Integrated detection system for in-situ monitoring of moisture content of living combustible

Abstract

An integrated detection system for in-situ monitoring of the moisture content of living combustible materials is provided with a monitoring device, and the monitoring device comprises a microwave transmitter and a microwave receiver which are oppositely arranged; the microwave transmitter and the microwave receiver are connected with a central processing and communication module, and the central processing and communication module is connected with a micro weather station and a remote server; the microwave transmitter and the microwave receiver are used for collecting microwave signals penetrating through living combustible materials; the micro weather station is used for collecting an environment parameter set of a target area; a preprocessing module and a lightweight model prediction module are integrated in the central processing and communication module, and the preprocessing module is used for performing preprocessing operation on microwave signals and an environment parameter set to construct an input feature vector; and the lightweight model prediction module is used for performing moisture content prediction according to the input feature vector and sending a moisture content prediction result to the remote server. The method has the effect that real-time, lossless and high-precision monitoring on the moisture content of the living combustible can be realized.

Description

Technical Field

[0001] This invention relates to the field of microwave sensing technology, and in particular to an integrated detection system for in-situ monitoring of the moisture content of living combustibles. Background Technology

[0002] Live fuel moisture content (LFMC) is a key physiological parameter influencing the occurrence, spread, and intensity of forest fires. Real-time and accurate acquisition of LFMC is crucial for building precise fire risk level early warning models.

[0003] Currently, there are several main methods for obtaining LFMC:

[0004] 1. Traditional laboratory method (drying and weighing method): This is currently recognized as the "gold standard", but its process is destructive sampling and cannot achieve continuous and real-time monitoring; at the same time, it is cumbersome, time-consuming and labor-intensive, and cannot meet the needs of large-scale and high-frequency monitoring.

[0005] 2. Remote sensing estimation method: This method uses satellite or UAV remote sensing data to establish an LFMC estimation model. While this method can achieve large-scale monitoring, it has low spatial resolution, is easily affected by clouds, atmospheric conditions, and canopy obstruction, and struggles to reflect the true state of individual trees or specific areas under the forest canopy. Its temporal resolution is also limited by the satellite revisit cycle.

[0006] 3. Existing Ground Sensor Methods: Some existing point sensors, such as time-domain reflectometer (TDR) probes, can measure dielectric constant to estimate moisture content, but they typically require invasive insertion into the tree trunk, causing damage to the tree, and the measurement points are very limited. Furthermore, there is a lack of standardized, low-power, and easily deployable equipment on the market that highly integrates signal acquisition, data processing, and intelligent analysis functions, specifically designed for trunk penetration measurements.

[0007] Disadvantages of existing technology: Existing LFMC monitoring devices have disadvantages such as high destructiveness, poor timeliness, and low monitoring accuracy. Summary of the Invention

[0008] The present invention provides an integrated detection system for in-situ monitoring of the moisture content of live combustibles, which can realize real-time, non-destructive and high-precision monitoring of the moisture content of live combustibles.

[0009] To achieve the above objectives, the present invention provides an integrated detection system for in-situ monitoring of the moisture content of live combustibles, the key feature of which is: a monitoring device for installation on the live combustibles, the monitoring device including a microwave transmitter and a microwave receiver arranged opposite to each other;

[0010] The microwave transmitter and microwave receiver are connected to a central processing and communication module via wires. The central processing and communication module is connected to a micro weather station via wires. The central processing and communication module is also connected to a remote server via wireless communication.

[0011] The microwave transmitter and microwave receiver are used to collect microwave signals that penetrate living combustibles. And transmit it to the central processing and communication module;

[0012] The micro weather station is used to collect environmental parameter set E of the target area and transmit it to the central processing and communication module;

[0013] The central processing and communication module integrates a preprocessing module and a lightweight model prediction module. The preprocessing module is used to process the microwave signal. The input feature vector is constructed by preprocessing the environmental parameter set E. The lightweight model prediction module is used to predict based on the input feature vector. Perform moisture content prediction and analyze the obtained moisture content prediction results. Perform atmospheric disturbance compensation, and then calculate the compensated moisture content results. Send to the remote server.

[0014] Through the above design, this invention enables microwaves to penetrate tree trunks or shrubs in situ, achieving completely non-destructive, real-time monitoring of living plants. Simultaneously, it works with a central processing and communication module and a miniature weather station to predict the moisture content of living combustibles. The monitored data is then transmitted wirelessly to a remote server, allowing staff to view the data in real time and implement timely protective measures in case of anomalies.

[0015] Preferably, the microwave transmitter is mounted on the left fixed frame, and the microwave receiver is mounted on the right fixed frame. The left fixed frame and the right fixed frame are connected to each other and form a ring structure.

[0016] The left fixed frame is provided with a left mounting bracket. The upper end of the left mounting bracket is fixed to the middle of the left fixed frame by a left mounting bolt. The left mounting bracket is perpendicular to the left fixed frame. The microwave transmitter is mounted on the lower end of the left mounting bracket.

[0017] The right fixed frame is provided with a right mounting bracket. The upper end of the right mounting bracket is fixed to the middle of the right fixed frame by a right mounting bolt. The right mounting bracket is perpendicular to the right fixed frame, and the microwave receiver is mounted on the lower end of the right mounting bracket.

[0018] Preferably, the right mounting bracket and the left mounting bracket are parallel to each other;

[0019] Both ends of the left fixing frame are bent and both face the right fixing frame;

[0020] Both ends of the right fixing frame are bent and both face the left fixing frame;

[0021] The bent portions of the left and right fixed frames are connected by fixing bolts to form the ring structure.

[0022] When the monitoring part is the tree trunk, the bottom of the screw of the fixing bolt is brought into contact with the two sides of the tree trunk of the living combustible material to achieve non-destructive installation and fixation.

[0023] Multiple fixing holes are evenly opened at both ends of the fixing frame. The installation position of the fixing bolts can be adjusted according to the thickness of the live combustible material, so that the device can be adapted to trees of different thicknesses and has stronger practical performance.

[0024] Preferably: the microwave signal It is in complex form, containing amplitude and phase information, and is expressed as follows:

[0025]

[0026] in, It is a microwave frequency domain signal. It is microwave amplitude information (energy absorption loss). It is microwave phase information. It is a microwave frequency. It is the base of the natural logarithm function. It is the imaginary unit. , It is a complex number;

[0027] The central processing and communication module is equipped with an onboard temperature sensor.

[0028] The environmental parameter set This includes temperature data collected from various locations within the target area by the aforementioned micro-weather station. ,humidity and pressure Data, including the onboard temperature collected from the onboard temperature sensor, and the environmental parameter set. Represented as:

[0029] in, For environmental parameter set; Represents humidity. Represents temperature. Represents pressure; subscript Indicates onboard; subscript Indicates the earth's surface; subscript Indicates air; subscript Indicates soil; subscript It refers to the atmosphere.

[0030] Preferably, the preprocessing module is provided with a temperature compensation unit, an IFFT transformation unit, a time-domain gate filtering unit, a feature extraction and reconstruction unit, and a data fusion unit connected in sequence.

[0031] The temperature compensation unit is used for the microwave signal. Temperature drift correction was performed to obtain the corrected microwave signal. ;

[0032] The IFFT transform unit is used to transform the corrected microwave signal through inverse fast Fourier transform. The microwave time-domain signal is obtained by converting the frequency domain data into the time domain data. ;

[0033] The time-domain gate filter unit is used to filter the microwave time-domain signal. Filtering is performed to obtain the filtered signal. ;

[0034] The feature extraction and reconstruction unit is used to extract features from the filtered signal. Mid-time domain features, to obtain time domain feature vectors Then, the filtered signal is transformed using a Fast Fourier Transform. Converted to filtered frequency domain signal and from the filtered frequency domain signal Extracting frequency domain features yields the frequency domain feature vector. ;

[0035] The data fusion unit is used to process the environmental parameter set E and the time-domain feature vector. and frequency domain eigenvectors Normalization is performed, and then the input feature vectors are concatenated to obtain the input feature vectors. .

[0036] The monitoring device continuously scans the microwave energy absorption loss (amplitude information) and phase change (phase information) of live combustibles in a 1-6 GHz wideband. The preprocessing module uses onboard temperature to compensate for temperature drift in the measurement data, forming a feature dataset of wideband measurements. Then, it performs IFFT transformation and uses a time-domain gate filtering algorithm to eliminate reflection or scattering interference caused by differences in the type, shape, density, and texture of live combustibles. It extracts time-domain feature parameters such as group delay, envelope integral, and peak shift. Then, it uses FFT transformation to obtain feature parameters such as absorbed energy loss, phase change, and energy integral. Finally, the processed microwave feature data and the environmental parameter set E are normalized together to construct a multi-dimensional feature vector. .

[0037] Preferably, the temperature compensation unit employs a second-order polynomial temperature compensation function. For the microwave signal Temperature drift correction is performed using the following expression:

[0038]

[0039]

[0040] in, This indicates the real-time onboard temperature. This is the compensation coefficient; The microwave signal before correction; This is the corrected microwave signal.

[0041] This invention fundamentally solves the problem of physical measurement drift caused by changes in ambient temperature by setting a pre-temperature compensation unit in the preprocessing module. The data correction module corrects the core microwave input signal in real time through a pre-calibrated polynomial temperature compensation function to ensure the fidelity and reliability of the input data source.

[0042] This invention uses the onboard temperature obtained by the onboard temperature sensor to correct the temperature drift of the original microwave signal, effectively eliminating the error caused by temperature changes in the equipment itself, and is the first line of defense to ensure long-term measurement stability.

[0043] Preferably, the IFFT transform unit converts the corrected microwave signal using an inverse fast Fourier transform. The microwave time-domain signal is obtained by converting the frequency domain data into the time domain data. The expression is:

[0044]

[0045] in, It is a microwave time-domain signal, which includes signals arriving at different times (such as directly transmitted signals, reflected signals, etc.). This is the inverse fast Fourier transform;

[0046] The time-domain gate filter unit filters the microwave time-domain signal. Filtering is performed to obtain the filtered signal. The expression is:

[0047]

[0048] Where W(t) is the filtering window function, which is 1 only during the time period when the expected direct transmission signal arrives, and 0 at other times.

[0049] The preprocessing module performs IFFT transformation on the microwave time-domain signal to the time domain using an IFFT transformation unit, and applies a time-domain gate filter unit to accurately filter out interference caused by surface reflection and multipath effects, extracting a pure transmission signal.

[0050] Preferably: the time-domain feature vector The features include, but are not limited to, the group delay feature, envelope integral feature, and peak shift quantity feature of microwave signals;

[0051] The feature extraction and reconstruction unit uses a fast Fourier transform to convert the filtered signal... Converted to filtered frequency domain signal The expression is:

[0052]

[0053] The frequency domain feature vector The characteristics of microwave signals include, but are not limited to, amplitude characteristics, phase characteristics, and energy integral characteristics.

[0054] Preferably, the data fusion unit concatenates the normalized features in sequence to obtain an input feature vector, which is represented as follows:

[0055]

[0056] in, This represents the input feature vector. The normalized time-domain feature vector, The normalized frequency domain feature vector, For environmental parameter set; Represents humidity. Represents temperature. Represents pressure; subscript Indicates onboard; subscript Indicates the earth's surface; subscript Indicates air; subscript Indicates soil; subscript The superscript "'" indicates the atmosphere, and the superscript "'" indicates the normalized characteristics.

[0057] This invention adopts a multi-source data fusion architecture. Its input features not only include various climate parameters reflecting the macroscopic environmental state, but also, more importantly, in-situ microwave signals that directly penetrate living combustibles, thus achieving a comprehensive characterization of the internal state and external environment of living combustibles.

[0058] Preferably, the model expression of the lightweight model prediction module is:

[0059]

[0060] in, The water content prediction result of the model at time t; Input the feature vector into the model at time t; This indicates the lightweight model prediction module.

[0061] The lightweight model prediction module provides the moisture content prediction result. Atmospheric disturbance compensation is performed using the following expression:

[0062]

[0063] in, This indicates the predicted moisture content after compensation. It is a compensation function based on environmental parameters. It is a set of environmental parameters.

[0064] The lightweight model prediction module uses a pre-trained measurement model to measure and output the water content. Finally, it uses an environmental parameter fusion algorithm to compensate for atmospheric interference in the measured values. Its measurement accuracy is more precise with single-frequency and multi-frequency microwave measurement methods, and it has high robustness to adapt to both northern and southern climates.

[0065] The beneficial effects of this invention are as follows: By setting up a monitoring device, microwaves can be transmitted in situ through tree trunks or shrubs, achieving completely non-destructive in-situ real-time monitoring of living plants. Simultaneously, in conjunction with a central processing and communication module and a miniature weather station, the moisture content of living combustibles can be predicted, effectively avoiding the influence of clouds, atmospheric conditions, and canopy obstruction on the monitoring data, thus significantly improving the accuracy of the prediction results. Furthermore, this invention transmits the monitored data to a remote server via a wireless network, allowing staff to view the monitoring data in real time and promptly implement protective measures when abnormal monitoring data occurs.

[0066] This invention provides a reliable data source for the model by performing temperature compensation on the monitored microwave signal and combining it with environmental parameters from multiple sources as input data for the lightweight model prediction module, thereby effectively improving the model's prediction accuracy.

[0067] The lightweight model prediction module integrates broadband microwave characteristics and real-time environmental parameters, making it not only accurate in measurement but also capable of self-adjusting to adapt to different regional climate conditions, exhibiting extremely high robustness. Attached Figure Description

[0068] Figure 1 This is a block diagram of the integrated detection system in the embodiment;

[0069] Figure 2This is a top view of the monitoring device applied to trees in the embodiment;

[0070] Figure 3 This is a three-dimensional view of the structure when the monitoring device is applied to trees in the embodiment;

[0071] Figure 4 This is a three-dimensional view of the structure of the monitoring device when it is applied to a shrubland in the embodiment;

[0072] Figure 5 This is a schematic diagram of the monitoring device in the embodiment;

[0073] Figure 6 This is a flowchart illustrating the overall workflow of the integrated detection system in this embodiment. Detailed Implementation

[0074] The present invention will be further described in detail below with reference to the accompanying drawings and specific examples. The following embodiments or drawings are used to illustrate the present invention, but are not intended to limit the scope of the present invention.

[0075] like Figure 1 , Figure 2 As shown, an integrated detection system for in-situ monitoring of the moisture content of live combustibles is provided with a monitoring device 1 for installation on the live combustibles. The monitoring device 1 includes a microwave transmitter 2 and a microwave receiver 3 arranged opposite to each other.

[0076] like Figure 5 As shown, the microwave transmitter 2 is mounted on the left fixed frame 1a, and the microwave receiver 3 is mounted on the right fixed frame 1b. The left fixed frame 1a and the right fixed frame 1b are connected to each other and form a ring structure.

[0077] The left fixed frame 1a is provided with a left mounting frame 1a1. The upper end of the left mounting frame 1a1 is fixed to the middle of the left fixed frame 1a by a left mounting bolt. The left mounting frame 1a1 is perpendicular to the left fixed frame 1a. The microwave transmitter 2 is mounted on the lower end of the left mounting frame 1a1.

[0078] The right fixed frame 1b is provided with a right mounting bracket 1b1. The upper end of the right mounting bracket 1b1 is fixed to the middle of the right fixed frame 1b by a right mounting bolt. The right mounting bracket 1b1 is perpendicular to the right fixed frame 1b. The microwave receiver 3 is installed at the lower end of the right mounting bracket 1b1.

[0079] The right mounting bracket 1b1 is parallel to the left mounting bracket 1a1;

[0080] Both ends of the left fixing frame 1a are bent and both face the right fixing frame 1b;

[0081] Both ends of the right fixing frame 1b are bent and both face the left fixing frame 1a;

[0082] The bent portions of the left fixing frame 1a and the right fixing frame 1b are connected by fixing bolts to form the ring structure.

[0083] The microwave transmitter 2 and microwave receiver 3 are connected to a central processing and communication module 6 via wires. The central processing and communication module 6 is connected to a micro weather station via wires. The central processing and communication module 6 is also connected to a remote server via wireless communication.

[0084] The microwave transmitter 2 and microwave receiver 3 are used to collect microwave signals that penetrate living combustibles. And transmit it to the central processing and communication module 6;

[0085] The micro weather station is used to collect environmental parameter set E of the target area and transmit it to the central processing and communication module 6;

[0086] The central processing and communication module 6 integrates a preprocessing module and a lightweight model prediction module. The preprocessing module is used to process the microwave signal. The input feature vector is constructed by preprocessing the environmental parameter set E. The lightweight model prediction module is used to predict based on the input feature vector. Perform moisture content prediction and analyze the obtained moisture content prediction results. Perform atmospheric disturbance compensation, and then calculate the compensated moisture content results. Send to the remote server.

[0087] The microwave signal It is in complex form, containing amplitude and phase information, and is expressed as follows:

[0088]

[0089] in, It is a microwave frequency domain signal. It is microwave amplitude information (energy absorption loss). It is microwave phase information. It is a microwave frequency. It is the base of the natural logarithm function. It is the imaginary unit. , It is a complex number;

[0090] The central processing and communication module 6 is equipped with an onboard temperature sensor.

[0091] The environmental parameter set This includes temperature data collected from various locations within the target area by the aforementioned micro-weather station. ,humidity and pressure Data, including the onboard temperature collected from the onboard temperature sensor, and the environmental parameter set. Represented as:

[0092] in, For environmental parameter set; Represents humidity. Represents temperature. Represents pressure; subscript Indicates onboard; subscript Indicates the earth's surface; subscript Indicates air; subscript Indicates soil; subscript It refers to the atmosphere.

[0093] The preprocessing module is provided with a temperature compensation unit, an IFFT transformation unit, a time-domain gate filtering unit, a feature extraction and reconstruction unit, and a data fusion unit connected in sequence.

[0094] The temperature compensation unit is used for the microwave signal. Temperature drift correction was performed to obtain the corrected microwave signal. ;

[0095] The IFFT transform unit is used to transform the corrected microwave signal through inverse fast Fourier transform. The microwave time-domain signal is obtained by converting the frequency domain data into the time domain data. ;

[0096] The time-domain gate filter unit is used to filter the microwave time-domain signal. Filtering is performed to obtain the filtered signal. ;

[0097] The feature extraction and reconstruction unit is used to extract features from the filtered signal. Mid-time domain features, to obtain time domain feature vectors Then, the filtered signal is transformed using a Fast Fourier Transform. Converted to filtered frequency domain signal and from the filtered frequency domain signal Extracting frequency domain features yields the frequency domain feature vector. ;

[0098] The data fusion unit is used to process the environmental parameter set E and the time-domain feature vector. and frequency domain eigenvectors Normalization is performed, and then the input feature vectors are concatenated to obtain the input feature vectors. .

[0099] Furthermore, the temperature compensation unit employs a second-order polynomial temperature compensation function. For the microwave signal Temperature drift correction is performed using the following expression:

[0100]

[0101]

[0102] in, This indicates the real-time onboard temperature. This is the compensation coefficient; The microwave signal before correction; This is the corrected microwave signal.

[0103] Furthermore, the IFFT transform unit converts the corrected microwave signal using an inverse fast Fourier transform. The microwave time-domain signal is obtained by converting the frequency domain data into the time domain data. The expression is:

[0104]

[0105] in, It is a microwave time-domain signal, which includes signals arriving at different times (such as directly transmitted signals, reflected signals, etc.). This is the inverse fast Fourier transform;

[0106] The time-domain gate filter unit filters the microwave time-domain signal. Filtering is performed to obtain the filtered signal. The expression is:

[0107]

[0108] Where W(t) is the filtering window function, which is 1 only during the time period when the expected direct transmission signal arrives, and 0 at other times.

[0109] Furthermore, the time-domain feature vector The features include, but are not limited to, the group delay feature, envelope integral feature, and peak shift quantity feature of microwave signals;

[0110] The feature extraction and reconstruction unit uses a fast Fourier transform to convert the filtered signal... Converted to filtered frequency domain signal The expression is:

[0111]

[0112] The frequency domain feature vector The characteristics of microwave signals include, but are not limited to, amplitude characteristics, phase characteristics, and energy integral characteristics.

[0113] The data fusion unit concatenates the normalized features in sequence to obtain an input feature vector, which is represented as follows:

[0114]

[0115] in, This represents the input feature vector. The normalized time-domain feature vector, The normalized frequency domain feature vector, For environmental parameter set; Represents humidity. Represents temperature. Represents pressure; subscript Indicates onboard; subscript Indicates the earth's surface; subscript Indicates air; subscript Indicates soil; subscript The superscript "'" indicates the atmosphere, and the superscript "'" indicates the normalized characteristics.

[0116] Furthermore, the model expression of the lightweight model prediction module is:

[0117]

[0118] in, The water content prediction result of the model at time t; Input the feature vector into the model at time t; This indicates the lightweight model prediction module.

[0119] The lightweight model prediction module provides the moisture content prediction result. Atmospheric disturbance compensation is performed using the following expression:

[0120]

[0121] in, This indicates the predicted moisture content after compensation. It is a compensation function based on environmental parameters. It is a set of environmental parameters.

[0122] like Figure 2 , Figure 3 As shown, the monitoring device 1 is installed on the trunk 7 of the living combustible material. The bottom of the screws of the left and right fixing bolts abut against the two side surfaces of the trunk 7 of the living combustible material, so as to achieve non-destructive installation and fixation.

[0123] The left and right mounting brackets 1a and 1b form a ring-shaped fixing and support assembly that encircles the tree trunk, ensuring the microwave transmitter and receiver are positioned at opposite ends of the trunk's diameter. This open, encircling structure makes installation extremely convenient and adaptable to trees of varying thicknesses. The fixing and support assembly acts as an adjustable frame encircling the tree trunk 7. The microwave transmitter 2 and microwave receiver 3 are directly mounted at opposite ends of this assembly, ensuring they penetrate the trunk. The central processing and communication module 6 is suspended outside the assembly and connected to the microwave transmitter 2 and microwave receiver 3 via wires. This open frame structure not only ensures precise alignment of the microwave transceiver modules but also allows the entire device to adapt to trees of varying thicknesses.

[0124] like Figure 4 As shown, the monitoring device 1 can also be applied to shrubs 8. The open structure of this device allows it to be easily placed in the core area of ​​the shrub, enabling the microwave path to penetrate the densest foliage to measure its overall average moisture content. The central processing and communication module 6 is the "brain" of the device, internally encapsulating a high-performance embedded circuit board, algorithm model, and communication unit. The microwave transmitter 2 and microwave receiver 3 are responsible for transmitting and receiving broadband microwaves under the control of the central processing and communication module 6.

[0125] The fixing and support structure of the monitoring device 1 is made of stainless steel.

[0126] Furthermore, the workflow of the entire system is as follows: Figure 6 As shown, the specific operation process is as follows:

[0127] (1) Deployment and connection: Install the monitoring device 1 on the target live combustible material and connect it to the micro weather station via cable.

[0128] (2) Wake-up and Acquisition: After the system is woken up from hibernation mode, the central processing and communication module 6 immediately performs two tasks: ① Controls the microwave transceiver module to acquire the original microwave signal after penetrating the live combustible material. ; ② Receive environmental data E from the weather station via the interface.

[0129] (3) Internal processing: The central processing and communication module 6 of the module executes the internal data processing flow, namely the data preprocessing and moisture content prediction compensation flow.

[0130] (4) Reporting and Sleep Mode: After processing is completed, the central processing and communication module 6 determines whether data needs to be reported. If reporting is required, the result is sent to the remote server through the internal wireless communication unit; otherwise, it directly enters the low-power sleep mode and waits for the next measurement cycle.

[0131] The above description is merely a preferred embodiment of the present invention and is not intended to limit the invention. Various modifications and variations can be made to the present invention by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. An integrated detection system for in-situ monitoring of moisture content in live combustibles, characterized in that: A monitoring device (1) for installation on a live combustible material is provided, the monitoring device (1) including a microwave transmitter (2) and a microwave receiver (3) arranged opposite to each other. The microwave transmitter (2) and microwave receiver (3) are connected to a central processing and communication module (6) via wires. The central processing and communication module (6) is connected to a micro weather station via wires. The central processing and communication module (6) is also connected to a remote server via wireless communication. The microwave transmitter (2) and microwave receiver (3) are used to collect microwave signals that penetrate living combustibles. , and pass it to the central processing and communication module (6); The micro weather station is used to collect environmental parameter set E of the target area and transmit it to the central processing and communication module (6). The central processing and communication module (6) integrates a preprocessing module and a lightweight model prediction module. The preprocessing module is used to process the microwave signal. Preprocessing operations are performed on the environmental parameter set E to construct the input feature vector. The lightweight model prediction module is used to predict based on the input feature vector. Perform moisture content prediction and analyze the obtained moisture content prediction results. Perform atmospheric disturbance compensation, and then calculate the compensated moisture content results. Send to the remote server.

2. The integrated detection system for in-situ monitoring of moisture content in live combustibles according to claim 1, characterized in that: The microwave transmitter (2) is mounted on the left fixed frame (1a), and the microwave receiver (3) is mounted on the right fixed frame (1b). The left fixed frame (1a) and the right fixed frame (1b) are connected to each other and form a ring structure. The left fixed frame (1a) is provided with a left mounting frame (1a1). The upper end of the left mounting frame (1a1) is fixed to the middle of the left fixed frame (1a) by a left mounting bolt. The left mounting frame (1a1) is perpendicular to the left fixed frame (1a). The microwave transmitter (2) is installed at the lower end of the left mounting frame (1a1). The right fixed frame (1b) is provided with a right mounting bracket (1b1). The upper end of the right mounting bracket (1b1) is fixed to the middle of the right fixed frame (1b) by a right mounting bolt. The right mounting bracket (1b1) is perpendicular to the right fixed frame (1b). The microwave receiver (3) is installed at the lower end of the right mounting bracket (1b1).

3. The integrated detection system for in-situ monitoring of moisture content in live combustibles according to claim 2, characterized in that: The right mounting bracket (1b1) is parallel to the left mounting bracket (1a1); The two ends of the left fixing frame (1a) are bent and both face the right fixing frame (1b). Both ends of the right fixing frame (1b) are bent and both face the left fixing frame (1a). The bent portions of the left fixing frame (1a) and the right fixing frame (1b) are connected by fixing bolts to form the ring structure.

4. The integrated detection system for in-situ monitoring of moisture content in live combustibles according to claim 1, characterized in that: The microwave signal It is in complex form, containing amplitude and phase information, and is expressed as follows: ； in, It is a microwave frequency domain signal. It is microwave amplitude information. It is microwave phase information. It is a microwave frequency. It is the base of the natural logarithm function. It is the imaginary unit; The central processing and communication module (6) is equipped with an onboard temperature sensor. The environmental parameter set This includes temperature data collected from various locations within the target area by the aforementioned micro-weather station. ,humidity and pressure Data, including the onboard temperature collected from the onboard temperature sensor, and the environmental parameter set. Represented as: ; in, For environmental parameter set; Represents humidity. Represents temperature. Represents pressure; subscript Indicates onboard; subscript Indicates the earth's surface; subscript Indicates air; subscript Indicates soil; subscript It refers to the atmosphere.

5. An integrated detection system for in-situ monitoring of moisture content in live combustibles according to claim 1, characterized in that: The preprocessing module is provided with a temperature compensation unit, an IFFT transformation unit, a time-domain gate filtering unit, a feature extraction and reconstruction unit, and a data fusion unit connected in sequence. The temperature compensation unit is used for the microwave signal. Temperature drift correction was performed to obtain the corrected microwave signal. ; The IFFT transform unit is used to transform the corrected microwave signal through inverse fast Fourier transform. The microwave time-domain signal is obtained by converting the frequency domain data into the time domain data. ; The time-domain gate filter unit is used to filter the microwave time-domain signal. Filtering is performed to obtain the filtered signal. ; The feature extraction and reconstruction unit is used to extract features from the filtered signal. Mid-time domain features, to obtain time domain feature vectors Then, the filtered signal is transformed using a Fast Fourier Transform. Converted to filtered frequency domain signal and from the filtered frequency domain signal Extracting frequency domain features yields the frequency domain feature vector. ; The data fusion unit is used to process the environmental parameter set E and the time-domain feature vector. and frequency domain eigenvectors Normalization is performed, and then the input feature vectors are concatenated to obtain the input feature vectors. .

6. An integrated detection system for in-situ monitoring of moisture content in live combustibles according to claim 5, characterized in that: The temperature compensation unit employs a second-order polynomial temperature compensation function. For the microwave signal Temperature drift correction is performed using the following expression: ； ； in, This indicates the real-time onboard temperature. The compensation coefficient; The microwave signal before correction; This is the corrected microwave signal.

7. An integrated detection system for in-situ monitoring of moisture content in live combustibles according to claim 5, characterized in that: The IFFT transform unit converts the corrected microwave signal using an inverse fast Fourier transform. The microwave time-domain signal is obtained by converting the frequency domain data into the time domain data. The expression is: ； in, It is a microwave time-domain signal; This is the inverse fast Fourier transform; The time-domain gate filter unit filters the microwave time-domain signal. Filtering is performed to obtain the filtered signal. The expression is: ； Where W(t) is the filtering window function.

8. An integrated detection system for in-situ monitoring of moisture content in live combustibles according to claim 5, characterized in that: The time-domain feature vector The features include, but are not limited to, the group delay feature, envelope integral feature, and peak shift quantity feature of microwave signals; The feature extraction and reconstruction unit uses a fast Fourier transform to convert the filtered signal... Converted to filtered frequency domain signal The expression is: ； The frequency domain feature vector The characteristics of microwave signals include, but are not limited to, amplitude characteristics, phase characteristics, and energy integral characteristics.

9. An integrated detection system for in-situ monitoring of moisture content in live combustibles according to claim 5, characterized in that: The data fusion unit concatenates the normalized features in sequence to obtain an input feature vector, which is represented as follows: ； in, This represents the input feature vector. The normalized time-domain feature vector, The normalized frequency domain feature vector, For environmental parameter set; Represents humidity. Represents temperature. Represents pressure; subscript Indicates onboard; subscript Indicates the earth's surface; subscript Indicates air; subscript Indicates soil; subscript The superscript "' " indicates the atmosphere, and the superscript "' " indicates the normalized characteristics.

10. An integrated detection system for in-situ monitoring of moisture content in live combustibles according to claim 1, characterized in that: The model expression for the lightweight model prediction module is: ； in, The water content prediction result of the model at time t; Input the feature vector into the model at time t; This indicates a lightweight model prediction module; The lightweight model prediction module provides the moisture content prediction result. Atmospheric disturbance compensation is performed using the following expression: ； in, This indicates the predicted moisture content after compensation. It is a compensation function based on environmental parameters. It is a set of environmental parameters.

Images