frequency and heating power using the spatial correction factor and the local confidence interval, and correct to obtain the final concentration vector; the sending module is used to reconstruct hardware parameters based on the sampling frequency and the heating power, and send a cooperative broadcast frame containing the final concentration vector.

[0006] This application provides a distributed networked MEMS intelligent multi-parameter gas sensing terminal collaborative monitoring method and system. Firstly, by solving the initial concentration vector and local confidence interval based on the physical response equation, the method eliminates baseline drift caused by environmental fluctuations and quantifies the local measurement error boundary, thus providing a precise physical benchmark for subsequent spatial correction. Secondly, by constructing a spatial gradient field using neighbor node data and generating a spatial correction factor, the method identifies and corrects instantaneous interference and zero-point drift of local sensors based on network-wide collaborative information, thereby improving the accuracy of gas concentration monitoring in complex dynamic airflow environments. Thirdly, by dynamically adjusting the sampling frequency and heating power based on the spatial correction factor, the method achieves high-frequency rapid response during sudden gas leaks and enters a low-power operation mode when the environment is stable. This solves the response lag and energy waste problems caused by fixed sampling strategies in existing technologies, achieving a balance between real-time monitoring and low-power operation. Attached Figure Description

[0007] To more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings used in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0008] Figure 1 A flowchart illustrating the collaborative monitoring method for distributed networking MEMS intelligent multi-parameter gas sensing terminals provided in this application embodiment; Figure 2 A schematic block diagram of the structure of a distributed networked MEMS intelligent multi-parameter gas sensing terminal collaborative monitoring system provided in this application embodiment. Detailed Implementation

[0009] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0010] The flowchart shown in the attached diagram is for illustrative purposes only and does not necessarily include all content and operations / steps, nor does it necessarily have to be performed in the order described. For example, some operations / steps can be broken down, combined, or partially merged, so the actual execution order may change depending on the actual situation.

[0011] It should also be understood that the terminology used in this specification is for the purpose of describing particular embodiments only and is not intended to limit the scope of the application. As used in this specification and the appended claims, the singular forms “a,” “an,” and “the” are intended to include the plural forms unless the context clearly indicates otherwise.

[0012] It should also be further understood that the term “and / or” as used in this application specification and the appended claims means any combination of one or more of the associated listed items and all possible combinations, and includes such combinations.

[0013] The following detailed description of some embodiments of this application is provided in conjunction with the accompanying drawings. Unless otherwise specified, the following embodiments and features described herein can be combined with each other.

[0014] Please see Figure 1 , Figure 1 A flowchart illustrating the distributed networking MEMS intelligent multi-parameter gas sensing terminal collaborative monitoring method provided in this application embodiment is shown below. Figure 1 As shown, the method includes steps S1 to S4.

[0015] S1. Acquire the original response signal and environmental physical quantities, and calculate the initial concentration vector and local confidence interval. Specifically, a multi-channel gas sensor array synchronously receives the resistance change signals of various gases in the target area, and simultaneously reads the temperature, humidity, and air pressure values ​​output by the integrated sensor. The resistance change signal, along with the temperature, humidity, and air pressure values, are substituted into the conversion relationship describing the physical properties of the material to perform dynamic compensation calculations, thereby eliminating the interference of environmental fluctuations on the resistance signal and calculating the preliminary concentration value of each gas. Then, based on the fluctuation of heating power and the influence of air pressure on gas diffusion, an error propagation logic is constructed to calculate the combined result of the measurement uncertainty of each physical quantity in real time to determine the confidence range boundary of the concentration value. Finally, the set containing the preliminary concentration values ​​of all gases is defined as the initial concentration vector, and the determined confidence range boundary is defined as the local confidence interval, completing the quantitative calculation of the local state.

[0016] S2. Receive neighbor node data, construct a spatial gradient field based on the initial concentration vector and the local confidence interval, and generate a spatial correction factor. Specifically, receive data packets sent by surrounding nodes through a wireless communication interface and parse out the concentration vector, location coordinates, and time stamp of the neighbor nodes. Use the difference between the time stamp and the local clock, combined with the gas diffusion rate, to perform a time-dimensional shift on the concentration vector of the neighbor nodes to eliminate transmission lag. Substitute the time-aligned concentration vector and location coordinates into the discrete form of the physical equation describing the transport and diffusion of gas in the wind field. Combine the measured wind speed vector to solve for the theoretically existing background concentration distribution at the location of this terminal. Compare the initial concentration vector with the theoretical background concentration distribution. When the initial concentration vector exceeds the theoretical background concentration distribution and exceeds the local confidence interval, it is determined that there is a measurement deviation. Based on the direction and magnitude of the deviation, perform algebraic operations to derive a set of proportional coefficients or biases for correcting local measurements. Define this set of coefficients or biases as the spatial correction factor.

[0017] S3. Calculate the sampling frequency and heating power using the spatial correction factor and the local confidence interval, and correct to obtain the final concentration vector. Specifically, evaluate the magnitude of the spatial correction factor and its rate of change over time. When the rate of change exceeds a preset limit, indicating a drastic gradient change in the environment, set the sampling frequency to the highest level and increase the heater drive energy to improve the response speed. When the rate of change is below the preset limit and the local confidence interval is narrow, indicating a stable environment, reduce the sampling frequency and decrease the heater drive energy to maintain low-power operation. Simultaneously, read the initial concentration vector and the spatial correction factor, perform element-wise multiplication of each gas component value in the initial concentration vector with the corresponding proportionality coefficient in the spatial correction factor, or perform addition of the bias value to eliminate instantaneous interference or zero-point drift error of the local sensor. Output the set of concentration values ​​after network-wide collaborative correction as the final concentration vector, completing the dynamic adjustment of the monitoring strategy and improving data accuracy.

[0018] S4. Based on the sampling frequency and heating power, reconstruct the hardware parameters and send a cooperative broadcast frame containing the final concentration vector. Specifically, write the calculated sampling frequency configuration value into the timer control register and the heating power target value into the pulse width modulation control register, so that the hardware circuit operates according to the new parameter configuration to complete the working state reconstruction. Start the channel listening mechanism to detect the wireless spectrum occupancy. After confirming that the channel is in an idle state, encapsulate the data containing the final concentration vector, terminal identification, geographical location information and confidence interval into a standard format frame and transmit it through the radio frequency front end. During the transmission process and after the transmission is completed, listen to the feedback signal in the channel or the behavior characteristics of neighboring nodes. If a specific neighboring node is detected to be continuously unresponsive or the data packet verification error rate exceeds the standard, update the status mark of the node as invalid in the locally maintained neighbor list so as to exclude invalid data sources in the next monitoring cycle and complete the hardware reconstruction and information interaction closed loop. The method provided in this embodiment, on the one hand, eliminates baseline drift caused by environmental fluctuations and quantifies the local measurement error boundary by solving the initial concentration vector and local confidence interval based on the physical response equation, thereby providing an accurate physical reference for subsequent spatial correction; on the other hand, by constructing a spatial gradient field and generating a spatial correction factor using neighbor node data, it identifies and corrects the instantaneous interference and zero-point drift of the local sensor based on the network-wide collaborative information, thereby improving the accuracy of gas concentration monitoring in complex dynamic airflow environments; furthermore, by dynamically adjusting the sampling frequency and heating power based on the spatial correction factor, it achieves high-frequency rapid response in the event of a sudden gas leak and enters a low-power operation mode when the environment is stable, thereby solving the problem of response lag and energy waste caused by the fixed sampling strategy in the prior art, and achieving a balance between real-time monitoring and low-power operation.

[0019] In some embodiments, the acquisition of raw response signals and environmental physical quantities, and the calculation of the initial concentration vector and local confidence interval, include: S11. Acquire the original resistance response signal and environmental physical quantity data. Substitute the original resistance response signal and the environmental physical quantity data into the resistance-concentration conversion function for dynamic compensation to calculate the preliminary concentration value. Specifically, the analog voltage signal output by the gas sensor array is synchronously read through a multi-channel interface and converted from analog to digital to obtain the original resistance response signal. At the same time, the real-time values ​​output by the on-chip integrated temperature sensor, humidity sensor, and air pressure sensor are read as environmental physical quantity data. The resistance-to-concentration conversion formula based on physical laws, stored in the firmware, is called. The real-time measured temperature and humidity values ​​are substituted into the exponent and coefficient terms of the formula to dynamically adjust the original resistance response signal to offset the nonlinear changes in the material's sensitive characteristics under different temperature and humidity conditions and eliminate signal baseline drift caused by environmental fluctuations. Then, the dynamically adjusted resistance signal value is converted into the corresponding gas concentration value. The above conversion process is repeated for each target gas. All the calculated single gas concentration values ​​are summarized to form a preliminary concentration value set, completing the preliminary mapping from the original electrical signal to the physical concentration quantity.

[0020] S12. Based on the power stability fluctuation range and influencing factors, construct an error propagation model, and use the error propagation model to calculate the uncertainty of the preliminary concentration value in real time, and derive the local confidence interval. Specifically, the current feedback value and voltage feedback value of the heater drive circuit are monitored in real time, and their product is calculated to obtain the real-time heating power. The fluctuation range of the real-time heating power within a preset time window is defined as the power stability fluctuation range. At the same time, the air pressure value is read and substituted into the inverse relationship between the gas diffusion coefficient and the air pressure to obtain the gas diffusion coefficient influence factor. Based on the multivariate function error propagation law, an error propagation logic is constructed. The resistance to concentration conversion relationship is regarded as a multivariable function of the original resistance signal, temperature value, humidity value, power stability fluctuation range, and gas diffusion coefficient influence factor. The partial derivatives of this multivariable function with respect to each input variable are calculated respectively. The measurement uncertainty of each input variable is multiplied by the square of the corresponding partial derivative, and the sum is then taken to obtain the combined standard uncertainty. The coverage factor is determined according to the preset confidence probability level, and the combined standard uncertainty is multiplied by the coverage factor to obtain the expanded uncertainty value. The initial concentration value minus the expanded uncertainty value is used as the lower limit value, and the initial concentration value plus the expanded uncertainty value is used as the upper limit value. The lower limit value and the upper limit value constitute the local confidence interval.

[0021] S13. Combine the preliminary concentration values ​​into an initial concentration vector, and temporarily store the initial concentration vector and the local confidence interval in random access memory. Specifically, arrange the set of preliminary concentration values ​​containing multiple gas component values ​​obtained in the previous step into a multi-dimensional array according to a predefined gas type order. Define this multi-dimensional array as the initial concentration vector. Read the lower limit set and upper limit set of the local confidence interval obtained in the previous step. Package the initial concentration vector, the lower limit set, and the upper limit set into a complete data structure. Write this data structure into the designated address space of the high-speed random access memory through the bus interface so that subsequent steps can directly call the values ​​in this data structure as local reference and error boundary constraints when constructing the spatial gradient field and generating the spatial correction factor. This ensures the consistency of the reference to the local measurement state in each stage of the data processing flow and completes the preparation for persistent storage and sharing of the local solution results. The method provided in this embodiment, on the one hand, eliminates the interference of temperature and humidity fluctuations on the sensor baseline by dynamically compensating for environmental physical quantities in the resistance-to-concentration conversion function, thereby obtaining a preliminary concentration value unaffected by environmental noise; on the other hand, by constructing an error propagation model based on the influence of power fluctuations and air pressure and calculating the uncertainty in real time, a local confidence interval that dynamically changes with operating conditions is derived, thus providing a quantitative criterion for distinguishing measurement noise from the true concentration gradient; furthermore, by combining the preliminary concentration values ​​into a vector and binding them to the confidence interval for storage, standardized management of the data structure is achieved, thereby laying a solid data foundation for efficient reading and collaborative calculation in subsequent steps.

[0022] In some embodiments, receiving neighbor node data, constructing a spatial gradient field based on the initial concentration vector and the local confidence interval, and generating a spatial correction factor include: S21. Receive and parse the cooperative data packet to obtain the neighbor's initial concentration vector, geographical coordinates, and timestamp. Use the difference between the timestamp and the local system clock to perform time shift compensation on the neighbor's initial concentration vector to obtain time-aligned concentration data. Specifically, the wireless communication module listens for and receives cooperative data packets sent by surrounding neighbor nodes in the previous broadcast time slot. The data packet is parsed to extract the neighbor's initial concentration vector, neighbor node geographical coordinates, and data packet generation timestamp contained therein. The current time value of the local system clock is read and the time difference with the data packet generation timestamp is calculated. The time delay required for gas to propagate from the neighbor node to the local terminal is estimated by combining the average diffusion speed of gas in the air. The time difference and propagation delay are used to perform time-dimensional shift correction on the values ​​of each gas component in the neighbor's initial concentration vector. The concentration values ​​measured by the neighbor node in the past are extrapolated to the equivalent concentration values ​​at the current time, eliminating the spatiotemporal asynchrony error caused by wireless transmission delay and gas physical propagation lag. The set of concentration values ​​after shift correction is defined as time-aligned concentration data, providing time-consistent data input for constructing an accurate spatial gradient field.

[0023] S22. Substitute the time-aligned concentration data and the geographical coordinates into the discretized form of the convection-diffusion equation, and solve for the theoretical background concentration field distribution at this terminal location by combining the wind speed vector. Specifically, the parsed neighboring node geographic coordinates are mapped to a local coordinate system centered on the terminal to obtain a spatial position vector. The wind speed and wind direction measured by the local meteorological sensor are read and decomposed into components along each axis of the local coordinate system to form a wind speed vector. The finite difference discretization formula based on the convection-to-diffusion equation of the law of conservation of mass is called. The current ambient temperature and air pressure values ​​are substituted into the empirical formula to calculate the gas diffusion coefficient. The gas diffusion coefficient, wind speed vector, and preset time step and spatial grid step are substituted into the coefficient terms of the discretization formula to construct a set of linear equations describing the transmission of gas concentration in the field. The time-aligned concentration data is used as the boundary condition constraint of the linear equations at the corresponding spatial position vector. The linear equations are solved using an iterative algorithm to calculate the concentration estimate of each discrete grid point in the local coordinate system to form a set of discretized background concentration field data. The interpolation algorithm is used to extract and calculate the continuous concentration value at the location of the terminal from the set of discretized background concentration field data based on the precise coordinates of the terminal in the local coordinate system. The values ​​are arranged by gas type to form a theoretical background concentration field distribution.

[0024] S23. Compare the initial concentration vector with the theoretical background concentration field distribution. If the initial concentration vector exceeds the theoretical background concentration field distribution and exceeds the local confidence interval, then derive the spatial correction factor through algebraic operations based on the deviation direction and magnitude. Specifically, the initial concentration vector obtained from local calculation and the theoretical background concentration field distribution obtained from neighboring data are read. The concentration deviation vector is obtained by subtracting the corresponding gas component values ​​in the theoretical background concentration field distribution from the values ​​of each gas component in the initial concentration vector. The upper and lower limits of the local confidence interval are read and the difference is calculated to obtain the confidence interval width. It is determined whether the absolute value of each component in the concentration deviation vector is greater than half of the confidence interval width. Only when the absolute value of the deviation of a gas component is greater than half of the confidence interval width is it determined that the component has a systematic deviation. For components with systematic deviation, the corresponding component value of the concentration deviation vector is divided by the corresponding gas component value in the theoretical background concentration field distribution to obtain the relative deviation ratio. The unit value is added or subtracted according to the sign of the relative deviation ratio to obtain the proportional correction coefficient. The correction coefficients of all gas components are arranged in order to form a spatial correction factor vector. This vector is used to correct the local measurement data in subsequent steps. The method provided in this embodiment, on the one hand, eliminates the spatiotemporal asynchrony error caused by transmission and diffusion delays by performing time shift compensation on the data of neighboring nodes, thereby obtaining concentration data that is strictly aligned with the local time; on the other hand, by solving the discretized form of the convection-diffusion equation and combining it with the wind speed vector inversion to retrieve the theoretical background concentration field, a network-wide collaborative sensing benchmark based on physical diffusion laws is constructed, thus providing an objective basis for identifying local measurement deviations; furthermore, by comparing the initial concentration vector with the distribution of the theoretical background concentration field and deriving the spatial correction factor based on the deviation direction and magnitude, the method achieves accurate quantification and correction of systematic errors exceeding the confidence interval, thereby improving the consistency and reliability of single-point monitoring data in the distributed network.

[0025] In some embodiments, substituting the time-aligned concentration data and the geographic location coordinates into the discretized form of the convection-diffusion equation, and combining this with wind speed vector calculations to obtain the theoretical background concentration field distribution at the terminal location, includes: S221. Construct a local coordinate system with the terminal as the origin and map the geographic coordinates to obtain a spatial position vector. Decompose the measured wind speed data to obtain a local wind speed vector. Specifically, establish a three-dimensional Cartesian coordinate system with the physical location of the terminal as the origin as a local reference frame. Read the geographic latitude, longitude, and altitude values ​​of each neighboring node obtained from the analysis. Convert the geographic coordinates of each neighboring node into three-dimensional rectangular coordinate values ​​relative to the origin of the terminal using a coordinate transformation algorithm. Define the converted three-dimensional rectangular coordinate values ​​as the spatial position vector of each neighboring node. At the same time, read the wind speed scalar value and wind direction angle value output by the local meteorological sensor. Decompose the wind speed scalar value and wind direction angle value into three component values ​​along the X-axis, Y-axis, and Z-axis of the local coordinate system using trigonometric function relationships. Combine these three component values ​​into a local wind speed vector. This local wind speed vector describes the advection transport direction and intensity of the wind field in the local coordinate system at the current moment, providing the necessary flow field boundary conditions for subsequent substitution into the gas diffusion equation.

[0026] S222. Calculate the gas diffusion coefficient. Substitute the gas diffusion coefficient, the local wind speed vector, and the preset step size into the finite difference discretization formula to construct a linear equation system. Specifically, read the current ambient temperature and air pressure values, substitute them into the empirical relationship describing the gas molecule motion characteristics to calculate the current gas diffusion coefficient value, and call the finite difference discretization formula of the unsteady convection-to-diffusion equation stored in the firmware. This formula describes the spatiotemporal evolution law of gas under the combined action of advection transport and turbulent diffusion in the wind field. Substitute the calculated gas diffusion coefficient value, the values ​​of each component of the local wind speed vector, and the preset time step size and spatial grid division step size into the coefficient terms of the finite difference discretization formula. For each discrete grid point in the local coordinate system, list an algebraic equation describing its concentration change rate. Combine the algebraic equations of all grid points to form a large sparse linear equation system. The unknowns of this linear equation system are the gas concentration estimates of each grid point at the current time. The coefficient matrix is ​​determined by the diffusion coefficient, wind speed vector, and step size parameters.

[0027] S223. Solve the linear equations using the time-aligned concentration data as boundary conditions to obtain a discretized background concentration field data set. Specifically, the time-aligned concentration data of each neighbor node after time-shift compensation processing is mapped to the grid point or adjacent grid point where the spatial position vector of the corresponding neighbor node is located in the linear equations. The concentration values ​​of these mapped points are set as known boundary conditions or source term constraints of the linear equations. Numerical solutions are performed on the constructed linear equations using the Gauss-Seidel iteration method or direct matrix solving algorithm. After meeting the convergence accuracy requirements, the solution vector of the linear equations is obtained. This solution vector contains the gas concentration estimates of all discrete grid points in the local coordinate system at the current time. The coordinates of all grid points are paired and stored with their corresponding concentration estimates to form a discretized background concentration field data set covering the entire local monitoring area. This set reflects the spatial concentration distribution pattern derived only from neighbor data and physical diffusion laws.

[0028] S224. Based on the coordinates of the terminal in the local coordinate system, interpolation is performed on the background concentration field data set to obtain the theoretical background concentration field distribution. Specifically, the precise coordinate values ​​of the terminal in the established local coordinate system, i.e., the origin coordinates, are determined. Several adjacent grid points surrounding the origin coordinates of the terminal and their corresponding concentration estimates are selected from the discretized background concentration field data set. A bilinear interpolation algorithm or a trilinear interpolation algorithm is used. Based on the geometric distance ratio between the origin coordinates of the terminal and the selected adjacent grid point coordinates, the concentration estimates of the adjacent grid points are weighted and calculated to obtain the continuous gas concentration values ​​at the origin position of the terminal. The above interpolation calculation process is repeated for each target gas. The interpolated concentration values ​​of all gas types at the terminal position are arranged in a predetermined order to form a theoretical background concentration field distribution vector. This vector represents the theoretically existing gas concentration level at the terminal under the condition of no local sensor error interference.

[0029] The method provided in this embodiment, on the one hand, unifies the spatial reference system and quantifies the wind field driving factors by constructing a local coordinate system and decomposing the wind speed vector, thereby providing accurate geometric and dynamic parameters for solving the physical equations; on the other hand, by substituting the diffusion coefficient and the wind speed vector to construct and solve a system of linear equations, the discrete neighbor data is reconstructed into a continuous background concentration field, thereby realizing spatial information interpolation and completion based on physical laws; furthermore, by performing interpolation operations on the background concentration field dataset to extract the concentration at the current terminal location, a high-precision theoretical background concentration value is obtained, thereby providing an accurate benchmark without mesh error interference for subsequent deviation determination.

[0030] In some embodiments, the process of deriving the spatial correction factor through algebraic operations based on the direction and magnitude of the deviation includes: S231. Calculate the difference between the local initial concentration vector and the theoretical background concentration field distribution to obtain the concentration deviation vector, and calculate the difference between the upper and lower limits of the local confidence interval to obtain the confidence interval width. Specifically, read the values ​​of each gas component in the locally calculated initial concentration vector, read the component values ​​of the corresponding gas type in the theoretical background concentration field distribution vector constructed through the spatial gradient field, subtract the corresponding component value in the theoretical background concentration field distribution vector from each component value in the initial concentration vector to obtain a set of difference values ​​containing positive and negative signs. This set of difference values ​​is defined as the concentration deviation vector, where the magnitude of each component value represents the deviation amplitude, and the positive and negative signs represent the deviation direction. At the same time, read the upper and lower limit sets of the local confidence interval, subtract the corresponding value in the lower limit set from each value in the upper limit set to obtain a set of width values ​​characterizing the measurement uncertainty range. This set of width values ​​is defined as the confidence interval width, which is used to determine whether the deviation is significant in subsequent tests.

[0031] S232. Determine whether the absolute value of each component of the concentration deviation vector is greater than half the width of the confidence interval, and initialize the correction coefficient corresponding to the component that is not greater than half the width of the confidence interval to a unit value. Specifically, iterate through the value of each component in the concentration deviation vector, calculate the absolute value of each component, and simultaneously read the corresponding component value in the confidence interval width and calculate half of its value. Compare the absolute value of the concentration deviation component with the value of half the width of the confidence interval. When the absolute value of the deviation of a gas component is less than or equal to the value of half the width of the confidence interval, it is determined that the local measurement value of the gas component is within the reliable range and there is no significant systematic deviation. The correction coefficient value corresponding to the gas component is directly set to the unit value 1, indicating that no correction is needed for the component, and its original measurement result is retained. Perform the same initialization operation on all gas components that meet this condition to form a partial correction coefficient set.

[0032] S233. Divide the component value greater than half the width of the confidence interval by the corresponding gas component value in the theoretical background concentration field distribution to obtain the relative deviation ratio. Based on the sign of the relative deviation ratio, perform addition and subtraction operations on the unit value to obtain the proportional correction coefficient. Specifically, for the component in the concentration deviation vector whose absolute value is greater than half the width of the confidence interval, read the original deviation value of the component, read the corresponding gas component value in the theoretical background concentration field distribution, divide the deviation value by the gas component value in the theoretical background concentration field distribution to obtain a dimensionless relative deviation ratio value, determine the sign of the relative deviation ratio value. When the relative deviation ratio value is positive, it indicates that the local measurement value is higher than the theoretical background value. Subtract the relative deviation ratio value from the unit value 1 to obtain the proportional correction coefficient. When the relative deviation ratio value is negative, it indicates that the local measurement value is lower than the theoretical background value. Add the absolute value of the relative deviation ratio value to the unit value 1 to obtain the proportional correction coefficient. Calculate a specific proportional correction coefficient for each gas component with significant deviation through the above algebraic operations.

[0033] S234. Arrange the correction coefficients corresponding to all components and the proportional correction coefficients in order to obtain the spatial correction factor vector. Specifically, collect the correction coefficient values ​​corresponding to all gas components initialized to a unit value of 1 in the collection step, collect all proportional correction coefficient values ​​obtained through algebraic operations in the collection step, and arrange these two types of coefficient values ​​in order according to the definition order of gas types to form a complete set of coefficients with the same dimension as the initial concentration vector. Define this complete set of coefficients as the spatial correction factor vector. Each element in this vector is a purely numerical proportional coefficient, without any statistical weights or model prediction parameters. It is directly used in subsequent steps to perform element-wise multiplication with the initial concentration vector to eliminate measurement errors, thus completing the final construction and output of the spatial correction factor.

[0034] The method provided in this embodiment, on the one hand, quantifies the degree of deviation between local measurement and theoretical expectation and establishes a dynamic judgment threshold by calculating the concentration deviation vector and the confidence interval width, thereby providing an accurate data basis for deviation identification; on the other hand, it effectively filters random noise interference and initiates correction only for significant deviations by initializing the correction coefficient based on a logical judgment of half the confidence interval width, thereby avoiding the occurrence of miscorrection; furthermore, by calculating the relative deviation ratio and generating a proportional correction coefficient based on positive and negative values, it achieves adaptive quantification and correction of systematic deviations in different directions, thereby outputting a spatial correction factor vector that can accurately eliminate drift errors.

[0035] In some embodiments, the step of calculating the sampling frequency and heating power using the spatial correction factor and the local confidence interval, and correcting to obtain the final concentration vector, includes: S31. Evaluate the magnitude and rate of change of the spatial correction factor. If the rate of change is greater than a preset threshold, set the sampling frequency to the maximum value and increase the heating power. If the rate of change is not greater than the preset threshold and the width of the local confidence interval is less than the preset width, reduce the sampling frequency and decrease the heating power. Specifically, the system reads the values ​​of each component in the spatial correction factor vector, calculates the rate of change of each component value within the continuous monitoring period, and compares the rate of change value with a preset threshold value. When the rate of change value is greater than the preset threshold value, it indicates that there is a drastic gradient change between the local environment and the surrounding network environment or that a leakage source is approaching. The sampling frequency configuration parameter of the analog-to-digital converter is set to the maximum allowable value, and the driving duty cycle value of the MEMS heater is increased to improve the sensor operating temperature. When the rate of change value is less than or equal to the preset threshold value and the width of the local confidence interval read is less than the preset width value, it indicates that the environment is in a stable state and the measurement data is highly reliable. The sampling frequency configuration parameter of the analog-to-digital converter is lowered to the value required to maintain basic monitoring, and the driving duty cycle value of the MEMS heater is reduced to reduce energy consumption. This completes the dynamic calculation of the adaptive sampling strategy and heating power.

[0036] S32. Perform element-wise multiplication of the initial concentration vector with the spatial correction factor to obtain the final concentration vector after eliminating drift error. Specifically, read the values ​​of each gas component in the locally calculated initial concentration vector, read the corresponding proportional coefficient values ​​in the calculated spatial correction factor vector, multiply the first gas component value in the initial concentration vector with the first proportional coefficient value in the spatial correction factor vector to obtain the first corrected concentration value, and sequentially perform multiplication operations on each gas component value in the initial concentration vector with the corresponding proportional coefficient value in the spatial correction factor vector. Rearrange all the results of the multiplication operations according to the original gas type order to form a new set of concentration values. Define this set as the final concentration vector. The values ​​in this vector have eliminated the measurement errors caused by sensor zero-point drift or local instantaneous interference, and represent high-precision gas concentration data after network-wide collaborative correction.

[0037] The method provided in this embodiment, on the one hand, achieves adaptive matching of the monitoring strategy with the dynamics of gas diffusion by evaluating the rate of change of the spatial correction factor and dynamically adjusting the sampling frequency and heating power, thereby ensuring high-frequency response during sudden leaks and achieving energy-saving operation during stable periods; on the other hand, by performing element-wise multiplication of the initial concentration vector with the spatial correction factor, the measurement error caused by sensor zero-point drift or local instantaneous interference is directly eliminated, thereby outputting a high-precision final concentration vector that integrates information from the entire network; furthermore, by establishing a deterministic mapping between environmental conditions and hardware parameters, the response lag and energy waste caused by fixed strategies are avoided, thereby achieving an optimized balance between real-time monitoring and low-power operation.

[0038] In some embodiments, the step of reconstructing hardware parameters based on the sampling frequency and the heating power, and sending a cooperative broadcast frame containing the final concentration vector, includes: S41. Write the configuration value of the sampling frequency into the timer register and the value of the heating power into the pulse width modulation control register to complete the reconstruction of the hardware operating parameters. Specifically, read the sampling frequency configuration value calculated in the previous step, and write the value into the timer register that controls the working rhythm of the analog-to-digital converter through the microcontroller bus interface, so that the timer generates a trigger signal according to the new count value. Read the target value of the heating power calculated in the previous step, and write the value into the pulse width modulation control register that controls the MEMS heater drive circuit. Adjust the duty cycle of the output waveform to change the average input power of the heater. After the register writing operation is completed, the hardware circuit immediately collects data according to the new sampling frequency and performs temperature control according to the new heating power to realize the real-time reconstruction of the hardware operating parameters, ensuring that the terminal is in the optimal working state at the beginning of the next monitoring cycle and avoiding data aliasing or response lag during parameter switching.

[0039] S42. Detect the wireless channel status. After confirming that the channel is idle, transmit the cooperative broadcast frame through the radio frequency front end. Specifically, activate the carrier sensing mechanism, detect the signal energy level of the wireless channel through the radio frequency front end, and determine whether the channel is currently occupied by other nodes. When the channel signal energy is detected to be lower than a preset threshold value, it is determined that the channel is idle. Immediately encapsulate the data packet containing the final concentration vector, terminal identification, geographical location information, updated confidence interval, and time slot length suggestion into a cooperative broadcast frame conforming to the communication protocol standard. Activate the radio frequency transmission circuit to modulate the cooperative broadcast frame onto the carrier frequency and transmit it through the antenna, so that other neighboring nodes in the network can receive the high-precision data corrected by this terminal, completing the broadcast sharing of cooperative information. If the channel is detected to be busy, wait for a random duration and then re-detect until the channel is idle.

[0040] S43. Listening to acknowledgment signals or listening results in the monitoring channel, if a neighbor node is found to be unresponsive or the data packet verification error rate exceeds a threshold, the status of the neighbor node marked in the active neighbor topology list is updated to filter valid neighbor data in the next monitoring cycle. Specifically, after sending a cooperative broadcast frame, the wireless channel is continuously monitored to detect whether other nodes send acknowledgment signals or to observe the data forwarding behavior of neighbor nodes in subsequent time slots. The response status and data packet verification results of each neighbor node in multiple consecutive monitoring cycles are statistically analyzed. When a neighbor node is found to have failed to send acknowledgment signals multiple times consecutively or the data packet verification error rate of its sent data exceeds a preset threshold, it is determined that the communication link of the neighbor node is invalid or the data is untrustworthy. The record of the neighbor node is searched in the locally stored active neighbor topology list, and its status is changed to invalid or untrustworthy. When the neighbor data reception and parsing steps are performed in the next monitoring cycle, neighbor nodes marked as invalid or untrustworthy are automatically skipped to prevent erroneous data from polluting the construction of the spatial gradient field and to complete the dynamic self-healing update of the network topology.

[0041] The method provided in this embodiment, on the one hand, completes hardware parameter reconstruction by directly writing to registers, thereby realizing the rapid conversion of software policies to hardware execution, and thus ensuring the immediate effect of new working parameters and the continuity of data acquisition; on the other hand, it sends a cooperative broadcast frame after confirming idleness through a channel listening mechanism, thereby avoiding wireless data transmission conflicts and ensuring the reliable transmission and sharing of cooperative data across the entire network; furthermore, by listening to confirmation signals and dynamically updating the neighbor topology list status, it realizes the automatic identification and isolation of communication failure nodes, thereby ensuring the data fusion quality of the cooperative monitoring network and the stability of long-term system operation in the event of node failure.

[0042] Please see Figure 2 , Figure 2 A schematic block diagram of the distributed networked MEMS intelligent multi-parameter gas sensing terminal collaborative monitoring system provided in this application embodiment is shown below. Figure 2 As shown in the embodiment of this application, the distributed networked MEMS intelligent multi-parameter gas sensing terminal collaborative monitoring system includes: Acquisition module 110 is used to acquire raw response signals and environmental physical quantities, and calculate the initial concentration vector and local confidence interval. Receiving module 120 is used to receive data from neighboring nodes, construct a spatial gradient field based on the initial concentration vector and the local confidence interval, and generate a spatial correction factor. Calculation module 130 is used to calculate the sampling frequency and heating power using the spatial correction factor and the local confidence interval, and correct to obtain the final concentration vector. Transmitting module 140 is used to reconstruct hardware parameters based on the sampling frequency and the heating power, and transmit a cooperative broadcast frame containing the final concentration vector.

[0043] It should be noted that those skilled in the art will understand that, for the sake of convenience and brevity, the specific working processes of the system and its modules described above can be referred to the corresponding processes in the aforementioned distributed networking MEMS intelligent multi-parameter gas sensing terminal collaborative monitoring method embodiment, and will not be repeated here.

[0044] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any person skilled in the art can easily conceive of various equivalent modifications or substitutions within the technical scope disclosed in this application, and these modifications or substitutions should all be covered within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

Claims

1. A method for collaborative monitoring of distributed networked MEMS intelligent multi-parameter gas sensing terminals, characterized in that, include: The original response signal and environmental physical quantities are collected, and the initial concentration vector and local confidence interval are calculated. Receive neighbor node data, construct a spatial gradient field based on the initial concentration vector and the local confidence interval, and generate a spatial correction factor; calculate the sampling frequency and heating power using the spatial correction factor and the local confidence interval, and correct to obtain the final concentration vector; reconstruct hardware parameters according to the sampling frequency and the heating power, and send a cooperative broadcast frame containing the final concentration vector.

2. The distributed networking MEMS intelligent multi-parameter gas sensing terminal collaborative monitoring method according to claim 1, characterized in that, The process of collecting raw response signals and environmental physical quantities, and calculating the initial concentration vector and local confidence interval includes: The raw resistance response signal and environmental physical quantity data are collected. The raw resistance response signal and the environmental physical quantity data are substituted into the resistance-concentration conversion function for dynamic compensation to calculate the preliminary concentration value. An error propagation model is constructed based on the power stability fluctuation range and influencing factors. The uncertainty of the preliminary concentration value is calculated in real time using the error propagation model to derive the local confidence interval. The preliminary concentration values ​​are combined into an initial concentration vector, and the initial concentration vector and the local confidence interval are temporarily stored in random access memory.

3. The distributed networking MEMS intelligent multi-parameter gas sensing terminal collaborative monitoring method according to claim 1, characterized in that, The process of receiving neighbor node data, constructing a spatial gradient field based on the initial concentration vector and the local confidence interval, and generating a spatial correction factor includes: The system receives and parses the neighbor's initial concentration vector, geographic location coordinates, and timestamp. It then uses the difference between the timestamp and the local system clock to perform time-shift compensation on the neighbor's initial concentration vector, obtaining time-aligned concentration data. The time-aligned concentration data and the geographic location coordinates are substituted into the discretized form of the convection-diffusion equation, and combined with the wind speed vector, the theoretical background concentration field distribution at the terminal's location is obtained. The initial concentration vector is compared with the theoretical background concentration field distribution. If the initial concentration vector exceeds the theoretical background concentration field distribution and exceeds the local confidence interval, a spatial correction factor is derived through algebraic operations based on the deviation direction and magnitude.

4. The distributed networking MEMS intelligent multi-parameter gas sensing terminal collaborative monitoring method according to claim 3, characterized in that, The step of substituting the time-aligned concentration data and the geographical coordinates into the discretized form of the convection-diffusion equation, and combining this with wind speed vector calculations to obtain the theoretical background concentration field distribution at the terminal location, includes: A local coordinate system with the terminal as the origin is constructed and the geographical coordinates are mapped to obtain a spatial position vector. The measured wind speed data is decomposed to obtain a local wind speed vector. The gas diffusion coefficient is calculated, and the gas diffusion coefficient, the local wind speed vector, and the preset step size are substituted into the finite difference discretization formula to construct a system of linear equations. The time-aligned concentration data is used as boundary conditions to solve the system of linear equations to obtain a discretized background concentration field data set. Based on the coordinates of the terminal in the local coordinate system, the background concentration field data set is interpolated to obtain the theoretical background concentration field distribution.

5. The distributed networking MEMS intelligent multi-parameter gas sensing terminal collaborative monitoring method according to claim 3, characterized in that, The spatial correction factor, derived algebraically based on the direction and magnitude of the deviation, includes: The concentration deviation vector is obtained by calculating the difference between the local initial concentration vector and the theoretical background concentration field distribution, and the confidence interval width is obtained by calculating the difference between the upper and lower limits of the local confidence interval. It is then determined whether the absolute value of each component of the concentration deviation vector is greater than half the confidence interval width, and the correction coefficients corresponding to the components not greater than half the confidence interval width are initialized to unit values. The values ​​of the components greater than half the confidence interval width are divided by the corresponding gas component values ​​in the theoretical background concentration field distribution to obtain the relative deviation ratio. Based on the positive or negative sign of the relative deviation ratio, the unit values ​​are added or subtracted to obtain the proportional correction coefficient. Finally, the spatial correction factor vector is obtained by arranging the correction coefficients corresponding to all components and the proportional correction coefficients in order.

6. The distributed networking MEMS intelligent multi-parameter gas sensing terminal collaborative monitoring method according to claim 1, characterized in that, The step of calculating the sampling frequency and heating power using the spatial correction factor and the local confidence interval, and then correcting them to obtain the final concentration vector, includes: The magnitude and rate of change of the spatial correction factor are evaluated. If the rate of change is greater than a preset threshold, the sampling frequency is set to the maximum value and the heating power is increased. If the rate of change is not greater than the preset threshold and the width of the local confidence interval is less than the preset width, the sampling frequency is reduced and the heating power is decreased. The initial concentration vector and the spatial correction factor are multiplied element-wise to obtain the final concentration vector that eliminates drift error.

7. The distributed networking MEMS intelligent multi-parameter gas sensing terminal collaborative monitoring method according to claim 1, characterized in that, The step of reconstructing hardware parameters based on the sampling frequency and the heating power, and sending a cooperative broadcast frame containing the final concentration vector, includes: The configuration value of the sampling frequency is written into the timer register, and the value of the heating power is written into the pulse width modulation control register to complete the reconstruction of the hardware operating parameters; the wireless channel status is detected, and after confirming that the channel is idle, the cooperative broadcast frame is sent out through the radio frequency front end; the acknowledgment signal or listening result in the channel is monitored, and if a neighbor node is not responded or the data packet verification error rate exceeds the threshold, the status of the neighbor node marked in the active neighbor topology list is updated so as to filter valid neighbor data in the next monitoring cycle.

8. A distributed networked MEMS intelligent multi-parameter gas sensing terminal collaborative monitoring system, characterized in that, include: The acquisition module is used to acquire raw response signals and environmental physical quantities, and calculate the initial concentration vector and local confidence interval. The receiving module is used to receive neighbor node data, construct a spatial gradient field based on the initial concentration vector and the local confidence interval, and generate a spatial correction factor. The calculation module is used to calculate the sampling frequency and heating power using the spatial correction factor and the local confidence interval, and correct them to obtain the final concentration vector; The transmitting module is used to reconstruct hardware parameters based on the sampling frequency and the heating power, and transmit a cooperative broadcast frame containing the final concentration vector.

Images