Odor pollution monitoring and hierarchical control device and method for garbage disposal plant

Abstract

The invention provides a garbage disposal site odor pollution monitoring and hierarchical control device and method, and relates to the technical field of garbage disposal site odor control. The device comprises a multi-stage pretreatment module used for carrying out stage-by-stage dehumidification and dust removal treatment on to-be-monitored gas to obtain pretreated gas, an integrated monitoring module used for obtaining an odor component concentration electric signal of the pretreated gas, a first data processing module used for converting the odor component concentration electric signal to obtain a chemical concentration monitoring value, and a second data processing module used for sending the chemical concentration monitoring value to a server. The second data processing module is used for converting the chemical concentration monitoring value into an odor concentration monitoring value, the grading module is used for grading the odor concentration monitoring value to obtain different grades of odor pollution, and the deodorization module is used for generating different deodorization implementation schemes according to the odor concentration grades. The problems of low odor monitoring precision and high odor treatment cost in the prior art are solved.

Description

Technical Field

[0001] The invention relates to the technical field of odor control in garbage disposal sites, and in particular to a device and method for monitoring and grading odor pollution in garbage disposal sites. Background Art

[0002] As an important facility for urban waste treatment, the odor emitted by garbage disposal sites has a serious impact on the surrounding environment and residents' health. Existing technologies generally use methods that have not been accurately tested, and directly rely on single or combined technologies such as microbial degradation, activated carbon adsorption or chemical adsorption. This method fails to combine odor control with human perception of odor, and cannot directly respond to odor control plans based on the degree of odor stimulation to human sensory organs and the concentration of each pollutant. As a result, the odor removal efficiency is often low, and due to the lack of targeting and precise control, the economic cost remains high. In addition, the existing online monitoring system is easily disturbed in a high humidity and dusty environment, and the monitoring accuracy is low, which affects the control effect. Therefore, when dealing with odor pollution in garbage dumps, the existing technology can neither ensure the treatment effect nor achieve cost-effectiveness optimization. Summary of the invention

[0003] In order to overcome the deficiencies of the prior art, the present invention provides a device and method for monitoring and grading odor pollution in a garbage disposal site, which solves the problems of low odor monitoring accuracy and high odor control cost in the prior art.

[0004] To achieve the above object, the present invention provides the following solutions:

[0005] A device for monitoring and controlling odor pollution in a garbage disposal site, comprising:

[0006] A power module and a multi-stage pre-processing module, an integrated monitoring module, a first data processing module, a second data processing module, a grading module and a deodorization module, all of which are connected to the power module;

[0007] The power supply module is used to supply power to the multi-stage preprocessing module, the integrated monitoring module, the first data processing module and the second data processing module; the multi-stage preprocessing module is used to perform graded dehumidification and dust removal treatment on the monitored gas to obtain the pretreated gas; the integrated monitoring module is used to obtain the odor component concentration electrical signal of the pretreated gas; the first data processing module is used to convert the odor component concentration electrical signal to obtain a chemical concentration monitoring value; the grading module is used to grade the degree of human olfactory stimulation according to the odor concentration monitoring value to obtain different levels of odor pollution directly related to human olfactory perception; and the deodorization module is used to generate different deodorization implementation plans according to the odor concentration level.

[0008] Preferably, the multi-stage pre-processing module comprises:

[0009] A primary dust removal device, a primary dehumidification device, a final dust removal device and a final dehumidification device connected in sequence;

[0010] The primary dust removal device is used to perform primary dust removal on the gas to be monitored using a filter to obtain the first treated gas, the primary dehumidification device is used to perform primary dehumidification on the first treated gas using a condenser to obtain the second treated gas, the ultimate dust removal device is used to perform ultimate dust removal on the second treated gas using a gas purifier to obtain the third treated gas, and the ultimate dehumidification device is used to perform ultimate drying on the third treated gas using a desiccant to obtain the pre-treated gas.

[0011] Preferably, the integrated monitoring module comprises:

[0012] A hydrogen sulfide sensor, an ammonia sensor, a sulfur dioxide sensor, a trimethylamine sensor, and a volatile organic compound sensor, all of which are connected to the multi-stage pretreatment module;

[0013] The hydrogen sulfide sensor is used to obtain the hydrogen sulfide concentration of the pretreated gas, the ammonia sensor is used to obtain the ammonia concentration of the pretreated gas, the sulfur dioxide sensor is used to obtain the sulfur dioxide concentration of the pretreated gas, the trimethylamine sensor is used to obtain the trimethylamine concentration of the pretreated gas, and the volatile organic compound sensor is used to obtain the volatile organic compound concentration of the pretreated gas.

[0014] Preferably, the first data processing module includes:

[0015] Signal conversion submodule, denoising submodule, normalization submodule, modeling submodule and monitoring submodule;

[0016] The signal conversion submodule is used to perform data conversion on the electrical signal of the odor component concentration of the pretreated gas to obtain a digital signal. The denoising submodule is used to perform noise reduction on the digital signal to obtain a noise-reduced signal. The normalization submodule is used to perform normalization on the noise-reduced signal to obtain a normalized signal. The modeling submodule is used to construct a multivariate linear regression model and calculate the chemical concentration monitoring value based on the normalized signal. The monitoring submodule is used to obtain the chemical concentration monitoring value in real time and store it.

[0017] Preferably, the second data processing module includes:

[0018] A CNN conversion submodule and a threshold management submodule, a dynamic model update submodule and an output submodule, all of which are connected to the CNN conversion submodule;

[0019] The threshold management submodule is used to store the odor thresholds of various pollutants and update the thresholds in the CNN conversion submodule according to the current temperature, humidity and air pressure to obtain updated thresholds. The dynamic model update submodule is used to update the model parameters of the CNN conversion submodule to obtain the optimal CNN model. The CNN conversion submodule is used to transform the chemical concentration monitoring value using the optimal CNN model and the updated threshold to obtain the odor concentration monitoring value. The output submodule is used to output the odor concentration monitoring value to the classification module.

[0020] Preferably, the CNN conversion submodule includes:

[0021] Input build subunits, convolutional layers, activation layers, pooling layers, and fully connected layers;

[0022] The input construction subunit is used to construct a high-dimensional input tensor according to the chemical concentration monitoring value, the update threshold and the current environmental parameters. The convolution layer is used to extract the coupling characteristics of the high-dimensional input tensor. The activation layer applies a nonlinear activation function. The pooling layer is used to reduce the dimension of the coupling characteristics to obtain reduced-dimensional features. The fully connected layer is used to obtain the odor concentration monitoring value according to the reduced-dimensional features.

[0023] Preferably, the denoising submodule comprises:

[0024] Signal division unit, motion estimation unit, filtering unit, recognition unit, multi-level processing unit and feedback unit;

[0025] The signal division unit is used to perform block processing on the digital signal to obtain a plurality of digital block signals, the motion estimation unit is used to obtain motion estimation values ​​of the plurality of digital block signals, the filtering unit is used to perform a first noise reduction processing on the block signal according to a Kalman filtering method to obtain a first processed signal, the identification unit is used to construct a convolutional neural network model to perform a second noise reduction processing on the first processed signal to obtain a second processed signal, the multi-level processing unit is used to identify the frequency band of the second processed signal and perform denoising according to different frequency bands to obtain a denoised signal, and the feedback unit is used to evaluate the denoising effect of the denoised signal and adjust the parameters of the filtering unit, the identification unit, and the multi-level processing unit in real time according to the obtained evaluation results.

[0026] Preferably, the motion estimation unit comprises:

[0027] Data acquisition subunit, matching subunit, calculation subunit;

[0028] The data acquisition subunit is used to acquire image frames of denoised images of the plurality of digital block signals and n frames of images of historical denoised completed images; the matching subunit is used to determine different block sizes and corresponding historical denoised completed images according to an adaptive block size selection mechanism and perform matching calculations to obtain initial motion estimation values; and the calculation subunit is used to weight each initial motion estimation value according to the background of each digital block signal to obtain a final motion estimation value.

[0029] Preferably, the filtering unit comprises:

[0030] A spatial filtering subunit, a temporal filtering subunit and a weighting subunit;

[0031] The time filtering subunit is used to filter and reduce noise on the block signal using the Kalman filtering method in the time domain to obtain a denoised image after Kalman filtering. The spatial filtering subunit is used to perform bilateral filtering and denoise on the block signal using a bilateral filter to obtain a bilaterally filtered image. The weighting subunit is used to obtain a first processed signal based on the final motion estimation value of the block signal, the denoised image after Kalman filtering and the image after bilateral filtering.

[0032] A method for monitoring and hierarchical control of odor pollution in a garbage disposal site, comprising:

[0033] Perform graded dehumidification and dust removal treatment on the monitored gas to obtain pre-treated gas;

[0034] Obtaining an electrical signal of odor component concentration of the pretreated gas;

[0035] Converting the odor component concentration electrical signal to obtain a chemical concentration monitoring value;

[0036] Converting the chemical concentration monitoring value into an odor concentration monitoring value;

[0037] The degree of olfactory stimulation to humans is graded according to the odor concentration monitoring value, and different levels of odor pollution directly related to human olfactory perception are obtained;

[0038] Different deodorization implementation plans are generated according to the odor concentration level.

[0039] The present invention discloses the following technical effects:

[0040] The present invention provides a device and method for monitoring and hierarchical control of odor pollution in a garbage disposal site. The device comprises: a power module and a multi-stage pre-processing module, an integrated monitoring module, a first data processing module, a second data processing module, a grading module and a deodorization module, all of which are connected to the power module; the power module is used to power the multi-stage pre-processing module, the integrated monitoring module, the first data processing module and the second data processing module; the multi-stage pre-processing module is used to perform hierarchical dehumidification and dust removal on the monitored gas to obtain the pre-treated gas; the integrated monitoring module is used to obtain the odor component concentration electrical signal of the pre-treated gas; the first data processing module is used to convert the odor component concentration electrical signal to obtain a chemical concentration monitoring value; the second data processing module is used to convert the chemical concentration monitoring value into an odor concentration monitoring value; the grading module is used to grade the degree of human olfactory stimulation according to the odor concentration monitoring value to obtain different levels of odor pollution directly related to human olfactory perception; the deodorization module is used to generate different deodorization implementation plans according to the odor concentration level. The present invention saves the treatment cost by accurately measuring the odor concentration and performing hierarchical treatment according to the measured concentration. BRIEF DESCRIPTION OF THE DRAWINGS

[0041] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative labor.

[0042] Figure 1 A schematic diagram of the structure of a device for monitoring and grading odor pollution in a garbage disposal site provided in an embodiment of the present invention.

[0043] Description of reference numerals:

[0044] 1-power supply module, 2-multi-stage pre-processing module, 3-integrated monitoring module, 4-first data processing module, 5-classification module, 6-deodorization module, 7-second data processing module. DETAILED DESCRIPTION

[0045] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0046] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the present invention is further described in detail below with reference to the accompanying drawings and specific embodiments.

[0047] like Figure 1 As shown, the present invention provides a device for monitoring and controlling odor pollution in a garbage disposal site, comprising:

[0048] A power module 1 and a multi-stage pre-processing module 2, an integrated monitoring module 3, a first data processing module 4, a second data processing module 7, a grading module 5 and a deodorization module 6 which are all connected to the power module 1; the multi-stage pre-processing module 2, the integrated monitoring module 3, the first data processing module 4, the second data processing module 7, the grading module 5 and the deodorization module 6 are connected in sequence.

[0049] The power supply module 1 is used to supply power to the multi-stage pre-processing module 2, the integrated monitoring module 3, the first data processing module 4 and the second data processing module 7; the multi-stage pre-processing module is used to perform graded dehumidification and dust removal treatment on the monitored gas to obtain the pre-treated gas; the integrated monitoring module 3 is used to obtain the odor component concentration electrical signal of the pre-treated gas; the first data processing module 4 is used to convert the odor component concentration electrical signal to obtain the chemical concentration monitoring value; the second data processing module 7 is used to convert the chemical concentration monitoring value into the odor concentration monitoring value; the grading module 5 is used to grade the degree of human olfactory stimulation according to the odor concentration monitoring value to obtain different levels of odor pollution directly related to human olfactory perception; the deodorization module 6 is used to generate different deodorization implementation plans according to the odor concentration level.

[0050] Specifically, the odor concentration monitoring value can be classified into low concentration state, medium concentration state and high concentration state. Low concentration state (0-1000ouE / m 3 ): The polluted gas is directly introduced into the microbial treatment area, and the odor is treated by microbial adsorption technology. Medium concentration state (1000-2500ouE / m 3 ): The polluted gas first enters the microbial treatment area and then enters the activated carbon adsorption unit. According to the monitored characteristic pollutant concentration, the appropriate activated carbon module is selected for targeted treatment. High concentration state (>2500ouE / m 3 ): The polluted gas is sequentially introduced into the microbial treatment area, activated carbon adsorption unit and absorption tank for graded treatment to ensure that the odor is fully controlled.

[0051] More specifically, adsorbent selection: select activated carbon adsorbent with high efficiency filtration, and use different types of activated carbon (such as acidic, alkaline, and reductive gas treatment special activated carbon) according to the properties of different gas pollutants. Absorption tank design: use aqueous solutions such as sodium hydroxide, sodium hypochlorite and hydrogen peroxide for gas absorption. The absorption tank is equipped with intelligent connection gas path, which can select the corresponding absorption liquid for treatment according to different pollutants.

[0052] Furthermore, the multi-stage pre-processing module 2 includes:

[0053] A primary dust removal device, a primary dehumidification device, a final dust removal device and a final dehumidification device connected in sequence;

[0054] The primary dust removal device is used to perform primary dust removal on the gas to be monitored using a filter to obtain the first treated gas, the primary dehumidification device is used to perform primary dehumidification on the first treated gas using a condenser to obtain the second treated gas, the ultimate dust removal device is used to perform ultimate dust removal on the second treated gas using a gas purifier to obtain the third treated gas, and the ultimate dehumidification device is used to perform ultimate drying on the third treated gas using a desiccant to obtain the pre-treated gas.

[0055] Specifically, primary dehumidification device: usually a condenser or dehumidifier is used to pre-treat the gas to reduce the moisture content in the gas. The condenser achieves the purpose of dehumidification by lowering the gas temperature and condensing water vapor into water. Desiccant: Desiccant (such as silica gel or molecular sieve) can be used in the system for further drying to adsorb residual moisture in the gas to ensure that the final sample is dry. Filter: Before the gas enters the system, an effective filter (such as a HEPA filter or an activated carbon filter) is set to remove solid particles, dust and possible odor substances in the gas. This step can effectively prevent the influence of solid pollutants on sensors and other equipment. Gas purifier: Some models may include a gas purification module, which uses chemical reactions or adsorption technology to further remove pollutants in the gas, such as volatile organic compounds (VOCs) and other gas components. The structure can be modularly configured according to actual needs, flexibly responding to gas samples under different environmental conditions, and ensuring the applicability and effectiveness of the system.

[0056] Specifically, for primary filtration, you can choose a primary filter. The filter material is generally non-woven fabric, nylon mesh, activated carbon filter material, metal mesh, etc.; for secondary filtration, choose a medium-efficiency and high-efficiency filter to remove fine particles. The medium-efficiency filter material is generally synthetic fiber, glass fiber or non-woven fabric; high-efficiency filtration often uses glass fiber, polypropylene or other synthetic materials. Medium and high-efficiency filters can be selected according to actual conditions.

[0057] Furthermore, the integrated monitoring module 3 includes:

[0058] A hydrogen sulfide sensor, an ammonia sensor, a sulfur dioxide sensor, a trimethylamine sensor, and a volatile organic compound sensor, all of which are connected to the multi-stage pretreatment module 2;

[0059] The hydrogen sulfide sensor is used to obtain the hydrogen sulfide concentration of the pretreated gas, the ammonia sensor is used to obtain the ammonia concentration of the pretreated gas, the sulfur dioxide sensor is used to obtain the sulfur dioxide concentration of the pretreated gas, the trimethylamine sensor is used to obtain the trimethylamine concentration of the pretreated gas, and the volatile organic compound sensor is used to obtain the volatile organic compound concentration of the pretreated gas.

[0060] Furthermore, the first data processing module 4 includes:

[0061] Signal conversion submodule, denoising submodule, normalization submodule, modeling submodule and monitoring submodule;

[0062] The signal conversion submodule is used to perform data conversion on the electrical signal of the odor component concentration of the pretreated gas to obtain a digital signal. The denoising submodule is used to perform noise reduction on the digital signal to obtain a noise-reduced signal. The normalization submodule is used to perform normalization on the noise-reduced signal to obtain a normalized signal. The modeling submodule is used to construct a multivariate linear regression model and calculate the chemical concentration monitoring value based on the normalized signal. The monitoring submodule is used to obtain the chemical concentration monitoring value in real time and store it.

[0063] Specifically, an analog-to-digital converter (ADC) is used to convert the analog electrical signal into a digital signal and the filtered signal is normalized to ensure the comparability of data of different gas concentrations.

[0064] Specifically, the expression of the multivariate linear regression model is:

[0065] C = A + aB + d;

[0066] Among them, C is the chemical concentration monitoring value, A is the intercept, a is the regression coefficient, B is the normalized signal, and d is the error term.

[0067] Furthermore, the second data processing module 7 includes:

[0068] A CNN conversion submodule and a threshold management submodule, a dynamic model update submodule and an output submodule, all of which are connected to the CNN conversion submodule;

[0069] The threshold management submodule is used to store the odor thresholds of various pollutants and update the thresholds in the CNN conversion submodule according to the current temperature, humidity and air pressure to obtain updated thresholds. The dynamic model update submodule is used to update the model parameters of the CNN conversion submodule to obtain the optimal CNN model. The CNN conversion submodule is used to transform the chemical concentration monitoring value using the optimal CNN model and the updated threshold to obtain the odor concentration monitoring value. The output submodule is used to output the odor concentration monitoring value to the classification module.

[0070] Specifically, the CNN conversion submodule is used to receive the chemical concentration monitoring values ​​output by the first data processing module, and to construct a high-dimensional input tensor in combination with the odor thresholds of multiple pollutants and optional environmental parameters (such as temperature, humidity, etc.), automatically extract the correlation characteristics and nonlinear relationships between multiple pollutants, and output the odor concentration monitoring values ​​after integrating the chemical concentration monitoring values ​​of various pollutants.

[0071] Preferably, the CNN conversion submodule includes:

[0072] Input build subunits, convolutional layers, activation layers, pooling layers, and fully connected layers;

[0073] The input construction subunit is used to construct a high-dimensional input tensor according to the chemical concentration monitoring value, the update threshold and the current environmental parameters. The convolution layer is used to extract the coupling characteristics of the high-dimensional input tensor. The activation layer applies a nonlinear activation function. The pooling layer is used to reduce the dimension of the coupling characteristics to obtain reduced-dimensional features. The fully connected layer is used to obtain the odor concentration monitoring value according to the reduced-dimensional features.

[0074] Specifically, the CNN conversion submodule includes:

[0075] The input construction subunit is used to construct a high-dimensional input tensor according to the chemical concentration monitoring value, update threshold and current environmental parameters, and perform feature extraction on the chemical concentration data of different pollutants; the convolution layer, when the concentration values ​​of multiple pollutants are input at the same time, scans different channels or time series dimensions through the convolution kernel to learn the coupling effects of various pollutants on the overall odor concentration under different environmental conditions; the activation layer: uses activation functions such as ReLU, Sigmoid or Tanh to realize nonlinear mapping, which can capture the odor increment caused by a jump in the concentration of a single pollutant or the superposition of multiple pollutants; the pooling layer: performs dimensionality reduction operations on the intermediate features output by the convolution layer to reduce redundancy and enhance the robustness of the model, and avoid overfitting of small noise; the fully connected layer: before the final output layer, the features after pooling compression and convolution extraction are combined and regressed through the fully connected network, and the odor concentration monitoring value is output, which can be understood as the predicted value of the dilution multiple of "chemical concentration ÷ odor threshold".

[0076] More specifically, the conversion workflow is as follows: the CNN conversion submodule integrates the chemical concentration values ​​of multiple pollutants, odor thresholds and environmental parameters into high-dimensional inputs, and obtains the odor concentration monitoring values ​​of each or more combined pollutants through convolution operations and full-connection outputs; the threshold management submodule dynamically updates the threshold information required by the CNN conversion submodule when identifying the need to update the threshold database; if the system detects changes in sensor characteristics or large deviations in actual olfactory feedback results, the dynamic model update submodule starts an online or offline update process and iteratively trains model parameters; the output submodule sends the calculated odor concentration monitoring values ​​to the grading module, and can visualize or archive the results for subsequent statistical analysis and deodorization solution optimization.

[0077] The updating process of the threshold is:

[0078] Historical monitoring data: including chemical concentration monitoring values ​​of different pollutants under external environmental parameters such as different seasons, temperature and humidity, wind direction and speed, and predicted odor concentration values ​​for the same period; actual olfactory evaluation data: can come from subjective olfactory tests of on-site personnel or special odor analysis equipment such as electronic noses, to conduct quantitative or semi-quantitative olfactory evaluation of sample gases; regularly (such as at regular intervals or after collecting enough new samples) compare the error between the stored "odor concentration predicted by the CNN conversion submodule" and the "actual olfactory evaluation data". When the error exceeds the preset threshold (such as ±5%) or shows obvious deviations in multiple consecutive samplings, it means that the odor threshold or environmental correction factor may need to be updated.

[0079] Trend analysis of deviations: If the deviations are concentrated in a certain pollutant or a certain time period, it can be determined whether the threshold setting for the pollutant does not match the actual scenario, or whether the threshold is inaccurate due to significant changes in external temperature and humidity.

[0080] Furthermore, the denoising submodule includes:

[0081] Signal division unit, motion estimation unit, filtering unit, recognition unit, multi-level processing unit and feedback unit;

[0082] The signal division unit is used to perform block processing on the digital signal to obtain a plurality of digital block signals, the motion estimation unit is used to obtain motion estimation values ​​of the plurality of digital block signals, the filtering unit is used to perform a first noise reduction processing on the block signal according to a Kalman filtering method to obtain a first processed signal, the identification unit is used to construct a convolutional neural network model to perform a second noise reduction processing on the first processed signal to obtain a second processed signal, the multi-level processing unit is used to identify the frequency band of the second processed signal and perform denoising according to different frequency bands to obtain a denoised signal, and the feedback unit is used to evaluate the denoising effect of the denoised signal and adjust the parameters of the filtering unit, the identification unit, and the multi-level processing unit in real time according to the obtained evaluation results.

[0083] Specifically, the values ​​of these matrices are dynamically adjusted according to the noise level of the real-time signal to optimize the filtering effect. This can be achieved by monitoring the quality of the denoised signal (eg, signal-to-noise ratio SNR).

[0084] More specifically, initial denoising: first use the adaptive filtering method to perform preliminary noise reduction, which can significantly reduce the high-frequency noise in the signal; deep learning post-processing: introduce the adaptively filtered signal into the deep learning model to further optimize the denoising effect, especially for the details retained in the original signal; multi-level analysis and fusion: finally, analyze the different frequency components through multi-level processing to select the optimal denoising method, even for targeted processing of certain specific frequency bands or noise types.

[0085] Further, the motion estimation unit comprises:

[0086] Data acquisition subunit, matching subunit, calculation subunit;

[0087] The data acquisition subunit is used to acquire image frames of denoised images of the plurality of digital block signals and n frames of images of historical denoised completed images; the matching subunit is used to determine different block sizes and corresponding historical denoised completed images according to an adaptive block size selection mechanism and perform matching calculations to obtain initial motion estimation values; and the calculation subunit is used to weight each initial motion estimation value according to the background of each digital block signal to obtain a final motion estimation value.

[0088] Specifically, the pre-filtered image and the saved n-frame images before the current frame that have completed the denoising process are divided into several block images of the same size. An adaptive block size selection mechanism is implemented to dynamically adjust the block size according to the motion situation and image features. The edge strength or the complexity of the motion area can be used to determine the block size, thereby improving the accuracy of motion estimation and performing matching calculations on each block image to obtain n initial motion estimation values ​​of the block image. By weighted calculation of these n initial motion estimation values, a weighted mechanism is used to process the estimation values ​​of different areas. The motion estimation of the static background can be given a lower weight, and the dynamic area can be given a higher weight, ensuring that the final motion estimation value is more consistent with the actual movement situation. The motion estimation value of the block image in the image frame to be processed is obtained. The motion estimation value of each pixel in the image frame to be processed takes the motion estimation value of the block image where the pixel is located.

[0089] Furthermore, the filtering unit includes:

[0090] A spatial filtering subunit, a temporal filtering subunit and a weighting subunit;

[0091] The time filtering subunit is used to filter and reduce noise on the block signal using the Kalman filtering method in the time domain to obtain a denoised image after Kalman filtering. The spatial filtering subunit is used to perform bilateral filtering and denoise on the block signal using a bilateral filter to obtain a bilaterally filtered image. The weighting subunit is used to obtain a first processed signal based on the final motion estimation value of the block signal, the denoised image after Kalman filtering and the image after bilateral filtering.

[0092] This embodiment also provides a method for monitoring and hierarchical control of odor pollution in a garbage disposal site, including:

[0093] Perform graded dehumidification and dust removal treatment on the monitored gas to obtain pre-treated gas;

[0094] Obtaining an electrical signal of odor component concentration of the pretreated gas;

[0095] Converting the odor component concentration electrical signal to obtain a chemical concentration monitoring value;

[0096] Converting the chemical concentration monitoring value into an odor concentration monitoring value;

[0097] The degree of olfactory stimulation to humans is graded according to the odor concentration monitoring value, and different levels of odor pollution directly related to human olfactory perception are obtained;

[0098] Different deodorization implementation plans are generated according to the odor concentration level.

[0099] The various embodiments in this specification are described in a progressive manner, and each embodiment focuses on the differences from other embodiments. The same or similar parts between the various embodiments can be referenced to each other.

[0100] The principles and implementation methods of the present invention are described in this article using specific examples. The description of the above embodiments is only used to help understand the method and core idea of ​​the present invention. At the same time, for those skilled in the art, according to the idea of ​​the present invention, there will be changes in the specific implementation methods and application scope. In summary, the content of this specification should not be understood as limiting the present invention.

Claims

1. A device for monitoring and controlling odor pollution in a garbage disposal site, characterized in that: include: A power module and a multi-stage pre-processing module, an integrated monitoring module, a first data processing module, a second data processing module, a grading module and a deodorization module, all of which are connected to the power module; The power supply module is used to supply power to the multi-stage pre-processing module, the integrated monitoring module, the first data processing module and the second data processing module; the multi-stage pre-processing module is used to perform graded dehumidification and dust removal treatment on the monitored gas to obtain the pre-treated gas; the integrated monitoring module is used to obtain the odor component concentration electrical signal of the pre-treated gas; the first data processing module is used to convert the odor component concentration electrical signal to obtain the chemical concentration monitoring value; the second data processing module is used to convert the chemical concentration monitoring value to obtain the odor concentration monitoring value; the grading module is used to grade the degree of human olfactory stimulation according to the odor concentration monitoring value to obtain different levels of odor pollution directly related to human olfactory perception; the deodorization module is used to generate different deodorization implementation plans according to the odor concentration level.

2. The device for monitoring and controlling odor pollution in a garbage disposal site according to claim 1 is characterized in that: The multi-stage pre-processing module comprises: A primary dust removal device, a primary dehumidification device, a final dust removal device and a final dehumidification device connected in sequence; The primary dust removal device is used to perform primary dust removal on the gas to be monitored using a filter to obtain the first treated gas, the primary dehumidification device is used to perform primary dehumidification on the first treated gas using a condenser to obtain the second treated gas, the ultimate dust removal device is used to perform ultimate dust removal on the second treated gas using a gas purifier to obtain the third treated gas, and the ultimate dehumidification device is used to perform ultimate drying on the third treated gas using a desiccant to obtain the pre-treated gas.

3. The device for monitoring and controlling odor pollution in a garbage disposal site according to claim 1 is characterized in that: The integrated monitoring module comprises: A hydrogen sulfide sensor, an ammonia sensor, a sulfur dioxide sensor, a trimethylamine sensor, and a volatile organic compound sensor, all of which are connected to the multi-stage pretreatment module; The hydrogen sulfide sensor is used to obtain the hydrogen sulfide concentration of the pretreated gas, the ammonia sensor is used to obtain the ammonia concentration of the pretreated gas, the sulfur dioxide sensor is used to obtain the sulfur dioxide concentration of the pretreated gas, the trimethylamine sensor is used to obtain the trimethylamine concentration of the pretreated gas, and the volatile organic compound sensor is used to obtain the volatile organic compound concentration of the pretreated gas.

4. The device for monitoring and controlling odor pollution in a garbage disposal site according to claim 1, characterized in that: The first data processing module comprises: Signal conversion submodule, denoising submodule, normalization submodule, modeling submodule and monitoring submodule; The signal conversion submodule is used to perform data conversion on the electrical signal of the odor component concentration of the pretreated gas to obtain a digital signal. The denoising submodule is used to perform noise reduction on the digital signal to obtain a noise-reduced signal. The normalization submodule is used to perform normalization on the noise-reduced signal to obtain a normalized signal. The modeling submodule is used to construct a multivariate linear regression model and calculate the chemical concentration monitoring value based on the normalized signal. The monitoring submodule is used to obtain the chemical concentration monitoring value in real time and store it.

5. The device for monitoring and controlling odor pollution in a garbage disposal site according to claim 4 is characterized in that: The second data processing module includes: A CNN conversion submodule and a threshold management submodule, a dynamic model update submodule and an output submodule, all of which are connected to the CNN conversion submodule; The threshold management submodule is used to store the odor thresholds of various pollutants and update the thresholds in the CNN conversion submodule according to the current temperature, humidity and air pressure to obtain updated thresholds. The dynamic model update submodule is used to update the model parameters of the CNN conversion submodule to obtain the optimal CNN model. The CNN conversion submodule is used to transform the chemical concentration monitoring value using the optimal CNN model and the updated threshold to obtain the odor concentration monitoring value. The output submodule is used to output the odor concentration monitoring value to the classification module.

6. The device for monitoring and controlling odor pollution in a garbage disposal site according to claim 5, characterized in that: The CNN conversion submodule includes: Input build subunits, convolutional layers, activation layers, pooling layers, and fully connected layers; The input construction subunit is used to construct a high-dimensional input tensor according to the chemical concentration monitoring value, the update threshold and the current environmental parameters. The convolution layer is used to extract the coupling characteristics of the high-dimensional input tensor. The activation layer applies a nonlinear activation function. The pooling layer is used to reduce the dimension of the coupling characteristics to obtain reduced-dimensional features. The fully connected layer is used to obtain the odor concentration monitoring value according to the reduced-dimensional features.

7. The device for monitoring and controlling odor pollution in a garbage disposal site according to claim 4 is characterized in that: The denoising submodule comprises: Signal division unit, motion estimation unit, filtering unit, recognition unit, multi-level processing unit and feedback unit; The signal division unit is used to perform block processing on the digital signal to obtain a plurality of digital block signals, the motion estimation unit is used to obtain motion estimation values ​​of the plurality of digital block signals, the filtering unit is used to perform a first noise reduction processing on the block signal according to a Kalman filtering method to obtain a first processed signal, the identification unit is used to construct a convolutional neural network model to perform a second noise reduction processing on the first processed signal to obtain a second processed signal, the multi-level processing unit is used to identify the frequency band of the second processed signal and perform denoising according to different frequency bands to obtain a denoised signal, and the feedback unit is used to evaluate the denoising effect of the denoised signal and adjust the parameters of the filtering unit, the identification unit, and the multi-level processing unit in real time according to the obtained evaluation results.

8. The device for monitoring and controlling odor pollution in a garbage disposal site according to claim 7, characterized in that: The motion estimation unit comprises: Data acquisition subunit, matching subunit, calculation subunit; The data acquisition subunit is used to acquire image frames of denoised images of the plurality of digital block signals and n frames of images of historical denoised completed images; the matching subunit is used to determine different block sizes and corresponding historical denoised completed images according to an adaptive block size selection mechanism and perform matching calculations to obtain initial motion estimation values; and the calculation subunit is used to weight each initial motion estimation value according to the background of each digital block signal to obtain a final motion estimation value.

9. The device for monitoring and controlling odor pollution in a garbage disposal site according to claim 8, characterized in that: The filtering unit comprises: A spatial filtering subunit, a temporal filtering subunit and a weighting subunit; The time filtering subunit is used to filter and reduce noise on the block signal using the Kalman filtering method in the time domain to obtain a denoised image after Kalman filtering. The spatial filtering subunit is used to perform bilateral filtering and denoise on the block signal using a bilateral filter to obtain a bilaterally filtered image. The weighting subunit is used to obtain a first processed signal based on the final motion estimation value of the block signal, the denoised image after Kalman filtering and the image after bilateral filtering.

10. A method for monitoring and hierarchical control of odor pollution in a garbage disposal site, characterized in that: include: Perform graded dehumidification and dust removal treatment on the monitored gas to obtain pre-treated gas; Obtaining an electrical signal of odor component concentration of the pretreated gas; Converting the odor component concentration electrical signal to obtain a chemical concentration monitoring value; Converting the chemical concentration monitoring value into an odor concentration monitoring value; The degree of olfactory stimulation to humans is graded according to the odor concentration monitoring value, and different levels of odor pollution directly related to human olfactory perception are obtained; Different deodorization implementation plans are generated according to the odor concentration level.

Images