Greenhouse gas release monitoring method and system in sewage treatment process

Abstract

The invention discloses a greenhouse gas release monitoring method and system in a sewage treatment process, and the method comprises the steps: obtaining greenhouse gas concentration data and environment data of a monitoring point in each process stage, and generating multi-modal time sequence data through a data fusion strategy; inputting the data into a pre-trained space-time diagram neural network model to extract conversion features; constructing a causal graph model of greenhouse gas emission in each process stage based on the features, and calculating contribution degrees; when the contribution degree of a certain process stage exceeds a threshold value, sequencing and weighted fusion are carried out on greenhouse gas concentration data to obtain a comprehensive emission intensity sequence; and calculating the change gradient of the sequence in the sliding time window, and generating an early warning signal when the gradient exceeds a preset threshold value. According to the invention, multi-source data fusion and intelligent analysis are realized, an abnormal discharge process link can be accurately identified, the monitoring accuracy and timeliness are significantly improved, and a reliable basis is provided for emission reduction optimization in a sewage treatment process.

Description

Technical Field

[0001] This invention belongs to the field of gas monitoring technology, and in particular relates to a method and system for monitoring greenhouse gas emissions during wastewater treatment processes. Background Technology

[0002] With the continuous expansion of wastewater treatment scale and the increasing complexity of processes, the emission of greenhouse gases (such as CH4, CO2, and N2O) generated during wastewater treatment has gradually attracted widespread attention. These gases not only directly exacerbate global climate change but may also reflect operational anomalies or inefficiencies in the treatment processes.

[0003] Currently, monitoring greenhouse gas emissions during wastewater treatment mainly employs traditional methods, such as periodically sampling using single gas sensors installed at specific locations, or analyzing gas composition through offline laboratory testing. However, these methods have significant limitations: First, monitoring points are sparse, resulting in insufficient data representativeness and making it difficult to comprehensively reflect the emission characteristics of different process stages; second, they only focus on the gas concentration itself, failing to integrate it with process parameters (such as temperature, pH, and dissolved oxygen) for in-depth analysis; third, data analysis methods are simplistic, relying heavily on threshold alarms or statistical mean comparisons, lacking in-depth analysis of emission sources, transmission paths, and causal relationships; and fourth, response is delayed, failing to achieve real-time early warning and rapid control.

[0004] Furthermore, existing technologies generally overlook the differences in the global warming potential (GWP) of various greenhouse gases and lack a scientific assessment of the overall emission intensity; they also fail to effectively incorporate the spatiotemporal characteristics of wastewater treatment processes, resulting in insufficient ability to trace the sources of abnormal emission events. Therefore, developing a greenhouse gas monitoring method capable of multi-source data fusion, intelligent feature extraction, and dynamic causal inference is of great significance for improving the environmental benefits and operational efficiency of wastewater treatment processes. Summary of the Invention

[0005] This invention provides a method and system for monitoring greenhouse gas emissions during wastewater treatment processes, which addresses the technical problem of insufficient source tracing capabilities for abnormal emission events due to the failure to effectively integrate the spatiotemporal characteristics of wastewater treatment processes.

[0006] In a first aspect, the present invention provides a method for monitoring greenhouse gas emissions during a wastewater treatment process, comprising: Greenhouse gas concentration data at various monitoring points during different process stages are acquired, along with environmental data of the areas where the different process stages are located. The greenhouse gas concentration data and environmental data are then fused according to a preset data fusion strategy to obtain multimodal time-series data. The multimodal time series data is input into a pre-trained spatiotemporal graph neural network model, and the spatiotemporal graph neural network model outputs transformation features corresponding to the multimodal time series data. Based on the aforementioned transformation characteristics, a causal graph model of greenhouse gas emissions at each process stage is constructed, and the contribution of each process stage to the total emissions is calculated based on the causal graph model. Determine whether the contribution of each process stage to the total emissions is greater than a preset contribution threshold; If the contribution of a certain process stage to the total emissions is greater than a preset contribution threshold, then the greenhouse gas concentration data of each greenhouse gas in the certain process stage are sorted based on the location relationship of the monitoring points to obtain at least one greenhouse gas concentration data sequence. Based on the weight of each gas in the global warming potential, the concentration data sequences of at least one greenhouse gas are weighted and fused according to a preset weighting strategy to obtain a comprehensive greenhouse gas emission intensity sequence. Calculate the gradient of the emission intensity sequence within a sliding time window, and determine whether the gradient is greater than a preset threshold. If the emissions exceed a preset threshold, an early warning signal for abnormal greenhouse gas emissions at a certain process stage is generated and sent to the user terminal.

[0007] Secondly, the present invention provides a greenhouse gas emission monitoring system for a wastewater treatment process, comprising: The acquisition module is configured to acquire greenhouse gas concentration data at various monitoring points in different process stages, as well as environmental data of the areas where the different process stages are located, and to fuse the greenhouse gas concentration data and environmental data according to a preset data fusion strategy to obtain multimodal time series data. The output module is configured to input the multimodal time series data into a pre-trained spatiotemporal graph neural network model, and the spatiotemporal graph neural network model outputs transformation features corresponding to the multimodal time series data. The calculation module is configured to construct a causal graph model of greenhouse gas emissions at each process stage based on the transformation characteristics, and calculate the contribution of each process stage to the total emissions based on the causal graph model. The first judgment module is configured to determine whether the contribution of each process stage to the total emissions is greater than a preset contribution threshold. The sorting module is configured to sort the greenhouse gas concentration data of each process stage based on the location relationship of the monitoring points if the contribution of a certain process stage to the total emissions is greater than a preset contribution threshold, thereby obtaining at least one greenhouse gas concentration data sequence. The fusion module is configured to perform weighted fusion of the at least one greenhouse gas concentration data sequence based on the weight of each gas in the global warming potential value and according to a preset weighting strategy to obtain a comprehensive greenhouse gas emission intensity sequence. The second judgment module is configured to calculate the change gradient of the emission intensity sequence within a sliding time window and determine whether the change gradient is greater than a preset threshold. The generation module is configured to generate an early warning signal for abnormal greenhouse gas emissions in a certain process stage if the emissions exceed a preset threshold, and then send the early warning signal to the user terminal.

[0008] Thirdly, an electronic device is provided, comprising: at least one processor, and a memory communicatively connected to the at least one processor, wherein the memory stores instructions executable by the at least one processor, the instructions being executed by the at least one processor to enable the at least one processor to perform the steps of the greenhouse gas emission monitoring method for a wastewater treatment process according to any embodiment of the present invention.

[0009] Fourthly, the present invention also provides a computer-readable storage medium having a computer program stored thereon, wherein when the program instructions are executed by a processor, the processor performs the steps of the greenhouse gas emission monitoring method for a wastewater treatment process according to any embodiment of the present invention.

[0010] This application presents a method and system for monitoring greenhouse gas emissions during wastewater treatment processes. Through the organic combination of multi-source data fusion, intelligent feature extraction, and causal reasoning, it achieves precise monitoring and intelligent early warning of greenhouse gas emissions during wastewater treatment. Specifically, by fusing greenhouse gas concentrations and environmental parameters from multiple process stages, a unified multimodal time-series data is constructed, effectively improving data integrity and consistency. Using a spatiotemporal graph neural network to extract transformation features, the spatiotemporal evolution of gas emissions can be captured. Furthermore, a causal graph model is used to quantify the contribution of each process stage to total emissions, achieving precise location of emission sources. Based on weighted fusion and gradient analysis of global warming potential values, abnormal emission trends can be sensitively identified. Finally, a multi-level early warning mechanism provides timely control measures for operators. This method significantly improves the accuracy and timeliness of monitoring, overcoming the shortcomings of traditional methods that rely on single indicators and have large time lags, providing reliable technical support for achieving low-carbon operation of wastewater treatment processes. Attached Figure Description

[0011] To more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings used in the following description of the embodiments will be briefly introduced. Obviously, the drawings described below are some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0012] Figure 1 A flowchart illustrating a method for monitoring greenhouse gas emissions during a wastewater treatment process, as provided in an embodiment of the present invention; Figure 2 This is a structural block diagram of a greenhouse gas emission monitoring system for a wastewater treatment process, provided in an embodiment of the present invention. Figure 3 This is a schematic diagram of the structure of an electronic device provided in an embodiment of the present invention. Detailed Implementation

[0013] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0014] Please see Figure 1 The diagram shows a flowchart of a method for monitoring greenhouse gas emissions during a wastewater treatment process according to this application.

[0015] like Figure 1 As shown, the method for monitoring greenhouse gas emissions during wastewater treatment specifically includes the following steps: Step S101: Obtain greenhouse gas concentration data at each monitoring point in different process stages, as well as environmental data of the area where the different process stages are located, and fuse the greenhouse gas concentration data and environmental data according to a preset data fusion strategy to obtain multimodal time series data.

[0016] In this step, greenhouse gas concentration data and environmental data at different sampling frequencies are acquired, wherein the environmental data includes temperature, humidity, and air pressure parameters. Using the data with the highest sampling frequency as a benchmark, the low-frequency data is aligned in the time dimension using a moving window linear interpolation method. For the aligned data sequence, a data completion algorithm based on spatiotemporal correlation is used to fill in the missing values. The data completion algorithm utilizes the data correlation of adjacent monitoring points in the same process stage and the data continuity of the same monitoring point at different times. The supplemented greenhouse gas concentration data and environmental data are merged according to timestamps to generate a multimodal time series data matrix with a unified time base; The multimodal time series data matrix is ​​standardized to eliminate the influence of different units, resulting in the final multimodal time series data.

[0017] Step S102: Input the multimodal time series data into a pre-trained spatiotemporal graph neural network model, and the spatiotemporal graph neural network model outputs transformation features corresponding to the multimodal time series data.

[0018] In this step, the spatiotemporal graph neural network model adopts a structure that combines graph convolutional networks (GCN) and long short-term memory networks (LSTM) to simultaneously capture spatial dependencies and temporal dynamics.

[0019] Step S103: Based on the transformation characteristics, construct a causal graph model of greenhouse gas emissions at each process stage, and calculate the contribution of each process stage to the total emissions based on the causal graph model.

[0020] In this step, key node features are extracted from the transformation features. These key nodes include greenhouse gas concentration, temperature, pH value, dissolved oxygen concentration, and hydraulic residence time at each process stage. Calculate the time delay mutual information between the features of each key node to determine whether there is a causal relationship between the nodes and the direction of causation. An initial causal network skeleton is constructed based on the time delay mutual information. The PC algorithm is used to optimize the initial skeleton, remove redundant edges, and orient causal arrows. We introduce prior knowledge constraints on wastewater treatment processes, injecting known process flow directions and biochemical reaction paths as constraints into the cause-effect graph model. A score-based structure learning algorithm is adopted, using the Bayesian information criterion as the score function, to optimize the search for causal graph structures; The output is a weighted directed acyclic graph as a causal graph model, where nodes represent process parameters or gas concentrations, edges represent causal relationships, and weights represent causal strengths, thus obtaining a causal graph model of greenhouse gas emissions at each process stage.

[0021] It should be noted that the expression for calculating the contribution of each process stage to the total emissions is as follows: , In the formula, This represents the contribution of the i-th process segment. For its emissions, This represents total emissions.

[0022] Step S104: Determine whether the contribution of each process stage to the total emissions is greater than the preset contribution threshold.

[0023] Step S105: If the contribution of a certain process stage to the total emissions is greater than a preset contribution threshold, then the greenhouse gas concentration data of each process stage are sorted based on the location relationship of the monitoring points to obtain at least one greenhouse gas concentration data sequence.

[0024] Step S106: Based on the weight of each gas in the global warming potential, the concentration data sequence of at least one greenhouse gas is weighted and fused according to a preset weighting strategy to obtain a comprehensive greenhouse gas emission intensity sequence.

[0025] In this step, the expression for the weighted strategy is:

[0026] In the formula, Let be the overall emission intensity at time t. For each gas, the GWP weighting coefficient is... Let t be the concentration value at time t.

[0027] Step S107: Calculate the gradient of the emission intensity sequence within the sliding time window, and determine whether the gradient is greater than a preset threshold.

[0028] In this step, the formula for calculating the gradient is: , In the formula, Let be the gradient of change at time t. Let be the overall emission intensity at time t. This represents the time window step size.

[0029] Step S108: If the emissions exceed a preset threshold, an early warning signal for abnormal greenhouse gas emissions in a certain process stage is generated and the early warning signal is sent to the user terminal.

[0030] In summary, the method presented in this application achieves precise monitoring and intelligent early warning of greenhouse gas releases during wastewater treatment processes through the organic combination of multi-source data fusion, intelligent feature extraction, and causal reasoning. Specifically, by fusing greenhouse gas concentrations and environmental parameters from multiple process stages to construct unified multimodal time-series data, the completeness and consistency of the data are effectively improved. Transformation features are extracted using a spatiotemporal graph neural network, capturing the spatiotemporal evolution of gas emissions. Furthermore, the contribution of each process stage to total emissions is quantified through a causal graph model, enabling precise location of emission sources. Weighted fusion and gradient analysis based on global warming potential values ​​can sensitively identify abnormal emission trends. Finally, a multi-level early warning mechanism provides timely control measures for operators. This method significantly improves the accuracy and timeliness of monitoring, overcoming the shortcomings of traditional methods that rely on single indicators and have large time lags, providing reliable technical support for achieving low-carbon operation of wastewater treatment processes.

[0031] Please see Figure 2 The diagram shows a structural block diagram of a greenhouse gas emission monitoring system for a wastewater treatment process according to this application.

[0032] like Figure 2 As shown, the greenhouse gas emission monitoring system 200 includes an acquisition module 210, an output module 220, a calculation module 230, a first judgment module 240, a sorting module 250, a fusion module 260, a second judgment module 270, and a generation module 280.

[0033] The acquisition module 210 is configured to acquire greenhouse gas concentration data at various monitoring points during different process stages, as well as environmental data of the areas where the different process stages are located, and fuse the greenhouse gas concentration data and environmental data according to a preset data fusion strategy to obtain multimodal time-series data; the output module 220 is configured to input the multimodal time-series data into a pre-trained spatiotemporal graph neural network model, and the spatiotemporal graph neural network model outputs transformation features corresponding to the multimodal time-series data; the calculation module 230 is configured to construct a causal graph model of greenhouse gas emissions at each process stage based on the transformation features, and calculate the contribution of each process stage to the total emissions according to the causal graph model; the first judgment module 240 is configured to judge whether the contribution of each process stage to the total emissions is greater than a preset contribution threshold. The sorting module 250 is configured to sort the greenhouse gas concentration data of each process stage based on the location relationship of the monitoring points if the contribution of a certain process stage to the total emissions is greater than a preset contribution threshold, thereby obtaining at least one greenhouse gas concentration data sequence; the fusion module 260 is configured to perform weighted fusion of the at least one greenhouse gas concentration data sequence according to a preset weighting strategy based on the weight of each gas in the global warming potential, thereby obtaining a comprehensive greenhouse gas emission intensity sequence; the second judgment module 270 is configured to calculate the change gradient of the emission intensity sequence within a sliding time window and determine whether the change gradient is greater than a preset threshold; the generation module 280 is configured to generate an early warning signal for abnormal greenhouse gas emissions in the certain process stage if the gradient is greater than the preset threshold, and send the early warning signal to the user terminal.

[0034] It should be understood that Figure 2 The modules and references described in the document Figure 1 The steps described in the text correspond to those in the method described above. Therefore, the operations, features, and corresponding technical effects described above also apply to the method described in the text. Figure 2 The various modules in the document will not be described in detail here.

[0035] In other embodiments, the present invention also provides a computer-readable storage medium having a computer program stored thereon, wherein when the program instructions are executed by a processor, the processor performs the greenhouse gas emission monitoring method for the wastewater treatment process in any of the above method embodiments. In one embodiment, the computer-readable storage medium of the present invention stores computer-executable instructions, which are configured as follows: Greenhouse gas concentration data at various monitoring points during different process stages are acquired, along with environmental data of the areas where the different process stages are located. The greenhouse gas concentration data and environmental data are then fused according to a preset data fusion strategy to obtain multimodal time-series data. The multimodal time series data is input into a pre-trained spatiotemporal graph neural network model, and the spatiotemporal graph neural network model outputs transformation features corresponding to the multimodal time series data. Based on the aforementioned transformation characteristics, a causal graph model of greenhouse gas emissions at each process stage is constructed, and the contribution of each process stage to the total emissions is calculated based on the causal graph model. Determine whether the contribution of each process stage to the total emissions is greater than a preset contribution threshold; If the contribution of a certain process stage to the total emissions is greater than a preset contribution threshold, then the greenhouse gas concentration data of each greenhouse gas in the certain process stage are sorted based on the location relationship of the monitoring points to obtain at least one greenhouse gas concentration data sequence. Based on the weight of each gas in the global warming potential, the concentration data sequences of at least one greenhouse gas are weighted and fused according to a preset weighting strategy to obtain a comprehensive greenhouse gas emission intensity sequence. Calculate the gradient of the emission intensity sequence within a sliding time window, and determine whether the gradient is greater than a preset threshold. If the emissions exceed a preset threshold, an early warning signal for abnormal greenhouse gas emissions at a certain process stage is generated and sent to the user terminal.

[0036] Computer-readable storage media may include a stored program area and a stored data area, wherein the stored program area may store an operating system and an application program required for at least one function; the stored data area may store data created based on the use of the greenhouse gas emission monitoring system for the wastewater treatment process, etc. Furthermore, the computer-readable storage medium may include high-speed random access memory, and may also include memory, such as at least one disk storage device, flash memory device, or other non-volatile solid-state storage device. In some embodiments, the computer-readable storage medium may optionally include memory remotely configured relative to a processor, which can be connected to the greenhouse gas emission monitoring system for the wastewater treatment process via a network. Examples of such networks include, but are not limited to, the Internet, corporate intranets, local area networks, mobile communication networks, and combinations thereof.

[0037] Figure 3 This is a schematic diagram of the structure of the electronic device provided in the embodiment of the present invention, such as... Figure 3As shown, the device includes a processor 310 and a memory 320. The electronic device may also include an input device 330 and an output device 340. The processor 310, memory 320, input device 330, and output device 340 can be connected via a bus or other means. Figure 3 Taking a bus connection as an example, the memory 320 is the computer-readable storage medium described above. The processor 310 executes various server functions and data processing by running non-volatile software programs, instructions, and modules stored in the memory 320, thereby implementing the greenhouse gas emission monitoring method for the wastewater treatment process described in the above embodiment. The input device 330 can receive input digital or character information and generate key signal inputs related to user settings and function control of the greenhouse gas emission monitoring system for the wastewater treatment process. The output device 340 may include a display screen or other display device.

[0038] The aforementioned electronic device can execute the method provided in the embodiments of the present invention, and has the corresponding functional modules and beneficial effects for executing the method. Technical details not described in detail in this embodiment can be found in the method provided in the embodiments of the present invention.

[0039] In one embodiment, the above-described electronic device is applied to a greenhouse gas emission monitoring system for a wastewater treatment process, serving as a client, and includes: at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, the instructions being executed by the at least one processor to enable the at least one processor to: Greenhouse gas concentration data at various monitoring points during different process stages are acquired, along with environmental data of the areas where the different process stages are located. The greenhouse gas concentration data and environmental data are then fused according to a preset data fusion strategy to obtain multimodal time-series data. The multimodal time series data is input into a pre-trained spatiotemporal graph neural network model, and the spatiotemporal graph neural network model outputs transformation features corresponding to the multimodal time series data. Based on the aforementioned transformation characteristics, a causal graph model of greenhouse gas emissions at each process stage is constructed, and the contribution of each process stage to the total emissions is calculated based on the causal graph model. Determine whether the contribution of each process stage to the total emissions is greater than a preset contribution threshold; If the contribution of a certain process stage to the total emissions is greater than a preset contribution threshold, then the greenhouse gas concentration data of each greenhouse gas in the certain process stage are sorted based on the location relationship of the monitoring points to obtain at least one greenhouse gas concentration data sequence. Based on the weight of each gas in the global warming potential, the concentration data sequences of at least one greenhouse gas are weighted and fused according to a preset weighting strategy to obtain a comprehensive greenhouse gas emission intensity sequence. Calculate the gradient of the emission intensity sequence within a sliding time window, and determine whether the gradient is greater than a preset threshold. If the emissions exceed a preset threshold, an early warning signal for abnormal greenhouse gas emissions at a certain process stage is generated and sent to the user terminal.

[0040] Through the above description of the embodiments, those skilled in the art can clearly understand that each embodiment can be implemented by means of software plus necessary general-purpose hardware platforms, and of course, it can also be implemented by hardware. Based on this understanding, the above technical solutions, in essence or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., including several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute the methods of various embodiments or some parts of embodiments.

[0041] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for monitoring greenhouse gas emissions during wastewater treatment, characterized in that, include: Greenhouse gas concentration data at various monitoring points during different process stages are acquired, along with environmental data of the areas where the different process stages are located. The greenhouse gas concentration data and environmental data are then fused according to a preset data fusion strategy to obtain multimodal time-series data. The multimodal time series data is input into a pre-trained spatiotemporal graph neural network model, and the spatiotemporal graph neural network model outputs transformation features corresponding to the multimodal time series data. Based on the aforementioned transformation characteristics, a causal graph model of greenhouse gas emissions at each process stage is constructed, and the contribution of each process stage to the total emissions is calculated based on the causal graph model. Determine whether the contribution of each process stage to the total emissions is greater than a preset contribution threshold; If the contribution of a certain process stage to the total emissions is greater than a preset contribution threshold, then the greenhouse gas concentration data of each greenhouse gas in the certain process stage are sorted based on the location relationship of the monitoring points to obtain at least one greenhouse gas concentration data sequence. Based on the weight of each gas in the global warming potential, the concentration data sequences of at least one greenhouse gas are weighted and fused according to a preset weighting strategy to obtain a comprehensive greenhouse gas emission intensity sequence. Calculate the gradient of the emission intensity sequence within a sliding time window, and determine whether the gradient is greater than a preset threshold. If the emissions exceed a preset threshold, an early warning signal for abnormal greenhouse gas emissions at a certain process stage is generated and sent to the user terminal.

2. The method for monitoring greenhouse gas emissions during wastewater treatment according to claim 1, characterized in that, The process of fusing various greenhouse gas concentration data and environmental data according to a preset data fusion strategy to obtain multimodal time-series data includes: Acquire greenhouse gas concentration data and environmental data at different sampling frequencies, wherein the environmental data includes temperature, humidity, and air pressure parameters; Using the data with the highest sampling frequency as a benchmark, the low-frequency data is aligned in the time dimension using a moving window linear interpolation method. For the aligned data sequence, a data completion algorithm based on spatiotemporal correlation is used to fill in the missing values. The data completion algorithm utilizes the data correlation of adjacent monitoring points in the same process stage and the data continuity of the same monitoring point at different times. The supplemented greenhouse gas concentration data and environmental data are merged according to timestamps to generate a multimodal time series data matrix with a unified time base; The multimodal time series data matrix is ​​standardized to eliminate the influence of different units, resulting in the final multimodal time series data.

3. The method for monitoring greenhouse gas emissions during wastewater treatment according to claim 1, characterized in that, The construction of a causal graph model of greenhouse gas emissions at each process stage based on the transformation characteristics includes: Key node features are extracted from the transformation features, including greenhouse gas concentration, temperature, pH value, dissolved oxygen concentration, and hydraulic residence time at each process stage. Calculate the time delay mutual information between the features of each key node to determine whether there is a causal relationship between the nodes and the direction of causation. An initial causal network skeleton is constructed based on the time delay mutual information. The PC algorithm is used to optimize the initial skeleton, remove redundant edges, and orient causal arrows. We introduce prior knowledge constraints on wastewater treatment processes, injecting known process flow directions and biochemical reaction paths as constraints into the cause-effect graph model. A score-based structure learning algorithm is adopted, using the Bayesian information criterion as the score function, to optimize the search for causal graph structures; The output is a weighted directed acyclic graph as a causal graph model, where nodes represent process parameters or gas concentrations, edges represent causal relationships, and weights represent causal strengths, thus obtaining a causal graph model of greenhouse gas emissions at each process stage.

4. The method for monitoring greenhouse gas emissions during wastewater treatment according to claim 1, characterized in that, The expression for calculating the contribution of each process stage to the total emissions is as follows: ， In the formula, This represents the contribution of the i-th process segment. For its emissions, This represents total emissions.

5. The method for monitoring greenhouse gas emissions during wastewater treatment according to claim 1, characterized in that, The expression for the weighting strategy is: , In the formula, Let be the overall emission intensity at time t. For each gas, the GWP weighting coefficient is... Let t be the concentration value at time t.

6. The method for monitoring greenhouse gas emissions during wastewater treatment according to claim 1, characterized in that, The formula for calculating the gradient of change is: ， In the formula, Let be the gradient of change at time t. Let be the overall emission intensity at time t. This represents the time window step size.

7. A greenhouse gas emission monitoring system for a wastewater treatment process, characterized in that, include: The acquisition module is configured to acquire greenhouse gas concentration data at various monitoring points in different process stages, as well as environmental data of the areas where the different process stages are located, and to fuse the greenhouse gas concentration data and environmental data according to a preset data fusion strategy to obtain multimodal time series data. The output module is configured to input the multimodal time series data into a pre-trained spatiotemporal graph neural network model, and the spatiotemporal graph neural network model outputs transformation features corresponding to the multimodal time series data. The calculation module is configured to construct a causal graph model of greenhouse gas emissions at each process stage based on the transformation characteristics, and calculate the contribution of each process stage to the total emissions based on the causal graph model. The first judgment module is configured to determine whether the contribution of each process stage to the total emissions is greater than a preset contribution threshold. The sorting module is configured to sort the greenhouse gas concentration data of each process stage based on the location relationship of the monitoring points if the contribution of a certain process stage to the total emissions is greater than a preset contribution threshold, thereby obtaining at least one greenhouse gas concentration data sequence. The fusion module is configured to perform weighted fusion of the at least one greenhouse gas concentration data sequence based on the weight of each gas in the global warming potential value and according to a preset weighting strategy to obtain a comprehensive greenhouse gas emission intensity sequence. The second judgment module is configured to calculate the change gradient of the emission intensity sequence within a sliding time window and determine whether the change gradient is greater than a preset threshold. The generation module is configured to generate an early warning signal for abnormal greenhouse gas emissions in a certain process stage if the emissions exceed a preset threshold, and then send the early warning signal to the user terminal.

8. An electronic device, characterized in that, include: At least one processor, and a memory communicatively connected to the at least one processor, wherein the memory stores instructions executable by the at least one processor to enable the at least one processor to perform the method according to any one of claims 1 to 6.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the program is executed by a processor, it implements the method described in any one of claims 1 to 6.

Images