Environmental monitoring method and system for microbial decomposition of odorous gas

Abstract

The invention provides an environment monitoring method and system for microbial decomposition of peculiar smell gas, and relates to the technical field of automation control, and the method comprises the following steps: obtaining a threshold range of decomposition effect and decomposition efficiency of a microbial decomposition peculiar smell device, and defining the threshold range as a double-constraint target; constructing a quality monitoring database, and performing full-link environment quality feature extraction to obtain a quality monitoring key feature set; building a peculiar smell gas decomposition environment prediction network library, inputting a quality monitoring key feature set, and outputting a predicted quality prediction trend of a microbial decomposition environment; establishing a mapping relationship between the quality prediction trend and the key quality features, calculating the risk contribution degree of the key features on each process node, and generating a first optimization instruction; and optimizing the quality of the microbial decomposition environment based on the first optimization instruction. According to the method, decomposition environment parameters are monitored and optimized, and the technical effects of automation and precision of microbial decomposition environment regulation and control are achieved.

Description

Technical Field

[0001] The present application relates to the field of automated control technology, and in particular to an environmental monitoring method and system for microbial decomposition of odorous gases. Background Art

[0002] The microbial decomposition of odorous gases device is specially used to treat odorous gases. It uses the metabolic reaction of microorganisms to neutralize and decompose specific substances in the odorous gases to achieve the purpose of gas purification. Since microorganisms have the unique metabolic function of converting harmful or odor-producing components in odorous gases into harmless substances, the device provides an effective way to solve the problem of odorous gases and is of great significance in improving air quality and reducing odor pollution.

[0003] At present, the process of microbial decomposition of odorous gases is easily restricted by a variety of environmental factors, such as the number and activity of the microorganisms themselves, as well as decomposition environmental factors such as temperature, humidity, and pH value. Compared with other advanced environmental protection technologies, the degree of intelligence of microbial decomposition of odorous gases is relatively low, and its automatic control system has poor real-time performance and is difficult to accurately measure and control many influencing factors. This makes it difficult to ensure the effectiveness and efficiency of microbial decomposition of odorous gases in actual operation. This imbalance may lead to serious consequences and even make gas emissions fail to meet environmental protection standards. This is an urgent problem that microbial decomposition of odorous gases needs to be solved. Summary of the invention

[0004] The present application provides an environmental monitoring method and system for microbial decomposition of odorous gases, which solves the technical problem that the prior art is unable to cope with the environment of microbial decomposition of odorous gases that is restricted by various factors due to its low level of intelligence, and it is difficult to ensure both the effect and efficiency of microbial decomposition of odorous gases. The method provided in the present application takes the effect and efficiency of microbial decomposition of odorous gases as the goal, monitors and optimizes the decomposition environment parameters in real time, and achieves the technical effect of automated and precise decomposition environment regulation.

[0005] In view of the above problems, the present application provides an environmental monitoring method for microbial decomposition of odorous gases, the method comprising: The threshold range of the decomposition effect and decomposition efficiency of the microbial decomposition odor device is obtained and defined as a double constraint target. The decomposition effect is the direct basis for meeting the emission standards, and the decomposition efficiency is the instantaneous metabolic capacity of microorganisms to decompose odorous gases. The multi-dimensional monitoring data of the microbial decomposition odor device is collected through the environmental monitoring sensor group, including the intake concentration of odorous gas, intake flow rate, liquid flow rate, temperature, humidity, dissolved oxygen concentration, pH value of the microbial reaction unit, and residual concentration of odorous gas, to build a quality monitoring database of the microbial decomposition environment; Extract the quality characteristics of the whole-link microbial decomposition environment according to the quality monitoring database to obtain the key feature set for quality monitoring of the microbial decomposition environment; Build a microbial decomposition environment prediction network library, input the key feature set of quality monitoring, and output the predicted quality prediction trend of the microbial decomposition environment; Traverse the quality prediction trend through the process node-quality monitoring key feature set attribution table, establish a mapping relationship with the key quality features, calculate the risk contribution of the key features on each process node, prioritize the risk contribution, and generate a first optimization instruction, wherein the process node-quality monitoring key feature set attribution table is used to characterize the key features corresponding to each process node; Based on the first optimization instruction, the quality of the microbial decomposition environment is optimized.

[0006] The present application also provides an environmental monitoring system for microbial decomposition of odorous gases, the system comprising: The dual constraint module is used to obtain the threshold range of the decomposition effect and decomposition efficiency of the microbial decomposition odor device, which is defined as a dual constraint target. The decomposition effect is the direct basis for meeting the emission standards, and the decomposition efficiency is the instantaneous metabolic capacity of microorganisms to decompose odorous gases. The quality monitoring database module is used to collect multi-dimensional monitoring data of the microbial decomposition odor device through the environmental monitoring sensor group, including the intake concentration of odorous gas, intake flow rate, liquid flow rate, temperature, humidity, dissolved oxygen concentration, pH value of the microbial reaction unit, and residual concentration of odorous gas, so as to build a quality monitoring database of the microbial decomposition environment; The quality monitoring key feature set module and the quality monitoring multidimensional feature set module are used to extract the quality features of the whole-link microbial decomposition environment according to the quality monitoring database, and obtain the quality monitoring key feature set of the microbial decomposition environment; Environmental quality prediction trend module: The environmental quality prediction trend module is used to build a microbial decomposition environment prediction network library, input the key feature set of quality monitoring, and output the predicted quality prediction trend of the microbial decomposition environment; A first optimization instruction module, which is used to traverse the quality prediction trend through the attribution table of the process node-quality monitoring key feature set, establish a mapping relationship with the key quality features, calculate the risk contribution of the key features on each process node, prioritize the risk contribution, and generate a first optimization instruction, wherein the attribution table of the process node-quality monitoring key feature set is used to characterize the key features corresponding to each process node; The decomposition environment optimization module is used to optimize the quality of the microbial decomposition environment based on the first optimization instruction.

[0007] One or more technical solutions provided in this application have at least the following technical effects or advantages: This application achieves a transition from single-target passive response to dual-target active optimization by setting dual-constraint targets, ensures emission concentration compliance and maximizes decomposition efficiency through real-time monitoring and optimization of microbial decomposition environmental parameters, and achieves a win-win situation of safety and benefits; transforms from experience-driven to data-driven, improves the scientificity and reliability of the system, ensures the optimal operation of the microbial odor decomposition device under different working conditions, improves the accuracy, adaptability and robustness of system control, effectively improves the decomposition efficiency of odorous gases, and ensures the effect of exhausting gases after decomposition. BRIEF DESCRIPTION OF THE DRAWINGS

[0008] Figure 1 A schematic flow chart of an environmental monitoring method for microbial decomposition of odorous gases provided in an embodiment of the present application.

[0009] Figure 2 A schematic diagram of the structure of an environmental monitoring system for microbial decomposition of odorous gases provided in an embodiment of the present application.

[0010] Explanation of the accompanying drawings: dual constraint module 10, quality monitoring database module 20, quality monitoring key feature set module 30, environmental quality prediction trend module 40, first optimization instruction module 50, decomposition environment optimization module 60. DETAILED DESCRIPTION

[0011] The above description is only an overview of the technical solution of the present application. In order to more clearly understand the technical means of the present application, it can be implemented in accordance with the contents of the specification. In order to make the above and other purposes, features and advantages of the present application more obvious and easy to understand, the specific implementation methods of the present application are listed below.

[0012] In order to make the purpose, technical solutions and advantages of the present application clearer, the present application will be further described in detail below in conjunction with the accompanying drawings. The described embodiments should not be regarded as limiting the present application. All other embodiments obtained by ordinary technicians in the field without making creative work are within the scope of protection of this application.

[0013] In the following description, reference is made to "some embodiments", which describe a subset of all possible embodiments, but it is understood that "some embodiments" may be the same subset or different subsets of all possible embodiments, and may be combined with each other without conflict, and the terms "first\second" involved are merely to distinguish similar objects and do not represent a specific ordering of objects. The terms "including" and "having" and any variations are intended to cover non-exclusive inclusions, for example, a process, method, platform, product, or server that includes a series of steps or units is not necessarily limited to those steps or units that are clearly listed, but may include other steps or modules that are not clearly listed or inherent to these processes, methods, products, or devices. Unless otherwise defined, all technical and scientific terms used herein have the same meaning as those generally understood by technicians in the technical field of this application. The terms used herein are for the purpose of describing the embodiments of the present application only.

[0014] The present application embodiment provides an environmental monitoring system for microbial decomposition of odorous gases, such as Figure 1 As shown, the method includes: The threshold range of the decomposition effect and decomposition efficiency of the microbial decomposition odor device is obtained and defined as a double constraint target. The decomposition effect is the direct basis for meeting the emission standards, and the decomposition efficiency is the instantaneous metabolic capacity of microorganisms to decompose odorous gases.

[0015] In one embodiment, the microbial decomposition odor device uses the metabolism of microorganisms in the microbial reaction unit to convert odorous gases carried by substances such as sulfur-containing and nitrogen-containing compounds into harmless or low-toxic gases, thereby effectively removing odors and optimizing air quality. The decomposition effect of odorous gases by microorganisms is the direct basis for meeting emission standards, that is, the residual concentration of odorous gases monitored at the exhaust port of the device is the direct basis for ensuring the final purification capacity of the system and meeting emission standards; the microbial decomposition efficiency (where η is the decomposition efficiency, C in is the intake concentration of odorous gas, C outThe decomposition effect and decomposition efficiency of microbial decomposition of odorous gases directly reflect the instantaneous metabolic capacity of the microbial community and are the core indicators of process performance. The decomposition effect and decomposition efficiency of microbial decomposition of odorous gases transform the complex biochemical reaction process into a two-dimensional control parameter space that can be intuitively operated. The threshold range of the decomposition effect is the parameter space determined according to the emission standards and is an insurmountable hard constraint. The threshold range of the decomposition efficiency is the parameter space determined by long-term monitoring of the microbial decomposition odorous gas device and is a flexible optimization target. The decomposition effect and decomposition efficiency are the constraints of the dynamic adjustment parameters and the optimization direction of the decomposition environment regulation of microbial decomposition of odorous gases, that is, the dual optimization balance of decomposition effect and decomposition efficiency is the optimization direction. By real-time monitoring of the decomposition environment and dynamically adjusting relevant parameters according to the monitoring results, it is promoted to develop in the direction of dual optimization balance of decomposition effect and decomposition efficiency, which can not only ensure that the residual concentration of odorous gases meets the emission standards and achieve good decomposition effects, but also systematically improve the decomposition efficiency, and achieve the dual goals of combining soft and hard decomposition effect and decomposition efficiency.

[0016] Through the environmental monitoring sensor group, multi-dimensional monitoring data of the microbial decomposition odor device is collected, including the intake concentration of odorous gas, intake flow, liquid flow, temperature, humidity, dissolved oxygen concentration, pH value of the microbial reaction unit, and residual concentration of odorous gas, to build a quality monitoring database for the microbial decomposition environment.

[0017] In one embodiment, an environmental monitoring sensor group is arranged on the odor decomposition device to monitor the full-link working environment of the process path in the odor decomposition device in real time to ensure that the decomposition efficiency of the odor gas remains stable and complies with relevant regulations. The quality monitoring database is a database that stores multidimensional monitoring data with time series collected from the same odor decomposition device, which is used to analyze and extract key features of monitoring the full-link working environment. These monitoring data include the concentration of odorous gases at the air inlet, air intake flow, liquid flow, temperature, humidity, dissolved oxygen concentration, pH value of the microbial reaction unit, residual concentration of odorous gases at the air outlet and other key parameters. By obtaining these monitoring data, the full-link working environment of the process path in the odor decomposition device can be understood in real time. For example, the concentration of odorous gases at the air inlet end is the starting point of full-link monitoring and an important indicator for measuring the degree of pollution entering the odor decomposition device. By comparing it with the residual concentration of odorous gases at the air outlet end, the treatment effect of the odor decomposition device can be intuitively evaluated; air intake flow rate, the air intake flow rate directly affects the residence time of odor in the filter bed. Excessive concentration of odor may inhibit the metabolic activity of microorganisms. Monitoring the flow rate can dynamically adjust the load to prevent microorganisms from being inactivated due to overload; liquid flow rate, the water washing unit is often set after the air inlet end to remove particulate matter, volatile organic compounds (VOCs) and other pollutants in the gas to prevent them from directly entering the microbial decomposition unit and causing biofilm blockage or toxicity inhibition. Excessive flow rate will shorten the contact period between microorganisms and pollutants, resulting in insufficient degradation; too small flow rate may cause local anaerobic or nutrient deficiency; the microbial reaction unit is the core treatment part of microbial treatment of odorous gases, including a biological filler layer and a microbial flora. The specific microorganisms (such as nitrifying bacteria, sulfur oxidizing bacteria, etc.) that attach to the surface of the filler are the key points of full-link monitoring. The temperature is one of the key environmental factors that affect the metabolic activity of microorganisms. Sudden fluctuations in temperature may indicate problems with the insulation system of the device or abnormal heat production or heat absorption during the microbial metabolism process. The humidity helps to maintain the activity and normal physiological functions of microbial cells, affects the diffusion speed and mode of odorous gases in the microbial reaction unit, and affects the overall treatment effect. The dissolved oxygen concentration can reflect the respiration state of microorganisms. In the microbial metabolism dominated by aerobic respiration, a stable dissolved oxygen concentration means that the microorganisms breathe normally and can continuously and effectively decompose odorous gases. The pH value, the appropriate pH value helps to maintain the microbial community structure in the microbial reaction unit and can efficiently catalyze metabolic reactions. The residual concentration of odorous gases at the outlet is the terminal of full-link monitoring and a direct indicator for testing the treatment effect of the odor decomposition device. Through the quality monitoring database, multi-dimensional monitoring data can be obtained, which provides a comprehensive data basis for subsequent data analysis and model construction, and improves the accuracy and reliability of subsequent data analysis.

[0018] Furthermore, the present application provides a database for constructing a quality monitoring database of a microbial decomposition environment, including: Perform outlier identification and data cleaning on the intake concentration of odorous gas, intake flow, liquid flow, temperature, humidity, dissolved oxygen concentration, pH value of the microbial reaction unit, and residual concentration of odorous gas to obtain usable multi-dimensional monitoring data; Establish node mapping relationship between multi-dimensional monitoring data and process nodes of microbial decomposition of odorous gas; The node state timing difference of the process of microbial decomposition of odorous gas is obtained, and the multi-dimensional monitoring data are time-correlated based on the node state timing difference to build a quality monitoring database.

[0019] In one embodiment, the residual concentration and decomposition efficiency of odorous gases in the microbial decomposition odor device are affected by multi-parameter coupling. Multi-dimensional key data collection is a prerequisite. The multi-dimensional monitoring data collected on-site is processed at the edge, and the available multi-dimensional monitoring data after outlier identification and data cleaning is transmitted to the intelligent control system. A correspondence is established between the available multi-dimensional monitoring data and each node in the odorous gas decomposition process. For each process node, the monitoring parameters corresponding to the node can be determined, and the causal chain of parameter influence can be clarified. For example, the odor gas flow rate at the air inlet end is related to the residence time of the odor gas in the odor decomposition device, which affects the reaction time between the odor gas and the microbial reaction unit. A mapping relationship is established between the odor gas flow rate at the air inlet end and the air inlet end node-load impact; the temperature, humidity, dissolved oxygen concentration, pH value and microbial ATP concentration of the microbial reaction unit. The microbial reaction unit includes the first reaction zone, the second reaction zone and the third reaction zone. The temperature, humidity, dissolved oxygen concentration, pH value and microbial ATP concentration of each reaction zone directly affect the activity, community structure, metabolic function, etc. of the microorganisms in the reaction zone. The temperature, humidity, dissolved oxygen concentration, pH value and microbial ATP concentration of each reaction zone are respectively mapped with the corresponding reaction zone number-reaction efficiency. By traversing each operating node of the odor decomposition device and determining the monitoring parameters corresponding to each node, multiple node mapping relationships are established. Due to the different operating speeds or response times of different nodes, the state data of different nodes differ in time sequence, that is, the node state timing difference. According to the timing difference of the node status, the available multi-dimensional monitoring data is time-aligned, and the timestamps of different nodes and different monitoring data are unified to ensure the consistency of all data in time. For example, there is a gas residence time lag effect in microbial degradation, and the timing difference needs to be corrected to achieve data synchronization. The change in inlet concentration needs to be 2-3 times the empty bed residence time (EBRT) before it affects the outlet concentration. Therefore, when constructing a local monitoring database, it is necessary to take the time difference of the residence time into account, and arrange the gas flow data collected when the odor gas enters the odor gas decomposition device and the odor gas residual concentration data collected after the residence time in the correct time sequence. Select appropriate methods, such as transfer function model replacement and application of dynamic time warping for time alignment, so that the data in the constructed local monitoring database are coherent and consistent in time. This makes the subsequent analysis of the data more accurate and reliable, avoids data misreading or erroneous analysis caused by time asynchrony, and provides a guarantee for accurately mining useful information in the data.

[0020] The quality characteristics of the whole-link microbial decomposition environment are extracted according to the quality monitoring database to obtain the key feature set for quality monitoring of the microbial decomposition environment.

[0021] Furthermore, a set of key features for quality monitoring is obtained, including: Obtain the intake concentration of odorous gas and the residual concentration of odorous gas in the quality monitoring database, obtain the current decomposition effect and the current decomposition efficiency, evaluate the current decomposition environment of microorganisms through the current decomposition effect and the current decomposition efficiency, and obtain the comprehensive performance and dominant mode of the current decomposition environment of microorganisms. The comprehensive performance includes a high-efficiency zone, a medium-efficiency zone and a low-efficiency zone. The dominant mode includes an efficiency-dominant zone, an effect-dominant zone and a balanced zone. The decomposition effect is the residual concentration of odorous gas, and the decomposition efficiency is the difference between the intake concentration of odorous gas and the residual concentration of odorous gas divided by the intake concentration of odorous gas; According to the comprehensive performance and dominant mode of the current microbial decomposition environment, the expert experience method is used to assign initial weights to the air flow rate, liquid flow rate, temperature, humidity, dissolved oxygen concentration, and pH value of the microbial reaction unit; The air intake flow, liquid flow, temperature, humidity, dissolved oxygen concentration, pH value of the microbial reaction unit in the quality monitoring database and the decomposition effect are analyzed to obtain the correlation coefficient of each monitoring data, and the initial weight is adjusted according to the correlation coefficient to obtain the weight of each monitoring data after adjustment; The air intake flow, liquid flow, temperature, humidity, dissolved oxygen concentration, and pH value of the microbial reaction unit in the quality monitoring database are input into the decomposition effect prediction model to obtain the importance score of each monitoring data to the residual concentration of the odorous gas. The decomposition effect prediction model is constructed based on the random forest model. The optimal number of trees and splitting rules are determined by cross-validation. The model is trained to predict the residual concentration of the odorous gas and the initial importance score of each monitoring data is output synchronously. The initial importance score of each monitoring data output by the random forest model is multiplied by the weight of each monitoring data after adjustment and normalized to generate an importance score. A score threshold is preset, and monitoring data with importance scores greater than the score threshold is extracted as key monitoring data. The key monitoring data features are extracted to obtain a key feature set for quality monitoring.

[0022] In one embodiment, a matrix is ​​constructed based on decomposition efficiency and decomposition effect to obtain a matrix of comprehensive performance level and dominant mode combination, wherein the first vertical direction of the matrix is ​​the comprehensive performance level, including high efficiency area, medium efficiency area and low efficiency area, the second vertical direction is the dominant mode, including efficiency dominant, effect dominant and balanced area, the third vertical direction is the comprehensive performance level threshold interval, the fourth vertical direction is the dominant mode interval boundary, and the fifth vertical direction is the description of the state of the microbial decomposition environment by any combination of comprehensive performance level and dominant mode. According to the current decomposition effect and decomposition efficiency, the comprehensive performance and dominant mode of the current decomposition environment of the microorganism that match the matrix of the combination of comprehensive performance and dominant mode are found, and the current decomposition environment of the microorganism is in the medium efficiency area of ​​the comprehensive performance and the balanced area of ​​the dominant mode. According to the expert experience method, the original weights are assigned to the intake flow, liquid flow, temperature, humidity, dissolved oxygen concentration and pH value of the microbial reaction unit in the quality monitoring database according to the current decomposition environment of the microorganism in the medium efficiency area of ​​the comprehensive performance and the balanced area of ​​the dominant mode. The correlation strength between each monitoring data and the decomposition effect in the quantified quality monitoring database is measured, such as the Pearson correlation coefficient method. The correlation coefficient adjusts the initial weight. The higher the absolute value of the correlation coefficient, the more significant the weight is. The smaller the absolute value of the correlation coefficient, the lower the weight is. The adjustment formula is as follows: adjusted weight = initial weight × (1 + |r|), where r is the correlation coefficient. The subjective weight is corrected by data-driven to avoid expert experience bias. The air intake flow, liquid flow, temperature, humidity, dissolved oxygen concentration, and pH value of the microbial reaction unit are input into the decomposition effect prediction model to obtain the importance score of each monitoring data for predicting the decomposition effect. The importance score calculated by the model can quantify the contribution of each monitoring data in predicting the decomposition effect. The decomposition effect prediction model is constructed, and the monitoring data such as air intake flow, liquid flow, temperature, humidity, dissolved oxygen concentration, and pH value are integrated as the input feature set. These features cover the environmental factors and material flow factors in the operation of the microbial reaction unit, and have potential correlation with the prediction of the residual concentration of odorous gas. The input features are standardized using the Z-score standardization method. Construct a random forest regression model and set the model's hyperparameters, such as the number of trees and the number of features considered at each split. The number of trees affects the model's complexity and generalization ability, while the number of features considered at each split controls the model's randomness and search space when constructing a decision tree. The standardized input feature set and the target variable (odor gas residual concentration) are input into the model for training, and the initial importance score of each feature is calculated by the reduction in the mean square error. For example, when constructing each decision tree, a feature is used to split the node, and the difference in the mean square error before and after the split is calculated. Then, the difference in the mean square error of the feature in all decision trees is averaged to obtain the initial importance score of the feature, which reflects the relative importance of each feature in the model's prediction of the odor gas residual concentration.Multiply the initial importance score output by the model by the adjusted weight item by item, normalize the multiplied result, and obtain the importance score of each monitoring data, as follows:. , Among them, i represents the index of the currently calculated monitoring data. When there are multiple monitoring data such as air flow, liquid flow, temperature, etc., i=1 represents the monitoring data of air flow, i=2 represents the liquid flow, and so on; j represents the summation index, traversing all monitoring data from 1 to n. When calculating the final score i, it is necessary to sum the product of the initial importance score of all monitoring data (from the 1st to the nth) and the adjusted weight. j is the index variable used in this summation process; n represents the total number of monitoring data. The importance score reflects the actual contribution of each monitoring data to the prediction of the residual concentration of odorous gas. It not only takes into account the evaluation of the importance of features within the model, but also combines the external weight adjustment factors. According to the importance score, it can be determined which parameters are most critical to controlling the residual concentration of odorous gas, so that these parameters can be adjusted preferentially to improve the effect of microbial decomposition.

[0023] Furthermore, the current decomposition environment of the microorganisms is evaluated through the current decomposition effect and current decomposition efficiency, and the comprehensive performance and dominant mode of the current decomposition environment of the microorganisms are obtained, including: A sampling sliding window is preset, and the intake concentration of odorous gas and the residual concentration of odorous gas are obtained within the sampling sliding window to obtain the decomposition efficiency and decomposition effect; The decomposition efficiency and decomposition effect are normalized and mapped to the polar coordinate system. The radial distance of the polar coordinate system represents the comprehensive performance strength, and the polar angle represents the balance between the decomposition efficiency and the decomposition effect. The formula is as follows: Normalization: , , t=1,2,3...,T; Among them, η norm is the normalized decomposition efficiency, η (t) is the decomposition efficiency calculated at time t, η min is the minimum value in the parameter space of decomposition efficiency, η max is the maximum value in the parameter space of decomposition efficiency, E norm is the decomposition effect after normalization, It is the decomposition effect index, that is, the reciprocal of the residual concentration of odorous gas. is a very small constant, E min is the minimum value in the parameter space of the decomposition effect, E max is the maximum value in the parameter space of the decomposition effect; The polar coordinate conversion formula is as follows: , , t=1,2,3...,T; Among them, r(t) is the radial distance at time t, η norm is the normalized decomposition efficiency, E norm is the decomposition effect after normalization, θ(t) is the polar angle at time t, θ∈[-π,π], t=1,2,3...,T, t is each sampling moment in the sampling sliding window, and all sampling moments are sampling sliding window data; The density distribution of radial distance is analyzed by histogram method to determine the threshold range of comprehensive performance level, and the comprehensive performance level includes high efficiency, medium efficiency and low efficiency; The polar angle is divided into N sectors through polar angle coordinate transformation and clustering method to determine the interval boundary of the dominant mode. The multi-dimensional dominant mode includes decomposition efficiency dominant area, effect dominant area and equilibrium area. Cartesian combination of the level threshold interval of the comprehensive performance and the interval boundary of the dominant mode is performed to generate a matrix of the combination of the comprehensive performance and the dominant mode; The radial distance and polar angle corresponding to the current decomposition effect and the current decomposition efficiency are calculated, and the matrix of the combination of comprehensive performance and dominant mode is traversed to obtain the comprehensive performance and dominant mode of the current decomposition environment of microorganisms.

[0024] In one embodiment, during the microbial decomposition process, there may be a nonlinear correlation between the decomposition efficiency and the decomposition effect: 1) When the intake concentration of the odorous gas is fixed: the decomposition efficiency and the decomposition effect are strongly negatively correlated, and the data are distributed along the diagonal in the rectangular coordinate system. It is difficult to effectively divide the quadrants using traditional methods; 2) When the intake concentration of the odorous gas is changed, the two variables may present a complex relationship, increasing the difficulty of analysis. In order to solve the problem caused by dimensional correlation, the polar coordinate transformation is used to decouple the correlation, and the formula is as follows: Normalization: , , t=1,2,3...,T; Among them, η norm is the normalized decomposition efficiency, η (t) is the decomposition efficiency calculated at time t, η min is the minimum value in the parameter space of decomposition efficiency, η max is the maximum value in the parameter space of decomposition efficiency, E norm is the decomposition effect after normalization, is the decomposition effect index at time t, that is, the reciprocal of the residual concentration of odorous gas. is a very small constant, E min is the minimum value in the parameter space of the decomposition effect, E max is the maximum value in the parameter space of the decomposition effect; The polar coordinate conversion formula is as follows: , , t=1,2,3...,T; Among them, r(t) is the radial distance at time t, η norm is the normalized decomposition efficiency, E norm is the decomposition effect after normalization, θ(t) is the polar angle at time t, θ∈[-π,π], t=1,2,3...,T, t is each sampling moment in the sampling sliding window, and all sampling moments are sampling sliding window data. The sampling sliding window is preset, and the window size of the sliding window = the decomposition period of the microbial deodorization device divided by the sampling interval. The sampling interval is based on the high-frequency step size, and the step size is 1 / 3 of the window size. The normalization process uniformly scales the decomposition efficiency and decomposition results to the [0,1] interval, which is convenient for multi-dimensional joint analysis. The extreme values ​​of the parameter space make the model adapt to data fluctuations under different working conditions. Avoid C out =0, and enhance the robustness of the algorithm. The radial distance r(t) comprehensively characterizes the performance, weakens the dominance of a single dimension, and is a geometric synthesis of the normalized decomposition efficiency and the normalized decomposition effect. The larger the r(t) value, the higher the decomposition efficiency and the lower the residual concentration, that is, the overall performance is excellent. The smaller the r(t) value, the lower the decomposition efficiency or the higher the residual concentration, and the process needs to be adjusted. The polar angle θ(t) captures the proportional relationship between efficiency and effect. Different θ values ​​correspond to different efficiency-effect dominant modes. A larger θ indicates that the effect is prioritized. For example, a larger angle: the decomposition effect is dominant (the residual concentration is extremely low, but the efficiency may not be optimal), a smaller angle: the decomposition efficiency is dominant (the efficiency is high, but the residual concentration may be high), and a moderate angle: the decomposition efficiency and decomposition effect are balanced. The density distribution of the radial distance is analyzed by the histogram method or the kernel density estimation method, and the threshold boundaries of the high-efficiency, medium-efficiency, and low-efficiency areas are set, such as 0.8≤r≤1.2 in the high-efficiency area, 0.5≤r<0.8 in the medium-efficiency area, and r<0.5 in the low-efficiency area. The number of clusters N is preset, N=3 corresponds to efficiency-dominated, effect-dominated and balanced areas, and the corresponding relationship with the polar angle is established. A large angle corresponds to decomposition effect-dominated, a moderate angle corresponds to the balanced area, and a small angle corresponds to decomposition efficiency-dominated. The range of the polar angle is mapped to the two-dimensional coordinates on the unit circle, x=cos(θ) and y=sin(θ) to achieve the conversion, and the annular data of the polar angle is converted to plane coordinates. The converted plane coordinates (x, y) are used to calculate the center point (x) of each cluster according to the preset number of clusters. c ,y c ), calculate its polar angle , and then adjust the calculated polar angle to the interval [-π, π]. Sort the cluster center angles and take the median of the adjacent center angles as the boundary. For example, if there are 3 clusters, the cluster center angles are -30°, 45°, and 120° respectively. After sorting, they are still -30°, 45°, and 120°. The boundary medians are calculated as (-30°+45°) / 2=7.5° and (45°+120°) / 2=82.5°. The final fan-shaped areas are [-180°, 7.5°), [7.5°, 82.5°), and [82.5°, 180°]. Cartesian product combination is used to cross-combine the comprehensive performance level (high efficiency, medium efficiency, low efficiency) with the dominant mode (efficiency-dominated, effect-dominated, balanced) to form a comprehensive performance level-dominant mode matrix. The example matrix is:

[0025] Build a microbial decomposition environment prediction network library, input the key feature set of quality monitoring, and output the predicted quality prediction trend of the microbial decomposition environment.

[0026] In one embodiment, a model is established to predict the quality prediction trend of the microbial decomposition environment, and the key feature set of quality monitoring is input into the prediction network library of the working environment of the decomposition of odorous gas to obtain the prediction trend of the working environment of the microbial decomposition of odorous gas. The prediction provides a quantitative description of the future state, and the judgment is based on the preset rules and real-time target weights, and the prediction value is converted into a specific action instruction to solve the problem of "what to do after the prediction".

[0027] Furthermore, a network library for decomposing odorous gas environment prediction is built, including: Mining and obtaining the decomposition odor gas monitoring database, extracting the key feature data of the whole link monitoring; The key characteristic data of the whole-link monitoring are classified and identified according to the type of odorous gas to obtain the classification data set of decomposed odorous gas monitoring; Feature extraction is performed on the decomposition odor gas monitoring classification data set to obtain the decomposition odor gas monitoring classification feature set; The decomposed odor gas monitoring classification feature set is trained for working environment quality trend prediction respectively to obtain a multi-dimensional decomposed environment trend prediction branch network set; The multi-dimensional decomposition environmental trend prediction branch network set is classified and connected in series according to the odor gas type to generate multiple odor gas type decomposition environmental prediction network sets; The environmental prediction network sets of multiple odorous gas types are identified and integrated, and a network library for decomposing the working environment prediction of odorous gases is built.

[0028] In one embodiment, the full-link monitoring data is mined and key features are extracted. The key features may include temperature, pH, dissolved oxygen, etc. This is consistent with the previous acquisition. Odor gases may have different components, such as H2S, NH3, VOCs, etc. The decomposition mechanisms and required environmental parameters of different gases are different, so the classification process can optimize the model in a targeted manner, and feature extraction is performed on each classified data set to capture the dynamic changes in the decomposition process of different gases, and train the environmental trend prediction branch network. The branch network here should be a prediction model for a specific gas type, such as predicting the trend of environmental parameters in the future, such as changes in temperature, pH, etc. These trends may affect the decomposition efficiency and compliance. Each branch network focuses on one type of gas, which may improve the prediction accuracy. The neural network model is trained separately through the classification features of each odor gas and the corresponding data trends until the model training is completed to obtain a trend prediction branch network set. The multidimensional environmental trend prediction branch network set is used to predict future environmental parameter trends based on known environmental data. The multi-dimensional environmental trend prediction branch network set is classified and connected in series according to the odor gas type. That is, according to the environmental characteristics corresponding to the odor gas type, multiple multi-dimensional environmental trend prediction branches are connected in series to generate a prediction network set for the working environment of microbial decomposition of odor gas. The working environment prediction network set is identified and integrated, that is, the corresponding odor gas category is identified for each odor gas type environmental prediction network, and the branch networks are connected in series by type. The series connection refers to integrating the prediction networks of different gas types to form a comprehensive prediction system that can handle multiple gas situations at the same time, integrate the network set, and build a prediction network library. The final network library can flexibly call the prediction models of different gases to adapt to complex application scenarios.

[0029] The quality prediction trend is traversed through the attribution table of the process node-quality monitoring key feature set, a mapping relationship with the key quality features is established, the risk contribution of the key features on each process node is calculated, the risk contribution is prioritized, and the first optimization instruction is generated, wherein the attribution table of the process node-quality monitoring key feature set is used to characterize the key features corresponding to each process node.

[0030] In one embodiment, the quality prediction trend is traversed through the attribution table of the process node-quality monitoring key feature set and a mapping relationship is established, which helps to deeply understand the intrinsic relationship between the quality prediction trend and the key quality features in each process node. If the quality prediction trend indicates that the product quality may decline, the key quality features in which process nodes may be the cause of this trend can be determined through this mapping relationship. This mapping relationship can accurately locate the process link where the key factors that may affect the quality are located. Calculating the risk contribution of the key features on each process node is to quantify the degree to which each key feature affects the quality. Different key features may have different degrees of influence on the final quality results. By calculating the risk contribution, the contribution of each key feature to the quality risk in the entire process can be clarified. It helps to reasonably allocate resources in the process of quality control and improvement. If the risk contribution of a key feature is very high, then in the case of limited resources, the key feature can be improved or controlled first to improve the overall quality. Prioritizing the risk contribution can determine the order of improvement of the key features of each process node. When faced with multiple key features that need to be improved, processing them in order of priority can more efficiently improve the overall quality. Generating the first optimization instruction provides a specific direction and operational guidance for actual process improvement. This instruction can clarify how the automated control system should adjust the key features of which process nodes.

[0031] Furthermore, a table of process node-quality monitoring key feature sets is established to characterize the key features corresponding to each process node, including: Acquire available multi-dimensional monitoring data and establish a mapping relationship between process nodes for microbial decomposition of odorous gases, and establish an attribution table between process nodes and available multi-dimensional monitoring data; Obtain key monitoring data and quality monitoring key feature sets, extract the attribution table of process nodes and key monitoring data from the attribution table of process nodes and available multidimensional monitoring data based on the key monitoring data, and establish the attribution table of process nodes-quality monitoring key feature sets based on the correspondence between key monitoring data and quality monitoring key feature sets.

[0032] In one embodiment, in the process of microbial decomposition of odorous gas, multidimensional monitoring data reflects the state of process nodes. By establishing this mapping relationship, the intrinsic connection between process nodes and monitoring data can be better understood. Establishing an attribution table of process nodes and available multidimensional monitoring data can systematically organize and record these relationships. This helps to quickly query and determine the monitoring data range corresponding to a specific process node in subsequent analysis and operation, and provides a data structure basis for a comprehensive understanding of the process of microbial decomposition of odorous gas. Obtaining key monitoring data and quality monitoring key feature sets, and then extracting the attribution table of process nodes and key monitoring data from the attribution table of existing process nodes and multidimensional monitoring data based on the key monitoring data, is based on the previously established relationship to perform data screening and extraction. Establishing an attribution table of process node-quality monitoring key feature sets based on the corresponding relationship between key monitoring data and quality monitoring key feature sets is to further integrate data relationships based on the aforementioned steps. This attribution table can characterize the key features corresponding to each process node, and provides an important data basis for in-depth analysis of the optimization of the decomposition environment during the process of microbial decomposition of odorous gas.

[0033] Further, generating a first optimization instruction includes: Decompose the quality prediction trend of the microbial decomposition environment into multi-dimensional independent trend factors, traverse the attribution table of the process node-quality monitoring key feature set, and establish a mapping relationship with the key features; The classification standard is preset to discretize the trend of multi-dimensional independent trend factors and generate the intensity vector of multi-dimensional trend factors; Pre-build discrete contribution function to calculate the discrete contribution of each key feature; Preset an optimization classification scale, classify the contribution of each key feature according to the optimization classification scale, and extract the first optimization instruction; Based on the first optimization instruction, the parameters of the key features corresponding to the process nodes are optimized to optimize the quality of the microbial decomposition environment.

[0034] In one embodiment, the quality prediction trend of the microbial decomposition environment is decomposed into multi-dimensional independent trend factors, the attribution table of the art node-quality monitoring key feature set is traversed, and each trend factor is matched and associated with the corresponding key feature. The mapping relationship between the trend factor and the key feature can be determined by statistical analysis, and the complex trend is disassembled into independent factors. The complex environmental quality change trend is disassembled into multiple independent trend factors (such as compliance factors, efficiency factors, environmental factors, etc.), reducing the complexity of the data dimension, facilitating targeted analysis, so as to better analyze and process. The preset classification standard selects threshold division or clustering algorithm, discretizes the trend factor, generates an intensity vector, and generates an intensity vector for each trend factor (for example, [0.8, 0.2, 0] represents "high" intensity), reflecting the intensity of changes in different factors in time or space. Pre-constructing a discrete contribution function and calculating the contribution of each key feature refers to evaluating the degree of influence of each factor on the overall trend. Finally, the optimization level is graded, multi-level optimization instructions are generated, and the optimal first optimization instruction is extracted. During the optimization process, the system can be fully monitored and evaluated, potential faults can be discovered in advance, and timely measures can be taken to repair or adjust them, avoiding system downtime or significant efficiency reduction caused by equipment failure or process abnormalities, and improving the system's operational stability and continuity. Ensuring that the decomposition efficiency of odorous gases is maintained above the preset threshold can ensure that the emitted gases meet strict environmental protection standards, reduce pollution to the environment, and meet increasingly stringent environmental protection regulations. In the process of microbial decomposition of odorous gases, the full-link process may involve multiple links such as microbial inoculation, cultivation, gas import and export. The first optimization instruction focuses on the most basic and critical links. By optimizing the key parameters at the full-link process nodes, the first optimized control parameters are obtained, approaching the optimal parameter combination that can meet the decomposition efficiency and decomposition effect, thereby improving the efficiency and effect of optimization.

[0035] The present application embodiment provides an environmental monitoring system for microbial decomposition of odorous gases, such as Figure 2 As shown, the system includes: The dual constraint module 10 is used to obtain the threshold range of the decomposition effect and decomposition efficiency of the microbial decomposition odor device, which is defined as a dual constraint target. The decomposition effect is the direct basis for meeting the emission standards, and the decomposition efficiency is the instantaneous metabolic capacity of the microorganisms to decompose the odor gas. The quality monitoring database module 20 is used to collect multi-dimensional monitoring data of the microbial decomposition odor device through the environmental monitoring sensor group, including the intake concentration of odorous gas, intake flow rate, liquid flow rate, temperature, humidity, dissolved oxygen concentration, pH value of the microbial reaction unit, and residual concentration of odorous gas, to build a quality monitoring database of the microbial decomposition environment; The quality monitoring key feature set module 30 is used to extract the quality features of the whole-link microbial decomposition environment according to the quality monitoring database to obtain the quality monitoring key feature set of the microbial decomposition environment; The environmental quality prediction trend module 40 is used to build a microbial decomposition environment prediction network library, input a quality monitoring key feature set, and output a predicted quality prediction trend of the microbial decomposition environment; The first optimization instruction module 50 is used to traverse the quality prediction trend through the attribute table of the process node-quality monitoring key feature set, establish a mapping relationship with the key quality features, calculate the risk contribution of the key features on each process node, prioritize the risk contribution, and generate a first optimization instruction, wherein the attribute table of the process node-quality monitoring key feature set is used to characterize the key features corresponding to each process node; The decomposition environment optimization module 60 is used to optimize the quality of the microbial decomposition environment based on the first optimization instruction.

[0036] In general, this application achieves a win-win situation of safety and efficiency by setting dual constraint targets, transforming from single-target passive response to multi-target active optimization; transforming from experience-driven to data-mechanism fusion-driven, improving the scientificity and reliability of the system, ensuring the optimal operation of the odor decomposition device under different working conditions, improving the accuracy, adaptability and robustness of system control, effectively improving the decomposition efficiency of odorous gases, and reducing odorous gas emissions.

[0037] The above description of the disclosed embodiments enables those skilled in the art to implement or use the present application. Various modifications to these embodiments will be apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present application. Therefore, the present application will not be limited to the embodiments shown herein, but will conform to the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. An environmental monitoring method for microbial decomposition of odorous gases, characterized in that: Methods include: The threshold range of the decomposition effect and decomposition efficiency of the microbial decomposition odor device is obtained and defined as a double constraint target. The decomposition effect is the direct basis for meeting the emission standards, and the decomposition efficiency is the instantaneous metabolic capacity of microorganisms to decompose odorous gases. The multi-dimensional monitoring data of the microbial decomposition odor device is collected through the environmental monitoring sensor group, including the intake concentration of odorous gas, intake flow rate, liquid flow rate, temperature, humidity, dissolved oxygen concentration, pH value of the microbial reaction unit, and residual concentration of odorous gas, to build a quality monitoring database of the microbial decomposition environment; Extract the quality characteristics of the whole-link microbial decomposition environment according to the quality monitoring database to obtain the key feature set for quality monitoring of the microbial decomposition environment; Build a microbial decomposition environment prediction network library, input the key feature set of quality monitoring, and output the predicted quality prediction trend of the microbial decomposition environment; Traverse the quality prediction trend through the process node-quality monitoring key feature set attribution table, establish a mapping relationship with the key quality features, calculate the risk contribution of the key features on each process node, prioritize the risk contribution, and generate a first optimization instruction, wherein the process node-quality monitoring key feature set attribution table is used to characterize the key features corresponding to each process node; Based on the first optimization instruction, the quality of the microbial decomposition environment is optimized.

2. The environmental monitoring method for microbial decomposition of odorous gases according to claim 1, characterized in that: Construct a quality monitoring database for microbial decomposition environment, including: Perform outlier identification and data cleaning on the intake concentration of odorous gas, intake flow, liquid flow, temperature, humidity, dissolved oxygen concentration, pH value of the microbial reaction unit, and residual concentration of odorous gas to obtain usable multi-dimensional monitoring data; Establish node mapping relationship between available multi-dimensional monitoring data and process nodes of microbial decomposition of odorous gas; The node state timing difference of the process of microbial decomposition of odorous gas is obtained, and the multi-dimensional monitoring data are time-correlated based on the node state timing difference to build a quality monitoring database.

3. The environmental monitoring method for microbial decomposition of odorous gases according to claim 1, characterized in that: Obtain a set of key characteristics for monitoring the quality of the microbial decomposition environment, including: Obtain the intake concentration of odorous gas and the residual concentration of odorous gas in the quality monitoring database, obtain the current decomposition effect and the current decomposition efficiency, evaluate the current decomposition environment of microorganisms through the current decomposition effect and the current decomposition efficiency, and obtain the comprehensive performance and dominant mode of the current decomposition environment of microorganisms. The comprehensive performance includes a high-efficiency zone, a medium-efficiency zone and a low-efficiency zone. The dominant mode includes an efficiency-dominant zone, an effect-dominant zone and a balanced zone. The decomposition effect is the residual concentration of odorous gas, and the decomposition efficiency is the difference between the intake concentration of odorous gas and the residual concentration of odorous gas divided by the intake concentration of odorous gas; According to the comprehensive performance and dominant mode of the current microbial decomposition environment, the expert experience method is used to assign initial weights to the air flow rate, liquid flow rate, temperature, humidity, dissolved oxygen concentration, and pH value of the microbial reaction unit; The air intake flow, liquid flow, temperature, humidity, dissolved oxygen concentration, pH value of the microbial reaction unit in the quality monitoring database and the decomposition effect are analyzed to obtain the correlation coefficient of each monitoring data, and the initial weight is adjusted according to the correlation coefficient to obtain the weight of each monitoring data after adjustment; The air intake flow, liquid flow, temperature, humidity, dissolved oxygen concentration, and pH value of the microbial reaction unit in the quality monitoring database are input into the decomposition effect prediction model to obtain the importance score of each monitoring data to the residual concentration of the odorous gas. The decomposition effect prediction model is constructed based on the random forest model. The optimal number of trees and splitting rules are determined by cross-validation. The model is trained to predict the residual concentration of the odorous gas and the initial importance score of each monitoring data is output synchronously. The initial importance score of each monitoring data output by the random forest model is multiplied by the weight of each monitoring data after adjustment and normalized to generate an importance score. A score threshold is preset, and monitoring data with importance scores greater than the score threshold is extracted as key monitoring data. The key monitoring data features are extracted to obtain a key feature set for quality monitoring.

4. The environmental monitoring method for microbial decomposition of odorous gases according to claim 3, characterized in that: The current decomposition environment of microorganisms is evaluated through the current decomposition effect and current decomposition efficiency, and the comprehensive performance and dominant mode of the current decomposition environment of microorganisms are obtained, including: A sampling sliding window is preset, and the intake concentration of odorous gas and the residual concentration of odorous gas are obtained within the sampling sliding window to obtain the decomposition efficiency and decomposition effect; The decomposition efficiency and decomposition effect are normalized and mapped to the polar coordinate system. The radial distance of the polar coordinate system represents the comprehensive performance strength, and the polar angle represents the balance between the decomposition efficiency and the decomposition effect. The formula is as follows: Normalization: , ,t=1,2,3...,T; Among them, η norm is the normalized decomposition efficiency, η (t) is the decomposition efficiency calculated at time t, η min is the minimum value in the threshold range of decomposition efficiency, η max is the maximum value in the threshold range of decomposition efficiency, E norm is the decomposition effect after normalization, It is the decomposition effect index, that is, the reciprocal of the residual concentration of odorous gas. is a very small constant, E min is the minimum value in the threshold range of the decomposition effect, E max is the maximum value in the threshold range of the decomposition effect; The polar coordinate conversion formula is as follows: , ,t=1,2,3...,T; Where r(t) is the radial distance at time t, η norm is the normalized decomposition efficiency, E norm is the decomposition effect after normalization, θ(t) is the polar angle at time t, θ∈[-π,π], t=1,2,3...,T, t is each sampling moment in the sampling sliding window, and all sampling moments are sampling sliding window data; The density distribution of radial distance is analyzed by histogram method to determine the threshold range of comprehensive performance level, and the comprehensive performance level includes high efficiency, medium efficiency and low efficiency; The polar angle is divided into N sectors through polar angle coordinate transformation and clustering method to determine the interval boundary of the dominant mode. The multi-dimensional dominant mode includes decomposition efficiency dominant area, effect dominant area and equilibrium area. Cartesian combination of the level threshold interval of the comprehensive performance and the interval boundary of the dominant mode is performed to generate a matrix of the combination of the comprehensive performance and the dominant mode; The radial distance and polar angle corresponding to the current decomposition effect and the current decomposition efficiency are calculated, and the matrix of the combination of comprehensive performance and dominant mode is traversed to obtain the comprehensive performance and dominant mode of the current decomposition environment of microorganisms.

5. The environmental monitoring method for microbial decomposition of odorous gases according to claim 1, characterized in that: Build a network library for decomposing odorous gas environment prediction, including: Mining and obtaining the decomposition odor gas monitoring database, extracting the key feature data of the whole link monitoring; The key characteristic data of the whole-link monitoring are classified and identified according to the type of odorous gas to obtain the decomposed odorous gas monitoring classification data set; Feature extraction is performed on the decomposition odor gas monitoring classification data set to obtain the decomposition odor gas monitoring classification feature set; The decomposed odor gas monitoring classification feature set is trained for working environment quality trend prediction respectively to obtain a multi-dimensional decomposed environment trend prediction branch network set; The multi-dimensional decomposition environmental trend prediction branch network set is classified and connected in series according to the odor gas type to generate multiple odor gas type decomposition environmental prediction network sets; The environmental prediction network sets of multiple odorous gas types are identified and integrated, and a network library for decomposing the working environment prediction of odorous gases is built.

6. The environmental monitoring method for microbial decomposition of odorous gas according to claim 2 or 3, characterized in that: Establish a table of process node-quality monitoring key feature sets to characterize the key features corresponding to each process node, including: Acquire available multi-dimensional monitoring data and establish a mapping relationship between process nodes for microbial decomposition of odorous gases, and establish an attribution table between process nodes and available multi-dimensional monitoring data; Obtain key monitoring data and quality monitoring key feature sets, extract the attribution table of process nodes and key monitoring data from the attribution table of process nodes and available multidimensional monitoring data based on the key monitoring data, and establish the attribution table of process nodes-quality monitoring key feature sets based on the correspondence between key monitoring data and quality monitoring key feature sets.

7. The environmental monitoring method for microbial decomposition of odorous gases according to claim 6, characterized in that: Generate a first optimization instruction, including: Decompose the quality prediction trend of the microbial decomposition environment into multi-dimensional independent trend factors, traverse the attribution table of the process node-quality monitoring key feature set, and establish a mapping relationship with the key features; The classification standard is preset to discretize the trend of multi-dimensional independent trend factors and generate the intensity vector of multi-dimensional trend factors; Pre-build discrete contribution function to calculate the discrete contribution of each key feature; Preset an optimization classification scale, classify the contribution of each key feature according to the optimization classification scale, and extract the first optimization instruction; Based on the first optimization instruction, the parameters of the key features corresponding to the process nodes are optimized to optimize the quality of the microbial decomposition environment.

8. Environmental monitoring system for microbial decomposition of odorous gases, characterized in that: The system includes: The dual constraint module is used to obtain the threshold range of the decomposition effect and decomposition efficiency of the microbial decomposition odor device, which is defined as a dual constraint target. The decomposition effect is the direct basis for meeting the emission standards, and the decomposition efficiency is the instantaneous metabolic capacity of microorganisms to decompose odorous gases. The quality monitoring database module is used to collect multi-dimensional monitoring data of the microbial decomposition odor device through the environmental monitoring sensor group, including the intake concentration of odorous gas, intake flow rate, liquid flow rate, temperature, humidity, dissolved oxygen concentration, pH value of the microbial reaction unit, and residual concentration of odorous gas, so as to build a quality monitoring database of the microbial decomposition environment; The quality monitoring key feature set module and the quality monitoring multidimensional feature set module are used to extract the quality features of the whole-link microbial decomposition environment according to the quality monitoring database, and obtain the quality monitoring key feature set of the microbial decomposition environment; Environmental quality prediction trend module: The environmental quality prediction trend module is used to build a microbial decomposition environment prediction network library, input the key feature set of quality monitoring, and output the predicted quality prediction trend of the microbial decomposition environment; A first optimization instruction module, which is used to traverse the quality prediction trend through the attribution table of the process node-quality monitoring key feature set, establish a mapping relationship with the key quality features, calculate the risk contribution of the key features on each process node, prioritize the risk contribution, and generate a first optimization instruction, wherein the attribution table of the process node-quality monitoring key feature set is used to characterize the key features corresponding to each process node; The decomposition environment optimization module is used to optimize the quality of the microbial decomposition environment based on the first optimization instruction.

Images