Intelligent emission control and diagnosis system for waste incineration power plant

Abstract

The invention relates to the technical field of waste incineration power generation, and discloses an intelligent emission control and diagnosis system for a waste incineration power plant, which comprises a sound wave transceiving array, an electrical capacitance tomography sensor array, a laser grid monitoring unit, a matrix type variable track spray gun array and a central control unit. The central control unit adopts time division multiplexing logic, and reconstructs a three-dimensional temperature field, a dielectric constant field and a gas concentration field in a hearth by using an acoustic-optical cross-modal coupling correction algorithm in a measurement time slot; and in the execution time slot, based on multi-field feature fusion, diagnosing and identifying a nitrogen oxide generation potential zone and a mixed dead zone, driving the matrix type variable track spray gun to perform targeted reducing agent injection on the potential zone, controlling the sound wave transceiving array to be switched to an excitation mode, and inducing sound-induced turbulence in the mixed dead zone to enhance mixing. According to the invention, accurate monitoring and zoned targeted treatment in a non-uniform combustion environment are realized, and consumption of a reducing agent and ammonia escape are effectively reduced.

Description

Technical Field

[0001] This invention relates to the field of waste incineration power generation and environmental control technology, specifically to an intelligent emission control and diagnostic system for waste incineration power plants. Background Technology

[0002] Waste-to-energy incineration, as a primary means of municipal solid waste treatment, involves a highly complex and unsteady-state combustion process. Due to the significant fluctuations in the composition, moisture content, and calorific value of the waste fed into the furnace, the temperature field, flow field, and chemical component concentration field within the furnace exhibit highly non-uniform and rapidly changing three-dimensional distribution characteristics. To meet increasingly stringent nitrogen oxide emission standards, selective non-catalytic reduction (SNR) technology is widely used in the flue gas denitrification process of waste-to-energy incineration plants. The core of this technology lies in injecting a reducing agent (such as ammonia or urea solution) into a specific temperature window region of the furnace, causing it to undergo a reduction reaction with nitrogen oxides in the flue gas.

[0003] However, existing control systems still have significant technical limitations in terms of monitoring accuracy and injection execution. In terms of monitoring, traditional furnace temperature monitoring mainly relies on wall thermocouples, which can only provide the temperature at a single point near the heated surface and cannot accurately reflect the temperature distribution of the furnace cross-section and core area. Although tunable semiconductor laser absorption spectroscopy has been introduced for gas concentration monitoring, its measurement principle is based on the integral absorption characteristics along the line-of-sight path, and the spectral intensity of gas molecules is strongly dependent on temperature. In the environment of a waste incinerator with a large temperature gradient, existing technologies often use path-averaged temperature or set temperature for inversion, ignoring the influence of temperature non-uniformity along the optical path on the spectral absorption rate, resulting in significant deviations between the measured concentrations of ammonia and nitrogen oxides and the true values.

[0004] In terms of control execution, due to the lack of high-resolution three-dimensional physical field data inside the furnace, existing reductant injection strategies typically rely on feedback adjustments based on emission indicators at the outlet section or coarse temperature zones. This control mode is inherently a lagging and uniform adjustment method, unable to accurately identify and track randomly generated local high-temperature core zones or high-concentration nitrogen oxide generation zones within the furnace. Furthermore, the arrangement and trajectory of the spray guns are usually fixed, making it difficult to adapt to combustion center shifts. This leads to excessive reductant injection in some areas causing ammonia escape, while insufficient coverage in other areas results in low denitrification efficiency. Simultaneously, in certain low-velocity or laminar flow regions within the furnace, the mixing of reductant and flue gas is limited by mass transfer resistance, forming reaction dead zones. Existing technologies lack effective non-invasive interference mechanisms to break this localized poor mixing. These problems collectively result in high reductant consumption, uncontrollable ammonia escape, and drastic fluctuations in emission indicators in existing systems. Therefore, this invention provides an intelligent emission control and diagnostic system for waste-to-energy incineration plants to address the shortcomings of existing technologies. Summary of the Invention

[0005] To address the shortcomings of existing technologies, this invention provides an intelligent emission control and diagnostic system for waste incineration power plants, solving the problems mentioned in the background section.

[0006] The first aspect of this invention provides an intelligent emission control and diagnostic system for waste incineration power plants.

[0007] The system includes an acoustic transceiver array, a capacitance tomography sensor array, a laser grid monitoring unit, a matrix-type variable trajectory spray gun array, and a central control unit. The central control unit is communicatively connected to each of the above arrays and units.

[0008] At the control logic level, the central control unit is configured with time-division multiplexing logic, dividing the system's working cycle into non-overlapping measurement time slots and execution time slots. This configuration allows the system to utilize the same acoustic transceiver hardware to perform low-power acoustic temperature measurement and high-power acoustic excitation in different time slots, avoiding interference between the excitation signal and the measurement signal.

[0009] At the data processing level, the central control unit has a built-in multi-physics inversion and fusion algorithm.

[0010] During the measurement time slot, the system simultaneously acquires acoustic, capacitance, and spectral measurement data. To address the spectral absorbance deviation caused by the non-uniform temperature field inside the furnace, the system employs an acoustic-optical cross-modal coupling correction strategy: using temperature field data reconstructed by acoustic tomography, temperature indexing is performed on discrete micro-elements along the laser beam path to correct spectral line intensity, and combined with dielectric constant distribution reconstructed by capacitance tomography, three-dimensional physical field data covering temperature, solid particle concentration, and gas component concentration are generated.

[0011] During the execution time slot, the system performs reaction dynamics region diagnosis and intervention based on the three-dimensional physical field data.

[0012] First, the system identifies potential regions for nitrogen oxide (NOx) generation. It calculates the spatial gradient modulus of the dielectric constant field to characterize the turbulence of solid particles, identifying regions that meet the conditions of high temperature and high dielectric constant gradient as NOx generation sources. For these regions, the system calculates the attitude angle of a matrix-type variable trajectory spray gun to achieve targeted spraying of the reducing agent.

[0013] Secondly, the system identifies mixing dead zones. Areas where ammonia escape concentration exceeds the standard and nitrogen oxide concentration does not meet the standard are identified as mass transfer-limited regions. For these regions, the system controls the acoustic transceiver array to switch to excitation mode, using a beamforming algorithm to focus acoustic energy in the mixing dead zone, inducing an acoustic flow effect in the fluid, thereby enhancing local mixing without relying on mechanical stirring.

[0014] A second aspect of the present invention provides an intelligent emission control and diagnosis method based on the above-described system.

[0015] This method includes data acquisition and reconstruction, multi-field feature fusion diagnosis, and partitioned collaborative control steps.

[0016] In the data acquisition and reconstruction steps, the method establishes a unified three-dimensional Cartesian coordinate system to map measurement data with different spatial resolutions to a unified voxel grid. The method involves mode switching of the drive circuit: during the measurement phase, a low-power pseudo-random encoded sequence is used to obtain the flight time; during the execution phase, the signal is switched to a high-power single-frequency or narrowband modulated signal to generate acoustic radiation force.

[0017] In the multi-field feature fusion diagnostic step, the method uses multi-parameter Boolean logic operations to jointly analyze the temperature field, the dielectric constant distribution reflecting the solid phase flow field state, and the gas concentration field reflecting the chemical component distribution, thereby locating the key reaction kinetic regions within the furnace.

[0018] In the partitioned collaborative control step, the method performs precise intervention based on spatial coordinates: for the potential region of nitrogen oxide generation, variable trajectory injection technology is used for source suppression; for the mixing dead zone, acoustic turbulence technology is used for mixing enhancement.

[0019] This invention provides an intelligent emission control and diagnostic system for waste-to-energy incineration plants. It offers the following advantages: 1. This invention solves the problem of large measurement deviation in complex temperature fields of traditional tunable semiconductor laser absorption spectroscopy by using an acoustic-optical cross-modal coupling correction mechanism. The system uses two-dimensional temperature field data reconstructed by acoustic tomography to index the temperature of discrete micro-elements through which the laser beam passes, and corrects the temperature dependence in the laser spectral intensity model accordingly. This effectively eliminates the interference of the drastic fluctuations in the non-uniform temperature distribution in the furnace on the gas concentration inversion calculation, and significantly improves the accuracy of monitoring the concentration field distribution of ammonia and nitrogen oxides in the furnace.

[0020] 2. This invention changes the traditional denitrification system's control mode, which can only perform uniform injection based on the average concentration at the outlet section. It realizes the diagnosis and intervention of reaction dynamics regions inside the furnace. Through the fusion of multi-physics field characteristics, the system can locate the potential zone for nitrogen oxide generation and the mixing dead zone in real time, and drive the matrix-type variable trajectory spray gun to perform fixed-point tracking injection. At the same time, it guides the acoustic array to generate acoustic turbulence at the dead zone position. This point-to-point intervention method based on spatial coordinates not only suppresses the generation of pollutants from the source, but also enhances the local mass transfer efficiency, reduces the amount of reducing agent used, and reduces the ammonia escape level.

[0021] 3. This invention achieves dual-function multiplexing of the acoustic transceiver array through a time-division multiplexing architecture and drive circuit switching design. The system strictly divides the working cycle into measurement time slots and execution time slots, enabling the same hardware to operate in low-power acoustic temperature measurement mode and high-power acoustic flow excitation mode respectively. This not only simplifies the sensor layout on the furnace wall and the system hardware topology, but also isolates the interference of high-intensity execution acoustic waves on weak measurement signals in the time domain, ensuring that the system maintains a high signal-to-noise ratio online monitoring capability while actively intervening. Attached Figure Description

[0022] Figure 1 This is a system architecture diagram of the present invention; Figure 2 This is a flowchart of the method of the present invention; Figure 3 This is a flowchart illustrating the multiphysics inverse reconstruction process of the present invention. Figure 4 This is a flowchart illustrating the multi-field feature fusion diagnosis process of the present invention. Detailed Implementation

[0023] The technical solutions in 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.

[0024] Please see the appendix Figure 1 -Appendix Figure 4 This invention provides an intelligent emission control and diagnostic system for waste incineration power plants, including an acoustic transceiver array, a capacitance tomography sensor array, a laser grid monitoring unit, a matrix-type variable trajectory spray gun, and a central control unit.

[0025] The system is installed on the furnace body of the waste incinerator, which is vertically divided into a lower combustion zone and an upper burnout zone. The acoustic transceiver array is arranged circumferentially along the sidewalls of the combustion zone and layered vertically. The acoustic transceiver array comprises several acoustic transceiver units, each containing an acoustic transducer, a sound driver, and a waveguide structure. Each acoustic transceiver unit has two operating modes: a measurement mode for receiving acoustic signals or transmitting probe acoustic waves, and an excitation mode for transmitting high-intensity modulated acoustic waves.

[0026] A capacitance tomography (CTT) sensor array is positioned on the outer side of the furnace wall in the main reaction section corresponding to the combustion zone. The array comprises several electrode units, which are evenly spaced along the circumference of the furnace and insulated. The CTT sensor array is used to measure the capacitance between the electrode units to reflect the distribution of the dielectric constant of the medium within the furnace cross-section.

[0027] The laser grid monitoring unit is located at the interface between the combustion zone and the burnout zone. The unit comprises opposing groups of laser emitters and receivers, forming a cross-optical grid across the furnace cross-section. The laser grid monitoring unit is used to measure the gas absorption spectral signal within this cross-section.

[0028] The matrix-type variable trajectory spray gun is installed on the furnace wall above the combustion zone. It is connected to the furnace wall via a dual-axis adjustment mechanism, which adjusts the horizontal deflection and vertical pitch angles of the spray gun. A flow control valve is installed on the pipeline of the matrix-type variable trajectory spray gun to regulate the injection flow rate of the reducing agent.

[0029] The central control unit establishes communication connections with the acoustic transceiver array, the capacitance tomography sensor array, the laser grid monitoring unit, and the matrix-type variable trajectory spray gun. The central control unit is used to perform timing control, physical field data calculation, and control command generation.

[0030] The central control unit divides a complete control cycle into measurement time slots and execution time slots, and performs control according to the following steps: During the measurement time slot, the central control unit controls the acoustic transceiver array to operate in measurement mode, controlling the acoustic transceiver unit to emit probe acoustic waves according to a preset sequence and acquiring the transit time data along each acoustic wave path. Simultaneously, the central control unit acquires capacitance measurement data output from the capacitance tomography sensor array and spectral absorption intensity data output from the laser grid monitoring unit.

[0031] Based on the collected flight time data and the thermodynamic relationship between sound velocity and temperature, the central control unit reconstructs the two-dimensional temperature field distribution data of the furnace cross-section. Based on the collected capacitance measurement data and an algorithm for the inverse electrostatic problem, the central control unit reconstructs the dielectric constant distribution data of the furnace cross-section. The dielectric constant distribution data characterizes the concentration distribution of solid particles.

[0032] The central control unit uses the reconstructed two-dimensional temperature field distribution data to correct the spectral line intensity parameters in the spectral absorption intensity data point by point, and then reconstructs the gas concentration field distribution data of the furnace cross-section based on the corrected parameters. The gas concentration field distribution data includes the concentration distributions of nitrogen oxides, ammonia, and carbon monoxide.

[0033] The central control unit maps the two-dimensional temperature field distribution data, dielectric constant distribution data, and gas concentration field distribution data to a unified three-dimensional spatial coordinate system. The central control unit determines whether there are regions that meet both a preset high-temperature threshold and a preset dielectric constant gradient threshold; if so, this region is marked as a nitrogen oxide generation potential region. The central control unit also determines whether there are regions where the ammonia concentration is higher than a preset escape threshold and the nitrogen oxide concentration is higher than a preset emission threshold; if so, this region is marked as a mixing dead zone.

[0034] During the execution time slot, the central control unit calculates the target deflection angle and target flow rate of the matrix variable trajectory spray gun based on the spatial coordinates of the nitrogen oxide generation potential zone, and sends the first control command to the dual-axis adjustment mechanism and flow control valve to make the reducing agent cover the nitrogen oxide generation potential zone.

[0035] When a mixing dead zone exists, the central control unit switches the acoustic transceiver array to excitation mode within the execution time slot. The central control unit calculates beamforming phase parameters based on the spatial location of the mixing dead zone and sends a second control command to the acoustic transceiver unit. The acoustic transceiver unit then transmits modulated acoustic waves according to the second control command, creating an acoustic energy gradient at the mixing dead zone location to generate an acoustic flow.

[0036] This invention provides an intelligent emission control method for waste-to-energy power plants based on the synergy of multimodal tomography and acoustic turbulence, comprising the following steps: S10, initialize system parameters, read furnace geometry parameters, sensor coordinate positions, reference sound velocity and reference capacitance value; S20, enter the measurement time slot, control the acoustic transceiver array to work in measurement mode to obtain the flight time data of each acoustic path, and simultaneously obtain the capacitance measurement data output by the capacitance tomography sensor array and the spectral absorption intensity data output by the laser grid monitoring unit. S30, reconstruct the two-dimensional temperature field distribution data of the furnace cross section based on the flight time data, reconstruct the dielectric constant distribution data characterizing the concentration distribution of solid particles based on the capacitance measurement data, and use the two-dimensional temperature field distribution data to correct the spectral absorption intensity data to reconstruct the gas concentration field distribution data. S40 maps two-dimensional temperature field distribution data, dielectric constant distribution data and gas concentration field distribution data to a three-dimensional spatial coordinate system. It identifies the potential region for nitrogen oxide generation based on preset high temperature threshold and dielectric constant gradient threshold, and identifies the mixing dead zone based on preset escape threshold and emission threshold. S50, enter the execution time slot, calculate the target deflection angle and target flow rate of the matrix variable trajectory spray gun according to the spatial coordinates of the nitrogen oxide generation potential zone, and control the matrix variable trajectory spray gun to cover the nitrogen oxide generation potential zone with reducing agent. S60, if a mixed dead zone exists within the execution time slot, the acoustic transceiver array is switched to excitation mode. The beamforming phase parameters are calculated based on the spatial location of the mixed dead zone, and modulated acoustic waves are emitted to form an acoustic energy gradient in the mixed dead zone to generate acoustic flow.

[0037] In this embodiment, steps S10 and S20 specifically involve system initialization and the multimodal data synchronous acquisition process based on a time-division multiplexing mechanism. In step S10, after the system is powered on, the central control unit first executes the initialization program. The central control unit reads the geometric parameters of the furnace from a preset storage module through the bus interface. These geometric parameters are used to construct the finite element mesh model for subsequent physical field inversion, specifically including the length, width, and height dimensions of the combustion zone and the burnout zone in the Cartesian coordinate system. Simultaneously, the central control unit reads the spatial position information of each sensor component, including the three-dimensional coordinates of each acoustic transceiver unit in the acoustic transceiver array. Geometric center coordinates of each electrode unit in the capacitance tomography sensor array And the coordinates of the installation origin of the matrix-type variable trajectory spray gun. In addition, the system loads environmental reference parameters, including the average background sound velocity in the furnace under cold air conditions. and the reference capacitance value between each electrode pair under no-load conditions Among them, the average background sound velocity The system pre-calculates and stores the values ​​based on the current ambient temperature and air gas constant, serving as a reference for subsequent acoustic temperature field reconstruction; the reference capacitance value... The inter-electrode capacitance value is measured when the furnace is filled with air and no combustion reaction is taking place. It is used to eliminate the basic influence of pipe wall and distributed capacitance.

[0038] In step S20, the system enters the measurement time slot. This invention employs a time-division multiplexing mechanism to coordinate the conflict between acoustic wave measurement and acoustic wave excitation functions in the acoustic channel. A high-precision timer is internally installed in the central control unit to set a complete control cycle. The control cycle is divided into non-overlapping measurement time slots. and execution slots In the measurement time slot Inside, the central control unit sends synchronous trigger signals to the acoustic transceiver array, the capacitance tomography sensor array, and the laser grid monitoring unit to initiate concurrent acquisition of multimodal data.

[0039] Step S20 further includes the following sub-steps: Step S21 involves the acquisition of acoustic wave data. The central control unit sends a first mode switching command to the acoustic transceiver array, controlling all acoustic transceiver units to operate in measurement mode. In this measurement mode, the acoustic transducer circuits in the array are configured for either low-power transmission or high-sensitivity reception. To accurately extract the test signal from the high-intensity background noise generated by furnace combustion, the system uses a binary pseudo-random coding sequence to modulate the transmitted signal. Each acoustic transceiver unit takes turns as a transmission source according to a preset logical timing sequence, transmitting acoustic pulses with specific coding characteristics into the furnace. The remaining acoustic transceiver units in reception mode acquire the acoustic signal after transmission through the furnace medium. After receiving the acoustic signal, the central control unit uses a cross-correlation algorithm to calculate the correlation peak between the received signal and the original transmitted coding sequence. Since background noise usually does not have the same autocorrelation characteristics as the transmitted sequence, by detecting the peak time of the correlation peak, the system can accurately lock the arrival time of the acoustic wave, thereby extracting the flight time on each effective acoustic path. Flight time The time required for a sound wave to travel through the furnace cross-section is characterized by its dependence on the temperature of the gas medium along the path. The specific circuit implementation for weak signal detection using the cross-correlation principle in this step is well-known in the art and will not be elaborated upon here. Finally, the system outputs an acoustic measurement vector containing the flight times of all valid paths. ; Step S22 involves the acquisition of electrical data. The central control unit controls the capacitance tomography sensor array to perform data scanning. The multi-channel data acquisition card in the capacitance tomography system sequentially selects a pair of electrode units as the excitation electrode and the measurement electrode, respectively, using a high-speed analog switch matrix in a polling manner. The remaining electrode units are grounded or in a floating state. During selection, the data acquisition card applies an AC sinusoidal excitation signal of a specific frequency (e.g., 100kHz to 1MHz) to the excitation electrode. Because the change in the dielectric constant of the medium inside the furnace causes a change in the coupling impedance between the electrodes, the induced current or voltage signal detected by the measurement electrode can be demodulated and converted into a capacitance value. The data acquisition card measures the mutual capacitance between the selected electrode pairs. This value reflects the equivalent capacitance determined by the gas-solid mixture within the space covered by the two electrode plates. The system traverses all independent electrode pair combinations to complete a full scan frame, obtaining the capacitance measurement vector. To eliminate the effects of distributed capacitance and circuit drift, the system further modifies the currently measured capacitance measurement vector. Compared with the reference capacitance value read in step S10 Perform differential processing to obtain the normalized capacitance vector used for subsequent inversion; Step S23 involves the acquisition of optical data. The central control unit controls the laser grid monitoring unit to perform spectral scanning. Each tunable semiconductor laser (TDLAS) in the laser emitter group is driven by a current controller, driving the laser diode in the form of a sawtooth wave superimposed with a high-frequency sine wave, so that its output wavelength periodically scans within a small band containing the characteristic absorption peaks of the gas to be measured. After the laser beam passes through the flue gas medium in the furnace, it is received by the laser receiver group on the opposite side. The photodetector in the laser receiver group converts the received transmitted light intensity signal into an electrical signal and transmits it to the lock-in amplifier. The lock-in amplifier demodulates using a reference signal with the same frequency as the modulation frequency to extract the second harmonic signal (2f signal). The amplitude of this second harmonic signal is proportional to the absorption intensity of the gas component and can suppress background noise interference. The system acquires the second harmonic signals on each cross optical path to construct a spectral measurement dataset. The spectral measurement dataset contains measurement data for the characteristic absorption peaks of nitrogen oxides, ammonia, and carbon monoxide. Through the above sub-steps S21 to S23, the system synchronously acquires the raw data of the three physical fields of sound, electricity and light within the measurement time slot, and transmits the acoustic measurement vector, capacitance measurement vector and spectral measurement dataset to the physical field solver to provide input data for subsequent three-dimensional reconstruction.

[0040] Step S30 in this embodiment specifically involves the inverse reconstruction and coupling correction process of the three-dimensional multiphysics field in the furnace. After acquiring the original multimodal data of sound, electricity, and light, the central control unit uses its built-in physics field solver to map the edge measurement data into the distribution of physical parameters inside the furnace through a mathematical inversion algorithm. Step S30 further includes the following sub-steps: Step S31 involves the inverse reconstruction of the acoustic temperature field. Based on the acoustic measurement vector obtained in step S20, the central control unit calculates the two-dimensional temperature field distribution data of each monitoring layer in the furnace. The propagation speed of sound waves in a gaseous medium has a thermodynamic dependence on the absolute temperature of the medium, which can be expressed as: ; in, Spatial location The speed of sound at that location; The adiabatic index of the gas; This is the universal gas constant; The absolute temperature distribution is to be determined. Let be the molar mass of the gas. To solve for this distribution, the system discretizes the measured cross-section of the furnace as follows: The grid cell, the first Time of flight measured by the acoustic wave path It equals the line integral of the slowness of sound (the reciprocal of the speed of sound) of each grid cell as the sound wave propagates along this path. The system constructs a system of linear equations describing the relationship between path length and slowness of sound. ,in This is a path length matrix. This represents the sound velocity vector. The system employs Algebraic Reconstruction Technique (ART) or Joint Algebraic Reconstruction Technique (SIRT) to iteratively solve the equations, obtaining the sound velocity values ​​of each grid cell within the cross-section, and then inversely deducing the two-dimensional temperature field distribution data. .

[0041] Step S32 involves the reverse reconstruction of the solid-phase flow field. The central control unit calculates the dielectric constant distribution data of the furnace cross-section based on the normalized capacitance vector obtained in step S20. The medium inside the furnace mainly consists of gaseous flue gas and solid fly ash particles. Since the dielectric constant of the solid particles is significantly higher than that of the gaseous medium, the equivalent dielectric constant of the furnace area can characterize the volume fraction of solid particles at that location. The material flow field follows the Poisson equation for the electrostatic field, and the system pre-calculates the sensitivity matrix. This matrix describes the weighted impact of minute variations in the dielectric constant of each grid cell on the boundary capacitance measurement. The system uses the Linear Back Projection (LBP) algorithm or the Landweber Iterative Algorithm to solve the matrix equation and reconstruct the dielectric constant distribution of the furnace cross-section. The dielectric constant distribution data characterizes the solid-phase flow field state within the furnace, where the high dielectric constant value region corresponds to the high concentration region where intense combustion leads to the large-scale aspiration of particulate matter; Step S33 involves gas concentration field reconstruction based on acoustic-optical coupling. The central control unit utilizes the two-dimensional temperature field distribution data reconstructed in sub-step S31. The spectral measurement dataset obtained in step S20 is corrected point-by-point to eliminate the inversion error of concentration caused by the non-uniform temperature field. According to Beer-Lambert's law, the absorption intensity of a gas component to a specific wavelength of laser light depends on the spectral line intensity, which is a function of temperature. The system records each optical path through which the laser passes. Discretize the segments based on the grid cells it passes through. A discrete infinitesimal element, the system obtains data from the two-dimensional temperature field distribution. The index displays the local temperature value corresponding to that location. Then, the spectral line intensity at that position is corrected using the following formula: ; in, The local temperature along the optical path is spectral line intensity at time; For reference temperature; The spectral line intensity is at the reference temperature; is the partition function of the molecule; It is Planck's constant; The speed of light; To transition to lower energy levels; Here, represents the Boltzmann constant. The corrected total absorption coefficient of the optical path is calculated through integration, constructing a coefficient matrix that includes temperature correction information. The system combines absorbance measurements from each intersecting optical path and uses either the Maximum Likelihood Expectation-Maximization (MLEM) algorithm or an algebraic reconstruction algorithm to calculate the gas concentration value within each grid cell, generating gas concentration field distribution data. The gas concentration field distribution data specifically includes the concentration distributions of nitrogen oxides, ammonia, and carbon monoxide. Step S34 involves the synthesis of three-dimensional physical field data. Sub-steps S31 to S33 are performed independently on the cross-sections corresponding to the sensor arrays at different height levels within the furnace. The central control unit synthesizes the two-dimensional temperature field distribution data, dielectric constant distribution data, and gas concentration field distribution data of each level according to their corresponding vertical height coordinates. Spatial stacking is performed. For the blank areas between sensor layers, the system uses a three-dimensional interpolation algorithm (such as trilinear interpolation) to complete the data, and finally generates a three-dimensional temperature field, a three-dimensional dielectric constant field, and a three-dimensional gas concentration field covering the entire furnace monitoring area.

[0042] Step S40 in this embodiment specifically involves a reaction kinetics region diagnosis process based on multi-field feature fusion. In this step, the central control unit uses the three-dimensional physical field data generated in step S30, which includes temperature, solid particle distribution, and gas component concentration, to identify key feature regions within the furnace that require physical intervention through spatial mapping and Boolean logic operations. Step S40 includes the following sub-steps: In step S41, the central control unit performs spatial registration and meshing of the multiphysics data. Due to the different spatial resolutions of acoustic thermometry, capacitance tomography, and laser mesh monitoring, the system pre-establishes a unified three-dimensional Cartesian coordinate system with the center of the furnace bottom as the origin, and divides the furnace interior space into uniformly sized voxels. The central control unit constructs a coordinate transformation matrix based on the geometric installation position and field of view of each sensor, mapping the three-dimensional temperature field, three-dimensional dielectric constant field, and three-dimensional gas concentration field generated in step S30 to this unified coordinate system. For mesh nodes of different resolutions, the system uses a trilinear interpolation algorithm for resampling, adjusting the coordinates of each voxel. The above forms a temperature scalar scalar of dielectric constant and gas concentration vector The data set; In step S42, the central control unit identifies and locates the potential region for nitrogen oxide (NOx) formation. The potential region for NOx formation is the area within the furnace where the temperature is higher than a set value and solid-phase particle turbulence is intense. The central control unit traverses the three-dimensional data matrix and calculates the spatial gradient magnitude of the dielectric constant at each voxel location. For discrete 3D mesh data, the system uses the central difference method to calculate the gradient magnitude, and the calculation formula is as follows: ; in, The current voxel is in The dielectric constant of adjacent voxels along the axial direction; Let be the physical dimension of a unit voxel along each coordinate axis. The system then applies the following Boolean logic criterion to identify potential regions: ; in, This is a binary identifier bit; The preset high temperature threshold; This is a preset dielectric constant gradient threshold. The system extracts all... Given a set of voxels, the spatial centroid coordinates of the region are calculated using the weighted centroid method. : ; in, For the first The position vectors of voxels that satisfy the conditions; This represents the temperature value of the voxel. Centroid coordinates in space. The coordinates of the aiming target point are stored as those of a matrix-type variable trajectory spray gun. In step S43, the central control unit identifies and locates the mixing dead zone. The mixing dead zone is an area where there is excess reducing agent but nitrogen oxides are not effectively removed. Based on gas concentration field data, the central control unit filters voxel regions that simultaneously meet the following two conditions: Condition 1: Ammonia concentration at this voxel location Greater than the preset ammonia escape threshold ; Condition 2: Nitrogen oxide concentration at this voxel location Emissions exceeding the preset threshold The system marks voxels that meet the above conditions as candidate points for the mixing dead zone. If multiple disconnected candidate regions exist in space, the system uses a connected component labeling algorithm to identify the largest connected region as the main mixing dead zone and calculates the geometric center coordinates of this main mixing dead zone. Geometric center coordinates The target point is stored as the beam focusing point of the acoustic array in excitation mode. Through the above steps, the system completes the calculation from physical field data to control target coordinates and sets the spatial center coordinates. and geometric center coordinates Transmitted to the execution control module.

[0043] In this embodiment, steps S50 and S60 specifically involve the acoustic-chemical co-location targeted suppression and hybrid enhancement process performed within the execution time slot. This occurs when the system clock enters the execution time slot. At this time, the central control unit actively intervenes in the physical field within the furnace based on the diagnostic results of step S40. Step S50 involves matrix-style variable trajectory precision injection targeting the nitrogen oxide generation potential zone. The central control unit first determines whether a marked nitrogen oxide generation potential zone exists. If it exists, the system determines the region's spatial centroid coordinates. From the matrix-type variable trajectory spray gun array, one or more spray guns with the closest Euclidean distance to the centroid coordinate are selected as the actuators. To achieve coverage of the high-temperature core region by the reducing agent, the system uses an inverse kinematics algorithm to calculate the target control parameters of the spray gun's dual-axis adjustment mechanism. A local coordinate system is established with the spray gun nozzle as the origin, and the nozzle installation coordinates of the selected spray gun are set as follows: Then its horizontal deflection angle and vertical pitch angle The calculation formula is as follows: ; ; Meanwhile, the central control unit uses the average temperature value within the potential zone. Based on the dielectric constant gradient value, the target injection flow rate of the reducing agent is calculated using a pre-stored stoichiometric model. The system sends the calculated angle and flow commands to the corresponding dual-axis adjustment mechanism and flow control valve. The dual-axis adjustment mechanism drives the spray nozzle to rotate to the target position, and the flow control valve opens to match the target spray flow rate, thereby delivering the high-concentration reducing agent solution to the most intense combustion zone and suppressing thermal combustion. The generation of the dual-axis regulating mechanism and the flow regulating valve is well-known in the field and will not be described in detail here.

[0044] Step S60 involves acoustic turbulence mixing enhancement for the mixing dead zone. If a mixing dead zone is identified in step S40, the central control unit sends a second mode switching command to the acoustic transceiver array, switching the array's operating state from "measurement mode" to "excitation mode". In excitation mode, the signal conditioning circuit at the front end of the acoustic transceiver unit switches to a high-power drive branch via a relay or analog switch. This branch includes a programmable gain power amplifier capable of boosting the drive voltage to 10 to 100 times that in measurement mode, enabling the acoustic transceiver unit to emit high-intensity single-frequency or narrowband modulated acoustic waves; To overcome the mass transfer constraint within the mixing dead zone without using a mechanical stirring device, this embodiment utilizes the principle of coherent superposition of sound waves and the acoustic flow effect. The central control unit determines the mixing dead zone based on its geometric center coordinates. The beamforming algorithm is used to calculate the acoustic wave emitting unit in the array. The amount of phase delay to be applied The calculation of this phase delay is intended to ensure that the wavefronts of the sound waves emitted by all transmitting units are at the target point. The formula for calculating the superposition of elements in phase is: ; in, The frequency of the excitation sound wave is preferably set to a low frequency range of 50Hz to 500Hz to reduce attenuation in the medium. For the first Each sound wave emitting unit reaches the target point. The straight-line distance The average sound velocity within the furnace is given. Each sound wave emitting unit emits sound waves according to the calculated phase delay, forming a high-intensity focusing zone at the mixing dead zone. Within this focusing zone, the spatially non-uniform distribution of sound wave energy density generates a non-linear acoustic radiation force, driving the fluid to produce macroscopic directional flow. This acoustic-flow driving force... With sound energy density The spatial gradient is proportional to the gradient, and can be expressed as: ; in, , This is the effective value of the sound pressure level; The density of the gas medium is denoted as ρ. This volume force, generated by the acoustic field gradient, acts on the flue gas medium, inducing secondary turbulent vortices within the mixing dead zone. These turbulent vortices disrupt the laminar boundary layer, enhancing momentum exchange and mass transport between the reducing agent and flue gas molecules, and promoting the reaction of unreacted ammonia with residual nitrogen oxides. After completing the above measurements and operations, the system ends the current control cycle and resets the timer to enter the measurement slot of the next cycle.

[0045] In summary, this embodiment constructs a spatiotemporal data base for acoustic-optical-electrical multimodal sensing, utilizes multi-physics field coupling correction and feature fusion diagnostic technology to achieve accurate analysis of the furnace combustion state, and drives the matrix spray gun and acoustic array to perform time-division multiplexing targeted intervention, thereby forming a complete closed-loop feedback mechanism from state perception to decision execution.

Claims

1. An intelligent emission control and diagnostic system for waste incineration power plants, characterized in that, include: Acoustic transceiver array, capacitance tomography sensor array, laser grid monitoring unit, matrix variable trajectory spray gun array, and central control unit; The central control unit is communicatively connected to the acoustic transceiver array, the capacitance tomography sensor array, the laser grid monitoring unit, and the matrix variable trajectory spray gun array, respectively. The central control unit is equipped with time-division multiplexing logic, which divides the system's working cycle into non-overlapping measurement time slots and execution time slots; During the measurement time slot, the central control unit sends a synchronous trigger signal to the acoustic transceiver array, the capacitance tomography sensor array, and the laser grid monitoring unit to acquire acoustic measurement vectors, capacitance measurement vectors, and spectral measurement datasets. It then uses a physical field inversion algorithm to reconstruct the three-dimensional temperature field distribution data, three-dimensional dielectric constant field distribution data, and three-dimensional gas concentration field distribution data within the furnace. During the execution time slot, the central control unit performs multi-field feature fusion diagnosis based on the three-dimensional temperature field distribution data, the three-dimensional dielectric constant field distribution data, and the three-dimensional gas concentration field distribution data to identify the nitrogen oxide generation potential zone and the mixing dead zone, and accordingly sends a spray control command to the matrix variable trajectory spray gun array and a vibration mode switching command to the acoustic transceiver array.

2. The intelligent emission control and diagnostic system for waste incineration power plants according to claim 1, characterized in that, When reconstructing the three-dimensional temperature field distribution data, the three-dimensional dielectric constant field distribution data, and the three-dimensional gas concentration field distribution data, the central control unit specifically performs the following operations: Based on the acoustic measurement vector, the sound velocity value of the furnace cross-section grid element is calculated using algebraic reconstruction technology or joint algebraic reconstruction technology, and the two-dimensional temperature field distribution data is obtained by solving the thermodynamic dependence of sound velocity on absolute temperature. Based on the capacitance measurement vector, the inverse problem of the electrostatic field Poisson equation is solved using the pre-calculated sensitivity matrix and the linear back projection algorithm, and the two-dimensional dielectric constant distribution data reflecting the volume fraction of solid particles is obtained. Based on the spectral measurement dataset, after acoustic-optical coupling correction in combination with the two-dimensional temperature field distribution data, the two-dimensional gas concentration field distribution data is obtained by using the maximum likelihood expectation maximization algorithm. The two-dimensional temperature field distribution data, the two-dimensional dielectric constant distribution data, and the two-dimensional gas concentration field distribution data of each monitoring layer are spatially stacked and three-dimensionally interpolated according to the vertical height coordinate to generate the three-dimensional temperature field distribution data, the three-dimensional dielectric constant field distribution data, and the three-dimensional gas concentration field distribution data.

3. The intelligent emission control and diagnostic system for waste incineration power plants according to claim 2, characterized in that, The acoustic-optical coupling correction based on the two-dimensional temperature field distribution data specifically includes: Each optical path of the laser grid monitoring unit is discretized and segmented according to the grid unit it passes through; The local temperature values ​​corresponding to the positions of each discrete micro-element in the optical path are indexed from the two-dimensional temperature field distribution data. The intensity of the laser spectral line at the corresponding location is corrected using the local temperature value. The correction formula covers the partition function term and exponential term of the spectral line intensity as a function of temperature. Based on the corrected spectral line intensities and the absorbance measurements in the spectral measurement dataset, a coefficient matrix containing temperature correction information is constructed for the inversion of the two-dimensional gas concentration field distribution data.

4. The intelligent emission control and diagnostic system for waste incineration power plants according to claim 1, characterized in that, The acoustic transceiver unit in the acoustic transceiver array is connected to a drive circuit that includes a mode switching switch. The central control unit is configured as follows: During the measurement time slot, the mode switching switch is controlled to connect the low-power transmission and high-sensitivity reception branches, the transmission signal is modulated using a binary pseudo-random coding sequence, and the flight time is extracted based on a cross-correlation algorithm. During the execution time slot, if the excitation mode switching command is received, the mode switching switch is controlled to connect the high-power drive branch, and the programmable gain power amplifier is used to drive the acoustic transceiver unit to transmit high-intensity acoustic waves with single frequency or narrowband modulation.

5. The intelligent emission control and diagnostic system for waste incineration power plants according to claim 1, characterized in that, When the central control unit identifies the potential region for nitrogen oxide formation, it performs the following operations: Establish a unified three-dimensional Cartesian coordinate system and divide the internal space of the furnace into unit voxels; The spatial gradient magnitude of the dielectric constant at each voxel location was calculated using the central difference method. Determine whether each voxel simultaneously meets the following conditions: the voxel's temperature value is greater than a preset high temperature threshold, and the voxel's dielectric constant spatial gradient magnitude is greater than a preset dielectric constant gradient threshold. The set of voxels that simultaneously meet the above conditions is marked as the nitrogen oxide generation potential region, and the spatial centroid coordinates of the region are calculated using the weighted centroid method.

6. The intelligent emission control and diagnostic system for waste incineration power plants according to claim 1, characterized in that, When identifying the mixed dead zone, the central control unit traverses the three-dimensional gas concentration field distribution data and determines whether each voxel simultaneously meets the following conditions: the ammonia concentration of the voxel is greater than the preset ammonia escape threshold, and the nitrogen oxide concentration of the voxel is greater than the preset emission threshold; voxels that simultaneously meet the above conditions are marked as candidate points for the mixed dead zone, and the connected component labeling algorithm is used to identify the largest connected region as the mixed dead zone, and the geometric center coordinates of the region are calculated.

7. The intelligent emission control and diagnostic system for waste incineration power plants according to claim 5, characterized in that, The central control unit sends spray control commands to the matrix-type variable trajectory spray gun array, including: From the matrix-type variable trajectory spray gun array, select the spray gun that is closest to the spatial centroid coordinate Euclidean distance of the nitrogen oxide generation potential zone; A local coordinate system is established with the nozzle of the spray gun as the origin. The horizontal deflection angle and vertical pitch angle of the spray gun are calculated based on the spatial centroid coordinates using an inverse kinematics algorithm. Based on the average temperature value within the nitrogen oxide generation potential zone, the target injection flow rate of the reducing agent is calculated using a stoichiometric model. The horizontal deflection angle, the vertical pitch angle, and the target spray flow rate are sent as spray control commands to the dual-axis adjustment mechanism and flow control valve of the spray gun.

8. The intelligent emission control and diagnostic system for waste incineration power plants according to claim 6, characterized in that, The central control unit sends an excitation mode switching command to the acoustic transceiver array, specifically including: Based on the geometric center coordinates of the hybrid dead zone, the phase delay of each acoustic transceiver unit in the acoustic transceiver array is calculated using a beamforming algorithm. The calculation of the phase delay is such that the wavefronts of the acoustic waves emitted by each acoustic transceiver unit are superimposed in phase at the geometric center coordinates. The acoustic transceiver array is controlled to emit high-intensity acoustic waves according to the phase delay, forming an acoustic energy density focusing region in the mixing dead zone, and using the acoustic flow driving force generated by the acoustic energy density gradient to induce secondary turbulent vortices.

9. The intelligent emission control and diagnostic system for waste incineration power plants according to claim 1, characterized in that, The capacitance tomography sensor array includes multiple pairs of electrode units; During the measurement time slot, the central control unit controls the multi-channel data acquisition card to sequentially select a pair of electrode units as excitation electrodes and measurement electrodes respectively through an analog switch matrix. An AC sinusoidal excitation signal is applied to the excitation electrode, and the mutual capacitance value of the measurement electrode is measured. The measured mutual capacitance value is differentially processed with the pre-stored reference capacitance value under no-load conditions to generate the normalized capacitance measurement vector.

10. The intelligent emission control and diagnostic system for waste incineration power plants according to claim 1, characterized in that, The laser grid monitoring unit includes a laser emitter group and a laser receiver group; The tunable semiconductor laser in the laser emitter group is configured to periodically scan the output wavelength within a band that includes the characteristic absorption peak of the gas to be measured. The laser receiver group is equipped with a lock-in amplifier for extracting the second harmonic signal from the received light intensity signal and transmitting the second harmonic signal as the spectral measurement dataset to the central control unit.

Images