Organic waste gas purification treatment system and method

Abstract

The present application relates to the field of intelligent control, and specifically discloses a system and method for purifying and treating organic waste gas, which uses a filter to preliminarily filter the waste gas, then utilizes an ozone catalyst to catalyze ozone and preliminarily oxidize the volatile organic compounds (VOCs) in the filtered waste gas, and then uses ultraviolet radiation nano-photocatalysts to further oxidize the preliminarily oxidized VOCs waste gas, converting it into small molecule organic matter that is more easily biodegradable, and monitors the concentration of VOCs in the waste gas in real time, and dynamically adjusts the fan speed according to the monitoring data to ensure the purification efficiency and the optimal operating state of the system.

Description

Technical Field

[0001] The present application relates to the field of intelligent control, and more specifically, to a system and method for purifying and treating organic waste gas. Background Art

[0002] With the acceleration of industrialization, the emission of organic compounds (VOCs) is becoming increasingly serious, posing a significant threat to the environment and human health. VOCs not only deteriorate air quality but also contribute to photochemical smog and ozone depletion. Therefore, effectively treating VOC waste gases has become a critical issue in environmental protection. Traditional VOC treatment methods, such as adsorption, absorption, and combustion, suffer from high energy consumption, low treatment efficiency, and secondary pollution.

[0003] In this regard, patent CN113731168A discloses a volatile organic waste gas purification system and method, which includes five steps: air intake filtration, ozone catalytic oxidation, nano-photocatalyst enhanced oxidation, ozone decomposition and BAF biodegradation. This process uses physical filtration to remove particulate matter, chemical oxidation to decompose difficult-to-degrade VOCs into easily biodegradable small molecules, and finally further degrades the remaining organic matter through a biological filter to solve traditional treatment problems and achieve efficient purification.

[0004] When dealing with volatile organic compounds (VOCs), precise control of fan speed is crucial to improving purification efficiency and reducing energy consumption. In the above-mentioned patent, the fan speed is adjusted through controller configuration. However, the traditional control method uses a preset fan speed, which causes the system to react slowly to changes in VOCs exhaust gas concentration and is difficult to quickly adapt to actual working conditions. When the exhaust gas concentration suddenly increases, the fixed low speed cannot provide sufficient airflow speed to handle the additional pollutants, thereby reducing the purification efficiency; and when the concentration decreases, the high speed will cause unnecessary energy consumption. In addition, due to the fixed parameters, the traditional system cannot be flexibly adjusted according to the real-time exhaust gas concentration fluctuations, making its performance unstable under different working conditions and difficult to always be in the optimal state.

[0005] Therefore, an optimized organic waste gas purification treatment solution is desired. Summary of the Invention

[0006] In order to solve the above technical problems, this application is proposed.

[0007] According to one aspect of the present application, an organic waste gas purification and treatment system is provided, which includes: a fan; a filter for filtering waste gas to obtain filtered waste gas; a primary oxidation module for using an ozone catalyst to catalyze ozone to perform primary oxidation of VOCs in the filtered waste gas to obtain primary oxidized VOCs waste gas; a secondary oxidation module for using ultraviolet radiation nano-photocatalysts to perform secondary oxidation on the primary oxidized VOCs waste gas to obtain biodegradable small molecular organic matter; a controller for real-time monitoring of VOCs waste gas concentration data in the waste gas during the waste gas purification process and adjusting the speed of the fan in real time; wherein the controller includes:

[0008] A VOCs exhaust gas concentration collection and transmission unit is used to collect a time series data set of VOCs exhaust gas concentration data at a predetermined sampling frequency using a VOCs exhaust gas concentration detector, and transmit the time series data set of VOCs exhaust gas concentration data to a fan speed control center through a wireless communication module;

[0009] A VOCs exhaust gas concentration time series analysis unit is used to perform phase space reconstruction and local time series extraction on the time series data set of the VOCs exhaust gas concentration data at the fan speed control center to obtain a set of VOCs exhaust gas concentration local time series implicit pattern features;

[0010] The VOCs exhaust gas concentration time series propagation unit is used to perform multi-gated feature dynamic propagation on the set of local time series implicit pattern features of the VOCs exhaust gas concentration to obtain a VOCs exhaust gas concentration time series feature propagation representation, and based on the VOCs exhaust gas concentration time series feature propagation representation, obtain an optimization instruction, and the optimization instruction includes a recommended value of the fan speed at the next time point.

[0011] According to another aspect of the present application, a method for purifying and treating organic waste gas is provided, which is used to implement the above-mentioned organic waste gas purification and treatment system. The method for purifying and treating organic waste gas comprises:

[0012] Using a VOCs exhaust gas concentration detector to collect a time series data set of VOCs exhaust gas concentration data at a predetermined sampling frequency, and transmitting the time series data set of VOCs exhaust gas concentration data to a fan speed control center through a wireless communication module;

[0013] At the fan speed control center, phase space reconstruction and local time series extraction are performed on the time series data set of the VOCs exhaust gas concentration data to obtain a set of implicit pattern features of the local time series of the VOCs exhaust gas concentration;

[0014] Multi-gated feature dynamic propagation is performed on the set of local time series implicit pattern features of the VOCs exhaust gas concentration to obtain a VOCs exhaust gas concentration time series feature propagation representation, and based on the VOCs exhaust gas concentration time series feature propagation representation, an optimization instruction is obtained, and the optimization instruction includes a recommended value of the fan speed at the next time point.

[0015] Compared with the prior art, the present application provides an organic waste gas purification and treatment system and method, which uses a VOCs waste gas concentration detector to collect a time series data set of VOCs waste gas concentration data at a predetermined sampling frequency, and transmits it to the fan speed control center through a wireless communication module. In the fan speed control center, a data analysis and encoding method based on deep learning is used to reconstruct the phase space and extract local time series implicit information from the time series data set of VOCs waste gas concentration data. In this way, the dynamic propagation representation obtained by multi-gated propagation based on the local time series implicit pattern characteristics of each VOCs waste gas concentration is used to intelligently obtain the recommended fan speed value for the next time point. In this way, the VOCs waste gas concentration can be monitored in real time, and the complex pattern of the VOCs waste gas concentration can be identified. The fan speed can be intelligently adjusted according to the current and predicted waste gas conditions, thereby improving the adaptability to different working conditions. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] The above and other purposes, features, and advantages of the present application will become more apparent through a more detailed description of the embodiments of the present application in conjunction with the accompanying drawings. The accompanying drawings are intended to provide a further understanding of the embodiments of the present application and constitute a part of the specification. Together with the embodiments of the present application, they are used to explain the present application and do not constitute a limitation of the present application. In the drawings, the same reference numerals generally represent the same components or steps.

[0017] Figure 1 4 is a block diagram of an organic waste gas purification system according to an embodiment of the present application.

[0018] Figure 2 This is a block diagram of a controller in an organic waste gas purification system according to an embodiment of the present application.

[0019] Figure 3 This is a block diagram of a VOCs waste gas concentration time series propagation unit in an organic waste gas purification system according to an embodiment of the present application.

[0020] Figure 4 This is a block diagram of a subunit for the time-series dynamic propagation of VOCs waste gas concentration in an organic waste gas purification and treatment system according to an embodiment of the present application.

[0021] Figure 5 Flowchart of a method for purifying organic waste gas according to an embodiment of the present application. DETAILED DESCRIPTION

[0022] The following describes embodiments of the present disclosure in more detail with reference to the accompanying drawings. While the drawings illustrate certain embodiments of the present disclosure, it should be understood that the present disclosure can be implemented in various forms and should not be construed as limited to the embodiments described herein. Rather, these embodiments are provided to provide a more thorough and complete understanding of the present disclosure. It should be understood that the drawings and embodiments of the present disclosure are for illustrative purposes only and are not intended to limit the scope of protection of the present disclosure.

[0023] It should be understood that the various steps described in the method embodiments of the present disclosure may be performed in a different order and / or in parallel. Furthermore, the method embodiments may include additional steps and / or omit the steps shown. The scope of the present disclosure is not limited in this respect.

[0024] In the description of the embodiments of the present disclosure, the term "including" and similar terms should be understood as open inclusion, i.e., "including but not limited to." The term "based on" should be understood as "based at least in part on." The term "one embodiment" or "the embodiment" should be understood as "at least one embodiment." The terms "first," "second," etc. may refer to different or the same objects. Other explicit and implicit definitions may also be included below.

[0025] With the acceleration of industrialization, the emission of organic compounds (VOCs) is becoming increasingly serious, posing a significant threat to the environment and human health. VOCs not only deteriorate air quality but also contribute to photochemical smog and ozone depletion. Therefore, effectively treating VOC waste gases has become a critical issue in environmental protection. Traditional VOC treatment methods, such as adsorption, absorption, and combustion, suffer from high energy consumption, low treatment efficiency, and secondary pollution.

[0026] In this regard, patent CN113731168A discloses a volatile organic waste gas purification system and method, which includes five steps: air intake filtration, ozone catalytic oxidation, nano-photocatalyst enhanced oxidation, ozone decomposition and BAF biodegradation. This process uses physical filtration to remove particulate matter, chemical oxidation to decompose difficult-to-degrade VOCs into easily biodegradable small molecules, and finally further degrades the remaining organic matter through a biological filter to solve traditional treatment problems and achieve efficient purification.

[0027] When dealing with volatile organic compounds (VOCs), precise control of fan speed is crucial to improving purification efficiency and reducing energy consumption. In the above-mentioned patent, the fan speed is adjusted through controller configuration. However, the traditional control method uses a preset fan speed, which causes the system to react slowly to changes in VOCs exhaust gas concentration and is difficult to quickly adapt to actual working conditions. When the exhaust gas concentration suddenly increases, the fixed low speed cannot provide sufficient airflow speed to handle the additional pollutants, thereby reducing the purification efficiency; and when the concentration decreases, the high speed will cause unnecessary energy consumption. In addition, due to the fixed parameters, the traditional system cannot be flexibly adjusted according to the real-time exhaust gas concentration fluctuations, making its performance unstable under different working conditions and difficult to always be in the optimal state.

[0028] Based on this, the present application proposes an organic waste gas purification and treatment system. Figure 1 The block diagram of the organic waste gas purification and treatment system according to an embodiment of the present application is shown. Specifically, the organic waste gas purification and treatment system 100 according to an embodiment of the present application includes: a fan 110; a filter 120 for filtering waste gas to obtain filtered waste gas; a primary oxidation module 130 for utilizing an ozone catalyst to catalyze ozone to perform primary oxidation of VOCs in the filtered waste gas to obtain primary oxidized VOC waste gas; a secondary oxidation module 140 for utilizing ultraviolet radiation nanophotocatalysts to perform secondary oxidation of the primary oxidized VOC waste gas to obtain biodegradable small molecule organic matter; and a controller 150 for monitoring VOC waste gas concentration data in real time during the waste gas purification process and adjusting the fan speed in real time.

[0029] Specifically, the fan 110. It should be understood that the fan plays a key role in the exhaust gas purification process, and it supports the effective implementation of the entire purification process by accurately controlling the gas flow rate of the VOCs exhaust gas.

[0030] Specifically, the filter 120 is used to filter the exhaust gas to obtain filtered exhaust gas.

[0031] That is, the filter is used to separate and filter particulate matter in VOCs exhaust gas, the diameter of which is ≥0.3 μm. In particular, the filter can be one of a coarse filter, a HEPA filter or an activated carbon filter.

[0032] Specifically, the preliminary oxidation module 130 utilizes an ozone catalyst to catalyze ozone to preliminarily oxidize the VOCs in the filtered exhaust gas to obtain preliminarily oxidized VOCs exhaust gas. It should be understood that the preliminary oxidation module is a key component specially designed for the efficient oxidation of difficult-to-degrade VOCs exhaust gas. The module consists of a built-in ozone generator module and an external ozone source module. The ozone catalyst used is a microbial charcoal-supported multi-metal nanoparticle catalyst. These catalysts use microbial cells or bacterial residues as carriers and contain transition metal and rare earth metal components. This design can significantly improve the reaction efficiency between ozone and VOCs.

[0033] Specifically, after preliminary filtration to remove particulate matter, the exhaust gas is introduced into the preliminary oxidation module, where a built-in or external ozone generator generates sufficient ozone. This ozone then comes into contact with the VOCs in the exhaust gas and reacts under specific conditions.

[0034] To accelerate and optimize this oxidation process, the system uses a microbial charcoal-supported multi-metal nanoparticle catalyst. This catalyst uses microbial cells or bacterial residue as a carrier and contains transition metal and rare earth metal components, which can significantly improve the reaction efficiency between ozone and VOCs. When ozone and VOCs meet, under the action of the catalyst, the ozone molecules become more active and can more effectively oxidize the VOCs. In this process, the carbon-hydrogen bonds in the VOCs are broken, forming a series of simpler intermediates, such as aldehydes and ketones. These intermediates are easier to be further decomposed by subsequent treatment steps than the original VOCs.

[0035] In the initial oxidation stage, the ozone catalyst not only promotes the oxidation reaction of VOCs, but also inhibits the formation of by-products to a certain extent, avoiding possible secondary pollution problems. For example, by controlling the reaction conditions, the production of nitrogen oxides and other harmful gases can be reduced. In addition, since ozone is a strong oxidant, it can also effectively remove some other pollutants in the exhaust gas, such as sulfides and ammonia, further improving the purification effect. In short, the ozone catalyst is used to catalyze ozone to carry out preliminary oxidation of VOCs in the filtered exhaust gas, and through efficient chemical reactions, complex organic pollutants are converted into small molecules that are easy to handle, laying the foundation for subsequent deep purification.

[0036] Specifically, the secondary oxidation module 140 utilizes ultraviolet radiation to irradiate nano-photocatalysts to secondary oxidize the pre-oxidized VOCs waste gas to produce biodegradable small-molecule organic compounds. It should be understood that the module is internally equipped with one to six UV lamps and a nano-photocatalyst mesh, where the catalyst is titanium dioxide doped with one or more metals selected from iron, cobalt, manganese, and cerium. Ultraviolet radiation irradiates the nano-photocatalysts, further enhancing the oxidation of VOCs waste gas. Its primary function is to break down or open the chains of typical pollutants in the VOCs waste gas, such as aromatic compounds and heterocyclic compounds, into biodegradable small-molecule organic compounds.

[0037] Specifically, first, after the ozone catalysis in the initial oxidation stage, most of the VOCs in the exhaust gas have been partially oxidized into relatively simple intermediates, such as aldehydes and ketones. However, these intermediates still have a certain degree of complexity and stability, and require further treatment to achieve higher purification standards. To this end, the system introduces ultraviolet radiation nano-photocatalyst technology for secondary oxidation. The core of this technology is to use ultraviolet light of a specific wavelength to irradiate a catalytic network containing nano-scale photocatalysts, thereby exciting the photocatalyst to produce highly oxidizing free radicals, thereby deeply oxidizing VOCs and their derivatives.

[0038] Specifically, the secondary oxidation module houses one to six UV lamps and a nano-photocatalytic mesh. The photocatalyst is typically titanium dioxide doped with one or more of the following metals: iron, cobalt, manganese, and cerium, forming nanoparticles with a highly active surface. When UV light strikes these nano-photocatalysts, electrons jump from the valence band to the conduction band, leaving holes in the valence band. These excited electrons and holes react with surrounding oxygen and water molecules to generate highly oxidizing hydroxyl radicals (·OH) and superoxide anion radicals (O2-). These radicals rapidly attack and break down carbon-hydrogen and other chemical bonds within the VOC molecular structure, gradually decomposing them into carbon dioxide, water, and other harmless small molecules. This secondary oxidation process not only further reduces the concentration of residual VOCs in the exhaust gas but also significantly improves their biodegradability. This is because small organic molecules are more easily decomposed by microorganisms than the original large VOC molecules, allowing for more thorough removal in the subsequent biological treatment stage. In addition, since the photocatalyst used in the entire process has good stability and recyclability, it can maintain a high catalytic efficiency throughout its service life, reducing the cost of frequent catalyst replacement.

[0039] Specifically, the controller 150 is used to monitor VOC concentration data in the exhaust gas in real time during the exhaust gas purification process and adjust the fan speed in real time. It should be understood that during the exhaust gas purification process, if the VOC concentration in the exhaust gas is high and not promptly addressed, the purification system may overload and fail to effectively remove all harmful substances, thereby affecting the ultimate purification effect. If the VOC concentration is low, the system may not need to operate at full capacity, and energy can be saved by reducing the fan speed, while still maintaining good purification results. Therefore, to ensure efficient, economical, and environmentally friendly exhaust gas treatment, by continuously monitoring VOC concentration, the system can quickly respond to changes in pollutants in the exhaust gas and automatically adjust the fan speed to optimize airflow, thereby improving purification efficiency. When the VOC concentration increases, increasing the fan speed can more effectively treat additional pollutants; when the concentration is low, reducing the speed saves energy. This intelligent control not only improves the overall treatment effect, but also significantly reduces energy consumption and operating costs. It also reduces the risk of secondary pollution caused by improper treatment, achieving a cleaner and more efficient industrial production environment. Such a closed-loop control mechanism is an important technical means to achieve sustainable development.

[0040] Accordingly, in the control module, the technical concept of the present application is to use a VOCs exhaust gas concentration detector to collect a time series data set of VOCs exhaust gas concentration data at a predetermined sampling frequency, and transmit it to the fan speed control center through a wireless communication module. In the fan speed control center, a data analysis and encoding method based on deep learning is used to perform phase space reconstruction and local time series implicit extraction on the time series data set of VOCs exhaust gas concentration data, so as to intelligently obtain the recommended value of the fan speed at the next time point by performing multi-gated propagation based on the local time series implicit pattern characteristics of each VOCs exhaust gas concentration. In this way, the VOCs exhaust gas concentration can be monitored in real time, and the complex pattern of the VOCs exhaust gas concentration can be identified. The fan speed can be intelligently adjusted according to the current and predicted exhaust gas conditions, thereby improving the adaptability to different working conditions.

[0041] Figure 2 FIG. 1 is a block diagram of a controller in an organic waste gas purification system according to an embodiment of the present application. Specifically, Figure 2As shown, the controller 150 includes: a VOCs exhaust gas concentration acquisition and transmission unit 151, which is used to use a VOCs exhaust gas concentration detector to collect a time series data set of VOCs exhaust gas concentration data at a predetermined sampling frequency, and transmit the time series data set of the VOCs exhaust gas concentration data to the fan speed control center through a wireless communication module; a VOCs exhaust gas concentration time series analysis unit 152, which is used to perform phase space reconstruction and local time series extraction on the time series data set of the VOCs exhaust gas concentration data at the fan speed control center to obtain a set of local time series implicit pattern features of the VOCs exhaust gas concentration; a VOCs exhaust gas concentration time series propagation unit 153, which is used to perform multi-gated feature dynamic propagation on the set of local time series implicit pattern features of the VOCs exhaust gas concentration to obtain a VOCs exhaust gas concentration time series feature propagation representation, and obtain an optimization instruction based on the VOCs exhaust gas concentration time series feature propagation representation, and the optimization instruction includes a recommended value of the fan speed at the next time point.

[0042] In an embodiment of the present application, the VOCs exhaust gas concentration collection and transmission unit 151 is used to collect a time series dataset of VOCs exhaust gas concentration data using a VOCs exhaust gas concentration detector at a predetermined sampling frequency, and transmit the time series dataset of VOCs exhaust gas concentration data to the fan speed control center via a wireless communication module. It should be understood that VOCs (volatile organic compound) exhaust gas concentration data refers to the content of VOCs in exhaust gas at a specific time and location. This data is usually expressed as a mass concentration (such as mg / m³ or ppm), reflecting the amount of VOCs in the exhaust gas. VOCs are a class of organic compounds widely present in industrial exhaust gas, including but not limited to benzene, toluene, xylene, formaldehyde, etc., which are potentially harmful to the environment and human health. Therefore, in order to achieve real-time monitoring and effective management of VOCs exhaust gas concentration, in the technical solution of the present application, a VOCs exhaust gas concentration detector is used to collect a time series dataset of VOCs exhaust gas concentration data at a predetermined sampling frequency.

[0043] Highly sensitive VOC exhaust gas concentration detectors are installed at exhaust outlets or key locations in the treatment process. These detectors are designed to continuously collect VOC concentration data in the exhaust gas at a predetermined sampling frequency. The sampling frequency is determined by the specific application scenario and can be set to every minute, every second, or even shorter intervals. This time-series dataset not only records the changes in VOC concentration over time in detail, but also provides comprehensive foundational information for subsequent data analysis.

[0044] VOCs exhaust gas concentration detectors typically utilize advanced sensing technologies, such as photoionization detectors (PIDs), flame ionization detectors (FIDs), or infrared absorption spectrometers. These devices can accurately measure the concentrations of different VOCs in exhaust gas. While their operating principles are based on different physical and chemical mechanisms, they share a common characteristic: rapid response and accurate, reliable concentration readings. By setting an appropriate sampling frequency, the system can capture subtle changes in VOC concentrations, which is crucial for identifying sudden pollution events and evaluating purification effectiveness.

[0045] Once the VOCs exhaust gas concentration data is collected, the next step is to transmit this data to the fan speed control center in real time. In this process, modern IoT technology plays a core role, ensuring stable and reliable data transmission even in complex industrial environments. The wireless communication module serves as a bridge between on-site sensors and the control center. Its selection depends on the specific network conditions and requirements of the site, and may include Wi-Fi, 4G / 5G cellular networks, or low-power wide area network technologies such as LoRa. Each technology has its advantages: Wi-Fi is suitable for high-speed data transmission over short distances; 4G / 5G cellular networks are suitable for application scenarios with wide coverage and high bandwidth requirements; and LPWAN (low-power wide area network) technologies such as LoRa are particularly suitable for long-distance, low-power environmental monitoring applications.

[0046] Wireless transmission greatly simplifies system deployment and maintenance, avoiding the wiring complexity and high maintenance costs associated with traditional wired connections. It also ensures real-time and secure data transmission. In actual operation, the wireless communication module regularly sends data packets containing the latest VOC concentration information to the control center. This allows the control system to obtain the latest environmental conditions and make appropriate adjustments. Furthermore, to enhance data transmission security, the system employs encryption technology and authentication mechanisms to protect data from unauthorized access.

[0047] Through this design, the entire process from the collection of VOCs concentration data to transmission and final analysis and decision-making forms a closed-loop feedback system, realizing intelligent management of the exhaust gas purification process.

[0048] In an embodiment of the present application, the VOCs exhaust gas concentration time series analysis unit 152 is used to perform phase space reconstruction and local time series extraction on the time series data set of the VOCs exhaust gas concentration data at the fan speed control center to obtain a set of VOCs exhaust gas concentration local time series implicit pattern features. Specifically, the VOCs exhaust gas concentration time series analysis unit includes: a VOCs exhaust gas concentration phase space reconstruction subunit, used to perform phase space reconstruction on the time series data set of the VOCs exhaust gas concentration data at the fan speed control center to obtain a set of VOCs exhaust gas concentration data time series subsequences; a VOCs exhaust gas concentration local time series encoding subunit, used to input each VOCs exhaust gas concentration data time series subsequence in the set of VOCs exhaust gas concentration data time series subsequences into a sequence encoder based on a forward LSTM model to obtain a set of VOCs exhaust gas concentration local time series implicit pattern feature vectors as the set of VOCs exhaust gas concentration local time series implicit pattern features.

[0049] Specifically, the VOCs exhaust gas concentration phase space reconstruction subunit is used to perform phase space reconstruction on the time series data set of the VOCs exhaust gas concentration data in the fan speed control center to obtain a set of VOCs exhaust gas concentration data time series subsequences. Accordingly, considering that in VOCs exhaust gas treatment, VOCs exhaust gas concentration usually exhibits complex nonlinear dynamic behavior, that is, there are complex interactions between the variables, making its behavior difficult to describe with a simple linear model. In addition, some may exhibit chaotic behavior, that is, behavior that appears random but is actually driven by deterministic rules. In other words, in the VOCs treatment process, even slight changes in the initial conditions of the VOCs exhaust gas concentration may lead to significantly different results. This sensitivity makes prediction and control difficult. Therefore, in the technical solution of the present application, phase space reconstruction is introduced to reconstruct the time series data set of the VOCs exhaust gas concentration data to convert the one-dimensional time series into a trajectory in a high-dimensional phase space, thereby better representing the dynamic characteristics of the VOCs exhaust gas concentration. It is worth mentioning that the phase space is a multidimensional space, in which each dimension represents a state variable of the system. For a complex system, a point in phase space represents the state of the system at a specific moment. Takens' embedding theorem is the basis for phase space reconstruction. This theorem states that for an infinitely long, noiseless d-dimensional dynamical system, by choosing an appropriate time lag τ and embedding dimension m, the one-dimensional time series can be reconstructed as a trajectory in an m-dimensional phase space, thereby recovering the system's dynamic characteristics.

[0050] Specifically, in a specific embodiment of the present application, the phase space reconstruction is performed on the time series data set of the VOCs exhaust gas concentration data to obtain a set of VOCs exhaust gas concentration data time series subsequences. The specific steps are as follows:

[0051] Suppose there is a time series data of VOCs exhaust gas concentration x(t) = [x(1), x(2), ..., x(n)], where n is the number of data in the time series of VOCs exhaust gas concentration.

[0052] First, choose an appropriate time lag τ, typically determined by the autocorrelation function or mutual information method. For example, by calculating the autocorrelation function and finding the first local minimum, we can use this as the time lag τ. Next, choose the embedding dimension m, typically determined using the false nearest neighbor method. Starting with an embedding dimension of m = 2, we gradually increase m until the number of false nearest neighbors stabilizes.

[0053] Next, the original one-dimensional VOC exhaust concentration time series data is reconstructed into a trajectory in a high-dimensional phase space, based on the selected time lag τ and embedding dimension m. Each point in the reconstructed phase space consists of m consecutive data points separated by a time interval τ. Specifically, for each starting time point i, the new data point Y(i) consists of x(i), x(i+τ), x(i+2τ), and so on to x(i+(m-1)τ).

[0054] Finally, the reconstructed phase space trajectory is divided into multiple subsequences, that is, a collection of VOCs exhaust gas concentration data time series subsequences, each of which represents the dynamic behavior of VOCs exhaust gas concentration data over a period of time. The number of subsequences depends on the length of the original time series and the embedding dimension m. For example, if the original time series has n data points and the embedding dimension is m, then n - (m-1)τ subsequences can be generated. Through the above steps, the original one-dimensional VOCs exhaust gas concentration data set can be reconstructed into a trajectory in a high-dimensional phase space, and a collection of VOCs exhaust gas concentration data time series subsequences is generated, laying the foundation for subsequent feature extraction and prediction.

[0055] Specifically, the VOCs exhaust gas concentration local time series encoding subunit is used to input each VOCs exhaust gas concentration data time series subsequence in the set of VOCs exhaust gas concentration data time series subsequences into a sequence encoder based on a forward LSTM model to obtain a set of VOCs exhaust gas concentration local time series implicit pattern feature vectors as the set of VOCs exhaust gas concentration local time series implicit pattern features. It should be understood that, considering that there are different degrees of temporal correlation relationships between different local time periods in each VOCs exhaust gas concentration data time series subsequence, in order to extract and represent the complex temporal dynamic correlation patterns in each VOCs exhaust gas concentration data sequence data, in the technical solution of the present application, each VOCs exhaust gas concentration data time series subsequence in the set of the VOCs exhaust gas concentration data time series subsequences is respectively input into a sequence encoder based on a forward LSTM model to utilize the forward LSTM model to capture the long-term dependency correlation relationship in the time series through its internal memory unit, and obtain a set of VOCs exhaust gas concentration local time series implicit pattern feature vectors, so as to better reflect the inherent local structure and dynamic characteristic changes of the VOCs exhaust gas concentration data.

[0056] In an embodiment of the present application, the VOCs exhaust gas concentration time series propagation unit 153 is used to perform multi-gated feature dynamic propagation on the set of local time series implicit pattern features of the VOCs exhaust gas concentration to obtain a VOCs exhaust gas concentration time series feature propagation representation, and based on the VOCs exhaust gas concentration time series feature propagation representation, obtain an optimization instruction, and the optimization instruction includes a recommended value of the fan speed at the next time point.

[0057] Figure 3 FIG. 1 is a block diagram of a VOCs waste gas concentration time series propagation unit in an organic waste gas purification system according to an embodiment of the present application. Specifically, Figure 3 As shown, the VOCs exhaust gas concentration time series propagation unit 153 includes: a VOCs exhaust gas concentration time series dynamic propagation subunit 1531, which is used to perform multi-gated feature dynamic propagation on the set of the VOCs exhaust gas concentration local time series implicit pattern feature vectors to obtain a VOCs exhaust gas concentration time series feature propagation representation vector as the VOCs exhaust gas concentration time series feature propagation representation; an optimization instruction generation subunit 1532, which is used to obtain an optimization instruction based on the VOCs exhaust gas concentration time series feature propagation representation vector, and the optimization instruction includes a recommended value of the fan speed at the next time point.

[0058] Specifically, the VOCs exhaust gas concentration time series dynamic propagation subunit 1531 is configured to perform multi-gated feature dynamic propagation on the set of VOCs exhaust gas concentration local time series implicit pattern feature vectors to obtain a VOCs exhaust gas concentration time series feature propagation representation vector as the VOCs exhaust gas concentration time series feature propagation representation. Furthermore, considering that each VOCs exhaust gas concentration local time series implicit pattern feature captures the local time series variation pattern of VOCs exhaust gas concentration within different specific time periods, including short-term increases, decreases, or fluctuations, and considering that changes in VOCs exhaust gas concentration typically have multi-scale temporal dependencies, including short-term fluctuations and long-term trends, it may be difficult to simultaneously capture these dependencies at different scales using a single LSTM model. Therefore, in order to further process the local time series implicit pattern features to generate a more comprehensive and richer feature representation, in the technical solution of the present application, multi-gated feature dynamic propagation is performed on the set of VOCs exhaust gas concentration local time series implicit pattern feature vectors to obtain a VOCs exhaust gas concentration time series feature propagation representation vector. In other words, dynamic propagation of multi-gated features can effectively transfer information between different time points, so that the features of each time point contain not only the concentration information of the current time point, but also the historical and future concentration information, so as to better understand the global patterns and trends of VOCs concentration changes.

[0059] Figure 4 FIG. 1 is a block diagram of a VOCs waste gas concentration time series dynamic propagation subunit in an organic waste gas purification system according to an embodiment of the present application. More specifically, Figure 4As shown, the VOCs exhaust gas concentration time series dynamic propagation subunit 1531 includes: a VOCs exhaust gas concentration feature mapping secondary subunit 15311, which is used to map each VOCs exhaust gas concentration local time series implicit pattern feature vector in the set of the VOCs exhaust gas concentration local time series implicit pattern feature vector to the Poincare space to obtain a set of VOCs exhaust gas concentration local time series heterogeneous modulation coding feature vectors; a VOCs exhaust gas concentration feature propagation secondary subunit 15312, which is used to perform forward LSTM encoding on the set of the VOCs exhaust gas concentration local time series heterogeneous modulation coding feature vectors to obtain a VOCs exhaust gas concentration heterogeneous propagation representation vector ; The VOCs exhaust gas concentration modulation secondary subunit 15313 is used to calculate the weight value of the set of the VOCs exhaust gas concentration local time series heterodomain modulation coding feature vectors based on the VOCs exhaust gas concentration heterodomain propagation representation vector to obtain the set of VOCs exhaust gas concentration transfer modulation weights; The VOCs exhaust gas concentration time series propagation secondary subunit 15314 is used to perform residual processing and mapping on the set of the VOCs exhaust gas concentration local time series heterodomain modulation coding feature vectors and the VOCs exhaust gas concentration heterodomain propagation representation vector based on the set of VOCs exhaust gas concentration transfer modulation weights to obtain the VOCs exhaust gas concentration time series feature propagation representation vector.

[0060] Specifically, the VOCs exhaust gas concentration feature mapping secondary subunit 15311 is used to map each VOCs exhaust gas concentration local time series implicit pattern feature vector in the set of VOCs exhaust gas concentration local time series implicit pattern feature vectors to the Poincare space to obtain a set of VOCs exhaust gas concentration local time series heterogeneous modulation coding feature vectors. That is, each VOCs exhaust gas concentration local time series implicit pattern feature vector is mapped to the Poincare space to use the unique geometric characteristics of the Poincare space to capture the nonlinear interaction between the exhaust gas concentration features and their hierarchical structure, and obtain a set of VOCs exhaust gas concentration local time series heterogeneous modulation coding feature vectors. That is, the Poincare space is a hyperbolic space that can project trajectories in a high-dimensional phase space into a low-dimensional space while retaining the semantic relevance between the features of each time point.

[0061] More specifically, the processing process of the VOCs exhaust gas concentration characteristic mapping secondary subunit 15311 can be expressed as follows: ;in, is the set of characteristic vectors of the local time series implicit pattern of the VOCs exhaust gas concentration, and are the first, second, and third characteristic vectors of the local time series implicit pattern of the VOCs exhaust gas concentration. and The local time series implicit pattern feature vector of VOCs exhaust gas concentration, and are the first weight matrix and the second weight matrix respectively, and They are the first, second, and third in the set of VOCs exhaust gas concentration local time series heterogeneous modulation coding feature vectors. and VOCs exhaust gas concentration local time series heterogeneous modulation coding feature vector, It is a set of local temporal heterogeneous modulation coding feature vectors of the VOCs exhaust gas concentration.

[0062] Specifically, the VOCs exhaust gas concentration feature propagation secondary subunit 15312 is configured to perform forward LSTM encoding on the set of VOCs exhaust gas concentration local temporal heterogeneous modulation coding feature vectors to obtain a VOCs exhaust gas concentration heterogeneous propagation representation vector. This forward LSTM encoding of the set of heterogeneous modulation coding feature vectors simulates transmission based on the Poincare semantic space, capturing long-term dependencies across multiple time points in the VOCs exhaust gas concentration data. The forget gate, input gate, and output gate work together to preserve key temporal feature information, resulting in a VOCs exhaust gas concentration heterogeneous propagation representation vector.

[0063] More specifically, the processing process of the VOCs exhaust gas concentration characteristic propagation secondary subunit 15312 can be expressed as follows: ;in, is the set of VOCs exhaust gas concentration local time series heterogeneous modulation coding feature vectors, is the forward LSTM encoding, It is the vector representing the heterogeneous spread of VOCs exhaust gas concentration.

[0064] More specifically, in an embodiment of the present application, the VOCs exhaust gas concentration modulation secondary subunit 15313 is used to: calculate the semantic association between each VOCs exhaust gas concentration local temporal heterodomain modulation coding feature vector in the set of the VOCs exhaust gas concentration local temporal heterodomain modulation coding feature vector and the VOCs exhaust gas concentration heterodomain propagation representation vector to obtain a set of VOCs exhaust gas concentration semantic associations; calculate the semantic space jump value of each VOCs exhaust gas concentration local temporal heterodomain modulation coding feature vector in the set of the VOCs exhaust gas concentration local temporal heterodomain modulation coding feature vector relative to the VOCs exhaust gas concentration heterodomain propagation representation vector to obtain a set of VOCs exhaust gas concentration semantic space jump values; perform multiple gated information transfer modulation on the set of VOCs exhaust gas concentration semantic associations and the set of VOCs exhaust gas concentration semantic space jump values ​​to obtain a set of VOCs exhaust gas concentration transfer modulation weights. Accordingly, the semantic association between each foreign modulation coding feature vector and the foreign propagation representation vector is calculated to represent the semantic association between the transfer features and each foreign modulation coding feature in Poincare space. This determines which local patterns significantly influence the overall trend and variation, resulting in a set of semantic associations for VOC exhaust gas concentrations. Next, the semantic space jump value between each foreign modulation coding feature vector and the foreign propagation representation vector is calculated. By comparing the spatial distance differences between the concentration transfer features and each foreign modulation coding feature in Poincare space, the trend and speed of concentration feature expression over time are evaluated, revealing the temporal dynamics of concentration. Subsequently, the semantic association and semantic space jump values ​​of each foreign modulation coding feature vector are modulated using multiple gated information transfer to obtain the modulation weight of each foreign modulation coding feature vector. This allows the weight of each foreign modulation coding feature vector to be dynamically adjusted based on the semantic association and semantic space jump value, helping to highlight important concentration temporal features and suppress unimportant or anomalous features. This allows for focusing on highly weighted events, achieving more refined and efficient control.

[0065] More specifically, the processing process of the VOCs exhaust gas concentration modulation secondary subunit 15313 can be expressed as follows: ;in, To calculate the square of the one-norm of a vector, is the inverse hyperbolic cosine function, and They are the first, second, and third in the set of semantic relevance of VOCs exhaust gas concentration. and The semantic correlation of VOCs exhaust gas concentration, is the set of semantic relevance of the VOCs exhaust gas concentration, and They are the first, second, and third values ​​in the semantic space jump value set of VOCs exhaust gas concentration. and VOCs exhaust gas concentration semantic space jump value, is the set of semantic space jump values ​​of the VOCs exhaust gas concentration, and are the semantic association weight matrix and the semantic space weight matrix respectively, is matrix multiplication, is the bias vector, yes function, It is a collection of VOCs exhaust gas concentration transfer modulation weights.

[0066] Specifically, the VOCs exhaust gas concentration time series propagation secondary subunit 15314 is used to: calculate the position-weighted sum of the set of VOCs exhaust gas concentration local time series heterogeneous modulation coding feature vectors based on the set of VOCs exhaust gas concentration transfer modulation weights to obtain a VOCs exhaust gas concentration heterogeneous feature significant propagation representation vector; input the VOCs exhaust gas concentration heterogeneous feature significant propagation representation vector and the VOCs exhaust gas concentration heterogeneous propagation representation vector into the residual unit to obtain a VOCs exhaust gas concentration feature significant propagation compensation representation vector; and perform Euclidean space mapping on the VOCs exhaust gas concentration feature significant propagation compensation representation vector to obtain the VOCs exhaust gas concentration time series feature propagation representation vector. That is, based on the transfer modulation weights of each VOCs exhaust gas concentration, the weighted fusion of each VOCs exhaust gas concentration local time series heterogeneous modulation coding feature vector is performed to better characterize and depict the criticality and temporal variation trend of each feature in a more fine-grained manner, thereby obtaining the VOCs exhaust gas concentration heterogeneous feature significant propagation representation vector. That is, the significance of the exhaust gas concentration characteristics and the change pattern in the time series are combined through the calculation method of position weighted summation, thereby ensuring that the generated feature vector can fully map the spatiotemporal properties of the data. Next, the generated significant propagation representation vector and the foreign propagation representation vector are input into the residual unit for residual connection, so as to utilize the jump connection in the residual unit to perform feature compensation on the significant propagation representation vector (emphasizing important features) and the foreign propagation representation vector (overall sequence features), while retaining the original features, so as to better capture the subtle changes and complex patterns in the concentration data, and obtain the VOCs exhaust gas concentration feature significant propagation compensation representation vector. Finally, the VOCs exhaust gas concentration feature significant propagation compensation representation vector is mapped to Euclidean space to obtain the VOCs exhaust gas concentration time series feature propagation representation vector. That is, by mapping the compensation representation vector to Euclidean space, the high-dimensional feature significant propagation compensation representation vector is reduced to a Euclidean space that is easier to handle and interpret, which facilitates subsequent data analysis and interpretation.

[0067] More specifically, the processing process of the VOCs exhaust gas concentration time series propagation secondary subunit 15314 can be expressed as follows: ;in, yes The corresponding VOCs exhaust gas concentration transfer modulation weight, is the number of vectors in the set of VOCs exhaust gas concentration local time series heterogeneous modulation coding feature vectors, is the vector representing the significant propagation of VOCs exhaust gas concentration heterogeneous characteristics, and are the third weight matrix and the fourth weight matrix respectively, is the Euclidean space mapping operation, is the significant propagation compensation representation vector of the VOCs exhaust gas concentration characteristic.

[0068] Specifically, the optimization instruction generation subunit 1532 is configured to obtain an optimization instruction based on the VOCs exhaust gas concentration time series characteristic propagation representation vector, wherein the optimization instruction includes a recommended fan speed value for the next time point. Specifically, in an embodiment of the present application, the optimization instruction generation subunit is configured to input the VOCs exhaust gas concentration time series characteristic propagation representation vector into a decoder-based fan speed dynamic optimization module to obtain the optimization instruction, wherein the optimization instruction includes a recommended fan speed value for the next time point.

[0069] More specifically, in an embodiment of the present application, the optimization instruction generation subunit is used to: multiply the decoding weight matrix of the decoder with the VOCs exhaust gas concentration time series feature propagation representation vector to obtain the decoded VOCs exhaust gas concentration time series feature propagation representation vector, and accumulate and sum all the eigenvalues ​​of the decoded VOCs exhaust gas concentration time series feature propagation representation vector to obtain the recommended value of the fan speed at the next time point. That is, the VOCs exhaust gas concentration time series feature propagation representation obtained by multi-gating the set of the VOCs exhaust gas concentration local time series implicit pattern feature vectors is used for decoding processing, so as to intelligently obtain the recommended value of the fan speed at the next time point. In this way, the VOCs exhaust gas concentration can be monitored in real time, and the complex pattern of the VOCs exhaust gas concentration can be identified. The fan speed can be intelligently adjusted according to the current and predicted exhaust gas conditions, thereby improving the adaptability to different working conditions.

[0070] It should be understood that here, when each VOCs exhaust gas concentration local temporal implicit pattern feature vector in the set of VOCs exhaust gas concentration local temporal implicit pattern feature vectors respectively represents the temporal correlation characteristics of the VOCs exhaust gas concentration data in the local phase space, when performing feature dynamic propagation based on a cross-domain multiple gate structure, the cross-domain gate structure differences caused by the temporal feature distribution differences in each local phase space are taken into account, and it is expected to improve the semantically consistent aggregation expression effect of the VOCs exhaust gas concentration temporal feature propagation representation vector obtained through feature dynamic propagation.

[0071] Preferably, inputting the VOCs exhaust gas concentration time series characteristic propagation representation vector into a decoder-based fan speed dynamic optimization module to obtain an optimization instruction includes:

[0072] The feature set of the VOCs exhaust gas concentration time series feature propagation representation vector is subjected to feature clustering to obtain a VOCs exhaust gas concentration time series feature propagation within-class feature set and a VOCs exhaust gas concentration time series feature propagation outside-class feature set, namely: ;in, is the intra-class feature set of the VOCs exhaust gas concentration time series feature propagation, is the characteristic value of each position in the feature set within the VOCs exhaust gas concentration time series feature propagation class, is the characteristic value of each position in the out-of-class feature set of the VOCs exhaust gas concentration time series feature propagation;

[0073] Calculate the ratio of the number of eigenvalues ​​in the feature set within the VOCs exhaust gas concentration time series feature propagation class to the number of eigenvalues ​​in the feature set of the VOCs exhaust gas concentration time series feature propagation representation vector, that is: ;in, is the number of eigenvalues ​​in the feature set within the VOCs exhaust gas concentration time series feature propagation class, is the number of eigenvalues ​​of the feature set of the VOCs exhaust gas concentration time series feature propagation representation vector, yes and The ratio of

[0074] Calculate the sum of the absolute values ​​of all feature values ​​in the feature set of the VOCs exhaust gas concentration time series feature propagation class The power and the sum of the absolute values ​​of all eigenvalues ​​in the feature set of the VOCs exhaust gas concentration time series feature propagation representation vector The power ratio is used to obtain the modulation weight of the VOCs exhaust gas concentration time series characteristic propagation, that is: ;in, is the modulation weight of the propagation of the time series characteristic of the VOCs exhaust gas concentration;

[0075] Calculate the sum of squares of all eigenvalues ​​in the feature set of the VOCs exhaust gas concentration time series feature propagation class The power and the VOCs exhaust gas concentration time series feature propagation represent the sum of the squares of all eigenvalues ​​in the feature set of the vector The power ratio is used to obtain the propagation harmonic weight of the VOCs exhaust gas concentration time series characteristics, that is: ;in, is the propagation harmonic weight of the time series characteristic of the VOCs exhaust gas concentration;

[0076] For each eigenvalue in the feature set within the VOCs exhaust gas concentration time series feature propagation class, the product of the eigenvalue and the VOCs exhaust gas concentration time series feature propagation harmonic weight is calculated, and then the product is added to the VOCs exhaust gas concentration time series feature propagation modulation weight to obtain the optimized eigenvalue, that is: ;in, It is the optimized characteristic value of each position of the feature set within the VOCs exhaust gas concentration time series feature propagation class;

[0077] For each eigenvalue in the out-of-class feature set of the VOCs exhaust gas concentration time series feature propagation, the product of the eigenvalue and the VOCs exhaust gas concentration time series feature propagation adjustment weight is calculated to obtain the optimized eigenvalue, that is: ;in, is the optimized characteristic value of each position of the out-of-class characteristic set of the VOCs exhaust gas concentration time series characteristic propagation;

[0078] The optimized VOCs exhaust gas concentration time series feature propagation representation vector composed of the optimized feature values ​​of the feature set within the VOCs exhaust gas concentration time series feature propagation class and the feature set outside the VOCs exhaust gas concentration time series feature propagation class is input into the fan speed dynamic optimization module based on the decoder to obtain the optimization instruction.

[0079] Therefore, while performing feature clustering on the VOCs exhaust gas concentration time series feature propagation representation vector, the interaction description of the key feature information of the VOCs exhaust gas concentration time series feature propagation representation vector in the clustering process is carried out, and the geometric equivariant topology of the feature is constructed by low-rank harmonic modulation based on the clustering feature of the VOCs exhaust gas concentration time series feature propagation representation vector and the equivariance of the feature as a whole, so as to obtain the graphical distribution translation and rotation symmetry of the clustering feature of the VOCs exhaust gas concentration time series feature propagation representation vector relative to the feature as a whole, thereby introducing geometric message passing into the feature expression of the VOCs exhaust gas concentration time series feature propagation representation vector and realizing the cluster mapping symmetry of the VOCs exhaust gas concentration time series feature propagation representation vector through irreducible low-rank order coefficient manipulation, thereby improving the clustering-based feature representation consistency of the VOCs exhaust gas concentration time series feature propagation representation vector, thereby improving the accuracy of the optimization instruction obtained by the VOCs exhaust gas concentration time series feature propagation representation vector input to the decoder-based fan speed dynamic optimization module.

[0080] In summary, the controller 150 is explained, which uses a VOCs exhaust gas concentration detector to collect a time series data set of VOCs exhaust gas concentration data at a predetermined sampling frequency, and transmits it to the fan speed control center through a wireless communication module. In the fan speed control center, a data analysis and encoding method based on deep learning is used to reconstruct the phase space and extract local time series implicit information from the time series data set of VOCs exhaust gas concentration data. In this way, the dynamic propagation representation obtained by multi-gated propagation based on the local time series implicit pattern characteristics of each VOCs exhaust gas concentration is used to intelligently obtain the recommended fan speed value for the next time point. In this way, the VOCs exhaust gas concentration can be monitored in real time, and the complex pattern of the VOCs exhaust gas concentration can be identified. The fan speed can be intelligently adjusted according to the current and predicted exhaust gas conditions, thereby improving the adaptability to different working conditions.

[0081] In summary, the organic waste gas purification and treatment system 100 based on the embodiment of the present application is explained, which uses a filter to perform preliminary filtration of the waste gas, and then uses an ozone catalyst to catalyze ozone to perform preliminary oxidation of volatile organic compounds (VOCs) in the filtered waste gas. Subsequently, ultraviolet radiation nanophotocatalysts are used to further oxidize these preliminarily oxidized VOCs waste gases, converting them into small molecular organic matter that is more easily biodegradable, and the concentration of VOCs in the waste gas is monitored in real time, and the fan speed is dynamically adjusted according to the monitoring data to ensure the purification efficiency and the optimal operating state of the system.

[0082] As described above, the organic waste gas purification and treatment system 100 according to the embodiment of the present application can be implemented in various wireless terminals, such as a server with an organic waste gas purification and treatment algorithm. In one possible implementation, the organic waste gas purification and treatment system 100 according to the embodiment of the present application can be integrated into the wireless terminal as a software module and / or hardware module. For example, the organic waste gas purification and treatment system 100 can be a software module in the operating system of the wireless terminal, or it can be an application developed for the wireless terminal; of course, the organic waste gas purification and treatment system 100 can also be one of the many hardware modules of the wireless terminal.

[0083] Alternatively, in another example, the organic waste gas purification and treatment system 100 and the wireless terminal may also be separate devices, and the organic waste gas purification and treatment system 100 may be connected to the wireless terminal through a wired and / or wireless network and transmit interactive information in accordance with an agreed data format.

[0084] Figure 5 Flowchart of the organic waste gas purification method according to the embodiment of the present application. Figure 5 As shown, the organic waste gas purification method according to the embodiment of the present application is used to execute the above-mentioned organic waste gas purification system, and the organic waste gas purification method includes: S110, using a VOCs waste gas concentration detector to collect a time series data set of VOCs waste gas concentration data at a predetermined sampling frequency, and transmitting the time series data set of the VOCs waste gas concentration data to the fan speed control center through a wireless communication module; S120, at the fan speed control center, performing phase space reconstruction and local time series extraction on the time series data set of the VOCs waste gas concentration data to obtain a set of local time series implicit pattern features of the VOCs waste gas concentration; S130, performing multi-gated feature dynamic propagation on the set of local time series implicit pattern features of the VOCs waste gas concentration to obtain a VOCs waste gas concentration time series feature propagation representation, and based on the VOCs waste gas concentration time series feature propagation representation, obtaining an optimization instruction, the optimization instruction including a recommended value of the fan speed at the next time point.

[0085] Here, those skilled in the art will understand that the specific operations of each step in the above-mentioned organic waste gas purification method have been described in detail above. Figures 1 to 4 The description of the organic waste gas purification treatment system has been introduced in detail, and therefore, its repeated description will be omitted.

[0086] While various implementations of the present disclosure have been described above, the foregoing description is intended to be illustrative and not exhaustive. The disclosure is not limited to the disclosed implementations, and numerous modifications and variations will be apparent to those skilled in the art without departing from the scope and spirit of the described implementations. The terminology used herein is selected to best explain the principles of the implementations, their practical applications, or improvements to existing technologies, or to enable others skilled in the art to understand the various implementations disclosed herein.

Claims

1. An organic waste gas purification and treatment system, comprising: A fan; a filter for filtering the exhaust gas to obtain filtered exhaust gas; a preliminary oxidation module for using an ozone catalyst to catalyze ozone to perform preliminary oxidation of VOCs in the filtered exhaust gas to obtain preliminary oxidized VOCs exhaust gas; a secondary oxidation module for using ultraviolet radiation nano-photocatalysts to perform secondary oxidation on the preliminary oxidized VOCs exhaust gas to obtain biodegradable small molecular organic matter; characterized in that the organic waste gas purification and treatment system further includes: a controller for real-time monitoring of VOCs exhaust gas concentration data in the exhaust gas during the exhaust gas purification process and adjusting the speed of the fan in real time; wherein the controller includes: A VOCs exhaust gas concentration collection and transmission unit is used to collect a time series data set of VOCs exhaust gas concentration data at a predetermined sampling frequency using a VOCs exhaust gas concentration detector provided at the exhaust gas discharge port, and transmit the time series data set of VOCs exhaust gas concentration data to a fan speed control center via a wireless communication module; A VOCs exhaust gas concentration time series analysis unit is configured to perform phase space reconstruction on the time series data set of the VOCs exhaust gas concentration data in the fan speed control center to obtain a set of VOCs exhaust gas concentration data time series subsequences, and input each VOCs exhaust gas concentration data time series subsequence in the set of VOCs exhaust gas concentration data time series subsequences into a sequence encoder based on a forward LSTM model to obtain a set of VOCs exhaust gas concentration local time series implicit pattern feature vectors; A VOCs exhaust gas concentration time series propagation unit includes: a VOCs exhaust gas concentration time series dynamic propagation subunit, configured to perform multi-gated feature dynamic propagation on a set of VOCs exhaust gas concentration local time series implicit pattern features to obtain a VOCs exhaust gas concentration time series feature propagation representation, and obtain an optimization instruction based on the VOCs exhaust gas concentration time series feature propagation representation, wherein the optimization instruction includes a recommended fan speed value at a next time point; The VOCs exhaust gas concentration time series dynamic propagation subunit includes: A VOCs exhaust gas concentration feature mapping secondary subunit is used to map each VOCs exhaust gas concentration local time series implicit pattern feature vector in the set of VOCs exhaust gas concentration local time series implicit pattern feature vectors to a Poincare space to obtain a set of VOCs exhaust gas concentration local time series heterogeneous modulation coding feature vectors; A VOCs exhaust gas concentration feature propagation secondary subunit is used to perform forward LSTM encoding on the set of VOCs exhaust gas concentration local time series heterogeneous modulation coding feature vectors to obtain a VOCs exhaust gas concentration heterogeneous propagation representation vector; A VOCs exhaust gas concentration modulation secondary subunit is configured to calculate, based on the VOCs exhaust gas concentration heterogeneous propagation representation vector, a weight value of a set of VOCs exhaust gas concentration local temporal heterogeneous modulation coding feature vectors to obtain a set of VOCs exhaust gas concentration transfer modulation weights; The VOCs exhaust gas concentration time series propagation secondary sub-unit is used to perform residual processing and mapping on the set of VOCs exhaust gas concentration local time series heterodomain modulation coding feature vectors and the VOCs exhaust gas concentration heterodomain propagation representation vector based on the set of VOCs exhaust gas concentration transfer modulation weights to obtain the VOCs exhaust gas concentration time series feature propagation representation vector.

2. The organic waste gas purification system according to claim 1, characterized in that: The VOCs exhaust gas concentration time series transmission unit further includes: The optimization instruction generation subunit is used to obtain an optimization instruction based on the VOCs exhaust gas concentration time series characteristic propagation representation vector, and the optimization instruction includes a recommended value of the fan speed at the next time point.

3. The organic waste gas purification system according to claim 2, characterized in that: The VOCs waste gas concentration modulation secondary subunit is used to: Calculating the semantic association between each VOCs exhaust gas concentration local time series heterogeneous modulation coding feature vector in the set of VOCs exhaust gas concentration local time series heterogeneous modulation coding feature vectors and the VOCs exhaust gas concentration heterogeneous propagation representation vector to obtain a set of VOCs exhaust gas concentration semantic associations; Calculating the semantic space jump value of each VOCs exhaust gas concentration local time series heterogeneous modulation coding feature vector in the set of the VOCs exhaust gas concentration local time series heterogeneous modulation coding feature vector relative to the VOCs exhaust gas concentration heterogeneous propagation representation vector to obtain a set of VOCs exhaust gas concentration semantic space jump values; Multiple gated information transfer modulation is performed on the set of VOCs exhaust gas concentration semantic associations and the set of VOCs exhaust gas concentration semantic space jump values ​​to obtain the set of VOCs exhaust gas concentration transfer modulation weights.

4. The organic waste gas purification system according to claim 3, characterized in that: The VOCs exhaust gas concentration time series propagation secondary sub-unit is used to: Based on the set of VOCs exhaust gas concentration transfer modulation weights, calculating the position-weighted sum of the set of VOCs exhaust gas concentration local time series heterogeneous modulation coding feature vectors to obtain a VOCs exhaust gas concentration heterogeneous feature significant propagation representation vector; Inputting the VOCs exhaust gas concentration heterogeneous feature significant propagation representation vector and the VOCs exhaust gas concentration heterogeneous propagation representation vector into a residual unit to obtain a VOCs exhaust gas concentration feature significant propagation compensation representation vector; The VOCs exhaust gas concentration characteristic significant propagation compensation representation vector is mapped into Euclidean space to obtain the VOCs exhaust gas concentration time series characteristic propagation representation vector.

5. The organic waste gas purification system according to claim 4, characterized in that: The optimization instruction generation subunit is used to: input the VOCs exhaust gas concentration time series feature propagation representation vector into the decoder-based fan speed dynamic optimization module to obtain the optimization instruction, and the optimization instruction includes the recommended value of the fan speed at the next time point.

6. The organic waste gas purification system according to claim 5, characterized in that: The optimization instruction generation sub-unit is used to: multiply the decoding weight matrix of the decoder with the VOCs exhaust gas concentration time series characteristic propagation representation vector to obtain the decoded VOCs exhaust gas concentration time series characteristic propagation representation vector, and accumulate and sum all the eigenvalues ​​of the decoded VOCs exhaust gas concentration time series characteristic propagation representation vector to obtain the recommended value of the fan speed at the next time point.

7. A method for purifying organic waste gas, applied to the organic waste gas purification system according to claim 1, characterized in that: The organic waste gas purification method comprises: Using a VOCs exhaust gas concentration detector to collect a time series data set of VOCs exhaust gas concentration data at a predetermined sampling frequency, and transmitting the time series data set of VOCs exhaust gas concentration data to a fan speed control center through a wireless communication module; At the fan speed control center, phase space reconstruction and local time series extraction are performed on the time series data set of the VOCs exhaust gas concentration data to obtain a set of implicit pattern features of the local time series of the VOCs exhaust gas concentration; Multi-gated feature dynamic propagation is performed on the set of local time series implicit pattern features of the VOCs exhaust gas concentration to obtain a VOCs exhaust gas concentration time series feature propagation representation, and based on the VOCs exhaust gas concentration time series feature propagation representation, an optimization instruction is obtained, and the optimization instruction includes a recommended value of the fan speed at the next time point.

Images