Pipeline online ultrasonic particle size distribution inversion method and system

Abstract

The present application relates to the technical field of online ultrasonic particle size measurement, and provides a pipeline online ultrasonic particle size distribution inversion method and system, comprising: adopting Tikhonov regularization and interior point method coupling inversion, constructing a constraint optimization objective function between a measured attenuation spectrum vector and a theoretical attenuation matrix and solving, to obtain a particle size distribution and characteristic particle size parameters; adopting one or more center frequency ultrasonic transducers for separate or combined measurement, performing hierarchical configuration on the transducer spacing, to realize multi-frequency attenuation measurement and particle size distribution inversion of a particle system of tens of nanometers to thousands of nanometers; integrating a flow channel module and a transducer module on an industrial pipeline, so that the sample to be measured passes through the measurement area in a continuous flow state, and combining a turbulence prevention structure and a transducer installation mode optimization, to realize online continuous measurement of the industrial pipeline. The present application solves the problems of low inversion calculation efficiency, insufficient detection sensitivity of the nanometer particle system and difficulty in applying to continuous online measurement of the industrial pipeline in ultrasonic particle size measurement.

Description

Technical Field

[0001] This invention relates to the technical field of online ultrasonic particle size measurement, and more particularly to a method and system for online ultrasonic particle size distribution inversion in pipelines. Background Technology

[0002] Particle size and its distribution are crucial parameters affecting material properties, product quality, and process stability, and have significant application value in fields such as new energy materials, electronic pastes, nanopowders, pigments and coatings, biomedicine, food, and chemical production. In particular, real-time and accurate measurement of particle size distribution is essential for product development, process control, online monitoring, and quality evaluation in nanoparticle suspension systems and continuously conveyed slurry systems.

[0003] Currently, particle size measurement methods mainly include laser scattering, image analysis, and ultrasonic particle size measurement.

[0004] Laser scattering is one of the most widely used particle size analysis techniques. It inverts the particle size distribution by measuring the scattering characteristics of particles to laser light. This method has advantages such as fast measurement speed and mature application. However, the measurement accuracy is easily affected by high-concentration suspension systems, multiple scattering systems, and opaque media. It usually requires sample dilution, making it difficult to achieve in-situ online measurement.

[0005] Image analysis methods acquire particle images using microscopes, CCD cameras, or electron microscopes, and then calculate particle size using image processing algorithms. While this method can intuitively obtain particle morphology information, it suffers from limitations such as limited sampling volume, insufficient statistical representativeness, complex sample preparation, and difficulties in online measurement, particularly restricting its application in nanoscale particles and high-concentration systems.

[0006] Ultrasonic particle size distribution (UPD) utilizes the scattering, absorption, and thermoviscous loss characteristics of ultrasonic waves by particles to invert particle size distribution by measuring the ultrasonic attenuation spectrum or sound velocity spectrum. Because ultrasonic waves can penetrate high-concentration, opaque media, they offer advantages such as no need for dilution, suitability for online detection, ease of continuous measurement, and applicability to industrial process monitoring, and have attracted widespread attention in recent years. Ultrasonic particle size distribution technology shows promising application prospects, particularly in scenarios such as high-concentration slurries, nanoparticle suspensions, and online monitoring of industrial pipelines.

[0007] However, existing ultrasonic particle size analysis techniques still have the following shortcomings: (1) Particle size inversion calculation is complex and lacks real-time performance.

[0008] Existing ultrasonic particle size inversion methods typically employ gradient-free optimization techniques such as Genetic Algorithm (GA) and Particle Swarm Optimization (PSO) to solve for particle size distribution. These methods require numerous iterative searches, resulting in high computational complexity, slow convergence speed, and high processor performance requirements, making them unsuitable for real-time online measurement needs.

[0009] (2) The ability to measure nanoparticles is limited.

[0010] For nanoparticle systems with a diameter less than 100 nm, the ultrasonic signal amplitude attenuation sensitivity is low due to the significantly weakened particle scattering effect, resulting in insufficient inversion discernibility. Furthermore, the transducer frequency range, acoustic path design, and inversion models of existing measurement systems are typically optimized for micrometer-sized particles, and their detection sensitivity and inversion stability are insufficient for particles with diameters of tens of nanometers or even smaller, limiting the application of ultrasonic particle size analysis in the field of nanomaterials.

[0011] (3) Existing online measurement devices cannot meet the requirements of miniaturization and online pipeline integration.

[0012] While existing online systems can achieve online detection, some devices are large in size and have complex installation structures, usually requiring a large installation space or supporting auxiliary structures, which is not conducive to direct deployment in existing industrial pipelines.

[0013] On the other hand, some miniaturized measuring devices use an insertion probe structure, which requires the measuring probe to be installed inside the pipe, which may increase the difficulty of installation and maintenance and affect the flow field.

[0014] Existing technologies still struggle to balance miniaturization, ease of installation, and compatibility with industrial pipelines, lacking a compact ultrasonic particle size measurement device suitable for long-term online measurement of industrial pipelines.

[0015] Therefore, there is an urgent need to propose an online ultrasonic particle size distribution inversion method and system for pipelines, which can improve particle size inversion efficiency and enhance the detection capability of nanoparticle systems while ensuring inversion accuracy, and realize the miniaturization of measuring devices and online integration of industrial pipelines to meet the needs of online monitoring of industrial processes. Summary of the Invention

[0016] To address the shortcomings of existing ultrasonic particle size distribution measurement technologies, such as low inversion calculation efficiency, insufficient detection sensitivity of nanoparticle systems, and difficulty in applying them to continuous online measurement of industrial pipelines, this invention aims to provide an online ultrasonic particle size distribution inversion method and system for pipelines. Through optimization of the inversion algorithm, multi-frequency ultrasonic measurement parameters, and pipeline flow channel structure, real-time online measurement and stable acquisition of nanoparticle size distribution are achieved.

[0017] The above-mentioned objective of this invention is achieved through the following technical solutions: A method for online ultrasonic particle size distribution inversion in pipelines includes the following steps: The particle size distribution and characteristic particle size parameters are obtained by coupling inversion with Tikhonov regularization and interior point method, constructing and solving a constrained optimization objective function between the measured attenuation spectrum vector and the theoretical attenuation matrix; One or more ultrasonic transducers with center frequencies are used for individual or combined measurements, and the transducer spacing is configured in a graded manner to achieve multi-frequency attenuation measurement and particle size distribution inversion of particle systems ranging from tens of nanometers to thousands of nanometers. By integrating the flow channel module and transducer module onto the industrial pipeline, the sample under test can pass through the measurement area in a continuous flow state. Combined with the optimization of the anti-turbulence structure and transducer installation method, online continuous measurement of industrial pipelines can be realized.

[0018] Furthermore, the process of using Tikhonov regularization coupled with the interior point method to obtain particle size distribution and characteristic particle size parameters specifically includes: The ultrasonic attenuation inversion problem is constructed as an ill-conditioned inverse problem described by the first type of Fredholm integral equation. A mapping relationship between particle size distribution and multi-frequency attenuation response is established using a theoretical acoustic scattering model combined with heat dissipation and viscous loss mechanisms. Tikhonov regularization is introduced to superimpose the smoothness constraint of the solution in the least squares objective to suppress numerical oscillations and non-physical oscillations of the solution caused by measurement noise and model errors. At the same time, physical prior conditions such as the non-negativity constraint of particle size distribution are introduced to transform the inversion problem into a constrained convex optimization problem. Then, the interior point method is used to iteratively approximate the global optimum along the gradient direction within the feasible region, so that the inversion process can achieve fast convergence and stable solution while maintaining the accuracy of the solution.

[0019] Furthermore, the coupled inversion using Tikhonov regularization and interior point method, by constructing and solving a constrained optimization objective function between the measured attenuation spectrum vector and the theoretical attenuation matrix, yields the particle size distribution and characteristic particle size parameters, specifically including: Transmission signals of the sample were acquired at multiple ultrasonic center frequencies, and the ultrasonic attenuation coefficients at the corresponding frequencies were calculated to construct the measured attenuation spectrum vector. ; Based on the particle acoustic scattering model, and combined with the heat dissipation model and viscous loss model, a multi-parameter theoretical attenuation response matrix is ​​established. ,in Indicates the measured frequency. Indicates discrete particle size range, Indicates the relevant physical parameters of the material; The theoretical attenuation matrix is ​​pre-calculated, discretized, and stored in the parameter database, and can be retrieved by indexing according to the operating conditions during the measurement process; The ultrasonic attenuation inversion problem is modeled as a first-type Fredholm integral discretization inverse problem. A Tikhonov regularization method is introduced to stabilize the original least squares problem. The objective function is as follows:

[0020] in, This is based on the multi-parameter theoretical decay matrix obtained from pre-calculated database calls. This is the measured attenuation spectral vector. Let be the particle size distribution vector to be determined. For the difference operator constrained by particle size distribution smoothing, This is the regularization coefficient.

[0021] Furthermore, the coupled inversion using Tikhonov regularization and interior point method, which constructs and solves a constrained optimization objective function between the measured attenuation spectrum vector and the theoretical attenuation matrix to obtain the particle size distribution and characteristic particle size parameters, also includes: Introducing nonnegativity constraints on particle size distribution into the constrained optimization model This transforms the problem into a constrained convex optimization problem. The original-dual interior point method is used to solve the problem. By constructing a barrier function, the non-negativity constraint is transformed into a logarithmic barrier term, and a nonlinear equation system is established in combination with KKT conditions. During the iteration process, the original and dual variables are updated simultaneously, and the search direction is solved based on gradient information and the Hessian matrix. The particle size distribution vector is updated iteratively step by step. Continue until the convergence condition is met; The final output is the particle size distribution curve, and the characteristic particle size parameters of Dv10, Dv50, and Dv90 are calculated.

[0022] Furthermore, the method of using one or more ultrasonic transducers with different center frequencies for individual or combined measurements, and configuring the transducer spacing in a graded manner to achieve multi-frequency attenuation measurement and particle size distribution inversion, specifically includes: Based on the differentiated sensitivity of different frequencies to particle size, one or more center frequencies are selected for individual or combined measurements according to the target particle size range to expand the coverage of particle size detection and improve the resolution of different particle size intervals. At the same time, the corresponding sound path configuration is matched according to the attenuation strength characteristics corresponding to the particle size, so that the attenuation spectrum maintains a detectable dynamic range and measurement sensitivity in different particle size intervals.

[0023] Furthermore, the use of one or more ultrasonic transducers with different center frequencies for individual or combined measurements to achieve multi-frequency attenuation measurement and particle size distribution inversion of particle systems ranging from tens of nanometers to thousands of nanometers specifically includes: The measurement system uses one or more ultrasonic transducers with center frequencies for individual or combined measurements. The center frequencies include one or more of 10MHz, 20MHz, 30MHz, 50MHz, 80MHz, 100MHz, 150MHz and 200MHz, thereby realizing multi-frequency attenuation measurement of particle systems ranging from tens of nanometers to thousands of nanometers. Based on the target particle size range, one or more center frequencies are selected for individual or combined measurements. Different frequencies correspond to different particle scattering sensitive ranges, thereby improving the recognition and resolution capabilities of nanoparticles of different sizes. For nanoparticle systems, one or more high-frequency transducers of 50MHz, 80MHz, 100MHz, 150MHz and 200MHz are used for individual or combined measurements. The ultrasonic signal enters the medium under test through the transmitting transducer. After being scattered, dissipated by heat and viscous loss by the particles, the transmitted signal is received by the receiving transducer. The system calculates the attenuation coefficient at each measurement frequency and constructs the measured attenuation spectrum vector. ; Based on the measurement frequency range, material parameters, and target particle size range, the corresponding pre-calculated theoretical attenuation matrix is ​​retrieved from the theoretical attenuation matrix database. The particle size distribution is then input into the particle size distribution inversion model for solution, thereby obtaining the particle size distribution results.

[0024] Furthermore, the step-by-step configuration of transducer spacing to achieve multi-frequency attenuation measurement and particle size distribution inversion for particle systems ranging from tens of nanometers to thousands of nanometers specifically includes: The transducer spacing is designed in stages according to the target nanoparticle size, ranging from 0.5 mm to 100 mm, to match the acoustic propagation attenuation characteristics under different nanoparticle systems. The transducer spacing is available in three configurations: short path, medium path, and long path. Short sound path of one of 0.5mm, 1mm, 1.5mm, 2mm, 2.5mm, 3mm, 3.5mm, 4mm, 4.5mm, and 5mm, used for measurement of particle systems in the tens of nanometer scale; The medium path length is one of 5mm, 10mm, 15mm, 20mm, 25mm, or 30mm, used for the measurement of particle systems in the hundreds of nanometer scale; The long path length is one of 30mm, 40mm, 50mm, 60mm, 70mm, 80mm, 90mm, or 100mm, used for the measurement of nanometer-scale particle systems; By coordinating the configuration of measurement frequency combinations, acoustic path structure, and theoretical attenuation matrix parameters, continuous particle size distribution measurement of particle systems ranging from tens of nanometers to thousands of nanometers can be achieved.

[0025] Furthermore, the integration of the flow channel module and transducer module onto the industrial pipeline allows the sample to pass through the measurement area in a continuous flow state. Combined with optimized anti-turbulence structures and transducer installation methods, this enables online continuous measurement of the industrial pipeline, specifically including: The ultrasonic measurement channel is integrated with the industrial delivery pipeline, enabling ultrasonic measurement of the sample during continuous flow. Depending on the operating conditions, either direct measurement via the main flow path or a branch flow approach is used to match the installation conditions and flow states of different industrial sites. The channel module uses standardized interfaces for connection and maintenance with existing pipeline systems. A flow field rectification structure is set up within the measurement area to suppress interference from turbulence and bubbles on ultrasonic propagation, ensuring that the ultrasonic propagation path traverses the stable flow region within the channel. Furthermore, the transducer acoustic field morphology is adapted according to measurement requirements to ensure stable acquisition and repeatability of the ultrasonic attenuation spectrum under flow conditions.

[0026] Furthermore, the integration of the flow channel module and the transducer module onto the industrial pipeline, enabling the sample to pass through the measurement area in a continuous flow state, specifically includes: The measuring device includes a flow channel module and a transducer module. The flow channel module includes a sample inlet, a measuring chamber, and a sample outlet. The sample to be tested passes through the measuring area in a continuous flow state, realizing online real-time measurement. The flow channel structure is a direct-connection pipe structure, which is directly integrated into the main industrial conveying pipeline to realize in-situ online measurement, so that the ultrasonic measurement path is synchronized with the industrial fluid conveying process; Alternatively, a bypass measurement structure can be set up to introduce the sample to be tested through a diversion method, so as to reduce the impact of mainstream disturbance on the measurement signal or adapt to complex flow conditions, thereby improving measurement stability and repeatability; The flow channel module is connected to the existing conveying pipeline through a standard industrial pipeline connection structure. The pipeline connection structure includes one or more of the following: standard flange connection structure, quick-connect clamp connection structure, and quick-connect coupling connection structure, so as to realize rapid installation, disassembly and maintenance with different industrial pipeline systems. Among them, the standard flange connection structure is used for long-term stable online installation; the quick-connect clamp connection structure is suitable for applications that require frequent disassembly or sanitary applications, and features convenient installation, reliable sealing, and easy cleaning and maintenance; the quick-connect coupling structure is used to realize the rapid docking and replacement between the measurement module and the pipeline system, so as to reduce system maintenance time.

[0027] Furthermore, the optimization of the anti-turbulence structure and transducer installation method enables continuous online measurement of industrial pipelines, specifically including: The transducer transmitter and receiver are respectively located on both sides of the measurement cavity, so that the ultrasonic propagation path passes through the stable flow area in the center of the flow channel, thereby reducing the influence of boundary reflection and flow disturbance on the measurement signal; An anti-turbulence structure is installed inside the measurement cavity to improve the flow field stability within the measurement area, enabling the measured medium to form a relatively stable flow state in the ultrasonic propagation path region. This reduces the impact of flow velocity fluctuations, particle agglomeration, and bubble entrainment on the measurement results, thereby improving the stability and repeatability of ultrasonic attenuation measurement. The anti-turbulence structure is installed upstream of the measurement area, inside the measurement area, or downstream of the measurement area, depending on the fluid characteristics, pipe diameter, and operating conditions. The transducer can be a focused transducer or a non-focused transducer. The non-focused transducer is used to form an approximately parallel or diffuse sound field, while the focused transducer is used to increase the local sound energy density, thereby enhancing the signal strength and measurement sensitivity. During the measurement process, the sample under test passes through the detection area in a continuous flow state. The transducer continuously emits and receives ultrasonic signals. The system obtains attenuation spectrum information at different frequencies and uses it for subsequent particle size distribution inversion processing, thereby realizing online continuous measurement of industrial pipelines.

[0028] An online ultrasonic particle size distribution inversion system for performing the above-described online ultrasonic particle size distribution inversion method for pipelines includes: The regular inversion module is used to perform inversion by coupling Tikhonov regularization with the interior point method. It obtains the particle size distribution and characteristic particle size parameters by constructing and solving a constrained optimization objective function between the measured attenuation spectrum vector and the theoretical attenuation matrix. The multi-frequency measurement module is used to perform individual or combined measurements using one or more ultrasonic transducers with one center frequency, and to configure the transducer spacing in a graded manner to achieve multi-frequency attenuation measurement and particle size distribution inversion of particle systems ranging from tens of nanometers to thousands of nanometers. The flow channel integration module is used to integrate the flow channel module and the transducer module onto industrial pipelines, allowing the sample to pass through the measurement area in a continuous flow state. Combined with the optimization of the anti-turbulence structure and transducer installation method, it enables online continuous measurement of industrial pipelines.

[0029] A computer device includes a memory and one or more processors, the memory storing computer code that, when executed by the one or more processors, causes the one or more processors to perform the method described above.

[0030] A computer-readable storage medium storing computer code that, when executed, performs the method described above.

[0031] Compared with the prior art, the present invention has at least one of the following beneficial effects: (1) This invention achieves the organic unity of rapid and stable inversion of particle size distribution, high-sensitivity measurement of nanoscale particles and online continuous monitoring of industrial pipelines through the synergistic cooperation of inversion algorithm optimization, multi-frequency sound path classification measurement and pipeline flow channel integration.

[0032] Specifically, the invention employs constrained Tikhonov regularization coupled with the interior-point method for inversion, significantly improving computational efficiency and inversion stability while ensuring the physical rationality of the solution. It utilizes a wideband multi-frequency combination and a three-stage sound path classification configuration to significantly enhance the detection sensitivity of particles in the tens of nanometers range and achieve continuous measurement over a wide range from tens to thousands of nanometers. Through a direct-connection pipeline structure integration and anti-turbulence design, it enables in-situ online continuous measurement and improves measurement stability and repeatability in high-concentration flow systems. The synergistic effect of these three elements fundamentally solves the complex problems in existing technologies, such as the difficulty in balancing inversion computational efficiency and accuracy, insufficient nanoparticle detection sensitivity, and poor industrial online adaptability. This invention meets the requirements for industrial online applications in terms of algorithm efficiency, measurement capability, and engineering applicability.

[0033] (2) Significantly improves the detection sensitivity of nanoparticles and realizes the measurement of continuous particle size distribution over a wide range.

[0034] This invention addresses the challenges of weak scattering effects and low sensitivity of attenuated signals in nanoparticle systems. It employs one or more 50MHz–200MHz high-frequency transducers, either individually or in combination, to enhance the scattering and thermoviscous response of ultrasound to particles in the tens of nanometers. Simultaneously, it utilizes one or more 10MHz–200MHz wideband multi-frequency combination schemes, combined with a three-level acoustic path configuration (short, medium, and long path), to match the sound propagation distance with the attenuation characteristics of different particle size ranges. Combined with the collaborative access to a theoretical attenuation matrix database, this invention enables continuous and accurate particle size distribution measurement of systems ranging from tens of nanometers to thousands of nanometers, while also improving the signal-to-noise ratio and detectability of weak attenuated signals.

[0035] (3) Realize in-situ online continuous measurement of industrial pipelines to improve measurement stability and repeatability.

[0036] This invention employs a direct-connection pipeline structure to directly embed the measurement unit into the main industrial conveying pipeline. The ultrasonic measurement path operates synchronously with the industrial fluid conveying, eliminating the need for sampling, dilution, or offline analysis, thus achieving in-situ online real-time continuous measurement. Simultaneously, an anti-turbulence structure is installed within the measurement chamber to effectively suppress fluid turbulence, local eddies, and bubble disturbances, ensuring a relatively stable flow state of the measured medium in the ultrasonic propagation path region. This reduces the interference of flow velocity fluctuations, particle agglomeration, and bubble entrainment on the measurement results, significantly improving the measurement stability and repeatability under high-concentration slurries and continuous conveying systems.

[0037] (4) The device has a compact structure, is compatible with various industrial pipeline connection methods, and takes into account the flexibility of bypass measurement.

[0038] The flow channel module of this invention adopts a standard industrial pipeline connection structure (flange connection, quick-connect clamp connection, quick-connect coupling connection), which can be quickly integrated into existing industrial pipeline systems. It meets the needs of different scenarios such as long-term stable installation, frequent disassembly and maintenance, and sanitary applications. The device is small in size and compact in structure, without the need for major modifications to existing pipelines, thus reducing deployment difficulty and installation and maintenance costs. In addition, when it is necessary to further reduce mainstream disturbance or adapt to complex flow conditions, it can be flexibly switched to a bypass measurement structure to introduce the sample to be tested through a diversion method, thereby improving the system's environmental adaptability and measurement accuracy. Attached Figure Description

[0039] Figure 1 This is an overall flowchart of the pipeline online ultrasonic particle size distribution inversion method of the present invention; Figure 2 This is a flowchart of the particle size inversion method based on Tikhonov regularization and interior point method of the present invention. Figure 3 This is a schematic diagram of the online ultrasonic particle size measurement device for pipelines according to the present invention; Figure 4 This is a schematic diagram of the anti-turbulence structure of the present invention; Figure 5 This is a diagram showing the overall structure of the pipeline online ultrasonic particle size distribution inversion system of the present invention.

[0040] Figure Labels 1: Inlet; 2: Flange / Clamp / Quick Connector; 3: Antenna; 4: Main Unit; 5: Transmitting Transducer; 6: Receiving Transducer; 7: Outlet; 8: Pipe wall; 9: Anti-turbulence structure; 10: Transducer connection structure. Detailed Implementation

[0041] To make the objectives, technical solutions, and advantages of the embodiments of this application clearer, the technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0042] Those skilled in the art will understand that, unless specifically stated otherwise, the singular forms “a,” “an,” “the,” and “the” used herein may also include the plural forms. It should be further understood that the term “comprising” as used in this specification means the presence of the stated features, integers, steps, operations, elements, and / or components, but does not exclude the presence or addition of one or more other features, integers, steps, operations, elements, components, and / or groups thereof.

[0043] The improvements involved in this invention are as follows: (1) To address the problems of large computational load and slow convergence speed of traditional gradient-free optimization methods such as genetic algorithm (GA) and particle swarm optimization (PSO), this invention constructs an inversion model between ultrasonic attenuation spectrum and particle size distribution, and uses the Tikhonov regularization method to establish a constrained inversion objective function to improve the stability and noise resistance of solving ill-conditioned inverse problems.

[0044] Since ultrasonic particle size distribution inversion is a typical ill-conditioned inverse problem, measurement noise and model errors can easily lead to oscillations, multiple peaks, or instability in the inversion results. Tikhonov regularization, by introducing a regularization term into the objective function, constrains the smoothness and stability of the particle size distribution solution, thereby suppressing noise amplification while ensuring data fitting accuracy and improving the stability and robustness of the inversion results.

[0045] Furthermore, the Interior Point Method is introduced as a constraint optimization solver to solve the non-negativity constraints, normalization constraints, and other physical constraints of the particle size distribution. Compared with traditional heuristic optimization algorithms, the Interior Point Method can fully utilize the gradient information of the objective function, iteratively search for the optimal solution within the feasible region, achieve fast convergence and stable solution, thereby reducing the computation time required for particle size distribution inversion.

[0046] This invention introduces the interior point method in the ultrasonic particle size distribution inversion process, combines the ultrasonic attenuation model and the Tikhonov regularization method to establish a particle size distribution inversion model, and uses the interior point method for optimization solution. While ensuring the inversion accuracy and result stability, it improves the computational efficiency and real-time performance.

[0047] (2) Ultrasonic measurement scheme for small-diameter nanoparticles To address the problem of insufficient sensitivity of existing ultrasonic particle size analyzers for measuring ultrafine nanoparticles, this invention proposes a high-sensitivity ultrasonic measurement scheme.

[0048] By optimizing the ultrasonic frequency range and adopting a wideband ultrasonic measurement mode to cover ultrasonic excitation conditions from low to high frequencies, such as single-frequency, multi-frequency, or frequency combination measurement methods, the response sensitivity of ultrasonic signals to nanoparticles is improved, and the attenuation characteristics caused by particles are enhanced.

[0049] Preferably, the ultrasonic measurement is performed using transducers with different center frequencies, including one or more of 10MHz, 20MHz, 30MHz, 50MHz, 80MHz, 100MHz, 150MHz, and 200MHz. Each transducer operates within its corresponding effective bandwidth, and one or more center frequencies can be selected individually or in combination according to the target particle size range to adapt to the attenuation response characteristics of particle systems of different sizes, thereby improving measurement sensitivity and applicability.

[0050] Meanwhile, the transducer spacing and acoustic propagation path length are optimized according to the target particle size range and concentration conditions, so that the effective propagation distance of ultrasound in the sample matches the attenuation characteristics caused by particles, thereby improving the detectability and signal-to-noise ratio of weak attenuation signals under different concentrations and particle sizes.

[0051] Preferably, the transducer spacing is configured in three ways: short spacing, medium spacing, and long spacing, depending on the measurement frequency and the target particle size range. Preferably, the short spacing is one of 0.5, 1, 1.5, 2, 2.5, 3, 3.5, 4, 4.5, or 5 mm; the medium spacing is one of 5, 10, 15, 20, 25, or 30 mm; and the long spacing is one of 30, 40, 50, 60, 70, 80, 90, or 100 mm.

[0052] (3) Optimization design of flow channel structure for online measurement of industrial pipelines To address the problems of particle sedimentation, uneven flow, and local blockage that easily occur in high-concentration suspended systems during industrial pipeline transportation, this invention designs an ultrasonic measurement flow channel structure suitable for online inspection of industrial pipelines.

[0053] The flow channel structure includes a sample inlet, a measuring chamber, and a sample outlet. The sample to be tested passes directly through the measuring area during the continuous flow of the industrial conveying pipeline, realizing in-situ online detection and real-time measurement.

[0054] Preferably, the flow channel structure adopts a direct-connection pipeline integration method, which allows the measurement unit to be directly embedded in the industrial main pipeline, so as to realize the synchronous operation of the ultrasonic measurement path and the fluid transportation process, thereby meeting the needs of continuous online monitoring.

[0055] In some operating conditions, to improve measurement stability and reduce the impact of mainstream disturbances, a bypass measurement structure can be used to introduce the sample to be measured through a diversion method, so that the measurement can be completed under relatively stable flow conditions.

[0056] Preferably, the flow channel module is connected to the existing conveying pipeline through a standard industrial pipeline connection structure. The connection structure includes one or more of the following: flange connection structure, quick-connect clamp connection structure, and quick-connect coupling connection structure, so as to realize the rapid integration and maintenance of the industrial pipeline system.

[0057] By optimizing the flow channel structure design, the measurement stability and anti-interference ability under high concentration and continuous flow systems are improved, providing a structural foundation for pipeline-based online ultrasonic particle size distribution measurement.

[0058] First Embodiment like Figure 1 As shown, this embodiment provides a method for online ultrasonic particle size distribution inversion in pipelines, including: (1) Tikhonov regularization coupled with interior point method inversion The ultrasonic attenuation inversion problem is constructed as an ill-conditioned inverse problem described by the first type of Fredholm integral equation. A mapping relationship between particle size distribution and multi-frequency attenuation response is established using a theoretical acoustic scattering model combined with heat dissipation and viscous loss mechanisms. Tikhonov regularization is introduced to superimpose the smoothness constraint of the solution in the least squares objective to suppress numerical oscillations and non-physical oscillations of the solution caused by measurement noise and model errors. At the same time, physical prior conditions such as the non-negativity constraint of particle size distribution are introduced to transform the inversion problem into a constrained convex optimization problem. Then, the interior point method is used to iteratively approximate the global optimum along the gradient direction within the feasible region, so that the inversion process can achieve fast convergence and stable solution while maintaining the accuracy of the solution.

[0059] like Figure 2 As shown, this embodiment employs a coupled inversion method using Tikhonov regularization and interior point method. By constructing and solving a constrained optimization objective function between the measured attenuation spectrum vector and the theoretical attenuation matrix, the particle size distribution and characteristic particle size parameters are obtained, specifically including: Transmission signals of the sample were acquired at multiple ultrasonic center frequencies, and the ultrasonic attenuation coefficients at the corresponding frequencies were calculated to construct the measured attenuation spectrum vector. ; Based on the particle acoustic scattering model, and combined with the heat dissipation model and viscous loss model, a multi-parameter theoretical attenuation response matrix is ​​established. ,in Indicates the measured frequency. Indicates discrete particle size range, Indicates the relevant physical parameters of the material; The theoretical attenuation matrix is ​​pre-calculated, discretized, and stored in the parameter database, and can be retrieved by indexing according to the operating conditions during the measurement process; The ultrasonic attenuation inversion problem is modeled as a first-type Fredholm integral discretization inverse problem. Since this type of problem typically exhibits ill-posedness and solution instability, direct solutions are easily affected by measurement noise amplification, leading to oscillations or non-physical solutions in the particle size distribution. Therefore, the Tikhonov regularization method is introduced to stabilize the original least squares problem. The objective function is as follows:

[0060] in, This is based on the multi-parameter theoretical decay matrix obtained from pre-calculated database calls. This is the measured attenuation spectral vector. Let be the particle size distribution vector to be determined. For the difference operator constrained by particle size distribution smoothing, This is the regularization coefficient.

[0061] Introducing nonnegativity constraints on particle size distribution into the constrained optimization model This transforms the problem into a constrained convex optimization problem. The original-dual interior point method is used to solve the problem. By constructing a barrier function, the non-negativity constraint is transformed into a logarithmic barrier term, and a nonlinear equation system is established in combination with KKT conditions. During the iteration process, the original and dual variables are updated simultaneously, and the search direction is solved based on gradient information and the Hessian matrix. The particle size distribution vector is updated iteratively step by step. Continue until the convergence condition is met; The final output is the Particle Size Distribution (PSD) curve, and the characteristic particle size parameters of Dv10, Dv50, and Dv90 are calculated.

[0062] In practical applications of ultrasonic particle size inversion, the measured ultrasonic attenuation spectrum and the particle size distribution to be determined constitute a typical ill-conditioned inverse problem. The core challenge lies in two aspects: firstly, noise and errors are unavoidable during measurement, and directly solving the inversion equation leads to non-physical phenomena such as severe oscillations, multiple peaks, or even negative values ​​in the particle size distribution results, significantly deviating from the true distribution; secondly, industrial online measurements have strict requirements for the speed of inversion calculations, and traditional heuristic algorithms relying on extensive iterative searches are insufficient to meet real-time demands. To address these issues, this invention transforms the particle size inversion problem into a constrained convex optimization problem. Specifically, by introducing regularization constraints to reasonably limit the smoothness of the particle size distribution solution, the noise amplification effect is effectively suppressed while ensuring the accuracy of fitting the measured data, achieving a good balance between numerical stability and physical rationality in the inversion results; simultaneously, non-negativity constraints are introduced to ensure that the inverted particle size distribution does not exhibit physically unrealizable negative values. In terms of solution strategy, this invention employs the interior-point method as the core optimization solver. It iteratively approaches the optimal solution within the feasible region, fully utilizing the gradient and second derivative information (Hessian matrix) of the objective function in each iteration to determine the optimal search direction, unlike traditional genetic algorithms or particle swarm optimization algorithms which rely on extensive random searches and trial and error. This allows the inversion process to converge with far fewer iterations than traditional methods, significantly reducing computation time and meeting the stringent real-time requirements of industrial environments. Through the synergistic combination of regularized modeling and efficient interior-point solution, this invention achieves stable and rapid acquisition of accurate particle size distribution curves and characteristic parameters such as Dv10, Dv50, and Dv90 in noisy environments, providing reliable inversion calculation support for industrial online ultrasonic particle size measurement.

[0063] (2) Measurement of particles ranging from tens of nanometers to thousands of nanometers has been achieved. Based on the differentiated sensitivity of different frequencies to particle size, one or more center frequencies are selected for individual or combined measurements according to the target particle size range to expand the coverage of particle size detection and improve the resolution of different particle size intervals. At the same time, the corresponding sound path configuration is matched according to the attenuation strength characteristics corresponding to the particle size, so that the attenuation spectrum maintains a detectable dynamic range and measurement sensitivity in different particle size intervals.

[0064] Using one or more ultrasonic transducers with different center frequencies for individual or combined measurements, and with graded configuration of transducer spacing, multi-frequency attenuation measurement and particle size distribution inversion of particle systems ranging from tens of nanometers to thousands of nanometers are achieved. Specifically, this includes: The measurement system uses one or more ultrasonic transducers with center frequencies for individual or combined measurements. The center frequencies include one or more of 10MHz, 20MHz, 30MHz, 50MHz, 80MHz, 100MHz, 150MHz and 200MHz, thereby realizing multi-frequency attenuation measurement of particle systems ranging from tens of nanometers to thousands of nanometers. Based on the target particle size range, one or more center frequencies are selected for individual or combined measurements. Different frequencies correspond to different particle scattering sensitive ranges, thereby improving the recognition and resolution capabilities of nanoparticles of different sizes. For nanoparticle systems, one or more high-frequency transducers of 50MHz, 80MHz, 100MHz, 150MHz and 200MHz are used for individual or combined measurements. The ultrasonic signal enters the medium under test through the transmitting transducer. After being scattered, dissipated by heat and viscous loss by the particles, the transmitted signal is received by the receiving transducer. The system calculates the attenuation coefficient at each measurement frequency and constructs the measured attenuation spectrum vector. ; Based on the measurement frequency range, material parameters, and target particle size range, the corresponding pre-calculated theoretical attenuation matrix is ​​retrieved from the theoretical attenuation matrix database. The particle size distribution is then input into the particle size distribution inversion model for solution, thereby obtaining the particle size distribution results.

[0065] The transducer spacing is designed in stages according to the target nanoparticle size, ranging from 0.5 mm to 100 mm, to match the acoustic propagation attenuation characteristics under different nanoparticle systems. The transducer spacing is available in three configurations: short path, medium path, and long path. Short sound path of one of 0.5mm, 1mm, 1.5mm, 2mm, 2.5mm, 3mm, 3.5mm, 4mm, 4.5mm, and 5mm, used for measurement of particle systems in the tens of nanometer scale; The medium path length is one of 5mm, 10mm, 15mm, 20mm, 25mm, or 30mm, used for the measurement of particle systems in the hundreds of nanometer scale; The long path length is one of 30mm, 40mm, 50mm, 60mm, 70mm, 80mm, 90mm, or 100mm, used for the measurement of nanometer-scale particle systems; By coordinating the configuration of measurement frequency combinations, acoustic path structure, and theoretical attenuation matrix parameters, continuous particle size distribution measurement of particle systems ranging from tens of nanometers to thousands of nanometers can be achieved.

[0066] In ultrasonic measurements of nanoparticles, the smaller the particle size, the weaker the scattering effect of ultrasonic waves. The resulting attenuated signal is often very weak and easily drowned out by noise, which is the fundamental reason why traditional ultrasonic particle size analyzers are ineffective in measuring nanoparticles. To overcome this technical bottleneck, this invention systematically and synergistically optimizes both ultrasonic frequency and sound propagation path.

[0067] In terms of frequency, this invention breaks away from the limitations of traditional devices that rely solely on a single or a few low-frequency transducers, employing a wide-band, multi-frequency measurement scheme covering 10MHz to 200MHz, either individually or in combination. The physical basis for this is that different frequencies of ultrasound have different sensitivity ranges to particle size—low-frequency ultrasound is sensitive to micron-sized particles, while high-frequency ultrasound, with its shorter wavelength, interacts more strongly with nanoscale particles, exciting more significant scattering and thermoviscous effects. Therefore, for ultrafine particle systems in the tens of nanometer range, this invention prioritizes the use of multiple high-frequency transducers, such as 50MHz, 80MHz, 100MHz, 150MHz, and 200MHz, for individual or combined measurements. For complex systems with a wider particle size range, encompassing nanometers to submicrometers and even micrometers, the flexible combination of multiple frequencies within the 10MHz to 200MHz range allows the ultrasonic attenuation spectrum to carry size information of particles at different scales across different frequency bands, thereby improving the identification and resolution capabilities of particles at various scales throughout the entire particle size range.

[0068] Regarding the sound propagation path, this invention recognizes that for nanoscale particles, due to the inherently weak attenuation signal, if the sound path is too long, the ultrasonic energy will be excessively attenuated during propagation due to medium absorption and scattering. The signal reaching the receiver will be too weak or even completely obscured by noise, resulting in ineffective detection. Conversely, for micron-sized or larger particle systems, the attenuation effect of particles on ultrasonic waves is significantly enhanced. If the sound path is too short, the amplitude of the attenuation signal change will be insufficient to be sensitively captured, also affecting measurement accuracy. Therefore, this invention classifies the transducer spacing according to the size range of the target particles, setting three configuration schemes: short sound path (0.5mm to 5mm), medium sound path (5mm to 30mm), and long sound path (30mm to 100mm). Specifically, a short path configuration is used when measuring particles in the tens of nanometers range to ensure that the ultrasonic signal reaches the receiver before being excessively attenuated, thereby effectively extracting the subtle attenuation changes caused by nanoparticles; a medium path configuration is used when measuring particles in the hundreds of nanometers range to allow the attenuated signal to reach a detectable dynamic range at a moderate propagation distance; and a long path configuration is used when measuring particles in the kilo-nanometer (submicron to micrometer) range to allow the attenuation effect caused by the particles to accumulate sufficiently, ensuring that the attenuation spectrum has sufficient resolution and signal-to-noise ratio.

[0069] The optimization of the frequency and sound path dimensions mentioned above is not independent, but rather a collaborative configuration based on pre-calculation of a theoretical attenuation matrix database. For specific target particle size ranges and concentration conditions, the system calls upon the pre-established theoretical attenuation matrix database, selecting a measurement frequency combination and sound path configuration scheme that matches the operating conditions, ensuring that the measured attenuation spectrum is always within the optimal detectable range. Through this strategy of multi-frequency combination and multi-level sound path collaborative optimization, this invention achieves continuous and accurate particle size distribution measurement of particle systems ranging from tens of nanometers to thousands of nanometers.

[0070] (3) Implementation of flow channel and transducer structure The ultrasonic measurement channel is integrated with the industrial delivery pipeline, enabling ultrasonic measurement of the sample during continuous flow. Depending on the operating conditions, either direct measurement via the main flow path or a branch flow approach is used to match the installation conditions and flow states of different industrial sites. The channel module uses standardized interfaces for connection and maintenance with existing pipeline systems. A flow field rectification structure is set up within the measurement area to suppress interference from turbulence and bubbles on ultrasonic propagation, ensuring that the ultrasonic propagation path traverses the stable flow region within the channel. Furthermore, the transducer acoustic field morphology is adapted according to measurement requirements to ensure stable acquisition and repeatability of the ultrasonic attenuation spectrum under flow conditions.

[0071] In this embodiment, the flow channel module and transducer module are integrated onto the industrial pipeline, allowing the sample to pass through the measurement area in a continuous flow state. Combined with optimized anti-turbulence structures and transducer installation methods, online continuous measurement of the industrial pipeline is achieved, specifically including: The measuring device includes a flow channel module and a transducer module, used for online ultrasonic particle size detection in industrial pipelines. (e.g.) Figure 3 As shown, the online ultrasonic particle size measurement device for pipelines includes a flow channel module and transducer modules installed on both sides of it. The flow channel module is directly connected to the industrial pipeline or connected via a bypass. The transmitter and receiver ends of the transducers are respectively arranged on both sides of the measurement cavity. The ultrasonic propagation path passes through the central region of the flow channel. Figure 3 In the measurement device, the sample to be tested enters through inlet 1. The flow channel module is connected to the industrial pipeline via flange / clamp / quick connector 2 for quick installation and disassembly maintenance. Antenna 3 is used for wireless transmission and remote communication of measurement data. The host unit 4 controls the transmission, reception, and data processing of ultrasonic signals. Transmitting transducer 5 is located on one side of the measurement cavity to transmit ultrasonic signals to the sample to be tested, and receiving transducer 6 is located on the other side of the measurement cavity to receive transmitted ultrasonic signals. Particle size distribution is detected by measuring the attenuation characteristics of the ultrasonic signal in the particle system. After the measurement is completed, the sample to be tested is discharged from the measurement device through outlet 7. The figure also shows the positional relationship of sample inlet 1, measurement cavity, and sample outlet 7, as well as the assembly structure between the components.

[0072] To ensure the compatibility of the ultrasonic propagation path with the particle size inversion model, this embodiment designs the flow channel structure and transducer installation method according to the flow state of the measured object and the acoustic characteristics of the medium.

[0073] The flow channel module includes a sample inlet, a measurement chamber, and a sample outlet. The sample to be tested flows through the measurement area in a continuous flow state, enabling online real-time measurement. The flow channel structure is a direct-connection pipe structure, which is directly integrated into the main industrial conveying pipeline to realize in-situ online measurement, so that the ultrasonic measurement path is synchronized with the industrial fluid conveying process; Alternatively, a bypass measurement structure can be set up to introduce the sample to be tested through a diversion method, so as to reduce the impact of mainstream disturbance on the measurement signal or adapt to complex flow conditions, thereby improving measurement stability and repeatability; The flow channel module is connected to the existing conveying pipeline through a standard industrial pipeline connection structure. The pipeline connection structure includes one or more of the following: standard flange connection structure, quick-connect clamp connection structure, and quick-connect coupling connection structure, so as to realize rapid installation, disassembly and maintenance with different industrial pipeline systems. Among them, the standard flange connection structure is used for long-term stable online installation; the quick-connect clamp connection structure is suitable for applications that require frequent disassembly or sanitary applications, and features convenient installation, reliable sealing, and easy cleaning and maintenance; the quick-connect coupling structure is used to realize the rapid docking and replacement between the measurement module and the pipeline system, so as to reduce system maintenance time.

[0074] To further reduce the impact of fluid turbulence, local eddies, and bubble disturbances on the ultrasonic propagation process, preferably, the transducer transmitter and receiver are respectively located on both sides of the measurement cavity, so that the ultrasonic propagation path passes through the stable flow region in the center of the flow channel, thereby reducing the impact of boundary reflections and flow disturbances on the measurement signal. An anti-turbulence structure is installed inside the measurement chamber to improve the flow field stability within the measurement area, enabling the measured medium to form a relatively stable flow state in the ultrasonic propagation path region. This reduces the impact of velocity fluctuations, particle agglomeration, and bubble entrainment on the measurement results, thereby improving the stability and repeatability of ultrasonic attenuation measurements. (e.g.) Figure 4 As shown, the anti-turbulence structure is installed upstream, inside, or downstream of the measurement area, depending on the fluid characteristics, pipe diameter, and operating conditions. Figure 4The following example illustrates the use of an anti-turbulence structure 9 positioned upstream of the measurement area. The pipe wall 8 forms the outer boundary of the measurement cavity, carrying the internal fluid and providing structural support. A transducer connection structure 10 is located inside the pipe wall 8, used to fix the transducer module to both sides of the measurement cavity and ensure acoustic coupling and sealing performance between the transducer transmitter and receiver and the measurement cavity. The anti-turbulence structure 9 is located inside the measurement cavity, on the side facing the incoming flow direction of the ultrasonic propagation path. Its shape and size are designed to match the pipe diameter and fluid characteristics. By rectifying the fluid before the measurement area, the velocity distribution of the measured medium tends to be uniform when entering the ultrasonic propagation path area, thereby effectively suppressing the interference of turbulent pulsations and eddies on the measurement signal. The figure also shows a schematic diagram of the flow direction of the measured medium entering from the inflow side, being rectified by the anti-turbulence structure 9, and leaving from the outflow side, as well as the relative positional relationship and assembly structure between the pipe wall 8, the anti-turbulence structure 9, and the transducer connection structure 10.

[0075] The anti-turbulence structure is installed upstream of the measurement area, inside the measurement area, or downstream of the measurement area, depending on the fluid characteristics, pipe diameter, and operating conditions. The transducer can be a focused transducer or a non-focused transducer. The non-focused transducer is used to form an approximately parallel or diffuse sound field, while the focused transducer is used to increase the local sound energy density, thereby enhancing the signal strength and measurement sensitivity. During the measurement process, the sample under test passes through the detection area in a continuous flow state. The transducer continuously emits and receives ultrasonic signals. The system obtains attenuation spectrum information at different frequencies and uses it for subsequent particle size distribution inversion processing, thereby realizing online continuous measurement of industrial pipelines.

[0076] In actual industrial production processes, the slurry or suspension to be tested is usually in a continuous flow state, often with a high solid content and a certain flow velocity. This presents unique challenges to online ultrasonic measurement, unlike static or offline measurements. On the one hand, industrial sites have compact pipeline layouts and limited space, requiring the measuring device to be small in size and easily integrated into existing pipeline systems, while also meeting the practical needs of frequent cleaning and maintenance and adapting to different connection methods. On the other hand, turbulence, eddies, and bubble entrainment are inevitable phenomena during fluid transport through pipelines. These factors interfere with the sound field distribution along the ultrasonic propagation path, causing random fluctuations in the received signal, which in turn affects the stability and repeatability of the measurement results. To address these practical engineering problems, this invention provides a systematic adaptation design for the flow channel structure and transducer installation.

[0077] Regarding pipeline integration, this invention offers two flexible structural solutions: direct connection and bypass. The direct connection structure embeds the measurement unit directly into the main industrial pipeline, allowing the ultrasonic measurement path to synchronize with the fluid transport process. The ultrasonic signal penetrates the flowing sample in real time, making it suitable for most conventional online monitoring scenarios and providing the most accurate reflection of the actual state of the medium within the pipeline. The bypass structure, on the other hand, diverts a small portion of the sample from the main pipeline into the measurement chamber, completing the measurement under relatively independent and controllable flow conditions. This method effectively reduces the non-acoustic interference of severe turbulence, high flow velocities, or complex flow patterns in the main pipeline on the ultrasonic signal. It is suitable for conditions with harsher flow conditions or higher requirements for measurement stability. Users can flexibly choose the appropriate solution based on the actual flow conditions and measurement needs.

[0078] Regarding pipeline connection and maintenance, this invention fully considers the installation conditions and usage habits in different industrial scenarios, employing multiple interface methods such as standard flange connections, quick-connect clamp connections, and quick-connect couplings. Flange connections are suitable for applications requiring long-term fixed installation and infrequent disassembly, ensuring a robust and reliable connection. Quick-connect clamp connections are specifically designed for hygienic applications in food and pharmaceutical industries that require frequent disassembly and cleaning, allowing for quick installation and disassembly without specialized tools, while also providing excellent sealing performance and facilitating regular cleaning and maintenance. Quick-connect couplings further simplify the docking process between the measurement module and the pipeline system, enabling replacement within seconds and significantly reducing system maintenance downtime. Through the compatible design of these multiple connection methods, this invention can flexibly adapt to the actual installation conditions of different industrial sites.

[0079] Regarding measurement stability, this invention addresses the interference of fluid flow conditions in industrial pipelines on ultrasonic measurements by optimizing both transducer arrangement and flow field rectification. The transducer transmitter and receiver are installed on opposite sides of the measurement cavity, ensuring the ultrasonic propagation path passes through the stable flow zone in the central region of the channel. This region experiences minimal interference from the pipe wall boundary layer effect and reflections, facilitating the acquisition of a cleaner acoustic signal. More importantly, this invention incorporates an anti-turbulence structure within the measurement cavity. The core function of this structure is to rectify and homogenize the fluid entering the measurement area, resulting in a more uniform velocity distribution and consistent flow direction. This suppresses disturbances to the ultrasonic propagation path caused by turbulent pulsations, local eddies, and bubble entrainment. Depending on different fluid characteristics, pipe diameters, and site conditions, the anti-turbulence structure can be flexibly positioned upstream, inside, or downstream of the measurement area to achieve optimal flow field stability.

[0080] Furthermore, in terms of transducer selection, this invention is compatible with both focused and unfocused transducers. Unfocused transducers generate approximately parallel or diffuse sound fields, suitable for measurement scenarios with short sound paths or high requirements for sound field uniformity; focused transducers, on the other hand, can concentrate ultrasonic energy at a specific focal point within the measurement area, forming a high energy density region near the focal point, thereby significantly enhancing the ability to penetrate weak attenuated signals, especially suitable for nanoscale particle systems or low-concentration samples with extremely weak attenuated signals. Users can flexibly select the appropriate transducer based on the actual sample characteristics and measurement sensitivity requirements.

[0081] Through the comprehensive technical solutions of the above-mentioned flow channel structure design, compatibility of multiple connection methods, optimization of anti-turbulence flow field, and flexible selection of transducers, this invention effectively solves engineering problems such as online installation and adaptation of industrial pipelines, suppression of flow state interference, and long-term operation and maintenance while ensuring measurement accuracy. It provides a reliable structural foundation and engineering guarantee for ultrasonic particle size measurement technology to move from the laboratory to the industrial field.

[0082] Second Embodiment like Figure 5 As shown, this embodiment provides a pipeline online ultrasonic particle size distribution inversion system for performing the pipeline online ultrasonic particle size distribution inversion method as described above, comprising: The regular inversion module is used to perform inversion by coupling Tikhonov regularization with the interior point method. It obtains the particle size distribution and characteristic particle size parameters by constructing and solving a constrained optimization objective function between the measured attenuation spectrum vector and the theoretical attenuation matrix. The multi-frequency measurement module is used to perform individual or combined measurements using one or more ultrasonic transducers with one center frequency, and to configure the transducer spacing in a graded manner to achieve multi-frequency attenuation measurement and particle size distribution inversion of particle systems ranging from tens of nanometers to thousands of nanometers. The flow channel integration module is used to integrate the flow channel module and the transducer module onto industrial pipelines, allowing the sample to pass through the measurement area in a continuous flow state. Combined with the optimization of the anti-turbulence structure and transducer installation method, it enables online continuous measurement of industrial pipelines.

[0083] A computer-readable storage medium stores computer code that, when executed, performs the methods described above. Those skilled in the art will understand that all or part of the steps in the various methods of the above embodiments can be implemented by a program instructing related hardware. This program can be stored in a computer-readable storage medium, which may include: read-only memory (ROM), random access memory (RAM), a magnetic disk, or an optical disk, etc.

[0084] The above description is merely a preferred embodiment of the present invention. The scope of protection of the present invention is not limited to the above embodiments. All technical solutions falling within the scope of the present invention's concept are within the scope of protection of the present invention. It should be noted that for those skilled in the art, any improvements and modifications made without departing from the principles of the present invention should also be considered within the scope of protection of the present invention.

[0085] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0086] It should be noted that the above embodiments can be freely combined as needed. The above description is only a preferred embodiment of the present invention. It should be pointed out that for those skilled in the art, several improvements and modifications can be made without departing from the principle of the present invention, and these improvements and modifications should also be considered within the scope of protection of the present invention.

Claims

1. A method for online ultrasonic particle size distribution inversion in pipelines, characterized in that, Includes the following steps: The particle size distribution and characteristic particle size parameters are obtained by coupling inversion with Tikhonov regularization and interior point method, constructing and solving a constrained optimization objective function between the measured attenuation spectrum vector and the theoretical attenuation matrix; One or more ultrasonic transducers with center frequencies are used for individual or combined measurements, and the transducer spacing is configured in a graded manner to achieve multi-frequency attenuation measurement and particle size distribution inversion of particle systems ranging from tens of nanometers to thousands of nanometers. By integrating the flow channel module and transducer module onto the industrial pipeline, the sample under test can pass through the measurement area in a continuous flow state. Combined with the optimization of the anti-turbulence structure and transducer installation method, online continuous measurement of industrial pipelines can be realized.

2. The pipeline online ultrasonic particle size distribution inversion method according to claim 1, characterized in that, The process employs a coupled inversion method using Tikhonov regularization and interior point method to obtain particle size distribution and characteristic particle size parameters, specifically including: The ultrasonic attenuation inversion problem is constructed as an ill-conditioned inverse problem described by the first type of Fredholm integral equation. A mapping relationship between particle size distribution and multi-frequency attenuation response is established using a theoretical acoustic scattering model combined with heat dissipation and viscous loss mechanisms. Tikhonov regularization is introduced to superimpose the smoothness constraint of the solution in the least squares objective to suppress numerical oscillations and non-physical oscillations of the solution caused by measurement noise and model errors. At the same time, physical prior conditions such as the non-negativity constraint of particle size distribution are introduced to transform the inversion problem into a constrained convex optimization problem. Then, the interior point method is used to iteratively approximate the global optimum along the gradient direction within the feasible region, so that the inversion process can achieve fast convergence and stable solution while maintaining the accuracy of the solution.

3. The pipeline online ultrasonic particle size distribution inversion method according to claim 1, characterized in that, The method employs a coupled inversion technique using Tikhonov regularization and the interior-point method. By constructing and solving a constrained optimization objective function between the measured attenuation spectral vector and the theoretical attenuation matrix, the particle size distribution and characteristic particle size parameters are obtained. Specifically, this includes: Transmission signals of the sample were acquired at multiple ultrasonic center frequencies, and the ultrasonic attenuation coefficients at the corresponding frequencies were calculated to construct the measured attenuation spectrum vector. ; Based on the particle acoustic scattering model, and combined with the heat dissipation model and viscous loss model, a multi-parameter theoretical attenuation response matrix is ​​established. ,in Indicates the measured frequency. Indicates discrete particle size range, Indicates the relevant physical parameters of the material; The theoretical attenuation matrix is ​​pre-calculated, discretized, and stored in the parameter database, and can be retrieved by indexing according to the operating conditions during the measurement process; The ultrasonic attenuation inversion problem is modeled as a first-type Fredholm integral discretization inverse problem. A Tikhonov regularization method is introduced to stabilize the original least squares problem. The objective function is as follows: ； in, This is based on the multi-parameter theoretical decay matrix obtained from pre-calculated database calls. This is the measured attenuation spectral vector. Let be the particle size distribution vector to be determined. For the difference operator constrained by particle size distribution smoothing, This is the regularization coefficient.

4. The pipeline online ultrasonic particle size distribution inversion method according to claim 3, characterized in that, The method employs a coupled inversion technique using Tikhonov regularization and the interior-point method. This involves constructing and solving a constrained optimization objective function between the measured attenuation spectral vector and the theoretical attenuation matrix to obtain the particle size distribution and characteristic particle size parameters. The method also includes: Introducing nonnegativity constraints on particle size distribution into the constrained optimization model This transforms the problem into a constrained convex optimization problem. The original-dual interior point method is used to solve the problem. By constructing a barrier function, the non-negativity constraint is transformed into a logarithmic barrier term, and a nonlinear equation system is established in combination with KKT conditions. During the iteration process, the original and dual variables are updated simultaneously, and the search direction is solved based on gradient information and the Hessian matrix. The particle size distribution vector is updated iteratively step by step. Continue until the convergence condition is met; The final output is the particle size distribution curve, and the characteristic particle size parameters of Dv10, Dv50, and Dv90 are calculated.

5. The pipeline online ultrasonic particle size distribution inversion method according to claim 1, characterized in that, The method employs one or more ultrasonic transducers with different center frequencies for individual or combined measurements, and hierarchically configures the transducer spacing to achieve multi-frequency attenuation measurement and particle size distribution inversion. Specifically, this includes: Based on the differentiated sensitivity of different frequencies to particle size, one or more center frequencies are selected for individual or combined measurements according to the target particle size range to expand the coverage of particle size detection and improve the resolution of different particle size intervals. At the same time, the corresponding sound path configuration is matched according to the attenuation strength characteristics corresponding to the particle size, so that the attenuation spectrum maintains a detectable dynamic range and measurement sensitivity in different particle size intervals.

6. The pipeline online ultrasonic particle size distribution inversion method according to claim 1, characterized in that, The method employs one or more ultrasonic transducers with different center frequencies for individual or combined measurements to achieve multi-frequency attenuation measurement and particle size distribution inversion of particle systems ranging from tens of nanometers to thousands of nanometers. Specifically, this includes: The measurement system uses one or more ultrasonic transducers with center frequencies for individual or combined measurements. The center frequencies include one or more of 10MHz, 20MHz, 30MHz, 50MHz, 80MHz, 100MHz, 150MHz and 200MHz, thereby realizing multi-frequency attenuation measurement of particle systems ranging from tens of nanometers to thousands of nanometers. Based on the target particle size range, one or more center frequencies are selected for individual or combined measurements. Different frequencies correspond to different particle scattering sensitive ranges, thereby improving the recognition and resolution capabilities of nanoparticles of different sizes. For nanoparticle systems, one or more high-frequency transducers of 50MHz, 80MHz, 100MHz, 150MHz and 200MHz are used for individual or combined measurements. The ultrasonic signal enters the medium under test through the transmitting transducer. After being scattered, dissipated by heat and viscous loss by the particles, the transmitted signal is received by the receiving transducer. The system calculates the attenuation coefficient at each measurement frequency and constructs the measured attenuation spectrum vector. ; Based on the measurement frequency range, material parameters, and target particle size range, the corresponding pre-calculated theoretical attenuation matrix is ​​retrieved from the theoretical attenuation matrix database. The particle size distribution is then input into the particle size distribution inversion model for solution, thereby obtaining the particle size distribution results.

7. The pipeline online ultrasonic particle size distribution inversion method according to claim 6, characterized in that, The step-by-step configuration of transducer spacing enables multi-frequency attenuation measurement and particle size distribution inversion for particle systems ranging from tens of nanometers to thousands of nanometers. Specifically, this includes: The transducer spacing is designed in stages according to the target nanoparticle size, ranging from 0.5 mm to 100 mm, to match the acoustic propagation attenuation characteristics under different nanoparticle systems. The transducer spacing is available in three configurations: short path, medium path, and long path. Short sound path of one of 0.5mm, 1mm, 1.5mm, 2mm, 2.5mm, 3mm, 3.5mm, 4mm, 4.5mm, and 5mm, used for measurement of particle systems in the tens of nanometer scale; The medium path length is one of 5mm, 10mm, 15mm, 20mm, 25mm, or 30mm, used for the measurement of particle systems in the hundreds of nanometer scale; The long path length is one of 30mm, 40mm, 50mm, 60mm, 70mm, 80mm, 90mm, or 100mm, used for the measurement of nanometer-scale particle systems; By coordinating the configuration of measurement frequency combinations, acoustic path structure, and theoretical attenuation matrix parameters, continuous particle size distribution measurement of particle systems ranging from tens of nanometers to thousands of nanometers can be achieved.

8. The pipeline online ultrasonic particle size distribution inversion method according to claim 1, characterized in that, The integration of the flow channel module and transducer module onto the industrial pipeline allows the sample to pass through the measurement area in a continuous flow state. Combined with optimized anti-turbulence structures and transducer installation methods, this enables online continuous measurement of industrial pipelines, specifically including: The ultrasonic measurement channel is integrated with the industrial delivery pipeline, enabling ultrasonic measurement of the sample during continuous flow. Depending on the operating conditions, either direct measurement via the main flow path or a branch flow approach is used to match the installation conditions and flow states of different industrial sites. The channel module uses standardized interfaces for connection and maintenance with existing pipeline systems. A flow field rectification structure is set up within the measurement area to suppress interference from turbulence and bubbles on ultrasonic propagation, ensuring that the ultrasonic propagation path traverses the stable flow region within the channel. Furthermore, the transducer acoustic field morphology is adapted according to measurement requirements to ensure stable acquisition and repeatability of the ultrasonic attenuation spectrum under flow conditions.

9. The pipeline online ultrasonic particle size distribution inversion method according to claim 1, characterized in that, The integration of the flow channel module and the transducer module onto the industrial pipeline, allowing the sample to pass through the measurement area in a continuous flow state, specifically includes: The measuring device includes a flow channel module and a transducer module. The flow channel module includes a sample inlet, a measuring chamber, and a sample outlet. The sample to be tested passes through the measuring area in a continuous flow state, realizing online real-time measurement. The flow channel structure is a direct-connection pipe structure, which is directly integrated into the main industrial conveying pipeline to realize in-situ online measurement, so that the ultrasonic measurement path is synchronized with the industrial fluid conveying process; Alternatively, a bypass measurement structure can be set up to introduce the sample to be tested through a diversion method, so as to reduce the impact of mainstream disturbance on the measurement signal or adapt to complex flow conditions, thereby improving measurement stability and repeatability; The flow channel module is connected to the existing conveying pipeline through a standard industrial pipeline connection structure. The pipeline connection structure includes one or more of the following: standard flange connection structure, quick-connect clamp connection structure, and quick-connect coupling connection structure, so as to realize rapid installation, disassembly and maintenance with different industrial pipeline systems. Among them, the standard flange connection structure is used for long-term stable online installation; the quick-connect clamp connection structure is suitable for applications that require frequent disassembly or sanitary applications, and features convenient installation, reliable sealing, and easy cleaning and maintenance; the quick-connect coupling structure is used to realize the rapid docking and replacement between the measurement module and the pipeline system, so as to reduce system maintenance time.

10. The pipeline online ultrasonic particle size distribution inversion method according to claim 9, characterized in that, The optimized combination of anti-turbulence structure and transducer installation method enables continuous online measurement of industrial pipelines, specifically including: The transducer transmitter and receiver are respectively located on both sides of the measurement cavity, so that the ultrasonic propagation path passes through the stable flow area in the center of the flow channel, thereby reducing the influence of boundary reflection and flow disturbance on the measurement signal; An anti-turbulence structure is installed inside the measurement cavity to improve the flow field stability within the measurement area, enabling the measured medium to form a relatively stable flow state in the ultrasonic propagation path region. This reduces the impact of flow velocity fluctuations, particle agglomeration, and bubble entrainment on the measurement results, thereby improving the stability and repeatability of ultrasonic attenuation measurement. The anti-turbulence structure is installed upstream of the measurement area, inside the measurement area, or downstream of the measurement area, depending on the fluid characteristics, pipe diameter, and operating conditions. The transducer can be a focused transducer or a non-focused transducer. The non-focused transducer is used to form an approximately parallel or diffuse sound field, while the focused transducer is used to increase the local sound energy density, thereby enhancing the signal strength and measurement sensitivity. During the measurement process, the sample under test passes through the detection area in a continuous flow state. The transducer continuously emits and receives ultrasonic signals. The system obtains attenuation spectrum information at different frequencies and uses it for subsequent particle size distribution inversion processing, thereby realizing online continuous measurement of industrial pipelines.

11. A pipeline online ultrasonic particle size distribution inversion system for performing the pipeline online ultrasonic particle size distribution inversion method as described in any one of claims 1-10, characterized in that, include: The regular inversion module is used to perform inversion by coupling Tikhonov regularization with the interior point method. It obtains the particle size distribution and characteristic particle size parameters by constructing and solving a constrained optimization objective function between the measured attenuation spectrum vector and the theoretical attenuation matrix. The multi-frequency measurement module is used to perform individual or combined measurements using one or more ultrasonic transducers with one center frequency, and to configure the transducer spacing in a graded manner to achieve multi-frequency attenuation measurement and particle size distribution inversion of particle systems ranging from tens of nanometers to thousands of nanometers. The flow channel integration module is used to integrate the flow channel module and the transducer module onto industrial pipelines, allowing the sample to pass through the measurement area in a continuous flow state. Combined with the optimization of the anti-turbulence structure and transducer installation method, it enables online continuous measurement of industrial pipelines.

12. A computer device, characterized in that, The device includes a memory and one or more processors, wherein the memory stores computer code that, when executed by the one or more processors, causes the one or more processors to perform the method as described in any one of claims 1 to 10.

13. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer code, and when the computer code is executed, the method as described in any one of claims 1 to 10 is performed.

Images