Garbage incinerator combustion control method and device based on frequency domain disturbance analysis, equipment and medium

Abstract

The application discloses a waste incinerator combustion control method and device based on frequency domain disturbance analysis, equipment and medium, relates to the technical field of waste incinerator combustion control, extracts the region of interest image from the waste incinerator hearth flame image;The region of interest image is pretreated, the two-dimensional fast Fourier transform is carried out to the pretreated image, the frequency domain matrix is obtained, and the power spectrum is calculated;The frequency spectrum region in the power spectrum is divided, the frequency spectrum region in different frequency ranges is obtained, the single frame flame energy ratio in the preset frequency range is determined by using the frequency spectrum region in different frequency ranges;Based on the single frame flame energy ratio in the statistical window, the window statistical characteristics are determined;The combustion disturbance grade is determined based on the flame frequency domain disturbance index, the control instruction is generated by using the combustion disturbance grade, the waste incinerator is controlled, the lag control mode is avoided, the single frame image misjudgment is avoided, and the complete dependence on artificial experience is avoided, and the stability of waste incinerator combustion control is improved.

Description

Technical Field

[0001] This invention relates to the field of waste incinerator combustion control technology, and in particular to a waste incinerator combustion control method, device, equipment and medium based on frequency domain perturbation analysis. Background Technology

[0002] The operation of municipal solid waste incinerators is inherently unstable. Even within the same incineration line, the waste entering the furnace at different times can vary significantly. For example, a higher proportion of kitchen waste leads to increased moisture content, while a higher proportion of plastics and paper results in a higher instantaneous calorific value. Large pieces of wet waste entering the grate can easily cause localized flame weakening or temporary flameout. These fluctuations directly affect the combustion state in the furnace, easily causing flame flickering, incomplete combustion in certain areas, furnace temperature fluctuations, and unstable steam parameters.

[0003] Currently, common control parameters for incinerators include furnace temperature, flue gas oxygen content, carbon monoxide concentration, steam flow rate, grate velocity, and primary air volume. While these parameters reflect the combustion state, most are combustion outcome parameters. For example, a drop in furnace temperature indicates that incomplete combustion has persisted for some time; a significant increase or decrease in flue gas oxygen content indicates a deviation between the oxygen supply and combustion state within the furnace. Therefore, relying solely on these parameters for control easily leads to a lag control pattern of "anomaly has occurred—parameters change—system readjusts."

[0004] In actual operation, operators will judge the combustion status by observing the flame pattern, such as whether the flame is continuous, whether it jumps violently, whether there are local dark areas, and whether the flame is shooting upwards. However, this judgment mainly relies on human experience and it is difficult to form a stable, quantifiable and repeatable control logic.

[0005] As can be seen from the above, how to avoid the formation of a lag control mode, misjudgment of single-frame images, and complete reliance on human experience, and improve the stability of combustion control in waste incinerators are problems that need to be solved in this field. Summary of the Invention

[0006] In view of this, the purpose of this invention is to provide a method, apparatus, equipment, and medium for combustion control of waste incinerators based on frequency domain perturbation analysis, which can avoid the formation of lag-type control modes, misjudgment of single-frame images, and complete reliance on human experience, thereby improving the stability of combustion control in waste incinerators. The specific solution is as follows: In a first aspect, this application discloses a combustion control method for a waste incinerator based on frequency domain perturbation analysis, comprising: Acquire images of the furnace flames of a waste incinerator, and extract regions of interest from the furnace flame images; The region of interest image is preprocessed to obtain a preprocessed image. A two-dimensional fast Fourier transform is performed on the preprocessed image to obtain a frequency domain matrix. The power spectrum is calculated based on the frequency domain matrix. Based on the spectral center point and each frequency domain point of the power spectrum, the spectral region in the power spectrum is divided to obtain spectral regions of different frequency ranges. Using the spectral regions of different frequency ranges, the single-frame flame energy ratio under the preset frequency range is determined. Based on the flame energy ratio of a single frame within the statistical window, the window statistical characteristics are determined; the window statistical characteristics include the mean energy, standard deviation of energy, and rate of change between adjacent frames within a preset frequency range. Using the window statistical features and preset benchmark values, the flame frequency domain disturbance index is determined, the combustion disturbance level is determined based on the flame frequency domain disturbance index, control commands are generated using the combustion disturbance level, and the combustion control of the waste incinerator is performed using the control commands.

[0007] Optionally, acquiring an image of the furnace flame of a waste incinerator and extracting a region of interest image from the furnace flame image includes: An industrial camera is used to capture video images of the furnace flames; the industrial camera is installed at the furnace observation port or above the grate of the waste incinerator. Frame images are extracted and analyzed from the video images of the furnace flame at preset time intervals. Extract the combustion zone image above the front section of the grate from the analysis frame image; The image of the combustion zone above the front section of the grate is taken as the region of interest.

[0008] Optionally, the image of the region of interest is preprocessed to obtain a preprocessed image, including: Calculate the proportion of effective pixels in the region of interest image; If the effective pixel ratio is greater than the preset pixel threshold, then the region of interest image is subjected to grayscale processing, median filtering, size scaling processing, and brightness normalization processing to obtain a normalized image. A two-dimensional window function is applied to the normalized image to obtain the preprocessed image.

[0009] Optionally, based on the spectral center point and each frequency domain point of the power spectrum, the spectral region in the power spectrum is divided, including: Calculate the radial distance from each frequency point in the power spectrum to the center point of the spectrum; The spectral regions in the power spectrum are divided based on the radial distance and the preset maximum effective radius.

[0010] Optionally, determining the single-frame flame energy ratio within a preset frequency range using spectral regions of different frequency ranges includes: Determine the target frequency range and the effective frequency range within the preset frequency range from different frequency range frequency ranges; The target spectral energy within a preset frequency range is determined using the power spectral values ​​at each frequency point within the target spectral region. The total effective spectral energy is determined using the power spectral values ​​at each frequency point within the effective spectral region. The single-frame flame energy ratio is calculated based on the target spectral energy and the total effective spectral energy.

[0011] Optionally, the flame frequency domain disturbance index is determined using the window statistical characteristics and a preset benchmark value, including: When the waste incinerator is in a stable operating state, a preset reference value is determined based on the furnace flame image within a preset time period; the preset reference value includes the average reference energy, the reference energy fluctuation value, and the reference adjacent frame change rate; Calculate the first ratio between the average energy value within the preset frequency range and the average reference energy value; Calculate the second ratio between the energy standard deviation within the preset frequency range and the reference energy fluctuation value; Calculate the third ratio between the rate of change of adjacent frames within the preset frequency range and the rate of change of adjacent frames at the reference frequency range; The flame frequency domain perturbation index is obtained by weighting the first ratio, the second ratio, and the third ratio.

[0012] Optionally, the combustion disturbance level is determined based on the flame frequency domain disturbance index, control commands are generated using the combustion disturbance level, and combustion control of the waste incinerator is performed using the control commands, including: The flame frequency domain disturbance index is compared with a preset disturbance threshold, and the combustion disturbance level is determined based on the comparison result; Control commands are generated based on the combustion disturbance level; the control commands include a first control command, a second control command, or a third control command; wherein, the first control command is a command for adjusting the primary air fan frequency and grate speed; the second control command is a command for adjusting the primary air fan frequency, grate speed, and feeding rhythm; and the third control command is a command for coordinating the adjustment of the primary air fan frequency, grate speed, feeding rhythm, and secondary air frequency.

[0013] Secondly, this application discloses a waste incinerator combustion control device based on frequency domain perturbation analysis, comprising: The image extraction module is used to acquire images of the furnace flames of a waste incinerator and extract the region of interest from the furnace flame images. The preprocessing module is used to preprocess the image of the region of interest to obtain a preprocessed image, perform a two-dimensional fast Fourier transform on the preprocessed image to obtain a frequency domain matrix, and calculate the power spectrum based on the frequency domain matrix. The energy ratio determination module is used to divide the spectral region in the power spectrum based on the spectral center point and each frequency domain point of the power spectrum to obtain spectral regions of different frequency ranges, and use the spectral regions of different frequency ranges to determine the single-frame flame energy ratio under a preset frequency range. The feature determination module is used to determine the window statistical features based on the flame energy ratio of the single frame within the statistical window; the window statistical features include the energy mean, energy standard deviation and adjacent frame change rate within a preset frequency range; The combustion control module is used to determine the flame frequency domain disturbance index using the window statistical features and preset benchmark values, determine the combustion disturbance level based on the flame frequency domain disturbance index, generate control commands using the combustion disturbance level, and use the control commands to control the combustion of the waste incinerator.

[0014] Thirdly, this application discloses an electronic device, including: Memory, used to store computer programs; A processor is used to execute the computer program to implement the aforementioned waste incinerator combustion control method based on frequency domain perturbation analysis.

[0015] Fourthly, this application discloses a computer-readable storage medium for storing a computer program; wherein, when the computer program is executed by a processor, it implements the steps of the aforementioned disclosed waste incinerator combustion control method based on frequency domain perturbation analysis.

[0016] Beneficial Effects: This application provides a method for combustion control of waste incinerators based on frequency domain perturbation analysis. This includes acquiring images of the furnace flames of the waste incinerator, extracting regions of interest (ROIs) from the furnace flame images, thus improving the targeting and accuracy of combustion state judgment and effectively reducing interference from factors such as furnace wall reflection and observation window contamination on image analysis. The ROI images are preprocessed to obtain preprocessed images, and a two-dimensional fast Fourier transform is performed on the preprocessed images to obtain a frequency domain matrix. The power spectrum is calculated based on the frequency domain matrix, thus suppressing noise such as smoke particles and instantaneous bright spots, and reducing the impact of camera exposure conditions, observation window contamination, and overall flame intensity changes on the analysis results, ensuring the comparability of image features at different times. Based on the spectral center point and each frequency domain point of the power spectrum, the spectral regions in the power spectrum are divided to obtain spectral regions of different frequency ranges. This method utilizes... Within the same frequency range, the single-frame flame energy ratio is determined within a preset frequency range, thus eliminating the influence of overall flame brightness fluctuations and achieving precise quantification of the degree of single-frame flame disturbance. Based on the single-frame flame energy ratio within a statistical window, window statistical characteristics are determined, effectively avoiding misjudgments caused by instantaneous flame jumps and single-frame image anomalies, improving the stability and reliability of combustion state recognition. Using window statistical characteristics and preset benchmark values, the flame frequency domain disturbance index is determined, and the combustion disturbance level is determined based on the flame frequency domain disturbance index. The combustion disturbance level is used to generate control commands, which are then used to control the combustion of the waste incinerator. This eliminates reliance on manual experience, avoids excessive adjustments that could cause new combustion fluctuations or control oscillations, overcomes the lag of existing thermal parameter feedback adjustments, suppresses problems such as furnace flame shaking, furnace temperature fluctuations, and unstable steam parameters, and improves the combustion stability of the waste incinerator. Attached Figure Description

[0017] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on the provided drawings without creative effort.

[0018] Figure 1 This is a flowchart of a waste incinerator combustion control method based on frequency domain perturbation analysis disclosed in this application; Figure 2 This is a schematic diagram of a waste incinerator combustion control device based on frequency domain perturbation analysis disclosed in this application; Figure 3 This is a schematic diagram of the structure of an electronic device disclosed in this application. Detailed Implementation

[0019] 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.

[0020] The operation of municipal solid waste incinerators exhibits significant instability. Even within the same incineration line, the waste entering the furnace at different times can vary considerably. For example, a higher proportion of kitchen waste leads to increased moisture content, while a higher proportion of plastics and paper results in a higher instantaneous calorific value. Large pieces of wet waste entering the grate can easily cause localized flame weakening or temporary flameout. These fluctuations directly affect the combustion state in the furnace, easily causing flame flickering, incomplete combustion, furnace temperature fluctuations, and unstable steam parameters. Currently, common control parameters for incinerators mainly include furnace temperature, flue gas oxygen content, carbon monoxide concentration, steam flow rate, grate velocity, and primary air volume. While these parameters reflect the combustion state, most are combustion outcome parameters. For example, a drop in furnace temperature indicates that incomplete combustion has persisted for some time; a significant increase or decrease in flue gas oxygen content indicates a deviation between the oxygen supply and combustion state within the furnace. Therefore, relying solely on these parameters for control easily leads to a lag control mode of "anomaly has occurred—parameters change—system readjusts." In actual operation, operators judge the combustion status by observing the flame pattern, such as whether the flame is continuous, whether it flickers violently, whether there are local dark areas, and whether the flame is shooting upwards. However, this judgment mainly relies on human experience and is difficult to form a stable, quantifiable, and repeatable control logic. As can be seen from the above, how to avoid the formation of a lagging control mode, misjudgment of single-frame images, and complete reliance on human experience, and improve the stability of combustion control in waste incinerators, is a problem that needs to be solved in this field.

[0021] See Figure 1 As shown in the figure, this invention discloses a combustion control method for a waste incinerator based on frequency domain perturbation analysis, which may specifically include: Step S11: Obtain an image of the furnace flame of the waste incinerator, and extract the region of interest image from the furnace flame image.

[0022] In this embodiment, an industrial camera is used to acquire video images of the furnace flames. The industrial camera is positioned at the furnace observation port or above the grate of the waste incinerator. Analytical frames are extracted from the video images of the furnace flames at preset time intervals. The combustion area image above the front section of the grate is extracted from the analytical frames. The combustion area image above the front section of the grate is used as the region of interest image.

[0023] In this step, a high-temperature resistant industrial camera is installed at a visible position above the grate or through the observation port of the waste incinerator. The industrial camera is protected by a quartz glass shield, and a compressed air purging structure is installed outside the shield. The purging pressure is preferably 0.2 MPa to 0.5 MPa to reduce the obstruction of the viewing window by fly ash, soot, and coking. The industrial camera uses a fixed exposure mode, with an exposure time preferably set to 3 ms to 8 ms to avoid automatic exposure causing changes in flame brightness to be misinterpreted as changes in combustion status. The sampling frame rate of the industrial camera is set to 15 fps to 30 fps, preferably 20 fps, and the image resolution is set to 1280×720 or 1920×1080. Image data is transmitted to the edge computing controller via an industrial Ethernet network.

[0024] In this embodiment, during the commissioning phase, operators or engineers mark the main combustion area in the image. The flame area above the front section of the grate can be selected as the ROI (Region of Interest), avoiding the furnace wall, the edge of the observation window, obvious reflective points, and fixed structural components. The ROI can be set to 50% to 80% of the original image width and 40% to 70% of the height. For example, in a 1280×720 image, an area with horizontal coordinates from 200 to 1080 and vertical coordinates from 180 to 620 can be selected as the initial ROI. The system allows adjustment of the ROI position according to different furnace types, but the ROI position remains fixed within the same operating cycle to ensure comparability of frequency domain feature calculations.

[0025] Although the industrial camera captures video at 20fps, the system does not need to perform full frequency domain calculations for every frame. For example, the system extracts one frame as an analysis frame every 0.5 seconds, processing 2 frames per second. With a 10-second analysis window, each window contains 20 frames for analysis. The system updates the analysis window every 2 seconds, meaning the current window partially overlaps with the previous window. This reduces computational load while ensuring good continuity in combustion disturbance identification.

[0026] Step S12: Preprocess the image of the region of interest to obtain a preprocessed image, perform a two-dimensional fast Fourier transform on the preprocessed image to obtain a frequency domain matrix, and calculate the power spectrum based on the frequency domain matrix.

[0027] In this embodiment, the effective pixel ratio in the region of interest image is calculated; if the effective pixel ratio is greater than a preset pixel threshold, the region of interest image is subjected to grayscale processing, median filtering, size scaling, and brightness normalization to obtain a normalized image; a two-dimensional window function is applied to the normalized image to obtain a preprocessed image.

[0028] In this embodiment, the system first calculates the effective brightness ratio within the region of interest. Pixels with grayscale values ​​greater than 40 and less than 245 are defined as effective pixels. The proportion of effective pixels to the total number of pixels in the region of interest is calculated and denoted as the effective pixel ratio. .

[0029] when If the image is severely obscured by smoke or dust, the lens is contaminated, or the flame area is not visible, this frame will not participate in the frequency domain perturbation calculation, and a "insufficient image quality" flag will be output to the system; when When the frame is in a certain state, it can participate in the calculation, but its weight in the window statistics is reduced; when... At this point, the frame is considered a valid analysis frame. This step can reduce the impact of lens contamination or instantaneous smoke on the judgment results.

[0030] In this embodiment, the system performs grayscale processing on the effective region of interest image. The grayscale processing formula is as follows: ; in, For pixels The brightness value after grayscale conversion For pixels The red channel value, For pixels The green channel value, For pixels The blue channel value, x The horizontal coordinates of the image pixels. y This represents the vertical coordinate of a pixel in the image.

[0031] After grayscale conversion, a 3×3 median filter is applied to the image to reduce the impact of smoke particles and transient bright spots on the calculation results. The image is then uniformly scaled to 256×256 pixels. Standardizing the image size ensures that the frequency domain calculation results at different times and resolutions have a consistent frequency scale, facilitating subsequent threshold comparisons.

[0032] Because the overall brightness of the flame may be affected by camera window contamination, flame intensity variations, and exposure conditions, the system performs brightness normalization processing on the grayscale image. Let the average grayscale value of the current region of interest be... The standard deviation of grayscale is Then the normalized image can be represented as: ; in, For pixels The corresponding normalized image values, For pixels The brightness value after grayscale conversion The mean gray level of the current region of interest. The standard deviation of the grayscale values ​​of the current region of interest. It is a very small constant.

[0033] This processing makes the system pay more attention to changes in flame texture, rather than just changes in overall brightness.

[0034] To reduce frequency domain leakage caused by region-of-interest truncation, the system applies a two-dimensional Hanning window to the normalized image, resulting in a preprocessed image: ; in, The image values ​​after preprocessing by applying a two-dimensional Hanning window. This represents the normalized image value corresponding to pixel (x, y). This is a two-dimensional Hanning window function. This step can make the frequency domain energy distribution more stable and reduce the impact of boundary abrupt changes on high-frequency energy calculations.

[0035] In this embodiment, the system performs a two-dimensional FFT (Fast Fourier Transform) on the preprocessed image to obtain the frequency domain matrix. The formula for the two-dimensional Fast Fourier Transform is as follows: ; in, For frequency domain points The corresponding frequency domain matrix values, For pixels The corresponding preprocessed image values, M The horizontal dimension of the image. N The vertical dimension of the image. j The imaginary unit, Pi e It is a natural constant.

[0036] The power spectrum is calculated based on the frequency domain matrix, and the power spectrum formula is as follows: ; in, For frequency domain points The corresponding power spectral density value. After calculation, the low-frequency center is moved to the center of the spectrum to facilitate subsequent division of frequency intervals according to radial distance.

[0037] Step S13: Based on the spectral center point and each frequency domain point of the power spectrum, divide the spectral region in the power spectrum to obtain spectral regions of different frequency ranges, and use the spectral regions of different frequency ranges to determine the single-frame flame energy ratio under the preset frequency range.

[0038] In this embodiment, the radial distance from each frequency point in the power spectrum to the center point of the spectrum is calculated; based on the radial distance and the preset maximum effective radius, the spectral region in the power spectrum is divided.

[0039] In this embodiment, a target frequency range and an effective frequency range are determined from the frequency range regions of different frequency ranges; the target frequency range energy is determined using the power spectrum values ​​of each frequency point in the target frequency range; the total effective frequency range energy is determined using the power spectrum values ​​of each frequency point in the effective frequency range; and the single-frame flame energy ratio is calculated based on the target frequency range energy and the total effective frequency range energy.

[0040] In this embodiment, the center point of the spectrum is set as The radial distance from any frequency domain point to the center point is: ; in, The radial distance from a point in the frequency domain to the center point of the spectrum. The horizontal coordinates of the frequency domain points are... The vertical coordinate of the frequency domain point. The horizontal coordinates of the center point of the spectrum. The vertical coordinates of the center point of the spectrum.

[0041] Let the maximum effective radius be The system divides the spectrum into different frequency ranges, such as low-frequency, mid-frequency, and high-frequency regions. The low-frequency region is... This mainly corresponds to changes in the overall shape of the flame; the mid-frequency region is... This mainly corresponds to changes in the main texture of the flame; the high-frequency area is... This mainly corresponds to flickering, localized breakage, and rapid flashing changes at the flame edge. The frequency is less than... The region contains a large amount of overall brightness and background trend information, with a frequency greater than [missing information]. The area is easily affected by sensor noise and dust particles, so it is not considered as the main judgment range in this embodiment.

[0042] Then calculate the energy in the low-frequency region separately. Mid-frequency energy and high frequency energy : ; ; ; in, For frequency domain points The corresponding power spectral density value, The radial distance from a point in the frequency domain to the center point of the spectrum. Low frequency region For the intermediate frequency region, This is the high-frequency region.

[0043] For the i For a single frame image, the flame energy ratio can be represented by the high-frequency energy ratio, and its calculation formula is as follows: ; in, Indicates the first i The flame energy ratio per frame of the image. i To analyze the frame image sequence number, To prevent extremely small constants with a denominator of zero.

[0044] Under normal circumstances, during stable combustion, Fluctuations are slight within the baseline range; when the flame edge shows continuous breaking, jumping, or localized unstable combustion... It will increase. By using the flame energy ratio in a single frame, the system can convert flame edge jitter, local fragmentation, and rapid flickering changes into quantifiable frequency domain parameters.

[0045] Step S14: Based on the single-frame flame energy ratio within the statistical window, determine the window statistical characteristics; the window statistical characteristics include the energy mean, energy standard deviation, and adjacent frame change rate within a preset frequency range.

[0046] In this embodiment, the system uses a 10-second statistical window to calculate the flame energy ratio of all valid frames within the window. If the number of valid frames within the window is less than 60% of the total number of frames sampled, the window is marked as having insufficient image quality, automatic control is not triggered, and only a prompt is output. If the number of valid frames meets the requirements, the system calculates the mean high-frequency energy, energy standard deviation, and rate of change between adjacent frames within the window.

[0047] In this embodiment, the formula for calculating the average high-frequency energy within the window is as follows: ; in, To calculate the average high-frequency energy within the statistical window, To count the number of valid frames within a window.

[0048] The formula for calculating the standard deviation of high-frequency energy within the window is as follows: ; in, This represents the standard deviation of high-frequency energy within the statistical window.

[0049] The formula for calculating the high-frequency change rate between adjacent frames is as follows: ; in, For the first i Frame image relative to the first i -1 frame image high frequency change rate of adjacent frames, For the first i -1 frame image of flame energy ratio per frame.

[0050] Therefore, the average rate of change within the window can be obtained. This is used to indicate whether the flame state changes rapidly within a short period of time.

[0051] Step S15: Using the window statistical features and preset benchmark values, determine the flame frequency domain disturbance index, determine the combustion disturbance level based on the flame frequency domain disturbance index, generate control commands using the combustion disturbance level, and use the control commands to control the combustion of the waste incinerator.

[0052] In this embodiment, when the waste incinerator is operating stably, a preset benchmark value is determined based on the furnace flame image within a preset time period. The preset benchmark value includes a benchmark average energy value, a benchmark energy fluctuation value, and a benchmark adjacent frame change rate. A first ratio between the energy average value within a preset frequency range and the benchmark energy average value is calculated. A second ratio between the energy standard deviation within the preset frequency range and the benchmark energy fluctuation value is calculated. A third ratio between the adjacent frame change rate within the preset frequency range and the benchmark adjacent frame change rate is calculated. The first ratio, the second ratio, and the third ratio are weighted and calculated to obtain the flame frequency domain perturbation index.

[0053] In this embodiment, the flame frequency domain disturbance index is compared with a preset disturbance threshold, and the combustion disturbance level is determined based on the comparison result; a control command is generated according to the combustion disturbance level; the control command includes a first control command, a second control command, or a third control command; wherein, the first control command is a command for adjusting the primary air fan frequency and grate speed; the second control command is a command for adjusting the primary air fan frequency, grate speed, and feeding rhythm; and the third control command is a command for coordinating the adjustment of the primary air fan frequency, grate speed, feeding rhythm, and secondary air frequency.

[0054] In this embodiment, the system selects a stable operating period to establish a baseline value during the commissioning phase. The selection criteria for a stable operating period may include: furnace temperature within a set range, such as 850℃ to 950℃; flue gas oxygen content within the normal on-site control range, such as 5% to 9%; steam flow rate fluctuation within 10 minutes being less than a set value, such as less than 3% of the rated evaporation capacity; and no significant manual adjustments to the feeding, grate, and blower. The system continuously acquires flame images for no less than 30 minutes during this stable period and calculates the baseline value. , and .in, Available stable samples the median of Available stable samples the median of Available stable samples The median. Using the median instead of the simple mean can reduce the impact of individual outliers on the baseline value.

[0055] In this embodiment, the system calculates the flame frequency domain perturbation index of the current window according to the following formula. : ; In the above formula, the weights of 0.55, 0.30 and 0.15 correspond to the high-frequency disturbance intensity, disturbance fluctuation amplitude and disturbance change rate, respectively.

[0056] In this embodiment, the high-frequency disturbance intensity is the primary judgment factor and therefore has the highest weight; the disturbance fluctuation amplitude is used to determine whether the flame is in a continuously unstable state; and the adjacent frame change rate is used to assist in identifying sudden disturbances. These weights can be adjusted within a small range during on-site debugging. For example, the high-frequency disturbance intensity weight can be set to 0.50 to 0.65, the disturbance fluctuation amplitude weight can be set to 0.20 to 0.35, and the adjacent frame change rate weight can be set to 0.10 to 0.20.

[0057] In this embodiment, the system is based on the disturbance index. Combustion disturbance level. For example, when When, it is determined to be a stable state; when When, it is judged as a slight disturbance; when When, it is judged as a moderate disturbance; when If a disturbance occurs, it is considered a relatively strong disturbance. To avoid misjudgment in a single window, the system requires that a mild disturbance triggers Level 1 control only after at least 3 consecutive rolling calculation points, a moderate disturbance triggers Level 2 control only after at least 2 consecutive rolling calculation points, and a strong disturbance triggers Level 3 control and a running prompt after lasting for more than 10 seconds.

[0058] The flame frequency domain disturbance index is used to identify combustion fluctuations in advance, but control actions are still constrained by parameters such as furnace temperature, flue gas oxygen content, and carbon monoxide concentration. For example, when the flue gas oxygen content is below the set lower limit, it indicates a high probability of insufficient oxygen supply, and the system allows for an increase in the primary air fan frequency. When the flue gas oxygen content is already above the set upper limit, the system no longer prioritizes increasing primary air, but instead prioritizes reducing grate speed or delaying material feeding to prevent excessive air from further lowering the furnace temperature. When the furnace temperature approaches the upper limit, the system restricts further increases in air supply or acceleration of combustion. When the carbon monoxide concentration rises abnormally, the system can appropriately increase the secondary air turbulence intensity, but an upper limit must still be set to avoid a rapid drop in furnace temperature.

[0059] When the system determines the combustion disturbance level to be mild, it indicates that the high-frequency flame disturbance has exceeded the stability baseline, but has not yet formed a significant combustion anomaly. At this time, the system implements small-amplitude control: the primary air fan frequency is increased by 2% to 3%, the grate speed is decreased by 1% to 2%, and the feeding mechanism maintains its original rhythm. The system continues to observe for 20 to 30 seconds. If the disturbance index falls below 1.20, the original parameters are gradually restored; if the disturbance index continues to rise, the system enters the secondary control stage. This primary control strategy corresponds to the first control command, which is used to adjust the primary air fan frequency and grate speed.

[0060] When the system determines the combustion disturbance level to be moderate, it indicates that flame edge shaking and texture fragmentation have persisted, possibly corresponding to low-calorific-value waste clumps, wet waste areas, or incomplete combustion in parts of the grate. At this time, the system implements moderate-amplitude control: the primary air fan frequency is increased by 4% to 6%, the grate speed is reduced by 3% to 4%, and the next push action is delayed by 5 to 10 seconds. If the flue gas oxygen content is within the normal range, the secondary air frequency can be increased by 2% to 4% to enhance mixing in the upper part of the furnace. The control objective at this stage is not to forcibly increase combustion intensity, but to prolong the effective combustion time of waste on the grate and reduce the entry of new waste disturbances into the main combustion zone in a short period. This secondary control strategy corresponds to the second control command, which is used to adjust the primary air fan frequency, grate speed, and push rhythm.

[0061] When the system determines that the combustion disturbance level is a relatively strong disturbance, it indicates that the flame frequency domain disturbance index has significantly deviated from the stable baseline, and the disturbance has persisted for a certain period of time. At this time, the system executes limited-amplitude linkage control: the primary air fan frequency is increased by 6% to 8%, the grate speed is reduced by 4% to 6%, the feeding action is delayed by 10 to 15 seconds, and the secondary air frequency is increased by 3% to 5% depending on the oxygen and carbon monoxide levels. Simultaneously, the system outputs a warning message to the operators, indicating possible concentrated entry of wet waste, abnormal waste layer thickness, or incomplete combustion in certain areas of the grate. The three-level control actions must not be continuously and infinitely superimposed; the cumulative increase in primary air fan speed within the same disturbance event should preferably not exceed 10%, and the cumulative decrease in grate speed should preferably not exceed 8%, to avoid over-control. This three-level control strategy corresponds to the third control command, which is used to coordinate and adjust the primary air fan frequency, grate speed, feeding rhythm, and secondary air frequency.

[0062] After the control action is executed, the system continues to update the disturbance index every 2 seconds. When S < 1.20 and remains so for more than 30 seconds, and the furnace temperature and flue gas oxygen content are both within the normal range, the system enters the recovery phase. Recovery does not restore all parameters at once, but rather uses a step-by-step recovery. For example, the primary blower frequency decreases by 1% every 30 seconds, the grate speed increases by 0.5% to 1% every 30 seconds, and the pusher delay is gradually eliminated. If the disturbance index rises again during the recovery process, the recovery is paused and the system re-enters the disturbance level judgment phase. This recovery logic is used to reduce fluctuations in control parameters.

[0063] When the system determines that the image quality is insufficient in three consecutive analysis windows—for example, due to a low effective pixel ratio, image obstruction by smoke or dust, or severe camera window contamination—the system stops image-based automatic control output, retaining only alarm or notification functions, and switches back to the original furnace temperature, oxygen, and steam flow control logic. Frequency domain disturbance analysis is re-enabled only after the image quality recovers. This protection logic prevents erroneous control caused by image quality issues.

[0064] Because the incinerator may experience changes in camera perspective, slight lens contamination, and seasonal variations in waste during long-term operation, the system allows for slow updates to baseline values. However, baseline updates are only performed under stable operating conditions, such as when the disturbance index is below 1.20 for 20 consecutive minutes, furnace temperature, oxygen levels, and steam flow are all within normal ranges, and there is no significant human intervention. Weighted updates can be used. ; ; ; in, This is the updated baseline energy mean. The baseline energy mean before the update. The baseline energy average determined under current stable operating conditions. This is the updated baseline energy fluctuation value. The baseline energy fluctuation value before the update. The baseline energy fluctuation value determined under current stable operating conditions. Updated baseline adjacent frame change rate, The rate of change between adjacent frames before the update. This is the baseline rate of change between adjacent frames determined under the current stable operating conditions.

[0065] By updating slowly, the system can adapt to long-term operational changes in the field, while avoiding the incorrect writing of short-term abnormal operating conditions into the baseline value.

[0066] The innovation of this application lies in transforming flame state judgment from simple brightness, color, or manual observation into a quantifiable frequency domain perturbation index. The high-frequency energy ratio of the flame, the window standard deviation, and the rate of change between adjacent frames all have clearly defined calculation formulas, which can clearly explain how the system obtains the basis for judging combustion fluctuations from flame images. The control logic of this application is more suitable for industrial field use. This invention does not directly and drastically take over the control of the incinerator with image results, but rather adopts a perturbation grading, thermal parameter constraint, incremental adjustment, and step-by-step recovery approach to intervene in the primary air, grate speed, and feeding rhythm with limited amplitude, reducing the risk of excessive control actions. The deployment method of this application is relatively restrained, mainly adding cameras, edge computing modules, and control algorithms. It does not require changes to the main structure of the incinerator, nor does it require reconstruction of the original DCS control system, making it more suitable as an auxiliary optimization module for existing incinerator combustion control systems.

[0067] The overall implementation path of this application's technical solution can be summarized into four consecutive stages. First, a flame image acquisition and effective area determination stage is established, where a high-temperature resistant industrial camera is installed at the furnace observation port, and dustproof, heatproof, and anti-coking structures ensure the camera can acquire effective images for an extended period. Second, a flame frequency domain disturbance calculation stage is established, where the image of the region of interest is processed by grayscale conversion, filtering, normalization, window function processing, and two-dimensional fast Fourier transform, and the power spectrum and single-frame flame energy ratio are calculated. Third, a combustion fluctuation grading judgment stage is established, where a benchmark value is established during the debugging phase, and during formal operation, the current flame frequency domain disturbance index is compared with the benchmark value and the preset disturbance threshold. Fourth, a combustion linkage control stage is established, where the disturbance level is converted into control suggestions or control commands for the primary air fan, grate, pusher mechanism, and secondary air system.

[0068] The core advantage of this solution lies in transforming the "shaking, breaking, and jumping" phenomena in flame conditions, which were originally observed through manual experience, into a calculable frequency domain perturbation index. This index has a clear data source, calculation formula, threshold range, and control output, making it more suitable as an auxiliary judgment variable in industrial control systems. Compared to simply relying on furnace temperature feedback, the triggering basis of this solution is closer to the combustion phenomenon itself. Parameters such as furnace temperature and steam flow rate reflect the combustion result, while changes in flame texture reflect the instantaneous state during the combustion process. By using the flame frequency domain perturbation as a feedforward judgment basis, the system can make small adjustments when the combustion state just begins to deviate, rather than waiting until the temperature, oxygen content, or steam volume deviates significantly before making larger corrections.

[0069] Meanwhile, this application retains the constraining role of traditional thermal parameters. The flame frequency domain perturbation index is responsible for identifying trends in advance, while furnace temperature, flue gas oxygen content, and CO concentration are responsible for limiting the safety boundaries of control actions. For example, when the flue gas oxygen content is already too high, the system will not blindly continue to increase the primary air volume, but will prioritize reducing the grate speed or delaying the feeding; when the furnace temperature is already close to the upper limit, the system will not continue to make adjustments that may push up the furnace temperature. This allows the frequency domain perturbation analysis to complement the existing control system, rather than conflict with it.

[0070] The design focus of this application is to make flame state judgment calculable, calibrable, and reproducible. For example, a reference frequency domain feature is established by acquiring flame images under stable operating conditions. Then, the proportion of high-frequency energy in the current flame image is compared with the reference value. Control actions are triggered only when the deviation reaches a set threshold and persists for a certain period of time. This avoids misjudgment based on a single frame image and also avoids complete reliance on human experience.

[0071] As can be seen, this application provides a method for combustion control of a waste incinerator based on frequency domain perturbation analysis. This method includes acquiring images of the furnace flame of the waste incinerator, extracting regions of interest (ROIs) from the furnace flame images, thus improving the targeting and accuracy of combustion state judgment and effectively reducing interference from factors such as furnace wall reflection and observation window contamination on image analysis. The ROI images are preprocessed to obtain preprocessed images, and a two-dimensional fast Fourier transform is performed on the preprocessed images to obtain a frequency domain matrix. The power spectrum is calculated based on the frequency domain matrix, thus suppressing noise such as smoke particles and instantaneous bright spots, and reducing the impact of camera exposure conditions, observation window contamination, and overall flame intensity variations on the analysis results, ensuring the comparability of image features at different times. Based on the spectral center point and each frequency domain point of the power spectrum, the spectral regions in the power spectrum are divided to obtain spectral regions of different frequency ranges. Different frequencies are then used to... By defining the spectral range within a preset frequency range, the single-frame flame energy ratio is determined, thus eliminating the influence of overall flame brightness fluctuations and achieving precise quantification of the degree of single-frame flame disturbance. Based on the single-frame flame energy ratio within the statistical window, window statistical characteristics are determined, effectively avoiding misjudgments caused by instantaneous flame jumps and single-frame image anomalies, improving the stability and reliability of combustion state recognition. Using window statistical characteristics and preset benchmark values, the flame frequency domain disturbance index is determined, and the combustion disturbance level is determined based on the flame frequency domain disturbance index. The combustion disturbance level is used to generate control commands, and the control commands are used to control the combustion of the waste incinerator. Therefore, it eliminates the reliance on human experience, avoids excessive adjustments that may cause new combustion fluctuations or control oscillations, overcomes the lag of existing thermal parameter feedback adjustments, suppresses problems such as furnace flame shaking, furnace temperature fluctuations, and unstable steam parameters, and improves the combustion stability of the waste incinerator.

[0072] See Figure 2 As shown in the figure, this invention discloses a waste incinerator combustion control device based on frequency domain perturbation analysis, which may specifically include: Image extraction module 11 is used to acquire images of the furnace flames of a waste incinerator and extract regions of interest from the furnace flame images; Preprocessing module 12 is used to preprocess the region of interest image to obtain a preprocessed image, perform a two-dimensional fast Fourier transform on the preprocessed image to obtain a frequency domain matrix, and calculate the power spectrum based on the frequency domain matrix. The energy ratio determination module 13 is used to divide the spectral region in the power spectrum based on the spectral center point and each frequency domain point of the power spectrum to obtain spectral regions of different frequency ranges, and to determine the single-frame flame energy ratio under the preset frequency range using the spectral regions of different frequency ranges. Feature determination module 14 is used to determine window statistical features based on the flame energy ratio of the single frame within the statistical window; the window statistical features include the mean energy, standard deviation of energy, and rate of change between adjacent frames within a preset frequency range; The combustion control module 15 is used to determine the flame frequency domain disturbance index using the window statistical features and preset benchmark values, determine the combustion disturbance level based on the flame frequency domain disturbance index, generate control commands using the combustion disturbance level, and use the control commands to control the combustion of the waste incinerator.

[0073] In some specific embodiments, the image extraction module 11 may specifically include: An industrial camera acquisition module is used to acquire video images of furnace flames using an industrial camera; the industrial camera is installed at the furnace observation port or above the grate of the waste incinerator. The analysis frame image extraction module is used to extract analysis frame images from the furnace flame video image at preset time intervals. The combustion zone image extraction module is used to extract the combustion zone image above the front section of the grate from the analysis frame image; The region of interest image determination module is used to determine the combustion zone image above the front section of the grate as the region of interest image.

[0074] In some specific embodiments, the preprocessing module 12 may specifically include: The effective pixel ratio calculation module is used to calculate the effective pixel ratio in the region of interest image. The normalized image acquisition module is used to perform grayscale processing, median filtering, size scaling processing, and brightness normalization processing on the region of interest image if the effective pixel ratio is greater than a preset pixel threshold, so as to obtain a normalized image. The preprocessed image acquisition module is used to apply a two-dimensional window function to the normalized image to obtain the preprocessed image.

[0075] In some specific embodiments, the energy ratio determination module 13 may specifically include: The radial distance calculation module is used to calculate the radial distance from each frequency domain point in the power spectrum to the center point of the spectrum; The spectrum region division module is used to divide the spectrum region in the power spectrum based on the radial distance and the preset maximum effective radius; The spectrum region determination module is used to determine the target spectrum region and the effective spectrum region within a preset frequency range from spectrum regions of different frequency ranges; The target spectrum energy determination module is used to determine the target spectrum energy within a preset frequency range by utilizing the power spectrum values ​​of each frequency point in the target spectrum region. The total effective spectrum energy determination module is used to determine the total effective spectrum energy using the power spectrum values ​​of each frequency point within the effective spectrum region. The single-frame flame energy ratio calculation module is used to calculate the single-frame flame energy ratio based on the target spectral energy and the total effective spectral energy.

[0076] In some specific embodiments, the combustion control module 15 may specifically include: The preset benchmark value determination module is used to determine preset benchmark values ​​based on furnace flame images within a preset time period when the operation of the waste incinerator is stable; the preset benchmark values ​​include the average benchmark energy, the benchmark energy fluctuation value, and the benchmark adjacent frame change rate; The first ratio calculation module is used to calculate the first ratio between the average energy value within a preset frequency range and the average reference energy value; The second ratio calculation module is used to calculate the second ratio between the energy standard deviation under a preset frequency range and the reference energy fluctuation value. The third ratio calculation module is used to calculate the third ratio between the adjacent frame change rate under the preset frequency range and the reference adjacent frame change rate. The flame frequency domain perturbation index acquisition module is used to perform weighted calculation on the first ratio, the second ratio, and the third ratio to obtain the flame frequency domain perturbation index; The combustion disturbance level determination module is used to compare the flame frequency domain disturbance index with a preset disturbance threshold and determine the combustion disturbance level based on the comparison result. A control command generation module is used to generate control commands based on the combustion disturbance level; the control commands include a first control command, a second control command, or a third control command; wherein, the first control command is a command for adjusting the primary air fan frequency and grate speed; the second control command is a command for adjusting the primary air fan frequency, grate speed, and feeding rhythm; and the third control command is a command for coordinating the adjustment of the primary air fan frequency, grate speed, feeding rhythm, and secondary air frequency.

[0077] Figure 3 This is a schematic diagram of an electronic device provided in an embodiment of this application. The electronic device 20 may specifically include: at least one processor 21, at least one memory 22, a power supply 23, a communication interface 24, an input / output interface 25, and a communication bus 26. The memory 22 stores a computer program, which is loaded and executed by the processor 21 to implement the relevant steps in the frequency domain perturbation analysis-based waste incinerator combustion control method disclosed in any of the foregoing embodiments.

[0078] In this embodiment, the power supply 23 is used to provide operating voltage for each hardware device on the electronic device 20; the communication interface 24 can create a data transmission channel between the electronic device 20 and external devices, and the communication protocol it follows can be any communication protocol applicable to the technical solution of this application, and is not specifically limited here; the input / output interface 25 is used to acquire external input data or output data to the outside world, and its specific interface type can be selected according to specific application needs, and is not specifically limited here.

[0079] In addition, the memory 22, as a carrier for resource storage, can be a read-only memory, random access memory, disk or optical disk, etc. The resources stored on it include operating system 221, computer program 222 and data 223, etc., and the storage method can be temporary storage or permanent storage.

[0080] The operating system 221 manages and controls the various hardware devices on the electronic device 20 and the computer program 222 to enable the processor 21 to perform calculations and processing on the data 223 in the memory 22. It can be Windows, Unix, Linux, etc. The computer program 222, in addition to including a computer program capable of performing the frequency domain perturbation analysis-based waste incinerator combustion control method executed by the electronic device 20 as disclosed in any of the foregoing embodiments, may further include computer programs capable of performing other specific tasks. The data 223 may include data received by the frequency domain perturbation analysis-based waste incinerator combustion control device from external devices, and may also include data collected by its own input / output interface 25.

[0081] The steps of the methods or algorithms described in conjunction with the embodiments disclosed herein can be implemented directly by hardware, a software module executed by a processor, or a combination of both. The software module can be located in random access memory (RAM), main memory, read-only memory (ROM), electrically programmable ROM, electrically erasable programmable ROM, registers, hard disk, removable disk, CD-ROM, or any other form of storage medium known in the art.

[0082] Furthermore, this application also discloses a computer-readable storage medium storing a computer program. When the computer program is loaded and executed by a processor, it implements the steps of the waste incinerator combustion control method based on frequency domain perturbation analysis disclosed in any of the foregoing embodiments.

[0083] Finally, it should be noted that in this document, relational terms such as "first" and "second" are used only 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 one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.

[0084] The present invention provides a detailed description of a method, apparatus, equipment, and medium for combustion control of a waste incinerator based on frequency domain perturbation analysis. Specific examples have been used to illustrate the principles and implementation methods of the present invention. The descriptions of the above embodiments are only for the purpose of helping to understand the method and core ideas of the present invention. At the same time, those skilled in the art will recognize that there will be changes in the specific implementation methods and application scope based on the ideas of the present invention. Therefore, the content of this specification should not be construed as a limitation of the present invention.

Claims

1. A method for combustion control of a waste incinerator based on frequency domain perturbation analysis, characterized in that, include: Acquire images of the furnace flames of a waste incinerator, and extract regions of interest from the furnace flame images; The region of interest image is preprocessed to obtain a preprocessed image. A two-dimensional fast Fourier transform is performed on the preprocessed image to obtain a frequency domain matrix. The power spectrum is calculated based on the frequency domain matrix. Based on the spectral center point and each frequency domain point of the power spectrum, the spectral region in the power spectrum is divided to obtain spectral regions of different frequency ranges. Using the spectral regions of different frequency ranges, the single-frame flame energy ratio under the preset frequency range is determined. Based on the flame energy ratio of a single frame within the statistical window, the statistical characteristics of the window are determined; The window statistical features include the mean energy, standard deviation of energy, and rate of change between adjacent frames within a preset frequency range; Using the window statistical features and preset benchmark values, the flame frequency domain disturbance index is determined, the combustion disturbance level is determined based on the flame frequency domain disturbance index, control commands are generated using the combustion disturbance level, and the combustion control of the waste incinerator is performed using the control commands.

2. The waste incinerator combustion control method based on frequency domain perturbation analysis according to claim 1, characterized in that, The process of acquiring an image of the furnace flame of a waste incinerator and extracting a region of interest image from the furnace flame image includes: An industrial camera is used to capture video images of the furnace flames; the industrial camera is installed at the furnace observation port or above the grate of the waste incinerator. Frame images are extracted and analyzed from the video images of the furnace flame at preset time intervals. Extract the combustion zone image above the front section of the grate from the analysis frame image; The image of the combustion zone above the front section of the grate is taken as the region of interest.

3. The waste incinerator combustion control method based on frequency domain perturbation analysis according to claim 1, characterized in that, The image of the region of interest is preprocessed to obtain a preprocessed image, including: Calculate the proportion of effective pixels in the region of interest image; If the effective pixel ratio is greater than the preset pixel threshold, then the region of interest image is subjected to grayscale processing, median filtering, size scaling processing, and brightness normalization processing to obtain a normalized image. A two-dimensional window function is applied to the normalized image to obtain the preprocessed image.

4. The waste incinerator combustion control method based on frequency domain perturbation analysis according to claim 1, characterized in that, Based on the spectral center point and each frequency domain point of the power spectrum, the spectral regions in the power spectrum are divided, including: Calculate the radial distance from each frequency point in the power spectrum to the center point of the spectrum; The spectral regions in the power spectrum are divided based on the radial distance and the preset maximum effective radius.

5. The waste incinerator combustion control method based on frequency domain perturbation analysis according to claim 1, characterized in that, The method of determining the single-frame flame energy ratio within a preset frequency range using spectral regions of different frequency ranges includes: Determine the target frequency range and the effective frequency range within the preset frequency range from different frequency range frequency ranges; The target spectral energy within a preset frequency range is determined using the power spectral values ​​at each frequency point within the target spectral region. The total effective spectral energy is determined using the power spectral values ​​at each frequency point within the effective spectral region. The single-frame flame energy ratio is calculated based on the target spectral energy and the total effective spectral energy.

6. The waste incinerator combustion control method based on frequency domain perturbation analysis according to claim 1, characterized in that, Using the aforementioned window statistical characteristics and preset benchmark values, the flame frequency domain perturbation index is determined, including: When the waste incinerator is in a stable operating state, a preset reference value is determined based on the furnace flame image within a preset time period; the preset reference value includes the average reference energy, the reference energy fluctuation value, and the reference adjacent frame change rate; Calculate the first ratio between the average energy value within the preset frequency range and the average reference energy value; Calculate the second ratio between the energy standard deviation within the preset frequency range and the reference energy fluctuation value; Calculate the third ratio between the rate of change of adjacent frames within the preset frequency range and the rate of change of adjacent frames at the reference frequency range; The flame frequency domain perturbation index is obtained by weighting the first ratio, the second ratio, and the third ratio.

7. The waste incinerator combustion control method based on frequency domain perturbation analysis according to any one of claims 1 to 6, characterized in that, The combustion disturbance level is determined based on the flame frequency domain disturbance index, control commands are generated using the combustion disturbance level, and combustion control of the waste incinerator is performed using the control commands, including: The flame frequency domain disturbance index is compared with a preset disturbance threshold, and the combustion disturbance level is determined based on the comparison result; Control commands are generated based on the combustion disturbance level; the control commands include a first control command, a second control command, or a third control command; wherein, the first control command is a command for adjusting the primary air fan frequency and grate speed; the second control command is a command for adjusting the primary air fan frequency, grate speed, and feeding rhythm; and the third control command is a command for coordinating the adjustment of the primary air fan frequency, grate speed, feeding rhythm, and secondary air frequency.

8. A waste incinerator combustion control device based on frequency domain perturbation analysis, characterized in that, include: The image extraction module is used to acquire images of the furnace flames of a waste incinerator and extract the region of interest from the furnace flame images. The preprocessing module is used to preprocess the image of the region of interest to obtain a preprocessed image, perform a two-dimensional fast Fourier transform on the preprocessed image to obtain a frequency domain matrix, and calculate the power spectrum based on the frequency domain matrix. The energy ratio determination module is used to divide the spectral region in the power spectrum based on the spectral center point and each frequency domain point of the power spectrum to obtain spectral regions of different frequency ranges, and use the spectral regions of different frequency ranges to determine the single-frame flame energy ratio under a preset frequency range. The feature determination module is used to determine the statistical features of the window based on the single-frame flame energy ratio within the statistical window; The window statistical features include the mean energy, standard deviation of energy, and rate of change between adjacent frames within a preset frequency range; The combustion control module is used to determine the flame frequency domain disturbance index using the window statistical features and preset benchmark values, determine the combustion disturbance level based on the flame frequency domain disturbance index, generate control commands using the combustion disturbance level, and use the control commands to control the combustion of the waste incinerator.

9. An electronic device, characterized in that, include: Memory, used to store computer programs; A processor is configured to execute the computer program to implement the waste incinerator combustion control method based on frequency domain perturbation analysis as described in any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that, Used to store computer programs; wherein, when the computer programs are executed by a processor, they implement the waste incinerator combustion control method based on frequency domain perturbation analysis as described in any one of claims 1 to 7.

Images