Solar powered underground jammer waveform detection method and system

Abstract

The present application relates to the technical field of electrical signal detection, and specifically discloses a solar-powered underground interference source waveform recording method and system. The present application aims at the monitoring demand of long-distance underground metal pipeline electrochemical corrosion protection, and solves the problem of low power consumption and high sensitivity in capturing transient interference. The present application constructs a dual-mode front-end sensing circuit containing an analog comparator and a high-precision analog-to-digital converter, adopts a two-stage triggering mechanism of hardware change rate preliminary screening and software feature snapshot recheck, effectively reduces the false triggering rate and system power consumption; when the system detects that the voltage change rate is out of limit, the main control unit is woken up for data acquisition and preprocessing, the waveform feature vector is extracted and differentially reported, the key alarm information is preferentially transmitted, and the wireless communication bandwidth occupation is reduced; the present application prolongs the self-operation cycle of the solar-powered equipment, improves the data transmission efficiency, and is suitable for the deployment of large-scale monitoring networks in remote areas.

Description

Technical Field

[0001] This invention belongs to the field of electrical signal detection technology, and relates to a method and system for detecting waveforms of underground interference sources powered by solar energy. Background Technology

[0002] Currently, monitoring the electrochemical corrosion protection status of long-distance underground metal pipelines is an important task. Effectively capturing and analyzing abnormal potential fluctuations caused by stray currents from external electromagnetic interference sources, such as high-voltage direct current transmission lines and electrified railways, is a crucial step in evaluating the effectiveness of the protection system. These interference signals are typically transient, sporadic, and have complex waveforms; accurate recording of these signals is fundamental for subsequent interference source identification and adjustment of protection strategies. However, since pipelines are often laid in remote areas with inadequate power and communication infrastructure, monitoring equipment must operate stably for extended periods without human intervention, relying on limited self-supplied energy. This poses a dual challenge to the power consumption and event capture capabilities of the monitoring methods.

[0003] To address these challenges, several technical solutions exist in the industry. One common approach is to employ a periodic wake-up and sampling mechanism. The monitoring terminal remains in deep sleep most of the time, only waking up the main processor and high-precision acquisition unit at preset time intervals, such as every few minutes or hours, to perform a brief data measurement and record, thereby reducing average power consumption. Another approach is to set a fixed voltage amplitude threshold. When the absolute value of the monitored pipe potential signal exceeds this preset threshold, the system is triggered to wake up and perform a complete waveform recording. Compared to periodic sampling, this approach theoretically possesses event-driven characteristics, aiming to reduce the probability of missing interference events.

[0004] However, the aforementioned traditional methods have certain limitations in practical applications. Periodic wake-up and sampling mechanisms have a high probability of missing transient interference events with durations much shorter than the sampling interval, potentially leading to the loss of crucial interference event information. Furthermore, triggering schemes based on fixed voltage amplitude thresholds need improvement in adaptability to slow baseline drift in pipeline protection potentials. Such baseline drifts may be caused by seasonal changes in environmental factors such as soil moisture and temperature, and are not actual interference events, but they can still cause voltage amplitudes to exceed the fixed threshold, resulting in a large number of invalid wake-ups and data records. This not only consumes valuable electrical energy and storage space but also increases the burden on backend data analysis. In addition, this scheme may be insensitive to interference signals with small amplitudes but extremely high voltage change rates, which can also affect the pipeline system.

[0005] Based on the above problems, the present invention aims to solve the problem of how to reliably capture occasional transient interference events and reduce the probability of false triggering caused by environmental factors under strict low power consumption constraints. Summary of the Invention

[0006] In view of this, in order to solve the problems mentioned in the background technology, a method and system for detecting waveforms of underground interference sources powered by solar energy is proposed.

[0007] The objective of this invention can be achieved through the following technical solutions: The first aspect of this invention provides a method for detecting waveforms of underground interference sources powered by solar energy, comprising: S1, constructing an adaptive sensing environment based on solar power, initializing a dual-mode front-end sensing circuit, setting an initial hardware threshold corresponding to the upper limit of the voltage change rate of the pipeline protection potential signal, marking it as a dynamic trigger reference, and causing the main control unit to enter a deep sleep state that retains only the analog comparator.

[0008] S2. Use an analog comparator to monitor the pipeline potential signal in real time. When the voltage change rate of the input signal exceeds the dynamic trigger reference, the output level is flipped to generate a hardware wake-up interrupt signal indicating the occurrence of potential interference.

[0009] S3. In response to the hardware wake-up interrupt signal, activate the main control unit to collect signal snapshot data within a preset short time window, perform feature calculations on the signal snapshot data and compare it with the noise model. If the comparison result is determined to be effective interference, generate a high-precision waveform recording command.

[0010] S4. Execute high-precision waveform recording instructions to control the dual-mode front-end sensing circuit to perform full-bandwidth data acquisition, combine the acquired discrete voltage sequences and write them into the local memory to generate an original waveform data frame containing complete interference event information.

[0011] S5. Perform edge calculation on the original waveform data frame, extract key physical indicators in the time and frequency domains, and encode and compress them to generate waveform feature vectors for low-bandwidth transmission.

[0012] S6. Based on the waveform feature vector analysis of the interference event level, dynamically schedule the wireless transmission strategy, and send the waveform feature vector or the original waveform data frame to complete the differentiated reporting of monitoring data.

[0013] The second aspect of the present invention provides a solar-powered underground interference source waveform detection system, comprising: a system initialization and management module, which constructs an adaptive sensing environment based on solar power, initializes a dual-mode front-end sensing circuit, sets an initial hardware threshold corresponding to the upper limit of the voltage change rate of the pipeline protection potential signal, marks it as a dynamic trigger reference, and puts the main control unit into a deep sleep state that retains only the analog comparator operation.

[0014] The hardware wake-up interrupt signal generation module uses an analog comparator to monitor the pipeline potential signal in real time. When the voltage change rate of the input signal exceeds the dynamic trigger reference, the output level is flipped to generate a hardware wake-up interrupt signal indicating the occurrence of potential interference.

[0015] The event determination and control module, in response to the hardware wake-up interrupt signal, activates the main control unit to collect signal snapshot data within a preset short time window, performs feature calculations and noise model comparisons on the signal snapshot data, and if the comparison result determines that it is effective interference, it generates a high-precision waveform recording command.

[0016] The original waveform data frame generation module executes high-precision waveform recording instructions, controls the dual-mode front-end sensing circuit to perform full-bandwidth data acquisition, combines the acquired discrete voltage sequences and writes them into the local memory to generate an original waveform data frame containing complete interference event information.

[0017] The waveform feature vector generation module performs edge calculations on the original waveform data frame, extracts key physical indicators in the time and frequency domains, and encodes and compresses them to generate waveform feature vectors for low-bandwidth transmission.

[0018] The differentiated data reporting module analyzes the interference event level based on the waveform feature vector, dynamically schedules the wireless transmission strategy, and sends the waveform feature vector or the original waveform data frame to complete the differentiated reporting of monitoring data.

[0019] Compared with the prior art, the embodiments of the present invention have at least the following advantages or beneficial effects: (1) The present invention constructs a dual-mode front-end sensing circuit containing an analog comparator and a high-precision analog-to-digital converter, and executes a control strategy combining hardware wake-up and software prediction by the main control unit. This enables the system to maintain high sensitivity to transient changes in pipeline potential while achieving microampere-level static power consumption. The system relies on an ultra-low power analog comparator to continuously monitor the signal change rate most of the time, while the main power-consuming units such as the main control unit and the high-precision analog-to-digital converter are in a deep sleep state. The main control unit is only woken up by a hardware interrupt when the signal change rate exceeds the dynamically set hardware benchmark. This event-driven working mode avoids the energy waste caused by periodic sampling. Compared with the scheme of continuous high-frequency digital sampling, it reduces the average operating current of the system, thereby extending the autonomous operation cycle of the equipment under solar power conditions.

[0020] (2) This invention employs a two-stage triggering and discrimination mechanism: initial screening using hardware change rate and verification using software feature snapshots. This improves the accuracy of interference event identification. The hardware triggering stage, composed of analog comparators, directly responds to the voltage change rate of the signal, effectively filtering out false triggers caused by slowly changing potential baseline drift, thus initially improving the effectiveness of the triggering. After hardware wake-up, the main control unit does not immediately initiate high-power full waveform recording, but instead performs a short-time signal snapshot acquisition and rapid feature calculation. By analyzing the effective value and zero-crossing rate of the signal snapshot and comparing them with a preset noise model, a secondary software judgment is made on the effectiveness of the triggering event, thereby filtering out false triggers caused by invalid noise sources such as high-frequency spikes. This mechanism reduces the full data acquisition, storage, and transmission operations performed due to invalid events, improving the utilization rate of system resources.

[0021] (3) This invention optimizes data transmission efficiency under low-bandwidth communication conditions by performing edge computing on the terminal device to extract features from the collected raw waveform data and implementing differentiated data reporting strategies based on the feature vectors. The device performs time-domain and frequency-domain analysis of the raw waveform data frames locally, extracts key physical parameters such as waveform peak value and main harmonic frequency, and combines them into a waveform feature vector with a small data volume. The system performs local hazard level assessment of interference events based on this waveform feature vector. For general events, only the feature vector is reported, while for events judged to be dangerous, the feature vector is reported as an alarm first, and then the asynchronous transmission of the raw waveform data frames is initiated. This method reduces the occupation of wireless communication bandwidth, ensures low-latency transmission of key alarm information, and makes it possible to deploy large-scale monitoring networks in areas with limited network conditions. Attached Figure Description

[0022] To more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0023] Figure 1 This is a schematic diagram of the method steps of the present invention.

[0024] Figure 2 This is a schematic diagram of the system structure connection of the present invention. Detailed Implementation

[0025] 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 embodiments of the present invention, and not all embodiments. 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.

[0026] Please see Figure 1 The first aspect of the present invention provides a method for detecting waveforms of underground interference sources powered by solar energy, comprising: S1, constructing an adaptive sensing environment based on solar power, initializing a dual-mode front-end sensing circuit, setting an initial hardware threshold corresponding to the upper limit of the voltage change rate of the pipeline protection potential signal, marking it as a dynamic trigger reference, and putting the main control unit into a deep sleep state that retains only the analog comparator operation.

[0027] In a specific embodiment of the present invention, initializing the dual-mode front-end sensing circuit includes: charging the energy storage module using solar cell units in conjunction with a maximum power point tracking algorithm.

[0028] A unidirectional switching circuit provides a microampere-level operating current to the dual-mode front-end sensing circuit.

[0029] The topology of the dual-mode front-end sensing circuit is configured such that the analog comparator and the high-precision analog-to-digital converter are connected in parallel to the same sensor signal input port through an electronic switch, and the signal path is preferentially connected to the analog comparator during the initialization phase.

[0030] In a specific embodiment of the present invention, an initial hardware threshold corresponding to the upper limit of the voltage change rate of the pipeline protection potential signal is set and marked as a dynamic triggering reference, including: obtaining the maximum voltage change rate of the pipeline protection potential signal under normal operating conditions.

[0031] The upper limit of the voltage change rate is calculated by multiplying the maximum voltage change rate by a preset safety factor.

[0032] An analog reference voltage corresponding to the upper limit of the voltage change rate is generated by a digital-to-analog converter or a programmable reference source, applied to the negative phase input of the analog comparator, and the value of the analog reference voltage is stored in non-volatile memory as a dynamic trigger reference.

[0033] Specifically, the main control unit first activates its energy management unit. The system calls the maximum power point tracking algorithm to optimize the output of the solar cell unit, manages the charging of the energy storage module through a DC-DC converter circuit with unidirectional conduction characteristics, and converts a stable voltage into a voltage level that is within a certain range. A microamp-range operating current is continuously supplied to the power supply pin of the dual-mode front-end sensing circuit. The dual-mode front-end sensing circuit consists of a low-power analog comparator and a high-precision analog-to-digital converter connected in parallel to the same sensor signal input port via an electronic switch. During initialization, the electronic switch connects the signal path to the positive input of the analog comparator. Then, the main control unit applies an initial reference voltage to the negative input of the analog comparator through its digital-to-analog converter port or a programmable reference voltage source. This reference voltage corresponds to the upper limit of the voltage change rate of the pipeline protection potential signal to be monitored. After completing the setup, the main control unit will set the initial reference voltage. The numerical value and the upper limit of the voltage change rate it represents The mapping relationship is stored in a specific address segment of non-volatile memory, and this address segment is marked as the dynamic trigger reference. Finally, the main control unit switches its core clock to a low-speed mode, shuts down the power domain of the high-precision analog-to-digital converter and related peripherals, and enters a deep sleep state by executing a specific sleep instruction. At this time, only the analog comparator and its necessary pull-up resistors and reference source circuit are retained. Powered by microampere-level current, maintaining real-time response mode.

[0034] Solar cells are semiconductor devices that convert light energy into electrical energy. Their output voltage, after being adjusted by a DC-DC converter, charges the energy storage module. The energy storage module is typically a rechargeable lithium battery or a supercapacitor, used to power the system in the absence of sunlight. A dual-mode front-end sensing circuit is a signal conditioning and acquisition circuit whose parallel design allows for signal monitoring using only an analog comparator in low-power mode, switching to an analog-to-digital converter for waveform digitization in high-precision mode. An analog comparator is a circuit with only two discrete output levels, used to quickly compare the magnitudes of two analog input voltages. A high-precision analog-to-digital converter is a circuit that converts continuous analog signals into discrete digital signals; its accuracy is determined by both resolution and sampling rate. The initial hardware threshold is the initial reference voltage. The initial hardware threshold is set as follows: the system pre-stores a standard steady-state waveform of a pipeline protection potential, calculates the maximum voltage change rate of this waveform under normal operating conditions, and multiplies it by a safety factor. ,in The value ranges from 1.2 to 2.0. This safety factor is used to avoid triggering by normal environmental noise during the initialization phase. Assuming the application scenario is a suburban pipeline with stable industrial interference, the initial threshold can be set to a relatively conservative, higher value, such as the upper limit of the corresponding voltage change rate. It can be set to 50 millivolts per second. The dynamic trigger reference is a data structure stored in non-volatile memory, containing the initial reference voltage. The numerical value and its physical meaning. Deep sleep state refers to a low-power mode of the main control unit. In this mode, the central processing unit core clock stops, most memory and peripheral functions are turned off, and it can only be woken up by a specific external interrupt signal. Microampere-level operating current. This refers to the quiescent current required to maintain the operation of the analog comparator and necessary reference circuitry, with a typical value ranging from 10 microamps to 100 microamps.

[0035] For example, assume the open-circuit voltage of the solar cell unit under standard illumination conditions is 5V, and the energy storage module is a 3.7V, 1000mAh lithium polymer battery. After system startup, the MPPT algorithm in the energy management unit starts working, driving a buck DC-DC converter to extract power from the solar cell at maximum efficiency and charge the energy storage module with a constant current of 500mA. Simultaneously, the system generates a stable 3.3V voltage from the output of the energy storage module through a low-dropout linear regulator, and provides this voltage to the dual-mode front-end sensing circuit via a constant current source with an accuracy of one percent. The continuous operating current is 25 microamps. In the dual-mode front-end sensing circuit, the analog comparator is a TI TLV7011, and the high-precision analog-to-digital converter is an ADI AD7091R. During initialization, the main control unit generates... A reference voltage of 1.65V is applied to the negative input of the analog comparator. This 1.65V reference voltage corresponds to the upper limit of the preset pipeline potential signal voltage change rate in the system. The preset value of 50 millivolts per second is based on statistical analysis of historical, interference-free data, taking 1.5 times the maximum volatility. The main control unit then writes the value 1.65 and its mapping relationship into the flash memory region starting at address 0x0800F000, and marks this region as the dynamic trigger reference storage area. Finally, the main control unit executes a deep sleep command, shutting down the power supply to the main clock and AD7091R, leaving only the TLV7011 comparator and its external 1.65V reference source circuit powered by 25 microamps, and the system immediately enters standby mode.

[0036] S2. Use an analog comparator to monitor the pipeline potential signal in real time. When the voltage change rate of the input signal exceeds the dynamic trigger reference, the output level is flipped to generate a hardware wake-up interrupt signal indicating the occurrence of potential interference.

[0037] In a specific embodiment of the present invention, generating a hardware wake-up interrupt signal indicating the occurrence of potential interference includes: using a passive differentiating circuit composed of resistors and capacitors to convert the pipeline protection potential signal into an analog voltage proportional to its rate of change.

[0038] The analog voltage is input to the positive input terminal of the analog comparator and compared in real time with the dynamic trigger reference at the negative input terminal.

[0039] When the analog voltage is higher than the dynamic trigger reference, the analog comparator output pin generates a level transition, and the transition edge is directly transmitted to the external interrupt pin of the main control unit as a hardware wake-up interrupt signal.

[0040] Specifically, while the main control unit is in deep sleep mode, the analog comparator in the dual-mode front-end sensing circuit remains continuously active. The pipeline protection potential signal passes through a passive high-pass filter composed of resistors and capacitors. This passive high-pass filter is functionally equivalent to a differentiator, outputting an analog voltage proportional to the rate of change of the pipeline protection potential signal in real time. This analog voltage is continuously input to the positive input of the analog comparator, while the negative input is stably connected to a voltage value set by the dynamic trigger reference. The analog comparator, through its internal differential amplifier circuit, compares the voltage levels of the positive and negative inputs in real time with a microsecond-level response speed. Under normal interference conditions, the rate of change of the pipeline protection potential signal remains within a preset range, ensuring that the voltage at the differentiator output remains consistently lower than the dynamic trigger reference voltage. At this time, the output pin of the analog comparator remains at a stable logic low level. When a strong interference source enters, the voltage of the pipeline protection potential signal undergoes a drastic jump, with its rate of change instantaneously exceeding the upper limit corresponding to the dynamic trigger reference, causing the voltage at the differentiator output to instantaneously exceed the dynamic trigger reference voltage. At this moment, the internal circuitry of the analog comparator flips, causing the output pin to immediately switch from logic low to logic high. This voltage transition edge is defined as the hardware wake-up interrupt signal, which is transmitted directly to a pre-configured external interrupt pin of the main control unit via physical wiring, thereby triggering the main control unit's hardware interrupt response mechanism.

[0041] Formula Explanation: The relationship between the voltage at the output of the differentiator and the rate of change of the pipeline protection potential signal voltage can be represented by the following formula: ,in, This is the real-time voltage output from the differentiator to the non-inverting input of the analog comparator. This is the instantaneous voltage of the pipeline protection potential signal. For time. The differential coefficients, determined by the resistance and capacitance parameters of the differentiating circuit, are in the dimension of time. This formula indicates the voltage that the analog comparator actually compares. Rate of change relative to the original signal They exhibit a linear proportional relationship.

[0042] By verifying the dimensional consistency of the above formula, the left side of the formula... The unit is volts (V). On the right side of the formula, The unit is volt (V). The unit is seconds (s), therefore The unit is volts per second (V / s). To ensure dimensional consistency on both sides of the equation, the differential coefficients... The unit must be seconds (s). In circuit implementation, this coefficient is equivalent to the RC time constant, and its unit is indeed seconds, thus the formula is dimensionally consistent.

[0043] Transient changes refer to non-periodic, sharp fluctuations in the pipeline protection potential signal caused by external high-current sources, typically lasting from milliseconds to seconds, distinct from background noise or slow daily variations. Hardware-level screening refers to the process of using a purely hardware circuit—an analog comparator—to make preliminary judgments about signal characteristics. This process requires no software intervention, can be completed while the main control unit is in sleep mode, and has extremely low power consumption and no response delay. The transition edge of the output level specifically refers to the effective triggering form of the hardware wake-up interrupt signal. It can be a rising edge from logic low to logic high or a falling edge from logic high to logic low, determined by the configuration of the external interrupt controller of the main control unit. The hardware wake-up interrupt signal is a physical electrical signal that can forcibly wake the main control unit from low-power modes such as deep sleep. The external interrupt pin of the main control unit is a physical pin on the microcontroller chip specifically designed to receive such external asynchronous event signals.

[0044] For example, continuing from the previous embodiment, the reference voltage set for the dynamic triggering reference is 1.65V, corresponding to a maximum voltage change rate of 50 millivolts per second. The differentiating circuit at the system front end consists of a 1-megohm resistor and a 33-microfarad capacitor connected in series, with a differentiating factor... That is, the RC product, which is 33 seconds. At a certain moment, the pipeline system is subjected to a continuous, stable power frequency disturbance with a rate of change of 20 millivolts per second. At this time, the output voltage of the differentiating circuit is... The calculation is 33 seconds multiplied by 0.02 volts per second, equaling 0.66V. The analog comparator TLV7011 compares the 0.66V at the positive input with the 1.65V at the negative input. Since 0.66V is less than 1.65V, its output remains at a logic low level, i.e., 0V, and the main control unit remains in sleep mode. Subsequently, a transient ground fault occurs in a high-voltage direct current transmission system, causing the pipeline potential to rise sharply by 8 millivolts within 0.1 seconds, with a transient voltage change rate of 80 millivolts per second. At the instant the fault occurs, the output voltage of the differentiating circuit... The calculation is 33 seconds multiplied by 0.08 volts per second, which equals 2.64V. The analog comparator compares the 2.64V at the positive input with the 1.65V at the negative input. Since 2.64V is greater than 1.65V, the comparator's output immediately flips from 0V to its power rail voltage of 3.3V. This rising edge from 0V to 3.3V serves as a hardware wake-up interrupt signal and is directly sent to the EXTI_5 pin of the main control unit, which is pre-configured to trigger an interrupt on a rising edge. Upon detecting this rising edge, the main control unit's hardware interrupt logic immediately starts the internal clock, waking the main control unit from deep sleep.

[0045] S3. In response to the hardware wake-up interrupt signal, activate the main control unit to collect signal snapshot data within a preset short time window, perform feature calculations on the signal snapshot data and compare it with the noise model. If the comparison result is determined to be effective interference, generate a high-precision waveform recording command.

[0046] In a specific embodiment of the present invention, the main control unit is activated to collect signal snapshot data within a preset short time window, and the signal snapshot data is subjected to feature calculation and compared with a noise model. If the comparison result is determined to be effective interference, a high-precision waveform recording instruction is generated, including: the main control unit responds to the hardware wake-up interrupt signal to reset from the sleep state, and controls the high-precision analog-to-digital converter to collect signal snapshot data within the preset short time window at a first sampling rate.

[0047] Perform fast feature operations on the signal snapshot data to calculate the signal RMS value and zero-crossing rate parameters, and compare the calculation results with the preset noise model.

[0048] If the comparison result is determined to be invalid noise, a suppression feedback instruction containing threshold increase parameters is generated to dynamically increase the dynamic trigger benchmark and return the system to sleep.

[0049] If the comparison result determines that it is effective interference, a high-precision waveform recording command is generated to maintain the wake-up state of the main control unit.

[0050] Specifically, upon detecting a valid level transition edge of the hardware wake-up interrupt signal, the external interrupt controller of the main control unit immediately resets from deep sleep and starts the internal high-speed clock oscillator. The main control unit jumps to the preset interrupt service routine entry address via its internal program counter and begins executing a rapid signal validity prediction process. First, the interrupt service routine outputs a high level to the enable pin of the high-precision analog-to-digital converter (ADC) through an I / O port, activating it from the power-off state. Subsequently, the main control unit configures the operating parameters of the ADC via the SPI bus, sets its sampling clock to the first sampling rate, and starts an internal timer. The timer is configured to continuously generate an analog-to-digital conversion trigger signal within a preset short time window. During this window, the ADC continuously samples the pipeline protection potential signal and transmits the converted digital value without delay to a designated buffer in the main control unit's internal RAM via a DMA (Direct Memory Access) channel, forming signal snapshot data. After data acquisition, the main control unit immediately performs rapid feature operations on the signal snapshot data. It calculates the effective value of the signal by traversing the sample points in the buffer and calculates the zero-crossing rate parameter by detecting the number of sign changes in the sample sequence. The calculated effective signal value and zero-crossing rate parameter are combined into a two-dimensional feature point and compared with a preset noise model stored in non-volatile memory. If the two-dimensional feature point falls within the judgment area defined by the preset noise model, it is determined to be invalid noise. At this time, the main control unit generates a suppression feedback instruction containing a threshold boosting parameter. This suppression feedback instruction is used to update the dynamic triggering reference in S1, and then the sleep instruction is executed again to return to the deep sleep state. If the two-dimensional feature point falls outside the judgment area defined by the preset noise model, it is determined to be valid interference. At this time, the main control unit generates a high-precision waveform recording instruction, maintains the wake-up state, and transfers the program execution to the subsequent high-precision data acquisition module.

[0051] Formula explanation: The root mean square algorithm is used to calculate the effective value of the signal. ,in, The effective value of the signal is the value of the snapshot data. This represents the total number of sample points contained in the signal snapshot data. For the first The voltage values ​​at each sample point. The zero-crossing rate parameter is calculated using the sign change counting method. , where ZCR is the zero-crossing rate parameter. The duration of the preset short time window. (Function) For a sign function, when When it is greater than 0, it takes the value of 1. When less than 0, take negative 1; when When the value is 0, take 0.

[0052] The first sampling rate is a medium sampling frequency set for rapid prediction, ranging from 1kSPS to 5kSPS. This setting controls the amount of data in the signal snapshot while ensuring the basic shape of the captured signal, thus shortening the execution time of the prediction algorithm. The preset short time window is the duration of the snapshot acquisition, typically ranging from 50ms to 100ms, sufficient to cover several power frequency interference cycles and providing support for the effective calculation of the zero-crossing rate parameter. The signal snapshot data is a one-dimensional discrete voltage sequence acquired at the first sampling rate within the preset short time window. The signal RMS value is the root mean square value of the signal snapshot data, used to characterize the signal's energy intensity. The zero-crossing rate parameter characterizes the frequency at which the signal crosses zero level, used to quickly assess the main frequency components of the signal. The preset noise model is a decision boundary established based on empirical data. For example, a two-dimensional judgment condition is defined as classifying the signal as invalid noise when the signal RMS value is less than 30mV and the zero-crossing rate parameter is greater than 1000Hz. This model is based on statistical analysis of a large amount of data acquired during interference-free periods in the field. The feedback suppression command is an internal control command used to trigger the adjustment logic of the dynamic trigger reference. The threshold increase parameter is a preset voltage increment value, such as five percent of the current reference voltage, used to appropriately raise the hardware screening threshold after a noise-induced false trigger. The high-precision waveform recording command is another internal control command used to initiate the subsequent full-bandwidth data acquisition process.

[0053] For example, continuing from the previous embodiment, the main control unit is awakened from deep sleep by a hardware wake-up interrupt signal. The interrupt service routine immediately starts the high-precision analog-to-digital converter AD7091R and configures it to sample at a first sampling rate of 2kSPS. Simultaneously, a timer is set to a preset short time window of 50ms. During this period, the AD7091R collects 2000 samples per second multiplied by 0.05 seconds, equaling 100 sample points. These 100 sample points constitute the signal snapshot data. Next, the main control unit processes these 100 sample points, calculating the effective signal value to be 150mV, and calculates the zero-crossing rate parameter to be 100Hz by statistically analyzing the symbol changes of the samples. Subsequently, the system compares this two-dimensional feature point composed of 150mV and 50Hz with a preset noise model. The criteria for determining the noise model are an effective signal value less than 30mV and a zero-crossing rate parameter greater than 1000Hz. Since the calculated effective signal value of 150mV is not less than 30mV, it does not meet the noise judgment condition. Therefore, the system determines that this wake-up is suspected effective interference. Finally, the main control unit generates a high-precision waveform recording instruction, which is a hexadecimal code 0xAA. The system writes this instruction into the internal status register and maintains the continuous working state of the central processing unit core and related peripherals, preparing to execute the next step.

[0054] S4. Execute high-precision waveform recording instructions to control the dual-mode front-end sensing circuit to perform full-bandwidth data acquisition, combine the acquired discrete voltage sequences and write them into the local memory to generate an original waveform data frame containing complete interference event information.

[0055] In a specific embodiment of the present invention, the collected discrete voltage sequences are combined and written into a local memory to generate an original waveform data frame containing complete interference event information, including: switching the high-precision analog-to-digital converter to a second sampling rate and continuously collecting the pipe potential signal.

[0056] The moving average of the signal amplitude is calculated in real time. When the moving average remains below the preset event end threshold for an extended period of time, data acquisition is stopped.

[0057] All the collected discrete voltage sequences are given a frame header and a frame tail. The frame header contains a timestamp and a geographic location tag obtained from the Global Positioning System. The original waveform data frame is then generated by combining these elements.

[0058] Specifically, upon receiving the high-precision waveform recording command, the main control unit's program flows to the full-bandwidth data acquisition module. First, the main control unit writes a new control word to the high-precision analog-to-digital converter's internal configuration register via its serial peripheral interface. This operation adjusts the clock division factor of the high-precision analog-to-digital converter, switching its operating mode from the first sampling rate to a preset second sampling rate. Next, the main control unit initiates a monitoring task, continuously acquiring the pipeline protection potential signal at the second sampling rate. To ensure the acquired data completely covers the entire interference event, the main control unit simultaneously executes a lightweight real-time signal analysis algorithm. This algorithm calculates the moving average of the signal amplitude within a short time window while acquiring data. When this moving average continuously falls below a preset event termination threshold within a silent acknowledgment period, the system determines that the interference event has ended and immediately stops data acquisition. All acquired discrete voltage sequences are sequentially stored in a dedicated data buffer within the main control unit's internal RAM. After data acquisition terminates, the main control unit calls the data encapsulation program to combine the discrete voltage sequences in the dedicated data buffer and add a data frame header and a data frame trailer to form a structured raw waveform data frame. Finally, the main control unit, through the memory interface controller, writes the raw waveform data frame as a whole data block to the next available sector of the local non-volatile memory, and simultaneously activates the GPS module to obtain the current timestamp and geographic location tag, and writes this metadata into the data frame header of the raw waveform data frame.

[0059] The second sampling rate is a high-speed sampling frequency set to achieve high-fidelity waveform recording. Its value must satisfy the Nyquist sampling theorem and is usually set to more than twice the highest harmonic frequency of the target interference signal, typically ranging from 10kSPS to 50kSPS. The duration of the interference event is a variable time length, adaptively determined by the main control unit through real-time monitoring of signal energy changes. The event termination threshold is a signal amplitude threshold used to determine whether the interference has ended. Its value is usually set to 1.5 to 3 times the system background noise level, which can be obtained by short-term statistical analysis of environmental signals during system initialization. The raw waveform data frame is a standardized data structure, typically containing a frame start symbol for synchronization and identification, a frame header containing metadata such as device ID, sampling rate, timestamp, geographic location tag, and data length, a data payload containing the actual acquired discrete voltage sequence, and a frame tail for data integrity verification, such as a cyclic redundancy check (CRC) code. Local non-volatile memory is a storage medium used to permanently store data, such as NAND Flash chips, characterized by data retention after power loss. A timestamp records the precise time information of the end of data acquisition, typically provided by a real-time clock (RTC) or a GPS module, and can be in UNIX timestamp format. A geolocation tag records the coordinate information of the device's location, provided by a GPS module, and is usually in latitude and longitude coordinates.

[0060] For example, continuing from the previous embodiment, the main control unit reads the high-precision waveform recording instruction 0xAA from its internal status register, and then writes configuration data to the AD7091R via the SPI bus, increasing its sampling rate from 2kSPS to a second sampling rate of 10kSPS. Simultaneously, the system activates the GPS module to obtain positioning and timing information. After data acquisition begins, the main control unit calculates the moving average of the signal amplitude within a 2-millisecond window (every 20 sampling points) in real time. The system's preset event termination threshold is 25mV, and the silence confirmation period is 30ms. A strong interference event causes the signal amplitude to fluctuate above 100mV for 450ms. After 450ms, the interference disappears, and the signal amplitude quickly drops back to a background noise level of 10mV. The main control unit detects that the moving average is below 25mV and remains below this threshold for the next 30ms, thus stopping data acquisition after a total of 480ms. This process collected 10,000 samples per second, which multiplied by 0.48 seconds, equaling 4,800 voltage sample points with 12-bit precision. After data acquisition, the GPS module reported a timestamp of 1698417015 (UNIX timestamp) and a geographic location tag of 39.9042 degrees North latitude and 116.4074 degrees East longitude. Based on this, the main control unit constructed a raw waveform data frame. The payload of this frame consisted of 9,600 bytes of data composed of 4,800 sample points, while the frame header contained metadata such as the device ID, timestamp, and location tag. The frame tail contained a 16-bit CRC checksum. Finally, this complete raw waveform data frame was written as a data file into the onboard NAND Flash memory chip.

[0061] S5. Perform edge calculation on the original waveform data frame, extract key physical indicators in the time and frequency domains, and encode and compress them to generate waveform feature vectors for low-bandwidth transmission.

[0062] In a specific embodiment of the present invention, generating a waveform feature vector for low-bandwidth transmission includes: reading the original waveform data frame and extracting waveform peak value, pulse width, and rising edge slope parameters in the time domain dimension.

[0063] Perform a Fast Fourier Transform on the original waveform data frame to extract the main harmonic frequency and total harmonic distortion parameters in the frequency domain.

[0064] The extracted time-domain and frequency-domain parameters are linearly quantized and fixed-pointed, and then combined to generate a multi-dimensional waveform feature vector.

[0065] Specifically, after data acquisition is complete, the main control unit calls the edge computing processing program. This program first reads the raw waveform data frame stored in the previous step from the local non-volatile memory through the file system interface and loads it into a processing buffer in the internal RAM. Next, the program enters the time-domain parameter extraction stage. The main control unit traverses all discrete voltage sequences in the processing buffer, determining the maximum value as the waveform peak by comparing the absolute values ​​of each sampling point. Then, the system uses a preset percentage of the waveform peak as a threshold, searching for the index of the sample point that first exceeds the threshold and the index of the sample point that last falls below the threshold in the voltage sequence, and calculates the pulse width based on the difference between the two indices and the sampling period. Simultaneously, the system locates the main rising edge of the waveform, calculates its time span and voltage increment within a preset voltage range, thereby obtaining the rising edge slope parameter. Subsequently, the program enters the frequency-domain parameter extraction stage. The main control unit calls an embedded Fast Fourier Transform library function to perform a transform on the discrete voltage sequence in the processing buffer, generating spectral data containing the amplitude and phase of each frequency component. The main control unit ignores the DC component in the spectrum data and searches for the frequency point with the largest amplitude, determining it as the main harmonic frequency. Based on this main harmonic frequency, the system identifies the amplitude of each harmonic component and calculates the total harmonic distortion (THD) parameter according to a standard formula. Finally, the program enters the encoding and compression stage, converting the extracted time-domain parameters (waveform peak value, pulse width, and rise slope) and frequency-domain parameters (main harmonic frequency and THD) into compact integer data through linear quantization and fixed-point conversion. These integer data are then arranged and combined in a predetermined order to generate a multi-dimensional waveform feature vector.

[0066] Formula Explanation: The formula for calculating the total harmonic distortion parameter is as follows: ,in, The total harmonic distortion parameter is a dimensionless ratio. The main harmonic frequency is the effective voltage value of the fundamental component. For the first Effective voltage value of the subharmonic component. To calculate the highest harmonic order under consideration, the effective voltage value of each harmonic component is directly derived from the amplitude at the corresponding frequency point in the Fast Fourier Transform result.

[0067] By verifying the dimensionality consistency of the above formula, the numerator is the square root of the sum of the squares of the effective voltage values ​​of each harmonic component, with the unit being volts (V). The denominator is the effective voltage value of the fundamental component, also with the unit being volts (V). The units of the numerator and denominator cancel each other out. Therefore... It is a dimensionless value, which conforms to its physical definition, and the formula is dimensionally consistent.

[0068] Edge computing refers to the process of directly analyzing and processing data at the device end, i.e., the main control unit, close to the data source, rather than uploading all the raw data to the cloud server. Time-domain parameters are features extracted directly from the original waveform showing the relationship between time and voltage. These include the waveform peak value (the maximum absolute voltage of the signal throughout the entire event cycle); pulse width (the duration for which the signal amplitude exceeds 50% of the waveform peak value); and rise slope (the rate of voltage change required for the signal to rise from 10% to 90% of the peak value). Frequency-domain parameters are features extracted from the frequency-energy spectrum after Fast Fourier Transform (FFT). These include the dominant harmonic frequency (the frequency component with the most concentrated signal energy) and total harmonic distortion (THD), which measures the degree of waveform distortion relative to an ideal sine wave. Encoding compression converts multiple floating-point or long integer feature parameters into shorter integer or fixed-point formats with smaller storage requirements through scaling and translation operations, facilitating transmission in low-bandwidth channels. The waveform feature vector is a fixed-length one-dimensional array composed of all the extracted and compressed feature parameters mentioned above. It summarizes the core physical characteristics of the original interference event with a small data volume.

[0069] For example, continuing from the previous embodiment, the main control unit reads a raw waveform data frame containing 4800 sample points from the NAND Flash. First, in time-domain analysis, the system traverses the data and finds that the maximum sample value is +250mV, thus determining the waveform peak value to be 250mV. Using 50% of the peak value, i.e., 125mV, as a threshold, the system finds that the signal first exceeds 125mV at the 100th sample point and last falls below 125mV at the 600th sample point. Since the sampling rate is 10kSPS and the interval between each sample point is 0.1ms, the pulse width is calculated to be 50ms. Subsequently, the system calculates that the signal rises from 10% of the peak value, i.e., 25mV, to 90%, i.e., 225mV, at the 50th and 150th sample points, respectively, with a time span of 10ms and a voltage increment of 200mV. Therefore, the rise edge slope is calculated to be 20V per second. Next, in the frequency domain analysis, the system performed a 4096-point Fast Fourier Transform on 4800 sample points and found that the highest energy spectral peak appeared in the 50Hz frequency range, thus determining the main harmonic frequency to be 50Hz. Based on this fundamental frequency, the system read its effective voltage value from the spectrum. The effective voltage value is 150mV with a second harmonic frequency of 100Hz. The effective voltage value is 20mV, with a third harmonic frequency of 150Hz. The value is 15mV; higher harmonic energy is negligible. The total harmonic distortion (THD) is calculated based on this. The value is approximately 0.167. Finally, the main control unit encodes and compresses these five parameters: 250mV is stored as a 16-bit integer 250, 50ms as a 16-bit integer 50, 20V per second as a 16-bit integer 20, 50Hz as a 16-bit integer 50, and 0.167 as a 16-bit integer (0.167 multiplied by 10000), resulting in 1670. These five 16-bit integers are arranged sequentially to form a 10-byte waveform feature vector.

[0070] S6. Based on the waveform feature vector analysis of the interference event level, dynamically schedule the wireless transmission strategy, and send the waveform feature vector or the original waveform data frame to complete the differentiated reporting of monitoring data.

[0071] In a specific embodiment of the present invention, the differentiated reporting of monitoring data includes: inputting the waveform feature vector into a preset event classification logic to determine the hazard level of the current interference event.

[0072] If the severity level of the current interference event is determined to be at the level of concern, then only the waveform feature vector is transmitted through the low-power wireless communication link.

[0073] If the current interference event is determined to be at a dangerous level, a waveform feature vector is immediately sent as an alarm, and the original data readback mechanism is activated at the same time.

[0074] In a specific embodiment of the present invention, the dynamic scheduling wireless transmission strategy further includes performing the following steps when the hazard level of the current interference event is determined to be dangerous: immediately sending waveform feature vectors as alarm information through a low-power wireless communication link.

[0075] The original data readback mechanism is activated, and the target packet length that can accommodate an integer number of signal cycles is calculated based on the main harmonic frequency parameter in the waveform feature vector.

[0076] Based on the target packet length, the locally stored original waveform data frame is divided into multiple data packets, and the broadband wireless communication link is activated for asynchronous transmission.

[0077] Specifically, after generating the waveform feature vector, the main control unit calls the event classification and data reporting program. This program takes the waveform feature vector as input and passes it to an event classification logic pre-stored in non-volatile memory. This logic, based on a set of rules defined by expert experience, performs a composite conditional judgment on the parameters of each dimension in the waveform feature vector to determine the severity level of the current interference event. If the severity level of the current interference event is determined to be at the concern level, the main control unit initializes its low-power wireless communication link and encapsulates the waveform feature vector in a lightweight data packet, asynchronously sending it to the remote monitoring center through the transmission interface of the low-power wireless communication link. After completing the reporting, it immediately enters a low-power standby state. If the severity level of the current interference event is determined to be at the danger level, the main control unit immediately executes a dual reporting strategy. First, it prioritizes sending the waveform feature vector as an immediate alarm message through the low-power wireless communication link to ensure that the monitoring center is aware of the occurrence of a serious event as soon as possible. At the same time, the main control unit initiates the raw data readback mechanism. The raw data readback mechanism first calls a packet length calculation function based on the main harmonic frequency parameter in the waveform feature vector to calculate the target packet length that can fully accommodate several interference cycles. Subsequently, the raw data readback mechanism segments the raw waveform data frame stored in the local non-volatile memory into multiple data packets, using the calculated target packet length as the unit. Finally, the main control unit activates its broadband wireless communication link and asynchronously and sequentially sends these data packets to the remote server via a reliable transmission protocol that supports retransmission acknowledgments, until the entire raw waveform data frame has been transmitted.

[0078] The event classification logic is a decision tree or set of conditional statements embedded in the program code. For example, a rule could be defined as follows: when the waveform peak value exceeds 500mV or the total harmonic distortion (THD) is greater than 0.3, the event is classified as hazardous; otherwise, when the waveform peak value is between 100mV and 500mV, it is classified as concerning. The hazard level is a classification of the potential impact of the interference event. For example, "concern level" indicates that it needs to be recorded for trend analysis, while "hazard level" indicates that it may pose an immediate threat to pipeline safety and requires immediate human intervention. Low-power wireless communication links refer to communication technologies designed specifically for IoT scenarios. They are characterized by long transmission distances and extremely low power consumption, but lower data rates, making them suitable for transmitting small data packets. The raw data readback mechanism is a background data transmission task triggered after a hazardous level is determined. The target packet length is a data block size calculated based on signal characteristics, striking a balance between transmission efficiency and data integrity. For example, for 50Hz interference, one period is 20ms. If the sampling rate is 10kSPS, then one period contains 200 sampling points. The target packet length can be set to an integer multiple of several periods, such as 1000 sampling points, which facilitates signal reconstruction and periodic analysis at the receiver. Broadband wireless communication links refer to communication technologies that provide high data transmission rates, such as 4G LTE or 5G NR, suitable for transmitting large raw data files. Asynchronous transmission means that data transmission does not require an immediate response, allowing the entire transmission task to be completed in batches and non-real-time when network conditions permit.

[0079] For example, continuing from the previous embodiment, the main control unit inputs a waveform feature vector containing five parameters, namely [250, 50, 20, 50, 1670], into the event classification logic. One rule of this logic is "if the waveform peak value is between 100 and 500 and the total harmonic distortion (THD) is less than 0.2, it is classified as a concern level." Since the peak value in the waveform feature vector is 250mV and the THD is 0.167, it fully meets this rule, therefore the current interference event is classified as a concern level. Subsequently, the main control unit activates its onboard narrowband IoT communication module, appends the device ID and timestamp to this 10-byte waveform feature vector to form a data packet of approximately 30 bytes in length, and sends it to the designated cloud platform server via the narrowband IoT network. After the transmission is completed, the communication module is shut down, and the main control unit re-enters deep sleep mode, waiting for the next interruption to wake it up. If, in another scenario, the extracted waveform feature vector shows a peak value of 600mV and a THD of 0.35, the event is classified as a danger level. The main control unit immediately sends an alarm message via narrowband IoT. Simultaneously, it initiates the raw data readback mechanism. Based on the main harmonic frequency of 50Hz in the waveform feature vector, the system calculates that each cycle contains 200 sampling points. Therefore, the target packet length is set to the amount of data for 5 cycles, i.e., 1000 sampling points (2000 bytes). Since there are a total of 4800 sampling points, the system divides the raw waveform data frame into 5 data packets, with the first 4 packets each containing 1000 sampling points and the last containing 800 sampling points. Next, the system activates the 4G LTE module, sequentially sending these 5 data packets to the server's data archiving interface via TCP / IP protocol.

[0080] Reference Figure 2 The second aspect of the present invention provides a solar-powered underground interference source waveform detection system, comprising: a system initialization and management module, a hardware wake-up interrupt signal generation module, an event determination and control module, a raw waveform data frame generation module, a waveform feature vector generation module, and a differentiated data reporting module.

[0081] The system initialization and management module is connected to the hardware wake-up interrupt signal generation module. The hardware wake-up interrupt signal generation module is connected to the event determination and control module. The event determination and control module is connected to the raw waveform data frame generation module. The raw waveform data frame generation module is connected to the waveform feature vector generation module. Both the raw waveform data frame generation module and the waveform feature vector generation module are connected to the differentiated data reporting module.

[0082] The system initialization and management module constructs an adaptive sensing environment based on solar power, initializes the dual-mode front-end sensing circuit, sets an initial hardware threshold corresponding to the upper limit of the voltage change rate of the pipeline protection potential signal, marks it as a dynamic trigger reference, and puts the main control unit into a deep sleep state that only retains the analog comparator.

[0083] The hardware wake-up interrupt signal generation module uses an analog comparator to monitor the pipeline potential signal in real time. When the voltage change rate of the input signal exceeds the dynamic trigger reference, the output level is flipped to generate a hardware wake-up interrupt signal indicating the occurrence of potential interference.

[0084] The event determination and control module, in response to the hardware wake-up interrupt signal, activates the main control unit to collect signal snapshot data within a preset short time window, performs feature calculations and noise model comparisons on the signal snapshot data, and if the comparison result determines that it is effective interference, it generates a high-precision waveform recording command.

[0085] The original waveform data frame generation module executes high-precision waveform recording instructions, controls the dual-mode front-end sensing circuit to perform full-bandwidth data acquisition, combines the acquired discrete voltage sequences and writes them into the local memory to generate an original waveform data frame containing complete interference event information.

[0086] The waveform feature vector generation module performs edge calculations on the original waveform data frame, extracts key physical indicators in the time and frequency domains, and encodes and compresses them to generate waveform feature vectors for low-bandwidth transmission.

[0087] The differentiated data reporting module analyzes the interference event level based on the waveform feature vector, dynamically schedules the wireless transmission strategy, and sends the waveform feature vector or the original waveform data frame to complete the differentiated reporting of monitoring data.

[0088] The above content is merely an example and illustration of the concept of the present invention. Those skilled in the art can make various modifications or additions to the specific embodiments described, or use similar methods to replace them, as long as they do not deviate from the concept of the invention or exceed the scope defined by the present invention, and all such modifications and additions should fall within the protection scope of the present invention.

Claims

1. A method for detecting waveforms of underground interference sources powered by solar energy, characterized in that, include: S1. Construct an adaptive sensing environment based on solar power, initialize the dual-mode front-end sensing circuit, set an initial hardware threshold corresponding to the upper limit of the voltage change rate of the pipeline protection potential signal, mark it as a dynamic trigger reference, and put the main control unit into a deep sleep state that only retains the operation of the analog comparator. S2. Use an analog comparator to monitor the pipeline potential signal in real time. When the voltage change rate of the input signal exceeds the dynamic trigger reference, the output level is flipped to generate a hardware wake-up interrupt signal indicating the occurrence of potential interference. S3. In response to the hardware wake-up interrupt signal, activate the main control unit to collect signal snapshot data within a preset short time window, perform feature calculations on the signal snapshot data and compare it with the noise model. If the comparison result is determined to be effective interference, generate a high-precision waveform recording command. S4. Execute high-precision waveform recording instructions to control the dual-mode front-end sensing circuit to perform full-bandwidth data acquisition, combine the acquired discrete voltage sequences and write them into the local memory to generate an original waveform data frame containing complete interference event information. S5. Perform edge calculation on the original waveform data frame, extract key physical indicators in the time and frequency domains, and encode and compress them to generate waveform feature vectors for low-bandwidth transmission. S6. Based on the waveform feature vector analysis of the interference event level, dynamically schedule the wireless transmission strategy, and send the waveform feature vector or the original waveform data frame to complete the differentiated reporting of monitoring data.

2. The method for detecting waveforms of underground interference sources powered by solar energy according to claim 1, characterized in that, The initialization of the dual-mode front-end sensing circuit includes: The energy storage module is charged using solar cell units in conjunction with a maximum power point tracking algorithm; A microampere-level operating current is provided to the dual-mode front-end sensing circuit through a unidirectional switching circuit; The topology of the dual-mode front-end sensing circuit is configured such that the analog comparator and the high-precision analog-to-digital converter are connected in parallel to the same sensor signal input port through an electronic switch, and the signal path is preferentially connected to the analog comparator during the initialization phase.

3. The method for detecting waveforms of underground interference sources powered by solar energy according to claim 1, characterized in that, The setting of an initial hardware threshold corresponding to the upper limit of the rate of change of the pipeline protection potential signal voltage, and marking it as a dynamic triggering reference, includes: Obtain the maximum voltage change rate of the protection potential signal of the pipeline under normal operating conditions; The upper limit of the voltage change rate is calculated by multiplying the maximum voltage change rate by the preset safety factor. An analog reference voltage corresponding to the upper limit of the voltage change rate is generated by a digital-to-analog converter or a programmable reference source, applied to the negative phase input of the analog comparator, and the value of the analog reference voltage is stored in non-volatile memory as a dynamic trigger reference.

4. The method for detecting waveforms of underground interference sources powered by solar energy according to claim 1, characterized in that, The generation of the hardware wake-up interrupt signal indicating the occurrence of potential interference includes: By using a passive differentiating circuit composed of resistors and capacitors, the pipeline protection potential signal is converted into an analog voltage proportional to its rate of change. The analog voltage is input to the positive input terminal of the analog comparator and compared in real time with the dynamic trigger reference at the negative input terminal. When the analog voltage is higher than the dynamic trigger reference, the analog comparator output pin generates a level transition, and the transition edge is directly transmitted to the external interrupt pin of the main control unit as a hardware wake-up interrupt signal.

5. The method for detecting waveforms of underground interference sources powered by solar energy according to claim 1, characterized in that, The activated main control unit collects signal snapshot data within a preset short time window, performs feature calculations on the signal snapshot data, and compares it with a noise model. If the comparison result determines it to be effective interference, a high-precision waveform recording command is generated, including: The main control unit responds to the hardware wake-up interrupt signal to reset from the sleep state and controls the high-precision analog-to-digital converter to collect signal snapshot data within a preset short time window at the first sampling rate; Perform fast feature operations on the signal snapshot data to calculate the signal RMS value and zero-crossing rate parameters, and compare the calculation results with the preset noise model; If the comparison result is determined to be invalid noise, a suppression feedback instruction containing threshold increase parameters is generated to dynamically increase the dynamic trigger benchmark and return the system to sleep. If the comparison result determines that it is effective interference, a high-precision waveform recording command is generated to maintain the wake-up state of the main control unit.

6. The method for detecting waveforms of underground interference sources powered by solar energy according to claim 1, characterized in that, The step of combining the acquired discrete voltage sequences and writing them into the local memory to generate a raw waveform data frame containing complete interference event information includes: Switch the high-precision analog-to-digital converter to the second sampling rate to continuously acquire the pipeline potential signal; The moving average of the signal amplitude is calculated in real time. When the moving average remains below the preset event end threshold for an extended period of time, data acquisition is stopped. All the collected discrete voltage sequences are given a frame header and a frame tail. The frame header contains a timestamp and a geographic location tag obtained from the Global Positioning System. The original waveform data frame is then generated by combining these elements.

7. The method for detecting waveforms of underground interference sources powered by solar energy according to claim 1, characterized in that, The generation of waveform feature vectors for low-bandwidth transmission includes: Read the raw waveform data frame and extract the waveform peak value, pulse width, and rising edge slope parameters in the time domain. Perform a fast Fourier transform on the original waveform data frame to extract the main harmonic frequency and total harmonic distortion parameters in the frequency domain. The extracted time-domain and frequency-domain parameters are linearly quantized and fixed-pointed, and then combined to generate a multi-dimensional waveform feature vector.

8. The method for detecting waveforms of underground interference sources powered by solar energy according to claim 1, characterized in that, The completion of differentiated reporting of monitoring data includes: Input the waveform feature vector into the preset event classification logic to determine the severity level of the current interference event; If the severity level of the current interference event is determined to be at the level of concern, then only the waveform feature vector is transmitted through the low-power wireless communication link; If the current interference event is determined to be at a dangerous level, a waveform feature vector is immediately sent as an alarm, and the original data readback mechanism is activated at the same time.

9. A method for detecting waveforms of underground interference sources powered by solar energy according to claim 8, characterized in that, The dynamic scheduling wireless transmission strategy also includes the following steps to be executed when the hazard level of the current interference event is determined to be dangerous: Immediately transmit waveform feature vectors as alarm information via a low-power wireless communication link; The original data readback mechanism is activated, and the target packet length that can accommodate an integer number of signal cycles is calculated based on the main harmonic frequency parameter in the waveform feature vector. Based on the target packet length, the locally stored original waveform data frame is divided into multiple data packets, and the broadband wireless communication link is activated for asynchronous transmission.

10. A solar-powered underground interference source waveform detection system, characterized in that, include: The system initialization and management module constructs an adaptive sensing environment based on solar power, initializes the dual-mode front-end sensing circuit, sets an initial hardware threshold corresponding to the upper limit of the voltage change rate of the pipeline protection potential signal, marks it as a dynamic trigger reference, and puts the main control unit into a deep sleep state that only retains the operation of the analog comparator. The hardware wake-up interrupt signal generation module uses an analog comparator to monitor the pipeline potential signal in real time. When the voltage change rate of the input signal exceeds the dynamic trigger reference, the output level is flipped to generate a hardware wake-up interrupt signal indicating the occurrence of potential interference. The event determination and control module responds to the hardware wake-up interrupt signal, activates the main control unit to collect signal snapshot data within a preset short time window, performs feature calculations and compares the signal snapshot data with a noise model, and if the comparison result determines that it is effective interference, it generates a high-precision waveform recording command. The original waveform data frame generation module executes high-precision waveform recording instructions, controls the dual-mode front-end sensing circuit to perform full-bandwidth data acquisition, combines the acquired discrete voltage sequences and writes them into the local memory to generate an original waveform data frame containing complete interference event information. The waveform feature vector generation module performs edge calculations on the original waveform data frame, extracts key physical indicators in the time and frequency domains, and encodes and compresses them to generate waveform feature vectors for low-bandwidth transmission. The differentiated data reporting module analyzes the interference event level based on the waveform feature vector, dynamically schedules the wireless transmission strategy, and sends the waveform feature vector or the original waveform data frame to complete the differentiated reporting of monitoring data.

Images