Pulse ultrasonic sensing device and method for fluid anomaly detection

Abstract

The invention relates to the technical field of measurement and testing, and discloses a pulse ultrasonic sensing device and method for fluid anomaly detection, and the method comprises the steps: a programmable pulse generator in an application-specific integrated circuit generates a short pulse sequence with preset parameters, and the short pulse sequence is amplified by a power amplifier and then is transmitted to a fluid medium by an ultrasonic transducer. In a transmitting gap, the transmitting-receiving switching circuit switches the system to a receiving mode, and echo signals are conditioned through the low-noise preamplifier, the programmable band-pass filter and the automatic gain control circuit in sequence. The conditioned signals are digitalized through an analog-to-digital converter array, pulse shaping, time sequence alignment, signal accumulation and digital filtering operation are carried out in a digital processing module, then signal envelopes are extracted by an envelope detector, feature vectors are calculated, and finally abnormity judgment is carried out by a judgment logic unit. According to the invention, rapid ultrasonic detection and accurate judgment of abnormal signals are effectively realized.

Description

Technical Field

[0001] This invention relates to the field of fluid detection device technology, and more specifically, to a pulsed ultrasonic sensing device and method for detecting fluid anomalies. Background Technology

[0002] Ultrasonic sensing technology, due to its non-contact nature, high precision, and good adaptability to various fluid media, has been widely used in fields such as gas transportation, industrial fluid metering, and safety monitoring. In fluid pipeline systems, ultrasonic waves measure flow rate by measuring the time difference between upstream and downstream propagation, which has become the core technology of modern smart gas meters.

[0003] However, existing ultrasonic detection systems are primarily designed for steady-state flow measurement and lack the ability to identify transient anomalies characterized by short duration, weak amplitude, and low signal-to-noise ratio, such as minute leaks, bubble inclusions, local turbulence, and sudden flow velocity anomalies in pipelines. Traditional solutions are limited by poor excitation pulse stability, weak front-end noise suppression, and insufficient sampling rate and resolution, making it difficult to achieve high-sensitivity, real-time, and reliable anomaly capture in complex industrial environments. Existing devices generally use general-purpose microcontrollers, discrete analog circuits, or low-integration signal conditioning modules to complete excitation, reception, and processing tasks. Limited by bandwidth, timing accuracy, and analog-to-digital conversion speed, this results in significant bottlenecks in weak echo extraction and time delay measurement. Furthermore, the lack of dedicated integrated circuits optimized for pulsed ultrasonic detection means that pulse shaping, high-speed echo capture, feature extraction, and real-time decision-making largely rely on software implementation. This leads to large delays, high power consumption, and weak anti-interference capabilities, making it difficult to synchronously analyze the multi-dimensional ultrasonic characteristics of rapidly changing fluid states, thus easily resulting in missed detections or false positives.

[0004] Therefore, how to accelerate the entire process from excitation to decision through a highly integrated hardware architecture, and build a pulsed ultrasonic sensing system with high timing accuracy, strong and weak signal sensing capabilities, and low latency response characteristics, has become an urgent technical challenge to improve the reliability and practicality of fluid anomaly detection. Summary of the Invention

[0005] In view of this, the present invention proposes a pulsed ultrasonic sensing device and method for fluid anomaly detection, aiming to solve the problem that existing ultrasonic detection systems in the current technology cannot achieve high sensitivity, low latency and high reliability real-time anomaly identification when facing short-term, weak amplitude and low signal-to-noise ratio transient fluid anomalies due to insufficient excitation, reception and processing link performance and lack of dedicated integrated circuit support.

[0006] This invention proposes a pulsed ultrasonic sensing device for fluid anomaly detection, comprising: The pulse excitation generator is configured to generate an ultrasonic excitation signal with a preset pulse width, repetition frequency, and amplitude shaping, and then amplify and transduc the excitation signal to emit pulses into the fluid medium. An echo receiving conditioning array is electrically connected to a pulse excitation generator. The echo receiving conditioning array is configured to switch between excitation and reception, and to amplify, filter, and automatically adjust the gain of the echo signal from the fluid medium to form a conditioning signal with preset requirements. The data acquisition module is electrically connected to the echo receiving conditioning array. The data acquisition module is configured to perform analog-to-digital conversion on the conditioning signal at a sampling rate no lower than a preset high sampling rate, and convert the conditioned signal after analog-to-digital conversion into a digital signal sequence. The signal processing module, integrated within an application-specific integrated circuit, is configured to perform pulse shaping, timing alignment, signal accumulation, digital filtering, envelope extraction, and feature parameter extraction on digital signal sequences, and to complete anomaly detection based on preset logic. The communication power management module is electrically connected to the pulse excitation generator, the echo receiving conditioning array, the data acquisition module, and the signal processing module, respectively. The communication power management module is configured to upload abnormal events and characteristic data to the early warning PTZ and provide power management for the pulse excitation generator, the echo receiving conditioning array, the data acquisition module, and the signal processing module.

[0007] Furthermore, the pulse excitation generator includes: The programmable pulse generation unit is configured to generate ultrasonic excitation pulses based on preset pulse width, repetition frequency and amplitude shaping method; The power amplifier unit is electrically connected to the programmable pulse generator unit and is configured to amplify the excitation pulse to the drive level. A transducer drive unit is coupled to the power amplifier unit and is configured to convert the amplified excitation pulse into an ultrasonic pulse and output it to the fluid medium. The impedance matching and shaping array, electrically connected to the programmable pulse generation unit, is configured to match the output impedance of the drive path and maintain the edge and amplitude of the excitation pulse.

[0008] Furthermore, the echo receiver conditioning array includes: The receive switching unit is electrically connected to the programmable pulse generating unit and is configured to switch between excitation and reception. The preamplifier unit is electrically connected to the receiver switching unit and is used to amplify the received weak echo signal to a preset level. The programmable filter unit, electrically connected to the preamplifier unit, is used to filter out spurious frequencies in the echo signal to improve the signal-to-noise ratio. An automatic gain control unit, electrically connected to a programmable filter unit, is used to dynamically adjust the signal gain based on the echo signal amplitude.

[0009] Furthermore, the data acquisition module is electrically connected to the output terminal of the automatic gain control unit, wherein the data acquisition module includes: The high-speed sampling control unit is configured to generate a sampling clock, and is also configured to perform sampling on the conditioned signal after step-by-step conditioning at a rate not lower than a preset high sampling rate; The analog-to-digital conversion unit is electrically connected to the high-speed sampling control unit. The analog-to-digital conversion unit is configured to perform high-resolution analog-to-digital conversion on the conditioning signal according to the sampling clock to obtain digital sampling data. The buffer unit is electrically connected to the analog-to-digital conversion unit and is configured to buffer the digital sampled data after analog-to-digital conversion. The data packing unit is electrically connected to the buffer unit. The data packing unit is configured to perform timestamp marking, frame formatting and serialization processing on the digital sampled data in the buffer unit, and generate a digital signal sequence based on the processed digital sampled data. The acquisition unit is configured to control the acquisition window based on the trigger signal from the pulse excitation generator.

[0010] Furthermore, when the data packaging unit performs timestamp marking, frame formatting, and serialization processing on the digital sampled data in the buffer unit, it includes: The data packaging unit is also configured to determine the timestamp information of each digital sampled data in the buffer unit based on the system clock; The data packaging unit is also configured to assemble each digital sampled data into a data frame containing a frame header, payload, and check field according to a preset frame based on timestamp information; The data packing unit is also configured to perform CRC or parity checks and redundancy encoding on the data frames; The data packing unit is also configured to serialize the verified data frames.

[0011] Furthermore, the signal processing module includes: The pulse shaping unit is electrically connected to the data packing unit and is also configured to perform amplitude shaping and edge sharpening processing on the digital signal sequence; the pulse shaping unit is also configured to perform time base alignment on the digital sequence based on the excitation trigger signal; The signal accumulation unit is electrically connected to the pulse shaping unit and is configured to accumulate and average multiple acquired digital signal sequences. The digital filtering unit is electrically connected to the signal accumulation unit. The digital filtering unit is configured to perform bandpass, low-pass, or high-pass filtering on the accumulated digital signal based on a preset digital filter. The envelope detection and extraction unit is electrically connected to the digital filtering unit and is configured to perform envelope extraction on the filtered digital signal. The envelope detection and extraction unit is also configured to calculate key feature quantities such as peak value, energy, and time difference. The feature modeling unit is electrically connected to the envelope detection and extraction unit. The feature modeling unit is configured to extract feature indicators based on the amplitude attenuation, spectral change, and time parameters of the envelope features. The feature modeling unit is also configured to perform normalization and modeling processing on the feature indicators. An anomaly determination unit is electrically connected to the feature modeling unit. The anomaly determination unit is configured to compare the feature parameters after modeling with a preset determination rule or threshold to determine whether the fluid medium is in an abnormal state.

[0012] Furthermore, when the feature modeling unit performs normalization and modeling processing on the feature indicators, it includes: The feature modeling unit is also configured to perform normalization on multiple feature metrics, wherein the normalization includes at least one of Min-Max interval normalization, Z-Score standardization, logarithmic transformation or window normalization; The feature modeling unit is also configured to construct a multidimensional feature vector based on the normalized feature index. The feature modeling unit is also configured to extract the offset trend of features changing over time in the multidimensional feature vector based on sliding window statistics, mean-variance trend analysis or cumulative sum statistical model. The feature modeling unit is also configured to determine a measure of the difference between the current feature vector and the reference feature vector based on Euclidean distance, Mahalanobis distance, or dynamic time warping methods. The feature modeling unit is also configured to model the difference measure and statistical results based on a lightweight decision tree.

[0013] Furthermore, when the anomaly detection unit compares the modeled feature parameters with preset detection rules or thresholds to determine whether the fluid medium is in an abnormal state, it includes: The anomaly decision unit is also configured to compare the amplitude of the modeled feature parameters with a preset threshold to generate a first decision flag; The anomaly judgment unit is also configured to perform cumulative statistical calculations on the feature parameters across periods and generate a second judgment flag based on whether the cumulative trend exceeds the limit; The anomaly judgment unit is also configured to analyze the volatility, rate of change and degree of mutation of the characteristic parameters within a preset sliding window, and output a third judgment flag based on the analysis results. The anomaly detection unit is also configured to perform anomaly pattern recognition on feature parameters based on a preset rule base, and generate corresponding category labels and their confidence levels; The anomaly determination unit is also configured to perform weighted fusion processing on the first determination flag, the second determination flag, the third determination flag, and the category label to determine the final anomaly determination result for the fluid medium.

[0014] Furthermore, when the anomaly determination unit performs weighted fusion processing on the first determination flag, the second determination flag, the third determination flag, and the category label to determine the final anomaly determination result for the fluid medium, it includes: The anomaly judgment unit is also configured to assign corresponding weights to the first judgment flag, the second judgment flag, the third judgment flag, and the anomaly category label, and the anomaly judgment unit is also configured to adaptively adjust the weights based on historical data; The anomaly decision unit is also configured to obtain the confidence index of each decision flag, and the anomaly decision unit is also configured to obtain the classification probability of the anomaly category label, and obtain the confidence vector based on the classification probability; The anomaly detection unit is also configured to perform normalization and fusion operations on the confidence vector based on weighted summation, and determine the fusion score based on the operation results; The anomaly decision unit is also configured to compare the fusion score with a preset anomaly threshold and determine the final anomaly decision result according to a preset priority rule in case of conflict.

[0015] Compared with existing technologies, the advantages of this invention are as follows: By employing a pulse excitation generator to generate an ultrasonic excitation signal with preset pulse width, repetition frequency, and amplitude shaping characteristics, and by amplifying and transducing the excitation signal at the hardware level, this device can significantly improve pulse stability and temporal resolution, thereby enhancing the ability to detect weak disturbances in fluids. Compared with traditional solutions relying on general-purpose circuits, this excitation link can effectively reduce pulse jitter and improve energy consistency, laying a stable signal source foundation for subsequent weak echo detection. Secondly, by setting up an echo receiving conditioning array, the device can quickly and without interference switch between excitation and receiving modes, and perform multi-stage amplification, filtering, and automatic gain control on the echo. Through hardware-based noise suppression and dynamic gain adjustment, this device can obtain a purer, moderately amplitude conditioned signal under low signal-to-noise ratio conditions, significantly improving the ability to capture slight echo distortions caused by micro-leakage, bubble doping, etc. Meanwhile, by using a high-sampling-rate data acquisition module to perform high-speed analog-to-digital conversion on the conditioned signal, the system can accurately record the time-domain details of the echo signal, avoiding time delay measurement errors caused by insufficient sampling frequency in traditional methods. Furthermore, the structured output of the digital signal sequence provides a high-fidelity input foundation for subsequent feature extraction and pattern analysis. In addition, integrating the signal processing module into a dedicated integrated circuit enables key algorithms such as pulse shaping, timing alignment, signal accumulation, digital filtering, envelope extraction, and feature parameter extraction to be executed at the hardware level with low latency. Compared to traditional software-dependent processing methods, this structure significantly reduces processing latency and power consumption, improves anti-interference capabilities, and enables real-time feature analysis even with rapid changes in fluid conditions, thereby enhancing the stability and accuracy of anomaly detection. Finally, the communication power management module allows each module to operate stably in low-power mode and uploads abnormal events and feature data to the early warning PTZ in real time, achieving coordinated operation between edge detection and cloud monitoring. Through centralized power management and communication scheduling, the system's reliability, security, and long-term maintainability are further improved.

[0016] On the other hand, this application also provides a pulsed ultrasonic method for fluid anomaly detection, comprising: The control pulse excitation generator generates an ultrasonic excitation signal with preset pulse width, repetition frequency, and amplitude shaping, and then amplifies and transduces the excitation signal to emit pulses into the fluid medium; The echo signal from the fluid medium is amplified, filtered, and automatically gain-adjusted to form a conditioning signal with preset requirements; The conditioning signal is converted from analog to digital at a sampling rate no lower than the preset high rate, and the conditioned signal after analog-to-digital conversion is converted into a digital signal sequence. The system performs pulse shaping, timing alignment, signal accumulation, digital filtering, envelope extraction, and feature parameter extraction on digital signal sequences, and completes anomaly detection based on preset logic.

[0017] It is understood that the pulsed ultrasonic sensing device and method for fluid anomaly detection in the above embodiments of the present invention have the same beneficial effects, and will not be described again. Attached Figure Description

[0018] Various other advantages and benefits will become apparent to those skilled in the art upon reading the following detailed description of preferred embodiments. The accompanying drawings are for illustrative purposes only and are not intended to limit the invention. Furthermore, the same reference numerals denote the same parts throughout the drawings. In the drawings: Figure 1 This is a structural block diagram of a pulsed ultrasonic sensor for fluid anomaly detection provided in an embodiment of the present invention; Figure 2 This is a circuit connection diagram of a pulsed ultrasonic sensor for fluid anomaly detection provided in an embodiment of the present invention; Figure 3 A schematic diagram illustrating the use of a pulsed ultrasonic sensor for fluid anomaly detection provided in an embodiment of the present invention; Figure 4 This is a flowchart of a pulsed ultrasonic method for detecting fluid anomalies provided in an embodiment of the present invention. Detailed Implementation

[0019] Exemplary embodiments of the present disclosure will now be described in more detail with reference to the accompanying drawings. While exemplary embodiments of the present disclosure are shown in the drawings, it should be understood that the present disclosure may be implemented in various forms and should not be limited to the embodiments set forth herein. Rather, these embodiments are provided to enable a more thorough understanding of the present disclosure and to fully convey the scope of the disclosure to those skilled in the art. It should be noted that, unless otherwise specified, embodiments and features in the embodiments of the present invention can be combined with each other. The present invention will now be described in detail with reference to the accompanying drawings and embodiments.

[0020] like Figure 2 The diagram shows a circuit connection schematic of a pulsed ultrasonic sensor for fluid anomaly detection according to the present invention: A pulse excitation generator may include a programmable pulse generation unit, a power amplification unit, a transducer driving unit, and an impedance matching and shaping array.

[0021] It is understood that the structures illustrated in the embodiments of the present invention do not constitute a specific limitation on the pulse excitation generator. In other embodiments of this application, the pulse excitation generator may include more or fewer components than illustrated, or combine some components, or split some components, or have different component arrangements. The illustrated components may be implemented in hardware, software, or a combination of software and hardware.

[0022] Programmable pulse generator units can include FPGAs or CPLDs: they can flexibly generate high-precision timing pulse signals and modify parameters in real time. Programmable logic pulse generator chips (Pulse Generator ICs) are specifically designed for high-precision pulse output. Microcontrollers (MCUs) paired with high-speed timer modules generate low-frequency or medium-frequency pulses, with amplitude controllable via PWM. In embedded systems, programmable pulse generator units control GPIO or DAC output pulses via software, achieving adjustable amplitude and frequency.

[0023] The power amplifier unit may include a high-bandwidth power amplifier (Class D, AB, or linear amplifier), MOSFET or BJT drive circuitry to provide high current drive capability, and a high-voltage pulse amplification module for use when high-voltage drive of piezoelectric transducers is required. The power amplifier unit controls the amplifier gain or PWM duty cycle to achieve amplitude regulation and protection functions.

[0024] The transducer drive unit may include a piezoelectric transducer (the most commonly used ultrasonic signal generating device), an electroacoustic transducer (used for specific fluid or pipeline detection), and a drive circuit (which matches and excites the transducer with the amplified pulse signal). The transducer drive unit provides drive timing and amplitude control to ensure that the transducer output waveform is synchronized with the preset pulse.

[0025] Impedance matching and shaping arrays can include LC network matching circuits (using inductors and capacitors for impedance matching), impedance transformers or matching transformers (improving power transmission efficiency), and signal shaping circuits (edge ​​shaping / limiting circuits) (maintaining clear pulse waveforms). Impedance matching and shaping arrays are based on digital pulse shaping algorithms to correct pulse waveforms in DACs or FPGAs.

[0026] The echo receiver conditioning array may include a receive switching unit, a pre-amplifier unit, a programmable filter unit, and an automatic gain control (AGC) unit.

[0027] It is understood that the structures illustrated in the embodiments of the present invention do not constitute a specific limitation on the echo reception conditioning array. In other embodiments of this application, the echo reception conditioning array may include more or fewer components than illustrated, or combine some components, or split some components, or have different component arrangements. The illustrated components may be implemented in hardware, software, or a combination of software and hardware.

[0028] The receiver switching unit may include an analog switch or relay for high-frequency signal switching, a semiconductor radio frequency switch for fast, high-frequency switching, and a bidirectional current limiting / isolation circuit to protect the front-end amplifier from high-voltage pulse damage. The receiver switching unit controls the switching timing of the switches, generating control signals via an MCU or FPGA.

[0029] Preamplifier units may include low-noise amplifiers (LNAs): commonly used for weak signal acquisition, differential amplifier or operational amplifier configurations; providing high gain and low distortion, and high-frequency / wideband amplifiers; suitable for ultrasonic frequency bands (tens of kHz to MHz). Preamplifier units are based on ADC digital gain control, amplifying or adjusting the acquired signal via software. They are also used for signal amplitude adjustment before digital filtering.

[0030] Programmable filtering units may include tunable band-pass filters, filter arrays (LC or RC networks) that can switch different center frequencies and bandwidths, FPGAs or dedicated filter chips to implement programmable analog filtering, and digital filtering (FIR / IIR filters) implemented through DSPs or FPGAs, with software controlling the center frequency, bandwidth, or filter coefficients.

[0031] The automatic gain control unit (AGU) may include a variable gain amplifier (VGA) and an analog AGC circuit (automatically adjusting bias voltage or gain). After ADC sampling, the AUG calculates the signal amplitude and adjusts the VGA gain using software or a DSP. The AUG controls the upper and lower limits of the gain and the dynamic response speed based on algorithms.

[0032] The data acquisition module may include a high-speed sampling control unit, an analog-to-digital conversion unit (ADC unit), a buffering unit, a data packaging unit, and an acquisition unit.

[0033] It is understood that the structures illustrated in the embodiments of the present invention do not constitute a specific limitation on the data acquisition module. In other embodiments of this application, the data acquisition module may include more or fewer components than illustrated, or combine some components, or split some components, or have different component arrangements. The illustrated components may be implemented in hardware, software, or a combination of software and hardware.

[0034] The high-speed sampling control unit may include an FPGA or ASIC to generate a precise sampling clock, a high-speed timer or pulse generator to control ADC triggering, and an MCU / embedded processor to output control signals through peripheral timers.

[0035] The analog-to-digital conversion unit may include a high-speed, high-resolution ADC chip (12-bit, 16-bit or higher resolution), a parallel / serial interface for transmitting digital data, an ADC driver to configure the sampling rate, reference voltage and data format, and an FPGA or DSP to perform operations such as data synchronization and channel selection.

[0036] The buffer unit may include a FIFO buffer (hardware FIFO), SRAM / SDRAM cache, double buffer design, for parallel acquisition and processing, and DMA (Direct Memory Access) combined with buffer to achieve high-speed data transmission.

[0037] The data packaging unit may include FPGA / ASIC to implement data packaging logic, timestamp counter and frame builder, MCU or DSP to package data according to the sampling timestamp, and perform frame serialization through communication protocols (such as SPI, UART, Ethernet).

[0038] The acquisition unit may include an FPGA / timer to generate a synchronous acquisition window, a trigger input to directly control the sampling switch, an MCU or DSP to start acquisition in response to an external trigger signal, and software control of the sampling window width and data alignment.

[0039] The signal processing module may include a pulse shaping unit, a signal accumulation unit, a digital filtering unit, an envelope detection and feature extraction unit, a feature modeling unit, and an anomaly decision unit.

[0040] It is understood that the structures illustrated in the embodiments of the present invention do not constitute a specific limitation on the signal processing module. In other embodiments of this application, the signal processing module may include more or fewer components than illustrated, or combine some components, or split some components, or have different component arrangements. The illustrated components may be implemented in hardware, software, or a combination of software and hardware.

[0041] The pulse shaping unit may include FPGA or ASIC built-in digital filter / shaping circuitry, high-speed logic unit for sequence alignment and edge sharpening, DSP or MCU software algorithm for digital amplitude adjustment, edge detection, and time base alignment controlled by trigger signal to control buffer index.

[0042] The signal accumulation unit may include an accumulator or register array in the FPGA, SRAM / register buffer multiple sampled data, and MCU / DSP perform cumulative averaging on multiple frames of data. Sliding window accumulation or exponential averaging can be set.

[0043] The digital filtering unit may include an FPGA / ASIC implementing an FIR or IIR filter, a programmable filter coefficient array, a digital filtering algorithm implemented in DSP / Firmware, and the ability to dynamically adjust the filter type and cutoff frequency.

[0044] The envelope detection and extraction unit may include an FPGA-built-in multiplier and squaring / accumulating unit to perform energy calculation. Envelope detection uses Hilbert transform or analog-digital mixed logic, and DSP / MCU calculates the signal envelope, peak value, and time difference.

[0045] The feature modeling unit is used to normalize, parameterize, and construct feature vectors from the multidimensional features output by the envelope detection extraction subunit, forming a standardized feature set that can be used for anomaly identification. In some embodiments, this subunit can implement feature normalization, feature vector concatenation, statistical modeling, or lightweight machine learning modeling through embedded algorithm modules, DSP digital processing programs, or FPGA logic units. This subunit can generate feature vectors in a unified format based on original features such as amplitude attenuation, spectral change, and time parameters, and can further calculate feature weights, feature scores, or mode distances to support the subsequent anomaly decision unit in performing fluid medium state assessment and anomaly determination.

[0046] The anomaly determination unit may include a subunit for performing multi-dimensional comprehensive analysis based on the modeled feature parameters to determine whether the fluid medium is in an abnormal state. In some embodiments, this subunit can achieve real-time determination of the feature parameters through threshold comparison logic circuits, embedded decision algorithm modules, rule matching engines, or lightweight classification models. This subunit can perform threshold judgment, pattern matching, sliding window trend analysis, or statistical deviation scoring on the amplitude, spectrum, and time-related feature parameters output by the feature modeling unit, thereby generating a primary anomaly flag. Furthermore, it can weightedly fuse multiple determination results to form a final anomaly state determination result, which can be used to drive an early warning mechanism or report to a cloud-based early warning platform.

[0047] The communication power management module may include a communication transceiver unit, a protocol parsing and data packaging unit, a power regulation unit, an energy management and distribution unit, a status monitoring unit, and a safety protection unit. It is understood that the structures illustrated in the embodiments of the present invention do not constitute a specific limitation on the communication power management module. In other embodiments of this application, the communication power management module may include more or fewer components than illustrated, or combine some components, or split some components, or have different component arrangements. The illustrated components may be implemented in hardware, software, or a combination of software and hardware.

[0048] The communication transceiver unit is configured to remotely transmit system status, characteristic data, and abnormal events. In some embodiments, the communication transceiver unit can support communication methods such as RS485, CAN, Modbus-TCP, Ethernet, LoRa, Wi-Fi, NB-IoT, or Bluetooth. This unit can perform real-time data reporting, remote command reception, heartbeat maintenance, and link status monitoring, and supports multiple transmission rates and adaptive retransmission mechanisms to ensure reliable delivery of critical data.

[0049] The protocol parsing and data encapsulation unit is electrically connected to the communication transceiver unit and is used to perform formatting, encryption, verification, and serialization processing on data frames. In some embodiments, this unit can encapsulate characteristic parameters and anomaly judgment results from the signal processing module into standard data frames, and can generate instruction feedback messages based on application protocols. This unit can also perform protocol parsing, field verification, instruction distribution, and error recovery on received remote commands to ensure consistency and interoperability in inter-system communication.

[0050] The power supply conditioning unit is configured to perform buck, voltage regulation, filtering, and current limiting processing on the external input power supply. In some embodiments, the unit may include a high-efficiency DC-DC converter, a linear regulator, a low-noise filter, and a surge suppressor to provide a stable power supply with low ripple and high transient response to the pulse excitation generator, echo receiver conditioning array, and data acquisition module. This unit can effectively suppress power supply noise interference on weak echo signal paths.

[0051] The energy management and distribution unit is electrically connected to the power conditioning unit and is used to intelligently schedule energy distribution according to the system operation phase. In some embodiments, this unit can detect transient peak loads during pulse excitation phases and execute tiered power supply, dynamic power allocation, or power priority strategies to avoid power drops or system resets caused by excitation pulses. This unit can also manage backup power supplies, supercapacitors, or energy storage units to achieve power redundancy and continuous operation capability.

[0052] The safety protection unit is configured to implement a comprehensive safety protection strategy for the power supply path. In some embodiments, the unit may include overvoltage protection circuitry, overcurrent protection circuitry, short-circuit protection circuitry, thermal protection circuitry, and ESD protection structures to prevent damage to the system from external power fluctuations, lightning surges, or equipment failures. The unit can also perform emergency power outages or switch to backup power supply upon detecting severe power anomalies to ensure system safety and data integrity.

[0053] based on Figure 2 The diagram shows a circuit connection diagram of a pulsed ultrasonic sensor for fluid anomaly detection. This invention provides a pulsed ultrasonic sensor for fluid anomaly detection, such as... Figure 1 and Figure 3The pulsed ultrasonic sensing device shown for fluid anomaly detection includes: a pulse excitation generator, an echo receiving conditioning array, a data acquisition module, a signal processing module, and a communication power management module.

[0054] Specifically, a pulse excitation generator is configured to generate an ultrasonic excitation signal with a preset pulse width, repetition frequency, and amplitude shaping, and amplifies and transduces the excitation signal before transmitting it into the fluid medium. An echo receiving and conditioning array is electrically connected to the pulse excitation generator and is configured to switch between excitation and reception, and to amplify, filter, and automatically adjust the gain of the echo signal from the fluid medium to form a pre-defined conditioning signal. A data acquisition module is electrically connected to the echo receiving and conditioning array and is configured to perform analog-to-digital conversion on the conditioning signal at a sampling rate not lower than a preset high sampling rate, and then transmit the converted signal. The signal is converted into a digital signal sequence. The signal processing module is integrated into the application-specific integrated circuit. The signal processing module is configured to perform pulse shaping, timing alignment, signal accumulation, digital filtering, envelope extraction, and feature parameter extraction on the digital signal sequence, and to complete the anomaly judgment based on preset logic. The communication power management module is electrically connected to the pulse excitation generator, echo receiving conditioning array, data acquisition module, and signal processing module. The communication power management module is configured to upload abnormal events and feature data to the early warning PTZ and provide power management for the pulse excitation generator, echo receiving conditioning array, data acquisition module, and signal processing module.

[0055] Specifically, by coordinating pulsed ultrasonic excitation, echo reception and conditioning, high-speed data acquisition, and signal processing within a dedicated integrated circuit, accurate detection and anomaly identification of fluid medium conditions can be achieved. The overall principle is based on the quantifiable characteristics of ultrasonic pulse propagation in fluids and its amplitude attenuation, time delay, and spectral changes caused by changes in the medium state, which are used to construct a stable, repeatable, and interference-resistant detection link.

[0056] For example, the pulse excitation generator generates an ultrasonic excitation signal with stable waveform parameters based on preset pulse width, repetition frequency, and amplitude shaping methods. Through power amplification and transducer drive, this excitation signal is converted into an ultrasonic pulse by the transducer and injected into the fluid medium, forming a propagation wave field that can be used for detection. Changes in the density, bubble content, impurities, and flow disturbances of the fluid medium will directly cause differences in the amplitude, phase, and time difference of the echo signal, thus providing a physical basis for subsequent anomaly identification.

[0057] For example, the echo receiving conditioning array, based on an excitation / reception switching mechanism, switches the same transducer path from transmission mode to reception mode and performs conditioning operations such as pre-amplification, filtering, and automatic gain control on the weak ultrasonic echoes returning from the fluid medium. Through this link, a conditioning signal with moderate dynamic range, improved signal-to-noise ratio, and controllable composition can be acquired in high-noise environments, ensuring that the echo waveform can be sent to the acquisition link with high quality.

[0058] For example, the data acquisition module performs analog-to-digital conversion on the conditioned signal at a rate no lower than a preset sampling rate based on a high-speed sampling clock, and converts it into a time-consistent, frame-structured digital signal sequence. High-speed sampling technology ensures the ability to capture echo details, enabling subsequent signal shaping, envelope calculation, and feature modeling to maintain high accuracy.

[0059] For example, within an application-specific integrated circuit (ASIC), the signal processing module performs pulse shaping, timing alignment, cumulative averaging, digital filtering, envelope extraction, and key feature parameter extraction on the digital signal sequence to establish characteristic indicators sensitive to the state of the fluid medium. The accumulation and filtering processes improve echo stability, envelope and peak extraction reveal waveform energy and propagation characteristics, and feature modeling further standardizes multidimensional features, enabling efficient parsing by rules or models. Based on these processing results, the module performs anomaly detection according to preset decision logic or thresholds, thereby identifying whether the fluid medium exhibits conditions such as leakage, blockage, gas entrainment, or abnormal disturbances.

[0060] For example, the communication power management module is responsible for uploading abnormal events, characteristic data, and operating status to the early warning PTZ platform, enabling remote monitoring and data management. Simultaneously, this module provides stable, low-noise power management for each functional unit within the system, ensuring reliable operation of the excitation, reception, acquisition, and signal processing links under varying load conditions. This unified power and communication management mechanism also enhances the overall system security, robustness, and maintainability.

[0061] Specifically, the pulse excitation generator includes: a programmable pulse generation unit configured to generate an ultrasonic excitation pulse based on a preset pulse width, repetition frequency, and amplitude shaping method; a power amplification unit electrically connected to the programmable pulse generation unit, configured to amplify the excitation pulse to a drive level; a transducer drive unit coupled to the power amplification unit, configured to convert the amplified excitation pulse into an ultrasonic pulse and output it to a fluid medium; and an impedance matching and shaping array electrically connected to the programmable pulse generation unit, configured to match the output impedance of the drive path and maintain the edge and amplitude of the excitation pulse.

[0062] Specifically, the pulse excitation generator is based on the principle of pulsed ultrasound detection. By generating excitation pulses with stable shapes and controllable parameters, it enables the ultrasonic transducer to transmit high-fidelity ultrasonic signals into the fluid medium, thus providing a reliable signal source for subsequent echo reception and feature extraction. Its working mechanism relies on multiple collaborative technologies, including programmable pulse parameter control, power amplification, acoustic-to-electrical conversion, and impedance matching, to ensure the stability and effectiveness of the transmitted pulses.

[0063] Specifically, the programmable pulse generation unit flexibly generates excitation pulse waveforms by controlling pulse width, repetition frequency, and amplitude shaping. Precise adjustment of the pulse width controls the transducer's frequency band excitation characteristics; adjustment of the repetition frequency allows setting the transmission rhythm according to detection cycle requirements; and amplitude shaping technology reduces pulse sidelobes and waveform distortion, improving the spectral purity and energy concentration of the transmitted signal. This unit serves as the time and waveform reference for the entire excitation chain, laying the foundation for high-quality ultrasonic transmission.

[0064] In one specific embodiment, the programmable pulse generation unit generates excitation pulses for driving the transducer based on set pulse width, repetition frequency, and amplitude shaping parameters. By precisely controlling the pulse width, the transducer achieves optimal excitation within the target frequency band; by setting the repetition frequency, the excitation period is ensured to match the system detection period; and an amplitude shaping algorithm is used to correct the pulse edges, suppressing sidelobes and waveform distortion, thereby improving the spectral purity and energy concentration of the excitation pulse. With the programmable control capability of this unit, the excitation link can obtain stable, repeatable, and waveform-quality-controlled ultrasonic pulses, providing a consistent and high signal-to-noise ratio detection basis for subsequent fluid state analysis.

[0065] Specifically, the power amplifier unit amplifies the amplitude of the excitation pulse from the programmable pulse generator unit to achieve the voltage and current levels required to drive the transducer. Transducers typically have high excitation requirements, and direct driving by the waveform generation circuit is often insufficient to generate sufficiently strong ultrasonic waves. Therefore, the power amplifier unit can provide a stronger energy input, ensuring the transducer receives adequate electroacoustic conversion driving force. Furthermore, the linearity and transient response characteristics of this unit also have a crucial impact on pulse waveform preservation.

[0066] In some embodiments, the power amplification unit enhances the energy of the excitation pulse output by the programmable pulse generator unit, so that the pulse signal reaches the voltage and current levels required to drive the ultrasonic transducer. Since the transducer typically requires a high amplitude excitation during operation, and the waveform generation circuit itself cannot provide sufficient energy, the power amplification unit improves the driving capability of the excitation signal through linear amplification, current boosting, or pulse power enhancement, ensuring that the transducer can still obtain stable electroacoustic output under high load conditions. Simultaneously, the transient response speed and linearity of this unit have a significant impact on the fidelity of the excitation pulse waveform, effectively reducing waveform distortion and ensuring the consistency and controllability of the ultrasonic signal.

[0067] Specifically, the transducer drive unit is responsible for converting electrical excitation into ultrasonic pulses. When the amplified pulse acts on the piezoelectric crystal of the transducer, its mechanical vibration forms an ultrasonic front in the fluid medium. The design of the transducer drive unit determines the acoustic energy transmission efficiency, frequency response, and directional characteristics. The ultrasonic pulses emitted by the transducer will pass through the fluid medium, and their propagation characteristics will change with the fluid state, resulting in amplitude attenuation, time delay drift, or spectral distortion, providing a physical basis for subsequent detection.

[0068] In some embodiments, the transducer drive unit converts the electrically excited signal, amplified by the power amplification unit, into an ultrasonic pulse that can propagate in the medium. When a high-amplitude pulse is applied to the piezoelectric crystal inside the transducer, the crystal generates transient mechanical vibrations under electromechanical coupling, thereby forming an ultrasonic front with a specific frequency and directionality in the fluid medium. The structure and matching design of the transducer drive unit determine the acoustic energy transmission efficiency, bandwidth response, and directional distribution, thus affecting the stability of the ultrasonic field. Since the ultrasonic pulse is affected by factors such as fluid density, temperature, and flow velocity during propagation, its amplitude, propagation delay, and spectral structure will change accordingly. These changes provide crucial physical characteristics for subsequent signal detection, state identification, and parameter inversion.

[0069] Specifically, the impedance matching and shaping array is used to ensure that the output impedance of the excitation link is consistent with the input impedance of the transducer, thereby reducing energy reflection, improving power transmission efficiency, and ensuring that the pulse waveform is not distorted due to impedance mismatch. This array can also further correct the pulse rise edge, fall edge, and overall amplitude stability, ensuring that the transmitted pulse maintains good controllability and consistency in both the time and frequency domains, guaranteeing that each excitation has high repeatability and stable characteristics.

[0070] In some embodiments, an impedance matching and shaping array is used to adjust the output impedance of the excitation link to match the input impedance of the transducer, thereby reducing energy reflection at the drive port and improving power transmission efficiency. By configuring a matching network and waveform shaping module in the circuit, this array not only suppresses pulse distortion caused by impedance mismatch but also finely controls the rise time, fall time, and amplitude stability of the excitation pulse, ensuring a fast and smooth transition in the time domain and good energy concentration and consistency in the frequency domain. Through the shaping and compensation effects of this array, standardized ultrasonic pulses with high repeatability, stable morphology, and adapted to transducer characteristics can be obtained for each excitation, providing a reliable excitation reference for subsequent detection processes.

[0071] Specifically, the echo receiving conditioning array includes: a receive switching unit electrically connected to a programmable pulse generating unit, the receive switching unit being configured to switch between excitation and reception; a preamplifier unit electrically connected to the receive switching unit, used to amplify the received weak echo signal to a preset level; a programmable filter unit electrically connected to the preamplifier unit, used to filter out spurious frequencies in the echo signal to improve the signal-to-noise ratio; and an automatic gain control unit electrically connected to the programmable filter unit, used to dynamically adjust the signal gain based on the echo signal amplitude.

[0072] Specifically, the receive switching unit is used to quickly switch between excitation pulse transmission and echo reception to ensure minimal time delay when the system switches between transmission and reception modes, avoiding signal aliasing or missed sampling. This unit can achieve mode switching through high-speed electronic switches or relay circuits, providing a foundation for accurate acquisition of echo signals.

[0073] In some embodiments, the receiving switching unit rapidly switches between pulse excitation transmission and echo reception via a high-speed electronic switch, minimizing mode switching delay and avoiding aliasing of excitation and echo signals. At the same time, it provides a stable receiving path for the preamplifier unit, thereby ensuring that the echo signal can be accurately captured and used for subsequent signal conditioning and analysis.

[0074] Specifically, the preamplifier unit amplifies the received weak echo signal to a level that can be recognized and analyzed by subsequent processing modules. Since echo signals are usually extremely weak and susceptible to noise interference, the linearity, bandwidth, and noise performance of the preamplifier unit directly affect the signal fidelity and detection sensitivity.

[0075] In some embodiments, the preamplifier unit amplifies the weak echo signal output by the receiving switching unit with high fidelity, so that the signal reaches the processing level of the subsequent programmable filtering unit. At the same time, by optimizing linearity, bandwidth and low noise characteristics, the original characteristics of the echo signal are effectively preserved, thereby improving the detection sensitivity and signal reliability of the system.

[0076] Specifically, the programmable filter unit is used to filter out spurious frequencies and interference signals in the echo signal, improving the signal-to-noise ratio. By setting bandpass, low-pass, or high-pass filter parameters, this unit can selectively retain echo signals in specific frequency bands, suppressing environmental noise and non-target frequency interference, providing a clear signal source for subsequent signal analysis.

[0077] In some embodiments, the programmable filtering unit performs frequency-selective processing on the echo signal output by the preamplifier unit. By setting bandpass, lowpass, or highpass filtering parameters, it effectively filters out spurious frequencies and environmental noise, thereby improving the signal-to-noise ratio and ensuring that the subsequent automatic gain control unit and signal processing module can perform feature extraction and anomaly detection based on clear and stable signals.

[0078] Specifically, the automatic gain control unit dynamically adjusts the gain based on the echo signal amplitude to adapt to changes in echo intensity and prevent signal overload or weakness. This unit can maintain the amplitude stability of the output signal under different echo intensities, ensuring a reliable and consistent signal level during data acquisition and digital processing, thereby improving the accuracy and robustness of anomaly detection.

[0079] In some embodiments, the automatic gain control unit dynamically adjusts the signal gain based on the echo signal amplitude output by the programmable filter unit, ensuring that the output signal remains at a stable level regardless of changes in echo intensity. This provides a reliable and consistent input signal to the data acquisition module, improving the accuracy and robustness of subsequent digital processing and anomaly detection.

[0080] Specifically, the data acquisition module is electrically connected to the output of the automatic gain control unit. The data acquisition module includes: a high-speed sampling control unit configured to generate a sampling clock, and further configured to sample the conditioned signal after step-by-step conditioning at a sampling rate not lower than a preset high sampling rate; an analog-to-digital conversion unit electrically connected to the high-speed sampling control unit, configured to perform high-resolution analog-to-digital conversion on the conditioned signal according to the sampling clock to obtain digital sampled data; a buffer unit electrically connected to the analog-to-digital conversion unit, configured to buffer the digital sampled data after analog-to-digital conversion; a data packing unit electrically connected to the buffer unit, configured to timestamp, format, and serialize the digital sampled data in the buffer unit, and generate a digital signal sequence based on the processed digital sampled data; and an acquisition unit configured to control the acquisition window based on a trigger signal from a pulse excitation generator.

[0081] Specifically, when the data packaging unit performs timestamp marking, frame formatting, and serialization processing on the digital sampled data in the buffer unit, the following steps are included: the data packaging unit is further configured to determine the timestamp information of each digital sampled data in the buffer unit based on the system clock; the data packaging unit is further configured to assemble each digital sampled data into a data frame containing a frame header, payload, and check field according to a preset frame based on the timestamp information; the data packaging unit is further configured to perform CRC or parity check and redundancy encoding on the data frame; and the data packaging unit is further configured to serialize the checked data frame.

[0082] Specifically, the data acquisition module, electrically connected to the automatic gain control unit, acquires stable echo signals and samples the conditioning signal at a rate no lower than a preset high sampling rate based on the sampling clock generated by the high-speed sampling control unit, achieving high-precision capture of ultrasonic echo signals in the fluid medium. The analog-to-digital conversion unit performs high-resolution conversion on the sampled analog signals to generate digital sampled data, providing accurate input for subsequent digital processing.

[0083] For example, in practical applications, the data acquisition module can connect to an automatic gain control unit to receive echo signals with dynamically adjusted amplitude. A high-speed sampling control unit samples the signal at a frequency of millions of times per second, thereby capturing weak and transient ultrasonic echoes in the fluid medium. Subsequently, an analog-to-digital converter converts these high-precision sampled analog signals into 16-bit or higher resolution digital data, enabling subsequent signal processing modules to perform pulse shaping, filtering, and feature extraction based on the accurate digital signals, thus achieving reliable detection of fluid state anomalies.

[0084] Specifically, the buffer unit temporarily stores and manages the digital sampling data after analog-to-digital conversion, so that high-speed continuous sampling data can be processed in an orderly manner, avoiding the loss of key information due to processing delays.

[0085] For example, in practical applications, the buffer unit can temporarily store the high-speed digital sampling data generated by the analog-to-digital conversion unit, ensuring that each sampling frame is arranged sequentially. This guarantees that subsequent data packaging and signal processing modules can continuously and stably read the data. Through this buffer management, even if there is a slight delay in signal processing, critical echo information will not be lost, thereby ensuring the real-time performance and reliability of the ultrasonic anomaly detection system in a high-speed sampling environment.

[0086] Specifically, the data packaging unit timestamps, formats, and serializes the digital sampled data in the buffer unit. Based on the system clock, it determines the timestamp of each sampled data point, assembles the data into data frames containing a frame header, payload, and checksum field according to preset frames, and performs CRC or parity checks and redundancy encoding to ensure the integrity and reliability of data transmission. The serialization process transforms the processed data frames into a continuous digital signal sequence, providing a standardized input interface for the signal processing module.

[0087] For example, in practical applications, the data packaging unit can generate timestamps for the digital sampled data stored in the buffer unit according to the system clock, and assemble these sampled data into data frames containing frame headers, payloads, and check fields according to a preset frame format. Simultaneously, CRC checks and redundancy encoding are performed to ensure data integrity. Subsequently, the data frames are serialized into a continuous digital signal sequence for direct reading and analysis by the signal processing module, thereby achieving highly reliable and standardized transmission and processing of the echo signal.

[0088] Specifically, the acquisition unit controls the acquisition window based on the trigger signal from the pulse excitation generator to achieve time alignment of signal acquisition, ensuring that each sampling is synchronized with the excitation pulse, thereby providing an accurate time reference for subsequent signal shaping, feature extraction and anomaly detection.

[0089] For example, in practical applications, the acquisition unit can open the acquisition window based on the trigger signal emitted by the pulse excitation generator, ensuring that high-speed sampling is strictly synchronized with the excitation pulse, thereby guaranteeing that each digital sample precisely corresponds to the ultrasonic pulse emission time. This time alignment mechanism provides a reliable time reference for subsequent pulse shaping, signal accumulation, feature extraction, and anomaly detection, ensuring the accuracy and repeatability of the detection results.

[0090] Specifically, the signal processing module includes: a pulse shaping unit electrically connected to a data packaging unit, wherein the pulse shaping unit is further configured to perform amplitude shaping and edge sharpening processing on the digital signal sequence; the pulse shaping unit is further configured to perform time base alignment of the digital sequence based on an excitation trigger signal; a signal accumulation unit electrically connected to the pulse shaping unit, wherein the signal accumulation unit is configured to accumulate and average the digital signal sequence acquired multiple times; a digital filtering unit electrically connected to the signal accumulation unit, wherein the digital filtering unit is configured to perform bandpass, low-pass, or high-pass filtering on the accumulated digital signal based on a preset digital filter; and an envelope detection and extraction unit and a digital filtering unit. Electrically connected, the envelope detection and extraction unit is configured to perform envelope extraction on the filtered digital signal; the envelope detection and extraction unit is also configured to calculate key feature quantities such as peak value, energy, and time difference; the feature modeling unit is electrically connected to the envelope detection and extraction unit, and the feature modeling unit is configured to extract feature indicators based on envelope feature attenuation, spectral change, and time parameters; the feature modeling unit is also configured to normalize and model the feature indicators; the anomaly judgment unit is electrically connected to the feature modeling unit, and the anomaly judgment unit is configured to compare the modeled feature parameters with preset judgment rules or thresholds to determine whether the fluid medium has an abnormal state.

[0091] Specifically, the signal processing module systematically processes the digital signal sequence from the data acquisition module to extract features and determine anomalies in the ultrasonic echoes within the fluid medium. The pulse shaping unit first performs amplitude shaping and edge sharpening on the digital signal sequence, and then aligns it with the time reference based on the excitation trigger signal, thereby ensuring consistency in both time and amplitude, providing high-quality input for subsequent accumulation and analysis.

[0092] For example, in an ultrasonic fluid monitoring system, the pulse shaping unit of the signal processing module can perform amplitude limiting and rising / falling edge sharpening on the acquired digital echo sequence, making the echo pulse shape clearly distinguishable. It also uses the trigger signal from the pulse excitation generator to precisely align each echo data to the excitation pulse moment, thereby ensuring that the signals of different acquisition cycles are consistent in time and amplitude, providing high-fidelity input for subsequent signal accumulation, filtering and feature extraction.

[0093] Specifically, the signal accumulation unit accumulates and averages the digital signal sequences acquired multiple times to enhance the stability and signal-to-noise ratio of the target echo signal and suppress random noise interference. Subsequently, the digital filtering unit filters the accumulated signal according to a preset bandpass, low-pass, or high-pass filter to further remove interference signals in non-target frequency bands, providing a clean signal basis for feature extraction.

[0094] For example, in an ultrasonic liquid level monitoring system, the signal accumulation unit of the signal processing module can accumulate and average the digital echo sequences acquired multiple times to enhance the stable target echo while suppressing the influence of random noise. Subsequently, the digital filtering unit can remove interference from non-target frequency bands from the accumulated signal according to preset bandpass filtering parameters, ensuring that the filtered signal is clear and has a concentrated spectrum, providing a reliable data foundation for subsequent envelope extraction and feature analysis.

[0095] Specifically, the envelope detection and extraction unit performs envelope extraction on the filtered signal and calculates key feature quantities such as peak value, energy, and time difference, providing quantitative indicators for the feature modeling unit.

[0096] For example, in an ultrasonic fluid monitoring system, the envelope detection and extraction unit can extract the envelope of the digitally filtered echo signal and calculate key features such as signal peak value, energy, and time difference between different echoes. These quantitative indicators can be used as input by the feature modeling unit to analyze signal amplitude attenuation, spectral changes, and propagation time characteristics, thereby providing basic data for fluid state determination and anomaly detection.

[0097] Specifically, the feature modeling unit calculates amplitude attenuation, spectral change and time parameters based on envelope features, and normalizes and models the feature indicators to form standardized feature parameters that can be used to determine the fluid state.

[0098] Specifically, when the feature modeling unit performs normalization and modeling processing on feature indicators, it includes: the feature modeling unit is further configured to perform normalization processing on multiple feature indicators, wherein the normalization processing includes at least one of Min-Max interval normalization, Z-Score standardization, logarithmic transformation, or window normalization; the feature modeling unit is further configured to construct a multi-dimensional feature vector based on the normalized feature indicators; the feature modeling unit is further configured to extract the offset trend of features changing over time in the multi-dimensional feature vector based on sliding window statistics, mean-variance trend analysis, or cumulative sum statistical models; the feature modeling unit is further configured to determine the difference measure between the current feature vector and the reference feature vector according to Euclidean distance, Mahalanobis distance, or dynamic time warping methods; and the feature modeling unit is further configured to model the difference measure and statistical results based on a lightweight decision tree.

[0099] For example, in an ultrasonic fluid monitoring system, the feature modeling unit, based on the amplitude, energy, and time features provided by the envelope detection extraction unit, first performs normalization processing on these feature indicators, such as Min-Max interval normalization, Z-Score standardization, logarithmic transformation, or window normalization, to eliminate the influence of differences in dimensions and amplitudes on subsequent analysis. Subsequently, the normalized feature indicators are combined to construct a multi-dimensional feature vector, ensuring that each acquired signal can be represented in a unified feature space. Based on the multi-dimensional feature vector, the feature modeling unit can further analyze the trend of feature changes over time. For example, through sliding window statistics, mean-variance trend analysis, or cumulative sum statistical models, it extracts information on feature shifts, fluctuations, and abrupt changes over time, thereby reflecting the dynamic changes in the fluid medium state. Simultaneously, the unit can quantify the difference between the current feature vector and the historical reference feature vector based on Euclidean distance, Mahalanobis distance, or dynamic time warping (DTW) methods, providing measurable anomaly offset indicators. Finally, the feature modeling unit uses lightweight decision trees or other fast modeling methods to model the difference measures and statistical trend information, generating standardized feature parameter outputs. This output can be used not only as input to the anomaly detection unit for real-time anomaly detection, but also for historical data analysis, state trend prediction and system optimization, thereby improving the reliability, accuracy and response speed of fluid medium monitoring.

[0100] Specifically, the anomaly judgment unit determines the abnormal state of the fluid medium by comparing the modeled feature parameters with preset judgment rules or thresholds.

[0101] Specifically, when the anomaly judgment unit compares the modeled feature parameters with preset judgment rules or thresholds to determine whether the fluid medium is in an abnormal state, the following steps are included: the anomaly judgment unit is further configured to compare the amplitude of the modeled feature parameters with the preset thresholds to generate a first judgment flag; the anomaly judgment unit is further configured to perform cross-period cumulative statistical calculations on the feature parameters and generate a second judgment flag based on whether the cumulative trend exceeds the limit; the anomaly judgment unit is further configured to analyze the volatility, rate of change, and degree of abrupt change of the feature parameters within a preset sliding window and output a third judgment flag based on the analysis results; the anomaly judgment unit is further configured to perform anomaly pattern recognition on the feature parameters based on a preset rule base and generate corresponding category labels and their confidence levels; the anomaly judgment unit is further configured to perform weighted fusion processing on the first judgment flag, the second judgment flag, the third judgment flag, and the category labels to determine the final anomaly judgment result of the fluid medium.

[0102] Specifically, when the anomaly judgment unit performs weighted fusion processing on the first judgment flag, the second judgment flag, the third judgment flag, and the category label to determine the final anomaly judgment result for the fluid medium, the following steps are included: the anomaly judgment unit is further configured to assign corresponding weights to the first judgment flag, the second judgment flag, the third judgment flag, and the anomaly category label respectively; the anomaly judgment unit is further configured to adaptively adjust the weights based on historical data; the anomaly judgment unit is further configured to obtain the confidence index of each judgment flag; the anomaly judgment unit is further configured to obtain the classification probability of the anomaly category label and obtain the confidence vector based on the classification probability; the anomaly judgment unit is further configured to perform normalization and fusion operations on the confidence vector based on weighted summation and determine the fusion score based on the operation result; the anomaly judgment unit is further configured to compare the fusion score with a configured preset anomaly threshold and determine the final anomaly judgment result according to a preset priority rule in case of conflict.

[0103] For example, in an ultrasonic fluid monitoring system, the anomaly decision unit takes the standardized feature parameters output by the feature modeling unit as input. It first compares these parameters with a preset amplitude threshold to generate a first decision flag, used to quickly identify abnormal signals whose amplitude significantly deviates from the normal range, enabling timely capture of sudden or severe abnormal events. Simultaneously, the anomaly decision unit performs cross-cycle cumulative statistical calculations on the feature parameters. By analyzing the cumulative trend of the features over time, it generates a second decision flag to detect progressive or slowly drifting abnormal behaviors, thereby avoiding misleading judgments caused by occasional fluctuations in a single sampling.

[0104] For example, the anomaly judgment unit analyzes the volatility, rate of change, and magnitude of abrupt changes of characteristic parameters within a preset sliding window, and generates a third judgment flag to capture short-term sudden anomalies or local anomalies. Simultaneously, the anomaly judgment unit combines a preset rule base to perform anomaly pattern recognition on the characteristic parameters, generating anomaly category labels and their corresponding confidence levels through statistical analysis and rule matching. This provides a quantitative basis for anomaly type determination, enabling the differentiation of different types of fluid medium anomalies, such as bubble formation, sediment accumulation, or abnormal flow fluctuations.

[0105] For example, in the fusion decision-making stage, the anomaly judgment unit performs weighted fusion processing on the first, second, and third judgment flags and anomaly category labels. Each judgment flag and category label is assigned a corresponding weight, and the weights can be adaptively adjusted based on historical data and system operating status to optimize the judgment strategy. The unit also obtains the confidence index of each judgment flag and the classification probability of the anomaly category label, and generates a comprehensive fusion score through weighted summation, normalization, and fusion operations. Finally, the anomaly judgment unit compares the fusion score with a preset anomaly threshold, and in the case of conflict or contradiction, determines the final anomaly judgment result according to preset priority rules, achieving a comprehensive, accurate, and robust judgment of the abnormal state of the fluid medium.

[0106] In the above embodiments, by employing a pulse excitation generator to generate an ultrasonic excitation signal with preset pulse width, repetition frequency, and amplitude shaping characteristics, and by amplifying and transducing the excitation signal at the hardware level, this device can significantly improve pulse stability and temporal resolution, thereby enhancing the ability to detect weak disturbances in fluids. Compared to traditional solutions relying on general-purpose circuits, this excitation link can effectively reduce pulse jitter and improve energy consistency, laying a stable signal source foundation for subsequent weak echo detection. Secondly, an echo receiving conditioning array is set up, enabling the device to quickly and without interference switch between excitation and receiving modes, and to perform multi-stage amplification, filtering, and automatic gain control on the echo. Through hardware-based noise suppression and dynamic gain adjustment, this device can obtain a purer, moderately amplitude conditioned signal under low signal-to-noise ratio conditions, significantly improving the ability to capture slight echo distortions caused by micro-leakage, bubble doping, etc. Meanwhile, by using a high-sampling-rate data acquisition module to perform high-speed analog-to-digital conversion on the conditioned signal, the system can accurately record the time-domain details of the echo signal, avoiding time delay measurement errors caused by insufficient sampling frequency in traditional methods. Furthermore, the structured output of the digital signal sequence provides a high-fidelity input foundation for subsequent feature extraction and pattern analysis. In addition, integrating the signal processing module into a dedicated integrated circuit enables key algorithms such as pulse shaping, timing alignment, signal accumulation, digital filtering, envelope extraction, and feature parameter extraction to be executed at the hardware level with low latency. Compared to traditional software-dependent processing methods, this structure significantly reduces processing latency and power consumption, improves anti-interference capabilities, and enables real-time feature analysis even with rapid changes in fluid conditions, thereby enhancing the stability and accuracy of anomaly detection. Finally, the communication power management module allows each module to operate stably in low-power mode and uploads abnormal events and feature data to the early warning PTZ in real time, achieving coordinated operation between edge detection and cloud monitoring. Through centralized power management and communication scheduling, the system's reliability, security, and long-term maintainability are further improved.

[0107] In another preferred embodiment based on the above embodiments, such as Figure 4 As shown, this embodiment provides a pulsed ultrasound method for fluid anomaly detection, including: Step S100: Control the pulse excitation generator to generate an ultrasonic excitation signal with preset pulse width, repetition frequency and amplitude shaping, and amplify and transduce the excitation signal to emit pulses into the fluid medium.

[0108] Step S200: Amplify, filter and automatically adjust the gain of the echo signal from the fluid medium to form a conditioning signal with preset requirements.

[0109] Step S300: Perform analog-to-digital conversion on the conditioning signal at a sampling rate no lower than the preset high sampling rate, and convert the conditioned signal after analog-to-digital conversion into a digital signal sequence.

[0110] Step S400: Perform pulse shaping, timing alignment, signal accumulation, digital filtering, envelope extraction, and feature parameter extraction on the digital signal sequence, and complete the anomaly judgment based on preset logic.

[0111] It is understood that the pulsed ultrasonic sensing device and method for fluid anomaly detection in the above embodiments of the present invention have the same beneficial effects, and will not be described again.

[0112] Those skilled in the art will understand that embodiments of this application can be provided as methods, systems, or computer program goods. Therefore, this application can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, this application can take the form of a computer program goods embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0113] This application is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program goods according to embodiments of this application. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart... Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0114] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0115] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0116] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and not to limit it. Although the present invention has been described in detail with reference to the above embodiments, those skilled in the art should understand that modifications or equivalent substitutions can still be made to the specific implementation of the present invention. Any modifications or equivalent substitutions that do not depart from the spirit and scope of the present invention should be covered within the scope of protection of the claims of the present invention.

Claims

1. A pulsed ultrasonic sensing device for fluid anomaly detection, characterized in that, include: The pulse excitation generator is configured to generate an ultrasonic excitation signal with a preset pulse width, repetition frequency, and amplitude shaping, and then amplify and transduc the excitation signal to emit pulses into the fluid medium. An echo receiving conditioning array is electrically connected to a pulse excitation generator. The echo receiving conditioning array is configured to switch between excitation and reception, and to amplify, filter, and automatically adjust the gain of the echo signal from the fluid medium to form a conditioning signal with preset requirements. The data acquisition module is electrically connected to the echo receiving conditioning array. The data acquisition module is configured to perform analog-to-digital conversion on the conditioning signal at a sampling rate no lower than a preset high sampling rate, and convert the conditioned signal after analog-to-digital conversion into a digital signal sequence. The signal processing module, integrated within an application-specific integrated circuit, is configured to perform pulse shaping, timing alignment, signal accumulation, digital filtering, envelope extraction, and feature parameter extraction on digital signal sequences, and to complete anomaly detection based on preset logic. The communication power management module is electrically connected to the pulse excitation generator, the echo receiving conditioning array, the data acquisition module, and the signal processing module, respectively. The communication power management module is configured to upload abnormal events and characteristic data to the early warning PTZ and provide power management for the pulse excitation generator, the echo receiving conditioning array, the data acquisition module, and the signal processing module.

2. The pulsed ultrasonic sensing device for fluid anomaly detection as described in claim 1, characterized in that, A pulse excitation generator, comprising: The programmable pulse generation unit is configured to generate ultrasonic excitation pulses based on preset pulse width, repetition frequency and amplitude shaping method; The power amplifier unit is electrically connected to the programmable pulse generator unit and is configured to amplify the excitation pulse to the drive level. A transducer drive unit is coupled to the power amplifier unit and is configured to convert the amplified excitation pulse into an ultrasonic pulse and output it to the fluid medium. The impedance matching and shaping array, electrically connected to the programmable pulse generation unit, is configured to match the output impedance of the drive path and maintain the edge and amplitude of the excitation pulse.

3. The pulsed ultrasonic sensing device for fluid anomaly detection as described in claim 2, characterized in that, The echo receiver conditioning array includes: The receive switching unit is electrically connected to the programmable pulse generating unit and is configured to switch between excitation and reception. The preamplifier unit, electrically connected to the receiver switching unit, is used to amplify the received weak echo signal to a preset level. The programmable filter unit, electrically connected to the preamplifier unit, is used to filter out spurious frequencies in the echo signal to improve the signal-to-noise ratio; An automatic gain control unit, electrically connected to a programmable filter unit, is used to dynamically adjust the signal gain based on the echo signal amplitude.

4. The pulsed ultrasonic sensing device for fluid anomaly detection as described in claim 3, characterized in that, The data acquisition module is electrically connected to the output terminal of the automatic gain control unit, wherein the data acquisition module includes: The high-speed sampling control unit is configured to generate a sampling clock, and is also configured to perform sampling on the conditioned signal after step-by-step conditioning at a rate not lower than a preset high sampling rate; The analog-to-digital conversion unit is electrically connected to the high-speed sampling control unit. The analog-to-digital conversion unit is configured to perform high-resolution analog-to-digital conversion on the conditioning signal according to the sampling clock to obtain digital sampling data. The buffer unit is electrically connected to the analog-to-digital conversion unit and is configured to buffer the digital sampled data after analog-to-digital conversion. The data packing unit is electrically connected to the buffer unit. The data packing unit is configured to perform timestamp marking, frame formatting and serialization processing on the digital sampled data in the buffer unit, and generate a digital signal sequence based on the processed digital sampled data. The acquisition unit is configured to control the acquisition window based on the trigger signal from the pulse excitation generator.

5. The pulsed ultrasonic sensing device for fluid anomaly detection as described in claim 4, characterized in that, When the data packaging unit performs timestamp marking, frame formatting, and serialization processing on the digital sampled data in the buffer unit, it includes: The data packaging unit is also configured to determine the timestamp information of each digital sampled data in the buffer unit based on the system clock; The data packaging unit is also configured to assemble each digital sampled data into a data frame containing a frame header, payload, and check field according to a preset frame based on timestamp information; The data packing unit is also configured to perform CRC or parity checks and redundancy encoding on the data frames; The data packing unit is also configured to serialize the verified data frames.

6. The pulsed ultrasonic sensing device for fluid anomaly detection as described in claim 5, characterized in that, The signal processing module includes: The pulse shaping unit is electrically connected to the data packing unit and is also configured to perform amplitude shaping and edge sharpening processing on the digital signal sequence; the pulse shaping unit is also configured to perform time base alignment on the digital sequence based on the excitation trigger signal; The signal accumulation unit is electrically connected to the pulse shaping unit and is configured to accumulate and average multiple acquired digital signal sequences. The digital filtering unit is electrically connected to the signal accumulation unit. The digital filtering unit is configured to perform bandpass, low-pass, or high-pass filtering on the accumulated digital signal based on a preset digital filter. The envelope detection and extraction unit is electrically connected to the digital filtering unit and is configured to perform envelope extraction on the filtered digital signal. The envelope detection and extraction unit is also configured to calculate key feature quantities such as peak value, energy, and time difference. The feature modeling unit is electrically connected to the envelope detection and extraction unit. The feature modeling unit is configured to extract feature indicators based on the amplitude attenuation, spectral change, and time parameters of the envelope features. The feature modeling unit is also configured to perform normalization and modeling processing on the feature indicators. An anomaly determination unit is electrically connected to the feature modeling unit. The anomaly determination unit is configured to compare the feature parameters after modeling with a preset determination rule or threshold to determine whether the fluid medium is in an abnormal state.

7. The pulsed ultrasonic sensing device for fluid anomaly detection as described in claim 6, characterized in that, When the feature modeling unit performs normalization and modeling processing on feature indicators, it includes: The feature modeling unit is also configured to perform normalization on multiple feature metrics, wherein the normalization includes at least one of Min-Max interval normalization, Z-Score standardization, logarithmic transformation or window normalization; The feature modeling unit is also configured to construct a multidimensional feature vector based on the normalized feature index. The feature modeling unit is also configured to extract the offset trend of features changing over time in the multidimensional feature vector based on sliding window statistics, mean-variance trend analysis or cumulative sum statistical model. The feature modeling unit is also configured to determine a measure of the difference between the current feature vector and the reference feature vector based on Euclidean distance, Mahalanobis distance, or dynamic time warping methods. The feature modeling unit is also configured to model the difference measure and statistical results based on a lightweight decision tree.

8. The pulsed ultrasonic sensing device for fluid anomaly detection as described in claim 7, characterized in that, When the anomaly detection unit compares the modeled feature parameters with preset detection rules or thresholds to determine whether the fluid medium is in an abnormal state, it includes: The anomaly decision unit is also configured to compare the amplitude of the modeled feature parameters with a preset threshold to generate a first decision flag; The anomaly judgment unit is also configured to perform cumulative statistical calculations on the feature parameters across periods and generate a second judgment flag based on whether the cumulative trend exceeds the limit; The anomaly judgment unit is also configured to analyze the volatility, rate of change and degree of mutation of the characteristic parameters within a preset sliding window, and output a third judgment flag based on the analysis results. The anomaly detection unit is also configured to perform anomaly pattern recognition on feature parameters based on a preset rule base, and generate corresponding category labels and their confidence levels; The anomaly determination unit is also configured to perform weighted fusion processing on the first determination flag, the second determination flag, the third determination flag, and the category label to determine the final anomaly determination result for the fluid medium.

9. The pulsed ultrasonic sensing device for fluid anomaly detection as described in claim 8, characterized in that, When the anomaly determination unit performs weighted fusion processing on the first determination flag, the second determination flag, the third determination flag, and the category label to determine the final anomaly determination result for the fluid medium, it includes: The anomaly judgment unit is also configured to assign corresponding weights to the first judgment flag, the second judgment flag, the third judgment flag, and the anomaly category label, and the anomaly judgment unit is also configured to adaptively adjust the weights based on historical data; The anomaly decision unit is also configured to obtain the confidence index of each decision flag, and the anomaly decision unit is also configured to obtain the classification probability of the anomaly category label, and obtain the confidence vector based on the classification probability; The anomaly detection unit is also configured to perform normalization and fusion operations on the confidence vector based on weighted summation, and determine the fusion score based on the operation results; The anomaly decision unit is also configured to compare the fusion score with a preset anomaly threshold and determine the final anomaly decision result according to a preset priority rule in case of conflict.

10. A pulsed ultrasonic method for fluid anomaly detection, applicable to a pulsed ultrasonic sensing device for fluid anomaly detection as described in any one of claims 1-9, characterized in that, include: The control pulse excitation generator generates an ultrasonic excitation signal with preset pulse width, repetition frequency, and amplitude shaping, and then amplifies and transduces the excitation signal to emit pulses into the fluid medium; The echo signal from the fluid medium is amplified, filtered, and automatically gain-adjusted to form a conditioning signal with preset requirements; The conditioning signal is converted from analog to digital at a sampling rate no lower than the preset high rate, and the conditioned signal after analog-to-digital conversion is converted into a digital signal sequence. The system performs pulse shaping, timing alignment, signal accumulation, digital filtering, envelope extraction, and feature parameter extraction on digital signal sequences, and completes anomaly detection based on preset logic.

Images