Solid waste treatment process control method and system thereof

Abstract

This application relates to the technical field of solid waste crushing and dust treatment, and discloses a method and system for controlling the solid waste treatment process. The system includes a data acquisition module, a dust prediction module, a first control module, a dust monitoring module, and a second control module. Specifically: the data acquisition module collects data on the crushing state of solid waste; the dust prediction module calculates the expected dust concentration during solid waste crushing; the dust monitoring module collects dust removal status data; the first control module sets dust treatment parameters based on the expected dust concentration and adjusts these parameters in real time based on the dust removal status data; the dust monitoring module also identifies whether the dust treatment status is abnormal; if the dust treatment status is abnormal, the second control module adjusts the solid waste crushing parameters based on feedback from the crushing status data. This application achieves coordinated optimization of the solid waste crushing and dust removal processes, effectively improving dust control efficiency.

Description

Technical Field

[0001] This application relates to the technical field of solid waste crushing and dust treatment, specifically to a solid waste treatment process control method and system. Background Technology

[0002] Solid waste crushing systems mostly rely on downstream dust collection equipment for dust removal. Due to the highly transient and nonlinear fluctuations in dust release, statically set dust collection parameters are insufficient to effectively adapt to dynamically changing dust loads. This unidirectional suction and fixed-frequency cleaning control logic has some shortcomings. For example, during solid waste crushing, the type, volume, hardness, and impact intensity of the material all affect the amount of dust released. For instance, the instantaneous rupture of large-volume materials and high-intensity impact crushing with low moisture content are often accompanied by severe airflow disturbances and dust peaks. Traditional control systems are mostly based on time-driven logic and lack the means to perceive coupled disturbances during the crushing process, making it difficult to accurately predict high dust trends. Traditional systems struggle to detect the current dust release intensity in a timely manner, often resulting in a delayed response or over-suction. When dust collection is delayed, dust accumulates, increasing the load on the filter bags and inducing equipment overload. Currently, most baghouse dust collectors use constant or periodic cleaning. Cleaning is still performed when the dust concentration is not high, and the frequent impact of compressed air on the filter bags shortens the filter bag life. When the dust concentration rises abnormally, if the cleaning frequency is not increased in time, it will cause the filter bags to become clogged and the dust collection to fail.

[0003] For example, Chinese Patent CN117181786B discloses a solid waste treatment system and a solid waste treatment method based on the system. The solid waste treatment system includes a pulverized coal furnace, a flue, a feeding device, and a crushing device. The flue is connected to the pulverized coal furnace to supply high-temperature flue gas from the furnace. The flue includes a descending section and an ascending section to supply the high-temperature flue gas to flow from the descending section to the ascending section. The feeding device is connected to the flue and to the upper end of the descending section to discharge solid waste rods from the upper end of the descending section into the flue for pyrolysis. The crushing device is located between the descending section and the ascending section to crush the unpyrolyzed solid waste rods to generate solid waste dust. The solid waste dust can flow with the high-temperature flue gas in the flue and return to the pulverized coal furnace for incineration. The solid waste treatment system proposed in this scheme has the advantages of simple and reliable structure and high economy, but it still has the problems mentioned in the background technology: the lack of means to sense the coupled disturbances during the crushing process makes it difficult to accurately predict the high dust trend.

[0004] The information disclosed in this background section is intended only to enhance the understanding of the overall background of this application and should not be construed as an admission or in any way implying that the information constitutes prior art known to those skilled in the art. Summary of the Invention

[0005] The technical problem to be solved by this application is to overcome the defects of the prior art and provide a solid waste treatment process control method and system to realize the linkage optimization of solid waste crushing and dust removal processes and effectively improve dust control efficiency.

[0006] To solve the above-mentioned technical problems, this application provides the following technical solution: On one hand, this application provides a solid waste treatment process control system, including a data acquisition module, a dust prediction module, a first control module, a dust monitoring module, and a second control module; wherein: The data acquisition module is used to collect data on the crushing status of solid waste during the crushing process.

[0007] The dust prediction module calculates the expected dust concentration for solid waste crushing treatment based on the crushing status data.

[0008] The first control module sets dust treatment parameters based on the expected dust concentration.

[0009] The dust removal monitoring module is used to collect dust removal status data; the first control module adjusts the dust treatment parameters in real time based on the dust removal status data.

[0010] The dust removal monitoring module is also used to identify whether the dust treatment status is abnormal based on the dust removal status data; if the dust treatment status is abnormal, the second control module adjusts the solid waste crushing parameters based on the crushing status data feedback.

[0011] As a preferred embodiment of the solid waste treatment process control system described in this application, the crushing state data includes spindle torque data, cavity micro-pressure data, and discharge image.

[0012] The dust prediction module includes a feature extraction unit and a prediction model unit; the feature extraction unit is used to extract features from the breakage state data and construct a feature matrix for each breakage state data; the prediction model unit calculates the expected dust concentration based on the feature matrix of each breakage state data.

[0013] The feature extraction unit is configured with a data processing strategy for extracting features from the spindle torque data and constructing a corresponding feature matrix.

[0014] The prediction model unit is equipped with a concentration prediction model for calculating the expected dust concentration. The input of the concentration prediction model includes the spindle torque data, cavity micro-pressure data, and feature matrix of the discharge image. The output is the expected dust concentration for K consecutive future time periods, where K is a positive integer.

[0015] As a preferred embodiment of the solid waste treatment process control system described in this application, the data processing strategy specifically includes: Continuously acquire spindle torque data and organize the time series of spindle torque data.

[0016] Set up a sliding window; initialize the feature matrix of the spindle torque data.

[0017] Based on the sliding window, the spindle torque data time series is slidably truncated; after each sliding truncation, a set of characteristic indicators of the spindle torque data is calculated based on the spindle torque data in the sliding window, and the characteristic indicators are filled into the corresponding columns of the characteristic matrix of the spindle torque data.

[0018] Each row of the feature matrix of the spindle torque data corresponds to the same feature index of the spindle torque data at different times, and the timestamp of each column is the time corresponding to the center position of the sliding window during the corresponding sliding intercept.

[0019] The characteristic indicators of any set of spindle torque data include at least the mean, standard deviation, and peak frequency of the spindle torque data; wherein, the method for extracting the peak frequency is as follows: perform peak detection on the spindle torque data, record the number of peaks and divide by the length of the sliding window to obtain the peak frequency.

[0020] As a preferred embodiment of the solid waste treatment process control system described in this application, the data processing strategy is further used to extract features from the cavity micro-pressure data and construct a corresponding feature matrix; the feature matrix of the cavity micro-pressure data includes feature indicators of the cavity micro-pressure data at different times; the feature indicators of the cavity micro-pressure data include at least the fluctuation amplitude, skewness, and gradient change rate of the cavity micro-pressure data; wherein, the fluctuation amplitude is the difference between the maximum and minimum values ​​of the cavity micro-pressure data.

[0021] The gradient change rate is the maximum value of the gradient of different cavity micro-pressure data; the gradient of any cavity micro-pressure data is the difference between the corresponding cavity micro-pressure data and the adjacent previous cavity micro-pressure data in the time series divided by the sampling interval.

[0022] In a preferred embodiment of the solid waste treatment process control system described in this application, the feature extraction unit is further configured with an image processing strategy for extracting features from the discharge image and constructing a corresponding feature matrix; the image processing strategy specifically includes: Initialize the feature matrix of the discharge image; after each sliding capture of the spindle torque data time series through the sliding window, synchronously acquire the continuous frame discharge images within the time range currently covered by the sliding window, calculate a set of feature indicators of the discharge images based on the continuous frame discharge images, and fill them into the corresponding columns of the feature matrix of the discharge images.

[0023] Any set of feature indicators for the discharge image includes at least the differential change rate and the local gray-level variance; wherein, the method for calculating the local gray-level variance is as follows: divide any frame of the discharge image into different image blocks, calculate the pixel value variance of the pixels in each image block, and calculate the mean to obtain the local pixel variance of the corresponding frame of the discharge image; take the mean of the local pixel variance of each frame of the discharge image as the local gray-level variance.

[0024] As a preferred embodiment of the solid waste treatment process control system described in this application, the method for calculating the differential rate of change is as follows: Construct a difference map of any two consecutive output images; the size of the difference map is the same as that of the two corresponding output images, and the pixel value of any pixel is the absolute value of the difference between the pixel values ​​of two corresponding pixels in the two consecutive output images.

[0025] Set a pixel value threshold and perform binarization processing on each difference image based on the pixel value threshold; those skilled in the art can set the specific value of the pixel value threshold according to actual needs; for any pixel in the difference image, set the pixel value greater than the pixel value threshold to 1, and set the pixel value less than or equal to the pixel value threshold to 0.

[0026] The proportion of pixels with a value of 1 in any difference image is used as the pixel change rate of the corresponding difference image.

[0027] Calculate the mean of the pixel change rates of all difference maps, and use it as the difference change rate.

[0028] As a preferred embodiment of the solid waste treatment process control system described in this application, the first control module includes a dust removal control unit and a dust removal adjustment unit.

[0029] The dust removal control unit is equipped with a dust removal control strategy, which is used to set dust treatment parameters according to the expected dust concentration; the dust treatment parameters include fan speed, damper opening, and cleaning pulse interval.

[0030] The dust control strategy specifically includes: obtaining the expected dust concentration and calculating the desired dust concentration; the desired concentration is the average of the expected dust concentration over the next K time points; setting the fan speed and damper opening based on the desired dust concentration, and both the fan speed and damper opening are positively correlated with the desired concentration.

[0031] Obtain the expected concentration of M consecutive dust particles; M is a positive integer; fit the expected concentration curve based on the expected concentration of the M consecutive dust particles, and extract the slope of the expected concentration curve as a concentration trend index; set a specific value for the dust removal pulse interval based on the concentration trend index, and the dust removal pulse interval is negatively correlated with the concentration trend index.

[0032] The dust removal adjustment unit is configured with a dust removal adjustment strategy for real-time adjustment of dust treatment parameters. The dust removal adjustment strategy specifically includes: setting a fan load threshold and a differential pressure recovery rate threshold; if the differential pressure recovery rate after at least m consecutive pulse cleanings is less than the differential pressure recovery rate threshold and the fan load is greater than the fan load threshold, then the cleaning pulse interval is adjusted to a preset minimum value; m is a positive integer.

[0033] As a preferred embodiment of the solid waste treatment process control system described in this application, the dust removal monitoring module includes a data monitoring unit and a status identification unit; the data monitoring unit is used to collect dust removal status data; the dust removal status data includes fan load and differential pressure recovery rate.

[0034] The data monitoring unit includes differential pressure sensors installed at both ends of each filter bag in the baghouse dust collector. At any given time, the differential pressure sensors collect the pressure difference at both ends of each filter bag and calculate the average value to obtain the average pressure difference at the corresponding time.

[0035] Mark the end time of any pulse cleaning as the target time, and mark the time with a lag time of t relative to the target time as the observation time; calculate the difference between the average pressure difference at the observation time and the average pressure difference at the target time and divide by t to obtain the pressure difference recovery rate after the corresponding pulse cleaning.

[0036] The status recognition unit is configured with a status recognition strategy to identify whether the dust handling status is abnormal; the status recognition strategy specifically includes: If the cleaning pulse interval is a preset minimum value, then after each pulse cleaning, the pressure difference between the two ends of each bag at the target time and the observation time is collected and divided by t to obtain the pressure difference recovery rate of each bag after each pulse cleaning; bags with pressure difference recovery rates less than the pressure difference recovery rate threshold are marked as dust accumulation bags; a dust accumulation ratio threshold is set; if the proportion of dust accumulation bags to all bags is greater than the dust accumulation ratio threshold after at least n consecutive pulse cleanings, then the dust treatment status is abnormal.

[0037] As a preferred embodiment of the solid waste treatment process control system described in this application, the second control module includes a first control unit and a second control unit.

[0038] The first control unit is configured with a crushing adjustment strategy to determine the target parameter to be adjusted in the crushing processing parameters; the second control unit is used to perform feedback adjustment on the target parameter.

[0039] The crushing parameters include spindle speed and feed rate; the crushing adjustment strategy specifically includes: The peak frequency of the spindle torque data over the most recent N moments is extracted from the feature matrix of the spindle torque data and the average value is calculated, which is then used as the first adjustment index.

[0040] The gradient rate of change of the spindle torque data over the most recent N moments is extracted from the feature matrix of the cavity micro-pressure data and the mean value is used as the second adjustment index.

[0041] The first adjustment index and the second adjustment index are weighted and summed to obtain the comprehensive adjustment index.

[0042] Set the adjustment index threshold; if the comprehensive adjustment index is greater than or equal to the adjustment index threshold, the target parameter is the spindle speed; otherwise, the target parameter is the feeding speed.

[0043] Secondly, this application provides a method for controlling a solid waste treatment process, comprising the following steps: Collect data on the crushing state of solid waste during crushing treatment.

[0044] The expected dust concentration for solid waste crushing treatment is calculated based on the crushing status data.

[0045] Dust treatment parameters are set based on the expected dust concentration.

[0046] Collect dust removal status data; adjust dust treatment parameters in real time based on the dust removal status data.

[0047] The dust removal status data is used to identify whether the dust treatment status is abnormal; if the dust treatment status is abnormal, the crushing treatment parameters of solid waste are adjusted based on the crushing status data feedback.

[0048] Compared with the prior art, the beneficial effects achieved by this application are as follows: This application uses multivariate data to predict dust release trends, enabling prediction before dust reaches its peak. This allows for advance setting of key operating parameters for the dust removal system, mitigating the risk of sudden increases in filter bag pressure and dust leakage caused by delayed dust removal response.

[0049] By monitoring dust removal status data in real time, dust treatment parameters can be adaptively optimized, avoiding energy waste and filter bag wear caused by fixed-frequency cleaning and static air volume settings, and realizing refined operation and energy-saving control of the dust removal system under different dust loads.

[0050] When abnormal dust removal efficiency is detected, the crushing and processing parameters are adjusted in reverse to suppress dust generation and enhance the stability and synergy of system operation, while ensuring that the system's processing capacity is not overloaded. Attached Figure Description

[0051] To more clearly illustrate the technical solutions of the embodiments of this application, the drawings used in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort. Wherein: Figure 1 This is a schematic diagram of the solid waste treatment process control system provided in this application.

[0052] Figure 2 A flowchart of the solid waste treatment process control method provided in this application. Detailed Implementation

[0053] The technical solution of this application will be described in detail below with reference to the accompanying drawings and specific embodiments. It should be understood that the embodiments and specific features in the embodiments are detailed descriptions of the technical solution of this application, rather than limitations thereof. In the absence of conflict, the embodiments and technical features in the embodiments can be combined with each other.

[0054] Example 1 This embodiment describes a solid waste treatment process control system, referring to... Figure 1 The system includes a data acquisition module, a dust prediction module, a first control module, a dust monitoring module, and a second control module; among which: The data acquisition module is used to collect data on the crushing status of solid waste during the crushing process.

[0055] The crushing status data includes spindle torque data, cavity micro-pressure data, and discharge image; the data acquisition module includes a first acquisition unit, a second acquisition unit, and a third acquisition unit.

[0056] The first acquisition unit includes a torque sensor installed on the main shaft of the crusher or the motor shaft to acquire main shaft torque data in real time. The main shaft torque data reflects the force changes during solid waste crushing and is a core parameter for determining crushing intensity. When the main shaft torque fluctuates drastically, it usually corresponds to a stage of intense crushing, accompanied by the release of a large amount of dust.

[0057] The second acquisition unit includes a micro-pressure sensor installed inside the crushing chamber for real-time acquisition of chamber micro-pressure data. This micro-pressure data reflects subtle changes in the internal air pressure of the crushing chamber, characterizing the gas disturbance state inside the crushing chamber during the solid waste crushing process. The crushing and fracturing of solid waste is accompanied by airflow disturbances, leading to a transient increase or periodic oscillation of the micro-pressure value.

[0058] The third acquisition unit includes an industrial camera installed in the discharge area to acquire discharge images. These images are visual images of the discharge area, recording visual information about the aerosol concentration at the discharge port. Changes in image brightness and grayscale mean fluctuations reflect the changing trend of dust concentration.

[0059] The dust prediction module calculates the expected dust concentration for solid waste crushing treatment based on the crushing status data.

[0060] The dust prediction module includes a feature extraction unit and a prediction model unit.

[0061] The feature extraction unit is used to extract features from the broken state data and construct a feature matrix for each broken state data; the prediction model unit calculates the expected dust concentration based on the feature matrix of each broken state data.

[0062] The feature extraction unit is configured with a data processing strategy for extracting features from the spindle torque data and constructing a corresponding feature matrix; the data processing strategy specifically includes: Continuously acquire spindle torque data and organize the time series of spindle torque data.

[0063] Set up a sliding window; initialize the feature matrix of the spindle torque data.

[0064] Based on the sliding window, the spindle torque data time series is slidably truncated; after each sliding truncation, a set of characteristic indicators of the spindle torque data is calculated based on the spindle torque data in the sliding window, and the characteristic indicators are filled into the corresponding columns of the characteristic matrix of the spindle torque data.

[0065] Each row of the feature matrix of the spindle torque data corresponds to the same feature index of the spindle torque data at different times, and the timestamp of each column is the time corresponding to the center position of the sliding window during the corresponding sliding intercept.

[0066] The characteristic indicators of any set of spindle torque data include at least the mean, standard deviation, and peak frequency of the spindle torque data; wherein, the method for extracting the peak frequency is as follows: perform peak detection on the spindle torque data, record the number of peaks and divide by the length of the sliding window to obtain the peak frequency.

[0067] The average value of the spindle torque data reflects the load baseline of solid waste crushing, the standard deviation reflects the material hardness fluctuation, and the peak frequency reflects the frequency of impact crushing. These characteristic indicators reflect the intensity changes and fluctuations in the solid waste crushing process, which are particularly important for predicting impact-release dust.

[0068] The data processing strategy is also used to extract features from the cavity micropressure data and construct a corresponding feature matrix; the feature matrix of the cavity micropressure data includes feature indicators of the cavity micropressure data at different times; the feature indicators of the cavity micropressure data include at least the fluctuation amplitude, skewness, and gradient rate of change of the cavity micropressure data; wherein, the fluctuation amplitude is the difference between the maximum and minimum values ​​of the cavity micropressure data; the gradient rate of change is the maximum value of the gradients of different cavity micropressure data; the gradient of any cavity micropressure data is the difference between the corresponding cavity micropressure data and the adjacent previous cavity micropressure data in the time series divided by the sampling interval; the sampling interval is the time difference between the two.

[0069] The fluctuation amplitude of the cavity micro-pressure data reflects the intensity of dust airflow disturbance, the skewness reflects the trend of pressure fluctuation deviating from the center, and the gradient change rate is used to capture local dust concentration abrupt changes; the above characteristic indicators together reflect the intensity of dust release.

[0070] Similar to constructing the feature matrix of the spindle torque data, the time series of the cavity micropressure data is first organized and aligned with the time series of the spindle torque data through interpolation, sampling, and other processing. A sliding window is then used to truncate the cavity micropressure data time series. After each truncation, a set of feature indicators for the cavity micropressure data is calculated based on the cavity micropressure data within the sliding window, and these feature indicators are filled into the corresponding columns of the cavity micropressure data feature matrix. When truncation is performed on the time series of both the spindle torque data and the cavity micropressure data using the sliding window, the length of the sliding window, the sliding step size, the number of sliding steps, and the timestamp corresponding to the first sliding start position are all equal, ensuring that the feature matrices of the spindle torque data and the cavity micropressure data are time-series aligned.

[0071] The feature extraction unit is also configured with an image processing strategy for extracting features from the discharge image and constructing a corresponding feature matrix; the image processing strategy specifically includes: Initialize the feature matrix of the discharge image; after each sliding capture of the spindle torque data time series through the sliding window, synchronously acquire the continuous frame discharge images within the time range currently covered by the sliding window, calculate a set of feature indicators of the discharge images based on the continuous frame discharge images, and fill them into the corresponding columns of the feature matrix of the discharge images.

[0072] Any set of feature indicators for the discharge image includes at least the differential change rate and the local gray-level variance. The method for calculating the local gray-level variance is as follows: Divide any frame of the discharge image into different image blocks, calculate the pixel value variance of each pixel within each image block, and calculate the mean to obtain the local pixel variance of the corresponding frame of the discharge image; take the mean of the local pixel variance of each frame of the discharge image as the local gray-level variance. A large local gray-level variance indicates uneven distribution of dust particles, which often occurs during peak crushing impact periods or the dust release period of large solid waste crushing.

[0073] The method for calculating the differential rate of change is as follows: Construct a difference map of any two consecutive output images; the size of the difference map is the same as that of the two corresponding output images, and the pixel value of any pixel is the absolute value of the difference between the pixel values ​​of two corresponding pixels in the two consecutive output images.

[0074] A pixel value threshold is set, and each difference image is binarized based on the pixel value threshold. Those skilled in the art can set the specific value of the pixel value threshold according to actual needs. In one possible implementation, the pixel values ​​of all pixels in the difference image are first statistically analyzed to obtain the mean pixel value and the standard deviation of the pixel value. The pixel value threshold can be set as the sum of the mean pixel value and three times the standard deviation of the pixel value, which is convenient for detecting significant pixel changes and is suitable for working conditions where impact crushing leads to rapid dust release.

[0075] For any pixel in the difference map, pixel values ​​greater than the pixel value threshold are set to 1, and pixel values ​​less than or equal to the pixel value threshold are set to 0.

[0076] The proportion of pixels with a value of 1 in any difference image is used as the pixel change rate of the corresponding difference image.

[0077] The mean of the pixel change rates of all difference maps is calculated as the difference rate of change. The difference rate of change reflects the mobility of dust particles and is used to determine whether dust particles are in a state of rapid emission. A large difference rate of change indicates active dust release.

[0078] The spatial morphological distribution characteristics of dust particle activity can be extracted from the aforementioned feature indicators of the discharge images, capturing the actual dust discharge effect that is difficult to reflect with sensors.

[0079] The prediction model unit is equipped with a concentration prediction model for calculating the expected dust concentration. The input of the concentration prediction model includes the spindle torque data, cavity micro-pressure data, and feature matrix of the discharge image. The output is the expected dust concentration for K consecutive future time periods, where K is a positive integer.

[0080] Optionally, the concentration prediction model is built on LSTM (Long Short-Term Memory Network). The LSTM captures the temporal patterns in the feature matrix of the fragmented state data and introduces an attention mechanism, enabling the model to dynamically weigh the weights of different dimensions of data, so as to maintain good prediction accuracy under different dust concentration change scenarios, such as adapting to gradual accumulation or instantaneous impact dust concentration changes.

[0081] In one possible implementation, the concentration prediction model includes an input layer, a feature fusion layer, a temporal modeling layer, and an output layer.

[0082] The input layer receives feature matrices from spindle torque data, cavity micro-pressure data, and discharge images, and aligns and concatenates these feature matrices according to timestamps to form a unified multi-dimensional time series input vector. The time step length for each processing step of the input layer is set to 10 to 50, and the dimension of the input vector corresponding to each time step is set to 16 to 32.

[0083] The feature fusion layer is used to perform linear transformation and scaling on the multi-dimensional time series input vector to generate feature vectors. Specifically, a fully connected transformation is used to weight and combine features from different sources, highlighting feature components that are highly correlated with changes in dust concentration. The feature vector dimension is set to 16 to 64.

[0084] The temporal modeling layer is used to extract temporal dependencies in the multidimensional time series. Single-layer or multi-layer LSTM units are used to recursively process the feature vectors of consecutive time steps, enabling the model to remember historical state changes during the fragmentation process, thereby capturing the cumulative effect and sudden change characteristics of dust release. For each time step, the LSTM unit outputs a corresponding hidden state vector, which represents the comprehensive dynamic characteristics at that moment. The hidden state dimension of the LSTM unit is set to 32 to 128.

[0085] The output layer is used to map the hidden state vector to the expected dust concentration at K consecutive future time points. Specifically, the hidden state vector is converted into a prediction vector of length K through a fully connected mapping, where each element of the prediction vector corresponds to the expected dust concentration at a future time point.

[0086] The first control module sets dust treatment parameters based on the expected dust concentration.

[0087] The first control module includes a dust removal control unit and a dust removal adjustment unit.

[0088] The dust removal control unit is equipped with a dust removal control strategy, which is used to set dust treatment parameters according to the expected dust concentration; the dust treatment parameters include fan speed, damper opening, and cleaning pulse interval; The fan speed refers to the motor speed of the induced draft fan; the induced draft fan is used to draw dust generated from solid waste crushing into the bag filter dust collector. The higher the fan speed, the stronger the dust removal effect. The damper opening is the valve opening at the air inlet of the induced draft fan, used to control the air volume entering the dust removal channel. The cleaning pulse interval is the working time interval of the back-blowing pulses used for cleaning in the bag filter dust collector. A shorter cleaning pulse interval results in a higher cleaning frequency, which can improve the self-cleaning ability of the filter bags and cope with high dust loads, but it will also increase energy consumption and filter bag fatigue wear.

[0089] The dust control strategy specifically includes: Obtain the expected dust concentration and calculate the desired dust concentration; the desired concentration is the average of the expected dust concentration over the next K time points; set the fan speed and damper opening based on the desired dust concentration, and both the fan speed and damper opening are positively correlated with the desired concentration.

[0090] Optionally, a desired concentration threshold can be set according to actual needs, and different value levels for fan speed and damper opening can be set. For example, within the normal operating range where the fan load is lower than the fan load threshold and the differential pressure recovery rate is greater than the differential pressure recovery rate threshold, the corresponding desired dust concentration is statistically analyzed, and the average value of the desired concentration within this range is taken as the desired concentration threshold. If the desired dust concentration is greater than the desired concentration threshold, the induced draft fan is turned on, and the corresponding value levels for fan speed and damper opening are selected according to the magnitude of the desired concentration; the higher the desired concentration, the larger the fan speed and damper opening. Driving the start and stop of the dust removal equipment with the desired dust concentration can avoid ineffective ventilation and delayed dust removal, improve dust removal efficiency, and reduce energy consumption and equipment wear.

[0091] Obtain the expected concentration of M consecutive dust particles; M is a positive integer; fit an expected concentration curve based on the expected concentration of the M consecutive dust particles, and extract the slope of the expected concentration curve as a concentration trend index; set a specific value for the dust removal pulse interval based on the concentration trend index, and the dust removal pulse interval is negatively correlated with the concentration trend index.

[0092] Optionally, the range of values ​​for the cleaning pulse interval can be set based on actual needs. If the induced draft fan is turned on, a specific value for the cleaning pulse interval can be selected within the range according to the concentration trend index; the larger the concentration trend index, the smaller the cleaning pulse interval.

[0093] The dust removal monitoring module is used to collect dust removal status data; the first control module adjusts the dust treatment parameters in real time based on the dust removal status data.

[0094] The dust monitoring module includes a data monitoring unit and a status recognition unit.

[0095] The data monitoring unit is used to collect dust removal status data; the dust removal status data includes fan load and differential pressure recovery rate.

[0096] The data monitoring unit includes differential pressure sensors installed at both ends of each filter bag in the baghouse dust collector. At any given time, the differential pressure sensors collect the pressure difference at both ends of each filter bag and calculate the average value to obtain the average pressure difference at the corresponding time.

[0097] Mark the end time of any pulse cleaning as the target time, and mark the time with a lag time of t relative to the target time as the observation time; calculate the difference between the average pressure difference at the observation time and the average pressure difference at the target time and divide by t to obtain the pressure difference recovery rate after the corresponding pulse cleaning.

[0098] Before pulse cleaning, the filter bag surface has a heavy dust accumulation, resulting in high pressure at the inlet and low pressure at the outlet, with a large pressure difference. After cleaning, the filter bag's permeability increases, and the pressure difference decreases. If the pressure difference recovery rate is too low, the cleaning effect will be poor, and problems such as scaling and clogging may occur.

[0099] Optionally, the fan load is the motor current or torque of the induced draft fan, which can be obtained through the frequency converter or motor driver interface; an increase in fan load indicates increased ventilation resistance and severe dust accumulation in the filter bags.

[0100] The dust removal adjustment unit is configured with a dust removal adjustment strategy for real-time adjustment of dust treatment parameters. The dust removal adjustment strategy specifically includes: setting a fan load threshold and a differential pressure recovery rate threshold; if the differential pressure recovery rate after at least m consecutive pulse cleanings is less than the differential pressure recovery rate threshold and the fan load is greater than the fan load threshold, then the cleaning pulse interval is adjusted to a preset minimum value; m is a positive integer.

[0101] This embodiment uses dust removal control and adjustment strategies to adjust the cleaning pulse interval in real time, thereby controlling the frequency and intensity of cleaning to maximize dust removal efficiency and avoid filter bag wear caused by ineffective cleaning. The fan load threshold characterizes the upper limit of the ventilation resistance of the dust removal system. It can be set as a percentage of the rated load, for example, 0.7 to 0.9 times the rated load, to trigger enhanced cleaning control when operating near high load. The differential pressure recovery rate threshold characterizes the lower limit of the filter bag's permeability recovery capability after pulse cleaning. It can be calculated based on the average differential pressure recovery rate from historical operating data, and the threshold can be set to 0.5 to 0.8 times this average value, allowing for timely identification when the cleaning effect significantly decreases.

[0102] The dust removal monitoring module is also used to identify whether the dust treatment status is abnormal based on the dust removal status data; if the dust treatment status is abnormal, the second control module adjusts the solid waste crushing parameters based on the crushing status data feedback.

[0103] The status recognition unit is configured with a status recognition strategy to identify whether the dust handling status is abnormal; the status recognition strategy specifically includes: If the cleaning pulse interval is a preset minimum value, then after each pulse cleaning, the pressure difference between the two ends of each bag at the target time and the observation time is collected and divided by t to obtain the pressure difference recovery rate of each bag after each pulse cleaning; bags with pressure difference recovery rates less than the pressure difference recovery rate threshold are marked as dust accumulation bags; a dust accumulation ratio threshold is set; if the proportion of dust accumulation bags to all bags is greater than the dust accumulation ratio threshold after at least n consecutive pulse cleanings, then the dust treatment status is abnormal.

[0104] The dust accumulation ratio threshold is used to determine whether the proportion of dust-accumulating filter bags in all filter bags has reached a level that affects the overall dust removal efficiency. The ratio can be set based on the structural scale of the bag filter, for example, set to 0.2 to 0.3. When the proportion of dust-accumulating filter bags exceeds this range, it indicates that the filter bag group has shown a significant decrease in permeability, and it needs to be judged as an abnormal dust treatment state.

[0105] The second control module includes a first control unit and a second control unit.

[0106] The first control unit is configured with a crushing adjustment strategy to determine the target parameters to be adjusted in the crushing process parameters; the crushing process parameters include the spindle speed and the feeding speed.

[0107] The spindle speed is the spindle speed of the crusher, which affects the intensity of solid waste crushing and dust release; the feeding speed is the speed at which solid waste is supplied to the crusher, which affects the dust output per unit time.

[0108] The fragmentation adjustment strategy specifically includes: The peak frequency of the spindle torque data over the most recent N moments is extracted from the feature matrix of the spindle torque data and the average value is calculated, which is then used as the first adjustment index.

[0109] The gradient rate of change of the spindle torque data over the most recent N moments is extracted from the feature matrix of the cavity micro-pressure data and the mean value is used as the second adjustment index.

[0110] The first adjustment index and the second adjustment index are weighted and summed to obtain the comprehensive adjustment index.

[0111] When calculating the comprehensive adjustment index, those skilled in the art can set the weights of the first adjustment index and the second adjustment index based on actual needs, and both the first adjustment index and the second adjustment index have been normalized and dimensionless.

[0112] Set the adjustment index threshold; if the comprehensive adjustment index is greater than or equal to the adjustment index threshold, the target parameter is the spindle speed; otherwise, the target parameter is the feeding speed.

[0113] In this embodiment, the first adjustment index reflects the intensity of mechanical impact fluctuations within the crushing chamber, and the second adjustment index reflects the intensity of airflow impacts within the crushing chamber. Together, they reflect the disturbance intensity of dust released during solid waste crushing. When excessive dust concentration leads to abnormal dust handling, this embodiment uses the comprehensive adjustment index to determine the main driving factor for dust rise. When the comprehensive adjustment index is high, the crushing impact and dust release are more intense, and the dominant factor for dust rise is solid waste crushing impact; when the comprehensive adjustment index is low, the excessive dust concentration is caused by dust accumulation and overflow due to the accumulation of solid waste materials. In one implementation, during system operation, multiple sets of crushing state data are collected, and the comprehensive adjustment index at the corresponding time is recorded synchronously, with the dust handling state manually labeled. When the dust handling state is abnormal and accompanied by severe fluctuations in spindle torque and a high peak frequency, it is marked as an impact-dominant condition. When the dust handling state is abnormal but the spindle torque fluctuations are relatively slow and the chamber micro-pressure changes are small but continuous accumulation occurs, it is marked as an accumulation-dominant condition. The value ranges of the comprehensive adjustment index under the two conditions are statistically analyzed, and the boundary between the two ranges is used as the adjustment index threshold. The boundary position is the average of the lower limit of the interval with the larger comprehensive adjustment index and the upper limit of the interval with the smaller comprehensive adjustment index among the two value intervals.

[0114] The second control unit is used to adjust the target parameter based on feedback. The second control unit is equipped with a PLC controller, which is used to adjust the target parameter based on feedback. The second control unit is configured with a target dust concentration. The expected dust concentration is obtained through the concentration prediction model, and the input of the PLC controller is constructed based on the difference between the expected dust concentration and the target dust concentration. The output of the PLC controller is the adjustment amount of the target parameter.

[0115] When the dust removal equipment provides feedback indicating that the dust removal efficiency has not met expectations, meaning the dust handling status is abnormal, this embodiment uses a PLC controller to adjust the working status of the crushing equipment, reducing the intensity of solid waste crushing and preventing excessively high dust concentrations. The target parameters are adjusted by regulating the output of the PLC controller, gradually bringing the expected dust concentration closer to the target dust concentration. This prevents an imbalance between the crushing rhythm and dust removal capacity, which could lead to equipment wear and dust accumulation or overflow.

[0116] Example 2 This embodiment is the second embodiment of this application; it is based on the same inventive concept as Embodiment 1, and refers to... Figure 2 This embodiment introduces a method for controlling a solid waste treatment process, including the following steps: Collect crushing status data during the crushing process of solid waste; the crushing status data includes spindle torque data, cavity micro-pressure data, and discharge images.

[0117] The expected dust concentration for solid waste crushing treatment is calculated based on the crushing state data. First, features are extracted from the crushing state data, and a feature matrix is ​​constructed for each crushing state data. The feature matrix of each crushing state data is processed by a trained concentration prediction model to obtain the expected dust concentration for K consecutive time periods.

[0118] Dust treatment parameters are set based on the expected dust concentration; wherein, the dust treatment parameters include fan speed, damper opening, and dust removal pulse interval.

[0119] Collect dust removal status data; adjust dust treatment parameters in real time based on the dust removal status data; wherein, the dust removal status data includes fan load and differential pressure recovery rate; if the differential pressure recovery rate after multiple consecutive pulse cleanings is less than a preset differential pressure recovery rate threshold and the fan load is greater than a preset fan load threshold, then adjust the cleaning pulse interval to a preset minimum value.

[0120] The dust removal status data is used to identify whether the dust treatment status is abnormal. If the dust treatment status is abnormal, the crushing parameters of solid waste are adjusted based on the crushing status data. If the dust removal pulse interval is adjusted to the preset minimum value, and the proportion of dust accumulation in the filter bag is too high after multiple consecutive pulse dust removals, the dust treatment status is abnormal. A first adjustment index is extracted based on the feature matrix of the spindle torque data, and a second adjustment index is extracted based on the feature matrix of the cavity micro-pressure data. The target parameters to be adjusted are determined by combining the first and second adjustment indicators, and the target parameters are adjusted by the PLC controller to limit the amount of dust released and prevent dust accumulation and overflow.

[0121] The specific functions of each module described above are as described in the relevant content of the solid waste treatment process control system in Example 1, and will not be repeated here.

[0122] Those skilled in the art will understand that embodiments of this application can be provided as methods, systems, or computer program products. Therefore, this application can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, this application can take the form of a computer program product embodied on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0123] The embodiments of this application have been described above with reference to the accompanying drawings. However, this application is not limited to the specific embodiments described above. The specific embodiments described above are merely illustrative and not restrictive. Those skilled in the art can make many other forms under the guidance of this application without departing from the spirit and scope of protection of this application, and these forms are all within the protection scope of this application.

Claims

1. A solid waste treatment process control system, characterized in that: It includes a data acquisition module, a dust prediction module, a first control module, a dust monitoring module, and a second control module; among which: The data acquisition module is used to collect data on the crushing status of solid waste during the crushing process. The crushing status data includes spindle torque data, cavity micro-pressure data, and discharge images; The dust prediction module calculates the expected dust concentration for solid waste crushing treatment based on the crushing status data. The dust prediction module includes a feature extraction unit and a prediction model unit; the feature extraction unit is used to extract features from the breakage state data and construct a feature matrix for each breakage state data; the prediction model unit calculates the expected dust concentration based on the feature matrix of each breakage state data. The feature extraction unit is configured with a data processing strategy for extracting features from the spindle torque data and constructing a corresponding feature matrix. The prediction model unit is equipped with a concentration prediction model for calculating the expected dust concentration. The input of the concentration prediction model includes the feature matrix of the spindle torque data, cavity micro-pressure data, and discharge image, and the output is the expected dust concentration for K consecutive future times, where K is a positive integer. The first control module sets dust treatment parameters based on the expected dust concentration. The dust removal monitoring module is used to collect dust removal status data; the first control module adjusts the dust treatment parameters in real time based on the dust removal status data. The dust removal monitoring module is also used to identify whether the dust treatment status is abnormal based on the dust removal status data; if the dust treatment status is abnormal, the second control module adjusts the solid waste crushing parameters based on the crushing status data feedback.

2. The solid waste treatment process control system as described in claim 1, characterized in that: The data processing strategy specifically includes: Continuously acquire spindle torque data and organize the time series of spindle torque data; Set up the sliding window; initialize the feature matrix of the spindle torque data; Based on the sliding window, a sliding truncation is performed on the time series of spindle torque data; after each sliding truncation, a set of characteristic indicators of spindle torque data is calculated based on the spindle torque data in the sliding window, and the characteristic indicators are filled into the corresponding columns of the characteristic matrix of spindle torque data. Each row of the feature matrix of the spindle torque data corresponds to the same feature index of the spindle torque data at different times, and the timestamp of each column is the time corresponding to the center position of the sliding window during the corresponding sliding intercept. The characteristic indicators of any set of spindle torque data include at least the mean, standard deviation, and peak frequency of the spindle torque data; wherein, the method for extracting the peak frequency is as follows: perform peak detection on the spindle torque data, record the number of peaks and divide by the length of the sliding window to obtain the peak frequency.

3. The solid waste treatment process control system as described in claim 2, characterized in that: The data processing strategy is also used to extract features from the cavity micro-pressure data and construct a corresponding feature matrix; the feature matrix of the cavity micro-pressure data includes feature indicators of the cavity micro-pressure data at different times; the feature indicators of the cavity micro-pressure data include at least the fluctuation amplitude, skewness, and gradient change rate of the cavity micro-pressure data; wherein, the fluctuation amplitude is the difference between the maximum and minimum values ​​of the cavity micro-pressure data; The gradient change rate is the maximum value of the gradient of different cavity micro-pressure data; the gradient of any cavity micro-pressure data is the difference between the corresponding cavity micro-pressure data and the adjacent previous cavity micro-pressure data in the time series divided by the sampling interval.

4. The solid waste treatment process control system as described in claim 3, characterized in that: The feature extraction unit is also configured with an image processing strategy for extracting features from the discharge image and constructing a corresponding feature matrix; the image processing strategy specifically includes: Initialize the feature matrix of the discharge image; after each sliding interception of the spindle torque data time series through the sliding window, synchronously acquire the continuous frame discharge images within the time range currently covered by the sliding window, calculate a set of feature indicators of the discharge images based on the continuous frame discharge images, and fill them into the corresponding columns of the feature matrix of the discharge images; Any set of feature indicators for the discharge image includes at least the differential change rate and the local gray-level variance; wherein, the method for calculating the local gray-level variance is as follows: divide any frame of the discharge image into different image blocks, calculate the pixel value variance of the pixels in each image block, and calculate the mean to obtain the local pixel variance of the corresponding frame of the discharge image; take the mean of the local pixel variance of each frame of the discharge image as the local gray-level variance.

5. The solid waste treatment process control system as described in claim 4, characterized in that: The method for calculating the differential rate of change is as follows: Construct a difference map of any two consecutive output images; the size of the difference map is the same as the corresponding two output images, and the pixel value of any pixel is the absolute value of the difference between the pixel values ​​of two corresponding pixels in the two consecutive output images; Set a pixel value threshold, and perform binarization processing on each difference image based on the pixel value threshold; for any pixel in the difference image, set the pixel value greater than the pixel value threshold to 1, and set the pixel value less than or equal to the pixel value threshold to 0. The proportion of pixels with a value of 1 in any difference image is used as the pixel change rate of the corresponding difference image. Calculate the mean of the pixel change rates of all difference maps, and use it as the difference change rate.

6. The solid waste treatment process control system as described in claim 5, characterized in that: The first control module includes a dust removal control unit and a dust removal adjustment unit; The dust removal control unit is equipped with a dust removal control strategy, which is used to set dust treatment parameters according to the expected dust concentration; the dust treatment parameters include fan speed, damper opening, and cleaning pulse interval; The dust control strategy specifically includes: obtaining the expected dust concentration and calculating the desired dust concentration; the desired dust concentration is the average of the expected dust concentrations over the next K time points; setting the fan speed and damper opening based on the desired dust concentration, and both the fan speed and damper opening are positively correlated with the desired dust concentration; Obtain the expected concentration of M consecutive dust particles; M is a positive integer; fit an expected concentration curve based on the expected concentration of the M consecutive dust particles, and extract the slope of the expected concentration curve as a concentration trend index; set a specific value for the cleaning pulse interval based on the concentration trend index, and the cleaning pulse interval is negatively correlated with the concentration trend index; The dust removal adjustment unit is configured with a dust removal adjustment strategy for real-time adjustment of dust treatment parameters. The dust removal adjustment strategy specifically includes: setting a fan load threshold and a differential pressure recovery rate threshold; if the differential pressure recovery rate after at least m consecutive pulse cleanings is less than the differential pressure recovery rate threshold and the fan load is greater than the fan load threshold, then the cleaning pulse interval is adjusted to a preset minimum value; m is a positive integer.

7. The solid waste treatment process control system as described in claim 6, characterized in that: The dust removal monitoring module includes a data monitoring unit and a status identification unit; the data monitoring unit is used to collect dust removal status data; the dust removal status data includes fan load and differential pressure recovery rate; The data monitoring unit includes differential pressure sensors installed at both ends of each filter bag in the baghouse dust collector. At any given time, the differential pressure sensors collect the pressure difference at both ends of each filter bag and calculate the average value to obtain the average pressure difference at the corresponding time. Mark the end time of any pulse cleaning as the target time, and mark the time with a lag time of t relative to the target time as the observation time; calculate the difference between the average pressure difference at the observation time and the average pressure difference at the target time and divide by t to obtain the pressure difference recovery rate after the corresponding pulse cleaning. The status recognition unit is configured with a status recognition strategy to identify whether the dust handling status is abnormal. The state recognition strategy specifically includes: If the cleaning pulse interval is a preset minimum value, then after each pulse cleaning, the pressure difference between the two ends of each bag at the target time and the observation time is collected and divided by t to obtain the pressure difference recovery rate of each bag after each pulse cleaning; bags with pressure difference recovery rates less than the pressure difference recovery rate threshold are marked as dust accumulation bags; a dust accumulation ratio threshold is set; if the proportion of dust accumulation bags to all bags is greater than the dust accumulation ratio threshold after at least n consecutive pulse cleanings, then the dust treatment status is abnormal.

8. The solid waste treatment process control system as described in claim 7, characterized in that: The second control module includes a first control unit and a second control unit; The first control unit is configured with a crushing adjustment strategy to determine the target parameter to be adjusted in the crushing processing parameters; the second control unit is used to perform feedback adjustment on the target parameter. The crushing parameters include the spindle speed and the feeding speed; The fragmentation adjustment strategy specifically includes: The peak frequency of the spindle torque data over the most recent N moments is extracted from the feature matrix of the spindle torque data and the mean value is calculated, which is then used as the first adjustment index. The gradient rate of change of the spindle torque data over the most recent N moments is extracted from the feature matrix of the cavity micro-pressure data and the mean value is used as the second adjustment index. The first adjustment index and the second adjustment index are weighted and summed to obtain the comprehensive adjustment index; Set the adjustment index threshold; if the comprehensive adjustment index is greater than or equal to the adjustment index threshold, the target parameter is the spindle speed; otherwise, the target parameter is the feeding speed.

9. A solid waste treatment process control method, implemented based on the solid waste treatment process control system according to any one of claims 1-8, characterized in that: Includes the following steps: Collect data on the crushing state during the crushing process of solid waste; The expected dust concentration for solid waste crushing treatment is calculated based on the crushing state data. Dust treatment parameters are set based on the expected dust concentration; Collect dust removal status data; adjust dust treatment parameters in real time based on the dust removal status data; The dust removal status data is used to identify whether the dust treatment status is abnormal; if the dust treatment status is abnormal, the crushing treatment parameters of solid waste are adjusted based on the feedback from the crushing status data.

Images