Rural domestic tail water cascade resource utilization regulation and control method and system

Abstract

The invention discloses a regulation and control method and system for gradient resource utilization of rural domestic tail water, and the requirements of reuse water in different seasons are met through regulation and control, that is, the quality of outlet water in the irrigation season meets the farmland water quality standard, and the quality standard of miscellaneous use of tail water or ecological water supplement is met in the non-irrigation season. According to the method, the state of tail water is monitored in real time through a sensor matrix, meanwhile, organic matter (or chemical oxygen demand) and ammonia nitrogen, total nitrogen, total phosphorus and microorganism effluent quality requirements input by a user are obtained, and time sequence semantic features of the state of the tail water are mined by adopting a neural network technology based on deep learning; and carrying out conjoint analysis on organic matters, ammonia nitrogen, total nitrogen and total phosphorus effluent to obtain associated characterization of multi-aspect tail water treatment requirements.

Description

Technical Field

[0001] This application relates to the field of intelligent regulation, and more specifically, to a regulation method and system for cascade resource utilization of rural domestic tail water. Background Art

[0002] The resource utilization of rural domestic sewage is the main way for rural domestic sewage treatment in China in the next few decades. Reusing the treated rural domestic sewage (tail water) for farmland is the most economical and effective way for the resource utilization of rural domestic sewage. Reusing for farmland does not require nitrogen and phosphorus removal, which can not only reduce the treatment difficulty, but also achieve the resource recycling rate of nitrogen and phosphorus resources. However, agricultural irrigation is seasonal. During the non-irrigation season, the tail water can be used for ecological water replenishment or miscellaneous water, but the sewage needs to be treated for nitrogen and phosphorus removal.

[0003] The rural domestic sewage treatment facilities are designed according to the irrigation standard, and cannot meet the miscellaneous water standard or discharge standard during the non-irrigation season, and black and odorous water bodies are also easily formed during storage. Designed according to the miscellaneous water or ecological water replenishment standard, it will cause waste of nitrogen and phosphorus resources during the irrigation season and increase the operation and maintenance management cost.

[0004] Therefore, in order to meet the water demand of farmland planting during the irrigation season and meet the water quality standards of tail water miscellaneous use or ecological water replenishment (up to standard discharge) during the non-irrigation season, a regulation method and system for cascade resource utilization of rural domestic tail water are expected to solve the imbalance between the tail water generation amount and the planting water demand in different seasons, reduce the operation cost, and avoid the generation of black and odorous water bodies. Summary of the Invention

[0005] In order to solve the above technical problems, this application is proposed. The embodiments of this application provide a regulation method and system for cascade resource utilization of rural domestic tail water, which can meet the reclaimed water demand in different seasons through regulation, that is, the effluent water quality meets the farmland water use standard during the irrigation season, and meets the tail water miscellaneous use or ecological water replenishment water quality standard during the non-irrigation season. It uses a sensor matrix to monitor the tail water state in real time, and at the same time obtains the organic matter (or chemical oxygen demand), ammonia nitrogen, total nitrogen, total phosphorus, and microbial effluent water quality requirements input by the user, and uses neural network technology based on deep learning to mine the temporal semantic features of the tail water state, and conducts a joint analysis of the organic matter, ammonia nitrogen, total nitrogen, and total phosphorus effluent to obtain the correlation representation of various aspects of tail water treatment requirements. Further, based on the query matching results of the temporal semantic features of the tail water state and the correlation features of rural tail water treatment requirements, the rotation speed value of the blower, the intermittent operation time of the reflux pump, the chemical dosage, the start and stop of the ultraviolet disinfection equipment, etc. are adaptively adjusted. In this way, it can intelligently solve the seasonal demand difference problem in the resource utilization of rural domestic sewage, so as to achieve the efficient utilization of resources, reduce the operation and maintenance cost, prevent the generation of black and odorous water bodies, and further achieve the goals of environmental protection and sustainable development.

[0006] According to one aspect of the present application, a method for regulating the cascade resource utilization of rural domestic tail water is provided, which includes:

[0007] Obtaining a data set of rural domestic tail water state parameters collected by a sensor matrix, and the required organic matter and ammonia nitrogen effluent water quality requirements for ecological water replenishment input by a user;

[0008] Performing data structuring processing on the data set of the rural domestic tail water state parameters to obtain a time series aggregation matrix of the rural domestic tail water state parameters;

[0009] Performing feature extraction on the time series aggregation matrix of the rural domestic tail water state parameters to obtain a time series correlation semantic coding feature of the rural domestic tail water state parameters;

[0010] Performing joint coding on the required organic matter for ecological water replenishment and the ammonia nitrogen effluent water quality requirements to obtain a joint coding feature of rural tail water treatment requirements;

[0011] Extracting local query hint information between the time series correlation semantic coding feature of the rural domestic tail water state parameters and the joint coding feature of the rural tail water treatment requirements, and performing cross-modal fusion coding guided by the local query hint information on the time series correlation semantic coding feature of the rural domestic tail water state parameters and the joint coding feature of the rural tail water treatment requirements to obtain a processing requirement - time series semantic joint coding feature of tail water state parameters;

[0012] Based on the processing requirement - time series semantic joint coding feature of tail water state parameters, obtaining an optimization instruction.

[0013] According to another aspect of the present application, a system for regulating the cascade resource utilization of rural domestic tail water is provided, which includes:

[0014] An information acquisition module, configured to obtain a data set of rural domestic tail water state parameters collected by a sensor matrix, and the required organic matter and ammonia nitrogen effluent water quality requirements for ecological water replenishment input by a user;

[0015] A data structuring processing module, configured to perform data structuring processing on the data set of the rural domestic tail water state parameters to obtain a time series aggregation matrix of the rural domestic tail water state parameters;

[0016] A rural domestic tail water state parameter feature extraction module, configured to perform feature extraction on the time series aggregation matrix of the rural domestic tail water state parameters to obtain a time series correlation semantic coding feature of the rural domestic tail water state parameters;

[0017] A joint coding module, configured to perform joint coding on the required organic matter for ecological water replenishment and the ammonia nitrogen effluent water quality requirements to obtain a joint coding feature of rural tail water treatment requirements;

[0018] A cross-modal fusion coding module is used to extract local query prompt information between the rural domestic tailwater status parameter temporal association semantic coding features and the rural tailwater treatment requirement joint coding features, and perform cross-modal fusion coding guided by local query prompt information on the rural domestic tailwater status parameter temporal association semantic coding features and the rural tailwater treatment requirement joint coding features to obtain the treatment requirement-tailwater status parameter temporal semantic joint coding features;

[0019] The optimization instruction generation module is used to obtain the optimization instruction based on the processing requirement-tailwater state parameter temporal semantic joint coding feature.

[0020] Compared with the prior art, the present application provides a method and system for regulating the cascade resource utilization of rural domestic tailwater, which meets the reuse water demand in different seasons through regulation, that is, the effluent water quality in the irrigation season meets the farmland water use standard, and meets the tailwater miscellaneous use or ecological water replenishment water quality standard in the non-irrigation season. It uses a sensor matrix to monitor the tailwater status in real time, and simultaneously obtains the organic matter (or chemical oxygen demand) and ammonia nitrogen, total nitrogen, total phosphorus, and microbial effluent water quality requirements input by the user, and uses deep learning-based neural network technology to mine the temporal semantic features of the tailwater status, and conducts a joint analysis of the organic matter and ammonia nitrogen, total nitrogen, and total phosphorus effluents to obtain the correlation representation of various tailwater treatment requirements. Furthermore, based on the query matching results of the temporal semantic features of the tailwater status and the associated features of the rural tailwater treatment requirements, the blower speed value, the intermittent operation time of the reflux pump, the dosage, the start and stop of the ultraviolet disinfection equipment, etc. are adaptively adjusted. In this way, the problem of seasonal demand differences in the resource utilization of rural domestic sewage can be solved intelligently, thereby realizing efficient utilization of resources, reducing operation and maintenance costs, preventing the generation of black and odorous water bodies, and achieving the goals of environmental protection and sustainable development. BRIEF DESCRIPTION OF THE DRAWINGS

[0021] By describing the embodiments of the present application in more detail in conjunction with the accompanying drawings, the above and other purposes, features and advantages of the present application will become more apparent. The accompanying drawings are used 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 accompanying drawings, the same reference numerals generally represent the same components or steps.

[0022] Figure 1 It is a flow chart of a method for regulating and controlling the cascade resource utilization of rural domestic tailwater according to an embodiment of the present application;

[0023] Figure 2 A data flow diagram of a method for regulating and controlling the cascade resource utilization of rural domestic tailwater according to an embodiment of the present application;

[0024] Figure 3 It is a flowchart of sub-step S5 of the rural domestic sewage cascade resource utilization regulation method according to an embodiment of the present application;

[0025] Figure 4 It is a block diagram of the rural domestic sewage cascade resource utilization regulation system according to an embodiment of the present application. Detailed implementation manners

[0026] Hereinafter, exemplary embodiments according to the present application will be described in detail with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all embodiments of the present application. It should be understood that the present application is not limited by the exemplary embodiments described herein.

[0027] As shown in the present application and the claims, unless the context clearly indicates an exception, words such as "a", "an", "one" and / or "the" are not specifically singular and may also include plural. Generally speaking, the terms "include" and "comprise" only indicate the inclusion of the steps and elements that have been clearly identified, and these steps and elements do not constitute an exclusive list. The method or device may also include other steps or elements.

[0028] Although the present application makes various references to certain modules in the system according to the embodiments of the present application, however, any number of different modules can be used and run on the user terminal and / or the server. The modules are only illustrative, and different aspects of the system and method can use different modules.

[0029] Flowcharts are used in the present application to illustrate the operations performed by the system according to the embodiments of the present application. It should be understood that the operations before or below are not necessarily executed precisely in order. On the contrary, various steps can be executed in reverse order or simultaneously as needed. At the same time, other operations can also be added to these processes, or one or several steps can be removed from these processes.

[0030] Hereinafter, exemplary embodiments according to the present application will be described in detail with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all embodiments of the present application. It should be understood that the present application is not limited by the exemplary embodiments described herein.

[0031] In the technical solution of the present application, a rural domestic sewage cascade resource utilization regulation method is proposed. Figure 1 It is a flowchart of the rural domestic sewage cascade resource utilization regulation method according to an embodiment of the present application. Figure 2 It is a schematic diagram of data flow of the rural domestic sewage cascade resource utilization regulation method according to an embodiment of the present application. As Figure 1 and Figure 2As shown, the rural domestic sewage cascade resource utilization regulation method according to an embodiment of the present application includes the steps of: S1, obtaining a dataset of rural domestic sewage state parameters collected by a sensor matrix, and the required organic matter and ammonia nitrogen effluent water quality requirements for ecological water replenishment input by a user; S2, performing data structuring processing on the dataset of the rural domestic sewage state parameters to obtain a rural domestic sewage state parameter time series aggregation matrix; S3, performing feature extraction on the rural domestic sewage state parameter time series aggregation matrix to obtain a rural domestic sewage state parameter time series correlation semantic coding feature; S4, jointly encoding the required organic matter for ecological water replenishment and the ammonia nitrogen effluent water quality requirements to obtain a rural sewage treatment requirement joint coding feature; S5, extracting local query prompt information between the rural domestic sewage state parameter time series correlation semantic coding feature and the rural sewage treatment requirement joint coding feature, and performing cross-modal fusion coding guided by the local query prompt information on the rural domestic sewage state parameter time series correlation semantic coding feature and the rural sewage treatment requirement joint coding feature to obtain a processing requirement - sewage state parameter time series semantic joint coding feature; S6, obtaining an optimization instruction based on the processing requirement - sewage state parameter time series semantic joint coding feature.

[0032] Specifically, in S1, a dataset of rural domestic sewage state parameters collected by a sensor matrix is obtained; and the required organic matter for ecological water replenishment and the ammonia nitrogen effluent water quality requirements input by a user. Among them, the rural domestic sewage state parameters include water volume, organic matter effluent concentration (or chemical oxygen demand), ammonia nitrogen effluent concentration, water temperature value, sludge concentration value, and dissolved oxygen concentration value. It should be understood that the treatment and utilization of rural domestic sewage need to be based on real-time and accurate data. Collecting a dataset of sewage state parameters through a sensor matrix can provide a reliable basis for subsequent data processing and analysis. And different application scenarios have different water quality requirements for sewage. By obtaining the required organic matter for ecological water replenishment and the ammonia nitrogen effluent water quality requirements input by a user, it can ensure that the treated sewage can meet the actual use requirements and avoid over-treatment or under-treatment. In one example, various sensors such as flow meters, sludge concentration sensors, dissolved oxygen sensors, ammonia nitrogen sensors, organic matter sensors, and temperature sensors can be installed at key positions (such as the inlet, treatment unit, outlet, etc.) of rural domestic sewage treatment facilities. The sensors regularly collect various parameters of the sewage, such as influent flow rate, sludge concentration, dissolved oxygen, ammonia nitrogen concentration, organic matter concentration, water temperature, etc., and transmit the data to the central data processing system to obtain a dataset of rural domestic sewage state parameters.

[0033] Specifically, in step S2, the data set of the rural domestic sewage state parameters is subjected to data structuring to obtain a time-series aggregation matrix of the rural domestic sewage state parameters. That is, in the technical solution of the present application, a data matrix is constructed, and the data set of the rural domestic sewage state parameters is filled into the data matrix according to the time dimension and the parameter sample dimension to obtain a time-series aggregation matrix of the rural domestic sewage state parameters. It should be understood that the original data is usually scattered and disordered. By constructing a data matrix, these data can be unified into a standard format, which is convenient for subsequent data processing and analysis.

[0034] Specifically, in step S3, feature extraction is performed on the time-series aggregation matrix of the rural domestic sewage state parameters to obtain time-series correlation semantic coding features of the rural domestic sewage state parameters. In a specific example of the present application, the time-series aggregation matrix of the rural domestic sewage state parameters is input into a state parameter time-series pattern feature extractor based on a depthwise separable convolutional neural network model to obtain a time-series correlation semantic coding feature map of the rural domestic sewage state parameters as the time-series correlation semantic coding features of the rural domestic sewage state parameters. It should be understood that the patterns of the tail water state parameters changing over time are often very complex, and traditional convolutional neural network models may not be able to effectively capture these patterns. The depthwise separable convolutional neural network (DS-CNN) can extract features of different scales through multiple convolutional operations, which helps to capture the local and global patterns of the tail water state parameters. Therefore, in the technical solution of the present application, by inputting the data matrix to obtain a time-series aggregation matrix of the rural domestic sewage state parameters into a state parameter time-series pattern feature extractor based on a depthwise separable convolutional neural network model, the time-varying patterns and internal correlations of the tail water state parameters are extracted, and a time-series correlation semantic coding feature map of the rural domestic sewage state parameters is obtained as the time-series correlation semantic coding features of the rural domestic sewage state parameters.

[0035] Specifically, in step S4, the organic matter required for ecological water replenishment and the water quality requirements for ammonia nitrogen effluent are jointly encoded to obtain the joint encoding feature of rural wastewater treatment requirements. That is, in the technical solution of the present application, first, one-hot encoding is performed on the organic matter required for ecological water replenishment and the water quality requirements for ammonia nitrogen effluent to obtain the one-hot encoding vector of the organic matter required for ecological water replenishment and the one-hot encoding vector of the water quality requirements for ammonia nitrogen effluent. It should be understood that the organic matter required for ecological water replenishment and the water quality requirements for ammonia nitrogen effluent are usually continuous numerical values or categorical labels. Directly using these numerical values or labels may make it difficult for the model to understand and process. Therefore, in the technical solution of the present application, one-hot encoding is performed on the organic matter required for ecological water replenishment and the water quality requirements for ammonia nitrogen effluent to convert these continuous numerical values or categorical labels into discrete binary vectors, so as to obtain the one-hot encoding vector of the organic matter required for ecological water replenishment and the one-hot encoding vector of the water quality requirements for ammonia nitrogen effluent. In this way, the input data can be made more standardized, facilitating the input and processing of machine learning models. Specifically, one-hot encoding is a method of converting categorical data (category labels) into binary vectors. Through one-hot encoding, each category is represented as a unique binary vector, where only one position is 1 and the rest are 0. This way enables categorical data to be effectively processed and understood by machine learning models. Then, the one-hot encoding vector of the organic matter required for ecological water replenishment and the one-hot encoding vector of the water quality requirements for ammonia nitrogen effluent are concatenated to obtain the joint encoding vector of rural wastewater treatment requirements as the joint encoding feature of rural wastewater treatment requirements. It should be understood that separately processing the water quality data information of each dimension may lead to data loss or incompleteness. Through the concatenation operation, the data information of different dimensions can be integrated, so as to more comprehensively reflect the comprehensive requirements of rural wastewater treatment. The joint encoding feature of rural wastewater treatment requirements obtained by concatenation contains all the information of the organic matter required for ecological water replenishment and the water quality requirements for ammonia nitrogen effluent, providing a comprehensive representation of the water quality comprehensive feature, which helps to ensure that the model can consider the wastewater treatment requirements of these two aspects simultaneously.

[0036] Specifically, in step S5, local query hint information between the temporal correlation semantic coding features of the rural domestic sewage tailwater state parameters and the joint coding features of the rural sewage treatment requirements is extracted, and cross-modal fusion coding guided by the local query hint information is performed on the temporal correlation semantic coding features of the rural domestic sewage tailwater state parameters and the joint coding features of the rural sewage treatment requirements to obtain the joint coding features of the treatment requirements - tailwater state parameter temporal semantics. Considering that the joint coding vector of the rural sewage treatment requirements and the temporal correlation semantic coding feature map of the rural domestic sewage tailwater state parameters have different feature dimensions and information densities, and traditional feature fusion methods may encounter information misalignment problems when processing different types of data. Therefore, to achieve effective information fusion, the present application proposes a cross-modal fusion coding method based on continuous hint guidance, which continuously modulates the temporal correlation semantic coding feature map of the rural domestic sewage tailwater state parameters by using the rural sewage treatment requirements as hint information to achieve deep integration of cross-modal features. In a specific example of the present application, as Figure 3 shown, step S5 includes: S51, extracting local query hint information between the joint coding vector of the rural sewage treatment requirements and the temporal correlation semantic coding feature map of the rural domestic sewage tailwater state parameters to obtain a set of local query hint semantic coding vectors of the rural sewage treatment requirements - state parameters; S52, performing cross-modal mask interaction based on the hint information on the joint coding vector of the rural sewage treatment requirements and each of the local query hint semantic coding vectors of the rural sewage treatment requirements - state parameters to obtain a set of cross-modal local feature mask weight matrices of the rural domestic sewage tailwater; S53, using the set of cross-modal local feature mask weight matrices of the rural domestic sewage tailwater as the set of weights, performing position-wise weighting on each feature matrix along the channel dimension in the temporal correlation semantic coding feature map of the rural domestic sewage tailwater state parameters and then aggregating to obtain a cross-modal hint-guided joint coding feature map of the rural domestic sewage tailwater as the joint coding features of the treatment requirements - tailwater state parameter temporal semantics.

[0037] Specifically, in step S51, local query hint information between the combined coding vector of rural domestic sewage treatment requirements and the temporal correlation semantic coding feature map of rural domestic sewage status parameters is extracted to obtain a set of local query hint semantic coding vectors for rural domestic sewage treatment requirements - status parameters. That is, in the technical solution of this application, first, local feature decomposition of the temporal correlation semantic coding feature map of rural domestic sewage status parameters along the channel dimension is performed to obtain a set of local coding feature matrices for temporal correlation of rural domestic sewage status parameters; by performing local feature decomposition of the temporal correlation semantic coding feature map of rural domestic sewage status parameters along the channel dimension, the feature resolution is enhanced, and more attention is paid to the cross-modal interaction of detailed features. Then, using the combined coding vector of rural domestic sewage treatment requirements as the query vector, and each local coding feature matrix for temporal correlation of rural domestic sewage status parameters in the set of local coding feature matrices for temporal correlation of rural domestic sewage status parameters as the key matrix, the query vector and the key matrix are input into the hint learning network to obtain a set of local query hint semantic coding vectors for rural domestic sewage treatment requirements - status parameters. Here, the hint learning network is used to perform attention query coding on the two, calculate the attention weights through fine-grained semantic similarity measurement, and select the tail water status feature part related to the semantics of rural domestic sewage treatment requirements from each temporal correlation semantic coding feature matrix of rural domestic sewage status parameters, so as to generate a set of local query hint semantic coding vectors for rural domestic sewage treatment requirements - status parameters, in order to enhance the understanding of the inter-modal association interaction relationship.

[0038] More specifically, the following local query hint information extraction formula is used to extract the local query hint information between the combined coding vector of rural domestic sewage treatment requirements and the temporal correlation semantic coding feature map of rural domestic sewage status parameters to obtain a set of local query hint semantic coding vectors for rural domestic sewage treatment requirements - status parameters; where the local query hint information extraction formula includes:

[0039] Decompose(F 2 )={M 1 ,M 2 ,...,M n}

[0040] M i ={v i1 ,v i2 ,...,v im}

[0041]

[0042] Among them, v 1 represents the combined coding vector of rural domestic sewage treatment requirements, F 2Represents the time-series correlation semantic coding feature map of the rural domestic sewage state parameters. Decompose(·) represents feature decomposition, M 1 、M 2 、M i and M n respectively represent the first, second, i-th, and n-th time-series correlation local coding feature matrices of the rural domestic sewage state parameters, that is, each feature matrix along the channel dimension of the time-series correlation semantic coding feature map of the rural domestic sewage state parameters. n is the number of channels of the time-series correlation semantic coding feature map of the rural domestic sewage state parameters, v i1 、v i2 、v ij and v im respectively represent the first, second, j-th, and m-th row vectors in the i-th time-series correlation local coding feature matrix of the rural domestic sewage state parameters, that is, the key vectors. m is the number of rows of the time-series correlation local coding feature matrix of the rural domestic sewage state parameters. Transformer(·,·) represents the prompting learning network, represents vector multiplication, (·) T represents the transpose of a vector, ‖·‖ represents the norm of a vector, s j and s k respectively represent the j-th and k-th local query prompt semantic similarities between the rural sewage treatment requirements and state parameters. exp(·) represents the exponential function with base e, e j represents the j-th normalized local query prompt semantic similarity between the rural sewage treatment requirements and state parameters. ⊙ represents element-wise multiplication, v ti represents the i-th local query prompt semantic coding vector of the rural sewage treatment requirements and state parameters.

[0043] Specifically, in S52, cross-modal masked interaction based on prompt information is performed on the joint coding vector of the rural sewage treatment requirements and each local query prompt semantic coding vector of the rural sewage treatment requirements - state parameters to obtain a set of cross-modal local feature masked weight matrices for rural domestic sewage. In the technical solution of this application, by inputting the joint coding vector of the rural sewage treatment requirements and each local query prompt semantic coding vector in the set of local query prompt semantic coding vectors of the rural sewage treatment requirements - state parameters into a cross-modal masked weaving network based on prompt information for further prompting learning, the cross-modal masked network processes the two based on the self-attention mechanism, which can effectively highlight the important interaction parts between the two, while suppressing irrelevant information and reducing the influence of noise, thereby creating a mask that emphasizes the most relevant parts of the modal interaction, that is, the cross-modal local feature masked weight matrix for rural domestic sewage based on prompt information.

[0044] More specifically, a cross-modal mask interaction is performed on the combined encoding vector of the rural domestic sewage treatment requirements and each of the rural domestic sewage treatment requirement-state parameter local query prompt semantic encoding vectors according to the following cross-modal mask interaction formula to obtain a set of rural domestic sewage cross-modal local feature mask weight matrices; wherein, the cross-modal mask interaction formula is:

[0045]

[0046] wherein, S is the feature scale of the rural domestic sewage treatment requirement-state parameter local query prompt semantic encoding vector, softmax(·) represents the normalized exponential function, S ti represents the i-th cross-modal local feature mask weight matrix of rural domestic sewage based on the prompt information, represents vector multiplication.

[0047] Specifically, for the S53, using the set of rural domestic sewage cross-modal local feature mask weight matrices as the set of weights, each feature matrix along the channel dimension in the rural domestic sewage state parameter temporal correlation semantic encoding feature map is weighted by position and then aggregated to obtain a rural domestic sewage cross-modal prompt-guided combined encoding feature map as the processing requirement-sewage state parameter temporal semantic combined encoding feature. That is, first, using each cross-modal local feature mask weight matrix of rural domestic sewage based on the prompt information as a weight, each rural domestic sewage state parameter temporal correlation local encoding feature matrix is weighted and modulated to emphasize the correlation between the sewage state parameter and the processing requirement, realizing the guided fusion of cross-modal data semantic information, thereby further improving the feature representation quality and prediction performance of the model. Furthermore, the set of each rural domestic sewage cross-modal prompt-guided combined encoding feature matrices is subjected to information aggregation along the channel dimension to restore the original feature map structure, generating a rural domestic sewage cross-modal prompt-guided combined encoding feature map. Through this continuous prompt-assisted learning process, the semantic association between cross-modal data can be better captured, effectively guiding the model to focus on the water quality requirement features in the rural domestic sewage state parameter temporal correlation semantic encoding feature map that match the rural domestic sewage treatment requirements, thereby improving the accuracy of sewage state processing.

[0048] More specifically, using the set of rural domestic sewage cross-modal local feature mask weight matrices as the set of weights, each feature matrix along the channel dimension in the rural domestic sewage state parameter temporal correlation semantic encoding feature map is weighted by position and then aggregated according to the following weighted aggregation formula to obtain a rural domestic sewage cross-modal prompt-guided combined encoding feature map as the processing requirement-sewage state parameter temporal semantic combined encoding feature; wherein, the weighted aggregation formula is;

[0049] F = Concat{M 1 ⊙ S t1 , M 2 ⊙ S t2 ,..., M i ⊙ S ti ..., M n ⊙ S tn}

[0050] where Concat{·, ·,..., ·} represents concatenation, ⊙ represents element-wise multiplication, S t1 , S t2 , S ti and S tn respectively represent the first, second, ith, and nth cross-modal local feature mask weight matrices of rural domestic sewage based on the hint information, and F represents the processing requirement - tail water state parameter temporal semantic joint encoding feature map.

[0051] In particular, the S6 obtains an optimization instruction based on the processing requirement - tail water state parameter temporal semantic joint encoding feature. In a specific example of the present application, the processing requirement - tail water state parameter temporal semantic joint encoding feature map is input into a decoder-based regulation and optimization module to obtain an optimization instruction, and the optimization instruction includes the decoded value of the recommended rotational speed value of the blower. That is, by using the fusion feature between the tail water state parameter and the processing requirement, an optimal control instruction, namely the recommended rotational speed value, is generated. By precisely controlling the operating parameters of the tail water treatment equipment, the treatment effect is ensured to be optimal, while resources are saved to the greatest extent and the operating cost is reduced. In particular, the decoder is a neural network model used to convert high-level feature representations into specific outputs. In a specific example, the decoding process of the decoder includes: First, multiple convolutional layers are used to extract local features in the processing requirement - tail water state parameter temporal semantic joint encoding feature map, and a pooling layer is used to reduce the spatial dimension of the processing requirement - tail water state parameter temporal semantic joint encoding feature map to retain the most important processing requirement - tail water state parameter temporal semantic joint encoding feature information; Then, after the processing requirement - tail water state parameter temporal semantic joint encoding feature map is unfolded into a set of processing requirement - tail water state parameter temporal semantic joint encoding feature vectors, multiple fully connected layers are used to further extract and fuse features to generate a comprehensive feature vector. Finally, the comprehensive feature vector is mapped to a specific recommended rotational speed value through the last fully connected layer to ensure the efficient operation of the tail water treatment equipment.

[0052] In particular, when the temporal correlation semantic coding feature map of the rural domestic sewage state parameters and the joint coding vector of the rural sewage treatment requirements respectively represent the implicit temporal correlation features among the parameters of the dataset of rural domestic sewage state parameters and the one-hot embedding coding cascade features of the required organic matter and ammonia nitrogen effluent water quality for ecological water replenishment, during the cross-modal joint coding assisted by continuous prompts, the temporal semantic joint coding feature map of the treatment requirements - sewage state parameters will also have insufficient long-distance joint perception representation of cross-modal information under multi-dimensional feature distributions, thereby reducing the expression effect of the temporal semantic joint coding feature map of the treatment requirements - sewage state parameters relative to the decoder-based regulation and optimization module. As a result, when the temporal semantic joint coding feature map of the treatment requirements - sewage state parameters is decoded and regressed through the decoder, the accuracy of the decoding result is affected.

[0053] Preferably, in the process of obtaining the optimization instruction by passing the temporal semantic joint coding feature map of the treatment requirements - sewage state parameters through the decoder-based regulation and optimization module, feature tuning is performed on the temporal semantic joint coding feature map of the treatment requirements - sewage state parameters, including:

[0054] Unfolding the temporal semantic joint coding feature map of the treatment requirements - sewage state parameters into a temporal semantic joint coding feature vector of the treatment requirements - sewage state parameters;

[0055] Deriving a parameter configuration tensor from the decoder-based regulation and optimization module;

[0056] Mapping the temporal semantic joint coding feature vector of the treatment requirements - sewage state parameters to the representation space of the parameter configuration tensor to generate a semantic joint simulation decoding mapping coding representation;

[0057] v p = M c v o

[0058] where v o represents the temporal semantic joint coding feature vector of the treatment requirements - sewage state parameters, M c represents the parameter configuration tensor, and v p represents the semantic joint simulation decoding mapping coding representation;

[0059] Applying an exponential reinforcement operation to the semantic joint simulation decoding mapping coding representation to output a calibrated-level semantic joint mapping coding representation;

[0060] v l = exp(v p )

[0061] where v l represents the calibrated-level semantic joint mapping coding representation;

[0062] Construct a semantic joint high-dimensional aggregation retrieval coding tensor based on the calibrated-level semantic joint mapping coding representation and the semantic joint simulation decoding mapping coding representation;

[0063]

[0064] where T represents the transpose operation, represents matrix multiplication, L represents the length of the calibrated-level semantic joint mapping coding representation, and M c represents the semantic joint high-dimensional aggregation retrieval coding tensor;

[0065] Perform a logarithmic transformation on the semantic joint high-dimensional aggregation retrieval coding tensor to generate a semantic joint smooth retrieval feature architecture;

[0066]

[0067] where M t represents the semantic joint smooth retrieval feature architecture;

[0068] Remap the processing requirement-tail water state parameter time series semantic joint coding feature vector to the representation domain of the semantic joint smooth retrieval feature architecture to obtain a semantic joint decoding auxiliary coding representation;

[0069] v b = M t v o

[0070] where v b represents the semantic joint decoding auxiliary coding representation;

[0071] By integrating the processing requirement-tail water state parameter time series semantic joint coding feature vector and the semantic joint decoding auxiliary coding representation, output an optimized processing requirement-tail water state parameter time series semantic joint coding feature vector;

[0072] v opt = α·v b + β·v o

[0073] where α and β represent weighted hyperparameters, and v opt represents the optimized processing requirement-tail water state parameter time series semantic joint coding feature vector.

[0074] Here, the original feature vector is guided to evolve in a more discriminative direction through the set of orthogonal basis determined by the implicit inference rules of the model, and the correlation strength with the decoded label calibration is significantly amplified by using the exponential enhancement operator. The high-dimensional aggregated retrieval smooth domain of the decoding features is further constructed as a representation distillation carrier to capture auxiliary representations that are beneficial to decoding, and dynamic representation tuning is implemented by fusing these decoding auxiliary representations. The optimized decoding features are more in line with the plastic calibration correction of the intrinsic parameter configuration of the decoder to improve the accuracy of the optimization instructions obtained by its input into the decoder-based control optimization module.

[0075] In summary, the regulation method for the cascade resource utilization of rural domestic tailwater according to the embodiment of the present application is explained, and the reuse water demand in different seasons is met through regulation, that is, the effluent water quality in the irrigation season meets the farmland water use standard, and the tailwater miscellaneous use or ecological replenishment water quality standard in the non-irrigation season is met. It uses a sensor matrix to monitor the tailwater status in real time, and simultaneously obtains the organic matter (or chemical oxygen demand) and ammonia nitrogen, total nitrogen, total phosphorus, and microbial effluent water quality requirements input by the user. The neural network technology based on deep learning is used to mine the temporal semantic features of the tailwater status, and the organic matter and ammonia nitrogen, total nitrogen, and total phosphorus effluents are jointly analyzed to obtain the correlation representation of various tailwater treatment requirements. Furthermore, based on the query matching results of the temporal semantic features of the tailwater status and the associated features of the rural tailwater treatment requirements, the blower speed value, the intermittent operation time of the reflux pump, the dosage, the start and stop of the ultraviolet disinfection equipment, etc. are adaptively adjusted. In this way, the problem of seasonal demand differences in the resource utilization of rural domestic sewage can be solved intelligently, thereby realizing efficient utilization of resources, reducing operation and maintenance costs, preventing the generation of black and odorous water bodies, and achieving the goals of environmental protection and sustainable development.

[0076] Furthermore, a rural domestic tail water cascade resource utilization and control system is also provided.

[0077] Figure 4 : is a block diagram of a rural domestic tailwater cascade resource utilization control system according to an embodiment of the present application. Figure 4As shown, the rural domestic sewage cascade resource utilization regulation system 300 according to an embodiment of the present application includes: an information acquisition module 310, configured to acquire a dataset of rural domestic sewage state parameters collected by a sensor matrix, and the required organic matter and ammonia nitrogen effluent water quality requirements for ecological water replenishment input by a user; a data structuring processing module 320, configured to perform data structuring processing on the dataset of the rural domestic sewage state parameters to obtain a rural domestic sewage state parameter time series aggregation matrix; a rural domestic sewage state parameter feature extraction module 330, configured to perform feature extraction on the rural domestic sewage state parameter time series aggregation matrix to obtain a rural domestic sewage state parameter time series correlation semantic coding feature; a joint coding module 340, configured to perform joint coding on the required organic matter for ecological water replenishment and the ammonia nitrogen effluent water quality requirements to obtain a rural sewage treatment requirement joint coding feature; a cross-modal fusion coding module 350, configured to extract local query prompt information between the rural domestic sewage state parameter time series correlation semantic coding feature and the rural sewage treatment requirement joint coding feature, and perform cross-modal fusion coding guided by the local query prompt information on the rural domestic sewage state parameter time series correlation semantic coding feature and the rural sewage treatment requirement joint coding feature to obtain a processing requirement - sewage state parameter time series semantic joint coding feature; an optimization instruction generation module 360, configured to obtain an optimization instruction based on the processing requirement - sewage state parameter time series semantic joint coding feature.

[0078] As described above, the rural domestic sewage cascade resource utilization regulation system 300 according to an embodiment of the present application can be implemented in various wireless terminals, such as a server with a rural domestic sewage cascade resource utilization regulation algorithm. In a possible implementation manner, the rural domestic sewage cascade resource utilization regulation system 300 according to an embodiment of the present application can be integrated into a wireless terminal as a software module and / or a hardware module. For example, the rural domestic sewage cascade resource utilization regulation system 300 can be a software module in the operating system of the wireless terminal, or can be an application program developed for the wireless terminal; of course, the rural domestic sewage cascade resource utilization regulation system 300 can also be one of the numerous hardware modules of the wireless terminal.

[0079] Alternatively, in another example, the rural domestic sewage cascade resource utilization regulation system 300 and the wireless terminal can also be separate devices, and the rural domestic sewage cascade resource utilization regulation system 300 can be connected to the wireless terminal through a wired and / or wireless network, and transmit interaction information in accordance with a predefined data format.

[0080] The resource utilization of rural domestic sewage is the main approach for rural domestic sewage treatment in China in the next few decades. Reusing the treated rural domestic sewage (effluent) for farmland is the most economical and effective way of resource utilization of rural domestic sewage. Reusing it for farmland does not require nitrogen and phosphorus removal, which can not only reduce the treatment difficulty but also achieve the recycling rate of nitrogen and phosphorus resources. However, agricultural irrigation is seasonal. During the non-irrigation season, the effluent can be used for ecological water replenishment or miscellaneous water use, but nitrogen and phosphorus removal treatment of the sewage is required.

[0081] The rural domestic sewage treatment facilities are designed according to the irrigation standard and cannot meet the miscellaneous water use standard or the discharge standard of sewage during the non-irrigation season, and black and odorous water bodies are also likely to form during the storage process. Designed according to the up-to-standard discharge standard, nitrogen and phosphorus resources will be wasted during the irrigation season, and the operation and maintenance management costs will be increased.

[0082] Based on this demand, this application provides "a stepped resource utilization regulation method and system for rural domestic effluent". During the irrigation season, it meets the water demand of farmland planting, and during the non-irrigation season, it meets the water quality standards for miscellaneous water use or sewage discharge of the effluent. It solves the imbalance problem between the effluent production volume and the water demand of planting industry in different seasons, reduces the operation cost, and avoids the generation of black and odorous water bodies.

[0083] (1) Through the integration of equipment and the optimized design of the automatic control system, key operating conditions such as the dissolved oxygen, reflux ratio, and chemical dosage of the sewage treatment device are regulated to achieve the regulation of the effluent quality as needed.

[0084] (2) According to the required effluent quality, it can respectively meet the farmland irrigation water quality standard, the miscellaneous water use quality standard of sewage, and the ecological water replenishment water quality standard (up-to-standard discharge), realizing stepped utilization.

[0085] (3) The regulation of the effluent quality is achieved through the optimized design of the automatic control system and the learning of artificial intelligence, realizing one-key regulation, which is fast, simple, and reliable.

[0086] (4) This system consists of an adjustment tank, an anoxic tank, an aerobic tank, a sedimentation tank, and an effluent tank. The main equipment includes 1) a blower; 2) a frequency converter; 3) a reflux pump (installed in the aerobic tank); 4) a multi-parameter instrument (online dissolved oxygen meter, water temperature, sludge concentration, installed in the aerobic tank); 5) an ultraviolet disinfection device (installed in the effluent tank); 6) a chemical dosing device; 7) control software; 8) a PLC control system.

[0087] Regulation method:

[0088] 1. Input the correlation between different effluent organic matter concentrations and operating condition parameters and equipment operating parameters for artificial intelligence learning. The operating parameters include sludge concentration, dissolved oxygen in the aerobic tank, water temperature, and water volume, and the equipment control parameters include the rotation speed of the blower.

[0089] 2. Input the correlation relationships between different effluent ammonia nitrogen concentrations and operating condition parameters and equipment operating parameters for artificial intelligence learning. The operating parameters include sludge concentration, dissolved oxygen in the aerobic tank, water temperature, and water volume, and the equipment control parameter includes the rotation speed of the blower.

[0090] 3. Input the correlation relationships between different effluent total nitrogen concentrations and operating condition parameters and equipment operating parameters for artificial intelligence learning. The operating parameters include reflux ratio, sludge concentration, water temperature, and water volume, and the equipment control parameter includes the intermittent operation time of the reflux pump.

[0091] 4. Input the correlation relationships between different effluent total phosphorus concentrations and operating condition parameters and equipment operating parameters for artificial intelligence learning. The operating parameter is water volume, and the equipment control parameter includes the chemical dosing amount of the chemical dosing pump.

[0092] 5. Start the ultraviolet disinfection device when there is a requirement for the effluent microorganism concentration.

[0093] The embodiments of the present disclosure have been described above. The above description is exemplary and not exhaustive, and is not limited to the disclosed embodiments. Many modifications and variations are obvious to those of ordinary skill in the art in the technical field without departing from the scope and spirit of the described embodiments. The selection of the terms used herein is intended to best explain the principles of the embodiments, practical applications, or improvements to the technologies in the market, or to enable other ordinary technical personnel in the technical field to understand the embodiments disclosed herein.

Claims

1. A method for regulating and controlling the cascade resource utilization of rural domestic tailwater, characterized in that: include: Obtain the data set of rural domestic tailwater status parameters collected by the sensor matrix, as well as the organic matter and ammonia nitrogen effluent quality requirements for ecological water replenishment input by the user; Performing data structuring processing on the data set of rural domestic tail water state parameters to obtain a rural domestic tail water state parameter time series aggregation matrix; Performing feature extraction on the rural domestic tail water status parameter time series aggregation matrix to obtain rural domestic tail water status parameter time series association semantic coding features; The organic matter required for ecological water replenishment and the ammonia nitrogen effluent water quality requirements are jointly coded to obtain the joint coding characteristics of rural tailwater treatment requirements; Extracting local query prompt information between the rural domestic tailwater status parameter temporal association semantic coding feature and the rural tailwater treatment requirement joint coding feature, and performing cross-modal fusion coding guided by local query prompt information on the rural domestic tailwater status parameter temporal association semantic coding feature and the rural tailwater treatment requirement joint coding feature to obtain the treatment requirement-tailwater status parameter temporal semantic joint coding feature; Based on the processing requirement-tailwater state parameter temporal semantic joint coding characteristics, an optimization instruction is obtained.

2. The method for regulating and controlling the cascade resource utilization of rural domestic tailwater according to claim 1, characterized in that: The rural domestic tail water status parameters include water volume, organic matter effluent concentration, ammonia nitrogen effluent concentration, water temperature value, sludge concentration value and dissolved oxygen concentration value.

3. The method for regulating and controlling the cascade resource utilization of rural domestic tailwater according to claim 2 is characterized in that: The data set of rural domestic tail water state parameters is subjected to data structuring processing to obtain a rural domestic tail water state parameter time series aggregation matrix, including: A data matrix is ​​constructed, and the data set of the rural domestic tail water status parameters is filled into the data matrix according to the time dimension and the parameter sample dimension to obtain the rural domestic tail water status parameter time series aggregation matrix.

4. The method for regulating and controlling the cascade resource utilization of rural domestic tailwater according to claim 3 is characterized in that: The rural domestic tail water status parameter time series aggregation matrix is ​​subjected to feature extraction to obtain rural domestic tail water status parameter time series associated semantic coding features, including: The rural domestic tail water state parameter time series aggregation matrix is ​​input into a state parameter time series pattern feature extractor based on a deep separable convolutional neural network model to obtain a rural domestic tail water state parameter time series association semantic coding feature map as the rural domestic tail water state parameter time series association semantic coding feature.

5. The method for regulating and controlling the cascade resource utilization of rural domestic tailwater according to claim 4 is characterized in that: The organic matter required for ecological water replenishment and the ammonia nitrogen effluent water quality requirements are jointly coded to obtain the joint coding characteristics of rural tailwater treatment requirements, including: One-hot encoding is performed on the organic matter required for ecological water replenishment and the ammonia nitrogen effluent water quality requirement to obtain a one-hot encoding vector of the organic matter required for ecological water replenishment and a one-hot encoding vector of the ammonia nitrogen effluent water quality requirement; The one-hot encoding vector of organic matter required for ecological water replenishment and the one-hot encoding vector of ammonia nitrogen effluent water quality requirement are cascaded to obtain a joint encoding vector of rural tailwater treatment requirements as the joint encoding feature of rural tailwater treatment requirements.

6. The method for regulating and controlling the cascade resource utilization of rural domestic tailwater according to claim 5, characterized in that: Extracting local query prompt information between the rural domestic tailwater status parameter temporal association semantic coding feature and the rural tailwater treatment requirement joint coding feature, and performing cross-modal fusion coding guided by local query prompt information on the rural domestic tailwater status parameter temporal association semantic coding feature and the rural tailwater treatment requirement joint coding feature to obtain the treatment requirement-tailwater status parameter temporal semantic joint coding feature, including: Extracting local query prompt information between the rural tailwater treatment requirement joint coding vector and the rural domestic tailwater status parameter temporal association semantic coding feature map to obtain a set of rural tailwater treatment requirement-status parameter local query prompt semantic coding vectors; Performing a cross-modal mask interaction based on the prompt information on the rural tailwater treatment requirement joint coding vector and each of the rural tailwater treatment requirement-state parameter local query prompt semantic coding vectors to obtain a set of rural domestic tailwater cross-modal local feature mask weight matrices; Taking the set of the rural domestic tailwater cross-modal local feature mask weight matrices as the set of weights, each feature matrix along the channel dimension in the rural domestic tailwater state parameter temporal association semantic coding feature map is weighted by position and then aggregated to obtain the rural domestic tailwater cross-modal prompt guidance joint coding feature map as the processing requirement-tailwater state parameter temporal semantic joint coding feature.

7. The method for regulating and controlling the cascade resource utilization of rural domestic tailwater according to claim 6 is characterized in that: Extracting local query prompt information between the rural tailwater treatment requirement joint coding vector and the rural domestic tailwater status parameter temporal association semantic coding feature map to obtain a set of rural tailwater treatment requirement-status parameter local query prompt semantic coding vectors, including: Performing local feature decomposition along the channel dimension on the rural domestic tail water state parameter temporal association semantic coding feature map to obtain a set of rural domestic tail water state parameter temporal association local coding feature matrices; The rural tailwater treatment requirement joint coding vector is used as the query vector, and each rural domestic tailwater status parameter time-series associated local coding feature matrix in the set of rural domestic tailwater status parameter time-series associated local coding feature matrices is used as a key matrix, and the query vector and the key matrix are input into the prompt learning network to obtain the set of rural tailwater treatment requirement-status parameter local query prompt semantic coding vectors.

8. The method for regulating and controlling the cascade resource utilization of rural domestic tailwater according to claim 7 is characterized in that: Based on the processing requirement-tailwater state parameter temporal semantic joint coding feature, an optimization instruction is obtained, including: The processing requirement-tailwater state parameter temporal semantic joint encoding feature map is input into a decoder-based control optimization module to obtain an optimization instruction, wherein the optimization instruction includes a decoded value of a recommended speed value of the blower.

9. A rural domestic tailwater cascade resource utilization control system, characterized in that: include: The information acquisition module is used to obtain the data set of rural domestic tailwater status parameters collected by the sensor matrix, as well as the organic matter and ammonia nitrogen effluent quality requirements required for ecological water replenishment input by the user; A data structured processing module, used for performing data structured processing on the data set of rural domestic tail water state parameters to obtain a rural domestic tail water state parameter time series aggregation matrix; A rural domestic tail water state parameter feature extraction module is used to extract features from the rural domestic tail water state parameter time series aggregation matrix to obtain rural domestic tail water state parameter time series associated semantic coding features; A joint coding module, used for jointly coding the organic matter required for ecological water replenishment and the ammonia nitrogen effluent water quality requirements to obtain a joint coding feature of rural tailwater treatment requirements; A cross-modal fusion coding module is used to extract local query prompt information between the rural domestic tailwater status parameter temporal association semantic coding features and the rural tailwater treatment requirement joint coding features, and perform cross-modal fusion coding guided by local query prompt information on the rural domestic tailwater status parameter temporal association semantic coding features and the rural tailwater treatment requirement joint coding features to obtain the treatment requirement-tailwater status parameter temporal semantic joint coding features; The optimization instruction generation module is used to obtain the optimization instruction based on the processing requirement-tailwater state parameter temporal semantic joint coding feature.

Images