Intelligent automobile pump cooperative control method and system based on digital twinning

Abstract

The invention provides an intelligent automobile pump cooperative control method and system based on digital twinning, and the method comprises the steps: firstly recognizing the rheological characteristic difference of new and old concrete through a hopper sensor array, and generating a digital model of a heterogeneous fluid interface; self-excited vibration generated when the pumping system works is used as a detection source, and the spatial position of an interface in a conveying pipeline is positioned in real time by analyzing a vibration echo signal. A dynamic evolution model is constructed based on the position, the future motion deformation trend of the interface is predicted, and the time when the interface arrives at the key position is accurately calculated and the pressure impact is estimated. And a set of nonlinear cooperative control curve containing various execution parameters is generated in advance according to a prediction result, and the pumping system is driven to operate according to the curve in a predicted time window. According to the invention, active avoidance and cooperative control of the pumping risk are realized, and the construction safety is improved.

Description

Technical Field

[0001] This invention belongs to the field of intelligent control technology for automotive pumps, specifically relating to an intelligent automotive pump collaborative control method and system based on digital twins. Background Technology

[0002] Concrete pump trucks are crucial equipment in modern construction, responsible for precisely pouring ready-mixed concrete to designated locations via long-distance pipelines. In large-scale continuous pouring operations, due to factors such as production scheduling at concrete batching plants, transportation delays, or minor adjustments to the formula, the pump truck hopper often receives different batches of concrete sequentially. Even if these batches of concrete have the same grade, their rheological properties, such as slump, cohesiveness, and aggregate gradation, can vary significantly. When concrete with these differing rheological properties enters the same pipeline successively, sudden changes in flow velocity and pressure disturbances can easily occur at the interface between the two fluids under pumping pressure. This is a major cause of pipe blockage and severe pressure shocks to the pumping system. Currently, handling this situation relies entirely on the subjective judgment and manual intervention of experienced operators, but this approach is not only unsuccessful but also carries significant risks. Summary of the Invention

[0003] This invention provides a method and system for collaborative control of intelligent automotive pumps based on digital twins, in order to solve the above-mentioned technical problems.

[0004] In a first aspect, the present invention provides a collaborative control method for an intelligent vehicle pump based on digital twins, applied to an intelligent vehicle pump, the intelligent vehicle pump including a hopper and a pumping system, the method comprising the following steps: When the presence of newly injected concrete in the hopper is detected by the hopper sensor array arranged in the hopper, the difference in rheological properties between the newly injected concrete and the existing concrete in the hopper is identified, and a heterogeneous fluid interface in the delivery pipeline of the pumping system is generated based on the difference in rheological properties. Using the self-excited vibration generated by the pumping system in operation as a detection source, the spatial position of the heterogeneous fluid interface in the delivery pipeline is located in real time by inverting and calculating the vibration echo signal formed after the self-excited vibration propagates in the concrete and is reflected by the heterogeneous fluid interface. A dynamic evolution model of the heterogeneous fluid interface is constructed based on the spatial location, and the motion and deformation trend of the heterogeneous fluid interface in the delivery pipeline is predicted by solving the dynamic evolution model. Based on the motion deformation trend, the arrival time of the heterogeneous fluid interface to the preset key position of the delivery pipeline and the estimated pressure impact are predicted using a dynamic evolution model. An adaptive pressure control curve is generated by combining the arrival time and the estimated pressure impact, which is time-coordinated and includes multiple execution parameters of the pumping system. The pumping system is driven to operate according to the adaptive pressure control curve within the preset time window corresponding to the arrival time.

[0005] Furthermore, the hopper sensor array includes a 3D vision sensor and an immersion rheological sensor. When the hopper sensor array, arranged within the hopper, detects the presence of newly injected concrete, the step of identifying the rheological property differences between the newly injected concrete and the existing concrete in the hopper, and generating a heterogeneous fluid interface within the pumping system's delivery pipeline based on these rheological property differences, includes the following steps: When the presence of newly injected concrete in the hopper is detected by the 3D vision sensor arranged above the hopper, the 3D vision sensor continuously collects three-dimensional point cloud data of the concrete surface in the hopper. The slump of the newly poured concrete is calculated based on the three-dimensional point cloud data. The viscosity and yield stress parameters of the freshly poured concrete are measured by the immersion rheological sensor installed inside the hopper. The rheological characteristic vector of the newly poured concrete is obtained by integrating the slump, viscosity and yield stress parameters. The rheological feature vector of the newly injected concrete is quantitatively compared with the pre-stored reference rheological feature vector of the original concrete to calculate the difference in rheological properties between the newly injected concrete and the original concrete. When the difference in rheological properties exceeds a preset risk threshold, it is determined that a heterogeneous fluid interface is generated in the delivery pipeline of the pumping system.

[0006] Furthermore, using the self-excited vibration generated by the pumping system during operation as a detection source, and calculating the vibration echo signal formed after the self-excited vibration propagates in the concrete and is reflected by the heterogeneous fluid interface, the spatial position of the heterogeneous fluid interface in the delivery pipeline in real time is determined by the following steps: Multiple piezoelectric accelerometers are arranged in the pump body, reversing valve group and delivery pipeline of the pumping system to form a vibration sensing array; The broadband vibration signal generated by the self-excited vibration of the pumping system during operation is continuously acquired through the vibration sensing array. The broadband vibration signal is processed by the signal processing unit of the pumping system to extract the vibration echo signal generated by the reflection from the interface of the heterogeneous fluids. A vibration propagation model of vibration waves in concrete medium is established, and the vibration echo signal is input into the vibration propagation model. Based on the propagation model, the vibration echo signal is inverted and calculated to obtain the spatial position of the heterogeneous fluid interface.

[0007] Furthermore, the spatial location of the heterogeneous fluid interface is obtained by inverting the vibration echo signal based on the propagation model, including the following steps: The vibration echo signal is subjected to adaptive filtering and processed using a beamforming algorithm. By combining the time difference of the same vibration echo signal received by different sensors in the vibration sensing array and the sound velocity parameter in the propagation model, the spatial position of the heterogeneous fluid interface in the delivery pipeline is calculated by the triangulation algorithm. By analyzing the energy attenuation characteristics of the vibration echo signal, the acoustic impedance difference between the concrete on both sides of the heterogeneous fluid interface is inverted, and the acoustic impedance difference is used as a quantitative indicator of the physical properties of the heterogeneous fluid interface and correlated with the spatial location.

[0008] Furthermore, constructing a dynamic evolution model of the heterogeneous fluid interface based on the spatial location, and predicting the motion and deformation trend of the heterogeneous fluid interface in the delivery pipeline by solving the dynamic evolution model includes the following steps: The delivery pipeline is constructed as a fixed three-dimensional Euler mesh in digital space; The located heterogeneous fluid interface is discretized into multiple virtual Lagrange particles carrying interface properties, and all the virtual Lagrange particles are mapped to the initial position of the three-dimensional Eulerian mesh. Based on the real-time pumping flow rate of the pumping system, the fluid dynamics equations are solved on the three-dimensional Eulerian grid to obtain the background flow field distribution of the entire pipeline in the delivery pipeline. Within each preset time step, the motion of the virtual Lagrange particles is driven by the background flow field distribution to track the macroscopic displacement of the heterogeneous fluid interface; Based on the velocity gradient and shear force of the background flow field distribution, the relative position changes between each virtual Lagrange particle are calculated simultaneously, and the stretching trend and dispersion deformation trend of the heterogeneous fluid interface are predicted.

[0009] Furthermore, the method also includes the following steps: Obtain the spatial location and acoustic impedance difference of the heterogeneous fluid interface; The heterogeneous fluid interface is discretized into multiple virtual Lagrange particles, and an initial spatial distribution of all the virtual Lagrange particles is generated in the three-dimensional Eulerian grid based on their spatial positions. The acoustic impedance difference is assigned as a physical interface attribute to all the virtual Lagrange particles. In the process of solving the dynamic evolution model, the physical behavior of the fluids on both sides of the heterogeneous fluid interface is distinguished by the relevant physical properties, and the dispersion rate of the heterogeneous fluid interface is calculated by combining the shear stress applied to the virtual Lagrange particle cloud by the background flow field.

[0010] Furthermore, based on the aforementioned motion deformation trend and using a dynamic evolution model to predict the arrival time of the heterogeneous fluid interface at the preset key location in the delivery pipeline, as well as to estimate the pressure impact, includes the following steps: Based on the geometric model of the delivery pipeline in digital space, virtual monitoring surfaces are set up at the bends, reducers and valve groups of the geometric model; In each iteration step of the dynamic evolution model, it is determined whether a target virtual Lagrange particle crosses the virtual monitoring surface; When the target virtual Lagrange particle crosses the virtual monitoring surface, the crossing time of the target virtual Lagrange particle is recorded, and the crossing time is used as the arrival time of the heterogeneous fluid interface to the preset key position of the delivery pipeline. The amplitude of the predicted pressure impact is calculated by combining the interface properties carried by the target virtual Lagrange particle and the pressure value of the background flow field.

[0011] Furthermore, the estimated magnitude of the pressure impact is calculated by combining the interface properties carried by the target virtual Lagrange particle and the pressure value of the background flow field, including the following steps: Define a monitoring control volume at the virtual monitoring surface; When the virtual Lagrange particle cloud enters the monitoring and control body, the average rheological parameters of the fluid within the monitoring and control body are updated according to the interface attributes carried by the virtual Lagrange particles. By combining the updated mean rheological parameters and the velocity information of the background flow field, the momentum change and viscous dissipation increment caused by the changes in fluid properties and pipe geometry are calculated. Substituting the momentum change and viscous dissipation increment into the momentum conservation equation, the momentum conservation equation is solved to obtain the pressure rise at the inlet section of the monitoring and control body, and the pressure rise is used as the amplitude of the predicted pressure impact.

[0012] The present invention also discloses an intelligent vehicle pump collaborative control system based on digital twins, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the method described in any of the above.

[0013] The present invention also discloses a storage medium storing computer instructions thereon, which, when executed by a processor, implement the steps of any of the methods described above.

[0014] The beneficial effects of this invention are: This invention overturns the outdated model of traditional concrete pumping operations, which relies on operator experience to passively respond to risks associated with alternating batches of concrete. Utilizing the self-excited vibration of the pumping system itself as a detection source, and combined with advanced inversion calculations, it achieves non-invasive, real-time dynamic tracking of the interface between heterogeneous fluids within the delivery pipeline, accurately presenting the previously invisible internal state of the pipe in a precise digital form. The core effect goes beyond positioning; it lies in constructing a dynamic evolution model of the fluid interface. This allows the control system to accurately predict the movement trend of the fluid interface in the pipeline, the time it takes to reach critical locations, and potential pressure shocks. Furthermore, it generates nonlinear collaborative control curves, upgrading pumping control from blind open-loop operation to closed-loop intelligent control, fundamentally and proactively mitigating the risks of pipe blockage and equipment impact. Attached Figure Description

[0015] Figure 1 This is a flowchart of the intelligent vehicle pump collaborative control method based on digital twins in this invention.

[0016] Figure 2 This is a velocity distribution cloud map of concrete with a yield stress of 150 Pa on the cross-section of a horizontal straight pipe in this invention.

[0017] Figure 3 This is a velocity distribution cloud map of concrete with a yield stress of 100 Pa on the cross-section of a horizontal straight pipe in this invention.

[0018] Figure 4 This is a velocity distribution cloud map of concrete with a yield stress of 80 Pa on the cross-section of a horizontal straight pipe in this invention.

[0019] Figure 5 This is a velocity cloud map of the pumped concrete within the delivery pipeline in this invention.

[0020] Figure 6 This is a schematic diagram of the virtual monitoring surface setting in the three-dimensional model of the delivery pipeline in this invention. Detailed Implementation

[0021] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of the present invention.

[0022] It should be noted that all directional indications (such as up, down, left, right, front, back, etc.) in the embodiments of the present invention are only used to explain the relative positional relationship and movement of each component in a certain specific posture (as shown in the figure). If the specific posture changes, the directional indication will also change accordingly.

[0023] Furthermore, in this invention, descriptions involving "first," "second," etc., are for descriptive purposes only and should not be construed as indicating or implying their relative importance or implicitly specifying the number of technical features indicated. Thus, a feature defined as "first" or "second" may explicitly or implicitly include at least one of that feature. In the description of this invention, "a plurality of" means at least two, such as two, three, etc., unless otherwise explicitly specified.

[0024] In this invention, unless otherwise explicitly specified and limited, the terms "connection," "fixed," etc., should be interpreted broadly. For example, "fixed" can mean a fixed connection, a detachable connection, or an integral part; it can mean a mechanical connection or an electrical connection; it can mean a direct connection or an indirect connection through an intermediate medium; it can mean the internal communication of two components or the interaction between two components, unless otherwise explicitly limited. Those skilled in the art can understand the specific meaning of the above terms in this invention according to the specific circumstances.

[0025] In the following description, the use of suffixes such as "module," "part," or "unit" to denote elements is solely for the purpose of illustrative purposes and has no specific meaning in itself. Therefore, "module," "part," or "unit" may be used interchangeably.

[0026] Furthermore, the technical solutions of the various embodiments of the present invention can be combined with each other, but only if they are feasible for those skilled in the art. If the combination of technical solutions is contradictory or cannot be implemented, it should be considered that such combination of technical solutions does not exist and is not within the scope of protection claimed by the present invention.

[0027] Reference Figure 1 The intelligent vehicle pump collaborative control method based on digital twins includes the following steps: S1: When the presence of newly injected concrete is detected by the hopper sensor array arranged in the hopper, the difference in rheological properties between the newly injected concrete and the existing concrete in the hopper is identified, and a heterogeneous fluid interface in the delivery pipeline of the pumping system is generated based on the difference in rheological properties.

[0028] S2: Using the self-excited vibration generated by the pumping system in operation as a detection source, the vibration echo signal formed by the propagation of the self-excited vibration in the concrete and its reflection at the interface of the heterogeneous fluid is calculated by inversion, and the spatial position of the interface of the heterogeneous fluid in the delivery pipeline is located in real time.

[0029] S3: Construct a dynamic evolution model of the heterogeneous fluid interface based on spatial location, and predict the motion and deformation trend of the heterogeneous fluid interface in the pipeline by solving the dynamic evolution model.

[0030] S4: Based on the motion deformation trend and using a dynamic evolution model, predict the arrival time of the heterogeneous fluid interface to the preset key position in the delivery pipeline and estimate the pressure impact.

[0031] S5: Generates an adaptive pressure control curve that is time-coordinated and includes multiple execution parameters of the pumping system by combining arrival time and estimated pressure shock.

[0032] S6: Drive the pumping system to operate according to the adaptive pressure control curve within the preset time window corresponding to the arrival time.

[0033] In the above embodiments, to achieve fully automatic or semi-automatic continuous material replenishment of the hopper, prior to the step of detecting newly injected concrete disclosed in this invention, the system also includes an intelligent material receiving and guidance module that works in conjunction with the pumping system. Specifically, this intelligent material receiving and guidance module monitors the amount of concrete in the hopper in real time through multiple laser or ultrasonic level sensors deployed on the edge of the hopper. When the concrete level is detected to be lower than a preset first replenishment threshold, the control system sends a material receiving request signal to the standby concrete mixer truck via a wireless communication unit. At the same time, a visual guidance system arranged above the hopper is activated. This system includes a wide-angle camera and a structured light emitter to capture the real-time position of the mixer truck's unloading chute and provides intuitive alignment guidance information to the mixer truck driver through a high-brightness LED indicator matrix installed on the pump truck body, or directly links with the mixer truck's unloading system with autonomous positioning function to automatically adjust the angle and position of the unloading chute to accurately align with the hopper's inlet. After concrete begins to be poured into the hopper, when the level sensor detects that the material level has reached the preset second overflow threshold, the system sends a signal to stop the material replenishment, thereby realizing an unmanned, precise, and safe automatic material receiving process. The initiation of this automatic material receiving process marks the start of a new round of concrete injection and immediately triggers the subsequent rheological property difference identification process.

[0034] When the hopper sensor array detects the injection of new concrete, it first acquires three-dimensional point cloud data of the concrete surface within the hopper using a 3D vision sensor to calculate the slump parameter of the newly injected concrete. Simultaneously, an immersion rheological sensor measures the viscosity and yield stress of the new concrete. The slump, viscosity, and yield stress are integrated into a new injection rheological feature vector, and the difference in rheological properties is quantified by calculating the Euclidean distance between the new injection rheological feature vector and a pre-stored reference rheological feature vector. When the difference exceeds a preset risk threshold, it is determined that a heterogeneous fluid interface will form within the pumping system's delivery pipeline. Then, using the self-excited vibration generated during pumping system operation as a detection source, a piezoelectric accelerometer array deployed in the pump body, reversing valve assembly, and delivery pipeline continuously acquires broadband vibration signals. The signal processing unit performs adaptive filtering and beamforming processing on the acquired vibration signals to extract the vibration echo signals reflected from the heterogeneous fluid interface. A propagation model of vibration waves in concrete medium is established. By combining the time difference of receiving the same echo signal from different sensors, the spatial position of the interface is calculated using a triangulation algorithm. At the same time, the energy attenuation characteristics of the echo signal are analyzed, and the acoustic impedance difference between the concrete on both sides of the interface is inferred as a quantitative indicator of the physical properties of the interface.

[0035] The delivery pipeline is constructed as a fixed three-dimensional Eulerian grid in digital space, and the located heterogeneous fluid interface is discretized into multiple virtual Lagrange particles carrying interface properties. Based on the real-time flow rate of the pumping system, the fluid dynamics equations are solved on the Eulerian grid. The background flow field distribution of the entire pipeline is obtained by solving the equations. Within each time step, the background flow field drives the motion of the virtual Lagrange particles, tracking the macroscopic displacement of the interface. Simultaneously, the relative positional changes between particles are calculated based on the velocity gradient and shear stress of the flow field, predicting the tensile and dispersion deformation trends of the interface. This Eulerian-Lagrange coupling method can accurately describe the dynamic evolution of heterogeneous fluid interfaces in complex pipelines.

[0036] Based on a digital geometric model of the delivery pipeline, virtual monitoring surfaces are deployed at key locations such as bends, reducers, and valve assemblies. In each iteration of the dynamic evolution model, it is monitored whether virtual Lagrange particles cross these monitoring surfaces. When a target particle crosses a monitoring surface, the crossing time is recorded as the arrival time of the heterogeneous fluid interface. A monitoring control volume is defined at the virtual monitoring surface. When a cloud of particles enters, the average rheological parameters of the fluid within the control volume are updated based on the interface properties carried by the particles. Combining the updated rheological parameters and the background flow field velocity, the momentum change and viscous dissipation increment caused by the change in fluid properties are calculated. Substituting these changes into the momentum conservation equation, the pressure rise at the inlet section of the monitoring control volume is obtained, serving as the estimated pressure impact amplitude. Next, combining the predicted arrival time of the heterogeneous fluid interface and the estimated pressure impact amplitude, an adaptive pressure control curve incorporating various execution parameters of the pumping system is generated. The control curve covers the time series of parameters such as pumping frequency, pumping pressure, directional valve switching sequence, and buffer device operation. By optimizing the algorithm to determine the nonlinear variation of each parameter on the time axis, the pumping system can smoothly transition the heterogeneous fluid interface within a preset time window before and after the interface reaches the critical position. The design principle of the control curve is to mitigate pressure shocks by appropriately reducing the pumping frequency and increasing the buffer time, while ensuring pumping continuity.

[0037] Finally, within the preset time window corresponding to the predicted arrival time of the heterogeneous fluid interface, the pumping system is driven to operate strictly according to the generated adaptive pressure control curve. The lead time of the time window is used to adjust the system to the optimal operating state in advance, and the follow time ensures that the system recovers smoothly after the interface has completely passed the critical position. During execution, various operating parameters of the pumping system, including pumping pressure, flow rate, and vibration amplitude, are monitored in real time and compared with the set values ​​of the control curve. When the actual parameters deviate from the set values ​​beyond the allowable range, a feedback correction mechanism is activated to dynamically adjust the control parameters to maintain system stability. At the same time, the actual position of the heterogeneous fluid interface is continuously tracked to verify the prediction accuracy and provide empirical data for the next control.

[0038] In another implementation, the stable pumping process achieved through the adaptive pressure control curve described above can be integrated into a higher-level automated pouring control process. Specifically, after a preset construction task is initiated, the central controller first resolves the three-dimensional spatial coordinate sequence of multiple target pouring points based on the input Building Information Model (BIM) or a preset construction path plan. The controller then drives the boom system of the truck-mounted pump, coordinating the hydraulic cylinder movements of each boom section to precisely move the end hose of the concrete delivery pipe to the first preset pouring point in the sequence. After positioning, the pumping system starts and begins executing the complete adaptive pressure control method described in S1 to S6 above to stably deliver concrete to the pouring area corresponding to the pouring point. The controller determines whether the task at the current point is completed based on the preset pouring duration or preset pouring volume associated with that pouring point. When the start time reaches the preset duration or the pumping volume reaches the preset volume, the controller instructs the pumping system to pause or stop. Subsequently, the controller queries whether there is a next preset pouring point in the task sequence. If no next pre-set pouring point exists, the entire automated pouring task is complete, and the controller will instruct the boom system to execute an automatic cleaning and retraction procedure. If a next pre-set pouring point exists, the controller will update this next pre-set pouring point to the current target pouring point and repeat the steps of controlling the concrete delivery pipe of the truck pump to reach the new pre-set pouring point and restarting the pumping system in adaptive pressure control mode, until the pouring task of all pre-set pouring points has been traversed and completed.

[0039] In another embodiment, the hopper sensor array includes a 3D vision sensor and an immersion rheological sensor. When the hopper sensor array, arranged within the hopper, detects the presence of newly injected concrete, the steps of identifying the rheological property differences between the newly injected concrete and the existing concrete in the hopper, and generating a heterogeneous fluid interface within the pumping system's delivery pipeline based on these rheological property differences, include the following steps: When a new concrete is detected being injected into the hopper by a 3D vision sensor positioned above the hopper, the 3D vision sensor continuously collects three-dimensional point cloud data of the concrete surface inside the hopper. The slump of the newly poured concrete was calculated based on the three-dimensional point cloud data. The viscosity and yield stress parameters of freshly poured concrete are measured by an immersion rheological sensor installed inside the hopper. The rheological characteristic vector of the newly poured concrete is obtained by integrating slump, viscosity and yield stress parameters; The rheological feature vector of the newly injected concrete is quantitatively compared with the pre-stored reference rheological feature vector of the original concrete to calculate the difference in rheological properties between the newly injected concrete and the original concrete. When the difference in rheological properties exceeds a preset risk threshold, a heterogeneous fluid interface is determined to be generated in the delivery pipeline of the pumping system.

[0040] In the above embodiment, when a 3D vision sensor installed above the hopper, such as a depth camera based on structured light or time-of-flight principles, detects new concrete being injected, it immediately activates and performs high-frequency scanning of the concrete surface within the hopper. This process is not a single photograph, but rather a continuous and dynamic capture of the entire process of concrete from injection to gradual spreading. By projecting specific grating patterns or emitting laser pulses onto the concrete surface and analyzing the shape or time difference of the reflected light, the sensor can accurately calculate the three-dimensional spatial coordinates of thousands of points on the surface, thereby generating a series of high-density three-dimensional point cloud data arranged in chronological order. Based on the acquired continuous three-dimensional point cloud data, the slump of the newly injected concrete can be automatically calculated, a key indicator for measuring concrete fluidity. The principle is similar to the traditional slump cone test, performed in digital space through an algorithm. First, the algorithm identifies the initial outline and highest point of the pile formed when the new concrete is first injected, recording it as the initial height. Subsequently, under the influence of gravity, the concrete naturally spreads out, and the point cloud data reflects this process in real time. By analyzing the point cloud at subsequent time points, the reduction in the pile height or the final diameter of the unfolded base can be calculated. The slump S is quantified through these geometric changes; for example, it can be directly calculated as the initial peak height. and the peak height after stabilization The difference, i.e. .

[0041] While performing geometric property analysis, an immersion rheological sensor, positioned inside the hopper and in direct contact with the newly injected concrete, simultaneously measures its intrinsic rheological parameters. This sensor typically employs a paddle or auger design, rotating within the concrete according to a pre-programmed sequence, sensing the concrete's resistance through built-in torque and tachometers. Its measurement principle is based on fluid shear theory; by applying different shear rates and measuring the corresponding shear stresses, a rheological curve of the concrete can be plotted. Based on this curve, using a specific fluid model, such as the Bingham model, two core parameters can be analyzed: viscosity, representing the viscous resistance of concrete flow; and yield stress, representing the minimum driving force required to initiate flow from a static state. Concrete with different yield stresses exhibits different flow velocities during subsequent pipeline transport. Figures 2 to 4 As shown, Figures 2 to 4 The velocity distribution cloud maps of concrete with yield stresses of 150 Pa, 100 Pa and 80 Pa on the cross section of a horizontal straight pipe are shown respectively.

[0042] Then, the previously measured independent parameters are integrated into a unified mathematical expression. This process is achieved by constructing a new injection rheological eigenvector. This vector is a vector that includes slump S, viscosity, and other parameters. and yield stress An ordered array of key parameters is defined as a three-dimensional column vector. Next, the newly generated rheological feature vector is quantitatively compared with the pre-stored baseline rheological feature vector of the original concrete in the system to accurately calculate the degree of difference between the two. The baseline rheological feature vector of the original concrete is collected, recorded, and integrated by the hopper sensor array when the original concrete is injected into the hopper. The degree of difference between the two vectors is calculated by calculating the distance between them in the multidimensional feature space. Specifically, a weighted Euclidean distance can be used. The formula for calculating the rheological property difference D is as follows: .in, , , These are the weighting coefficients for slump, viscosity, and yield stress, respectively.

[0043] Finally, the calculated rheological property difference D is compared with a pre-set risk threshold. A comparison is made to determine the risk threshold. This threshold is a critical value set based on extensive pumping test data and engineering experience, representing the upper limit of concrete performance fluctuations allowed for the safe and stable operation of the pumping system. When the comparison results meet... Under this condition, it is determined that the performance difference between the old and new concrete is too large and has exceeded the safe range. This significant difference will form a heterogeneous fluid interface with a sudden change in physical properties when the old concrete and the newly poured concrete meet in the delivery pipeline of the pumping system.

[0044] In another embodiment, the self-excited vibration generated by the pumping system during operation is used as a detection source. The spatial location of the heterogeneous fluid interface in the delivery pipeline is determined in real time by inversion calculation of the vibration echo signal formed after the self-excited vibration propagates in the concrete and is reflected by the heterogeneous fluid interface. The steps include the following: Multiple piezoelectric accelerometers are arranged in the pump body, reversing valve group and delivery pipeline of the pumping system to form a vibration sensing array; The broadband vibration signal generated by the self-excited vibration of the pumping system during operation is continuously collected through a vibration sensor array. The signal processing unit of the pumping system processes the broadband vibration signal and extracts the vibration echo signal generated by the reflection from the interface of heterogeneous fluids. A vibration propagation model of vibration waves in concrete medium is established, and the vibration echo signal is input into the vibration propagation model. Based on the propagation model, the vibration echo signal is inverted and calculated to obtain the spatial position of the heterogeneous fluid interface.

[0045] In the above embodiment, multiple highly sensitive piezoelectric accelerometers are tightly fixed to the outer walls of the pump body, hydraulic directional valve assembly, and delivery pipeline. The core principle of piezoelectric sensors is to utilize the piezoelectric effect of their internal crystals. When the pipeline wall experiences minute vibrations due to internal pressure fluctuations and mechanical motion, this mechanical stress is converted into a measurable electrical charge signal. By spatially distributing these sensors to form an array, not only can the initial signal of the vibration source be captured, but signals from the propagation and reflection of vibration waves in the pipeline can also be received from different directions and distances. After the pumping operation begins, the vibration sensor array operates continuously, constantly acquiring the self-excited vibration signals generated by the pumping system itself under normal operating conditions. This self-excited vibration originates from various physical processes such as the reciprocating motion of the pump core, the instantaneous opening and closing of hydraulic valves, and the friction of concrete flowing within the pipeline. Its characteristic is a wide frequency range, containing rich components from low to high frequencies, forming a broadband vibration signal. These vibration waves, with the pump body and valve assembly as the main sources, propagate to distant ends along the pipeline structure and the internal concrete medium.

[0046] However, the acquired raw broadband vibration signal is extremely complex, with the strong source vibration signal masking the extremely weak echo signal reflected from the heterogeneous fluid interface. Therefore, it is necessary to perform refined processing using a dedicated signal processing unit of the pumping system to extract the useful vibration echo. Specifically, an adaptive filtering technique is employed, using the sensor signal near the vibration source as reference noise. An algorithm is used to eliminate components related to the source vibration in other sensor signals in real time, thereby highlighting the abnormal signal portion caused by reflection. This separated weak signal is the vibration echo signal carrying information about the location and characteristics of the heterogeneous fluid interface. To calculate the exact spatial location of the heterogeneous fluid interface, inversion calculation is required. This process first establishes a mathematical model of the vibration wave propagating in concrete, describing the vibration wave's velocity, attenuation, and other characteristics. Then, the pure vibration echo signal extracted in the previous step is used as input. The core of the inversion calculation is utilizing the signal arrival time difference. Assume that the times at which two sensors i and j at different locations receive the same reflected echo are respectively... and The time difference is Based on the known wave propagation speed *c* in concrete and the precise position of the sensor in the pipe coordinate system, a set of hyperbolic equations about the interface location can be established. By solving these equations, the three-dimensional spatial coordinates of the reflection source, i.e., the interface between heterogeneous fluids, can be derived.

[0047] In another implementation, the spatial location of the heterogeneous fluid interface is obtained by inverting the vibration echo signal based on the propagation model, including the following steps: Adaptive filtering is applied to the vibration echo signal, and a beamforming algorithm is used to process the vibration echo signal. By combining the time difference of the same vibration echo signal received by different sensors in the vibration sensing array and the sound velocity parameter in the propagation model, the spatial position of the heterogeneous fluid interface in the delivery pipeline is calculated by the triangulation algorithm. By analyzing the energy attenuation characteristics of the vibration echo signal, the acoustic impedance difference between the concrete on both sides of the heterogeneous fluid interface is inferred, and the acoustic impedance difference is used as a quantitative indicator of the physical properties of the heterogeneous fluid interface and correlated with the spatial location.

[0048] In the above implementation, predictable and regular background noise, such as the periodic vibration generated by the main pump of the pumping system, is first eliminated through adaptive filtering. Then, a beamforming algorithm is used to spatially filter the multiple signals acquired by the sensor array. This algorithm applies a precisely calculated, minute time delay to each signal and then superimposes them. This results in signals from a specific direction—the reflection direction from the heterogeneous fluid interface—being amplified in phase due to phase alignment, while interference signals from other directions cancel each other out due to phase misalignment. The output signal of the beamforming algorithm is shown below. It can be represented as Where N is the number of sensors, These are the weighting coefficients for the i-th sensor. It is the signal it receives. This is to focus on direction The time delay applied to the i-th signal.

[0049] After obtaining a clear echo signal, the three-dimensional spatial position of the heterogeneous fluid interface can be accurately calculated using a triangulation algorithm. The core principle of this method is to utilize the time difference between the arrival times of the echo signal at sensors at different spatial positions. For any two sensors in the array, the time difference between receiving the same reflected echo... Multiplying this by the propagation speed *c* of the vibration wave in the concrete medium gives the difference in distance between the interface and the two sensors. This relationship can be expressed by the formula... To describe, among which and These represent the straight-line distances from the heterogeneous fluid interface to the i-th and j-th sensors, respectively. In three-dimensional space, the set of points satisfying this distance difference forms a hyperboloid. By simultaneously solving multiple hyperboloid equations generated by different sensor pairs and finding their unique intersection point, the precise spatial coordinates of the reflection source, i.e., the heterogeneous fluid interface, can be derived. This transforms multi-channel temporal information into precise physical spatial information, enabling non-contact three-dimensional positioning of the invisible interface inside the pipe.

[0050] When a vibrational wave is reflected at the interface between two different media, the energy intensity of the reflected wave directly depends on the difference in acoustic impedance between the two media. Acoustic impedance is the product of material density and sound velocity, and is an important physical quantity that measures the ability of a medium to propagate sound waves. The greater the difference in acoustic impedance, the stronger the reflection, and the stronger the energy of the echo signal. The reflection coefficient R can be obtained by calculating the ratio of the amplitude of the reflected echo to the estimated amplitude of the incident wave. According to acoustic theory, the reflection coefficient R is related to the acoustic impedance of the media on both sides of the interface. and The relationship between them is This formula can be used to inversely calculate the difference in acoustic impedance. This allows us to assign a key physical property quantification index to the previously located spatial points, namely the acoustic impedance difference, so that the interface in the digital model not only has a geometric position, but also a description of physical intensity.

[0051] In another implementation, constructing a dynamic evolution model of the heterogeneous fluid interface based on spatial location, and predicting the motion and deformation trend of the heterogeneous fluid interface in the delivery pipeline by solving the dynamic evolution model includes the following steps: The delivery pipeline is constructed as a fixed three-dimensional Eulerian mesh in digital space; The identified heterogeneous fluid interface is discretized into multiple virtual Lagrange particles carrying interface properties, and all virtual Lagrange particles are mapped to the initial position of the three-dimensional Eulerian mesh. Based on the real-time pumping flow rate of the pumping system, the fluid dynamics equations are solved on a three-dimensional Eulerian grid to obtain the background flow field distribution of the entire pipeline in the delivery pipeline. Within each preset time step, the motion of virtual Lagrange particles is driven by the background flow field distribution to track the macroscopic displacement of the heterogeneous fluid interface; Based on the velocity gradient and shear force of the background flow field distribution, the relative position changes between each virtual Lagrange particle are calculated simultaneously, and the stretching trend and dispersion deformation trend of the heterogeneous fluid interface are predicted.

[0052] In the above implementation, a computational framework is constructed for the physical pipeline in a digital twin space to achieve accurate simulation of the fluid dynamics within the pipeline. Specifically, the continuous three-dimensional geometric model of the pipeline is discretized into a set of numerous fixed small cubes or tetrahedral elements, forming a fixed three-dimensional Eulerian mesh. This is similar to setting up a static observation network in space, focusing on observing changes in physical quantities such as fluid velocity and pressure at each fixed mesh node rather than tracking individual fluid particles. The complex pipeline space is thus transformed into a structured discrete domain that facilitates numerical solutions by the computer. After constructing the fixed computational mesh, the physically heterogeneous fluid interfaces previously located using acoustic methods need to be transformed into virtual objects that can be tracked within the digital mesh. This is achieved by abstracting and discretizing this continuous interface into a point cloud composed of thousands of independent virtual Lagrange particles. Unlike the fixed nature of the Eulerian mesh, the Lagrange particles move with the fluid. Each virtual particle not only records its initial spatial coordinates but is also endowed with key interface physical properties, such as the previously calculated acoustic impedance difference. Subsequently, these attribute-carrying particles were precisely mapped onto a three-dimensional Eulerian grid, and their initial positions were completely consistent with the results of physical detection and localization.

[0053] Next, the background flow field distribution needs to be calculated to allow the virtual Lagrange particles to move realistically. This process is based on the principles of classical computational fluid dynamics (CFD). On an established three-dimensional Eulerian grid, the Navier-Stokes equations describing fluid motion are solved numerically. The input conditions for the calculation are boundary parameters such as pump flow rate, which are fed back in real time by sensors in the pumping system. Through iterative solving, the pressure value and velocity vector at each element and node in the grid can be calculated, thus obtaining a complete velocity and pressure distribution map of the concrete inside the entire delivery pipeline at a certain instant, i.e., the background flow field. Its core momentum equation can be simplified as follows: ,in It is the fluid density. It is a velocity vector, and P is pressure. It is dynamic viscosity. For physical strength.

[0054] With the background flow field, all virtual Lagrangian particles representing heterogeneous fluid interfaces can be precisely driven to move within each preset, tiny time step. At any given moment, each particle queries the local fluid velocity from its Eulerian grid position, and then moves at this velocity for a short time to reach a new position. This process is iteratively implemented using a simple numerical integration, where the position of any particle p at the next time step is determined. From current location and the flow velocity at the current location Decision, that is By repeating this process for all particles, the entire particle cloud migrates within the digital pipeline, enabling real-time tracking and prediction of the macroscopic overall displacement of the heterogeneous fluid interface. However, simply tracking macroscopic displacement is insufficient, as the fluid interface undergoes shape changes during motion. Therefore, it is also necessary to analyze the local differences in the surrounding background flow field velocity, i.e., the velocity gradient, while driving the particle motion. Figure 5 As shown, the flow velocity is high at the center of the pipe and slows down near the pipe wall. This velocity difference generates shear force, causing the interface to be stretched. By calculating the velocity gradient and shear force in the region where the particle cloud is located, this deformation effect can be quantified, and the relative positions between the virtual Lagrangian particles can be adjusted accordingly. For example, the shear stress generated by the velocity gradient is proportional to the rate of velocity change, and this force will cause adjacent particles to shift relative to each other.

[0055] In another embodiment, the method further includes the following steps: To obtain the spatial location and acoustic impedance difference at the interface of heterogeneous fluids; The interface of heterogeneous fluids is discretized into multiple virtual Lagrange particles, and the initial spatial distribution of all virtual Lagrange particles is generated in a three-dimensional Eulerian grid based on their spatial location. The acoustic impedance difference is assigned as a physical interface property to all virtual Lagrange particles. In solving the dynamic evolution model, the physical properties are used to distinguish the physical behavior of the fluids on both sides of the heterogeneous fluid interface, and the dispersion rate of the heterogeneous fluid interface is calculated by combining the shear stress applied to the virtual Lagrange particle cloud by the background flow field.

[0056] In the above implementation, the data for initiating the dynamic evolution model initially originates from the results of prior acoustic detection, namely, the precisely acquired three-dimensional spatial location information of the heterogeneous fluid interface and the acoustic impedance difference quantifying its physical properties. The spatial location information is a set of three-dimensional coordinate points describing the geometry of the interface within the pipeline, while the acoustic impedance difference is a single scalar value that profoundly reflects the degree of mismatch in density and elastic modulus between the concrete on both sides of the interface. These two parameters together constitute a complete snapshot of the actual interface state in the physical world. Next, based on the acquired spatial coordinates, an initial cloud of particles is generated within a fixed three-dimensional Eulerian mesh background, perfectly matching the shape of the actual interface. The position of each particle strictly corresponds to a point in the initial positioning data. Then, the previously calculated acoustic impedance difference is... This key physical quantity is assigned as a fixed attribute value to each newly generated virtual Lagrange particle in batches. This means that each particle p carries an attribute. This attribute value remains unchanged throughout the simulation.

[0057] During the iterative solution of the dynamic evolution model, the physical properties carried by the particles are activated and used. When calculating the background flow field, the solver will determine the location of the particle cloud and utilize its carried property values. This allows for local correction of fluid parameters within the computational domain, such as viscosity or yield stress, thereby distinguishing the different dynamic behaviors of the fluids on either side of the interface. More importantly, the background flow field imposes non-uniform shear stress on the particle cloud, which is the primary driving force for interfacial mixing and dispersion. This is achieved by establishing the dispersion rate... The relationship with shear stress and interface properties, such as ,in The dispersion coefficient allows for the real-time calculation of the interface mixing rate.

[0058] In another embodiment, predicting the arrival time of the heterogeneous fluid interface at a preset key location in the delivery pipeline and estimating the pressure impact based on the motion deformation trend and using a dynamic evolution model includes the following steps: Based on the geometric model of the delivery pipeline in digital space, virtual monitoring surfaces are set up at the bends, reducers and valve groups of the geometric model; In each iteration step of the dynamic evolution model, determine whether a target virtual Lagrange particle crosses the virtual monitoring surface; When the target virtual Lagrange particle crosses the virtual monitoring surface, the crossing time of the target virtual Lagrange particle is recorded, and the crossing time is used as the arrival time of the heterogeneous fluid interface to the preset key position of the delivery pipeline. The amplitude of the predicted pressure impact is calculated by combining the interface properties carried by the target virtual Lagrange particle and the pressure value of the background flow field.

[0059] In the above implementation, based on the existing three-dimensional digital geometric model of the delivery pipeline, in critical locations where pressure shocks are easily triggered due to abrupt changes in geometry, such as... Figure 6 As shown, a series of two-dimensional virtual monitoring surfaces are artificially arranged, such as at the corners of bends in pipes, the scaling sections of reducers, and inside valve assemblies. These monitoring surfaces are purely mathematical planes and do not exist in the physical world; their positions and normal directions are precisely defined, serving as tripwires in digital space. One way to define monitoring surfaces is using plane equations. ,in It is the unit normal vector of the monitored surface. It is the coordinate vector of any point on the surface. In each iterative calculation loop of the model, all virtual Lagrange particles representing the interface are traversed, and it is checked whether their positions in the old and new time steps cross any virtual monitoring surface. Specifically, the position of the particle before the start of the time step is calculated. and the position after the end The function values ​​relative to the equation of the same monitoring surface. If the product of two function values ​​is negative, that is... This indicates that the particle must have crossed the monitoring surface within this time step. Once a target virtual Lagrange particle is detected crossing a specific virtual monitoring surface, the model immediately performs a recording operation to determine the precise predicted time when the heterogeneous fluid interface reaches that critical location. The principle is very straightforward: when the crossing event is confirmed, the timestamp of the current simulation is captured. Typically, the crossing time of the first particle to cross the monitoring surface is defined as the arrival time of the entire interface front. This time is the time required from the current moment until the interface reaches the risk location in the future.

[0060] After determining the arrival time, it is also necessary to estimate the intensity of the pressure shock that this arrival event may trigger in order to determine the strength of the control strategy. Pressure shocks mainly arise from the abrupt change in momentum of the fluid at the point of geometric change in the pipe when fluid properties undergo abrupt changes (caused by the passage of heterogeneous fluid interfaces). In the calculation, the background pressure value at the monitoring surface location before the interface arrives is first extracted from the background flow field model. Then, combining the interface properties (such as acoustic impedance difference) carried by the virtual Lagrange particles crossing the monitoring surface, and the background flow field velocity at that time, the additional pressure increment caused by the change in fluid properties is calculated using the fluid momentum conservation equation. The final estimated peak pressure shock can be expressed as: .

[0061] In another implementation, calculating the estimated pressure impact amplitude by combining the interface properties carried by the target virtual Lagrange particle and the pressure value of the background flow field includes the following steps: Define the monitoring control volume at the virtual monitoring surface; When the virtual Lagrange particle cloud enters the monitoring and control volume, the average rheological parameters of the fluid within the monitoring and control volume are updated according to the interface attributes carried by the virtual Lagrange particles. By combining the updated average rheological parameters and the velocity information of the background flow field, the changes in momentum and the increment of viscous dissipation caused by changes in fluid properties and pipe geometry are calculated. Substituting the change in momentum and the increase in viscous dissipation into the momentum conservation equation, the momentum conservation equation is solved to obtain the pressure rise at the inlet section of the monitoring and control volume, and the pressure rise is used as the amplitude of the predicted pressure shock.

[0062] In the above implementation, a tiny three-dimensional region with a defined volume and boundaries, i.e., the monitoring and control volume, is first defined around each virtual monitoring surface in the digital twin model. This control volume is fixed in space and is typically set as a short pipe with clearly defined inlet and outlet sections and pipe wall boundaries. The principle is to extend the analysis from an infinitely thin two-dimensional plane to a three-dimensional space of finite thickness, thereby enabling the application of integral-form conservation laws in fluid mechanics, such as momentum and energy conservation. When the virtual Lagrange particle cloud representing the interface of heterogeneous fluids in the dynamic evolution model begins to enter this pre-defined monitoring and control volume, the physical properties of the fluid inside the control volume will dynamically change. The calculation principle is to calculate the volume ratio of the old and new concrete within the control volume in real time by statistically counting the number of particles entering the control volume. Based on this volume ratio, and using the rheological parameters associated with interface properties such as the difference in acoustic impedance carried by the particles, the average density, average viscosity, and other rheological parameters of the mixed fluid within the control volume are updated in real time using a weighted averaging method. For example, the average viscosity within the control volume... It is possible Calculation, where This represents the volume percentage of newly poured concrete within the control volume.

[0063] After obtaining the updated average rheological parameters, the key mechanical changes caused by the combined effects of fluid properties and pipe geometry can be accurately calculated. Combining the velocity distribution information provided by the background flow field, the momentum flux change when the fluid enters and exits the control volume is first calculated, i.e., the momentum change. Simultaneously, based on the updated average viscosity and the velocity gradient of the flow field within the control volume, the energy loss increment due to increased fluid viscosity, i.e., the viscous dissipation increment, is calculated; this part will be reflected as additional drag acting on the fluid. Finally, the calculated momentum change and the additional drag represented by the viscous dissipation increment are substituted into the integral form of the momentum conservation equation applied to the entire monitoring and control volume. This equation is essentially Newton's second law, which balances all forces acting on the control volume (pressure difference between inlet and outlet, shear force from the pipe wall) with the rate of momentum change of the fluid within the control volume. In the equation... China, export pressure Cross-sectional area A, viscous resistance and rate of change of momentum All terms are known or have been calculated. By solving this equation, the unique unknown inlet pressure can be determined. The increase in pressure relative to the initial background pressure is the estimated pressure shock amplitude. Ultimately, a precise, quantified pressure prediction is output, providing the highest priority decision-making basis for subsequent coordinated control systems.

[0064] The present invention also proposes an intelligent vehicle pump collaborative control system based on digital twins, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, it implements the method described in any of the above embodiments.

[0065] The processor can be a central processing unit (CPU). Of course, depending on the actual use, it can also be other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), off-the-shelf programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor can be a microprocessor or any conventional processor, etc., and this application does not limit it.

[0066] The memory can be an internal storage unit of a computer device, such as a hard disk or RAM, or an external storage device, such as a plug-in hard disk, smart memory card (SMC), secure digital card (SD), or flash memory card (FC) provided on the computer device. Furthermore, the memory can be a combination of internal storage units and external storage devices of a computer device. The memory is used to store computer programs and other programs and data required by the computer device. The memory can also be used to temporarily store data that has been output or will be output. This application does not limit this.

[0067] Furthermore, embodiments of the present invention also propose a storage medium storing a computer program that, when executed by a processor, implements the steps of the above-described method.

[0068] The logic and / or steps represented in the flowchart or otherwise described herein, for example, can be considered as a sequenced list of executable instructions for implementing logical functions, and can be embodied in any computer-readable medium for use by, or in conjunction with, an instruction execution system, apparatus, or device (such as a computer-based system, a processor-included system, or other system that can fetch and execute instructions from, an instruction execution system, apparatus, or device). For the purposes of this specification, "computer-readable medium" can be any means that can contain, store, communicate, propagate, or transmit programs for use by, or in conjunction with, an instruction execution system, apparatus, or device.

[0069] More specific examples of computer-readable media (a non-exhaustive list) include: electrical connections (electronic devices) having one or more wires, portable computer disk drives (magnetic devices), random access memory, read-only memory, erasable and editable read-only memory, fiber optic devices, and portable optical disc read-only memory. Furthermore, computer-readable media can even be paper or other suitable media on which the program can be printed, because the program can be obtained electronically, for example, by optically scanning the paper or other medium, followed by editing, interpreting, or otherwise processing as necessary, and then stored in computer memory.

[0070] It should be understood that various parts of the present invention can be implemented in hardware, software, firmware, or a combination thereof. In the above embodiments, multiple steps or methods can be implemented in software or firmware stored in memory and executed by a suitable instruction execution system. For example, if implemented in hardware, as in another embodiment, it can be implemented using any one or a combination of the following techniques known in the art: discrete logic circuits having logic gates for implementing logical functions on data signals, application-specific integrated circuits (ASICs) having suitable combinational logic gates, programmable gate arrays (FPGAs), field-programmable gate arrays (FPGAs), etc.

Claims

1. A collaborative control method for an intelligent vehicle pump based on digital twins, characterized in that, The method, applied to an intelligent vehicle pump, which includes a hopper and a pumping system, comprises the following steps: The feeding situation of the hopper is monitored by a hopper sensor array arranged in the hopper. When the presence of newly injected concrete is detected, the difference in rheological properties between the newly injected concrete and the original concrete in the hopper is identified, and a heterogeneous fluid interface in the delivery pipeline of the pumping system is generated based on the difference in rheological properties. Using the self-excited vibration generated by the pumping system in operation as a detection source, the spatial position of the heterogeneous fluid interface in the delivery pipeline is located in real time by inverting and calculating the vibration echo signal formed after the self-excited vibration propagates in the concrete and is reflected by the heterogeneous fluid interface. A dynamic evolution model of the heterogeneous fluid interface is constructed based on the spatial location, and the motion and deformation trend of the heterogeneous fluid interface in the delivery pipeline is predicted by solving the dynamic evolution model. Based on the motion deformation trend, the arrival time of the heterogeneous fluid interface to the preset key position of the delivery pipeline and the estimated pressure impact are predicted using a dynamic evolution model. An adaptive pressure control curve is generated by combining the arrival time and the estimated pressure impact, which is time-coordinated and includes multiple execution parameters of the pumping system. The pumping system is driven to operate according to the adaptive pressure control curve within the preset time window corresponding to the arrival time.

2. The method according to claim 1, characterized in that, The hopper sensor array includes a 3D vision sensor and an immersion rheological sensor. When the hopper sensor array, arranged within the hopper, detects newly injected concrete, the step of identifying the rheological property differences between the newly injected concrete and the existing concrete in the hopper, and generating a heterogeneous fluid interface within the pumping system's delivery pipeline based on these rheological property differences, includes the following steps: When the presence of newly injected concrete in the hopper is detected by the 3D vision sensor arranged above the hopper, the 3D vision sensor continuously collects three-dimensional point cloud data of the concrete surface in the hopper. The slump of the newly poured concrete is calculated based on the three-dimensional point cloud data. The viscosity and yield stress parameters of the freshly poured concrete are measured by the immersion rheological sensor installed inside the hopper. The rheological characteristic vector of the newly poured concrete is obtained by integrating the slump, viscosity and yield stress parameters. The rheological feature vector of the newly injected concrete is quantitatively compared with the pre-stored reference rheological feature vector of the original concrete to calculate the difference in rheological properties between the newly injected concrete and the original concrete. When the difference in rheological properties exceeds a preset risk threshold, it is determined that a heterogeneous fluid interface is generated in the delivery pipeline of the pumping system.

3. The method according to claim 1, characterized in that, The method of using the self-excited vibration generated by the pumping system during operation as a detection source, and calculating the vibration echo signal formed after the self-excited vibration propagates in the concrete and is reflected by the heterogeneous fluid interface to locate the spatial position of the heterogeneous fluid interface in the delivery pipeline in real time includes the following steps: Multiple piezoelectric accelerometers are arranged in the pump body, reversing valve group and delivery pipeline of the pumping system to form a vibration sensing array; The broadband vibration signal generated by the self-excited vibration of the pumping system during operation is continuously acquired through the vibration sensing array. The broadband vibration signal is processed by the signal processing unit of the pumping system to extract the vibration echo signal generated by the reflection from the interface of the heterogeneous fluids. A vibration propagation model of vibration waves in concrete medium is established, and the vibration echo signal is input into the vibration propagation model. Based on the propagation model, the vibration echo signal is inverted and calculated to obtain the spatial position of the heterogeneous fluid interface.

4. The method according to claim 3, characterized in that, The step of inverting the vibration echo signal based on the propagation model to obtain the spatial location of the heterogeneous fluid interface includes the following steps: The vibration echo signal is subjected to adaptive filtering and processed using a beamforming algorithm. By combining the time difference of the same vibration echo signal received by different sensors in the vibration sensing array and the sound velocity parameter in the propagation model, the spatial position of the heterogeneous fluid interface in the delivery pipeline is calculated by the triangulation algorithm. By analyzing the energy attenuation characteristics of the vibration echo signal, the acoustic impedance difference between the concrete on both sides of the heterogeneous fluid interface is inverted, and the acoustic impedance difference is used as a quantitative indicator of the physical properties of the heterogeneous fluid interface and correlated with the spatial location.

5. The method according to claim 4, characterized in that, The process of constructing a dynamic evolution model of the heterogeneous fluid interface based on the spatial location, and predicting the motion and deformation trend of the heterogeneous fluid interface in the delivery pipeline by solving the dynamic evolution model, includes the following steps: The delivery pipeline is constructed as a fixed three-dimensional Euler mesh in digital space; The located heterogeneous fluid interface is discretized into multiple virtual Lagrange particles carrying interface properties, and all the virtual Lagrange particles are mapped to the initial position of the three-dimensional Eulerian mesh. Based on the real-time pumping flow rate of the pumping system, the fluid dynamics equations are solved on the three-dimensional Eulerian grid to obtain the background flow field distribution of the entire pipeline in the delivery pipeline. Within each preset time step, the motion of the virtual Lagrange particles is driven by the background flow field distribution to track the macroscopic displacement of the heterogeneous fluid interface; Based on the velocity gradient and shear force of the background flow field distribution, the relative position changes between each virtual Lagrange particle are calculated simultaneously, and the stretching trend and dispersion deformation trend of the heterogeneous fluid interface are predicted.

6. The method according to claim 5, characterized in that, The method further includes the following steps: Obtain the spatial location and acoustic impedance difference of the heterogeneous fluid interface; The heterogeneous fluid interface is discretized into multiple virtual Lagrange particles, and an initial spatial distribution of all the virtual Lagrange particles is generated in the three-dimensional Eulerian grid based on their spatial positions. The acoustic impedance difference is assigned as a physical interface attribute to all the virtual Lagrange particles. In the process of solving the dynamic evolution model, the physical behavior of the fluids on both sides of the heterogeneous fluid interface is distinguished by the relevant physical properties, and the dispersion rate of the heterogeneous fluid interface is calculated by combining the shear stress applied to the virtual Lagrange particle cloud by the background flow field.

7. The method according to claim 5, characterized in that, The step of predicting the arrival time of the heterogeneous fluid interface to the preset key position in the delivery pipeline and estimating the pressure impact based on the motion deformation trend and using a dynamic evolution model includes the following steps: Based on the geometric model of the delivery pipeline in digital space, virtual monitoring surfaces are set up at the bends, reducers and valve groups of the geometric model; In each iteration step of the dynamic evolution model, it is determined whether a target virtual Lagrange particle crosses the virtual monitoring surface; When the target virtual Lagrange particle crosses the virtual monitoring surface, the crossing time of the target virtual Lagrange particle is recorded, and the crossing time is used as the arrival time of the heterogeneous fluid interface to the preset key position of the delivery pipeline. The amplitude of the predicted pressure impact is calculated by combining the interface properties carried by the target virtual Lagrange particle and the pressure value of the background flow field.

8. The method according to claim 7, characterized in that, The calculation of the estimated pressure impact amplitude by combining the interface properties carried by the target virtual Lagrange particle and the pressure value of the background flow field includes the following steps: Define a monitoring control body at the virtual monitoring surface; When the virtual Lagrange particle cloud enters the monitoring and control body, the average rheological parameters of the fluid within the monitoring and control body are updated according to the interface attributes carried by the virtual Lagrange particles. By combining the updated mean rheological parameters and the velocity information of the background flow field, the momentum change and viscous dissipation increment caused by the changes in fluid properties and pipe geometry are calculated. Substituting the momentum change and viscous dissipation increment into the momentum conservation equation, the momentum conservation equation is solved to obtain the pressure rise at the inlet section of the monitoring and control body, and the pressure rise is used as the amplitude of the predicted pressure impact.

9. A digital twin-based intelligent vehicle pump collaborative control system, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the computer program, it implements the method as described in any one of claims 1 to 8.

10. A storage medium storing computer instructions thereon, characterized in that, When executed by the processor, this instruction implements the steps of the method according to any one of claims 1 to 8.

Images