Drain pipe noise control system based on environmental data adjustment

Abstract

The invention provides a drainage pipe noise control system based on environmental data adjustment, which relates to the technical field of drainage pipe noise control and comprises a multi-dimensional data acquisition layer, a sound field and turbulence interactive modeling layer, an edge cooperative control layer and a central dynamic decision-making layer, the sound field and turbulence interaction modeling layer collects sound pressure data, vibration data, flow velocity data, temperature data, humidity data and air pressure data, and packages and transmits the data to the sound field and turbulence interaction modeling layer through the narrowband Internet of Things. According to the system, a sound pressure sensor, a vibration sensor and an ultrasonic flow velocity sensor are adopted, a phonon topology array and a sound field modulation transducer are combined, low-frequency sound wave characteristics are accurately captured, the lattice spacing of the phonon topology array is dynamically adjusted, anti-phase sound waves are emitted, low-frequency sound wave propagation is blocked, and pipeline resonance is weakened.

Description

Technical Field

[0001] This invention relates to the field of drainage pipe noise control technology, specifically a drainage pipe noise control system based on environmental data adjustment. Background Technology

[0002] With the acceleration of urbanization and the increase in building density, drainage pipes, as an important component of urban infrastructure, directly affect the quality of life of residents and the efficiency of industrial production. Low-frequency noise generated during the operation of drainage pipes not only disrupts residents' daily lives but may also cause pipe vibration, leading to structural fatigue and potential safety hazards. Therefore, effectively suppressing low-frequency noise within drainage pipes and improving the operational quality of pipeline systems has become an important research direction in the fields of building engineering and environmental engineering.

[0003] Currently, noise control in drainage pipes mainly relies on the following technical methods:

[0004] Sound-absorbing material wrapping: By wrapping the outer wall of the pipe with sound-absorbing materials, such as fiberglass or foam, some of the sound wave energy is absorbed, reducing noise transmission. This method is widely used in residential and commercial buildings, but its effect is mainly limited to mid-to-high frequency noise, and its ability to suppress low-frequency noise is limited.

[0005] Pipeline structure optimization: By designing sound-absorbing elbows or increasing pipe wall thickness, noise caused by fluid turbulence and vibration can be reduced. This method is effective in the initial stage of pipeline installation, but the cost of modifying existing pipelines is high, and it is difficult to adapt to dynamic changes in operating conditions.

[0006] Passive vibration isolation devices: Rubber pads or spring isolators are installed at pipe support points to reduce the transmission of vibration to the building structure. This method can partially reduce vibration noise, but it is not very effective at suppressing sound waves caused by the fluid inside the pipe.

[0007] Although existing noise control technologies have played a role in reducing noise in drainage pipes, they still have many shortcomings:

[0008] Problem 1: Sound-absorbing materials are not very effective at suppressing low-frequency noise. Low-frequency sound waves have long wavelengths and strong penetrating power, making it difficult for conventional sound-absorbing materials to effectively absorb them. As a result, the low-frequency noise problem in residential and commercial areas has not been completely solved.

[0009] Problem 2: Pipeline structure optimization and passive vibration isolation devices rely on fixed designs, making it difficult to adapt to changes in fluid velocity, pressure, or environmental conditions, such as a sudden increase in drainage or increased turbulence due to pipeline aging. This significantly reduces the noise reduction effect and the cost of retrofitting is high, making it difficult to implement on a large scale in existing pipelines.

[0010] Thirdly, existing technologies lack real-time monitoring and dynamic adjustment capabilities. Most methods are passive noise reduction and cannot adjust control strategies based on real-time changes in the sound field, turbulence, and environmental data within the pipeline. This results in unstable noise reduction effects under complex operating conditions, such as peak drainage or extreme weather, making it difficult to meet high-standard noise reduction requirements.

[0011] Therefore, a drainage pipe noise control system based on environmental data is needed to solve the above problems. Summary of the Invention

[0012] Technical problems to be solved

[0013] To address the shortcomings of existing technologies, this invention provides a drainage pipe noise control system based on environmental data adjustment, which solves the problems mentioned in the background.

[0014] Technical solution

[0015] To achieve the above objectives, the present invention provides the following technical solution: a drainage pipe noise control system based on environmental data adjustment, comprising a multi-dimensional data acquisition layer, a sound field and turbulence interaction modeling layer, an edge collaborative control layer, and a central dynamic decision-making layer. The specific configuration of the control system is as follows:

[0016] The multidimensional data acquisition layer deploys sensor arrays at bends and bifurcations of the drainage pipe to collect multidimensional data from the drainage pipe. This data is then encapsulated via narrowband IoT and transmitted to the sound field and turbulence interaction modeling layer. The sensor array includes sound pressure sensors, vibration sensors, ultrasonic flow velocity sensors, temperature sensors, humidity sensors, and air pressure sensors at the bends and bifurcations of the drainage pipe. These sensors collect sound pressure data, vibration data, flow velocity data, temperature data, humidity data, and air pressure data, respectively. The sound pressure sensors are arranged in a ring array with equal spacing along the inner wall of the pipe, with 4-6 sensors per group, fixed with M6 stainless steel bolts. The probes face the center of the pipe to capture low-frequency sound pressure fluctuations. The vibration sensors are arranged in a grid array along the outer edge of the inner wall of the pipe, with 3-5 sensors per group, spaced approximately 5 cm apart, fixed with M6 stainless steel bolts, and closely attached to the pipe surface to detect fluid flow. Vibrations caused by impact or turbulence; ultrasonic flow velocity sensors are symmetrically arranged at the inlet and outlet of straight pipe sections, with two sensors per group forming a through-beam array, fixed by stainless steel annular clamping brackets. The probes are parallel to the flow direction, with a spacing of 1.5 times the pipe diameter, measuring fluid velocity; temperature sensors are evenly distributed in a ring array along the outer circumference of the pipe, with 3-4 sensors per group, spaced approximately 15 cm apart, fixed by epoxy resin bonding, and the probes contact the pipe surface to monitor ambient temperature; humidity sensors are arranged in a dotted array on the outer wall of the pipe, close to the temperature sensors, with 2-3 sensors per group, spaced approximately 10 cm apart, fixed by epoxy resin bonding, and the probes are exposed to the air to collect humidity data; barometric pressure sensors are arranged in a dotted array on the top or side of the outer wall of the pipe, with 1-2 sensors per group, spaced approximately 20 cm apart, fixed by epoxy resin bonding, and the probes are connected to the outside air to monitor ambient air pressure. The array design, through a combination of ring, grid, through-beam, and dotted arrays, ensures comprehensive coverage of acoustic, vibration, fluid, and environmental parameters. The fixing method ensures stability, and the sensor spacing and number balance data coverage and deployment cost.

[0017] The acoustic field and turbulence interaction modeling layer receives data transmitted from the multidimensional data acquisition layer, processes the data through orthogonal matching tracking algorithm and mainstream shape analysis algorithm, constructs a three-dimensional tensor response model of acoustic-fluid-environment, encapsulates the model data through narrowband Internet of Things, and then transmits it to the edge collaborative control layer.

[0018] The edge collaborative control layer deploys distributed sensing and control nodes at bends and bifurcations of the drainage pipes, receives model data transmitted by the sound field and turbulence interaction modeling layer, generates phonon topology control signals, turbulence disturbance control signals, and sound field modulation control signals, and transmits the control signals to the central dynamic decision-making layer through narrowband Internet of Things.

[0019] The central dynamic decision-making layer receives control signals transmitted from the edge collaborative control layer, generates a global noise reduction control strategy through a multi-objective optimization algorithm, converts it into standard semantic control commands, and sends them to the distributed sensing and control nodes of the edge collaborative control layer through narrowband IoT to coordinate phonon topology control, turbulence disturbance control, and sound field modulation control.

[0020] Preferably, the overall control flow of the control system is as follows: the multi-dimensional data acquisition layer acquires data and transmits it to the sound field and turbulence interaction modeling layer; the sound field and turbulence interaction modeling layer generates a three-dimensional tensor response model and transmits it to the edge collaborative control layer; the edge collaborative control layer generates control signals and transmits them to the central dynamic decision-making layer; the central dynamic decision-making layer generates a global control strategy and issues control commands to the edge collaborative control layer; the edge collaborative control layer then drives the noise reduction device; and the edge collaborative control layer provides real-time feedback of execution data to the central dynamic decision-making layer to update the control strategy.

[0021] Preferably, the acoustic field and turbulence interaction modeling layer constructs a three-dimensional tensor response model of acoustic-fluid-environment through the following steps:

[0022] A1. The parser receives data transmitted from the multidimensional data acquisition layer and parses it into acoustic feature vectors, fluid feature vectors, and environmental feature vectors.

[0023] A2. Decompose the acoustic feature vectors using the orthogonal matching pursuit algorithm to generate the sound field distribution matrix;

[0024] A3. By compressing the fluid feature vector through mainstream shape analysis algorithms, a low-dimensional Riemannian manifold representation is generated, reducing the dimensionality to 5% of the original data;

[0025] A4. Based on the sound field distribution matrix, low-dimensional Riemannian manifold representation, and environmental feature vectors, a three-dimensional tensor response model of acoustic-fluid-environment is constructed.

[0026] Preferably, the distributed sensing and control nodes of the edge collaborative control layer operate as follows:

[0027] B1. A three-dimensional tensor response model transmitted by the acoustic field and turbulence interaction modeling layer via narrowband Internet of Things;

[0028] B2. Process the three-dimensional tensor response model using the covariance matrix decomposition algorithm to generate sound field distribution data and turbulence mode data;

[0029] B3. A phonon topology array, a miniature ultrasonic transducer, and a sound field modulation transducer are configured. All three components are noise reduction devices. The phonon topology array hardware includes an electromagnetic actuator. The distributed sensing and control node generates a phonon topology control signal based on sound field distribution data to drive the electromagnetic actuator of the phonon topology array. The phonon topology array is a periodic acoustic structure made of composite materials. A turbulence disturbance control signal is generated based on turbulence mode data to drive the miniature ultrasonic transducer, which is a ceramic transducer. A sound field modulation control signal is generated based on sound field distribution data to drive the sound field modulation transducer.

[0030] B4. Control signals are encapsulated and transmitted to the central dynamic decision-making layer via narrowband Internet of Things.

[0031] Preferably, the edge collaborative control layer generates control signals through the following steps:

[0032] C1. A three-dimensional tensor response model transmitted by the acoustic field and turbulence interaction modeling layer via narrowband Internet of Things;

[0033] C2. The three-dimensional tensor response model is decomposed by the covariance matrix decomposition algorithm to generate the dominant frequency and phase in the sound field distribution data, and the phonon topology control signal is generated to drive the electromagnetic actuator to adjust the lattice spacing of the phonon topology array to 0.05 mm.

[0034] C3. Extract boundary layer features from turbulence mode data using mainstream shape analysis algorithm, generate turbulence disturbance control signal to drive the micro ultrasonic transducer to apply disturbance to an amplitude of 0.1 mm;

[0035] C4. Extract the spatial distribution characteristics of the sound field distribution data through the spatiotemporal total variational regularization algorithm, and generate a sound field modulation control signal to drive the sound field modulation transducer to emit anti-phase sound waves.

[0036] C5. Control signals are encapsulated and transmitted to the central dynamic decision-making layer via narrowband Internet of Things.

[0037] Preferably, the central dynamic decision-making layer generates a global noise reduction control strategy through the following steps:

[0038] D1. Receive phonon topology control signals, turbulence disturbance control signals and sound field modulation control signals transmitted by the edge collaborative control layer through narrowband Internet of Things;

[0039] D2. Calculate the global noise reduction control strategy based on the control signal using a multi-objective optimization algorithm, and generate a control instruction set that includes phonon topology adjustment parameters, turbulence disturbance parameters, and sound field modulation parameters.

[0040] D3. Encapsulate the control instruction set through narrowband Internet of Things and transmit it to the central dynamic decision-making layer;

[0041] D4. Receive execution feedback data transmitted from the edge collaborative control layer, update the control instruction set, and update the control instruction set with an update period of 5 milliseconds.

[0042] Preferably, the central dynamic decision-making layer converts the control instruction set into standard semantic control instructions using a semantic conversion algorithm before encapsulating the control instruction set.

[0043] Preferably, the distributed sensing and control nodes of the edge collaborative control layer achieve inter-node interaction through the following steps:

[0044] E1. Each distributed sensing and control node receives phonon topology control signals, turbulence disturbance control signals, and sound field modulation control signals from neighboring nodes via narrowband Internet of Things.

[0045] E2. Analyze the control signals of adjacent nodes using the covariance matrix decomposition algorithm, and adjust the phonon topology control signal, turbulence disturbance control signal, and sound field modulation control signal of this node.

[0046] E3. The adjusted control signals are encapsulated and transmitted to adjacent nodes and the central dynamic decision-making layer via narrowband IoT;

[0047] E4. Receive the global noise reduction control strategy transmitted from the central dynamic decision-making layer and update the interaction parameters between nodes;

[0048] This is mainly used for adjusting multiple sets of drain pipes.

[0049] Preferably, the sound field and turbulence interaction modeling layer optimizes the model through the following steps:

[0050] F1. Update the three-dimensional tensor response model of acoustic-fluid-environment using tensor decomposition algorithm to extract new dominant interaction factors;

[0051] F2. The updated three-dimensional tensor response model was verified by the covariance matrix factorization algorithm, with the error controlled within 1%.

[0052] F3. The updated 3D tensor response model is encapsulated via narrowband IoT and transmitted to the edge collaborative control layer.

[0053] The phonon topology array is fixed to the inner wall of the bend and fork of the drainage pipe by bolts. The micro ultrasonic transducer and the sound field modulation transducer are fixed to the straight pipe section by stainless steel annular clamping brackets. The phonon topology array receives the phonon topology control signal transmitted by the edge cooperative control layer and adjusts the lattice spacing by the electromagnetic actuator. The phonon bandgap principle is used to block the propagation of sound waves caused by the vibration of the inner wall of the drainage pipe.

[0054] The miniature ultrasonic transducer receives the turbulence disturbance control signal transmitted by the edge collaborative control layer, applies ultrasonic disturbance with an amplitude of 0.1 mm, and uses the fluid dynamics boundary layer control principle to change the turbulence characteristics of the fluid boundary layer in the drainage pipe, thereby reducing vibration noise caused by turbulence.

[0055] The sound field modulation transducer receives the sound field modulation control signal transmitted by the edge collaborative control layer and emits anti-phase sound waves to weaken the resonant sound waves in the drainage pipe by using the principle of sound wave interference.

[0056] The execution feedback data transmitted by the edge collaborative control layer is received via narrowband IoT to update the lattice spacing parameters of the phonon topology array, the perturbation amplitude parameters of the micro ultrasonic transducer, and the anti-phase acoustic wave parameters of the acoustic field modulation transducer, with an adjustment period of 5 milliseconds.

[0057] Beneficial effects

[0058] This invention provides a drainage pipe noise control system based on environmental data adjustment. It has the following beneficial effects:

[0059] This invention effectively solves the problem of poor low-frequency noise suppression by traditional sound-absorbing materials through real-time monitoring of the sound field, vibration, and fluid characteristics within drainage pipes. The system employs sound pressure sensors, vibration sensors, and ultrasonic flow velocity sensors, combined with a phonon topology array and a sound field modulation transducer, to accurately capture low-frequency acoustic characteristics. It dynamically adjusts the lattice spacing of the phonon topology array and emits anti-phase sound waves, blocking the propagation of low-frequency sound waves and weakening pipe resonance. An orthogonal matched pursuit algorithm is used to analyze acoustic characteristics and generate a sound field distribution matrix, significantly improving the low-frequency noise suppression effect. Experimental verification shows that the system can significantly reduce low-frequency noise in residential, industrial, and chemical pipeline scenarios, outperforming traditional sound-absorbing materials. It solves the problem of strong penetration and difficulty in absorption of low-frequency sound waves, improving the quality of life for residents and the comfort of industrial environments.

[0060] The system dynamically generates control strategies by real-time acquisition of sound field, turbulence, and environmental data, combined with mainstream shape analysis and covariance matrix decomposition algorithms. This solves the problem of pipeline structure optimization and passive vibration isolation devices being unable to adapt to dynamic operating conditions. Sensors monitor parameters such as flow velocity, temperature, humidity, and air pressure, forming multi-dimensional feature vectors. A three-dimensional tensor response model is constructed through an interactive modeling layer of sound field and turbulence, driving a miniature ultrasonic transducer to optimize the fluid boundary layer and adapt to complex conditions such as sudden increases in drainage volume or pipeline aging. The algorithm automatically identifies turbulence modes and vibration characteristics, generating precise control signals without requiring large-scale pipeline modifications, thus reducing implementation costs. Experimental verification shows that the system maintains stable noise reduction performance under various pipeline materials and fluid conditions, overcoming the limitations of traditional fixed designs. Attached Figure Description

[0061] Figure 1 This is a system control flowchart of the present invention;

[0062] Figure 2 This is a system framework diagram of the present invention. Detailed Implementation

[0063] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention. Specific Implementation Example 1:

[0065] like Figure 1-2 As shown, this invention provides a drainage pipe noise control system based on environmental data adjustment. It aims to accurately suppress low-frequency noise (20-200 Hz) within drainage pipes through multi-dimensional environmental data acquisition, sound field and turbulence interaction modeling, distributed collaborative control, and global dynamic decision-making. The system utilizes a phonon topology array, a miniature ultrasonic transducer, and a sound field modulation transducer, respectively employing the phonon bandgap principle, the fluid dynamics boundary layer control principle, and the acoustic wave interference principle to achieve end-to-end noise reduction from noise source generation to propagation path. The following details the system's working principle, installation method, data acquisition, analysis, processing, noise reduction execution, and feedback process.

[0066] System overall working principle:

[0067] The core of this system lies in constructing a dynamic interaction model of acoustics, fluid, and environment within drainage pipes through real-time acquisition and analysis of acoustic, fluid, and environmental data. This model generates precise noise reduction control signals and drives the noise reduction device through distributed collaborative control and global dynamic decision-making, achieving dynamic suppression of low-frequency noise. The system consists of a multi-dimensional data acquisition layer, a sound field and turbulence interaction modeling layer, an edge collaborative control layer, and a central dynamic decision-making layer. It achieves efficient data interaction via narrowband IoT using JSON format data packets, forming a closed-loop control process. The specific workflow is as follows: The multi-dimensional data acquisition layer collects sound pressure, vibration, flow velocity, temperature, humidity, and air pressure data, and transmits them to the sound field and turbulence interaction modeling layer; the sound field and turbulence interaction modeling layer generates a three-dimensional tensor response model, which is transmitted to the edge collaborative control layer; the edge collaborative control layer generates phonon topology control signals, turbulence disturbance control signals, and sound field modulation control signals, drives the noise reduction device, and transmits them to the central dynamic decision layer; the central dynamic decision layer generates a global noise reduction control strategy, converts it into standard semantic control commands, and sends them to the edge collaborative control layer to drive the noise reduction device. At the same time, it receives execution feedback data to dynamically update the control strategy, ensuring that the system adapts to dynamic operating conditions such as flow rate changes and temperature fluctuations, achieving efficient noise reduction, and is suitable for noise regulation of multiple drainage pipes.

[0068] System hardware installation method:

[0069] The system's hardware installation includes a sensor array and noise reduction devices, precisely deployed at bends, bifurcations, and straight sections of the drainage pipe. The sensor array includes sound pressure sensors, vibration sensors, ultrasonic flow velocity sensors, temperature sensors, humidity sensors, and air pressure sensors. Sound pressure sensors are arranged in a ring array with equal axial spacing of approximately 10 cm along the inner wall of the pipe, with 4-6 sensors per group, fixed with M6 stainless steel bolts at a tightening torque of 10 N·m. The probes face the center of the pipe, capturing low-frequency sound pressure fluctuations of 20-200 Hz. Vibration sensors are arranged in a grid array along the outer edge of the inner wall of the pipe, with 3-5 sensors per group, spaced approximately 5 cm apart, fixed with M6 stainless steel bolts, close to the pipe surface, detecting vibration acceleration caused by fluid impact or turbulence. Ultrasonic flow velocity sensors are symmetrically arranged at the inlet and outlet of straight pipe sections, with 2 sensors per group forming a through-beam array, fixed with stainless steel ring clamping brackets at a locking torque of 8 N·m. The probes are parallel to the flow direction, with a spacing of 1.5 times the pipe diameter, measuring the fluid velocity. Temperature sensors are evenly distributed in a ring array along the outer circumference of the pipe, with 3-4 sensors per group and a spacing of approximately 15 cm. They are fixed with epoxy resin, with an adhesive layer thickness of 0.2 mm and a curing time of 24 hours. The probes contact the pipe surface and monitor ambient temperatures from -20°C to 60°C. Humidity sensors are arranged in a dotted array near the temperature sensors on the outer wall of the pipe, with 2-3 sensors per group and a spacing of approximately 10 cm. They are also fixed with epoxy resin. The probes are exposed to the air and collect humidity data from 0% to 100% relative humidity. Barometric pressure sensors are arranged in a dotted array on the top or side of the outer wall of the pipe, with 1-2 sensors per group and a spacing of approximately 20 cm. They are fixed with epoxy resin, and the probes are connected to the outside air to monitor ambient air pressure. All sensors are connected to an analog-to-digital converter (ADC) via waterproof connectors. The ADC converts the analog signals into 16-bit precision digital signals, which are then encapsulated into JSON format data packets via a narrowband IoT module. The noise reduction device includes a phonon topology array, a miniature ultrasonic transducer, and a sound field modulation transducer. The phonon topology array, made of composite materials such as polycarbonate and silicone, includes electromagnetic actuators and is fixed to the inner wall of the drain pipe bends and bifurcations with M6 stainless steel bolts, tightening torque of 10 N·m. The miniature ultrasonic transducers are ceramic transducers, and the sound field modulation transducers are ultrasonic transmitters, both fixed to the straight pipe section with stainless steel annular clamping brackets, clamping torque of 8 N·m, and the emission direction parallel to the pipe axis. The noise reduction device is connected to the distributed sensing and control nodes of the edge collaborative control layer via waterproof cables, receiving control commands in JSON format.

[0070] Data collection:

[0071] The multi-dimensional data acquisition layer collects acoustic, fluid, and environmental parameters through sensor arrays deployed at bends, bifurcations, and straight sections of the drainage pipes. Sound pressure sensors measure changes in sound pressure within the pipe walls, outputting low-frequency sound pressure signals in the 20-200 Hz range to reflect noise propagation characteristics. Vibration sensors measure the vibration acceleration of the pipe walls, capturing mechanical vibrations caused by fluid impact or turbulence. Ultrasonic flow velocity sensors measure the fluid velocity within the pipe using the Doppler effect, acquiring turbulence characteristics. Temperature, humidity, and air pressure sensors measure the external temperature, humidity, and air pressure of the pipes, respectively, reflecting the environmental impact on the fluid and sound field. Each sensor collects data at a 20 kHz sampling frequency, converting analog signals to 16-bit digital signals using an analog-to-digital converter. The digital signals are encapsulated into JSON format data packets via a narrowband IoT module, containing timestamps, sensor numbers, and data values ​​such as sound pressure in decibels, vibration acceleration in meters per second squared, and flow velocity in meters per second, and transmitted to the sound field and turbulence interaction modeling layer.

[0072] Data Analysis and Processing:

[0073] The acoustic field and turbulence interaction modeling layer receives JSON format data packets, which are parsed by a parser into acoustic feature vectors based on sound pressure and vibration data, fluid feature vectors based on flow velocity data, and environmental feature vectors based on temperature, humidity, and air pressure data. An orthogonal matching pursuit algorithm performs sparse decomposition on the acoustic feature vectors to generate a sound field distribution matrix, describing the spatial distribution and temporal variation of sound pressure within the pipe, with a focus on analyzing low-frequency noise characteristics in the 20-200 Hz range. A mainstream manifold analysis algorithm performs differential geometric compression on the fluid feature vectors, generating a low-dimensional Riemannian manifold representation, reducing the dimensionality to 5% of the original data, and capturing core features of the turbulent boundary layer, such as boundary layer thickness and turbulence intensity. Based on the sound field distribution matrix, the low-dimensional Riemannian manifold representation, and the environmental feature vectors, the system constructs a three-dimensional tensor response model of acoustics-fluid-environment. The dominant interaction factors are extracted using the CP decomposition method to quantify the coupling relationships between acoustics and turbulence, and the environment, such as the correlation between sound pressure and flow velocity, and the influence of temperature on turbulence. The three-dimensional tensor response model is encapsulated into a JSON format data packet via narrowband IoT and transmitted to the edge collaborative control layer. To adapt to dynamic operating conditions, the acoustic field and turbulence interaction modeling layer receives execution feedback data from the edge collaborative control layer, updates the three-dimensional tensor response model through tensor decomposition algorithm, extracts new dominant interaction factors, and verifies the model error through covariance matrix decomposition algorithm, keeping it within 1% to ensure model accuracy and real-time performance.

[0074] Control signal generation and noise reduction execution:

[0075] The edge collaborative control layer deploys distributed sensing and control nodes at bends and bifurcations in the drainage pipes. These nodes receive JSON-formatted 3D tensor response models and decompose the models using a covariance matrix decomposition algorithm. This generates acoustic field distribution data (containing dominant frequencies and phases in the 20-200 Hz range) and turbulence mode data (containing boundary layer features such as turbulence intensity). Based on the acoustic field distribution data, a phonon topology control signal is generated to drive the electromagnetic actuator of the phonon topology array to adjust the lattice spacing to 0.05 mm, utilizing the phonon bandgap principle to block the propagation of sound waves caused by vibrations within the pipe wall. Based on the turbulence mode data, a mainstream shape analysis algorithm is used to extract boundary layer features, generating a turbulence disturbance control signal. This signal drives a micro-ultrasonic transducer to apply ultrasonic disturbances with an amplitude of 0.1 mm, altering the turbulence characteristics of the fluid boundary layer within the pipe through the fluid dynamics boundary layer control principle, thus reducing vibration noise. Finally, based on the acoustic field distribution data, a spatiotemporal total variational regularization algorithm is used to extract spatial distribution features, generating an acoustic field modulation control signal. This signal drives the acoustic field modulation transducer to emit anti-phase sound waves, weakening resonant sound waves within the pipe through the acoustic wave interference principle. Distributed sensing and control nodes receive control signals from neighboring nodes via narrowband IoT, including phonon topology, turbulence disturbances, and acoustic field modulation. They analyze these signals using a covariance matrix factorization algorithm and adjust their own control signals accordingly, ensuring consistency between acoustic interference and flow field disturbances. The adjusted control signals are then encapsulated into JSON format data packets and transmitted to neighboring nodes and the central dynamic decision-making layer.

[0076] Global decision-making and command coordination:

[0077] The central dynamic decision-making layer receives JSON-formatted control signals from the edge collaborative control layer, including phonon topology control signals, turbulence disturbance control signals, and sound field modulation control signals. It then uses a multi-objective optimization algorithm based on weighted least squares to calculate a global noise reduction control strategy, generating a control instruction set containing parameters such as phonon topology adjustment (lattice spacing), turbulence disturbance (disturbance amplitude), and sound field modulation (anti-phase acoustic wave phase). This control instruction set is converted into standard semantic control instructions in JSON format using a semantic conversion algorithm, including execution time, device number, and control parameters. These instructions are then distributed to the distributed sensing and control nodes of the edge collaborative control layer via narrowband IoT. The nodes execute the control instructions, driving the noise reduction device. Execution feedback data, including device operating status such as lattice spacing, disturbance amplitude, acoustic wave phase, noise level (sound pressure decibels), and flow field parameters (flow velocity changes), is encapsulated into JSON-formatted data packets via narrowband IoT and transmitted to the central dynamic decision-making layer and the sound field and turbulence interaction modeling layer. The central dynamic decision-making layer updates the control instruction set based on the feedback data, with an update cycle of 5 milliseconds, ensuring the system adapts to dynamic operating conditions.

[0078] Definition and working principle of noise reduction device:

[0079] The noise reduction device comprises three components: a phonon topology array made of composite materials such as polycarbonate and silicone, containing an electromagnetic actuator, fixed to the inner wall of the drain pipe bends and bifurcations with M6 stainless steel bolts; a phonon topology control signal that adjusts the lattice spacing to 0.05 mm to form a phonon bandgap, blocking the propagation of low-frequency sound waves in the 20-200 Hz range and suppressing noise caused by vibration of the pipe inner wall; a miniature ultrasonic transducer, a ceramic transducer fixed to a straight pipe section with a stainless steel annular clamping bracket; a turbulence disturbance control signal that applies ultrasonic disturbance with an amplitude of 0.1 mm, altering the turbulence characteristics of the fluid boundary layer within the pipe through the principle of fluid boundary layer control, thereby reducing vibration noise; and a sound field modulation transducer, an ultrasonic transmitter fixed to a straight pipe section with a stainless steel annular clamping bracket, that receives a sound field modulation control signal and emits anti-phase sound waves, weakening resonant sound waves within the pipe through the principle of sound wave interference.

[0080] Feedback and dynamic optimization:

[0081] The execution feedback data is generated by the edge collaborative control layer, including device operating status such as lattice spacing, disturbance amplitude, acoustic phase, noise level (sound pressure decibels), and flow field parameters (velocity changes). This data is encapsulated into JSON format data packets via narrowband IoT and transmitted to the central dynamic decision-making layer and the acoustic field and turbulence interaction modeling layer. The central dynamic decision-making layer updates the control command set using a multi-objective optimization algorithm, adjusting the lattice spacing of the phonon topology array, the disturbance amplitude of the micro-ultrasonic transducer, and the acoustic phase of the acoustic field modulation transducer, with an update cycle of 5 milliseconds. The acoustic field and turbulence interaction modeling layer updates the three-dimensional tensor response model using a tensor decomposition algorithm, extracting new interaction factors. It then verifies the model error using a covariance matrix decomposition algorithm, controlling it to within 1%, ensuring the model adapts to dynamic conditions such as flow rate changes and temperature fluctuations.

[0082] The phonon topology array, as explained, is a periodic acoustic structure made of composite materials such as polycarbonate and silicone. It includes an electromagnetic actuator, and the lattice spacing is adjusted to 0.05 mm to create a phonon bandgap, blocking the propagation of low-frequency sound waves in the 20-200 Hz range and suppressing noise caused by vibrations within the pipe wall. The miniature ultrasonic transducer is a ceramic transducer that applies ultrasonic perturbations with an amplitude of 0.1 mm. By applying fluid boundary layer control principles, it alters the turbulence characteristics of the fluid boundary layer, reducing vibration noise. The sound field modulation transducer is an ultrasonic transmitter that emits anti-phase sound waves, weakening resonant sound waves within the pipe through sound wave interference. The three-dimensional tensor response model is a mathematical model based on the sound field distribution matrix, fluid eigenvectors, and environmental eigenvectors. It quantifies the dynamic interaction between acoustics, fluid, and the environment through CP decomposition to generate noise reduction control signals. The orthogonal matched pursuit algorithm is a sparse signal processing algorithm that extracts the sound field distribution matrix through multi-scale decomposition and analyzes the frequency and phase of low-frequency noise in the 20-200 Hz range. The mainstream manifold analysis algorithm is a differential geometric compression algorithm that generates a low-dimensional Riemannian manifold representation, capturing turbulent boundary layer characteristics and reducing the dimensionality to 5% of the original data. The covariance matrix factorization algorithm is a signal processing algorithm that analyzes the statistical characteristics of sound field distribution data and turbulence mode data, extracting dominant frequencies, phase, and boundary layer features. The spatiotemporal total variational regularization algorithm extracts the spatial distribution characteristics of the sound field distribution data and generates a sound field modulation control signal. The multi-objective optimization algorithm, based on weighted least squares, integrates phonon topology, turbulence disturbances, and sound field modulation parameters to generate a global noise reduction strategy. The standard semantic control instructions are JSON-formatted control instructions, containing execution time, device number, and control parameters, coordinating the unified execution of the noise reduction devices. Specific Implementation Example 2:

[0084] like Figure 1-2 As shown below, the inputs, outputs, operation steps, algorithm logic, and specific applications of each algorithm in the drainage pipe noise control system based on environmental data adjustment are described in detail below. The system achieves precise suppression of low-frequency noise (20-200 Hz) by utilizing the principles of phonon bandgap, fluid dynamics boundary layer control, and acoustic interference, through phonon topology arrays, miniature ultrasonic transducers, and sound field modulation transducers.

[0085] Orthogonal matching pursuit algorithm:

[0086] The orthogonal matching pursuit algorithm is used in the acoustic field and turbulence interaction modeling layer to process acoustic feature vectors, extract the sound field distribution matrix, and capture the sound pressure distribution characteristics within the drainage pipe. Its input is an acoustic feature vector containing sound pressure data (decibels) and vibration acceleration data (meters per second squared) collected by sound pressure and vibration sensors, in a time-series vector format with 20,000 samples per second, a sampling frequency of 20 kHz, and 16-bit digital signals. The output is a sound field distribution matrix, 100×100 in dimension, with a spatial resolution of 0.05 meters and a temporal resolution of 0.01 seconds, representing the spatial distribution and temporal variation of sound pressure within the pipe, with a focus on analyzing the low-frequency noise characteristics in the 20-200 Hz range. The operation steps include: receiving the acoustic feature vector parsed by the sound field and turbulence interaction modeling layer through a JSON parser; dividing the acoustic feature vector into 0.01-second time windows, with 200 samples per window; constructing a sparse base dictionary containing 500 sine, cosine, and exponentially decaying waves in the frequency range of 20-200 Hz; initializing the sound field distribution matrix as an empty matrix with a dimension of 100×100; for each window, iteratively selecting the waveform with the highest correlation to the acoustic feature vector (correlation coefficient > 0.9) and adding it to the matrix; updating the residual by subtracting the contribution of the selected waveform, repeating the iteration until the residual energy is less than 1% of the original signal energy or after 50 iterations; merging the matrices of all windows to form the final sound field distribution matrix; and encapsulating the matrix into a JSON format data packet via narrowband IoT and transmitting it to the edge collaborative control layer. The algorithm logic decomposes high-dimensional acoustic data through sparse representation, extracts key frequencies and phases, reduces computational load, and preserves the main features of the sound field. In the system, the sound field distribution matrix is ​​used to generate phonon topology control signals, which drive the phonon topology array to adjust the lattice spacing to 0.05 mm, blocking the propagation of low-frequency sound waves, and generate sound field modulation control signals, which drive the sound field modulation transducer to emit anti-phase sound waves to weaken resonant sound waves.

[0087] Mainstream shape analysis algorithm:

[0088] The mainstream manifold analysis algorithm is used for the acoustic field and turbulence interaction modeling layer. It compresses the fluid feature vector to generate a low-dimensional Riemannian manifold representation, capturing the core features of the turbulent boundary layer. Its input is a fluid feature vector containing velocity data (meters per second) collected by an ultrasonic flow sensor, in a time-series vector format with 20,000 samples per second and 16-bit digital signals. The output is a low-dimensional Riemannian manifold representation, with a dimension of 5% of the original data (approximately 100 dimensions), representing turbulent boundary layer features such as boundary layer thickness and turbulence intensity. The operation steps include: receiving the fluid feature vector parsed by the acoustic field and turbulence interaction modeling layer through a JSON parser; dividing the fluid feature vector into 0.01-second windows with 200 samples per window; constructing a high-dimensional data matrix with dimensions 200×500; calculating the covariance matrix of the high-dimensional data matrix, extracting the eigenvectors corresponding to the first 100 eigenvalues ​​to form the initial manifold basis; performing nonlinear mapping using kernel principal component analysis with a Gaussian kernel function, and adaptively adjusting the bandwidth parameter according to the standard deviation of the velocity data (range 0.1-1); projecting the high-dimensional data matrix onto the initial manifold basis to generate a 100-dimensional low-dimensional Riemannian manifold representation; removing noise through regularization (regularization parameter 0.01) to ensure feature stability; and encapsulating the low-dimensional representation into a JSON format data packet via narrowband IoT and transmitting it to the edge collaborative control layer. The algorithm logic compresses the high-dimensional velocity data through covariance analysis and nonlinear mapping, extracts the geometric features of the turbulent boundary layer, reduces computational complexity, and maintains feature robustness. In the system, a low-dimensional Riemannian manifold representation is used to generate a turbulence disturbance control signal, which drives a miniature ultrasonic transducer to apply ultrasonic disturbances with an amplitude of 0.1 mm, optimizes the fluid boundary layer, and reduces vibration noise caused by turbulence.

[0089] Covariance matrix decomposition algorithm:

[0090] The covariance matrix factorization algorithm is used in the distributed sensing and control nodes of the edge collaborative control layer to process the 3D tensor response model, generate sound field distribution data and turbulence mode data, and support inter-node interactive optimization. Its input is a JSON-formatted 3D tensor response model transmitted from the sound field and turbulence interaction modeling layer, containing acoustic features (sound pressure and vibration), fluid features (flow velocity), and environmental features (temperature, humidity, air pressure). The output is sound field distribution data (including dominant frequencies and phases, with errors of ±5 Hz and ±0.05 radians) and turbulence mode data (including boundary layer velocity gradients, with errors of ±0.01 m / s). The operational steps include: receiving a 3D tensor response model in JSON format via narrowband IoT; unfolding the model into a matrix with dimensions 100×100×10; calculating the covariance matrix of the matrix and extracting eigenvectors corresponding to the top 10 eigenvalues ​​using singular value decomposition (with a threshold set to 1% of the matrix trace); generating sound field distribution data (dominant frequencies and phases) and turbulence mode data (boundary layer velocity gradients) based on eigenvector decomposition; receiving JSON format control signals (phonon topology, turbulence disturbances, sound field modulation) from adjacent nodes during inter-node interactions, calculating the covariance matrix of the signals, and generating adjustment weights (range 0-1, based on a linear combination of phase and amplitude differences); adjusting the control signal of this node according to the weights; and encapsulating the adjusted control signal into a JSON format data packet via narrowband IoT and transmitting it to the central dynamic decision-making layer and adjacent nodes. The algorithm logic decomposes high-dimensional data through covariance analysis, extracts key features, and achieves inter-node collaboration through weight adjustment to ensure the consistency of control signals. In the system, sound field distribution data drives the phonon topology array to adjust the lattice spacing to 0.05 mm and the sound field modulation transducer to emit antiphase sound waves, while turbulence mode data drives the micro ultrasonic transducer to apply a 0.1 mm amplitude perturbation, and the interaction between nodes optimizes signal stability.

[0091] Spatiotemporal total variational regularization algorithm:

[0092] The spatiotemporal total variational regularization algorithm is used in the edge collaborative control layer to extract the spatial distribution features of sound field distribution data and generate a sound field modulation control signal. Its input is sound field distribution data, including a sound field distribution matrix (100×100 dimensions, spatial resolution 0.05 meters, temporal resolution 0.01 seconds). The output is a sound field modulation control signal containing the phase parameters of the out-of-phase sound wave (deviation 0.1 radians). The operation steps include: receiving the sound field distribution matrix; dividing the matrix into a spatial grid (100 points, spacing 0.05 meters) and a time window (100 points, interval 0.01 seconds); calculating the total variation of the time series for each grid point and extracting the sound pressure gradient as a spatial distribution feature; generating the phase parameters of the out-of-phase sound wave based on the gradient (deviation controlled within 0.1 radians); encapsulating the phase parameters into a sound field modulation control signal and transmitting it to the central dynamic decision-making layer via narrowband IoT. The algorithm logic smooths the sound field data through total variational regularization, extracts spatial distribution features, generates accurate phase parameters, and reduces noise interference. In the system, the sound field modulation control signal drives the sound field modulation transducer to emit anti-phase sound waves, which weakens the resonant sound waves inside the pipe.

[0093] CP decomposition algorithm:

[0094] The CP decomposition algorithm is used in the acoustic field and turbulence interaction modeling layer to optimize the 3D tensor response model and extract the dominant interaction factors. Its inputs are a 3D tensor response model (containing the acoustic field distribution matrix, low-dimensional Riemannian manifold representation, and environmental feature vectors, with dimensions 100×100×10) and JSON-formatted execution feedback data from the edge collaborative control layer (including device operating status, noise level, and flow field parameters). The output is an updated 3D tensor response model (decomposition rank 10). The operation steps include: receiving the 3D tensor response model and execution feedback data; initializing the decomposition rank to 10 and constructing three factor matrices: acoustic, fluid, and environmental; iteratively optimizing the factor matrices using alternating least squares (50 iterations, convergence condition: tensor reconstruction error <1%); extracting the dominant interaction factors and updating the 3D tensor response model; verifying the model error through covariance matrix decomposition (controlled within 1%); and encapsulating the updated model into a JSON-formatted data packet via narrowband IoT and transmitting it to the edge collaborative control layer. The algorithm logic simplifies the tensor structure through low-rank decomposition, extracts the coupling relationship between acoustics and turbulence, and dynamically adapts to changes in operating conditions. In the system, the updated model ensures the accuracy of the control signals and supports real-time adjustment of phonon topology arrays, micro ultrasonic transducers, and acoustic field modulation transducers.

[0095] Multi-objective optimization algorithm:

[0096] A multi-objective optimization algorithm is used in the central dynamic decision-making layer to generate a global noise reduction control strategy. Its input consists of JSON-formatted control signals (phonon topology control signal, turbulence disturbance control signal, and acoustic field modulation control signal) transmitted from the edge collaborative control layer. The output is the global noise reduction control strategy, which includes phonon topology adjustment parameters (lattice spacing 0.05 mm), turbulence disturbance parameters (disturbance amplitude 0.1 mm), and acoustic field modulation parameters (out-of-phase acoustic wave phase, deviation 0.1 radians). The operation steps include: receiving JSON-formatted control signals; constructing an objective function that integrates the noise reduction effects of phonon topology, turbulence disturbance, and sound field modulation, with weights dynamically adjusted according to pipeline conditions (sound field modulation is prioritized for residential pipelines, with a weight of 0.5; turbulence disturbance is prioritized for industrial pipelines, with a weight of 0.5; and all three are balanced for chemical plant pipelines, with weights of 0.33 each); optimizing the objective function using weighted least squares (iteration 20 times); generating a control instruction set, including phonon topology adjustment parameters, turbulence disturbance parameters, and sound field modulation parameters; converting the instruction set into standard semantic control instructions in JSON format using a semantic conversion algorithm, including execution time, device number, and control parameters; distributing the instructions to the edge collaborative control layer via narrowband IoT; receiving execution feedback data (including changes in sound pressure level and turbulence intensity), updating the weights, with an update cycle of 5 milliseconds. The algorithm logic dynamically balances the effects of each control signal through weights to generate a globally optimal strategy. In the system, the control instruction set drives the phonon topology array to adjust the lattice spacing, the micro-ultrasonic transducer to apply disturbance, and the sound field modulation transducer to emit anti-phase sound waves, achieving collaborative noise reduction.

[0097] These algorithms achieve efficient interaction through narrowband IoT and JSON format data packets, forming a closed-loop process from acoustic-fluid-environmental data analysis to control signal generation and then to the execution of noise reduction devices. This ensures that the system adapts to the dynamic working conditions within the drainage pipe and accurately suppresses low-frequency noise in the 20-200 Hz range. Specific Implementation Example 3:

[0099] like Figure 1-2 As shown, the following is a detailed description of each piece of hardware mentioned in Embodiment 1:

[0100] The sound pressure sensor is used in a multi-dimensional data acquisition layer to measure changes in sound pressure on the inner wall of drainage pipes, capturing the intensity and distribution characteristics of noise. Its structure is a miniature condenser microphone made of silicon-based material. The sensitive element is a 2mm diameter silicon diaphragm with a sensitivity of -40 dB, a measurement range of 20 Pascals to 2000 Pascals, and a frequency response range of 20 Hz to 20 kHz. The sensor housing is made of stainless steel, 5mm in diameter and 10mm in length, corrosion-resistant, with an IP68 protection rating, suitable for humid and water-flow environments within pipes. Installation involves fixing it to the inner wall of the drainage pipe at bends and forks using M6 stainless steel bolts. The bolt tightening torque is 10 N·m. The bolt holes are aligned with pre-drilled holes in the pipe inner wall (6.5mm diameter), ensuring the sensor diaphragm is parallel to the pipe inner wall and 5cm from the pipe's central axis for accurate sound pressure signal capture. The sensor connects to an analog-to-digital converter via a waterproof cable, outputting a 16-bit digital signal with a sampling frequency of 20 kHz. In the system, the sound pressure data collected by the sound pressure sensor is used to generate acoustic feature vectors, which are then used by the sound field and turbulence interaction modeling layer to construct the sound field distribution matrix through the orthogonal matching pursuit algorithm, guiding phonon topology control and sound field modulation.

[0101] A vibration sensor is used in the multi-dimensional data acquisition layer to measure the vibration acceleration of the inner wall of a drainage pipe, capturing the noise source caused by pipe wall vibration. Its structure is a piezoelectric accelerometer sensor, using lead zirconate titanate ceramic as the sensitive element, with a sensitivity of 100 mV / g, a measurement range of 0.1g to 50g, and a frequency response range of 10 Hz to 10 kHz. The sensor housing is made of stainless steel, with a diameter of 8 mm and a length of 12 mm, and an IP68 protection rating, adaptable to changes in humidity and temperature within the pipe. Installation involves fixing it to the inner wall of the drainage pipe's bends and bifurcations using M6 stainless steel bolts, with a bolt tightening torque of 10 N·m. The bottom of the sensor is in direct contact with the inner wall of the pipe, and the installation point is 5 cm away from the sound pressure sensor to avoid signal interference. The sensor is connected to an analog-to-digital converter via a waterproof cable, outputting a 16-bit digital signal with a sampling frequency of 20 kHz. In the system, the vibration acceleration data and sound pressure data collected by the vibration sensor together constitute an acoustic feature vector, which is used by an orthogonal matching pursuit algorithm to extract the sound field distribution matrix, supporting phonon topology arrays to block vibration-induced sound waves.

[0102] An ultrasonic flow velocity sensor is used in a multi-dimensional data acquisition layer to measure the flow velocity of fluid within a pipe through the Doppler effect, capturing turbulent characteristics. Its structure consists of an ultrasonic transmitter and receiver pair. The transmitter and receiver utilize 5 mm diameter piezoelectric ceramics, operating at a frequency of 1 MHz, with a measurement range of 0.01 m / s to 5 m / s and an accuracy of 0.01 m / s. The sensor housing is made of polytetrafluoroethylene (PTFE), 10 mm in diameter and 15 mm in length, corrosion-resistant, with an IP68 protection rating, and adaptable to water flow impact within the pipe. Installation involves fixing it to a straight section of the drainage pipe using a stainless steel ring clamp bracket. The bracket consists of two semi-circular stainless steel rings, with the inner diameter matching the outer diameter of the pipe (error less than 0.5 mm), and is secured with M5 stainless steel bolts with a torque of 8 N·m. The sensor probe is aligned with the central axis of the pipe, with a probe spacing of 10 cm to ensure the ultrasonic signal propagates along the fluid direction. The sensor connects to an analog-to-digital converter via a waterproof cable, outputting a 16-bit digital signal at a sampling frequency of 20 kHz. In the system, the flow velocity data collected by the ultrasonic flow velocity sensor generates a fluid feature vector, which is used by the manifold analysis algorithm to generate a low-dimensional Riemannian manifold representation to guide turbulence disturbance control.

[0103] A temperature sensor is used in the multi-dimensional data acquisition layer to measure the external temperature of the pipeline and capture the impact of environmental parameters on noise propagation. Its structure is that of a thermistor sensor, employing an NTC thermistor, with a measurement range of -20°C to 60°C, an accuracy of 0.1°C, and a response time of less than 1 second. The sensor housing is made of stainless steel, with a diameter of 3 mm and a length of 8 mm, an IP67 protection rating, and adaptability to humidity and temperature variations. Installation is achieved by bonding the sensor to the outer wall of the pipeline with epoxy resin; the adhesive layer is 0.2 mm thick, with a curing time of 24 hours. The installation point is located on a straight pipe section, 10 cm away from bends or bifurcations to avoid heat conduction interference at pipe connections. The sensor is connected to an analog-to-digital converter via a waterproof cable, outputting a 16-bit digital signal with a sampling frequency of 20 kHz. In the system, temperature data, along with humidity and air pressure data, constitute an environmental feature vector, which is used to construct a three-dimensional tensor response model and optimize acoustic and turbulence interaction analysis.

[0104] A humidity sensor is used in the multi-dimensional data acquisition layer to measure the external humidity of the pipeline and capture the effect of ambient humidity on acoustic propagation. Its structure is a capacitive humidity sensor using a polymer film as the sensing element. It measures relative humidity from 0% to 100%, with an accuracy of 2% relative humidity and a response time of less than 5 seconds. The sensor housing is made of polycarbonate, 3 mm in diameter and 8 mm in length, with an IP67 protection rating and resistance to moisture corrosion. Installation involves bonding it to the outer wall of the pipeline with epoxy resin; the adhesive layer is 0.2 mm thick and cures in 24 hours. The installation point is adjacent to the temperature sensor, 10 mm away, to ensure environmental data consistency. The sensor connects to an analog-to-digital converter via a waterproof cable, outputting a 16-bit digital signal with a sampling frequency of 20 kHz. In the system, humidity data is used to create an environmental feature vector, optimizing the three-dimensional tensor response model and enhancing the system's adaptability to humid environments.

[0105] A barometric pressure sensor is used in the multi-dimensional data acquisition layer to measure the external air pressure of the pipeline and capture the influence of ambient air pressure on sound wave propagation. Its structure is a MEMS piezoresistive sensor, employing a silicon-based pressure-sensitive element. The measurement range is 300 hPa to 1100 hPa, with an accuracy of 0.1 hPa and a response time of less than 1 millisecond. The sensor housing is made of stainless steel, with a diameter of 3 mm and a length of 8 mm, and an IP67 protection rating to adapt to changes in the pipeline's external environment. Installation involves bonding the sensor to the pipeline's outer wall with epoxy resin. The adhesive layer is 0.2 mm thick and cures in 24 hours. The installation point is adjacent to the temperature and humidity sensors, 10 mm away, to ensure synchronous environmental data acquisition. The sensor is connected to an analog-to-digital converter via a waterproof cable, outputting a 16-bit digital signal with a sampling frequency of 20 kHz. In the system, the barometric pressure data is used to create an environmental feature vector, optimizing the three-dimensional tensor response model and improving the system's adaptability to changes in air pressure.

[0106] An electromagnetic actuator is used in the edge-coordinated control layer to drive the phonon topology array to adjust the lattice spacing, thus blocking sound wave propagation. Its structure consists of a miniature electromagnetic coil, 1 cm in diameter and 2 cm in length, made of enameled copper wire with 500 turns, operating at 12 volts and 0.5 amps, with an adjustment accuracy of 0.05 mm. The housing is made of aluminum alloy with an IP68 protection rating, resistant to moisture and vibration inside pipes. It is integrated into the phonon topology array and secured to the inner wall of the drain pipe bends and bifurcations using M6 stainless steel bolts, with a tightening torque of 10 N·m. The electromagnetic actuator connects to the distributed sensing and control node via a waterproof cable to receive phonon topology control signals. In the system, the electromagnetic actuator adjusts the phonon topology array lattice spacing to 0.05 mm according to the control signals, forming a phonon bandgap of 1.5 cm width, thus blocking the propagation of low-frequency sound waves.

[0107] The analog-to-digital converter (ADC) is used in the multi-dimensional data acquisition layer to convert analog signals acquired by sensors into digital signals. Its structure consists of a 16-bit ADC chip, model ADS1115, with a sampling frequency of 20 kHz and six input channels, connecting to sensors for sound pressure, vibration, ultrasonic flow velocity, temperature, humidity, and air pressure. The housing is made of ABS plastic, measuring 5 cm × 3 cm × 1 cm, with an IP65 protection rating, suitable for the external environment of pipelines. Installation involves screwing the chip into a control box outside the pipeline, no more than 1 meter from the sensor. The cable length is 0.5 meters, using a waterproof connector to connect the sensor and the narrowband IoT module. Within the system, the ADC converts the sensor's analog signals into 16-bit digital signals, which are then encapsulated into JSON format data packets by the narrowband IoT module and transmitted to the sound field and turbulence interaction modeling layer.

[0108] The narrowband IoT module is used for multi-dimensional data acquisition, sound field and turbulence interaction modeling, edge collaborative control, and central dynamic decision-making, enabling the transmission of data and control commands. Its structure is an NB-IoT communication module, model BC95, operating at 850 MHz, with a data transmission rate of 250 kilobits per second and power consumption of less than 200 milliwatts. The housing is made of ABS plastic, measuring 4 cm × 2 cm × 1 cm, with an IP65 protection rating, and is resistant to humidity and temperature variations. Installation involves screwing it into a control box outside a pipe, integrating it with an analog-to-digital converter and a JSON parser, using waterproof connectors. The module supports JSON format data packet transmission, including timestamps, data types, and values. Within the system, the narrowband IoT module enables cross-module transmission of sensor data, model data, control signals, and command sets, ensuring real-time interaction.

[0109] The JSON parser is used in the acoustic field and turbulence interaction modeling layer and the central dynamic decision-making layer to parse JSON format data packets and extract data content. Its structure is an embedded software module running on an ARM Cortex-M4 microprocessor with a clock speed of 120 MHz and 256 kilobytes of memory. The parser supports JSON format data packets, has a processing capacity of 1000 data packets per second, and a data packet size of 1 kilobyte. It is integrated into the microprocessor motherboard within the control box and connects to the narrowband IoT module via an internal bus, requiring no separate enclosure. The JSON parser receives data packets transmitted from the narrowband IoT module, parses timestamps, sensor numbers, and data values, and generates acoustic feature vectors, fluid feature vectors, environmental feature vectors, or control signal parameters. Within the system, the JSON parser ensures efficient data packet parsing, supporting the generation of a 3D tensor response model in the acoustic field and turbulence interaction modeling layer and the generation of a global noise reduction control strategy in the central dynamic decision-making layer.

[0110] The microprocessor in the distributed sensing and control node serves as the edge collaborative control layer, processing the 3D tensor response model, generating control signals, and enabling inter-node interaction. Its architecture is a 32-bit ARM Cortex-M4 microprocessor with a 120 MHz clock speed, 512 kilobytes of memory, and 1 megabyte of storage. It supports covariance matrix factorization and spatiotemporal total variational regularization algorithms. The housing is made of aluminum alloy, measuring 6 cm × 4 cm × 2 cm, with an IP65 protection rating, resistant to external pipeline vibration and moisture. Installation involves screwing it into a control box outside the pipeline, connecting it to a narrowband IoT module and a 1-meter waterproof cable. The microprocessor receives the JSON-formatted 3D tensor response model, runs algorithms to generate phonon topology control signals, turbulence disturbance control signals, and acoustic field modulation control signals, and exchanges these control signals with neighboring nodes via narrowband IoT, calculating and adjusting weights. Within the system, the microprocessor supports distributed collaborative control, ensuring real-time generation of control signals and consistency between nodes.

[0111] The control box houses the microprocessor for the analog-to-digital converter, narrowband IoT module, JSON parser, and distributed sensing and control nodes, providing physical protection and an integrated environment. Its structure is a stainless steel enclosure, measuring 20 cm × 15 cm × 10 cm, with a thickness of 2 mm, and an IP66 protection rating, resistant to corrosion and external impact. The enclosure contains internal mounting brackets to support screw mounting of all hardware modules, and external waterproof connectors for connecting waterproof cables. Installation involves fixing the stainless steel brackets to the outside of the drainage pipe, 0.5 meters from the pipe surface. The brackets are secured to the wall or ground with M8 bolts, tightened to a torque of 15 N·m. The control box connects the various hardware modules internally via a bus, and externally connects sensors and noise reduction devices via waterproof cables. Within the system, the control box provides centralized management and protection of the hardware, ensuring stable system operation in the pipeline environment.

[0112] Waterproof cables are used to connect sensors, noise reduction devices, analog-to-digital converters, narrowband IoT modules, and control boxes, ensuring reliable signal transmission. They are multi-core shielded cables with PVC insulation and a polyurethane outer sheath, 8 mm in diameter, 6 cores, a single core cross-sectional area of ​​0.5 mm², a withstand voltage of 300 volts, an IP68 protection rating, and resistance to moisture and water flow impact. Cable lengths are customized according to installation requirements: 0.5 meters from sensor to control box, and 1 meter from noise reduction device to control box. Waterproof connectors are used for connections, with a tightening torque of 5 N·m. In the system, the waterproof cable transmits analog, digital, and control signals from sensors, ensuring stable data and command transmission in humid environments. Specific Implementation Example 4:

[0114] like Figure 1-2As shown, the drainage pipe noise control system based on environmental data adjustment provided by this invention has versatility and scalability, and can adapt to various pipe types, including ventilation pipes, gas pipes, and industrial fluid pipes. By adjusting the hardware configuration, installation method, and algorithm parameters, it can meet the noise reduction requirements of different pipe materials, fluid types, and operating conditions. The core architecture of the system (multi-dimensional data acquisition layer, sound field and turbulence interaction modeling layer, edge collaborative control layer, central dynamic decision layer, and semantic interaction bus module) remains unchanged. The hardware and algorithm are optimized according to the pipe characteristics to ensure accurate suppression of low-frequency noise. The following describes in detail the hardware adjustments, installation method modifications, algorithm parameter optimizations, and specific applications for adapting the system to other pipes.

[0115] The hardware adjustments primarily target the materials, dimensions, and fluid characteristics of different ducts. The structure and function of the acoustic pressure sensor, vibration sensor, ultrasonic flow velocity sensor, temperature sensor, humidity sensor, barometric pressure sensor, electromagnetic actuator, analog-to-digital converter, narrowband IoT module, JSON parser, microprocessor, control box, and waterproof cable remain unchanged, but their materials and parameters are optimized based on the duct environment. For example, for ventilation ducts (including those made of PVC or galvanized steel), the housings of the acoustic pressure sensor and vibration sensor have been changed from stainless steel to corrosion-resistant aluminum alloy to reduce weight and adapt to the lightweight structure of the ventilation duct, while maintaining the IP68 protection rating. The probe of the ultrasonic flow velocity sensor has been adjusted to a 3mm diameter piezoelectric ceramic, and the operating frequency has been increased to 2MHz to accommodate the lower airflow characteristics, while maintaining an accuracy of 0.01m / s. The polycarbonate and silicone composite material of the phonon topology array remains unchanged, but the grid size has been adjusted to 1.5cm × 1.5cm to adapt to the propagation characteristics of low-frequency sound waves (50Hz to 500Hz) within the ventilation duct. The driving frequency of the miniature ultrasonic transducer and sound field modulation transducer was adjusted from 40 kHz to 30 kHz, the power was reduced to 3 watts, the amplitude remained at 0.1 mm, and the phase deviation was 0.1 radians, to match the acoustic wave propagation impedance of the air medium. For gas pipelines (including stainless steel or carbon steel), the housing of the sensor and noise reduction device was upgraded to 316L stainless steel, resistant to high pressure (up to 10 bar) and corrosive gases, with an IP68 protection rating. The measurement range of the ultrasonic flow velocity sensor was extended to 0.05 m / s to 20 m / s to adapt to the high-speed flow characteristics of gas. The electromagnetic actuator driving voltage of the phonon topology array was increased to 15 volts and the current to 0.6 amps, ensuring that the lattice spacing was adjusted to 0.05 mm in high-pressure environments. Industrial fluid pipelines require consideration of high temperatures (up to 120 degrees Celsius) and corrosive liquids. The temperature sensor has been upgraded to a PT100 platinum resistance thermometer, extending the measurement range to -50 to 150 degrees Celsius with an accuracy of 0.1 degrees Celsius. The humidity sensor has been replaced with a chemically resistant ceramic-based sensor, measuring from 0% to 100% relative humidity with an accuracy of 2%. The control box material has been upgraded to 304 stainless steel, 3 mm thick, with an IP66 protection rating, resistant to high temperatures and chemical corrosion.

[0116] Installation methods are adjusted according to the pipe material and structure. For ventilation ducts, sound pressure sensors and vibration sensors are fixed to the inner wall of the pipe with M5 aluminum alloy bolts, a tightening torque of 8 N·m, a bolt hole diameter of 5.5 mm, and the sensor distance from the pipe's central axis is 3 cm, suitable for pipes with smaller diameters (10 cm to 50 cm). Ultrasonic flow velocity sensors, miniature ultrasonic transducers, and sound field modulation transducers are fixed to straight pipe sections with aluminum alloy ring clamping brackets. The inner diameter of the bracket matches the outer diameter of the pipe, with an error of less than 0.3 mm, and a locking torque of 6 N·m. Temperature, humidity, and air pressure sensors are fixed to the outer wall of the pipe with high-temperature resistant epoxy resin, with an adhesive layer thickness of 0.2 mm and a curing time of 24 hours. The installation point is 15 cm from the pipe connection. For gas pipelines, high-pressure sealing must be considered. Sound pressure sensors and vibration sensors are fixed with M6 stainless steel bolts, a tightening torque of 12 N·m, and high-pressure sealing washers with a thickness of 1 mm are used in the bolt holes. The phonon topology array and electromagnetic actuator are fixed to the inner wall of the pipe with M6 stainless steel bolts and equipped with explosion-proof sealing rings to ensure safety. The ultrasonic flow velocity sensor and transducer are fixed with stainless steel ring clamping brackets, which adopt a double-layer locking design with a torque of 10 N·m. The control box is fixed to the outside of the pipe with an M10 stainless steel bracket, 1 meter away from the pipe, and equipped with an explosion-proof junction box. The installation of industrial fluid pipelines requires high temperature and corrosion resistance. The sound pressure sensor and vibration sensor are fixed with M6 stainless steel bolts, which are coated with anti-corrosion coating and tightened to a torque of 12 N·m. The phonon topology array is fixed to the inner wall of the pipe with a high-temperature resistant silicone gasket. The ultrasonic flow velocity sensor and transducer are fixed with a 304 stainless steel ring clamping bracket, which is lined with a corrosion-resistant rubber gasket and tightened to a torque of 10 N·m. The temperature, humidity, and air pressure sensors are bonded with high-temperature resistant epoxy resin, with an adhesive layer thickness of 0.3 mm and a curing time of 36 hours. The waterproof cable has been upgraded to a high-temperature resistant PTFE outer sheath, with a diameter of 10 mm, 8 cores, and a withstand voltage of 500 volts.

[0117] The algorithm parameters are optimized based on the pipeline operating conditions. The sparse basis dictionary of the orthogonal matching pursuit algorithm is adjusted to 300 waveforms for ventilation ducts, including low-frequency sine waves (50 Hz to 500 Hz), adapting to the acoustic characteristics of air media; for gas pipelines, the number of waveforms is increased to 600, covering 100 Hz to 1 kHz; and for industrial fluid pipelines, 500 waveforms are maintained to optimize for high-temperature liquid environments. The low-dimensional Riemannian manifold representation dimension of the mainstream shape analysis algorithm is adjusted according to fluid characteristics: 50 dimensions for ventilation ducts (3% of the original data), 120 dimensions for gas pipelines (6% of the original data), and 100 dimensions for industrial fluid pipelines (5% of the original data). The number of eigenvalues ​​in the covariance matrix decomposition algorithm is adjusted to 8 for ventilation ducts, 12 for gas pipelines, and 10 for industrial fluid pipelines, with weight calculation based on the density and viscosity differences of the fluid in the pipeline. The spatial grid resolution of the spatiotemporal total variational regularization algorithm is adjusted to 0.03 meters for ventilation ducts, 0.06 meters for gas pipelines, and 0.05 meters for industrial fluid pipelines. The decomposition rank of the CP decomposition algorithm is adjusted to 8 for ventilation ducts, 12 for gas ducts, and 10 for industrial fluid ducts, with the computation time kept to be less than 4 milliseconds and the error within 1%. The objective function weights of the multi-objective optimization algorithm are adjusted according to the noise characteristics of the ducts: sound field modulation is prioritized for ventilation ducts (weight 0.5), turbulence disturbance is prioritized for gas ducts (weight 0.5), and the balance of the three factors is maintained for industrial fluid ducts (weight 0.33 for each).

[0118] In the system, the adapted hardware and algorithms work together to collect acoustic, fluid, and environmental data from inside and outside the pipeline. These data packets are transmitted via narrowband IoT in JSON format to generate a three-dimensional tensor response model. This model drives a phonon topology array (blocking sound waves, lattice spacing of 0.05 mm), a miniature ultrasonic transducer (optimizing turbulence, amplitude of 0.1 mm), and a sound field modulation transducer (reducing resonance, phase deviation of 0.1 radians), achieving low-frequency noise suppression. Through distributed collaborative control and global dynamic decision-making, the system adapts to the dynamic operating conditions of different pipelines (such as flow rate changes and pressure fluctuations) to ensure effective noise reduction.

[0119] The following are application cases of the system in three different scenarios, namely residential ventilation ducts, industrial gas pipelines and chemical plant fluid pipelines, respectively, clarifying the pipeline characteristics, system configuration and noise reduction effect.

[0120] Case 1: Residential ventilation ducts

[0121] Duct characteristics: The ventilation duct for residential buildings is made of PVC, with a diameter of 20 cm and a length of 10 m. The air velocity is 0.5 m / s to 2 m / s. The noise is mainly low-frequency noise (100 Hz to 500 Hz) caused by fan vibration and airflow turbulence, with a sound pressure level of about 60 decibels.

[0122] System Configuration: The sound pressure sensor and vibration sensor are housed in aluminum alloy casings and fixed to the inner wall of the pipe with M5 bolts, tightening torque 8 N·m, 3 cm from the central axis. The ultrasonic flow velocity sensor (frequency 2 MHz) is fixed to the straight pipe section with an aluminum alloy ring clamping bracket, torque 6 N·m. The phonon topology array (grid size 1.5 cm × 1.5 cm) is fixed with M5 bolts, and the miniature ultrasonic transducer and sound field modulation transducer (frequency 30 kHz, power 3 W) are fixed with clamping brackets. The temperature, humidity, and air pressure sensors are bonded to the outer wall of the pipe with epoxy resin. The control box (aluminum alloy, IP66) is fixed 0.5 meters outside the pipe. The algorithm parameters are adjusted as follows: orthogonal matching pursuit algorithm with 300 waveforms, mainstream shape analysis algorithm with 50 manifold dimensions, CP decomposition with rank 8, and multi-objective optimization algorithm prioritizing sound field modulation (weight 0.5).

[0123] Noise reduction effect: The system blocks the sound waves of fan vibration through phonon topology array, optimizes the airflow boundary layer with micro ultrasonic transducer, and weakens resonance with sound field modulation transducer, reducing noise to 40 decibels with a noise reduction rate of 33%, which meets the requirements of quiet residential environment.

[0124] Case 2: Industrial Gas Pipelines

[0125] Pipeline characteristics: Industrial gas pipeline, made of stainless steel, 50 cm in diameter, 50 m in length, gas flow velocity 5 m / s to 15 m / s, pressure 2 bar to 8 bar, noise is mainly low-frequency noise (200 Hz to 1 kHz) caused by high-speed airflow turbulence and pipeline vibration, sound pressure level is about 80 dB.

[0126] System configuration: The sound pressure sensor and vibration sensor are housed in 316L stainless steel, secured with M6 bolts to a tightening torque of 12 N·m, and equipped with high-pressure sealing gaskets. The ultrasonic flow velocity sensor (measuring range 0.05 m / s to 20 m / s) is secured with a double-layer stainless steel clamping bracket to a torque of 10 N·m. The phonon topology array is secured with M6 bolts, and the electromagnetic actuator operates at a driving voltage of 15 volts. The miniature ultrasonic transducer and sound field modulation transducer are secured with clamping brackets, operating at a frequency of 30 kHz and a power of 3 watts. The control box (304 stainless steel, IP66) is equipped with an explosion-proof junction box and is fixed 1 meter outside the pipe. The algorithm parameters are adjusted as follows: orthogonal matching pursuit algorithm with 600 waveforms, mainstream shape analysis algorithm with a manifold dimension of 120, CP decomposition rank of 12, and multi-objective optimization algorithm prioritizing turbulent disturbances (weight 0.5).

[0127] Noise reduction effect: The system reduces high-speed airflow noise through turbulence disturbance control, blocks vibration sound waves through phonon topology array, and weakens resonance through sound field modulation, reducing noise to 50 decibels with a noise reduction rate of 37.5%, meeting industrial safety standards.

[0128] Case 3: Fluid Piping in a Chemical Plant

[0129] Pipeline characteristics: Chemical plant fluid pipeline, made of polypropylene, 30 cm in diameter, 20 m in length, the fluid is a corrosive chemical liquid, the flow rate is 1 m / s to 5 m / s, the temperature is 50 degrees Celsius to 100 degrees Celsius, the noise is mainly low-frequency noise (100 Hz to 800 Hz) caused by liquid turbulence and pump vibration, with a sound pressure level of about 70 decibels.

[0130] System Configuration: The sound pressure sensor and vibration sensor are housed in 316L stainless steel and secured with M6 bolts coated with an anti-corrosion layer, with a torque of 12 N·m. The ultrasonic flow velocity sensor is secured with a 304 stainless steel clamping bracket lined with a corrosion-resistant rubber gasket, with a torque of 10 N·m. The phonon topology array is secured with M6 bolts and equipped with a high-temperature resistant silicone gasket. The miniature ultrasonic transducer and sound field modulation transducer are secured with clamping brackets, operating at a frequency of 30 kHz and a power of 3 W. The temperature sensor (PT100 platinum resistance, range -50°C to 150°C) and humidity sensor (ceramic substrate) are bonded with high-temperature resistant epoxy resin. The control box (304 stainless steel, IP66) is fixed 0.5 meters outside the pipeline. The algorithm parameters are adjusted as follows: orthogonal matching pursuit algorithm with 500 waveforms, mainstream shape analysis algorithm with 100 manifold dimensions, CP decomposition with rank 10, and multi-objective optimization algorithm balancing the three (weights 0.33 each).

[0131] Noise reduction effect: The system optimizes the chemical liquid boundary layer through turbulence disturbance, blocks the pump vibration sound waves through phonon topology array, and weakens resonance through sound field modulation, reducing noise to 45 decibels with a noise reduction rate of 35.7%, and is suitable for high-temperature and corrosive environments. Specific Implementation Example 5:

[0133] like Figure 1-2 As shown, the following are experimental data of a drainage pipe noise control system based on environmental data adjustment in three different pipeline scenarios:

[0134]

[0135] The above experimental data are from results collected in laboratory simulation environments and actual field tests from April to May 2025.

[0136] The following conclusions can be drawn from the above data:

[0137] The system achieves significant noise reduction in various pipeline scenarios. In residential ventilation ducts, the noise level is reduced from 60 dB to 40 dB, a reduction rate of 33.33%. In industrial gas pipelines, the noise level is reduced from 80 dB to 50 dB, a reduction rate of 37.50%. In chemical plant fluid pipelines, the noise level is reduced from 70 dB to 45 dB, a reduction rate of 35.71%. This meets the requirements for quiet residential environments, industrial safety, and chemical operations.

[0138] The system is adaptable to various fluids and operating conditions, with flow rates ranging from 1.2 m / s to 10 m / s, temperatures from 25°C to 80°C, humidity from 50% to 70%, and air pressures from 1000 hPa to 1010 hPa. It exhibits stable noise reduction performance, verifying the versatility of the hardware and algorithms.

[0139] With high hardware parameter consistency, the phonon topology array lattice spacing of 0.05 mm, the amplitude of the micro ultrasonic transducer of 0.1 mm, and the phase deviation of the sound field modulation transducer of 0.1 radians remain unchanged in all scenarios, and no major adjustments are required to adapt to different pipelines.

[0140] The noise reduction rate of fluid pipelines in chemical plants is slightly lower than that of industrial gas pipelines. High temperature of 80 degrees Celsius and high humidity of 70% increase the complexity of liquid turbulence, suggesting that the system can optimize the turbulence disturbance control weights in extreme environments.

[0141] The following is a comparative analysis of the drainage pipe noise control system based on environmental data adjustment and existing technologies, aiming to clarify the technical innovations and advantages of this invention. The analysis is based on existing literature, patents, and the current state of industry technology, combined with specific embodiments of this technical solution. The comparison focuses on system architecture, hardware configuration, algorithm design, noise reduction effect, and applicable scenarios, ensuring sufficient evidence and clear logic. This system, through multi-dimensional data acquisition, sound field and turbulence interaction modeling, distributed collaborative control, and global dynamic decision-making, utilizes phonon topology arrays, miniature ultrasonic transducers, and sound field modulation transducers to accurately suppress low-frequency noise (20-200 Hz) in drainage pipes, adapting to various pipe types (residential ventilation ducts, industrial gas pipelines, and chemical plant fluid pipelines).

[0142] System architecture comparison:

[0143] This invention employs a four-layer architecture (multidimensional data acquisition layer, sound field and turbulence interaction modeling layer, edge collaborative control layer, and central dynamic decision-making layer). It utilizes narrowband IoT with JSON-formatted data packets to achieve efficient data interaction, forming a closed-loop control process. The multidimensional data acquisition layer collects real-time data on sound pressure, vibration, flow velocity, temperature, humidity, and air pressure. The sound field and turbulence interaction modeling layer constructs a three-dimensional acoustic-fluid-environment tensor response model. The edge collaborative control layer generates phonon topology control signals, turbulence disturbance control signals, and sound field modulation control signals. The central dynamic decision-making layer generates a global noise reduction strategy through a multi-objective optimization algorithm and issues standard semantic control commands. Feedback data updates the model and strategy at 5-millisecond intervals. This distributed and centralized architecture ensures real-time performance and adaptability, making it suitable for dynamic operating conditions (such as flow rate changes and temperature fluctuations).

[0144] Existing technology:

[0145] Traditional silencer technology: This type of silencer uses passive silencers, which reduce pipe noise through sound-absorbing materials (such as fiberglass) or resonant cavities (such as Helmholtz resonators). It has a simple structure, requires no data acquisition or processing, and relies on a fixed geometric design. However, this technology has limited effectiveness in suppressing low-frequency noise (20-200 Hz), with noise reduction rates typically below 20%, and it cannot adapt to dynamic operating conditions.

[0146] Active noise cancellation technology: This technology collects sound pressure signals through microphones and generates anti-phase sound waves for interference. A typical device is a loudspeaker. The control unit is based on a simple digital signal processor (DSP) and lacks a distributed architecture. The system only targets single sound pressure data and lacks analysis of fluid and environmental parameters. The noise reduction rate is approximately 25-30%, and it is less effective against vibration noise caused by turbulence.

[0147] Intelligent noise reduction systems employ sensors to collect sound pressure and vibration data, generate control signals through a central controller to drive loudspeakers or vibration suppression devices, and some systems use narrowband IoT to transmit data. However, the system architecture is typically single-layered or dual-layered (acquisition layer and control layer), lacking sound field and turbulence interaction modeling and distributed collaborative control, achieving a noise reduction rate of approximately 30%, and exhibiting insufficient adaptability to complex operating conditions.

[0148] Innovation points and advantages:

[0149] This invention integrates multi-dimensional data (acoustics, fluid, and environment) through a four-layer architecture to construct a three-dimensional tensor response model of acoustics-fluid-environment. Compared with traditional silencers and active noise reduction technology that only processes sound pressure, this invention comprehensively captures the noise generation and propagation mechanism and improves the accuracy of low-frequency noise suppression.

[0150] The combination of the distributed collaborative control layer and the central dynamic decision-making layer, through inter-node interaction and global optimization, significantly improves the adaptability to dynamic operating conditions (such as flow velocity of 1.2-10 meters per second and temperature of 25-80 degrees Celsius) compared to the single-layer control of existing intelligent noise reduction systems.

[0151] The closed-loop interaction mechanism between narrowband IoT and JSON format data packets offers lower power consumption (<200 milliwatts) and more stable transmission compared to existing data transmission methods (such as Bluetooth or Wi-Fi), making it suitable for long-term operation in pipeline environments.

[0152] Based on the experimental data of Embodiment 5 of the present invention (noise reduction rate of 33.33% for residential ventilation ducts, 37.50% for industrial gas pipelines, and 35.71% for chemical plant fluid pipelines), it is significantly better than traditional silencers (<20%) and existing active noise reduction systems (25-30%).

[0153] It should be noted that, in this document, relational terms such as "first" and "second" are used merely to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising a reference structure" does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes the element.

[0154] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.

Claims

1. A drainage pipe noise control system based on environmental data adjustment, comprising a multi-dimensional data acquisition layer, a sound field and turbulence interaction modeling layer, an edge collaborative control layer, and a central dynamic decision-making layer, characterized in that: The specific configuration of the control system is as follows: The multidimensional data acquisition layer deploys sensor arrays at bends and bifurcations of the drainage pipe to collect multidimensional data of the drainage pipe, encapsulates the data through narrowband Internet of Things, and then transmits it to the sound field and turbulence interaction modeling layer. The acoustic field and turbulence interaction modeling layer receives data transmitted from the multidimensional data acquisition layer, processes the data through orthogonal matching tracking algorithm and mainstream shape analysis algorithm, constructs a three-dimensional tensor response model of acoustic-fluid-environment, encapsulates the model data through narrowband Internet of Things, and then transmits it to the edge collaborative control layer. The edge collaborative control layer deploys distributed sensing and control nodes at bends and bifurcations of the drainage pipes, receives model data transmitted by the sound field and turbulence interaction modeling layer, generates phonon topology control signals, turbulence disturbance control signals, and sound field modulation control signals, and transmits the control signals to the central dynamic decision-making layer through narrowband Internet of Things. The central dynamic decision-making layer receives control signals transmitted from the edge collaborative control layer, generates a global noise reduction control strategy through a multi-objective optimization algorithm, converts it into standard semantic control commands, and sends them to the distributed sensing and control nodes of the edge collaborative control layer through narrowband IoT to coordinate phonon topology control, turbulence disturbance control, and sound field modulation control.

2. The drainage pipe noise control system based on environmental data adjustment according to claim 1, characterized in that: The overall control flow of the control system is as follows: the multi-dimensional data acquisition layer acquires data and transmits it to the sound field and turbulence interaction modeling layer; the sound field and turbulence interaction modeling layer generates a three-dimensional tensor response model and transmits it to the edge collaborative control layer; the edge collaborative control layer generates control signals and transmits them to the central dynamic decision-making layer; the central dynamic decision-making layer generates a global control strategy and issues control commands to the edge collaborative control layer; the edge collaborative control layer then drives the noise reduction device; and the edge collaborative control layer feeds back execution data to the central dynamic decision-making layer in real time to update the control strategy.

3. The drainage pipe noise control system based on environmental data adjustment according to claim 1, characterized in that: The acoustic field and turbulence interaction modeling layer constructs a three-dimensional tensor response model of acoustic-fluid-environment through the following steps: A1. The parser receives data transmitted from the multidimensional data acquisition layer and parses it into acoustic feature vectors, fluid feature vectors, and environmental feature vectors. A2. Decompose the acoustic feature vectors using the orthogonal matching pursuit algorithm to generate the sound field distribution matrix; A3. By compressing the fluid feature vector through mainstream shape analysis algorithms, a low-dimensional Riemannian manifold representation is generated, reducing the dimensionality to 5% of the original data. A4. Based on the sound field distribution matrix, low-dimensional Riemannian manifold representation, and environmental feature vectors, a three-dimensional tensor response model of acoustic-fluid-environment is constructed.

4. The drainage pipe noise control system based on environmental data adjustment according to claim 1, characterized in that: The distributed sensing and control nodes of the edge collaborative control layer operate as follows: B1. A three-dimensional tensor response model transmitted by the acoustic field and turbulence interaction modeling layer via narrowband Internet of Things; B2. Process the three-dimensional tensor response model using the covariance matrix decomposition algorithm to generate sound field distribution data and turbulence mode data; B3. A phonon topology array, a micro-ultrasonic transducer, and a sound field modulation transducer are configured. All three are noise reduction devices. The phonon topology array hardware includes an electromagnetic actuator. The distributed sensing and control node generates a phonon topology control signal based on sound field distribution data to drive the electromagnetic actuator of the phonon topology array. The phonon topology array is a periodic acoustic structure made of composite materials. A turbulence disturbance control signal is generated based on turbulence mode data to drive the micro-ultrasonic transducer, which is a ceramic transducer. A sound field modulation control signal is generated based on sound field distribution data to drive the sound field modulation transducer. B4. Control signals are encapsulated and transmitted to the central dynamic decision-making layer via narrowband Internet of Things.

5. The drainage pipe noise control system based on environmental data adjustment according to claim 4, characterized in that: The edge collaborative control layer generates control signals through the following steps: C1. A three-dimensional tensor response model transmitted by the acoustic field and turbulence interaction modeling layer via narrowband Internet of Things; C2. The three-dimensional tensor response model is decomposed by the covariance matrix decomposition algorithm to generate the dominant frequency and phase in the sound field distribution data, and the phonon topology control signal is generated to drive the electromagnetic actuator to adjust the lattice spacing of the phonon topology array to 0.05 mm. C3. Extract boundary layer features from turbulence mode data using mainstream shape analysis algorithm, generate turbulence disturbance control signal to drive the micro ultrasonic transducer to apply disturbance to an amplitude of 0.1 mm; C4. Extract the spatial distribution characteristics of the sound field distribution data through the spatiotemporal total variational regularization algorithm, and generate a sound field modulation control signal to drive the sound field modulation transducer to emit anti-phase sound waves. C5. Control signals are encapsulated and transmitted to the central dynamic decision-making layer via narrowband Internet of Things.

6. The drainage pipe noise control system based on environmental data adjustment according to claim 1, characterized in that, The central dynamic decision-making layer generates a global noise reduction control strategy through the following steps: D1. Receive phonon topology control signals, turbulence disturbance control signals and sound field modulation control signals transmitted by the edge collaborative control layer through narrowband Internet of Things; D2. Calculate the global noise reduction control strategy based on the control signal using a multi-objective optimization algorithm, and generate a control instruction set that includes phonon topology adjustment parameters, turbulence disturbance parameters, and sound field modulation parameters. D3. Encapsulate the control instruction set through narrowband IoT and transmit it to the edge collaborative control layer; D4. Receive execution feedback data transmitted from the edge collaborative control layer, update the control instruction set, and update the control instruction set with an update period of 5 milliseconds.

7. The drainage pipe noise control system based on environmental data adjustment according to claim 6, characterized in that: Before encapsulating the control instruction set, the central dynamic decision-making layer converts the control instruction set into standard semantic control instructions using a semantic conversion algorithm.

8. The drainage pipe noise control system based on environmental data adjustment according to claim 1, characterized in that: The distributed sensing and control nodes of the edge collaborative control layer achieve inter-node interaction through the following steps: E1. Each distributed sensing and control node receives phonon topology control signals, turbulence disturbance control signals, and sound field modulation control signals from neighboring nodes via narrowband Internet of Things. E2. Analyze the control signals of adjacent nodes using the covariance matrix decomposition algorithm, and adjust the phonon topology control signal, turbulence disturbance control signal, and sound field modulation control signal of this node. E3. The adjusted control signals are encapsulated and transmitted to adjacent nodes and the central dynamic decision-making layer via narrowband IoT; E4. Receive the global noise reduction control strategy transmitted by the central dynamic decision-making layer and update the interaction parameters between nodes.

9. The drainage pipe noise control system based on environmental data adjustment according to claim 1, characterized in that: The acoustic field and turbulence interaction modeling layer optimizes the model through the following steps: F1. Update the three-dimensional tensor response model of acoustic-fluid-environment using tensor decomposition algorithm to extract new dominant interaction factors; F2. Verify the updated three-dimensional tensor response model using the covariance matrix factorization algorithm; F3. The updated 3D tensor response model is encapsulated via narrowband IoT and transmitted to the edge collaborative control layer.

Images